CONFIDENTIAL Evidence Dossier

PUBLISH DATE January 22, 2021

National Comprehensive Cancer Network® (NCCN) makes no warranties of any kind whatsoever regarding their content, use or application and disclaims any responsibility for their application or use in any way.

US-FDX-2000073 Jump to Table of Contents

Last update: January 22, 2021

1 EXECUTIVE SUMMARY FoundationOne®CDx is the first United States (US) Food and Drug Administration (FDA)-approved broad companion diagnostic that is clinically and analytically validated for all solid tumors. It functions as a companion diagnostic test to identify patients who may benefit from treatment in accordance with the approved therapeutic product labeling for 21 drug therapies in 7 tumor types and also for 2 drug therapies across solid tumor types (pan tumor). Additionally, FoundationOne CDx is intended to provide tumor mutation profiling to be used by qualified healthcare professionals in accordance with professional guidelines in oncology (FoundationOne CDx Product Description).1 FoundationOne CDx is part of a portfolio of comprehensive genomic profiling (CGP) assays that provides a CGP testing option for any patient with advanced cancer (Figure 1-1).

Figure 1-1. Foundation Medicine Portfolio

DNA, deoxyribonucleic acid; FDA, Food and Drug Administration; RNA, ribonucleic acid. Source: Foundation Medicine, Inc.

Identifying Appropriate Treatment Options in Advanced Cancer Represents Significant Unmet Need Approximately 540,000 people are diagnosed with advanced cancer annually in the US (Figure 1-2) (Epidemiology of Advanced Cancer).2,3 Prognosis remains poor for most types of metastatic solid tumors, with relatively low 5-year survival, ranging from 5% in patients with non-small cell lung cancer (NSCLC) to 39% in patients with head and neck cancer (Historical Treatment of Advanced Cancer).4,5

CONFIDENTIAL PAGE 2 DXUS-F -2000073 Jump to Table of Contents

Last update: January 22, 2021

Figure 1-2. Estimated 2020 Distribution of Stage of Disease Within the Select Solid Tumor Cancers

350,000

300,000

250,000

200,000

150,000

100,000

50,000

0

Stage III or IV All stages a Hepatobiliary only includes HCC for this calculation. CRC, colorectal cancer; HCC, hepatocellular carcinoma; NSCLC, non-small cell lung cancer. Source: Kantar Health 2020.3

Avoidance of therapies that are unlikely to benefit patients and potentially have serious side effects will maximize patient outcomes and reduce costs (Historical Treatment of Advanced Cancer).6-8 . Response rates (RRs) to chemotherapy, the historical standard of care, are typically poor among patients with advanced cancer, with complete responses (CRs) observed in 10% or fewer patients.9 . Further, chemotherapy-related adverse events result in hospitalization in up to 20% of advanced cancer patients.6-8 . The goal is to provide patients with therapies that have a potential to provide benefit, and although all therapies have toxicities, the benefit:risk ratio is of utmost importance when making therapeutic decisions for patients.7 A precision medicine approach can have advantages over cytotoxic regimens. . The use of biomarker-based targeted therapy has been shown to improve treatment response and survival outcomes in patients with actionable alterations for which there is targeted therapy available (either FDA approved or in clinical trials) compared with standard of care chemotherapy or best supportive care (BSC) (Figure 1-3) (Molecularly Matched Therapies Improve Clinical Outcomes).10-12

CONFIDENTIAL PAGE 3 US-FDX-2000073 Jump to Table of Contents

Last update: January 22, 2021

Figure 1-3. Response Rates (A) and Survival (B) With Genomically Matched Therapy vs Nonmatched Therapy Across Tumor Types

35 A 31 B 30

25

20

15

10 5

Overall Respone% Rate, 5

0

Matched therapy Nonmatched therapy

a Meta-analysis comparing patient outcomes from phase 1 studies that used a biomarker-based selection strategy vs those that did not. Source: Schwaederle 2016.12 b Prospective study comparing patients who received genomically matched therapy vs those who received unmatched therapy. Source: Kopetz 2019.13 CI, confidence interval; HR, hazard ratio.

However, the growing complexity of advanced cancer treatment is associated with significant drug spend. . Costs to treat advanced cancer are generally at least 2-fold higher than for earlier-stage disease.14-17 . The recent surge in new treatment options (and the resulting increasing numbers of cancers being treated with available medicines and for longer) is a key driver in spending and growth rates.18-20 It is essential to appropriately select patients for treatment with targeted and immunotherapies. . The majority of cancer types have ≥1 targetable biomarker; despite this, generally fewer than 50% of patients receive molecular testing (Real-World Molecular Testing Patterns).21-26 . According to NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®), molecular testing is recommended for certain patients with bladder cancer, breast cancer, colorectal cancer (CRC), gastric and esophageal cancer, glioblastoma, cutaneous melanoma, NSCLC, ovarian cancer, pancreatic cancer, prostate cancer, uterine cancer, and vulvar cancer; specific recommendations are outlined in Table 7-1.27-41 . As the knowledge of the molecular basis for tumors grows, more therapies are being studied utilizing a genetic-focused development approach as opposed to a tissue histology-based approach, leading to the approvals of several tumor agnostic therapies.42 o Molecular testing for microsatellite instability (MSI) or deficient mismatch repair (dMMR) is now recommended across solid tumors after the approval of pembrolizumab for the treatment of adult and pediatric patients with unresectable or metastatic microsatellite instability-high (MSI-H) or dMMR solid tumors that have progressed following prior treatment and who have no satisfactory alternative treatment options (or for patients with MSI-H or dMMR CRC that has progressed following treatment with

CONFIDENTIAL PAGE 4 US-FDX-2000073 Jump to Table of Contents

Last update: January 22, 2021

fluoropyrimidine, oxaliplatin, and irinotecan).43 o Tumor mutational burden (TMB) was the second tumor-agnostic indication pembrolizumab received; testing for high TMB (defined as ≥10 mutations/megabase [mut/Mb]) is now recommended across multiple solid tumor types as pembrolizumab has an indication for adult and pediatric patients with unresectable or metastatic TMB-H (≥10 mut/Mb) solid tumors, as determined by an FDA-approved test, that have progressed following prior treatment and who have no satisfactory alternative treatment options.43 o Similarly, with the tissue/site-agnostic approvals of the tropomyosin receptor kinase (TRK) inhibitors (larotrectinib, entrectinib), testing is now recommended across multiple solid tumor types (Table 3-1).44,45 . In addition to the many targeted therapies and immunotherapy agents now available, NCCN Guidelines® state that the best management of any patient with cancer is in a clinical trial27-41; approximately 40% of clinical trials utilize the presence of tumor genomic alterations or biomarkers for eligibility and/or stratification (Guideline Recommendations for Molecular Testing).20

A CGP Approach to Testing Optimizes Treatment Selection As the field of molecular profiling is rapidly evolving, there is a shift in focus from a few small, predictive, disease-specific tests to a broader panel testing that can analyze changes in a myriad of genes or gene products (Evidence of Improved Detection of Genomic Alterations With CGP).46 Overall, CGP testing provides valuable information on the presence of actionable biomarkers, which enables healthcare providers to make evidence-based treatment decisions regarding treatments that result in improved outcomes for patients with advanced cancer (Decision Impact of CGP in Clinical Practice).12,47,48 . CGP utilizes next-generation sequencing (NGS) technology to examine entire regions of cancer- relevant genes (in contrast to limited “hotspot” tests) and genes in established cancer pathways for all tumor types, identifying the 4 main classes of genomic alterations (base substitutions, insertions or deletions, copy number alterations [CNAs], gene rearrangements) and reporting complex biomarkers such as TMB and MSI, to inform cancer treatment decisions via a single assay.46,49-52 . Results from CGP can provide information about genomic alteration to guide uses of FDA- approved targeted therapies, potential eligibility for oncology clinical trials, and information that enables physicians to use chemotherapy more effectively (Decision Impact of CGP in Clinical Practice).13,20,53,54

CGP Allows for Improved Detection of Genomic Alterations . Of patients with advanced cancer who receive CGP, 51.7% to 99% will have an actionable alteration that can be matched to either a targeted therapy or to a genomically matched clinical trial.10,13,53,55-67 . In patients who previously underwent conventional testing methods (ie, fluorescence in situ hybridization [FISH], polymerase chain reaction [PCR], single-gene tests, and hotspot testing), CGP identified at least 1 actionable genomic alteration not previously identified in up to 84%

CONFIDENTIAL PAGE 5 US-FDX-2000073 Jump to Table of Contents

Last update: January 22, 2021

across multiple tumor types (Evidence of Improved Detection of Genomic Alterations With CGP).13,61 . Further, CGP has been shown to improve detection of genomic alterations within specific tumor types (Table 1-1).68-71

Table 1-1. Detection of More Patients With Actionable Genomic Alterations With CGP in Specific Tumor Types Melanoma CGP can identify up to 37% more patients with BRAF alterations compared (Boussemart 2019)68 with traditional PCR-based methods Colorectal cancer Of the 6.4% of patients who harbor potentially resistant KRAS mutations outside (Rankin 2016)69 of codons 12 and 13, CGP may be able to identify 88% of those resistance alterations not assessed by focused PCR-based testing Breast cancer CGP can identify patients who harbor multiple PIK3CA mutations that are (Vasan 2019)70 traditionally missed by hotspot testing NSCLC CGP has been shown to identify up to 35% more patients with ALK fusions (Rozenblum 2017)71 and 21% more patients with EGFR alterationsa compared with traditional methods in NSCLC a 41% of these EGFR mutations are common alterations targetable by an FDA-approved therapy in the patient’s tumor type. ALK, anaplastic lymphoma kinase; BRAF, v-raf murine sarcoma viral oncogene homolog B1; CGP, comprehensive genomic profiling; EGFR, epidermal growth factor receptor; KRAS, V-Ki-ras2 Kirsten rat sarcoma; NSCLC, non-small cell lung cancer; PCR, polymerase chain reaction.

Treatment Decisions Informed by CGP Testing Result in Better Outcomes for Patients Across Tumor Types Clinical utility establishes the net clinical benefit to the patient of incorporating CGP to the current standard of care decision making—in effect, answering the question: “Does the intervention (ie, the CGP test) improve patient outcomes?”72

CGP Informs Treatment Decisions in Clinical Practice . A recent national survey using data from the National Survey of Precision Medicine in Cancer Treatment reported that 75.6% of oncologists use multi-marker NGS tumor panels to guide treatment decisions.73 . Up to 50% of patients with a treatment plan informed by CGP testing pursue genomically matched therapy, including on-label and off-label FDA-approved therapies and clinical trial enrollment.53,59-62,64,73,74 . CGP testing has been associated with a 10% to 20% enrollment rate in clinical trials to date compared with a historical enrollment rate of ≤8%; based on a small cohort analysis from phase 1 clinical trials, this may save payers $25,000 per patient through diversion of drug costs to the study sponsor.53,59-61,75,76 . Even for those without or unable to pursue genomically matched options, the personalized treatment plan may confirm chemotherapy as the best option and/or help with discussions about palliative care, thereby avoiding the use of unnecessary therapies.

CONFIDENTIAL PAGE 6 US-FDX-2000073 Jump to Table of Contents

Last update: January 22, 2021

Patients Have Improved Outcomes Following Treatment Informed by CGP . Several pan-tumor and tumor-specific cohort studies have demonstrated substantial improvements in patient outcomes, including RR, progression-free survival (PFS), and overall survival (OS), associated with CGP testing (Evidence of Improved Clinical Outcomes With Treatment Informed by CGP).11,63,66,67,74,75,77-79 . Improvement in outcomes associated with treatment informed by Foundation Medicine CGP testing as reported in peer-reviewed published studies mirrors the improvement in outcomes associated with the use of FDA-approved targeted therapies as reported in respective drugs’ FDA labeling.11,63,66,67,74,75,77-95

Oncology Guidelines Recommend Molecular Testing Across Tumor Histologies According to NCCN Guidelines, molecular testing is recommended for certain patients across multiple tumor types based on the availability of genomically targeted therapies or immunotherapies that improve outcomes in select patients with advanced cancer.27-41 As such, the NCCN Guidelines provide an overview of which patients should undergo molecular testing; these testing recommendations are largely dependent on the tumor histology and stage of disease.27-41 Further, pan-tumor molecular testing is now recommended across solid tumor types for specific therapeutics. . MSI or dMMR status is now recommended across multiple tumor types after the approval of pembrolizumab, the first tissue/site-agnostic approved drug for the treatment of adult and pediatric patients with unresectable or MSI-H or dMMR solid tumors that have progressed following prior treatment and who have no satisfactory alternative treatment options (or for patients with MSI-H or dMMR CRC that has progressed following treatment with a fluoropyrimidine, oxaliplatin, and irinotecan).96 . Similarly, with the tissue/site-agnostic approvals of the neurotrophic tyrosine receptor kinase (NTRK) inhibitors (larotrectinib, entrectinib), testing is now recommended across multiple tumor types.44,45 Importantly, the NCCN Guidelines recommend broad molecular testing across several solid tumor types; the recommendations regarding broad molecular testing and/or the use of NGS for select cancer types are briefly outlined in Table 1-2.30,31,35,39 The National Comprehensive Cancer Network® (NCCN®) recommendations for the specific molecular biomarkers across select solid tumor types are outlined in Table 1-3.27,30,31,34-36,38,39 A more detailed overview of NCCN Guidelines recommendations for molecular testing can be found in Table 7-1.

Table 1-2. NCCN Guidelines Recommendations for Broad Molecular Testing/NGS NCCN Guidelines Recommendations for Broad Molecular Testing/NGS NSCLC V.2.202135 The NCCN Guidelines panel strongly advises broader molecular profiling with the goal of identifying rare driver mutations for which effective drugs may already be available, or to appropriately counsel patients regarding the availability of clinical trials. Broad molecular profiling is a key component of the improvement of the care of patients with NSCLC. Colon cancer V.1.202130 All patients with metastatic colorectal cancer should have tumor tissue genotyped Rectal cancer V.1.202139 for RAS (KRAS and NRAS), BRAF mutations, and HER2 amplifications, individually

CONFIDENTIAL PAGE 7 US-FDX-2000073 Jump to Table of Contents

Last update: January 22, 2021

or as part of an NGS panel. If known RAS/RAF mutation, HER2 testing is not indicated. Testing for MSI may be accomplished by PCR or a validated NGS panel, with the NGS panel especially useful for patients with metastatic disease who require genotyping of RAS and BRAF. Melanoma: Cutaneous Broader genomic profiling can be considered if feasible for patients with stage V.1.202131 III-IV metastatic melanoma or a clinical recurrence, especially if the test results might guide future treatment decisions or eligibility for participation in a clinical trial. MSI, microsatellite instability; NCCN, National Comprehensive Cancer Network; NGS, next generation sequencing; NSCLC, non-small cell lung cancer; PCR, polymerase chain reaction.

Table 1-3. NCCN Guidelines Recommended Biomarkers for Molecular Testing in Select Tumor Types NCCN Guidelines Recommended Biomarkers for Molecular Testing NSCLC V.2.202135 Stage 1B-IIIAa: . EGFR Advanced or metastatic: . EGFR . ALK . ROS1 . BRAF . NTRK1/2/3 . MET exon 14 skipping . RET Breast V.6.202027 At diagnosis and at progressionb: . HER2 Stage IV or recurrent: . MSI/dMMR . NTRK . BRCA1/2 (germline) . PIK3CAc Colon Cancer V.1.202130 At diagnosis: Rectal Cancer V.1.202139 . MSI/dMMR Metastatic: . BRAF . RAS (KRAS, NRAS) . HER2d Ovarian Cancer V.1.202036 At diagnosis: . BRCA1/2 Persistent/recurrent diseasee: . MSI/dMMR Prostate Cancer V.3.202038 Metastatic: . MSI/dMMRf

CONFIDENTIAL PAGE 8 US-FDX-2000073 Jump to Table of Contents

Last update: January 22, 2021

. Tumor testing: HRRg (BRCA1/2, ATM, PALB2, FANCA, RAD51D, CHEK2, CDK12) . Germline testing: BRCA1/2, ATM, PALB2, CHEK2, MLH1, MSH2, MSH6, and PMS2 Cutaneous Melanoma Stage III if future BRAF-directed therapy may be an option: V.1.202131 . BRAF Stage IV or clinical recurrence: . BRAF . KIT (in the appropriate clinical setting) Hepatobiliary Cancerh Unresectable or metastatic disease gallbladder cancer: 34 V.5.2020 . MSI/dMMR . NTRK Unresectable or metastatic intra- or extra-hepatic cholangiocarcinoma: . MSI/dMMR . NTRK . FGFR2 . IDH1 a Consider testing for EGFR mutation on surgical tissue or biopsy for stages IB-IIIA NSCLC. b HER2 testing should be performed on all new primary or newly metastatic breast cancers. c Testing recommended in HR-positive/HER-2 negative breast cancer. d If known RAS/RAF mutation, HER2 testing is not indicated. e Evaluation of HRD can be considered. Additional somatic tumor testing can be considered at the physician’s discretion to identify genetic alterations for which FDA-approved tumor-specific or tumor-agnostic therapy options exist. f Tumor MSI/MMR testing can be considered in patients with regional or castration-naïve metastatic prostate cancer and is recommended in patients with metastatic CRPC. g Germline testing is only for a subset of these genes: BRCA1/2, ATM, PALB2, CHEK2. h The molecular testing recommendations pertain to biliary tract cancer. CRPC, castration-resistant prostate cancer; dMMR, defective DNA mismatch repair; HRR, homologous recombination repair; MSI, microsatellite instability; PD-L1, programmed death-ligand 1.

Clinical Utility and Validity of FoundationOne CDx Across Tumor Types A considerable and increasing body of evidence supports the clinical utility of FoundationOne CDx to match patients with solid tumors to targeted therapies or immunotherapies based on their tumor’s genomic alterations and biomarkers (Table 1-4) (Evidence of Improved Clinical Outcomes With Treatment Informed by CGP); this evidence supports the evolving view of cancer as a disease of the genome as opposed to a disease of a single anatomic site.74,77,97,98 Several pan-tumor and tumor-specific cohort studies have demonstrated substantial improvements in outcomes associated treatment guided by FoundationOne® testing. FoundationOne CDx is the next evolution of FoundationOne and has been established as highly concordant with FoundationOne; therefore, outcomes are expected to be similar.

CONFIDENTIAL PAGE 9 US-FDX-2000073 Jump to Table of Contents

Last update: January 22, 2021

Table 1-4. Selected Outcomesa in Studies Showing the Clinical Utility of Foundation Medicine CGP Clinical impact Matched vs unmatched therapy Author/year Study design Outcome Outcomeb P-value measure Pan-tumor

Schwaederle Retrospective study of the utility of CGP DCRc 34.5% vs 16.1% P<0.02 201674 to match patients with advanced solid malignancies to a therapy (n=347) Kato 201863 Prospective study of the utility of tissue PFS 19.7 months vs 3.5 P=0.008 and liquid CGP in patients with rare months cancers (n=40)d Wheler 201667 Single-arm, nonrandomized study to TTF 2.7 months vs 1.9 P=0.001 prospectively investigate the clinical months utility of CGP in patients with advanced malignancies (N=500) NSCLC Madison A retrospective analysis including 6,491 OS 26.7 months vs 17.9 P=0.035 202099 NSCLC patients with a liquid biopsy months (N=937 tests) and/or tissue biopsy (N=5,582 tests) Singal 201979 A retrospective study to determine the OS 18.6 months vs 11.4 P<0.001 clinical utility of a clinico-genomic months database (using CGP) in patients with NSCLC (n=4,064)e Breast Ganesan Retrospective study of consecutive PFS 6.4 months vs 1.9 P=0.001 201477 patients with advanced or metastatic months triple-negative metastatic breast cancer (N=106) treated in a phase 1 clinic Pancreatic Pishvaian Prospective program (Know Your OS 2.57 years vs 1.51 years P=0.0004 202078 Tumor) using CGP to determine matched therapy in patients with pancreatic cancer (n=1,856) a For a complete representation of all outcomes from these studies, please refer to Table 7-10 and Table 7-11. b All outcomes for PFS, OS, and TTF outlined above are medians. c DCR is defined as stable disease ≥6 months, partial response, or complete response. d Unmatched therapy was last prior unmatched therapy (intrapatient comparison); only 12 patients were included in the PFS analysis. e The unmatched therapy group included only patients with targetable genomic alterations who did not receive matched targeted therapy. CGP, comprehensive genomic profiling; DCR, disease control rate; NSCLC, non-small cell lung cancer; OS, overall survival; PFS, progression-free survival; TTF, time to treatment failure.

CONFIDENTIAL PAGE 10 US-FDX-2000073 Jump to Table of Contents

Last update: January 22, 2021

Summary of Clinical Utility and Validity Data Supporting FoundationOne CDx Companion Diagnostic Claims In addition to providing tumor mutation profiling across a broad range of tumors, FoundationOne CDx is an FDA-approved companion diagnostic test to identify patients who may benefit from treatment in accordance with the approved therapeutic product labeling for 21 drug therapies in 7 tumor types, and also for 2 drug therapies across solid tumor types (pan tumor). FoundationOne CDx had FDA-approved companion diagnostic indications in NSCLC, breast, CRC, ovarian, prostate, melanoma, and cholangiocarcinoma; it also serves as a companion diagnostic across 2 pan-tumor indications (NTRK gene fusions and TMB-H) (Table 1-5). Across these specific tumor types, of the 34 therapies/regimens that require an FDA-approved test, FoundationOne CDx is FDA-approved as a companion diagnostic for 25 (74%).1,100 The sections below highlight the supporting data in patient populations for which FoundationOne CDx has a companion diagnostic claim. For some of the companion diagnostic claims, the FDA approval was based on concordance data with an approved comparator assay. As such, the clinical benefit associated with such claims is described as a part of the prescribing information for the associated targeted therapy. For an overview of the data supporting the specific companion diagnostic claims, please follow the link associated with each tumor type.

Table 1-5. FoundationOne CDx Companion Diagnostic Claims Tumor type Biomarker(s) detected Therapy Pan-tumor NTRK1/2/3 fusions Vitrakvi® (larotrectinib) TMB ≥ 10 mutations per megabase Keytruda® (pembrolizumab) NSCLC EGFR exon 19 deletions and EGFR exon Gilotrif® (afatinib), Iressa® (gefitinib), 21 L858R alterations Tagrisso® (osimertinib), or Tarceva® (erlotinib) EGFR exon 20 T790M alterations Tagrisso® (osimertinib) ALK rearrangements Alecensa® (alectinib), Xalkori® (crizotinib), or Zykadia® (ceritinib) BRAF V600E Tafinlar® (dabrafenib) in combination with Mekinist® (trametinib) MET exon 14 skipping Tabrecta™ (capmatinib) Breast cancer ERBB2 (HER2) amplification Herceptin® (trastuzumab), Kadcyla® (ado-trastuzumab-emtansine), or Perjeta® (pertuzumab) PIK3CA C420R, E542K, E545A, E545D Piqray® (alpelisib) [1635G>T only], E545G, E545K, Q546E, Q546R, H1047L, H1047R, and H1047Y alterations CRC KRAS wild-type (absence of mutations in Erbitux® (cetuximab) codons 12 and 13)

CONFIDENTIAL PAGE 11 US-FDX-2000073 Jump to Table of Contents

Last update: January 22, 2021

KRAS wild-type (absence of mutations in Vectibix® (panitumumab) exons 2, 3, and 4) and NRAS wild type (absence of mutations in exons 2, 3, and 4) Ovarian cancer BRCA1/2 alterations Lynparza® (olaparib) or Rubraca® (rucaparib) Prostate cancer HRR gene (BRCA1, BRCA2, ATM, Lynparza® (olaparib) BARD1, BRIP1, CDK12, CHEK1, CHEK2, FANCL, PALB2, RAD51B, RAD51C, RAD51D and RAD54L) alterations Melanoma BRAF V600E Tafinlar® (dabrafenib) or Zelboraf® (vemurafenib) BRAF V600E and V600K Mekinist® (trametinib) or Cotellic® (cobimetinib) in combination with Zelboraf® (vemurafenib) Cholangiocarcinoma FGFR2 fusions and select rearrangements Pemazyre™ (pemigatinib) CRC, colorectal cancer; HRR, homologous recombination repair; NSCLC, non-small cell lung cancer. Source: FoundationOne CDx Label1

Pan-tumor FoundationOne CDx is a validated, FDA-approved companion diagnostic assay for larotrectinib and pembrolizumab for pan-tumor indications. Specifically for pembrolizumab, FoundationOne CDx is FDA- approved to identify solid cancer patients with tumor mutational burden-high (TMB-H) tumors, defined as ≥10 mut/Mb. FoundationOne CDx consistently showed high positive percent agreement (PPA) and negative percent agreement (NPA) with validated assays for detection of the predictive pan-tumor biomarkers, TMB and NTRK gene fusions; results from concordance studies are summarized in Table 1-6. Further, the effectiveness of the FoundationOne CDx assay in identifying patients with TMB-H (≥10 mut/Mb) was demonstrated through a prospectively-planned retrospective analysis of clinical specimens from the patients enrolled in the KEYNOTE-158 clinical trial. In addition to the concordance data, FoundationOne CDx has clinical bridging data reporting clinical outcome results similar to those seen when the biomarker status was determined using the local clinical trial assay (CTA) in the pooled analysis of 3 clinical trials for larotrectinib (Table 1-7).1,43,101,102 For additional information concerning the clinical utility of FoundationOne CDx in pan-tumor companion diagnostic claims, please refer to Clinical Utility and Validity of FoundationOne CDx in Pan Tumors.

Table 1-6. Clinical Validity of FoundationOne CDx for Pan-tumor Companion Diagnostic Claims Biomarker PPA, % (n/N) NPA, % (n/N) Comparator assay TMB 87.28 (51/59)a 91.56 (138/159)a CLIA validated WES assay NTRK 84.1 (37/44)b 100 (226/226)b CTA a The overall PPA and NPA were calculated based on a weighted average of the results (Set A and Set B) in the TMB concordance analysis. b Concordance between FoundationOne CDx and the CTA for NTRK gene fusions excludes invalid FoundationOne CDx test results.

CONFIDENTIAL PAGE 12 US-FDX-2000073 Jump to Table of Contents

Last update: January 22, 2021

CLIA, Clinical Laboratory Improvement Amendments; CTA, clinical trial assay; NPA, negative percent agreement; PPA, positive percent agreement; TMB, tumor mutational burden; WES, whole exome sequencing. Source: FoundationOne CDx Label1

Table 1-7. Clinical Utility of FoundationOne CDx for Pan-tumor Companion Diagnostic Claims Clinical FoundationOne CTA results Biomarker Therapy endpoint CDx results N=91 N=102 ORRb, % (n/N) TMBa Pembrolizumab 33 (30/91) 29 (30/102) (95% CI) (24, 44) (21, 39) N=26 N=55 ORRc, % (n/N) NTRK Larotrectinib 77 (20/26) 75 (41/55) (95% CI) (56, 91) (61, 85) a FoundationOne CDx is a companion diagnostic for pembrolizumab for identifying TMB-H defined as ≥10 mut/Mb. b ORR was assessed per central radiology using RECIST v1.1. c ORR was assessed by independent review committee using RECIST v1.1. CI, confidence interval; CTA, clinical trial assay; ORR, overall response rate; RECIST, Response Evaluation Criteria in Solid Tumors; TMB, tumor mutational burden. Source: FoundationOne CDx Label1; FoundationOne CDx SSED 016103; FoundationOne CDx SSED 017104

NSCLC FoundationOne CDx is an FDA-approved companion diagnostic to determine treatment with afatinib, erlotinib, gefitinib, osimertinib, alectinib, crizotinib, ceritinib, dabrafenib and trametinib, and capmatinib in patients with NSCLC.1

. The companion diagnostic approvals for EGFR mutations (afatinib, erlotinib, gefitinib, osimertinib) and ALK rearrangements (alectinib, crizotinib, ceritinib) were based on high concordance to the clinically validated FDA-approved assays, cobas EGFR mutation Test v2 and Ventana ALK CDx assay, respectively (Table 1-8).1 . As a companion diagnostic for the detection of MET exon 14 skipping mutations for treatment with capmatinib, FoundationOne CDx was highly concordant to the CTA (Table 1-8).1 Additionally, FoundationOne CDx has clinical bridging data reporting clinical outcome results similar to those seen when the biomarker status was determined using the CTA for treatment with capmatinib in the pivotal GEOMETRY mono-1 clinical trial (Table 1-9).1,105 . The companion diagnostic approval for FoundationOne CDx for dabrafenib and trametinib in patients with BRAF V600E mutated NSCLC is based on concordance data extrapolated from melanoma. Please refer to Table 1-15 for an overview of this concordance data.1

For additional information concerning the clinical utility of FoundationOne CDx in NSCLC, please refer to Clinical Utility and Validity of FoundationOne CDx in NSCLC.

Table 1-8. Clinical Validity of FoundationOne CDx for NSCLC Companion Diagnostic Claims Biomarker PPA, % (n/N) NPA, % (n/N) Comparator assay

CONFIDENTIAL PAGE 13 US-FDX-2000073 Jump to Table of Contents

Last update: January 22, 2021

EGFR exon 19 98.1 (106/108) 99.4 (153/154) cobas® EGFR Mutation Test v2 deletions and exon 21 L858R EGFR T790M 98.9 (87/88) 86.1 (93/108) cobas® EGFR Mutation Test v2 ALK rearrangements 92.9 (78/84) 100 (75/75) Ventana ALK (D5F3) CDx Assay Vysis ALK Break Apart FISH Probe Kit MET exon 14 98.6 (72/73) 100 (125/125) reverse transcriptase-PCR CTA skipping CTA, clinical trial assay; FISH, fluorescence in situ hybridization; NPA, negative percent agreement; PCR, polymerase chain reaction; PPA, positive percent agreement. Source: FoundationOne CDx Label1

Table 1-9. Clinical Utility of FoundationOne CDx for NSCLC Companion Diagnostic Claims Clinical FoundationOne CTA results Biomarker Therapy endpoint CDx results Cohort 4b N=52 N=69 44.2 (23/52) 40.6 (28/69) MET exon 14 ORRa, % (n/N) (30.6, 58.7) (28.9, 53.1) Capmatinib skipping (95% CI) Cohort 5bc N=20 N=28 70 (14/20) 67.9 (19/28) (45.7, 88.1) (47.6, 84.1) a ORR as assessed by BICR according to RECIST v1.1. b Cohort 4: previously treated patients. c Cohort 5b: treatment-naïve patients. BICR, blinded independent central review; CI, confidence interval; CTA, clinical trial assay; ORR, overall response rate; RECIST, Response Evaluation Criteria in Solid Tumors. Source: FoundationOne CDx Label1; FoundationOne CDx SSED 011106

Breast cancer FoundationOne CDx is an FDA-approved companion diagnostic to determine treatment with trastuzumab, ado-trastuzumab-emtansine, pertuzumab, and alpelisib in patients with NSCLC.

. The companion diagnostic approvals for ERBB2 (HER2) amplifications (trastuzumab, ado- trastuzumab-emtansine, pertuzumab) were based on high concordance to the clinically validated, FDA-approved assay, Dako HER2 FISH PharmDx kit (Table 1-10).1 . As a companion diagnostic for the detection of PIK3CA mutations for treatment with alpelisib + fulvestrant, FoundationOne CDx was highly concordant to the PCR-based CTA (Table 1-10). Additionally, the clinical bridging data for FoundationOne CDx reported clinical outcome results similar to those seen when the biomarker status was determined using the CTA for treatment with

CONFIDENTIAL PAGE 14 US-FDX-2000073 Jump to Table of Contents

Last update: January 22, 2021

alpelisib + fulvestrant in the pivotal SOLAR-1 trial (Table 1-11).1,107 For additional information concerning the clinical utility of FoundationOne CDx in breast cancer companion diagnostic claims, please refer to Clinical Utility and Validity of FoundationOne CDx in Breast Cancer.

Table 1-10. Clinical Validity of FoundationOne CDx for Breast Cancer Companion Diagnostic Claims Biomarker PPA, % (n/N) NPA, % (n/N) Comparator assay ERBB2 (HER2) 89.4 (101/113) 98.4 (180/183) Dako HER2 FISH PharmDx™ Kit amplifications PIK3CA 93.8 (106/113) 98.8 (159/161) PCR-based PIK3CA hot-spot CTA 1 91.6 (197/215) 98.8 (162/164) PCR-based PIK3CA hot-spot CTA 2 CTA, clinical trial assay; FISH, fluorescence in situ hybridization; NPA, negative percent agreement; PCR, polymerase chain reaction; PPA, positive percent agreement. Source: FoundationOne CDx Label1

Table 1-11. Clinical Utility of FoundationOne CDx for Breast Cancer Companion Diagnostic Claims Biomarker Therapy Clinical FoundationOne CDx CTA results endpoint results N=56 N=169 11.2c 11.0 Alpelisib + PFSa, months 0.52 (0.29, 0.93) 0.65 (0.50, 0.85) PIK3CA fulvestrant HRb (95% CI) N=42 10.9d 0.35 (0.16, 0.77)

a PFS by investigator assessment in patients with PIK3CA alteration positive tumors. b The HR shown here for both the FoundationOne CDx results and the CTA results is for alpelisib + fulvestrant for risk of disease progression or death compared to placebo. c PFS in the CTA1-enrolled positive and FoundationOne CDx-positive patients. d PFS in the CTA2-enrolled positive and FoundationOne CDx-positive patients. CI, confidence interval; CTA, clinical trial assay; HR, hazard ratio; PFS, progression-free survival. Source: FoundationOne CDx Label1

CRC FoundationOne CDx is an FDA-approved companion diagnostic to determine treatment with cetuximab and panitumumab in patients with CRC. The companion diagnostic approvals for KRAS and NRAS wild- type were based on high concordance to the clinically validated, FDA-approved assay, therascreen KRAS RGQ polymerase chain reaction (PCR) kit (Table 1-12).1

CONFIDENTIAL PAGE 15 US-FDX-2000073 Jump to Table of Contents

Last update: January 22, 2021

For additional information concerning the clinical utility of FoundationOne CDx in CRC companion diagnostic claims, please refer to Clinical Utility and Validity of FoundationOne CDx in CRC.

Table 1-12. Clinical Validity of FoundationOne CDx for CRC Companion Diagnostic Claims Biomarker PPA, % (n/N) NPA, % (n/N) Comparator assay KRAS 100 (173/173) 100 (154/154) therascreen® KRAS RGQ PCR Kit NPA, negative percent agreement; PCR, polymerase chain reaction; PPA, positive percent agreement. Source: FoundationOne CDx Label1

Ovarian cancer FoundationOne CDx is an FDA-approved companion diagnostic to determine treatment with olaparib or rucaparib in patients with ovarian cancer.1

. FoundationOne CDx demonstrated equivalence to FoundationFocus CDxBRCA, the original companion diagnostic for detection of BRCA1/2 mutations in formalin-fixed paraffin-embedded (FFPE) ovarian cancer tumor specimens to select for rucaparib eligibility.1,108 FoundationOne CDx replaced FoundationFocus in the market in 2018. . For olaparib in ovarian cancer, the FoundationOne CDx CTA was used to identify somatic BRCA1/2 alterations on prospectively collected tumor samples for patients enrolled in the clinical trial (SOLO-1) based on local (germline or somatic) or central (germline) testing. In the SOLO-1 trial, after a median follow-up of 4.8 years in patients treated with maintenance olaparib and 5 years in placebo-treated patients, the median PFS was 56.0 months vs 13.8 months for patients treated with olaparib vs placebo, respectively (HR: 0.33; 95% CI: 0.25, 0.43).1,109,110 . Both olaparib and rucaparib showed improved clinical outcomes in those patients with BRCA1/2 mutations as determined by FoundationOne CDx as compared to those patients who received placebo (Table 1-13).1

For additional information concerning the clinical utility of FoundationOne CDx in ovarian cancer companion diagnostic claims, please refer to Clinical Utility and Validity of FoundationOne CDx in Ovarian Cancer.

Table 1-13. Clinical Utility of FoundationOne CDx for Ovarian Cancer Companion Diagnostic Claims Biomarker Therapy Clinical FoundationOne Full analysis set endpoint CDx results results BRCA1/2 Olaparib PFSa, months N=206 N=260 b HR (95% CI) Not reached Not reached 0.28 (0.20, 0.38) 0.30 (0.23, 0.41) N=124 NR PFSc, months Rucaparib 16.6 NR HRb (95% CI) 0.24 (0.16, 0.36) a Investigator assessed median PFS evaluated according to RECIST v1.1.

CONFIDENTIAL PAGE 16 US-FDX-2000073 Jump to Table of Contents

Last update: January 22, 2021

a HR for both FoundationOne CDx and full analysis set (olaparib only) compares olaparib or rucaparib to placebo for risk of disease progression or death. c Investigator assessed median PFS. CI, confidence interval; HR, hazard ratio; NR, not reported; PFS, progression-free survival; RECIST, Response Evaluation Criteria in Solid Tumors. Source: FoundationOne CDx Label1

Prostate cancer FoundationOne CDx is an FDA-approved companion diagnostic to determine treatment with olaparib in patients with prostate cancer.1 FoundationOne CTA was the registrational assay used in the PROfound trial.1,111 In the PROfound study, the median OS for olaparib was 17.4 months vs 13.6 months for placebo in patients with HRR mutations (BRCA1, BRCA2, ATM, BARD1, BRIP1, CDK12, CHEK1, CHEK2, FANCL, PALB2, RAD51B, RAD51C, RAD51D and RAD54L) (HR: 0.76, 95% CI: 0.58, 1.00). Olaparib showed improved clinical outcomes in those patients with HRR mutations as determined by FoundationOne CDx as compared to those patients who received placebo (Table 1-14).1

For additional information concerning the clinical utility of FoundationOne CDx in prostate cancer companion diagnostic claims, please refer to Clinical Utility and Validity of FoundationOne CDx in Prostate Cancer.

Table 1-14. Clinical Utility of FoundationOne CDx for Prostate Cancer Companion Diagnostic Claims Biomarker Therapy Clinical FoundationOne Full analysis set endpoint CDx results results N=248 N=256 HRR (BRCA1, BRCA2, ATM, BARD1, BRIP1, CDK12, rPFSa, months 6.2 5.8 CHEK1, CHEK2, FANCL, Olaparib b PALB2, RAD51B, RAD51C, HR (95% CI) 0.49 (0.38, 0.63) 0.49 (0.38, 0.63) RAD51D and RAD54L) a rPFS based on BICR using RECIST v1.1 and/or PCWG3, or death (by any cause in the absence of progression) regardless of whether the patient withdrew from randomized therapy or received another anticancer therapy prior to progression. b HR for both FoundationOne CDx and full analysis set compares olaparib to investigator’s choice of therapy (either enzalutamide 160 mg orally once daily or abiraterone acetate 1000 mg orally once daily with prednisone 5 mg orally twice daily [prednisolone was permitted for use instead of prednisone, if necessary]) for radiological disease progression or death. BICR, blinded independent central review; CI, confidence interval; HR, hazard ratio; HRR, homologous recombination repair; ORR, overall response rate; PCWG3, Prostate Cancer Working Group 3; RECIST, Response Evaluation Criteria in Solid Tumors; rPFS, radiological progression-free survival. Source: FoundationOne CDx Label1

Melanoma FoundationOne CDx is an FDA-approved companion diagnostic to determine treatment with dabrafenib ± trametinib and vemurafenib ± trametinib or cobimetinib in patients with melanoma. The companion diagnostic approvals for BRAF mutations were based on high concordance to the clinically validated FDA-approved assays, cobas BRAF V600 mutation Test and THxID BRAF kit (Table 1-15).1

CONFIDENTIAL PAGE 17 US-FDX-2000073 Jump to Table of Contents

Last update: January 22, 2021

For additional information concerning the clinical utility of FoundationOne CDx in melanoma companion diagnostic claims, please refer to Clinical Utility and Validity of FoundationOne CDx in Melanoma.

Table 1-15. Clinical Validity of FoundationOne CDx for Melanoma Companion Diagnostic Claims Biomarker PPA, % (n/N) NPA, % (n/N) Comparator assay BRAF V600 99.4 (166/167) 89.6 (121/135)a cobas BRAF V600 Mutation Test BRAF V600E 99.3 (149/150) 99.2 (121/122) BRAF V600 dinucleotide 96.3 (26/27) 100 (24/24) THxID™ BRAF kit a The reported difference in NPA values for BRAF V600 and BRAF V600E are likely attributed to known sensitivity differences in the cobas test, which has lower sensitivity for detection of dinucleotide V600 alterations than for the single nucleotide V600E c.1799T>A alteration, especially for samples in which FoundationOne CDx detected the nucleotides to be of lower than 40% mutational allele frequency, leading to low NPA values. NPA, negative percent agreement; PPA, positive percent agreement. Source: FoundationOne CDx Label1

Cholangiocarcinoma FoundationOne CDx is an FDA-approved companion diagnostic to determine treatment with pemigatinib in patients with cholangiocarcinoma.1 FoundationOne CDx has demonstrated concordance to FoundationOne, the original CTA, for the detection of FGFR2 fusions and rearrangements, and has since replaced it in the market (Table 1-16).1 Additionally, the clinical bridging data for FoundationOne CDx reported clinical outcome results similar to those seen when the biomarker status was determined using the FoundationOne CTA for treatment with pemigatinib in the phase 2 FIGHT-202 clinical trial (Table 1-17).1,112

For additional information concerning the clinical utility of FoundationOne CDx in cholangiocarcinoma companion diagnostic claims, please refer to Clinical Utility and Validity of FoundationOne CDx in Cholangiocarcinoma.

Table 1-16. Clinical Validity of FoundationOne CDx for Companion Diagnostic Claims Biomarker PPA, % (n/N) NPA, % (n/N) Comparator assay FGFR2 100 (84/84) 100 (97/97) FoundationOne CTA CTA, clinical trial assay; NPA, negative percent agreement; PPA, positive percent agreement. Source: FoundationOne CDx Label1

Table 1-17. Clinical Utility of FoundationOne CDx for Companion Diagnostic Claims Biomarker Therapy Clinical FoundationOne FoundationOne endpoint CDx results CTA results N=80 N=107 FGFR2 Pemigatinib ORRa, % 37.50 35.51 (95% CI) (26.92, 49.04) (26.50, 45.35) a ORR per central review per RECIST v1.1.

CONFIDENTIAL PAGE 18 US-FDX-2000073 Jump to Table of Contents

Last update: January 22, 2021

CI, confidence interval; CTA, clinical trial assay; ORR, overall response rate; RECIST, Response Evaluation Criteria in Solid Tumors. Source: FoundationOne CDx Label1

Economic Impact of CGP As shown by the economic modeling that has been reported to date for CGP, and specifically Foundation Medicine CGP, patients who utilize CGP may have slightly increased total costs in comparison to those who utilize non-CGP molecular testing; the increase primarily occurs because such testing achieves the ultimate goal in oncology—meaningful prolongation of life (Comprehensive genomic profiling compared with conventional testing).113,114 . The primary cost driver among patients utilizing CGP is the cost of using effective drugs for a longer period of time. o In particular, weekly patient costs remain similar for patients treated with matched therapies compared with traditional approaches to therapy, although patients receiving precision medicine-based treatments often incur higher medication costs that may be offset in part because of lower costs for managing toxicity.11 . Additionally, CGP facilitates identification of patients for clinical trials, thus potentially diverting substantial drug costs to study sponsors (Potential savings associated with clinical trial enrollment).59,61

For the most up-to-date information related to the economic value associated with FoundationOne CDx, please contact Foundation Medicine Payer Relations.

Conclusion As evidence shows molecularly targeted therapies are efficacious in subgroups of patients, it is increasingly important to define these subgroups using an accurate and efficient molecular testing method, such as CGP.46 . A comprehensive approach to testing with FoundationOne CDx leads to improved outcomes of treatment response and survival, with a manageable budget impact that is driven by longer duration of effective therapy for patients.115 . Another advantage of using FoundationOne CDx is that it aligns with nationally recognized guidelines by establishing molecular eligibility for clinical trials of investigational agents, thereby increasing the likelihood of deriving clinically meaningful benefit from anticancer therapy. . Preliminary economic analyses have shown that these important clinical benefits may be accompanied by a modest increase in cost. Overall, the use of FoundationOne CDx may improve the current treatment paradigm of patients with advanced cancer by helping physicians personalize selection of the most suitable treatment to maximize their likelihood of response and minimize treatment toxicity.

CONFIDENTIAL PAGE 19 US-FDX-2000073 Last update: January 22, 2021

TABLE OF CONTENTS

1 EXECUTIVE SUMMARY ...... 2

IDENTIFYING APPROPRIATE TREATMENT OPTIONS IN ADVANCED CANCER REPRESENTS SIGNIFICANT UNMET NEED ...... 2

A CGP APPROACH TO TESTING OPTIMIZES TREATMENT SELECTION...... 5 CGP Allows for Improved Detection of Genomic Alterations ...... 5

TREATMENT DECISIONS INFORMED BY CGP TESTING RESULT IN BETTER OUTCOMES FOR PATIENTS ACROSS TUMOR TYPES ...... 6 CGP Informs Treatment Decisions in Clinical Practice ...... 6 Patients Have Improved Outcomes Following Treatment Informed by CGP ...... 7 Oncology Guidelines Recommend Molecular Testing Across Tumor Histologies ...... 7

CLINICAL UTILITY AND VALIDITY OF FOUNDATIONONE CDX ACROSS TUMOR TYPES ...... 9 Summary of Clinical Utility and Validity Data Supporting FoundationOne CDx Companion Diagnostic Claims ...... 11

ECONOMIC IMPACT OF CGP ...... 19

CONCLUSION ...... 19 TABLE OF CONTENTS ...... 20 2 UNMET NEED AND RATIONALE FOR MOLECULAR TESTING ...... 23

EPIDEMIOLOGY OF ADVANCED CANCER ...... 23

HISTORICAL TREATMENT OF ADVANCED CANCER ...... 25

COMPLEXITY OF THE CURRENT TREATMENT PARADIGM ...... 26

NCCN GUIDELINES RECOMMENDATIONS FOR MOLECULAR TESTING ...... 26

MOLECULARLY MATCHED THERAPIES IMPROVE CLINICAL OUTCOMES ...... 29 Targeted Therapy and Immunotherapy ...... 29

REAL-WORLD MOLECULAR TESTING PATTERNS ...... 33 Barriers to Guideline Adherence for Molecular Testing ...... 35

IMPROVED DETECTION OF GENOMIC ALTERATIONS WITH COMPREHENSIVE GENOMIC PROFILING ...... 35 Evidence of Improved Detection of Genomic Alterations With CGP ...... 36 3 PRODUCT DESCRIPTION ...... 38

FOUNDATION MEDICINE PORTFOLIO OF CGP TESTS ...... 38

FOUNDATION MEDICINE DECISION SUPPORT ...... 38

FOUNDATIONONE CDX PRODUCT DESCRIPTION ...... 39

ANALYTIC VALIDITY OF FOUNDATIONONE CDX...... 43

CONFIDENTIAL PAGE 20 US-FDX-2000073 Last update: January 22, 2021

Sample validation and concordance ...... 43 TMB ...... 48 MSI ...... 49 LOH ...... 49

MEDICAL POLICY COVERAGE OF FOUNDATIONONE CDX ...... 49 4 CLINICAL UTILITY OF FOUNDATIONONE CDX ...... 51

EVIDENCE OF IMPROVED CLINICAL OUTCOMES WITH TREATMENT INFORMED BY CGP ...... 51 Decision Impact of CGP in Clinical Practice ...... 54 Guideline Recommendations for CGP ...... 54

EVIDENCE OF IMPROVED CLINICAL OUTCOMES WITH FOUNDATIONONE CDX ...... 56 5 FOUNDATIONONE CDX FOR COMPANION DIAGNOSTIC CLAIMS ...... 57

CGP IN PAN-TUMORS ...... 58 Unmet Need of CGP in Pan-Tumors ...... 58 Place of CGP in the Pan-Tumor Evaluation of TMB, MSI-H, and NTRK ...... 62 Clinical Utility and Validity of FoundationOne CDx in Pan Tumors ...... 65

CGP IN NSCLC ...... 70 Unmet Need for Molecular Testing in NSCLC ...... 70 Place of CGP in NSCLC ...... 71 Clinical Utility and Validity of FoundationOne CDx in NSCLC ...... 73

CGP IN BREAST CANCER ...... 77 Unmet Need for Molecular Testing in Breast Cancer ...... 77 Place of CGP in Breast Cancer...... 79 Clinical Utility and Validity of FoundationOne CDx in Breast Cancer ...... 80

CGP IN CRC ...... 83 Unmet Need for Molecular Testing in CRC ...... 83 Place of CGP in CRC ...... 85 Clinical Utility and Validity of FoundationOne CDx in CRC ...... 85

CGP IN OVARIAN CANCER ...... 87 Unmet Need for Molecular Testing in Ovarian Cancer ...... 87 Place of CGP in Ovarian Cancer ...... 88 Clinical Utility and Validity of FoundationOne CDx in Ovarian Cancer ...... 89

CGP IN PROSTATE CANCER ...... 92 Unmet Need for Molecular Testing in Prostate Cancer ...... 92

CONFIDENTIAL PAGE 21 US-FDX-2000073 Last update: January 22, 2021

Place of CGP in Prostate Cancer...... 94 Clinical Utility and Validity of FoundationOne CDx in Prostate Cancer ...... 95

CGP IN MELANOMA ...... 97 Unmet Need for Molecular Testing in Melanoma...... 97 Place of CGP in Melanoma ...... 98 Clinical Utility and Validity of FoundationOne CDx in Melanoma ...... 99

CGP IN CHOLANGIOCARCINOMA ...... 101 Unmet Need for Molecular Testing in Cholangiocarcinoma ...... 101 Place of CGP in Cholangiocarcinoma ...... 102 Clinical Utility and Validity of FoundationOne CDx in Cholangiocarcinoma ...... 102 6 ECONOMIC VALUE OF FOUNDATIONONE CDX ...... 105

ECONOMIC BENEFITS ASSOCIATED WITH CGP IN ADVANCED CANCER ...... 105 Potential savings associated with clinical trial enrollment ...... 105 Matched compared with unmatched therapy ...... 106 Comprehensive genomic profiling compared with conventional testing ...... 109 7 APPENDICES ...... 112

TABLES ...... 112

FIGURES ...... 115

TERMS AND DEFINITIONS ...... 116

LIST OF ABBREVIATIONS ...... 123

NCCN GUIDELINES RECOMMENDATIONS FOR MOLECULAR TESTING ...... 126

FOUNDATIONONE PORTFOLIO DESCRIPTION OF DECISION SUPPORT SERVICES ...... 145

FOUNDATIONONE CDX PRODUCT DESCRIPTION ...... 146 FoundationOne CDx Sample Report ...... 149

ANALYTIC VALIDITY OF FOUNDATIONONE CDX...... 153 Concordance of FoundationOne CDx to Externally Validated NGS Assays ...... 153

CLINICAL VALIDITY OF FOUNDATIONONE CDX ...... 154 Additional Studies Supporting Clinical Validity of FoundationOne CDx by Tumor Type ...... 154

CLINICAL VALIDITY OF PREVIOUS VERSIONS OF FOUNDATION MEDICINE TISSUE ASSAY ...... 156

CLINICAL UTILITY OF PREVIOUS VERSIONS OF FOUNDATION MEDICINE TISSUE ASSAY ...... 162

CLINICAL UTILITY OF NGS ...... 176

REFERENCES ...... 183

CONFIDENTIAL PAGE 22 US-FDX-2000073 Last update: January 22, 2021

2 UNMET NEED AND RATIONALE FOR MOLECULAR TESTING

. Approximately one-third of patients are diagnosed with their cancer in the advanced stage in the US; survival at 5 years is poor (ranging from 39% to <2% depending on cancer site).116,2,3,5 . Treatment in advanced tumors has been largely based on tumor site, histology, tumor stage, and prior response to therapy, with the majority of patients receiving chemotherapy; however, response rates (RRs) to chemotherapy are typically poor among patients with advanced cancer, with CRs observed in 10% or fewer of patients.9,117 . Both molecularly targeted therapies and immunotherapies have shown improved outcomes of RRs and survival in patients with a corresponding biomarker over standard of care therapy across multiple advanced solid tumors.10,11 118 12,48,119,52 . Even though guideline recommendations are in place and there is ample evidence supporting molecular testing and biomarker-based therapies, recent evidence shows the majority of advanced cancer patients do not receive molecular testing.113

Epidemiology of Advanced Cancer In the US, approximately 1.8 million people will be diagnosed with cancer and an estimated 606,520 cancer-related deaths will occur in 2020.116 Approximately one-third of patients diagnosed with a solid tumor will be classified as advanced (defined as stage III or IV cancer).2,3 Based on this estimation, approximately 540,000 people in the US will be diagnosed with advanced cancer per year.2,3,116 It is estimated the 10 most common solid tumor cancers in the US, with an incidence of 1,420,048, will comprise approximately 75% of all cancer diagnoses in 2020.3,116 Among these cancer types, approximately 33% of patients will have stage III or IV disease at diagnosis, equivalent to an incidence of 426,014.2,3 The distribution of stage of diagnosis within select cancer types is illustrated in Figure 2-1.

CONFIDENTIAL PAGE 23 US-FDX-2000073 Last update: January 22, 2021

Figure 2-1. Estimated 2020 Distribution of Stage of Disease Within the Select Solid Tumor Cancers

350,000

300,000

250,000

200,000

150,000

100,000

50,000

0

Stage III or IV All stages

a Hepatobiliary only includes HCC for this calculation. CRC, colorectal cancer; HCC, hepatocellular carcinoma; NSCLC, non-small cell lung cancer. Source: Kantar Health 2020.3

Although many cancers, including prostate cancer and breast cancer, are predominantly diagnosed early in the disease course, even with optimal treatment, up to 30% of patients will relapse and present with metastatic disease.120,121 Prognosis remains poor for most types of metastatic cancers, with relatively low 5-year survival when diagnosed as distant. Examples of 5-year relative survival rates in patients diagnosed at distant stage for select tumor types, according to Surveillance, Epidemiology, and End Results (SEER) are shown in Figure 2-2.

CONFIDENTIAL PAGE 24 US-FDX-2000073 Last update: January 22, 2021

Figure 2-2. 5-Year Survival Rates in Select Metastatic Solid Tumorsa

Prostate 31

Ovarian 29

Breast 27

Melanoma 25

CRC 14

Lung 5

Pancreas 3

Hepatobiliary 2

0 10 20 30 40 50 60 70 80 90 100

5-year relative survival rate, %

a Data based on the NCI SEER Program in the US using data from 2009 to 2015. CRC, colorectal cancer; NCI, National Cancer Institute; SEER, Surveillance, Epidemiology, and End Results; US, United States Source: Siegel 2020.5

Historical Treatment of Advanced Cancer The diagnosis and treatment selection of all tumor types has historically relied almost exclusively on clinical and pathologic features of the tumor. Prior to the introduction of immunotherapy and targeted therapy, standard of care treatment options for patients with advanced cancer included chemotherapy, radiotherapy, hormone therapy, or surgery. However, RRs to chemotherapy are typically poor among patients with advanced cancer, with CRs observed in 10% or fewer of patients.9 For example, a meta-analysis of 68 chemotherapy trials (2,732 patients) reported a CR rate following cytotoxic chemotherapy treatment of 7.4% (95% confidence interval [CI]: 6.3, 8.4) in patients with late- stage cancer, regardless of tumor type or drug regimen used.9 The PR rate in this analysis was 27.9%, meaning a total of 35.3% of patients were considered “responders” (ie, CR + PR) to chemotherapy.9 Conversely, 64.7% of patients were nonresponders to therapy (ie, did not achieve a PR or better with cytotoxic chemotherapy).9

CONFIDENTIAL PAGE 25 US-FDX-2000073 Last update: January 22, 2021

Despite significant differences between tumor types (P=0.028), the individual CR rates (defined as disappearance of all cancer as a result of chemotherapy) did not exceed a mean rate of 11% and included:9 . 3.2% for patients with pancreatic cancer; . 5.0% for patients with NSCLC; . 6.1% for patients with ovarian cancer; . 6.7% for patients with CRC; . 8.5% for patients with melanoma; . 10.1% for patients with breast cancer; and . 10.9% for patients with prostate cancer.

Complexity of the Current Treatment Paradigm

Treatment selection in advanced tumors has been largely based on tumor site, histology, tumor stage, and prior response to therapy.117 However, improvements in our understanding of the biology of cancers have led to the identification of new biomarkers to help guide treatment selection for individual patients. The natural history of solid tumors is a process of multistep carcinogenesis and tumor growth that is driven by changes in the genomic landscape. The genomic status of the tumor, in addition to its location, has substantial implications for effective patient management.122-126 As such, there has been a shift in the treatment approach of oncology patients toward biomarker-based therapies, the aim of which is to identify interventions likely to be of most benefit to patients based upon features of the individual or their disease.127 The discovery of multiple new genomic and other biomarkers for use with targeted therapy and immunotherapy has revolutionized the patient journey and treatment paradigm, shifting away from trial-and-error methods and toward biomarker-based selection of the most rational treatment to maximize the likelihood of treatment response and survival. Although therapy targeted to a biomarker has been shown to improve outcomes in patients with advanced cancer, there are many factors physicians must now consider and incorporate into treatment decision making, adding to the complexity of effectively treating these patients.10-12 Physician understanding of both appropriate molecular testing as well as the results of such testing are vital to ensuring patients receive appropriate treatment. In a recent survey of the 20 top US cancer centers and hospitals, all the respondents indicated genetic test results are reported in patients’ electronic medical records in some manner; however, 45% indicated that they don’t have decision support tools in place to access relevant genomic information when they need it in patient care.128 Although clinical guidelines are noted to be effective tools for uniformly and sustainably delivering optimal, quality-focused, patient-centric, safe care, adherence to guideline recommendations are variable.129 Precision medicine can be very beneficial as it effectively reduces overtreatment by removing more extensive treatment options from consideration if deemed by clinicians to be futile; however, these tests must be ordered, and the results must be understood.130

NCCN Guidelines Recommendations for Molecular Testing According to NCCN Guidelines, molecular testing is recommended for certain patients with breast cancer, CRC, NSCLC, cutaneous melanoma, ovarian cancer, prostate cancer, endometrial cancer, gastric cancer, esophageal and esophagogastric junction cancer, glioblastoma, pancreatic cancer, bladder cancer, hepatobiliary cancer, uterine cancer, and vulvar cancer.27-41 To review the NCCN Guidelines recommendations regarding all of these tumor types, please refer to Table 7-1.

CONFIDENTIAL PAGE 26 US-FDX-2000073 Last update: January 22, 2021

Table 2-1 summarizes a selection of applicable guidelines supporting molecular testing for select tumor types for which molecular testing is recommended which have a companion diagnostic indication for testing with FoundationOne CDx.27,30,31,34-36,38,39 Also, please refer to Table 3-1 for FoundationOne CDx companion diagnostic indications, where the test is intended to be a companion diagnostic to identify patients who may benefit from treatment with certain targeted therapies, in accordance with the approved therapeutic product labeling. For additional information concerning the molecular testing recommendations made by NCCN, please refer to Table 7-1.

Table 2-1. Molecular Testing Recommendations in Select Tumor Types per NCCN Guidelines Tumor type NCCN recommendationsa NSCLC To minimize tissue use and potential wastage, NCCN recommends that broad molecular NCCN Guidelines profiling be done in eligible patients with metastatic NSCLC as part of biomarker testing for NSCLC using a validated test(s) that assesses a minimum of the following potential genetic V.2.202135 variants: EGFR mutations (Category 1), ALK fusions (Category 1), BRAF mutations, METex14 skipping mutations, RET rearrangements, NTRK1/2/3, and ROS1 fusions. Both FDA and laboratory-developed test platforms are available that address the need to evaluate these and other analytes.b Broad molecular profiling is also recommended to identify rare driver mutations in eligible patients with metastatic NSCLC for which effective therapy may be available, such as high level MET amplification and ERBB2 mutations. If there is insufficient tissue to allow testing for all of EGFR, ALK, ROS1, BRAF, MET, NTRK1/2/3, and RET, repeat biopsy and/or plasma testing should be done. The use of plasma cell-free/circulating tumor DNA (plasma testing) can be considered in specific clinical circumstances, most notably: if a patient is medically unfit for invasive tissue sampling; or, if following pathologic confirmation of a NSCLC diagnosis, there is insufficient material for molecular analysis. It is recommended that when feasible, molecular testing be performed via a broad, panel-based approach, most typically performed by NGS. For patients who, in broad panel testing don’t have identifiable driver oncogenes (especially in never smokers), consider RNA-based NGS if not already performed, to maximize detection of fusion events. Breast cancerc HER2 testing at diagnosis and of a metastatic site at progression. NCCN Guidelines Assess for BRCA1/2 germline mutations in all patients with recurrent or metastatic for Breast Cancer breast cancer to identify candidates for PARP inhibitor therapy. 27 V.6.2020 For stage IV or recurrent TNBC, assess PD-L1 biomarker status on tumor-infiltrating immune cells to identify candidates for atezolizumab plus albumin-bound paclitaxel. For stage IV or recurrent breast cancer, assess for PIK3CA mutation with tumor or liquid biopsy if hormone receptor-positive/HER2-negative and if considering therapy with alpelisib. PIK3CA mutation testing can be done on tumor tissue or ctDNA in peripheral blood (liquid biopsy). If liquid biopsy is negative, tumor tissue testing is recommended. For stage IV or recurrent breast cancer, larotrectinib and entrectinib are FDA-approved therapies useful in certain circumstances for NTRK gene fusion positive patients. For stage IV or recurrent breast cancer, pembrolizumab is useful in certain circumstances for MSI-H/dMMR tumors. CRC All patients with metastatic CRC should have tumor tissue genotyped for RAS (KRAS and NRAS) and BRAF mutations, and HER2 amplifications individually or as part of an NGS panel. If known RAS/RAF mutation, HER2 testing is not indicated. NGS panels

CONFIDENTIAL PAGE 27 US-FDX-2000073 Last update: January 22, 2021

NCCN Guidelines have the ability to pick up rare and actionable mutations and fusions. All patients with for Colon Cancer CRC should be tested for MMR or MSI at diagnosis. MSI may be accomplished with a V.1.202130 validated NGS panel, especially in patients with metastatic disease who require NCCN Guidelines genotyping of RAS and BRAF. for Rectal Cancer NTRK fusions are extremely rare in CRC and data support limiting the subpopulation of V.1.202139 CRC that should be tested for NTRK fusions to those with wild-type KRAS, NRAS, BRAF, and arguably to those that are dMMR/MSI-H. If testing is done, selection of the appropriate assay for NTRK fusion detection depends on tumor type and genes involved, as well as consideration of other factors such as available material, accessibility of various clinical assays, and whether comprehensive genomic testing is needed concurrently. Ovarian cancerc Both somatic and germline BRCA1/BRCA2 testing is recommended at diagnosis for NCCN Guidelines patients with pathologically confirmed epithelial ovarian cancer/fallopian tube for Ovarian Cancer cancer/primary peritoneal cancer. V.1.202036 Tumor molecular testing is recommended prior to initiation of therapy for persistent/recurrent disease. Validated molecular testing should be performed in a CLIA- approved facility using the most recent available tumor tissue. Testing recommended to include at least: BRCA1/2 (via NGS) and MSI or DNA mismatch repair if not previously done. Evaluation of homologous recombination deficiency can be considered. Additional somatic tumor testing can be considered at the physician’s discretion to identify genetic alterations for which FDA-approved tumor-specific or tumor-agnostic targeted therapy options exist. Prostate cancer Recommend tumor testing for HRR gene mutations (HRRm) in all men diagnosed with metastatic prostate cancer, such as BRCA1, BRCA2, ATM, PALB2, FANCA, RAD51D, NCCN Guidelines for Prostate Cancer CHEK2, CDK12, and consider tumor testing for HRRm in men diagnosed with regional V.3.202038 prostate cancer. Tumor testing for MSI or dMMR is recommended for all men with metastatic CRPC and can be considered for men with castrate-naïve metastatic or regional prostate cancer. Cutaneous Stage III: melanoma BRAF mutation testing is recommended for patients with stage III for whom future NCCN Guidelines BRAF-directed therapy may be an option. for Melanoma: Stage IV or clinical recurrence: Cutaneous Obtain tissue to ascertain alterations in BRAF and in the appropriate clinical setting, KIT, V.1.202131 from either biopsy of the metastasis (preferred) or archival material if the patient is being considered for targeted therapy. Consider broader genomic profiling (eg, larger NGS panels, BRAF non-V600 mutations) if the test results might guide future treatment decisions or eligibility for participation in a clinical trial. If BRAF single-gene testing was the initial test performed and is negative, clinicians should strongly consider larger NGS panels to identify other potential genetic targets (eg, KIT, BRAF non-V600). Hepatobiliary Unresectable or metastatic gallbladder cancer: cancer For unresectable or metastatic gallbladder cancer, dMMR/MSI testing is recommended. NCCN Guidelines For patients with dMMR/MSI-H tumors or a family history suggestive of BRCA1/2 for Hepatobiliary mutations, consider germline testing. Cancer V.5.202034 Consider additional molecular testing in patients with unresectable or metastatic biliary tract cancers gallbladder cancer; this testing may include NTRK gene fusion testing. Unresectable or metastatic intra- or extra-hepatic cholangiocarcinoma: For unresectable or metastatic intrahepatic cholangiocarcinoma or extrahepatic cholangiocarcinoma, dMMR/MSI testing is recommended. For patients with

CONFIDENTIAL PAGE 28 US-FDX-2000073 Last update: January 22, 2021

dMMR/MSI-H tumors or a family history suggestive of BRCA1/2 mutations, consider germline testing. Consider additional molecular testing in patients with unresectable or metastatic intrahepatic cholangiocarcinoma or extrahepatic cholangiocarcinoma; this testing may include NTRK gene fusion testing. Testing for FGFR2 fusions or rearrangements and IDH1 mutations to determine targeted treatment for patients with progression on a preferred first-line regimen. a All NCCN recommendations are category 2A unless otherwise indicated. Additionally, NCCN states that the best management of any patient with cancer is in a clinical trial. b The NCCN Guidelines for NSCLC provide recommendations for individual biomarkers that should be tested and recommend testing techniques but do not endorse any specific commercially available biomarker assays. c Data are specific to women. CLIA, Clinical Laboratory Improvement Amendments; CNS, central nervous system; CPS, combined positive score; CRC, colorectal cancer; CRPC, castration-resistant prostate cancer; dMMR, deficient mismatch repair; DNA, deoxyribonucleic acid; ER, estrogen receptor; FDA, Food and Drug Administration; FFPE, formalin-fixed, paraffin-embedded; HBOC, hereditary breast and ovarian cancer; HRD, homologous recombination deficiency; IHC, immunohistochemistry; LOH, loss of heterozygosity; MSI, microsatellite instability; MMR, mismatch repair; NCCN, National Comprehensive Cancer Network; NGS, next-generation sequencing; NSCLC, non-small cell lung cancer; NTRK, neurotrophic receptor tyrosine kinase; PFS, progression-free survival; tBRCA, tumor BRCA; TMB, tumor mutational burden; TNBC, triple-negative breast cancer.

Molecularly Matched Therapies Improve Clinical Outcomes Biomarker-based therapy, or targeted therapy, involves treatment that is targeted specifically to the genomic profile of a patient’s tumor that may affect certain signaling pathways known to underpin disease progression.131 Immunotherapy (with immune checkpoint inhibitors such as pembrolizumab and nivolumab) refers to the use of monoclonal antibodies that block key molecules in immune checkpoint pathways, thereby eliciting an immune response that targets and destroys cancer cells.132 Each therapy type has demonstrated an established clinical benefit in advanced cancers across multiple tumor types.

Targeted Therapy and Immunotherapy As shown in Table 2-2, numerous pan-tumor cohort studies and meta-analyses have demonstrated improvements in outcomes for genomically matched therapy compared with unmatched therapy.10,11,118 12,48,119 Immunotherapies have also demonstrated significant improvements in outcomes such as overall response rate (ORR), PFS, and OS in patients with advanced cancer. Although not all trials were biomarker driven, treatment with checkpoint inhibitors has led to similar quality of life improvements as observed for targeted therapies compared with chemotherapy or investigator’s choice of therapy in phase 3 trials in multiple tumor types.133 134,135,136 Further, patients having high TMB or MSI-H/dMMR have been shown to have improved outcomes with immunotherapy compared to those without these biomarkers (Table 2-3).52

Table 2-2. Improved Clinical Outcomes With Genomically Matched Targeted Therapy, Pan-Tumor Studies Author/year Study design Clinical impact

CONFIDENTIAL PAGE 29 US-FDX-2000073 Last update: January 22, 2021

Populations compared Outcome P-value measurea Massard Single-center, single-arm, open- Intrapatient comparison PFS2/PFS1 ratio P=NA 201710 label MOSCATO 01 trial; 199 using PFS2/PFS1b >1.33: patients with advanced hard-to- 33% of patients treat solid cancers received therapy matched to a genomic alteration Haslem The Intermountain Healthcare Patients with genomic PFS (average): P=0.002 201711 study characterized patients alteration who received 22.9 weeks with advanced solid cancers targeted therapy who received genomic testing (n=36) followed by targeted therapy Historical control group PFS (average): who received 12.0 weeks chemotherapy or BSC (n=36) Schwaederle A meta-analysis of 346 phase 1 Genomically matched ORR: 30.6% P<0.001 201612 studies of patients with biomarker-based targeted refractory hematologic and therapy solid tumors (N=13,203) (n=58 trials) Non-matched therapy ORR: 4.9% (n=293 trials) Genomically matched PFS: 5.7 months P=0.049 biomarker-based targeted therapy (n=45 trials) Non-matched therapy PFS: 2.95 months (n=45 trials) Tsimberidou A prospective study of patients Matched therapy ORR: 16.4% P<0.0001 118 2019 with refractory cancers (n=711) (N=3,743) who were referred to Unmatched therapy phase 1 trials ORR: 5.4% (n=596) Matched therapy PFS: 4.0 months P<0.0001 (n=711) Unmatched therapy PFS: 2.8 months (n=596) Matched therapy OS: 9.3 months P<0.0001 (n=711) Unmatched therapy OS: 7.3 months (n=596) Jardim A meta-analysis of 112 Genomically matched ORR: 48% P<0.001 2015119 registrational trials (57 targeted therapy randomized [32% personalized] (n=44 trials) and 55 nonrandomized trials [47% personalized]; n=38,104 Unmatched therapy ORR: 23% (n=67 trials)

CONFIDENTIAL PAGE 30 US-FDX-2000073 Last update: January 22, 2021

patients) leading to FDA Genomically matched PFS: 8.3 months P=0.002 approval of 58 cancer therapies targeted therapy in patients with hematologic (n=28 trials) and solid tumor types Unmatched therapy PFS: 5.5 months (n=62 trials) Genomically matched OS: 19.3 moths P=0.04 targeted therapy (n=11 trials) Unmatched therapy OS: 13.5 months (n=49 trials) Schwaederle A meta-analysis of 570 phase 2 Matched therapy ORR: 31% P<0.001 201548 single-agent studies (N=32,149) Unmatched therapy ORR: 10.5% Matched therapy PFS: 5.9 months P<0.001 Unmatched therapy PFS: 2.7 months Matched therapy OS: 13.7 months P<0.001 Unmatched therapy OS: 8.9 months a Unless otherwise indicated, the outcome measures in the table are medians for PFS and OS; for ORR, only CRs and PRs were considered (Tsimberdou 2019 used RECIST criteria; Jardim 2015 used RECIST or WHO criteria; Schwaederle 2016 and Schwaederle 2015 did not specify the method of assessment). b The primary objective was to evaluate clinical benefit as measured by the percentage of patients presenting PFS on matched therapy (PFS2) 1.3-fold longer than the PFS on prior therapy (PFS1). CR, complete response; FDA, Food and Drug Administration; NA, not applicable; ORR, overall response rate; OS, overall survival; PFS, progression-free survival; PR, partial response; RECIST, Response Evaluation Criteria in Solid Tumors; WHO, World Health Organization.

Table 2-3. Improved Clinical Outcomes With Immunotherapy Matched to TMB or MSI-H/dMMR Author/year Study design Clinical impact Populations compared Outcome P-value measurea Gandara A retrospective analysis of 2 bTMB 16 treated with OS: 13.5 P=0.025 201850 large randomized trials of atezolizumab months atezolizumab in NSCLC as test (n=77) and validation studies for bTMB 16 treated with bTMB measurement as OS: 6.8 months docetaxel assessed by FoundationACT (n=81) Hellman A subgroup of high TMB Patients with high TMB 1-year PFS rate: P<0.001 201852 patients (N=299) from the treated with nivolumab + 42.6% phase 3, randomized, placebo- ipilimumab controlled CheckMate-227 trial (n=139) of patients with advanced Patients with high TMB 1-year PFS rate: NSCLC treated with chemotherapy 13.2% (n=160)

CONFIDENTIAL PAGE 31 US-FDX-2000073 Last update: January 22, 2021

Goodman A retrospective pan-tumor Patients with high TMB ORR: 58% P=0.0001 2017137 study of patients with locally treated with immunotherapy advanced or metastatic cancers (n=38) who were treated with various Patients with low TMB ORR: 20% immunotherapies with TMB treated with immunotherapy assessed by FoundationOne (n=113) (N=151) Patients with high TMB PFS: 12.8 P<0.0001 treated with immunotherapy months (n=38) Patients with low TMB PFS: 3.3 treated with immunotherapy months (n=113) Patients with high TMB OS: Not P=0.0036 treated with immunotherapy reached (n=38) Patients with low TMB OS: 16.3 treated with immunotherapy months (n=113) Goodman 60 patients seen at UCSD had Patients with high TMB and 26.8 months P=0.0173 2019138 tumor types analyzed by MSI-stable treated with FoundationOne for MSI and immunotherapy (n=15) TMB status and response to Patients with low- 4.3 months immunotherapy intermediate TMB and MSI- stable treated with immunotherapy (n=45) Patients with high TMB and NE P=0.0635 MSI-stable treated with immunotherapy (n=15) Patients with low- 16.3 months intermediate TMB and MSI- stable treated with immunotherapy (n=45) Le 2015139 Phase 2 study of the clinical Patients with dMMR CRC ORR: 40% NR activity of pembrolizumab in treated with pembrolizumab patients with metastatic cancer (n=10) with or without dMMR (n=41) Patients with dMMR non- ORR: 71% CRC treated with pembrolizumab (n=7) Patients pMMR CRC treated ORR: 0% with pembrolizumab (n=18) Patients with dMMR CRC PFS: NE P<0.001 treated with pembrolizumab (n=10) Patients pMMR CRC treated PFS: 2.2 with pembrolizumab (n=18) months

CONFIDENTIAL PAGE 32 US-FDX-2000073 Last update: January 22, 2021

Patients with dMMR CRC OS: NE P=0.03 treated with pembrolizumab (n=10) Patients pMMR CRC treated OS: 5.0 months with pembrolizumab (n=18) a Unless otherwise indicated, the outcome measures in the table are medians for PFS and OS; for ORR, only complete responses and partial responses were considered and this was as assessed by RECIST. bTMB, blood tumor mutational burden; CRC, colorectal cancer; dMMR, DNA mismatch repair; MSI-H, microsatellite instability–high; NE, not evaluable; NR, not reported; NSCLC, non-small cell lung cancer; ORR, overall response rate; OS, overall survival; PFS, progression-free survival; pMMR, proficient DNA mismatch repair; RECIST, Response Evaluation Criteria in Solid Tumors; TMB, tumor mutational burden; USCD, University of California San Diego.

Real-World Molecular Testing Patterns Although both randomized controlled trials and real-world evidence have shown that targeted therapies improve outcomes for patients as compared with standard of care therapy, current molecular testing rates still fall short of guideline recommendations. Recent evidence suggests that many patients with advanced cancer are not undergoing molecular tumor testing.21 . A real-world assessment from US labs and claims databases conducted in April 2017 suggest that approximately 6,514 patients monthly and 78,168 annually could be missed from targeted therapy due to suboptimal testing.25 These were patients whose tests may have led to targeted treatment, but likely did not because the test results were incorrect, too late, or inconclusive due to sample management issues.25 Importantly, these data are conservative in that they only include known testing issues with molecular testing and do not include patients not tested due to lags in test adoption, which would increase these numbers considerably.25 . A recent retrospective claims analysis in the US included a total of 8,193 adults with select metastatic cancers with a diagnosis between January 2010 and March 2015.21 The observed frequencies of molecular diagnostic tests among all patients were 52% for breast, 42% for NSCLC, 37% for CRC, 34% for head and neck, 41% for ovarian, and 42% for uterine cancer.21 In advanced NSCLC, a population representing the vast majority of testing data due to the availability for more than a decade of targeted therapies which have become standard of care, the testing rates remain suboptimal. . A survey conducted in 2015 of 157 practitioners (n=148 [94%], medical oncologists) responsible for treating NSCLC reported that in newly diagnosed, metastatic patients with NSCLC, only 72%, 69%, 38%, and 18% tested for EGFR, ALK, ROS1, and BRAF gene alterations, respectively.26 . A retrospective analysis within a large electronic health record database of patients with advanced NSCLC (n=1,203) receiving treatment within community practices (encompassing 289 oncologists) in 2017 and 2018 found that the testing rate for all biomarkers with an FDA- approved on-label drug (EGFR, ALK, ROS1, and BRAF) was 22%.140 Further, testing for all 7 guideline-recommended genes, excluding PD-L1, for associated therapies was 7%.140 . Another retrospective analysis of patients with advanced NSCLC reported that, as of mid-2017, 38% had no record of any biomarker testing.141 . A retrospective data analysis comparing biomarker testing rates at academic and community cancer programs and utilizing data input by providers into the Via Oncology clinical pathways software program (Via Portal) was conducted between January 1, 2017 and March 31, 2017.99 This analysis revealed that testing rates for ALK, EGFR, and ROS1 were 94% (n = 285), 95%

CONFIDENTIAL PAGE 33 US-FDX-2000073 Last update: January 22, 2021

(n = 288), and 88% (n = 267), respectively, in the overall nonsquamous NSCLC population (N = 304).99 The testing rate for all 3 biomarkers was 100% in the academic setting; however, in the community setting, the testing rates were lower (EGFR: 94%; ALK: 92%; ROS1: 85%). This study utilized clinical pathways software support that prompted clinicians for biomarker testing results, likely leading to increased compliance with molecular testing in this population.99 Within other solid tumor types, there is a paucity of data regarding real-world molecular testing rates. This gap in data may be due to the fact that the therapeutic options for some of targetable genomic alterations in these tumor types are relatively new comparatively to those in NSCLC. However, there is one analysis available for metastatic colon cancer patients that shows testing likely remains suboptimal in other tumor types as well.23 . A retrospective review of the COTA Real-World Data database was performed for 1,497 patients with metastatic colon cancer diagnosed between January 2013 and December 2017 and treated at 23 practiced in the US.23 Overall guideline-aligned biomarker testing was only completed in 40% of patients in this study; guideline-aligned biomarker testing rates for rat sarcoma (RAS), BRAF, and MSI/mismatch repair deficiency over this study period were 41%, 43%, and 51%, respectively.23 As studies show that a significant proportion of patients, up to 80%, do not receive guideline-based molecular testing, there is a missed opportunity to ensure patients are receiving optimal treatment.21- 23,25,26,140 Table 2-4 outlines this opportunity in terms of patient numbers.

Table 2-4. Biomarkers and Real-World Testing Patterns in Select Advanced Cancers Incidencea Prevalence of Real-world testing Estimate of patients biomarkers rates missed NSCLC 124,725 30%b 33.2%c 25,000 Prostate 43,157 15%–30%d 41%e 3,800–7,600 Breast 43,120 65%f 50%g 14,000 Ovarian 16,916 50%h 41%i 5,000 Colorectal 80,697 51%j 27-47.5%k 21,600−30,000 Total estimated number of unidentified patients potentially eligible for targeted 69,400−81,600 treatments a Incidence from Kantar Health CancerMPact with estimations for 2020 based on patients with stage III or IV disease at diagnosis. b In eligible patients with metastatic NSCLC, testing is recommended in the NCCN Guidelines for the following biomarkers: EGFR, ALK, ROS1, BRAF, MET exon 14 skipping, RET, NTRK1/2/3, and PD-L1 (NCCN Guidelines for NSCLC V.2.202135). Prevalence from Shepherd et al 2019. c Estimation determined by averaging the studies presented above, which present data on detection of >1 genomic alteration (Mason 2016 and Chawla 2018). d In certain patients with prostate cancer, testing is recommended in the NCCN Guidelines for the following biomarkers: HRR gene mutations, including BRCA1, BRCA2, ATM, PALB2, FANCA, RAD51D, CHEK2, CDK12, and MSI or dMMR (NCCN Guidelines for Prostate Cancer V.3.202038). Prevalence from Warner et al 2019 and Athie et al 2019. e No specific estimation available for prostate cancer. Estimation determined from Chawla et al 2018 by averaging the molecular testing rate across tumor types.

CONFIDENTIAL PAGE 34 US-FDX-2000073 Last update: January 22, 2021

f In patients with recurrent or metastatic breast cancer, testing is recommended in the NCCN Guidelines for the following biomarkers: HER2, germline BRCA1/2, PIK3CA, PD-L1, NTRK, and MSI-H or dMMR (NCCN Guidelines for Breast Cancer V.6.202027). Prevalence from Pauletti et al 2000, Yaziji et al 2004, and Kratz et al 2018. g Estimation of real-world testing rates taken from Chawla et al 2018 and Lux et al 2020. h In patients with recurrent ovarian cancer, testing is recommended in the NCCN Guidelines for the following biomarkers:, BRCA1/2, and MSI-H/dMMR; consider HRD (NCCN Guidelines for Ovarian Cancer V.1.202036). Prevalence from Konstantinopoulos et al 2015, Bonadio et al 2018, and Gee et al 2018. i Estimation of real-world testing rates taken from Chawla et al 2018. j Refers to prevalence of extended RAS (KRAS/NRAS) as reported by Cohen et al 2020. k Estimation of real-world testing rates (Rico 2016 and Landsman-Blumberg 2014). NCCN, National Comprehensive Cancer Network; NSCLC, non-small cell lung cancer. Source: Kantar Health3; Jordan 2017142; Shepherd 2019143; Athie 2019144; Warner 2019145; Mason 201622; Chawla 201821; Giermann 2019146; Kratz 2018147; Pauletti 2000148; Yaziji 2004149; Lux 20205; Konstantinopoulos 2015150; Bonadio 2018151; Gee 2018152; Cohen 2020153; Rico 2016154; Landsman-Blumberg 2014155;

Barriers to Guideline Adherence for Molecular Testing Molecular profiling should be considered standard practice for most patients with advanced cancer.46 However, patients may not undergo molecular testing for a number of reasons, including: . lack of physician knowledge about the benefits of results to help inform treatment decision making; . lack of access to testing; or . factors specific to the clinical scenario for a given patient, which make molecular testing utilizing a tissue-based test not feasible. In patients who do undergo non-CGP molecular testing methods, such as single-gene tests, hotspot panels, or cancer-specific focused panels (which typically rely on PCR or FISH methodology), there are considerable limitations including: . incomplete information as not all relevant genes and/or types of alterations are assessed71,156-161; . inefficiency as these methods may require sequential testing in certain cancer types156,157,162,182,187; and . risk of re-biopsy as multiple tests exhaust precious tissue.156,157,162 As tumor molecular profiling is essential to optimizing treatment in clinical practice, options that allow for more complete molecular testing will enable a more informed treatment plan.46,115

Improved Detection of Genomic Alterations with Comprehensive Genomic Profiling CGP utilizes NGS technology to examine entire regions of cancer-relevant genes (in contrast to limited hotspot tests) and genes in established cancer pathways for all tumor types, identifying the 4 main classes of genomic alterations (base substitutions, insertions or deletions, CNAs, gene rearrangements) and reporting complex biomarkers such as TMB and MSI, to inform cancer treatment decisions via a single assay.46,49-52 Evidence from CGP testing has demonstrated the additional value of using a CGP-based approach to match patients to therapy compared with standard genomic tests such as FISH and PCR.

CONFIDENTIAL PAGE 35 US-FDX-2000073 Last update: January 22, 2021

. By increasing the number of targetable genomic alterations identified as compared with single gene or hotspot testing, a tissue-based CGP approach has resulted in improved patient outcomes compared with standard of care (unmatched) therapy through the matching of genomic alterations to effective therapeutic options.11,63,67,74,163 . CGP not only allows for identification of genomic alterations but also accurate measurement of complex biomarkers, such as MSI and TMB.49 . CGP testing has been associated with a 10% to 20% enrollment rate in clinical trials to date compared with a historical enrollment rate of ≤8%.53,59-61,75,76 CGP streamlines testing for molecular biomarkers, provides information on genomic signatures that cannot be captured by single-gene tests or smaller panels, and provides context-based test results to allow for evidence-based clinical decision making. With the utilization of broad molecular profiling, such as CGP, treatment options will improve for an increasing number of patients while eventually emerging as a more cost-effective, generally beneficial option compared with the currently accepted trial-and-error treatment model.46

Evidence of Improved Detection of Genomic Alterations With CGP Identifying targetable biomarkers may inform therapeutic interventions that are likely to have a greater clinical benefit via a precision medicine approach as opposed to a “one-size-fits-all” approach.127 Up to 95% of patients with advanced cancer who undergo CGP have an actionable alteration that can be matched either to a targeted therapy or to a genomically matched clinical trial.10,13,53,55-62 Further, evidence from tissue-based CGP testing has demonstrated the additional value of using a CGP-based approach to match patients to therapy compared with standard genomic tests such as FISH, PCR, single-gene tests, and hotspot testing, as CGP identifies missed genomic alterations from other testing methods in 41% to 84% of previously tested patients (Table 2-5).13,61

Table 2-5. Improved Detection of Genomic Alterations With CGP Testing Percent of patients with ≥1 Percent of patients Author/year Study description missed genomic alteration who received identified with CGP targeted therapy Kopetz 201913 Prospective study of 521 patients 41% 19% with refractory cancers comparing a 46- or 50-gene NGS assay with a 409-gene whole exome assay Reitsma 201961 Retrospective analysis of medical 84% 19% records including 96 patients in community oncology practice who received CGP testing Subset of 32 patients who previously received conventional testing CGP, comprehensive genomic profiling; NGS, next-generation sequencing.

CGP has also been shown to improve detection of actionable genomic alterations within specific tumor types compared with traditional testing methods.68-71

CONFIDENTIAL PAGE 36 US-FDX-2000073 Last update: January 22, 2021

. CGP can identify up to 37% more melanoma patients with BRAF alterations compared with traditional PCR-based methods.68 . Of the 6.4% of CRC patients who harbor potentially resistant KRAS mutations outside of codons 12 and 13, CGP may be able to identify 88% of those resistance alterations not assessed by focused PCR-based testing.69 . CGP can identify breast cancer patients who harbor multiple PIK3CA mutations that are traditionally missed by hotspot testing.70 . CGP has been shown to identify up to 35% more patients with ALK fusions and 21% more patients with EGFR alterations (41% of these EGFR mutations are common alterations targetable by an FDA-approved therapy in the patient’s tumor type) compared with traditional methods in NSCLC.71 With hundreds of investigational agents currently in clinical trials and many existing therapies being tested for new indications, this list will continue to grow. Through August 2019, of the 30 new approval notices from the FDA for hematology/oncology agents, 21 involved targeted agents.164 Given that there are several other such therapies in development with anticipated approvals in upcoming years, a broad companion diagnostic, such as FoundationOne CDx, allows these updates to be incorporated into a single assay, potentially providing physicians with the opportunity to receive more information at the same time to better inform the treatment of their patients.

CONFIDENTIAL PAGE 37 US-FDX-2000073 Last update: January 22, 2021

3 PRODUCT DESCRIPTION

. FoundationOne CDx is an FDA-approved diagnostic test that targets 324 genes as well as the genomic signatures MSI, TMB, and loss of heterozygosity (LOH) (in ovarian cancer).1 . FoundationOne CDx is currently approved by the FDA as a broad companion diagnostic test for 23 drug therapies in 7 tumor types; further, FoundationOne CDx is the companion diagnostic TMB high across all solid tumors.1 . FoundationOne CDx is highly concordant with FoundationOne and externally validated NGS and non-NGS companion diagnostic assays.1

Foundation Medicine Portfolio of CGP Tests The FoundationOne portfolio facilitates a precision medicine approach for a broad spectrum of patients across both solid tumor and hematologic malignancies. . FoundationOne CDx is the first FDA-approved broad companion diagnostic that is clinically and analytically validated for solid tumors. It is approved for use as a companion diagnostic to identify patients who may benefit from treatment with the targeted therapies in accordance with the approved therapeutic product labeling. Additionally, FoundationOne CDx is intended to provide tumor mutation profiling to be used by qualified healthcare professionals in accordance with professional guidelines in oncology for patients with solid malignant neoplasms. FoundationOne CDx is currently an FDA-approved companion diagnostic for 23 drug therapies in 7 cancer types.1 For more information concerning FoundationOne CDx, please refer to the FoundationOne CDx label. . In August 2020, FoundationOne Liquid CDx became the first FDA-approved circulating tumor DNA (ctDNA)-based CGP assay (liquid biopsy). It is intended to be used as a companion diagnostic to identify patients who may benefit from treatment with targeted therapies in accordance with the approved therapeutic product labeling. FoundationOne Liquid CDx is intended to provide tumor mutation profiling to be used by qualified healthcare professionals in accordance with professional guidelines in oncology for patients with malignant neoplasms. FoundationOne Liquid CDx is currently an FDA-approved companion diagnostic for 7 drug therapies in 4 cancer types.165 For more information concerning FoundationOne Liquid CDx, please refer to the FoundationOne Liquid CDx label. . Together, these tests, with FoundationOne Heme and available programmed death ligand-1 (PD- L1) testing, provide an appropriate genomic testing option to inform precision medicine for any advanced cancer patient.

Foundation Medicine Decision Support Foundation Medicine’s services go “beyond the test” by providing a clear, in-depth report that supports clinical decision making as well as decision support services and technology solutions to help streamline patient care. Please refer to FoundationOne Portfolio Description of Decision Support Services for more specific information about the FoundationOne Portfolio and the Decision Support Services offered by Foundation Medicine.

CONFIDENTIAL PAGE 38 US-FDX-2000073 Last update: January 22, 2021

FoundationOne CDx Product Description Foundation Medicine designed and developed FoundationOne CDx based on the previous version of the assay, FoundationOne laboratory-developed test (LDT). While the FoundationOne LDT assay was not FDA-cleared- or -approved, it was the original clinically available Foundation Medicine CGP assay used to detect the presence of genomic alterations in tissue specimens.1 FoundationOne® CDx is a qualitative NGS based in vitro diagnostic test that uses targeted high throughput hybridization-based capture technology for detection of substitutions, insertion and deletion alterations (indels), and CNAs in 324 genes and select gene rearrangements, as well as genomic signatures including MSI and TMB using DNA isolated from formalin-fixed, paraffin-embedded (FFPE) tumor tissue specimens. The test is intended as a companion diagnostic to identify patients who may benefit from treatment with the targeted therapies listed in Table 3-1 in accordance with the approved therapeutic product labeling. Additionally, FoundationOne CDx is intended to provide tumor mutation profiling to be used by qualified health care professionals in accordance with professional guidelines in oncology for patients with solid malignant neoplasms. Genomic findings other than those listed in Table 3-1 are not prescriptive or conclusive for labeled use of any specific therapeutic product.1 Given the complexity and rapid growth of the precision oncology space, please refer to the FDA’s list of cleared or approved companion diagnostic devices for the most recent list of companion diagnostic indications.

Table 3-1. FoundationOne CDx Companion Diagnostic Indicationsa Indication Biomarker Therapies Pan-tumor TMB-H (≥ 10 mut/Mb) Pembrolizumab NTRK1/2/3 gene fusions Larotrectinib NSCLC EGFR exon 19 deletions and EGFR exon Afatinib, gefitinib, osimertinib, or erlotinib 21 L858R alterations EGFR exon 20 T790M alterations Osimertinib ALK rearrangements Alectinib, crizotinib, ceritinib BRAF V600E Dabrafenib in combination with trametinib MET SNVs and indels that lead to MET Capmatinib exon 14 skipping Breast cancer ERBB2 (HER2) amplification Trastuzumab, ado-trastuzumab-emtansine, or pertuzumab PIK3CA C420R, E542K, E545A, E545D, Alpelisib [1635G>T only, E545G, E545K, Q546E, Q546R, H1047L, and H1047Y alterations Colorectal cancer KRAS wild-type (absence of mutations in Cetuximab codons 12 and 13) KRAS wild-type (absence of mutations in Panitumumab exons 2, 3, and 4) and NRAS wild type (absence of mutations in exons 2, 3, and 4) Ovarian cancer BRCA1/2 alterations Olaparib or rucaparib Prostate cancer HRR gene (BRCA1, BRCA2, ATM, Olaparib BARD1, BRIP1, CDK12, CHEK1,

CONFIDENTIAL PAGE 39 US-FDX-2000073 Last update: January 22, 2021

CHEK2, FANCL, PALB2, RAD51B, RAD51C, RAD51D, and RAD54L) alterations Melanoma BRAF V600E Vemurafenib or dabrafenib BRAF V600E and V600K Trametinib or cobimetinib in combination with vemurafenib Cholangiocarcinoma FGFR2 fusions and select rearrangements Pemigatinib a This table does not contain, nor is it intended to represent, NCCN-recommended treatments for oncogenic driver mutations for individual tumor types; instead this represents those oncogenic driver mutations for which FoundationOne CDx has a companion diagnostic indication and the list of targeted treatments FDA-approved for those patients with alterations in these specific genes. ALK, anaplastic lymphoma kinase; BRCA, breast cancer gene; dMMR, mismatch repair deficient; EGFR, epidermal growth factor receptor; FDA, Food and Drug Administration; HER2, human epidermal growth factor receptor 2; HRR, Homologous recombination repair; KRAS, V-Ki-ras2 Kirsten rat sarcoma; Mb, megabase; mut, mutations; NSCLC, non-small cell lung cancer; NTRK, neurotrophic receptor tyrosine kinase; SNV, single nucleotide variant; TMB, tumor mutational burden-high. Source: FoundationOne CDx Label1

FoundationOne CDx is also used for detection of genomic LOH from FFPE ovarian tumor tissue.1 Positive HRD status (F1CDx HRD defined as tBRCA-positive and/or LOH high) in ovarian cancer patients is associated with increased PFS from Rubraca® maintenance therapy in accordance with the Rubraca product label.1 The F1CDx assay is performed at Foundation Medicine, Inc. sites located in Cambridge, MA and Morrisville, NC.1 The complete list of genes and noncoding regions tested by this assay is described in further detail in the Appendix (Table 7-4 and Table 7-5). The FoundationOne CDx platform employs whole-genome shotgun library construction and hybridization-based capture of DNA extracted from FFPE tumor tissue followed by uniform and deep sequencing. Following sequencing, customized software and algorithms are used to determine genomic variants including substitutions, indels, CNAs, select genomic rearrangements, MSI, TMB, and LOH, with LOH noted as specific to ovarian cancer.1 FoundationOne CDx report results are annotated by automated software with companion diagnostic- relevant information and are merged with patient demographic information; this is combined with any additional relevant information provided by Foundation Medicine as a professional service before the report is approved and released by the laboratory director or designee.1 An example test report is shown in the Appendix (FoundationOne CDx Sample Report). The output of the test includes: . A listing of all alterations in tested genes that are known or likely to be cancer-driver alterations and genomic signatures, some of which may also be associated with companion diagnostic information. . When indicated, the FoundationOne CDx report will include FDA-approved therapeutic options that may be considered based on detected alterations and tumor types for which the FoundationOne CDx is approved as a companion diagnostic. . If an identified genomic alteration or genomic signature may be associated with treatment resistance, the FoundationOne CDx report will include a note notifying of potential resistance.

CONFIDENTIAL PAGE 40 US-FDX-2000073 Last update: January 22, 2021

. If no genomic alteration or genomic signatures associated with companion diagnostic-relevant information are identified, the FoundationOne CDx report will note that there are no reportable alterations with companion diagnostic claims. . The professional services section of the FoundationOne CDx report provides a list of potential clinical trials and investigational options to consider for identified genomic alterations or genomic signatures. A rationale, targets, and location of potential clinical trials are described in detail. The report guide, which points out key features of the FoundationOne CDx report is shown in Figure 3-1.

CONFIDENTIAL PAGE 41 US-FDX-2000073 Last update: January 22, 2021

Figure 3-1. FoundationOne CDx Report Guide

CONFIDENTIAL PAGE 42 US-FDX-2000073 Last update: January 22, 2021

Analytic Validity of FoundationOne CDx

Sample validation and concordance The Limit of Detection (LoD) of alterations assessed by FoundationOne CDx was evaluated. The LoD of the 13 companion diagnostic biomarkers are summarized in Table 3-2 and Table 3-3 below. LoD indicates the median variant allele frequency (VAF) at which the test has shown 95% probability of detection (sensitivity). Further, the LoD for representative alterations detected by the FoundationOne CDx platform is summarized in Table 3-4.

CONFIDENTIAL PAGE 43 US-FDX-2000073 Last update: January 22, 2021

Table 3-2. Sample Validation for FoundationOne CDx Companion Diagnostic Claims Alteration LoDa allele fraction (%) (100% LoDb allele fraction (%) hit rate) (probit) EGFR L858R 2.4% <2.4% (all detected) EGFR Exon 19 deletion 5.1% 3.4% EGFR T790M 2.5% 1.8% KRAS G12/G13 2.3% <2.3% (all detected) BRAF V600E/K 2.0% <2.0% (all detected) MET Exon 14 SNVsc N/A <2.9% (all detected) MET Exon 14 insertion and N/A 5.7% deletionc PIK3CA E542K 4.9% Not calculated BRCA1/2d Alteration in non-repetitive or N/A 5.9% homopolymer <4 bp Deletion in 8 bp homopolymer N/A 15.3% HRR gene base substitutions 5.44%–6.33%e Not calculated HRR gene indels 5.22%–12.74% Not calculated a LoD calculations for the CDx variants were based on the hit rate approach, as there were less than three levels with hit rate between 10% and 90% for all CDx variants (not including BRCA1/2 variants). LoD from the hit rate approach is defined as the lowest level with 100% hit rate (worst scenario). b LoD calculations for the CDx variants based on the probit approach with 95% probability of detection. c For each sample, five levels of MAF, with 10 replicates per level, were evaluated for a total of 50 replicates per sample. d See Summary of Safety and Effectiveness Data for P160018. e LoD defined as the lowest level with 95% hit rate or greater. BRAF, v-Raf murine sarcoma viral oncogene homolog B; BRCA, breast cancer gene; EGFR, epidermal growth factor receptor; HRR, Homologous recombination repair; KRAS, V-Ki-ras2 Kirsten rat sarcoma; LoD, limits of detection; MET, mesenchymal- epithelial transition; PIK3CA, phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha. Source: FoundationOne CDx Label1

Table 3-3. Sample Validation for FoundationOne CDx Companion Diagnostic Claims Alteration Tumor purity (%) Tumor purity (%) (95% hit rate)a (probit)b ALK fusion 2.6%c 1.8% ERBB2 amplification 25.3%d 19.7% BRCA2 homozygous deletion (HD) 8.8%e Not calculated LOHf 35% 30% FGFR2 fusions 5.31%g 5.38%

CONFIDENTIAL PAGE 44 US-FDX-2000073 Last update: January 22, 2021

Alteration Tumor purity (%) Tumor purity (%) (95% hit rate)a (probit)b HRR gene rearrangementsh 20.1%g Not calculated HRR gene homozygous deletionsh 23.9%g Not calculated TMB ≥ 10 mut/Mbh 28.16%g Not calculated NTRK1 fusionsi,j 12.1% N/A NTRK2 fusionsi,k 11.5% N/A NTRK3 fusionsi,l 6.1% N/A a Sensitivity calculations for the CDx variants were based on the hit rate approach, as there were less than three levels with hit rate between 10% and 90%. LoD from the hit rate approach is defined as the lowest level with 100% hit rate (worst scenario). b Sensitivity calculations for the CDx variants based on the probit approach with 95% probability of detection. c The number of chimeric reads for the sample evaluated is 16 at the indicated tumor fraction. d The number of copy number amplifications for the sample evaluated is 6 at the indicated tumor fraction. e The LoD calculation for the BRCA2 HD was based on the hit rate approach, as there was a hit at every dilution level tested, making the probit regression not applicable. f Please refer the FoundationOne CDx label for the Summary of Safety and Effectiveness Data for P160018/S001. g Calculated using the 95% hit rate. h For each sample, five levels of tumor purity, with 20 replicates per level except for the highest level at which 14 replicates were tested, were evaluated for a total of 94 replicates per sample. i For each sample, a total of 94 tumor dilution replicates were assessed, including 20 replicates for each level of tumor purity, excluding the highest level, for which only 14 replicates were performed. j The LoD study included 2 samples with CDx NTRK1 fusion positive status: 1 NTRK1-LMNA fusion, and 1 NTRK1-TRP fusion. k The LoD study included 2 samples with CDx NTRK2 fusion positive status: 1 NTRK2-BCR fusion, and 1 NTRK2-GARNL3 fusion. l The LoD study included 3 samples with CDx NTRK3 fusion positive status: 3 NTRK3-ETV6 fusions. ALK, anaplastic lymphoma kinase; BRCA, breast cancer gene; ERBB2, Erb-B2 receptor tyrosine kinase 2; FGFR2, fibroblast growth factor receptor 2; HRR, homologous recombination repair; Mb, megabase; mut, mutations; N/A, not applicable; NTRK, neurotrophic receptor tyrosine kinase; TMB, tumor mutational burden-high. Source: FoundationOne CDx Label1

Table 3-4. Sample Validation for FoundationOne CDx Variant category Range LoDa Subcategory N Allele Fraction (%) Base substitutions Knownb 21c 1.8–7.9c Otherd 166 5.9–11.8 Indels at non-homopolymer context, including Known 3 4.5–6.5

insertions up to 42bp and deletions up to 276bp Other 17 6.0–10.2 5bp repeat 8 10.0–12.2 Indels at homopolymer context 6bp repeat 2 13.6–13.7 7bp repeat 4 16.3–20.4 8bp repeat 3 17.0–20.0

CONFIDENTIAL PAGE 45 US-FDX-2000073 Last update: January 22, 2021

a LoD calculations for the platform variants were based on the hit rate approach for variants with less than three levels with hit rate between 10% and 90% and probit approach for variants with at least three levels with hit rate between 10% and 90%. LoD from the hit rate approach is defined as the lowest level with 100% hit rate (worst scenario). b Data includes an alteration in the TERT promoter, 124C>T (LoD of 7.9%). TERT is the only promoter region interrogated and is highly enriched for repetitive context of poly-Gs, not present in coding regions. c Alterations classified as” known” are defined as those that are listed in COSMIC. d Alterations classified as “other” include truncating events in tumor suppressor genes (splice, frameshift and nonsense) as well as variants that appear in hotspot locations but do not have a specific COSMIC association, or are considered variants of unknown significance (VUS) due to lack of reported evidence and conclusive change in function. bp, base pair; LoD, limits to detection. Source: FoundationOne CDx Label1

The analytic performance of FoundationOne CDx has been validated across a series of concordance studies and a large-scale retrospective analysis of more than 80,000 patient specimens covering 43 tissue types.1 Concordance studies were performed between FoundationOne CDx and FoundationOne, an externally validated NGS assay, and multiple FDA-approved non-NGS companion diagnostic assays.1 The concordance study between FoundationOne CDx and FoundationOne (set as the reference method) was established so that the clinical utility data generated from prior studies that used FoundationOne could also be used to support the clinical utility of FoundationOne CDx. This concordance study evaluated a test set of 165 specimens covering 35 disease ontologies. A total of 2,325 variants, including 2,026 short variants, 266 CNAs, and 33 rearrangements, were included in the study. The PPA and NPA between FoundationOne CDx and FoundationOne were calculated for the genomic alterations, and the results are summarized in Table 3-5. Further, the concordance data associated with the specific FoundationOne CDx companion diagnostic claims at reported in Table 3-6.1 Of note, concordance data is not available for breast cancer (BRCA1/2 mutations) and prostate cancer (HRR gene alterations); this is due to Foundation Medicine being the registrational assay for the therapies used in these indications.1,108,109,111 For the clinical data associated with these indications, please refer to Clinical Utility and Validity of FoundationOne CDx in Breast Cancer and Clinical Utility and Validity of FoundationOne CDx in Prostate Cancer, respectively.

Table 3-5. Concordance Summary for Genomic Alterations Between FoundationOne CDx and FoundationOne Genomic F1CDx+ F1CDx- F1CDx+ F1CDx- PPA, % NPA, % alteration /F1+, n /F1+, n /F1-, n /F1-, n All variants 2,246 33 46 322,890 98.6 99.99 All short variants 1,984 19 23 299,099 99.1 99.99 Substitutions 1,692 10 19 254,854 99.4 99.99 Indels 292 9 4 44,245 97.0 99.99 All CNAs 230 14 22 19,204 94.3 99.9 Amplifications 157 10 12 14,671 94.0 99.9 Losses 73 4 10 4,533 94.8 99.8

CONFIDENTIAL PAGE 46 US-FDX-2000073 Last update: January 22, 2021

Rearrangements 32 0 1 4,587 100 99.98 CNA, copy number alteration; F1CDx, FoundationOne CDx; F1, FoundationOne; indel, insertion and deletion alteration; NPA, negative percent agreement; PPA, positive percent agreement. Source: FoundationOne CDx Label1

Table 3-6. Concordance data for FoundationOne CDx Claims Tumor Type Biomarker PPA, % (n/N) NPA, % (n/N) Comparator assay a Pan-tumor TMB-H (≥10 87.28 (51/59)a 91.56 (138/159) CLIA validated WES mut/Mb) assay NTRK gene 84.1 (37/44) 100 (226/226) Local CTA fusions NSCLC EGFR exon 19 98.1 (106/108) 99.4 (153/154) cobas® EGFR Mutation deletions and exon Test v2 21 L858R EGFR T790M 98.9 (87/88) 86.1 (93/108) cobas EGFR Mutation mutation Test v1 cobas EGFR Mutation Test v2 ALK 92.9 (78/84) 100 (75/75) Ventana ALK (D5F3) rearrangements CDx Assay Vysis ALK Break Apart FISH Probe Kit MET exon 14 98.6 (72/73) 100 (125/125) reverse transcriptase-PCR skipping CTA Breast ERBB2 (HER2) 89.4 (101/113) 98.4 (180/183) Dako HER2 FISH amplifications PharmDx™ Kit PIK3CA 93.8 (106/113) 98.8 (159/161) PCR-based PIK3CA hot- alterations spot CTA 1 91.6 (197/215) 98.8 (162/164) PCR-based PIK3CA hot- spot CTA 2 CRC KRAS WT 100 (173/173) 100 (154/154) therascreen® KRAS RGQ PCR Kit Ovarian BRCA1/2 NR NR NR alterations Prostate HRR gene NR NR NR (BRCA1, BRCA2, ATM, BARD1, BRIP1, CDK12, CHEK1, CHEK2, FANCL, PALB2, RAD51B, RAD51C, RAD51D, and

CONFIDENTIAL PAGE 47 US-FDX-2000073 Last update: January 22, 2021

RAD54L) alterations Melanoma BRAF V600 99.4 (166/167) 89.6 (121/135)b cobas BRAF V600 BRAF V600E 99.3 (149/150) 99.2 (121/122) Mutation Test BRAF V600 96.3 (26/27) 100 (24/24) THxID™ BRAF kit dinucleotide Cholangiocarcinoma FGFR2 fusions 100 (84/84) 100 (97/97) FoundationOne CTA and select rearrangements a The overall PPA and NPA were calculated based on a weighted average of the results (Set A and Set B) in the TMB concordance analysis. b The reported difference in NPA values for BRAF V600 and BRAF V600E are likely attributed to known sensitivity differences in the cobas test, which has lower sensitivity for detection of dinucleotide V600 alterations than for the single nucleotide V600E c.1799T>A alteration, especially for samples in which FoundationOne CDx detected the nucleotides to be of lower than 40% mutational allele frequency, leading to low NPA values. ALK, anaplastic lymphoma kinase; CDx, companion diagnostic; CRC, colorectal cancer; EGFR, epidermal growth factor receptor; FISH, fluorescence in situ hybridization; HER2, human epidermal growth factor receptor 2; KRAS, V-Ki-ras2 Kirsten rat sarcoma; Mb, megabase; mut, mutations; NPA, negative percent agreement; NSCLC, non-small cell lung cancer; PCR, polymerase chain reaction; PPA, positive percent agreement; WT, wild type. Source: FoundationOne CDx Label1

TMB TMB is a measure of the number of somatic mutations per Mb of sequenced DNA. TMB is a genomic signature biomarker that has emerged as a predictor of immune checkpoint inhibitor treatment outcomes, independent of PD-L1 status.51,166 TMB is optimally calculated by whole exome sequencing (WES), but CGP panels provide TMB estimates in a more time- and cost-effective manner.167 However, there is considerable complexity in calculating TMB using a panel and algorithms.167 A study that compared the results of panel TMB testing vs WES showed that there was variability within and between panel TMB values; despite these findings, this study found that panel TMB values were strongly correlated with WES TMB.167 TMB by FoundationOne CDx is determined by counting all synonymous and non-synonymous variants present at 5% allele frequency or greater (after filtering) and the total number is reported as mutations per megabase (mut/Mb) unit. Observed TMB is dependent on characteristics of the specific tumor focus tested for a patient (eg, primary vs metastatic, tumor content) and the testing platform used for the detection; therefore, observed TMB results may vary between different specimens for the same patient and between detection methodologies employed on the same sample. The TMB calculation may differ from TMB calculations used by other assays depending on variables such as the amount of genome interrogated, percentage of tumor, assay LoD, filtering of alterations included in the score, and the read depth and other bioinformatic test specifications. Refer to the SSED for a detailed description of these variables in FMI’s TMB calculation. The clinical validity of TMB defined by this panel has been established for TMB as a qualitative output for a cut-off of 10 mutations per megabase but has not been established for TMB as a quantitative score.1

CONFIDENTIAL PAGE 48 US-FDX-2000073 Last update: January 22, 2021

MSI FoundationOne CDx detects MSI status which may help inform decisions regarding the use of immunotherapy agents. NCCN Guidelines recommend universal MSI or dMMR testing in newly diagnosed patients with CRC; testing for MSI may be accomplished with a validated NGS panel, especially for patients with metastatic disease who require genotyping of RAS and BRAF.30,39 MSI is a genomic signature biomarker that has been associated with improved responses to certain immunotherapies across a range of tumor types. MSI is a condition of genetic hypermutability that arises from defects in the dMMR system, which then generates excessive amounts of short insertion/deletion mutations in the genome.168 The MSI-H/MSS designation by FoundationOne CDx test is based on genome wide analysis of 95 microsatellite loci and not based on the 5 or 7 MSI loci described in current clinical practice guidelines. Refer to the SSED for additional details on methodology. The threshold for MSI-H/MSS was determined by analytical concordance to comparator assays (IHC and PCR) using uterine, cecum and colorectal cancer FFPE tissue. Patients with microsatellite status of “Cannot Be Determined” should be retested with an orthogonal (alternative) method. The clinical validity of the qualitative MSI designation has not been established.1

LOH The LOH score is a profile of the percentage of the tumor genome that is under focal loss of 1 allele.108 Focal LOH events accumulate as genomic “scars” due to incorrect DNA double-strand break repair when the homologous recombination pathway is deficient.169-172 HRD and consequent genomic LOH occur as a result of genetic or epigenetic inactivation of 1 or more of the homologous recombination pathway genes, including BRCA1, BRCA2, RAD51C, ATM, PALB2, and BRIP1.171-174 The genomic LOH score may help identify patients with ovarian cancer who are more likely to benefit from treatment with PARP inhibitors that target the homologous recombination pathway, such as rucaparib. FoundationOne CDx can detect for genomic LOH from FFPE ovarian tumor tissue. Positive HRD status (defined as tBRCA-positive and/or LOH high) in ovarian cancer patients is associated with improved PFS from Rubraca (rucaparib) maintenance therapy in accordance with the Rubraca product label. It should be noted that alterations at allele frequencies below the established limit of detection may not be detected consistently. Detection of LOH has been verified only for ovarian cancer patients and the performance of the LOH classification has not been established for samples below 35% tumor content and with LOH scores near the cut-off of 16. There may be potential interference of ethanol with LOH detection. The interfering effects of xylene, hemoglobin, and triglycerides on the LOH score have not been demonstrated.1

Medical Policy Coverage of FoundationOne CDx CMS outlined coverage criteria for FDA-approved NGS-based in vitro companion diagnostic assays, like FoundationOne CDx, through an NCD.175 FoundationOne CDx is covered under the NCD when the patient has: a. Either recurrent, relapsed, refractory, metastatic, or advanced stages III or IV cancer; and

CONFIDENTIAL PAGE 49 US-FDX-2000073 Last update: January 22, 2021

b. Either not been previously tested using FoundationOne CDx for the same primary diagnosis of cancer or repeat testing using FoundationOne CDx only when a new primary cancer diagnosis is made by the treating physician; and c. Decided to seek further cancer treatment. Commercial coverage of Foundation Medicine testing continues to expand with more than 30 publicly available positive coverage policies, including 4 national payers and TRICARE. Also, the Veterans Affairs National Precision Oncology Program contract provides broad access to Foundation Medicine’s entire portfolio of products.

CONFIDENTIAL PAGE 50 US-FDX-2000073 Last update: January 22, 2021

4 CLINICAL UTILITY OF FOUNDATIONONE CDX

. FoundationOne CDx is an FDA-approved broad companion diagnostic CGP assay approved for use as a companion diagnostic to identify patients who may benefit from treatment with targeted therapies in accordance with the approved therapeutic product labeling.1 . Several pan-tumor and tumor-specific cohort studies have demonstrated substantial improvements in patient outcomes, including RR, PFS, and OS, associated with Foundation Medicine-based CGP testing.7922,74 . This use of CGP in clinical practice allows patients to receive genomically matched therapy with data reporting up to 50% of patients pursue genomically matched therapy, including FDA approved therapies and clinical trial enrollment.1,53,59-62,64,73,74 . Published data from clinical trials have shown that FoundationOne CDx identifies potentially actionable genomic alterations in many patients and can guide treatment decisions in a manner that leads to improved patient outcomes.1

Biomarker-based targeted therapies have led to considerable improvements in clinical outcomes including response rates and survival compared with traditional chemotherapy.10,55-58 CGP testing provides valuable information on the presence of actionable biomarkers, which enables healthcare providers to make evidence-based treatment decisions regarding treatments that result in these improved outcomes for patients with advanced cancer.12,47,48,176 Of patients with advanced cancer who undergo CGP, 51.7% to 99% will have an actionable alteration that can be matched to either a targeted therapy or to a genomically matched clinical trial.10,13,53,55-67 FoundationOne CDx is an FDA-approved broad companion diagnostic CGP assay approved for use as a companion diagnostic to identify patients who may benefit from treatment with targeted therapies in accordance with the approved therapeutic product labeling. The following sections review the clinical utility of CGP (Evidence of Improved Clinical Outcomes with CGP) and FoundationOne CDx (Evidence of Improved Clinical Outcomes With FoundationOne CDx), specifically in pan-tumors, NSCLC, CRC, breast cancer, prostate cancer, ovarian cancer, melanoma, and cholangiocarcinoma.

Evidence of Improved Clinical Outcomes With Treatment Informed by CGP Clinical utility establishes the net clinical benefit to the patient of adding CGP to the current standard of care decision making; in effect, does the intervention (ie, the CGP test) improve patient outcomes?72 As shown in Table 4-1, several pan-tumor and tumor-specific cohort studies have demonstrated substantial improvements in patient outcomes, including RR, PFS, and OS, associated with CGP testing.22,74

CONFIDENTIAL PAGE 51 US-FDX-2000073 Last update: January 22, 2021

Table 4-1. Clinical Utility of CGP Clinical impact Author/year Study design Populations Outcome P-value compared measurea Pan-tumor Kato 201863 Prospective study of the utility of Matched targeted PFS: 19.7 P=0.008 tissue and liquid CGP in patients therapy (n=12) months with rare cancers (n=40) Previous unmatched PFS: 3.5 therapy (n=12) months Schwaederle 201674 Retrospective study of the utility Matched therapy DCRb: 34.5% P≤0.02 of CGP to match patient with (n=87) advanced solid malignancies to a Unmatched therapy DCRb: 16.1% therapy (n=347) (n=93) Matched therapy PFS: 4.0 P=0.039 (n=87) months Unmatched therapy PFS: 3.0 (n=93) months Wheler 201667 Single-arm, nonrandomized study Matched therapy DCR: 19% P=0.61 to prospectively investigate the (n=122) clinical utility of CGP in patients Unmatched therapy DCR: 8% with advanced malignancies (n=66) (N=500) Matched therapy TTF: 2.7 P=0.001 (n=122) months Unmatched therapy TTF: 1.9 (n=66) months Matched therapy OS: 9.3 P=0.087 (n=122) months Unmatched therapy OS: 7.2 (n=66) months Sicklick 201966 Prospective navigation trial at 2 High matching score PFS: 6.5 P=0.046 I-PREDICT centers using tissue-based CGP to (n=28) months match patients to therapies based Low matching score PFS: 3.1 on a matching scorec (n=73) (n=55) months Breast Ganesan 201477 Retrospective study of Matched therapy PFS: 6.4 P=0.001 consecutive patients with (n=90) months advanced or metastatic triple- negative metastatic breast cancer Unmatched therapy PFS: 1.9 (N=106) treated in a phase 1 (n=16) months clinic Pancreatic Pishvaian 201875 Prospective program (Know Your Matched therapy PFS: 4.1 P=0.03 Tumor) using CGP to determine (n=17) months matched therapy in patients with Unmatched therapy PFS: 1.9 pancreatic cancer (n=640) (n=18) months Pishvaian 202078 Prospective program (Know Your Matched therapy OS: 2.58 years P=0.0004 Tumor) using CGP to determine (n=46)

CONFIDENTIAL PAGE 52 US-FDX-2000073 Last update: January 22, 2021

Clinical impact Author/year Study design Populations Outcome P-value compared measurea matched therapy in patients with Unmatched therapy OS: 1.51 years pancreatic cancer (n=1,856) (n=143) NSCLC Singal 201979 A retrospective study to determine Patients with driver OS: 18.6 P<0.001 the clinical utility of a clinico- alteration treated months genomic database (using CGP) in with targeted patients with NSCLC (n=4,064) therapy (n=575) Patients with driver OS: 11.4 alteration not treated months with targeted therapy (n=560) Madison 202099 A retrospective, real-world study Patients with rwPFS: 9.4 P=0.022 to determine clinical outcomes for genomic alteration months NSCLC patients following CGP matched to targeted with liquid biopsy or tissue biopsy therapy (n=287) (n=6,491) Patients with rwPFS: 6.9 genomic alteration months not treated with targeted therapy (n=130) Patients with OS: 26.7 P=0.035 genomic alteration months matched to targeted therapy (n=262) Patients with OS: 17.9 genomic alteration months not treated with targeted therapy (n=130) a PFS, TTF, and OS as presented above are all measured as medians. b DCR is the percentage of patients achieving a complete response, partial response, or stable disease for ≥6 months. c A “matching score” score system was then utilized for each patient. Blinded to patient outcomes, the investigators calculated the total number of molecular alterations matched to the drugs administered and divided that number by the total number of characterized genomic aberrations. CGP, comprehensive genomic profiling; DCR, disease control rate; NSCLC, non-small cell lung cancer; ORR, overall response rate; OS, overall survival; PFS, progression-free survival; rwPFS, real-world progression-free survival; TTF, time to treatment failure.

CONFIDENTIAL PAGE 53 US-FDX-2000073 Last update: January 22, 2021

Decision Impact of CGP in Clinical Practice CGP most commonly guides the use of genomically matched targeted therapies; in this regard, it can provide information about uses of FDA-approved therapies.13,53 As NCCN Guidelines state that the best management of any patient with cancer is in a clinical trial, CGP is also important for determination of eligibility of a substantial proportion of oncology clinical trials (~40%).27,30,31,34-36,38,39 CGP can also provide genomic information that enables physicians to use chemotherapy more effectively, as in the case of homologous recombination deficiency (HRD) ovarian cancer.54 Recent data show that the majority of oncologists are using CGP in their current clinical practice.73 . A recent national survey using data from the National Survey of Precision Medicine in Cancer Treatment reported that 75.6% of oncologists use CGP to guide treatment decisions.73 Of these physicians, CGP was used to guide the use of an FDA-approved therapy (33.5%), determine eligibility for clinical trial enrollment (29.1%), and/or make decisions about off-label use of FDA-approved therapies (17.5%).73 Further, CGP test results informed treatment decisions often and sometimes for 26.8% and 52.4% of respondents, respectfully.73 This use of CGP in clinical practice allows patients to receive genomically matched therapy; data show that up to 50% of patients pursue genomically matched therapy, including FDA approved therapies and clinical trial enrollment.53,59-62,64,73,74 As approximately 40% of oncology clinical trials require a biomarker for eligibility and/or stratification, CGP-based testing has been associated with a clinical trial enrollment rate between 10% and 20%, compared with a historical enrollment rate of ≤8%.53,59-61,75,76 Even in those without or unable to pursue genomically matched options, the personalized treatment plan may confirm chemotherapy as the best option and/or help with discussions about palliative care, thereby avoiding the use of unnecessary therapies. By matching more patients to effective therapeutic options, a CGP approach has resulted in improved patient outcomes, including significantly improved response rates and survival, compared with standard of care (unmatched) therapy.63,66,67,74,75,78,79

Guideline Recommendations for CGP According to NCCN Guidelines, molecular testing is recommended for certain patients with bladder, breast, CRC, cutaneous melanoma, gastric and esophageal cancer, glioblastoma, hepatobiliary cancer, NSCLC, ovarian cancer, pancreatic cancer, prostate cancer, uterine cancer, vulvar cancer; specific recommendations are outlined in Table 7-1.27-41 Additionally, molecular testing is now recommended for certain molecular alterations across solid tumors based on the pan-tumor indication approvals of pembrolizumab, larotrectinib, and entrectinib.44,45,96 . Molecular testing for MSI or dMMR status is now recommended across multiple tumor types after the approval of pembrolizumab, the first tissue/site-agnostic approved drug for the treatment of adult and pediatric patients with unresectable or metastatic MSI-H or dMMR solid tumors that have progressed following prior treatment and who have no satisfactory alternative treatment options (or for patients with MSI-H or dMMR CRC that has progressed following treatment with a fluoropyrimidine, oxaliplatin, and irinotecan).96 . Similarly, with the tissue/site agnostic approvals of the TRK inhibitors (larotrectinib, entrectinib), testing is now recommended across multiple tumor types (Table 7-1).44,45 In addition to the FDA approval of pan-tumor therapies requiring molecular testing across solid tumor types, the NCCN Guidelines recommend broad molecular testing across several solid tumor types; the

CONFIDENTIAL PAGE 54 US-FDX-2000073 Last update: January 22, 2021 recommendations regarding broad molecular testing and/or the use of NGS for select cancer types are briefly outlined in Table 4-2.30,31,35,39

Table 4-2. NCCN Guidelines Recommendations for Broad Molecular Testing/NGS NCCN Guidelines Recommendations for Broad Molecular Testing/NGS NSCLC V.2.202135 The NCCN Guidelines panel strongly advises broader molecular profiling with the goal of identifying rare driver mutations for which effective drugs may already be available, or to appropriately counsel patients regarding the availability of clinical trials. Broad molecular profiling is a key component of the improvement of the care of patients with NSCLC. It is recommended at this time that when feasible, molecular testing be performed via a broad, panel-based approach, most typically performed by NGS. For patients who, in broad panel testing don’t have identifiable driver oncogenes (especially in never smokers), consider RNA-based NGS if not already performed, to maximize detection of fusion events. Colon cancer V.1.202130 All patients with metastatic colorectal cancer should have tumor tissue genotyped Rectal cancer V.1.202139 for RAS (KRAS and NRAS) and BRAF mutations and HER2 amplifications, individually or as part of an NGS panel. If known RAS/RAF mutation, HER2 testing is not indicated. Testing for MSI may be accomplished by PCR or a validated NGS panel, with the NGS panel especially useful for patients with metastatic disease who require genotyping of RAS and BRAF. Cutaneous melanoma Broader genomic profiling can be considered if feasible for patients with stage V.4.202031 III-IV metastatic melanoma or a clinical recurrence, especially if the test results might guide future treatment decisions or eligibility for participation in a clinical trial. MSI, microsatellite instability; NCCN, National Comprehensive Cancer Network; NGS, next generation sequencing; NSCLC, non-small cell lung cancer; PCR, polymerase chain reaction.

CONFIDENTIAL PAGE 55 US-FDX-2000073 Last update: January 22, 2021

Evidence of Improved Clinical Outcomes With FoundationOne CDx Clinical trials of FoundationOne and concordance studies of FoundationOne to FoundationOne CDx have played a pivotal role in establishing the clinical utility of FoundationOne CDx across a broad range of cancer types. In particular, published data from clinical trials have shown that FoundationOne identifies potentially actionable genomic alterations in many patients and can guide treatment decisions in a manner that leads to improved patient outcomes. . Several pan-tumor and tumor-specific cohort studies have demonstrated substantial improvements in patient outcomes, including RR, PFS, and OS, associated with CGP testing (Evidence of Improved Clinical Outcomes With Treatment Informed by CGP).11,63,66,67,74,75,77-79 . Improvement in outcomes associated with treatment informed by Foundation Medicine CGP testing as reported in peer-reviewed published studies mirrors the improvement in outcomes associated with the use of FDA-approved targeted therapies as reported in each of the respective drug’s FDA-labeling.11,63,66,67,74,75,77-95 . Table 7-9 in the Appendix summarizes the available data on frequent genomic alterations and the clinical outcomes identified from major studies of FoundationOne in patients with advanced cancer of mixed tumor types. Findings demonstrated that FoundationOne effectively identified a large number of genomic alterations in each of the study populations, and there were several reports of statistically significant improvements in survival and response outcomes for matched compared with unmatched therapy. Because of high concordance between FoundationOne and FoundationOne CDx, these results are expected to be similar between the 2 assays. o Further details for each study and detailed clinical utility data for several individual tumor types are reported in the Table 7-10.

CONFIDENTIAL PAGE 56 US-FDX-2000073 Last update: January 22, 2021

5 FOUNDATIONONE CDX FOR COMPANION DIAGNOSTIC CLAIMS FoundationOne CDx is an FDA-approved companion diagnostic test to identify patients who may benefit from treatment in accordance with the approved therapeutic product labeling for 21 drug therapies in 7 tumor types, and also for 2 drug therapies across solid tumor types (pan-tumor). FoundationOne CDx has FDA-approved companion diagnostic indications in NSCLC, breast, CRC, ovarian, prostate, melanoma, and cholangiocarcinoma; it also serves as a companion diagnostic across 2 pan-tumor indications (NTRK gene fusions and TMB-H) (Table 3-1). Across these specific tumor types, of the 34 therapies/regimens that require an FDA-approved test, FoundationOne CDx is FDA-approved as a companion diagnostic for 25 (74%).1,100 The sections below highlight the supporting data in patient populations for which FoundationOne CDx has a companion diagnostic claim. For some of the companion diagnostic claims, the FDA approval was based on concordance data with an approved comparator assay. As such, the clinical benefit associated with such claims is described as a part of the prescribing information for the associated targeted therapy.

CONFIDENTIAL PAGE 57 US-FDX-2000073 Last update: January 22, 2021

CGP in Pan-Tumors

. Across solid tumors there are genomic signatures and abnormalities that guide treatment decisions regardless of the primary tumor site: o TMB is an independent biomarker for selecting immunotherapy in advanced cancer patients, and inclusion of TMB in the immunotherapy treatment selection may help optimize the identification of patients eligible for immunotherapy. . In a series of retrospective analyses in a large pan-tumor cohort of patients receiving Foundation Medicine testing during the course of routine clinical care, a consistent association of clinically relevant benefit from immunotherapy was observed with higher TMB.137,138,177 o NTRK gene fusions are implicated in approximately 0.3% of all solid tumors, with the frequency varying by tumor type, ranging from <1% to 3% in common tumor histologies (such as lung cancer and CRC) to approaching 100% in rare histologies (such as MASC and IFS).178,179 . Adult and pediatric patients with TRK fusion cancer have shown marked and durable responses to TRK inhibitors to-date.180 o MSI appears to be a generalized cancer phenotype in about 4% of all adult cancers in total and is another example of a predictive mutational phenotype. 46 . Patients with dMMR/MSI-H cancers have also been shown to be sensitive to an immune checkpoint blockade regardless of the cancer’s tissue of origin. . FoundationOne CDx is a clinically validated, FDA-approved companion diagnostic for detection of NTRK gene fusions and TMB-H across solid tumor types.1

Unmet Need of CGP in Pan-Tumors The mechanisms underlying cancer have been investigated for >100 years; however, the clinical management of cancer remains rooted in treatments that rely on surgery to remove tumors and chemotherapy and/or radiation therapy to stop uncontrolled cellular proliferation.181 Morphological and histopathological methods are largely used to diagnose patients and estimate prognosis.181 Additionally, treatment selection has been largely based on tumor site, histology, tumor stage, and prior response to therapy. . Cancer development is now understood to be driven by genomic alterations, such as mutations (ie, changes in deoxyribonucleic acid [DNA] sequence that make up a gene), structural variations or rearrangements (ie, changes in the orientation, location, or number of copies of larger DNA segments which include fusions, translocations, inversions, deletions, and duplications), or variation in the number of copies of a gene (ie, gene amplification).127,182 Precision medicine (also called personalized medicine) aims to prevent, diagnose, and treat disease based on an individual’s specific biological, genetic, environmental, and disease state characteristics.127 In cancer care, precision medicine can refer to the use of therapeutic interventions that are expected to benefit a subset of patients whose cancer shows specific molecular or cellular features, such as genetic mutations and other types of genomic alterations (such as gene fusions) or protein expression patterns.127 Advances in precision medicine have significantly evolved over the past several years, with the testing becoming both quicker and less costly and thus allowing more patients to benefit from the results of it.183

CONFIDENTIAL PAGE 58 US-FDX-2000073 Last update: January 22, 2021

Further, these sequencing technologies are detecting previously undetected genomic alteration across a multitude of cancers, thereby allowing the emergence of novel targets of therapy.184 Genomic analyses within oncology practice have shown that many genomic alterations are rare and are found across tumor histologies.185 Figure 5-1 shows that many genomic alterations are common in one tumor type, whereas they are rare in others.186 In particular, this is seen with NTRK gene fusions which have varying frequencies reported across solid tumor histologies; however, the overall prevalence of NTRK gene fusions within most of these histologies is very low as shown in Figure 5-2.187 This provides evidence of the need for molecular testing, regardless of the histologic type of cancer, as there may be actionable alterations found. CGP overcomes this challenge because it allows the potential ability to capture these rare events at the same time as testing for more common alterations without the sacrifice of additional time or tissue.

Figure 5-1. Spectrum of Genomic Alterations Across Tumor Types

Source: Rahal 2016186.

CONFIDENTIAL PAGE 59 US-FDX-2000073 Last update: January 22, 2021

Figure 5-2. Prevalence of NTRK Gene Fusions by Tumor Type

Inflammatory myofibroblastic tumor 17.70% Salivary gland carcinoma 5.08% Thyroid carcinoma 2.28% Sarcoma 0.68% Glioma 0.55% Appendiceal adenocarcinoma 0.48% Melanoma 0.36% Pancreatic adenocarcinoma 0.34% Tumor Type Colorectal carcinoma 0.31% Neuroendocrine tumor 0.31% Cholangiocarcinoma 0.25% Lung adenocarcinoma 0.23% Breast carcinoma 0.13%

Frequeny of NTRK gene fusions, %

NTRK, neurotrophic receptor tyrosine kinase. Source: Solomon 2020187. Facilitation of the identification highly relevant and actionable genomic alterations is becoming ever more important in oncology practice as there are currently 3 FDA-approved tumor agnostic therapies that have been shown in clinical trials to improve outcomes across tumor types based on a specific genetic alteration or protein expression pattern (Table 5-1).188 Further, identification of genomic alterations not only opens up FDA approved therapies, but also the potential for entry into a genomically matched study.188 Three of these studies, which are offering treatment across a variety of tumor types based on genomic matching, are the NCI-MATCH (NCT02465060), MyPathway (NCT02091141), and the American Society of Clinical Oncology (ASCO) Targeted Agent and Profiling Utilization Registry (TAPUR) (NCT02693535).188

Table 5-1. FDA-approved Tumor Agnostic Therapies FDA-approved therapy Indication Pembrolizumab . For the treatment of adult and pediatric patients with unresectable or metastatic, MSI-H or dMMR solid tumors that have progressed following prior treatment and who have no satisfactory alternative treatment options . For the treatment of adult and pediatric patients with unresectable or metastatic TMB-H [≥10 mutations/megabase] solid tumors, as determined by an FDA- approved test, that have progressed following prior treatment and who have no satisfactory alternative treatment options Larotrectinib . For the treatment of adult and pediatric patients with solid tumors that: have a NTRK gene fusion without a known acquired resistance mutation, are metastatic or where surgical resection is likely to result in severe morbidity

CONFIDENTIAL PAGE 60 US-FDX-2000073 Last update: January 22, 2021

and have no satisfactory alternative treatments or that have progressed following treatment. Entrectinib . For the treatment of adult and pediatric patients 12 years of age and older with solid tumors that: have a NTRK gene fusion without a known acquired resistance mutation, are metastatic or where surgical resection is likely to result in severe morbidity, and have no satisfactory alternative treatments or that have progressed following treatment. dMMR, mismatch repair deficient; FDA, Food and Drug Administration; MSI-H, microsatellite instability-high; NTRK, neurotrophic receptor tyrosine kinase; TMB, tumor mutational burden-high Source: Keytruda PI 202043; Vitrakvi PI 2019189; Rozlytrek PI 2019190.

As there is increased focus on precision medicine, tumour-agnostic therapies have emerged as a revolutionary new approach to cancer treatment.191 In recent years, tumor agnostic therapies have become a key focus area for an increasing number of small and large pharmaceutical companies.191 Table 5-2 shows select tumor agnostic therapies in development.

Table 5-2. Select Tumor Agnostic Therapies in Development Agent Biomarker Studied indication Trial Atezolizumab MSI-H, TMB-H Solid tumors with MSI-H, TMB-H, or alterations in Phase 3 DNA proofreading genes Merestinib MET, NTRK Solid tumors with NTRK rearrangements Phase 2 PLX8394 Mutant BRAF and Solid tumors with BRAF mutations Phase 1/2 wild-type CRAF PLX9486 KIT Solid tumors with KIT mutations Phase 1/2 LY3300054 MSI-H Monotherapy with MSI-H solid tumors; various other Phase 1/2 combination criteria LOXO-195 NTRK Solid tumors with NTRK fusions, including those Phase 1/2 resistant to larotrectinib TPX-0005 NTRK, ALK, ROS1 Solid tumors with NTRK, ALK, or ROS1 Phase 1/2 rearrangements LOXO-292 RET Solid tumors with RET rearrangements Phase 1/2 RXDX-105 RET Solid tumors with RET fusions Phase 1 ALK, anaplastic lymphoma kinase; DNA, deoxyribonuclease acid; MSI-H, microsatellite instability high; NTRK, neurotrophic tropomyosin receptor tyrosine kinase; PD-L1, programmed death-ligand 1; RET, rearranged during transfection proto-oncogene; TRK, tropomyosin receptor kinase. Source: Garber 2018192; ClinicalTrials.gov 2020193

Although there is currently a paucity of data for molecular testing across tumor types for tumor agnostic therapies and for the testing rates of rare genomic alterations, such as NTRK gene fusions, there is recently published data in NSCLC that identifies that the majority of patients do not receive testing for rare genomic alterations.146

CONFIDENTIAL PAGE 61 US-FDX-2000073 Last update: January 22, 2021

. A retrospective database study included 1,203 advanced NSCLC patients treated across 5 community oncology practices since January 1, 20179).146 The percentage of advanced NSCLC patients who received testing for EGFR mutations or ALK rearrangements was 54% and 51%, respectively.146 Both of these genomic alterations occur in >5% of NSCLC patients.194,195 However, for 2 other targetable alterations, ROS1 and BRAF, which occur in approximately 1% to 2% and 2% to 4%, respectively, patients less frequently received testing; 43% of patients received testing for ROS1 and only 29% of patients received testing for BRAF.146,195-197 As such, there is an unmet need for platforms that have the potential ability to capture rare events at the same time as testing for more common alterations without the sacrifice of additional time or tissue.

Place of CGP in the Pan-Tumor Evaluation of TMB, MSI-H, and NTRK TMB Neoantigens are protein products of non-driver alterations that are recognized by the immune system and can contribute to immune checkpoint inhibitor response. TMB is a surrogate measure for neoantigen burden and may help identify patients more likely to respond to immune checkpoint inhibitors. TMB by FoundationOne CDx is determined by counting all synonymous and non-synonymous variants present at 5% allele frequency or greater (after filtering) and the total number is reported as mutations per megabase (mut/Mb) unit. TMB was originally measured using whole exome sequencing (WES), a methodology that typically counts only nonsynonymous base substitutions that alter the amino acid sequence of a protein; this infers a direct link between protein coding changes and the number of potential neoantigens within a tumor genome. Whereas targeted NGS panels estimate TMB by including nonsynonymous and synonymous base substitutions and short insertion and deletion alterations in the calculation. Synonymous variants, which do not alter the amino acid sequence of a protein, are not assumed to generate neoantigens. However, the presence of synonymous variants is indicative of a mutational process also likely to result in nonsynonymous variants; therefore, their inclusion in the TMB algorithm effectively improves assay sensitivity by increasing the number of qualifying variants into the calculation.198 In addition to TMB, MSI and PD-L1 expression are also relevant, but not completely overlapping, biomarkers for immunotherapy selection (Figure 5-3).51,138,198-201 This was also demonstrated in KEYNOTE-158, where patients with TMB-high (TMB-H) (≥10 mut/Mb) tumors had better ORR, regardless of PD-L1 or MSI (Figure 5-4).

CONFIDENTIAL PAGE 62 US-FDX-2000073 Last update: January 22, 2021

Figure 5-3. Overlap of Immunotherapy Biomarkers Among PD-L1 Tested Patients Across Tumor Types

IHC, immunohistochemistry; MSI, microsatellite instability; TMB, tumor mutational burden. Source: Foundation Medicine database as of June 2017.

Figure 5-4. Overall Response Rate with Pembrolizumab in KEYNOTE-158

MSI-H, microsatellite instability-high; tTMB, tissue tumor mutational burden. Source: Sigornovitch 2017202

While TMB can be precisely quantified with WES, this is not currently a feasible approach to testing in the clinical setting due to availability, cost/reimbursement, and prolonged turnaround time. Targeted NGS panels interrogate a sufficient (yet considerably smaller) portion of the genome; however, these panels can efficiently and accurately determine TMB.51,198,203 With the pan-tumor approval of pembrolizumab across solid tumors with TMB-H (≥10 mut/Mb) status, certain NCCN Guidelines have followed with recommendations for TMB testing.40,41,43,204-208

CONFIDENTIAL PAGE 63 US-FDX-2000073 Last update: January 22, 2021

NTRK NTRK gene fusions have been reported in both the pediatric and adult patient populations, including the following histologies: CRC, NSCLC, thyroid carcinoma, spitzoid melanoma, glioblastoma multiforme (GBM), IFS, MASC of the salivary gland, invasive breast carcinoma, secretory breast carcinoma, CMN, GIST, STS, head and neck carcinomas, acute myeloid leukemia, intrahepatic cholangiocarcinoma, and many other tumors.178,179,184 NTRK gene fusions are implicated in approximately 0.3% of all solid tumors.178,179 However, the frequency of NTRK gene fusions varies by tumor type, ranging from <1% to 3% in common tumor histologies (such as lung cancer and CRC) to approaching 100% in rare histologies (such as MASC and IFS).179,209

The importance of finding patients with NTRK gene fusions cannot be understated, as the results of the trials for recently approved TRK inhibitors have shown impressive results. Adult and pediatric patients with TRK fusion cancer have shown durable response to TRK inhibitors to-date.180 . In 55 pediatric or adult patients with tumor harboring NTRK gene fusions, larotrectinib was shown to produce high RRs at the time of the primary analysis with 13% of patients having a CRs and 62% having a PR (ORR: 75%).102 Further, in a recent update of this analysis including 159 patients with a median follow-up of 13.9 months, the median overall survival was 44.4 months (95% CI: 36.5 months, NE) with an OS rate at 12 months of 88%.180 . In 54 adults with locally advanced or metastatic NTRK gene fusion-positive solid tumors treated with entrectinib the ORR was 57% with 7% of patients having a CR and 50% having a PR to entrectinib therapy.210 The median OS was 21 months (95% CI: 14.9, NE).210 Further, in patients with baseline CNS disease (n=12), the ORR was 50% with a median PFS of 7.7 months (95% CI: 4.7, NE).210 With the pan-tumor approval of larotrectinib and entrectinib across solid tumors harboring NTRK gene fusion, certain NCCN Guidelines have followed with recommendations for NTRK gene fusion testing.27,30,32-35,39-41,189,190,206-208,211,212 MSI MSI, which arises from defects in the DNA MMR system, is another example of a predictive mutational phenotype. MSI appears to be a generalized cancer phenotype in about 4% of all adult cancers in total.46 Pembrolizumab was the first tissue/site-agnostic drug approved by the FDA for treatment of MSI-H solid tumors. In the clinical trials supporting the approval, MSI status was determined by local laboratory- developed PCR tests for MSI-H status or IHC tests for dMMR.96 In particular, pembrolizumab was granted accelerated approval for adult and pediatric patients who have unresectable or metastatic MSI-H or dMMR solid tumors that have progressed following prior treatment and who have no satisfactory alternative treatment options (or for patients with MSI-H or dMMR CRC that has progressed following treatment with a fluoropyrimidine, oxaliplatin, and irinotecan).213 Patients with dMMR cancers have also been shown to be sensitive to an immune checkpoint blockade regardless of the cancer’s tissue of origin (Table 2-3).214 MMR or MSI testing are also recommended for certain types of cancer; NCCN Guidelines for Colon Cancer and Rectal Cancer recommend MSI/MMR testing in all newly diagnosed patients with CRC and NCCN Guidelines for Uterine Neoplasms recommend universal MMR and MSI testing of endometrial

CONFIDENTIAL PAGE 64 US-FDX-2000073 Last update: January 22, 2021

carcinomas.30,39,40 With the pan-tumor approval of pembrolizumab across solid tumors for MSI-H status, certain NCCN Guidelines have followed with recommendations for testing of MSI status.27,30,32-34,36- 41,43,204,206,208,215,216

Clinical Utility and Validity of FoundationOne CDx in Pan Tumors TMB On June 16, 2020, FoundationOne CDx received FDA-approval as the first companion diagnostic to identify TMB-H ( ≥10 mut/Mb) to inform treatment with pembrolizumab in patients with advanced solid tumors.1 Pembrolizumab is indicated for the treatment of adult and pediatric patients with unresectable or metastatic TMB-H (≥10 mut/Mb) solid tumors, as determined by an FDA-approved test, that have progressed following prior treatment and who have no satisfactory alternative treatment options.43 This indication received accelerated approvala based on tumor response rate and durability of response.43 While TMB was originally developed as a count of somatic mutations determined using WES, TMB can accurately be estimated using FoundationOne CDx by counting all synonymous and non-synonymous substitution and indel variants present at 5% allele frequency or greater and filtering out potential germline variants according to published databases of known germline polymorphisms and using a somatic-germline/zygosity algorithm, and reporting TMB as mut/Mb of sequenced DNA. Substantial variability in different panel-based estimates of TMB have been observed.217-221 TMB estimation using genomic panels requires validated and complex filtering methods to account for germline and driver alterations.198,218 Other sources of variation can include217,220: . quality and quantity of DNA input; . sequencing depth; . coverage; and . variant allele frequency cut-points. Together, the current evidence suggests that a TMB score reported in one lab may not directly correlate clinically to the TMB score reported by another lab. An industry-wide effort, led by the Friends of Cancer Research, is developing guidelines for standardizing the way TMB is calculated and reported (FoCR). This includes recommendations for standardizing the analytical validation of TMB scores estimated by commercially available genomic panels and clinical application of TMB broadly. Recently published consensus recommendations from this effort state that TMB assays should validate against WES or an FDA-approved companion diagnostic, if available.220 Thus, following FDA-approval as a companion diagnostic for pembrolizumab in TMB-H (≥10 mut/Mb) tumors, FoundationOne CDx is a recognized standard for validation of panel TMB measurement, having both FDA approval and robust validation vs WES.1,167 The gold standard for measuring TMB has historically been WES.221-223 FoundationOne CDx was validated against WES in a concordance study of 218 samples using a TMB threshold of 10 mut/Mb.224 . The overall PPA (95% CI) was 87.3% (64.4%, 96.2%) and the overall NPA (95% CI) was 91.6% (85.7%, 95.6%). Other validation studies have shown good correlation between TMB from Foundation Medicine with WES.51,225 Figure 5-5 below shows correlation of Foundation

a Continued approval for this indication may be contingent upon verification and description of clinical benefit in the confirmatory trials.

CONFIDENTIAL PAGE 65 US-FDX-2000073 Last update: January 22, 2021

Medicine TMB with WES (Panel A), the consistency of Foundation Medicine between replicates (Panel B), and the increasing correlation between TMB and WES as the total amount of DNA sequenced increases.

Figure 5-5. Correlation of Foundation Medicine TMB and WES (A), Reproducibility of Foundation Medicine TMB (B), and Correlation of Panel TMB and WES by DNA Sequence Amount (C)

Mb, megabase; mut, mutations; TMB, tumor mutation burden.

Source: Chalmers 201751

The pembrolizumab approval for solid tumors with TMB-H (≥10 mut/Mb) was based on the KEYNOTE- 158 study, a phase 2 basket trial of pembrolizumab in 1,066 patients with select advanced solid tumors enrolled in cohorts A through J.43,202,226 The assessment of antitumor activity according to TMB using a threshold of ≥10 mut/Mb was a prespecified analysis. This threshold was prespecified before the TMB results were correlated with outcomes and was analytically validated for accuracy, precision, and

CONFIDENTIAL PAGE 66 US-FDX-2000073 Last update: January 22, 2021

sensitivity on the FoundationOne CDx platform following guidelines from harmonization efforts as well as the FDA.1 Of the efficacy analysis population, 790 patients had sufficient tissue for testing with the FoundationOne CDx CTA; of these, 102 (13%) had TMB-H (≥10 mut/Mb).1 The primary efficacy outcome was ORR, defined as the proportion of patients in the analysis population who had a response (CR or PR) per central review as assessed by RECIST v1.1 (Table 5-3).1 . TMB-H (≥10 mut/Mb) was associated with a clinically meaningful improvement in patients with previously treated solid tumors, as demonstrated by an ORR of 29% (95% CI: 21%, 39%), compared to 6% (95% CI: 5%, 8%) in the non-TMB-H group (primary endpoint). At a median follow-up of approximately 3 years, the median duration of response was not reached in the TMB-H (≥10 mut/Mb) group and was 33.1 months in the non-TMB-H group.1

Table 5-3. Clinical Utility of FoundationOne CDx for Determining TMB-H Status (≥10 mut/Mb) in Solid Tumors in the Efficacy Analysis Population TMB ≥10 mut/Mb TMB <10 mut/Mb (n=102) (n=688) ORRa, % (95% CI) 29 (21, 39) 6 (5, 8) PR 26 (17, 35) 5 (3, 7) CR 4 (1, 10) 2 (1, 3) a Central radiology assessed responses per RECIST 1.1 (confirmed) CI, confidence interval; CR, complete response; mut/Mb, mutations per megabase; ORR, objective response rate; PR, partial response; TMB, tumor mutational burden. Source: FoundationOne CDx Label1. Sigornovitch 2017202.

NTRK For the intended use to identify NTRK1, NTRK2, or NTRK3 fusions in solid tumor cancer patients who may benefit from treatment with larotrectinib, the effectiveness of the FoundationOne CDx assay was demonstrated through a clinical bridging study using specimens from the patients enrolled in the LOXO- TRK-14001 (Bayer 20288, NCT02122913), -15002 (Bayer 20289, NAVIGATE, NCT02576431), and - 15003 (Bayer 20290, SCOUT, NCT02637687) (14001, 15002, and 15003) clinical trials with known NTRK fusion status and an additional set of NTRK fusion negative samples from the FMI archives.1,101,102 Study 14001 is an ongoing multicenter open-label Phase 1 dose escalation study in adult patients with advanced solid tumors (all comers) unselected for NTRK gene fusion cancer; study 15002 is an ongoing multicenter open-label Phase 2 “basket” study in patients age 12 and older with recurrent advanced solid tumors with a documented NTRK gene fusion; and study 15003 is an ongoing multicenter open-label Phase 1/2 study in pediatric patients aged from birth to 21 years with advanced solid or primary CNS tumors.1,101,102 The bridging study assessed the concordance of results for the NTRK gene fusion status between the FoundationOne CDx assay and the local CTAs and established the clinical validity of FoundationOne CDx in identifying NTRK fusion positive patients for treatment with larotrectinib. Local CTAs included DNA NGS, RNA NGS, FISH, and reverse transcriptase- polymerase chain reaction (RT-PCR) methods. The majority of 105 clinical trial patients with known NTRK fusion status enrolled into the trials had been tested with NGS methods (92%); 51% of the 105 patients had been tested with DNA NGS

CONFIDENTIAL PAGE 67 US-FDX-2000073 Last update: January 22, 2021

methods and 41% with RNA NGS methods. The clinical bridging study included 45 CTA-positive samples and 24 CTA-negative samples from patients enrolled in 14001, 15002, and 15003, supplemented with 206 NTRK fusion negative samples from FMI archives to demonstrate the safety and effectiveness of F1CDx for identification of patients with solid tumors who may be eligible for treatment with larotrectinib.1 . Test concordance between FoundationOne CDx and the CTAs in identifying NTRK gene fusion status was reported as positive percent agreement (PPA), negative percent agreement (NPA), and overall percent agreement (OPA), and the results were as follows: PPA: 84.1% (95% CI: 69.9, 93.4); NPA: 100.0% (95% CI: 98.4, 100.0); OPA: 97.4% (95% CI: 94.7, 99.0) (Table 5-4).1 . Clinical validity of determining NTRK gene fusions utilizing the FoundationOne CDx was evaluated by estimation of clinical efficacy in the FoundationOne CDx-positive patients and the CTA-positive population from the clinical trials as assessed by independent review committee using RECIST 1.1 criteria. ORR of larotrectinib in the FoundationOne CDx- positive patient population was 77% (95% CI: 56, 91); this was comparable to the efficacy for the NDA filing, where the ORR of larotrectinib was reported to be: 75% (95% CI: 61, 85) (Table 5-4).1

Table 5-4. Clinical Validity and Utility of FoundationOne CDx for Pan-tumor NTRK Gene Fusion Companion Diagnostic Claim Concordance Clinical efficacy ORRa, % (n/N) (95% CI) FoundationOne CTA Drug and PPA, % NPA, Comparator CDx results results alteration (n/N) % (n/N) assay n=26 n=55 Larotrectinib 84.1 100 77 (20/26) 75 (41/55) NTRK gene CTA (37/44)b (226/226)b (56, 91) (61, 85) fusions a ORR was assessed by independent review committee using RECIST v1.1. b Concordance between FoundationOne CDx and the CTA for NTRK gene fusions excludes invalid FoundationOne CDx test results. CI, confidence interval; CLIA, Clinical Laboratory Improvement Amendments; CTA, clinical trial assay; NPA, negative percent agreement; ORR, overall response rate; PPA, positive percent agreement. Source: FoundationOne CDx Label1; FoundationOne CDx SSED 017104

MSI The MSI-H/MSS designation by FoundationOne CDx test is based on genome wide analysis of 95 microsatellite loci and not based on the 5 or 7 MSI loci described in current clinical practice guidelines.1 227,228The qualitative output for MSI (MSI-H vs microsatellite stable [MSS]) in the FoundationOne LDT and FoundationOne CDx were evaluated. PPA, NPA and OPA of MSI status between the 2 assays was calculated for all 165 samples. Of the 165 samples, 5 were MSI-H by FoundationOne LDT and 160 were MSS by FoundationOne LDT; there was 1 discordant sample observed. The discordant sample was called

CONFIDENTIAL PAGE 68 US-FDX-2000073 Last update: January 22, 2021

MSS by FoundationOne LDT and MSI-H by FoundationOne CDx. After manual review, the discordant case had an MSI score close to the threshold used to classify MSI status.1 . For the concordance between FoundationOne LDT and FoundationOne CDx, the PPA was 100% (95% CI: 47.8%, 100%), NPA was 99.5% (95% CI: 96.6%, 99.98%), and OPA was 99.4% (95% CI: 96.7%, 99.98%).1

CONFIDENTIAL PAGE 69 US-FDX-2000073 Last update: January 22, 2021

CGP in NSCLC

. Lung cancer accounts for an estimated 13% of new cancer diagnoses and up to 22% of cancer- related deaths in 2020, with NSCLC accounting for 80% of cases.229,230 Survival for advanced metastatic lung cancer at 5 years is 5.8%, with over half of the patients (57%) diagnosed at this stage.229 . Historically, traditional chemotherapy has resulted in poor survival outcomes, with a median survival of less than 1 year in a real-world population with advanced NSCLC.231 . Conversely, outcomes of patients with advanced NSCLC with an oncogenic driver given genotype-directed therapy are much improved compared with standard of care chemotherapy.232 . Actionable mutations in NSCLC are numerous, with up to 1 in 3 patients having a mutation with an FDA-approved matched therapy (Figure 5-6).142 Figure 5-6. Prevalence of Actionable Genomic Alterations in NSCLC Patientsa233-239

a Additionally, PD-L1, an immune biomarker, is found in up to 66% of patients with NSCLC and may be regulated by certain oncogenic drivers.240 b EGFR driver mutation prevalence varies by ethnicity; for example, among East Asians the prevalence is 40%–55% and among Caucasians is 5%–15%.241 . Pertinent professional societies that provide clinical care guidance in metastatic NSCLC are aligned that broad molecular profiling panel tests are preferred over multiple single-gene tests to more broadly capture targeted treatment options outside of EGFR, ALK, BRAF, and ROS1.35,242,243 . FoundationOne CDx is an FDA-approved companion diagnostic for EGFR alterations (afatinib, erlotinib, gefitinib, osimertinib), ALK rearrangements (alectinib, ceritinib, crizotinib), BRAF V600E alterations (dabrafenib with trametinib), and MET exon 14 skipping alterations (capmatinib).1

Unmet Need for Molecular Testing in NSCLC There are numerous mutations in advanced NSCLC that affect treatment decisions; targetable genomic alterations that occur in NSCLC include EGFR mutations, ALK rearrangements, ROS1 rearrangements, BRAF mutations, and NTRK gene fusions.35,142 Outcomes of patients with advanced NSCLC with an oncogenic driver given genotype-directed therapy are much improved as compared with standard of care chemotherapy; this has been shown in randomized controlled trials as well as in real-world populations of patients with NSCLC.232

CONFIDENTIAL PAGE 70 US-FDX-2000073 Last update: January 22, 2021

. In a study that included 14 sites within the US that enrolled patients with metastatic NSCLC (n=1007) and tested their tumors for 10 oncogenic drivers (EGFR, ALK, KRAS, NRAS, BRAF, ERBB2, PIK3CA, MEK, AKT1, and MET), the median OS was significantly prolonged for patients with an oncogenic driver mutation that received genotype-directed targeted therapy (Table 5-5).232

Table 5-5. Overall Survival by Receipt vs No Receipt of Genotype-Directed Therapy in Metastatic NSCLC Patients Population n Median survival (95% CI) Oncogenic driver and genotype-directed therapya 260 3.5 years (3.0, 4.3) Oncogenic driver and no genotype-directed therapya 318 2.4 years (1.8, 2.9) No oncogenic driver detected 360 2.1 years (1.8, 2.5) a Propensity score-adjusted HR (driver mutation detected and genotype-directed therapy given vs driver mutation detected but no genotype-directed therapy given): 0.69 [95% CI: 0.53, 0.9]; P=0.006). CI, confidence interval; HR, hazard ratio; NSCLC, non-small cell lung cancer. Source: Kris 2014232

Despite the improvement of outcomes for NSCLC patients treated with therapy targeted to a specific oncogenic driver, not all eligible patients are being tested. . A retrospective analysis, within a large electronic health record database, of patients with advanced NSCLC (n=1,203) receiving treatment within community practices (encompassing 289 oncologists) found that the testing rate for all biomarkers with an FDA-approved on-label drug (EGFR, ALK, ROS1, and BRAF) was 22%, and testing for all 7 NCCN-recommended genes for associated therapies was 7%, excluding testing for PD-L1 as it is can be measured via immunohistochemistry (IHC) rather than via genomic testing.146

Place of CGP in NSCLC The NCCN Guidelines for NSCLC recommend molecular testing in eligible patients with metastatic NSCLC and strongly advise broad molecular profiling testing using a validated test(s) that assesses a minimum of the following potential genetic variants: EGFR mutations, ALK fusions, BRAF mutations, METex14 skipping mutations, RET rearrangements, NTRK1/2/3 gene fusions, and ROS1 fusions (Table 5-6). Broad molecular profiling is also recommended to identify rare driver mutations for which effective therapy may be available, such as high level MET amplification and ERBB2 mutations.35

Table 5-6. Review of Guideline-Recommended Biomarker Testing and FoundationOne CDx Assay Capabilities Biomarker included Applicable guidelines for Recommended in FoundationOne tumor profiling biomarker testing CDx assay NCCN Guidelines for EGFR exon 19 deletions or EGFR exon 21 L858R Yes NSCLCa,b (V.2.2021)35 alterations (Category 1 for both)

CONFIDENTIAL PAGE 71 US-FDX-2000073 Last update: January 22, 2021

Biomarker included Applicable guidelines for Recommended in FoundationOne tumor profiling biomarker testing CDx assay EGFR exon 20 T790M alterations Yes ALK rearrangements (Category 1) Yes BRAF V600E Yes ROS1 rearrangements Yes PD-L1 Noc MET exon 14 skipping mutation Yes High-level MET amplificationd Yes RET rearrangements Yes ERBB2 (HER2) mutationsd Yes NTRK1/2/3 gene fusions Yes IASLC/CAP/AMP guideline242 EGFR Yes ALK Yes ROS1 Yes ERBB2e Yes MET e Yes BRAFe Yes KRAS e Yes RET e Yes IASLC statement244 EGFR Yes ALK Yes ROS1 Yes BRAF Yes a The NCCN Guidelines for NSCLC provide recommendations for individual biomarkers that should be tested and recommend testing techniques but do not endorse any specific commercially available biomarker assays. b Category 1: Based upon high-level evidence, there is uniform NCCN consensus that the intervention is appropriate. Category 2A: Based upon lower-level evidence, there is uniform NCCN consensus that the intervention is appropriate. All NCCN recommendations are category 2A unless otherwise indicated. c PD-L1 testing is available from Foundation Medicine. d Defined as an emerging biomarker. e Molecular testing for this gene is not indicated as a routine stand-alone assay outside the context of a clinical trial. It is appropriate to include as part of larger testing panels performed either initially or when routine EGFR, ALK, and ROS1 testing are negative. ALK, anaplastic lymphoma kinase; AMP, Association of Molecular Pathologists; CAP, College of American Pathologists; EGFR, epidermal growth factor receptor; HER2, human epidermal growth factor receptor 2; IASLC, International Association for the Study of Lung Cancer; KRAS, V-Ki-ras2 Kirsten rat sarcoma; MET, mesenchymal epithelial transition factor receptor; NCCN, National Comprehensive Cancer Network; NSCLC, non-small cell lung cancer; NTRK, neurotrophic receptor tyrosine kinase; PD-L1, programmed death ligand-1.

CONFIDENTIAL PAGE 72 US-FDX-2000073 Last update: January 22, 2021

Clinical Utility and Validity of FoundationOne CDx in NSCLC FoundationOne CDx is an in vitro diagnostic device that targets 324 genes. The test also detects the genomic signatures TMB and MSI-H. FoundationOne CDx utilizes DNA isolated from FFPE from the tumor tissue of cancer patients. The test is intended to be used as a companion diagnostic to identify patients with NSCLC who may benefit from treatment with the targeted therapies listed in Table 5-7 in accordance with the approved therapeutic product labeling. Additionally, FoundationOne CDx is intended to provide tumor mutation profiling to be used by qualified healthcare professionals in accordance with professional guidelines in oncology for patients with malignant neoplasms, including NSCLC.1 As described in the Pan-tumor section, other biomarkers (NTRK, TMB) are relevant in all solid tumors.

Table 5-7. Companion Diagnostic Indications Pertinent for Patients With NSCLC Tumor Type Biomarker(s) Detected Therapy Gilotrif® (afatinib) EGFR exon 19 deletions and Iressa® (gefitinib) EGFR exon 21 L858R alterations Tagrisso® (osimertinib) Tarceva® (erlotinib) EGFR exon 20 T790M alterations Tagrisso® (osimertinib) ® NSCLC Alecensa (alectinib) ALK rearrangements Xalkori® (crizotinib) Zykadia® (ceritinib) Tafinlar® (dabrafenib) in combination BRAF V600E with Mekinist® (trametinib) MET exon 14 skipping Tabrecta™ (capmatinib) ALK, anaplastic lymphoma kinase; BRAF, v-Raf murine sarcoma viral oncogene homolog B; EGFR, epidermal growth factor receptor; MET, mesenchymal-epithelial transition; NSCLC, non-small cell lung cancer. Source: FoundationOne CDx Label1

EGFR

EGFR Exon19del/L858R Clinical validity of the FoundationOne CDx assay was evaluated as a companion diagnostic in identification of patients with advanced NSCLC who may be eligible for treatment with drugs targeting EGFR exon19 and L858R mutations (ie, afatinib, gefitinib, osimertinib, erlotinib). Retrospective samples (n=282) from NSCLC patients were included in this study, which were tested for EGFR alterations (specifically, exon 19 deletion and exon 21 L858R) by the FoundationOne CDx assay and the previously approved cobas® EGFR Mutation Test v2 (Roche Molecular Systems, referred to cobas assay). Samples were tested across 2 replicates by the cobas assay (denoted as CCD1 and CCD2) and 1 replicate by FoundationOne CDx.1 . The PPA and NPA between the combination of cobas replicates (CCD1 and CCD2) and FoundationOne CDx assays were as follows: PPA: 98.1%; NPA: 99.4% (Table 5-8).1 . Based on these results, FoundationOne CDx has been demonstrated to be noninferior to the cobas assay for the detection of EGFR exon 19 deletions and EGFR exon 21 L858R mutations.1

CONFIDENTIAL PAGE 73 US-FDX-2000073 Last update: January 22, 2021

EGFR T790M Clinical validity of the FoundationOne CDx assay was evaluated as a companion diagnostic in identification of patients with advanced NSCLC who may be eligible for treatment with drugs targeting EGFR T790M alterations (ie, osimertinib). Retrospective samples (n=227) from NSCLC patients were included in this study, which were tested for EGFR alterations (specifically, exon 20 T790M alterations) by the FoundationOne CDx assay and the previously approved cobas® EGFR Mutation Test v2 (Roche Molecular Systems, referred to cobas assay). Samples were tested across 2 replicates by the cobas assay (denoted as CCD1 and CCD2) and 1 replicate by FoundationOne CDx.1 . The PPA and NPA between the combination of cobas replicates (CCD1 and CCD2) and FoundationOne CDx assays were as follows: PPA: 98.9%; NPA: 86.1% (Table 5-8).1 . Based on these results, FoundationOne CDx has been demonstrated to be concordant to the cobas assay for the detection of EGFR T790M mutations.1

Table 5-8. Clinical Validity of FoundationOne CDx for NSCLC EGFR Mutation Companion Diagnostic Claims Biomarker PPA, % (n/N) NPA, % (n/N) Comparator assay EGFR exon 19 98.1 (106/108) 99.4 (153/154) cobas® EGFR Mutation Test v2 deletions and exon 21 L858R EGFR T790M 98.9 (87/88) 86.1 (93/108) cobas® EGFR Mutation Test v2 CTA, clinical trial assay; NPA, negative percent agreement; PPA, positive percent agreement. Source: FoundationOne CDx Label1

ALK The clinical validity of using FoundationOne CDx as a companion diagnostic to identify patients with ALK rearrangement-positive NSCLC and therefore be eligible for treatment with alectinib, crizotinib, or ceritinib was assessed through a phase 3 multicenter, open-label study (NCT02075840) to evaluate the efficacy and safety of alectinib compared with crizotinib in cancer patients with treatment-naïve ALK rearrangements. The ALK diagnostic results from the FoundationOne CDx panel were compared against those obtained from the FDA approved Ventana ALK (D5F3) CDx Assay (“Ventana IHC”, Ventana Medical Systems, Inc.) and Vysis ALK BreakApart FISH Probe Kit (“Vysis FISH”, Abbott Molecular) from NSCLC patient samples (n=175).1 . The PPA and NPA between the combination of Ventana ALK CDx assay and Vysis ALK break- apart FISH probe kit samples as the reference assays were as follows: PPA 92.9%; NPA is 100% (Table 5-9).1 . Based on these results, FoundationOne CDx results were concordant to the FDA approved IHC and FISH assays.1

Table 5-9. Clinical Validity of FoundationOne CDx for NSCLC ALK Rearrangement Companion Diagnostic Claims Biomarker PPA, % (n/N) NPA, % (n/N) Comparator assay

CONFIDENTIAL PAGE 74 US-FDX-2000073 Last update: January 22, 2021

ALK 92.9 (78/84) 100 (75/75) Ventana ALK (D5F3) CDx Assay Vysis ALK Break Apart FISH Probe Kit CTA, clinical trial assay; FISH, fluorescence in situ hybridization; NPA, negative percent agreement; PPA, positive percent agreement. Source: FoundationOne CDx Label1

BRAF V600E Clinical validity of the FoundationOne CDx assay was evaluated as a companion diagnostic in identification of patients with BRAF V600E alterations.1 . The PPA and NPA between the cobas BRAF V600E mutation test and FoundationOne CDx assays were as follows: PPA: 99.3%; NPA: 99.2% (Table 5-10).1 . Based on these results, FoundationOne CDx has been demonstrated to be concordant to the cobas BRAF V600 Mutation Test for the detection of BRAF V600E mutations.1

Table 5-10. Clinical Validity of FoundationOne CDx for NSCLC BRAF V600E Mutation Companion Diagnostic Claim Biomarker PPA, % (n/N) NPA, % (n/N) Comparator assay BRAF V600E 99.3 (149/150) 99.2 (121/122) cobas BRAF V600 Mutation Test NPA, negative percent agreement; PPA, positive percent agreement. Source: FoundationOne CDx Label1

MET exon 14 skipping The GEOMETRY-mono 1 is an open-label, single-arm, phase 2 study of capmatinib in stage IIIB/IV patients with ALK-negative, EGFR wild-type NSCLC.245,246 This was a multi-cohort study that enrolled patients into cohorts based on prior treatment and the presence of METexon14 skipping or MET amplification. Using a reverse transcriptase-PCR clinical trial assay (CTA), cohort 4 enrolled 69 patients with METexon14 skipping alteration and 1or 2 prior lines of therapy while cohort 5b enrolled 28 patients with METexon14 skipping alteration who were treatment naïve.1,105 FoundationOne CDx was retrospectively used to analyze samples from patients enrolled in the GEOMETRY-mono 1 trial.1,105 . Concordance between FoundationOne CDx and the CTA for GEOMETRY-mono 1 was established with the 97 relevant alteration positive samples from enrolled patients and 130 randomly selected CTA-negative samples. Of these 227 samples, 198 had FoundationOne CDx evaluable results.1 o The concordance between FoundationOne CDx and CTA in combined cohorts 4 and 5b from the GEOMETRY-mom 1 clinical trial, the PPA was 98.6% and the NPA was 100% (Table 5-11).1 . The clinical efficacy among FoundationOne CDx-positive patients was similar to the observed clinical efficacy among CTA-positive patients during the clinical trial (Table 5-11). The ORR in cohort 4 was 44.2% (95% CI: 30.6, 58.7) and 40.6% (95% CI: 28.9, 53.1) for FoundationOne CDx and the CTA, respectively. In cohort 5b, the ORR was 70% (95% CI: 45.7, 88.1) and 67.9% (95% CI; 47.6, 84.1) for FoundationOne CDx and the CTA, respectively.1

CONFIDENTIAL PAGE 75 US-FDX-2000073 Last update: January 22, 2021

Table 5-11. Clinical Validity and Utility of FoundationOne CDx for NSCLC MET Exon 14 Skipping Companion Diagnostic Claim Concordance Clinical efficacy ORRa, % (n/N) (95% CI) Drug and PPA, % NPA, % Comparator FoundationOne CTA results alteration (n/N) (n/N) assay CDx results Cohort 4b N=52 N=69

Capmatinib 44.2 (23/52) 40.6 (28/69) reverse MET exon 14 98.6 100 (30.6, 58.7) (28.9, 53.1) transcriptase-PCR skipping c (72/73) (125/125) CTA Cohort 5b

N=20 N=28 70.0 (14/20) 67.9 (19/28) (45.7, 88.1) (47.6, 84.1) a ORR as assessed by BICR according to RECIST v1.1. b Cohort 4: previously treated patients. c Cohort 5b: treatment-naïve patients. BICR, blinded independent central review; CI, confidence interval; CTA, clinical trial assay; ORR, overall response rate; RECIST, Response Evaluation Criteria in Solid Tumors. Source: FoundationOne CDx Label1; FoundationOne CDx SSED 011106

CONFIDENTIAL PAGE 76 US-FDX-2000073 Last update: January 22, 2021

CGP in Breast Cancer

. Breast cancer is the second-leading cause of cancer-related mortality among women in the US, with an estimated 42,170 deaths in 2020.247 . Localized and locally advanced breast cancer have high 5-year survival rates at 99% and 86%, respectively; however, the 5-year survival rate for metastatic breast cancer is 28%.116 . In advanced breast cancer, the NCCN Guidelines currently recommend patients have access to testing for HER2/ERBB2, BRCA1, BRCA2, PIK3CA, NTRK, PD-L1, and MSI-H/dMMR.27 . FoundationOne CDx is an FDA-approved companion diagnostic test for ERBB2 (HER2) amplifications to determine treatment with trastuzumab, ado-trastuzumab emtansine, or pertuzumab, and PIK3CA alterations to determine treatment with or alpelisib.1

Unmet Need for Molecular Testing in Breast Cancer Breast cancer is one of the leading causes of cancer-related mortality in the US, with over 40,000 deaths annually; as such, treatment in this setting remains as challenge.248 Treatment of metastatic breast cancer is largely based on hormone receptor status (ER/PR) and HER2 status.27 Targeting the ER and/or HER2 are the best established targeted treatment approaches in metastatic breast cancer.249 Still, with currently available therapies, metastatic breast cancer is considered an incurable disease, highlighting the need to define additional actionable targets for the treatment of these patients.248,250 Notably, there is a growing number of genomic alterations emerging, which may help facilitate a tailored approach to treatment of metastatic breast cancer.251 The most common targetable alterations in advanced breast cancer now include (Litton 2019)251: . Recurrent somatic mutations of ERBB2 occur in 2% to 4% of patients, most commonly in patients with HER2-negative breast cancer.147 . Activation mutations in the phosphoinositide 3 kinase (PI3K) pathway PIK3CA gene arise in nearly 40% of tumors (most common mutations in ER+ breast cancer).251 . Pathogenic mutations in BRCA1 and BRCA2 (BRCA1/2) arise in approximately 5% of breast cancers (most cases reflect an underlying germline mutation, although somatic mutations without a predisposing germline mutation can be found).251 . Across all breast cancer subtypes, 2% of tumors harbor MSI, reflecting underlying defects in mismatch DNA repair.251 Additionally, genomic testing results may also be relevant to determine eligibility for clinical trials with investigational agents.147 This approach may be particularly relevant for patients with advanced breast cancer, for whom standard treatments are minimally effective.147 There are several targeted therapies that are actively being studied in metastatic breast cancer, including those targeting AKT1 mutations or ESR1 mutations (Table 5-12).147

Table 5-12. Genomic Alterations With Targeted Therapies Available or in Clinical Trials Molecular alteration Prevalence Drug class Targets with FDA-approved therapies HER2 20−30% HER2 mAb or HER2 TKI

CONFIDENTIAL PAGE 77 US-FDX-2000073 Last update: January 22, 2021

BRCA1/BRCA2 5% PARP inhibitor therapy PIK3CA mutation 30%–40%; ER+/HER2−, 20%– PI3K inhibitor (α-isoform– 25%; TNBC, 10%–15% specific/selective) ERBB2 mutation 2%–4% HER2-negative HER2 TKI NTRK fusion Enriched in secretory (TNBC) TRK inhibitor High tumor mutational 1%–5% breast cancers Immune checkpoint inhibitor burden Mismatch repair deficiency <5% breast cancers Immune checkpoint inhibitor signature Targets under investigation in clinical trials AKT1 mutation 2%–5% breast cancer AKT inhibitor mTOR inhibitor ESR1 mutation 30%–40% ER+/HER2− after AI Oral selective estrogen receptor degrader AKT1, protein kinase B; BRCA, breast cancer gene; ER, estrogen receptor; ERBB2, Erb-B2 receptor tyrosine kinase 2; ESR1, estrogen receptor 1; HER2, human epidermal growth factor receptor 2; mAb, monoclonal antibody; mTOR, mammalian target of rapamycin; NTRK, neurotrophic tyrosine receptor kinase; PARP, poly (ADP-ribose) polymerase; PR, progesterone receptor; PIK3CA, phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha; SERM, selective estrogen receptor modulator; TKI, tyrosine kinase inhibitor; TNBC, triple negative breast cancer; TRK, tropomyosin receptor kinase. Source: Pauletti 2000148. Yaziji 2004149. Kratz 2018147.

The treatment of breast cancer has been largely determined by HR and HER2 status, as both hormonal and HER2-targeted therapies have greatly improved survival. However, several therapies based on novel (ie, non-HR or HER2) targets are creating opportunities for improved outcomes in women with metastatic breast cancer.249 These therapies provide potential opportunities, not only for women who are HR- and HER2-negative, but also for women who have exhausted all treatment options. Improvements in outcomes have been reported with the novel targeted therapies for patients with advanced breast cancer, including those targeting BRCA1/2 mutations, PI3KCA mutations, and PD-L1 expression. . Treatment with poly ADP ribose polymerase (PARP) inhibitors (olaparib, talazoparib) for the treatment of advanced breast cancer patients with BRCA1/2 mutations has results in an approximate 3 month improvement in PFS and a doubling a ORR (approximately 60% with PARP inhibitors vs 30% with standard of care).252253 . Advanced breast cancer patients with PIK3CA mutations treated with a PI3K inhibitor (alpelisib) had a 5 month improvement in PFS with more than double the ORR of patients treated with standard of care.107 . In PD-L1 positive advanced breast cancer, treating with a PD-L1 checkpoint inhibitor in addition to chemotherapy prolonged PFS by 2.5 months and improved ORR by approximately 30% vs chemotherapy alone.254 As targeted therapies have been shown to improve outcomes for women with breast cancer, there is a recognized need for molecular testing for genomic variations, as these have become integral in the management of advanced breast cancer.46,255 Although the determination of HR status and HER2 status is well known in the treatment paradigm of advanced breast cancer, other genomic alterations and

CONFIDENTIAL PAGE 78 US-FDX-2000073 Last update: January 22, 2021

biomarkers (ie, BRCA1/2 mutations, PIK3CA mutations, NTRK gene fusions, PD-L1 status, MSI- H/dMMR) are actionable either through an FDA-approved therapy or within a clinical trial and are equally important to test for in the management of advanced breast cancer.27,140 However, there are little data regarding the current clinical practice of genomic testing in patients with advanced breast cancer. The data currently available in advanced breast cancer primarily focus on germline cancer genetic testing.5,255 . A database registry study reviewed 77,085 women with breast cancer and 6,001 women with ovarian cancer diagnosed between 2013 and 2014 from the Georgia Cancer Registry and the California Cancer registry.255 The data from these patients were linked with germline genetic testing information from 4 laboratories that performed the majority of clinical testing.255 Only 24.1% of patients with breast cancer had testing results linked with any laboratory.255 . A real-world study sought to assess the somatic and/or germline BRCA1/2 testing rates in 1,285 HER2-negative adult women with advanced breast cancer in the US (Table 5-13).5 The BRCA1/2 testing rate observed for the overall sample was 50%, with significantly lower BRCA1/2 testing seen among HR+/HER2- vs TNBC patients (41% vs 75%; P<0.001).5 Among HR+/HER2-, lower BRCA1/2 testing rates were observed among patients known not to have a family history of breast or ovarian cancer.5

Table 5-13. BRCA1/2 Testing by HR Status and Known Family History of Breast or Ovarian Cancer

HR+/HER2- HR+/HER2- TNBC with TNBC without with FHx without FHx FHx FHx Patients tested, % (n/N) 69% (97/141) 41% (289/713) 75% (36/48) 69% (146/211) P-value <0.001 0.487 FHx, family history; HER2, human epidermal growth factor 2; HR, hormone receptor; TNBC, triple-negative breast cancer. Source: Lux 20205

Many factors are involved in limiting genetic testing in those advanced breast cancer patients with clinical indications, including patients’ and clinicians’ attitudes about the value of genetic testing and the challenges of integrating genetic testing into the cancer treatment workflow.255

Place of CGP in Breast Cancer Guidelines recommend testing for HER2 at diagnosis and at progression of metastatic sites and for BRCA1/2 mutations in patients with recurrent or metastatic disease. Furthermore, molecular testing for genomic variations have become integral in the management of advanced breast cancer.46,255 For CGP testing specifically in breast cancer, ERBB2 (HER2) and PIK3CA mutations are recommended to be tested via tumor tissue.27 It should be noted that the NCCN Guidelines recommend assessing for PIK3CA mutations in metastatic breast cancer with either tumor tissue or liquid biopsy (for more information on liquid biopsy please refer to the FoundationOne Liquid CDx Label). Table 5-14 reviews the recommended biomarker testing to be conducted in breast cancer and the ability of the FoundationOne CDx Assay to test for these recommended biomarkers.

CONFIDENTIAL PAGE 79 US-FDX-2000073 Last update: January 22, 2021

Table 5-14. Review of Guideline-Recommended Biomarker Testing and FoundationOne CDx Assay Capabilities in Breast Cancer Applicable guidelines for Recommended Biomarker included tumor profiling biomarker testing in FoundationOne CDx assay NCCN Guidelines for Breast ERBB2 (HER2) amplification Yes 27 Cancera (V.6.2020) BRCA1 mutation Yes BRCA2 mutation Yes PIK3CA mutation Yes NTRK fusion Yes MSI-H/dMMR Yes PD-L1 expression Nob HR status (ER or PR) No a Category 1: Based upon high-level evidence, there is uniform NCCN consensus that the intervention is appropriate. Category 2A: Based upon lower-level evidence, there is uniform NCCN consensus that the intervention is appropriate. All NCCN recommendations are category 2A unless otherwise indicated. b PD-L1 testing is available from Foundation Medicine. dMMR, mismatch repair deficient; ER, estrogen receptor; HER2, human epidermal growth factor receptor 2; HR, hormone receptor; NCCN, National Comprehensive Cancer Network; NTRK, neurotrophic receptor tyrosine kinase; PD-L1, programmed death ligand-1; PR, progesterone receptor; TMB, tumor mutational burden.

Clinical Utility and Validity of FoundationOne CDx in Breast Cancer FoundationOne CDx is an in vitro diagnostic device that targets 324 genes. The test also detects the genomic signatures TMB and MSI-H. FoundationOne CDx utilizes DNA isolated from FFPE from the tumor tissue of cancer patients. The test is intended to be used as a companion diagnostic to identify patients with breast cancer who may benefit from treatment with the targeted therapy listed in Table 5-15 in accordance with the approved therapeutic product labeling. Additionally, FoundationOne CDx is intended to provide tumor mutation profiling to be used by qualified healthcare professionals in accordance with professional guidelines in oncology for patients with malignant neoplasms, including breast cancer.1 As described in the Pan-tumor section, other biomarkers (NTRK, TMB) are relevant in all solid tumors.

Table 5-15. FoundationOne CDx Companion Diagnostic Indications Pertinent for Patients With Breast Cancer Tumor type Biomarker(s) detected Therapy Herceptin® (trastuzumab) ERBB2 (HER2) amplification Kadcyla® (ado-trastuzumab-emtansine) Perjeta® (pertuzumab) Breast PIK3CA C420R, E542K, E545A, E545D [1635G>T only], E545G, E545K, Q546E, Piqray® (alpelisib) Q546R, H1047L, H1047R, and H1047Y alterations

CONFIDENTIAL PAGE 80 US-FDX-2000073 Last update: January 22, 2021

ERBB2, erb-b2 receptor tyrosine kinase 2; HER2, human epidermal growth factor receptor 2; PIK3CA, phosphatidylinositol-4,5- bisphosphate 3-kinase catalytic subunit alpha. Source: FoundationOne CDx Label1

ERBB2 (HER2) The clinical validity of using FoundationOne CDx to identify breast cancer patients with ERBB2 (HER2) alterations eligible for treatment with HER-2 directed therapies (ie, trastuzumab, ado-trastuzumab, pertuzumab) was assessed through retrospective samples (n=317) from patients with advanced breast cancer. The ERBB2 amplification testing from FoundationOne CDx assay were compared with the previously approved HER2 FISH PharmDx® Kit (Dako Denmark, A/S).1 . Using the consensus calls between CCD1 and CCD2 as the reference standard, FoundationOne CDx PPA is 89.4%; NPA is 98.4% (Table 5-16).1 . FoundationOne CDx was found to be concordant to the approved HER2 FISH assay (HER2 FISH PharmDx® Kit) for the detection of ERBB2 amplification.1

Table 5-16. Clinical Validity of FoundationOne CDx for Breast Cancer HER2 Amplifications Companion Diagnostic Claims Biomarker PPA, % (n/N) NPA, % (n/N) Comparator assay ERBB2 (HER2) 89.4 (101/113) 98.4 (180/183) Dako HER2 FISH PharmDx™ Kit amplifications FISH, fluorescence in situ hybridization; NPA, negative percent agreement; PPA, positive percent agreement. Source: FoundationOne CDx Label1

PIK3CA The clinical validity of using FoundationOne CDx to identify breast cancer patients harboring PIK3CA alterations eligible for treatment with alpelisib was assessed through retrospective testing of samples collected prior to study treatment from advanced or metastatic breast cancer patients enrolled in the SOLAR-1 clinical trial.1,107 SOLAR-1 was a randomized, double-blind, placebo-controlled phase 3 clinical trial to evaluate the safety and efficacy of alpelisib in combination with fulvestrant for men and postmenopausal women with HR-positive, HER2-negative advanced breast cancer that had progressed on or after aromatase inhibitor treatment.107 In this study, all available samples collected at baseline prior to randomization were tested with FoundationOne CDx. Of the 572 SOLAR-1 randomized patients, 415 had sufficient residual samples available and were tested using FoundationOne CDx. Out of the of these 415 patients 296 were enrolled with CTA1 (119 PIK3CA alteration positive patients and 177 PIK3CA alteration negative patients), and 119 were enrolled with CTA2 (115 PIK3CA alteration positive patients and 4 PIK3CA alteration negative patients).1 . The point estimates of PPA and NPA between FoundationOne CDx and the CTA1 assay were: PPA (95% CI): 93.8% (87.7%, 97.5%); NPA (95% CI): 98.8% (95.6%, 99.8%) (Table 5-17).1 . The point estimates of PPA and NPA between FoundationOne CDx and the CTA2 assay were: PPA (95% CI): 91.6% (87.1%, 95.0%); NPA (95% CI): 98.8% (95.7%, 99.9%) (Table 5-17).1 . The median PFS in the CTA-1 enrolled and CTA-2 enrolled FoundationOne CDx-positive patients were 11.2 months and 10.9 months in the alpelisib + fulvestrant arm, respectively, and

CONFIDENTIAL PAGE 81 US-FDX-2000073 Last update: January 22, 2021

5.5 months and 4.2 months in the placebo treated arm, respectively. In comparison to placebo in the FoundationOne CDx-positive patients, those treated with alpelisib + fulvestrant , there was a lower risk for progression or death vs placebo (CTA-1 enrolled patients: HR=0.52, 95% CI: 0.29, 0.93; CTA-2 enrolled patients: HR=0.35, 95% CI: 0.16, 0.77) (Table 5-17).1 o In the SOLAR-1 clinical trial, alpelisib in combination with fulvestrant demonstrated an estimated 35% risk reduction of disease progression or death compared to the placebo plus fulvestrant arm (HR=0.65; 95% CI: 0.50, 0.85; P=0.00065) in the PIK3CA alteration cohort with a median PFS of 11.0 months in the alpelisib + fulvestrant arm vs 5.7 months in the placebo arm.1

Table 5-17. Clinical Validity and Utility of FoundationOne CDx for Breast Cancer PIK3CA Mutation Companion Diagnostic Claim Concordance Clinical efficacy PFSa, months HRb (95% CI) Drug and PPA, % NPA, % Comparator FoundationOne CTA results alteration (n/N) (n/N) assay CDx results Alpelisib + 93.8 98.8 PCR-based N=56 N=169 fulvestrant PIK3CA hot-spot (106/113) (159/161) 11.2b PIK3CA CTA 1 0.52 (0.29, 0.93) mutations 11.0 91.6 98.8 PCR-based N=42 0.65 (0.50, 0.85) (197/215) (162/164) PIK3CA hot-spot c CTA 2 10.9 0.35 (0.16, 0.77)

a PFS by investigator assessment in patients with PIK3CA alteration positive tumors. b The HR shown here for both the FoundationOne CDx results and the CTA results is for alpelisib + fulvestrant for risk of disease progression or death compared to placebo in the PIK3CA alteration-positive population. c PFS in the CTA1-enrolled positive and FoundationOne CDx-positive patients. d PFS in the CTA2-enrolled positive and FoundationOne CDx-positive patients. CI, confidence interval; CTA, clinical trial assay; HR, hazard ratio; NPA, negative percent agreement; PCR, polymerase chain reaction; PFS, progression-free survival; PPA, positive percent agreement. Source: FoundationOne CDx Label1

CONFIDENTIAL PAGE 82 US-FDX-2000073 Last update: January 22, 2021

CGP in CRC

. Colon and rectal cancers accounts for an estimated 8.2% of new cancer diagnoses and is the second leading cause of cancer-related death in the United States.256 Survival for metastatic colorectal cancer (mCRC) at 5 years is 14.8%, with approximately one-fifth of the patients (20.5%) diagnosed at this stage.256 . Historically, prior to the advent of biologics and targeted therapies, chemotherapy alone resulted in a median overall survival of ~21 months.257 . However, substantial advances in individualizing treatment strategies for patients with mCRC, based on the presence of actionable genomic alterations or the absence of mutations that render specific therapies as ineffective, have improved survival in sub-sets of patients ≥30 months.153,258 o For example, when the oncogene KRAS is mutated to become constitutionally active (in exon 2, codons 12 and 13), downstream signaling pathways were activated that could bypass EGFR signaling, leading to ineffective anti-EGFR therapy (ie, cetuximab and panitumumab).259 . There are several actionable biomarkers in mCRC, with approximately 50% of patients harboring a mutation in either KRAS or NRAS rendering them ineligible to receive anti-EGFR therapy (Figure 5-7).153 Figure 5-7. Prevalence of Actionable Genomic Alterations in mCRC Patients

KRAS 45% NRAS 6% BRAF 9% MSI 5% HER2 5% NTRK <1%

0 10 20 30 40 50

Source: Cohen 2020153 . NCCN Guidelines that all mCRC patients should undergo tumor tissue genotyping for RAS (KRAS and NRAS), BRAF mutations, and MSI (or universal mismatch repair [ie, MMR]). Testing for HER2 amplifications is also recommended, except in cases where there is a known RAS/RAF mutation. Further, NCCN Guidelines state that for patients with mCRC who require testing for all of the aforementioned alterations, a validated NGS panel may be especially useful.30,39 . FoundationOne CDx is an FDA-approved companion diagnostic for detection of KRAS and NRAS wild type patients for selection of therapy with an anti-EGFR therapy (cetuximab or panitumumab).1

Unmet Need for Molecular Testing in CRC The last decade has seen substantial progress in both the molecular understanding of mCRC and of the molecular testing approaches.260 The current standard of care biomarkers for management of a patient

CONFIDENTIAL PAGE 83 US-FDX-2000073 Last update: January 22, 2021

with mCRC include extended RAS (KRAS and NRAS), BRAF V600E mutation, MMR or MSI, HER2 amplification; additionally, and although rare, actionable mutations also exist for NTRK fusions.260 . Evaluation of mutations in the oncogenes, KRAS, NRAS are required for first-line treatment decisions, as both randomized trials and meta-analysis have highlighted the resistance to EGFR monoclonal antibodies (mAbs – cetuximab and panitumumab) in the metastatic setting for those with KRAS/NRAS mutations.261-263 Conversely, the benefit of both panitumumab and cetuximab in patients that are RAS wild-type has been reported in numerous clinical trials, and is summarized below in Figure 5-8 utilizing results from a meta-analysis of these trials.263

Figure 5-8. Anti-EGFR Treatment Benefit for Tumors Without any RAS Mutations (All RAS Wild Type)

CI, confidence interval; PFS, progression-free survival; RAS, rat sarcoma virus gene family (includes KRAS and NRAS). Source: Sorich 2015.263

. Further, BRAF mutational analysis is also needed at first-line decision-making, as the addition of encorafenib (a BRAF-inhibitor) to cetuximab was associated with a significant survival benefit in a recent phase III trial in patients presenting a BRAF V600E mutation.264 . Alterations in mismatch repair proteins (MLH1, MSH2, MSH6 and PMS2) or for MSI-H tumors is an area of clinical significance in mCRC because it predicts the efficacy of the immunotherapy agents, pembrolizumab and nivolumab in this setting.265,266 Finally, mCRC with HER2 amplifications and/or overexpression revealed significant responses with dual HER2 therapy in prospective studies.267,268 . Finally, TRK inhibitors showed high efficacy in pan-tumor trials in NTRK fusion-positive tumors (Place of CGP in the Pan-Tumor Evaluation of TMB, MSI-H, and NTRK). Despite the improvement of outcomes in mCRC with therapy guided by molecular testing, many patients who should be tested pursuant to guidance by the NCCN go untested; and while there is a paucity of data for testing utilizing a CGP test, data for RAS mutations alone reveal: . In a ten-state-population-based cancer registry study of 3608 patients diagnosed with mCRC in 2011, only 27% had a documented RAS test.154

CONFIDENTIAL PAGE 84 US-FDX-2000073 Last update: January 22, 2021

. In a retrospective analysis of adult patients with mCRC engaged in routine care across 12 community oncology sites, of the 1,363 patients included in the study with both evidence of metastatic disease and with no other primary cancer and with at least 2 visits to their oncologist, less than half (47.5%) were tested for KRAS.155

Place of CGP in CRC Given the number of required genomic alterations needed for the management of mCRC and the ability to detect other infrequent alterations, approaches using broad panel or CGP represent the most complete method for determining genomic alterations in mCRC.260 The NCCN Guidelines for Colon Cancer and for Rectal Cancer do not currently specify where CGP should be considered in this patient population; however, they do state that specific to capturing MSI/dMMR and RAS/BRAF mutations for patients with mCRC, a validated NGS panel may be especially useful.30,39 Table 5-18 reviews the recommended biomarker testing to be conducted in mCRC, as well as the ability of the FoundationOne Tissue CDx assay to test for these recommended biomarkers.

Table 5-18. Review of Guideline-Recommended Biomarker Testing for Patients With Metastatic Colon and Rectal Cancers and FoundationOne Tissue CDx Assay Capabilities Biomarker included in Applicable guidelines for Recommended FoundationOne Tissue CDx tumor profiling biomarker testing assay NCCN Guidelines for Colon RAS (KRAS and NRAS) and BRAF Yes Cancer V.1.202130 mutationsa NCCN Guidelines for Rectal dMMR (ie, MLH1, MSH2, MSH6, Yes 39 Cancer V.1.2021 and PMS2) or MSI-Hb HER2c Yes a Tumor testing is recommended in patients with metastatic CRC cancer. b Recommended in all newly diagnosed patients with CRC. c Anti-HER2 therapy is only indicated in HER2-amplified tumors that are also RAS and BRAF wild-type. BRAF, serine/threonine-protein kinase B-raf; CRC, colorectal; dMMR, mismatch repair deficiency resulting from a mutation or inactivation in the MMR genes (MLH1, PMS2, MSH2, MSH6); HER2, human epidermal growth factor receptor 2 MSI-H, microsatellite instability–high; NCCN, National Comprehensive Cancer Network; NTRK, neurotrophic receptor tyrosine kinase; RAS, rat sarcoma virus gene family (includes KRAS and NRAS).

Clinical Utility and Validity of FoundationOne CDx in CRC FoundationOne CDx is an in vitro diagnostic device that targets 324 genes. The test also detects the genomic signatures TMB and MSI-H. FoundationOne CDx utilizes DNA isolated from FFPE from the tumor tissue of cancer patients. The test is intended to be used as a companion diagnostic to identify patients with CRC who may benefit from treatment with the targeted therapies listed in Table 5-19 in accordance with the approved therapeutic product labeling. Additionally, FoundationOne CDx is intended to provide tumor mutation profiling to be used by qualified healthcare professionals in accordance with professional guidelines in oncology for patients with malignant neoplasms, including CRC.1 As described in the Pan-tumor section, other biomarkers (NTRK, TMB) are relevant in all solid tumors.

CONFIDENTIAL PAGE 85 US-FDX-2000073 Last update: January 22, 2021

Table 5-19. Companion Diagnostic Indications Pertinent for Patients With Colorectal Cancer Tumor type Biomarker(s) detected Therapy KRAS wild-type (absence of mutations in codons 12 and 13) Erbitux® (cetuximab) CRC KRAS wild-type (absence of mutations in exons 2, 3, and 4) Vectibix® (panitumumab) NRAS wild type (absence of mutations in exons 2, 3, and 4) CRC, colorectal cancer; KRAS, Kirsten rat sarcoma viral oncogene homolog; NRAS, neuroblastoma RAS viral (v-ras) oncogene homolog. Source: FoundationOne CDx Label1

KRAS/NRAS Clinical validity of the FoundationOne CDx assay was evaluated as a companion diagnostic in identification of patients with colorectal cancer who may not benefit from EGFR inhibitor treatments (eg, cetuximab or panitumumab) due to KRAS alterations. Retrospective samples (n=342) from advanced front-line or later-line CRC patients were included in this study, which were tested for KRAS alterations by the FoundationOne CDx assay and the previously approved therascreen® KRAS RGQ PCR Kit (QIAGEN). Samples were tested across 2 replicates of therascreen® KRAS assay (denoted as CCD1 and CCD2) and 1 replicate by FoundationOne CDx.1 . The PPA and NPA between the combination of therascreen® replicates (CCD1 and CCD2) and FoundationOne CDx assays were as follows: PPA: 100%; NPA: 100% (Table 5-20).1 . Based on these results, FoundationOne CDx has been demonstrated to be concordant to the therascreen® assay for the detection of KRAS mutations (Table 5-20).1

Table 5-20. Clinical Validity of FoundationOne CDx for CRC Companion Diagnostic Claims Biomarker PPA, % (n/N) NPA, % (n/N) Comparator assay KRAS 100% (173/173) 100% (154/154) therascreen® KRAS RGQ PCR Kit NPA, negative percent agreement; PCR, polymerase chain reaction; PPA, positive percent agreement. Source: FoundationOne CDx Label1

CONFIDENTIAL PAGE 86 US-FDX-2000073 Last update: January 22, 2021

CGP in Ovarian Cancer

. Although ovarian cancer is a rare cancer type among women (17th most common cancer diagnosis with an estimated 21,750 diagnosed in 2020 in the US), it contributes significantly to cancer-related deaths each year, as it is the fifth most common cause of cancer-related death.247,269 . With chemotherapy-based treatment, the vast majority (>70%) of ovarian cancer patients will recur within 2 years.270-273 As patients with ovarian cancer become platinum-resistant or refractory, the median survival becomes poor, ranging from 9 to 18 months.274,275 . Actionable alterations in patients with advanced ovarian cancer include BRCA1/2 mutations, HRD, and MSI-H/dMMR.36 o Therapies targeting these alterations in patients with ovarian cancer have shown improved outcomes in women with advanced or metastatic ovarian cancer.276-279,280,281 . In advanced ovarian cancer, the NCCN Guidelines currently recommend patients have access to testing for BRCA1/2 mutations, MSI-H, and dMMR.36 . FoundationOne CDx is an FDA-approved companion diagnostic for the detection of BRCA1/2 mutations for treatment with olaparib or rucaparib.1

Unmet Need for Molecular Testing in Ovarian Cancer Genomic testing has been an important part of the management of advanced ovarian cancer since BRCA1 and BRCA2 were identified 25 years ago.255 Approximately 15% to 20% of all patients with ovarian cancer harbor a BRCA mutation.282 Women with BRCA1 and BRCA2 mutations have an increased risk of 39% to 46% and 10% to 27%, respectively, of developing ovarian, fallopian tube, or peritoneal cancer by the age of 70 years.283 Further, HRD tumors, which include BRCA-mutated tumors, may serve as a marker for platinum sensitivity.46 Other HRD-related genes, in addition to BRCA1/2 mutations, include PALB2, BARD1, BRIP1, RAD51B, RAD51C, RAD51D, ATM, FAAP20, CHEK2, FAN1, FANCE, FANCM, and POLQ.255,284 Identifying mutation status may influence treatment recommendations; for example, certain targeted therapies such as PARP inhibitors have been developed specifically for the treatment of ovarian cancer with BRCA1 or BRCA2 mutations. Genomic alternations for which targeted therapies are currently available are summarized in Table 5-21.

Table 5-21. Genomic Alterations With Targeted Therapies Available Molecular Alteration Prevalencea Drug Class BRCA1/BRCA2 15%–20% PARP inhibitor HRD 40%–50%b PARP inhibitorc NTRK <1% TRK inhibitor MSI-H/dMMR <1% PD-1 mAbd BRCA, breast cancer susceptibility gene; dMMR, mismatch repair deficiency; HRD, homologous recombination deficiency; mAb, monoclonal antibody; NTRK, neurotrophic tyrosine receptor kinase; PARP, poly (ADP-ribose) polymerase; PD-1, programmed death-1; TRK, tropomyosin receptor kinase. a The prevalence estimates for molecular alterations are based on data primarily obtained in patients with epithelial ovarian cancer and in particular those with high-grade serous ovarian cancer.

CONFIDENTIAL PAGE 87 US-FDX-2000073 Last update: January 22, 2021

b HRD includes BRCA1/BRCA2 mutations, as these are HRD genetic alterations. c Niraparib is recommended in patients with cancer associated with HRD, defined as either a deleterious or suspected deleterious BRCA mutation or genomic instability and progression >6 months after response to last platinum-based chemotherapy. d Pembrolizumab. Source. Konstantinopoulos et al. 2015.150 Bonadio et al. 2018.151 Gee et al. 2018.152

Therapies directed at molecular alterations have improved outcomes in patients with advanced ovarian cancer. . For advanced ovarian cancer patients with BRCA1/2 mutations, PARP inhibitors significantly improved PFS by 11−16 months (vs placebo) as a maintenance therapy following a response to platinum-based chemotherapy.276278280 . Patients with HRD-positive advanced ovarian cancer had a ORR of 28% and a duration of response of 9.6 months with treated with a PARP inhibitor.279 . Advanced ovarian cancer patients with MSI-H tumors treated with a PD-1 inhibitor had an ORR of 33.3%.281 As targeted therapies have improved outcomes of patients with advanced ovarian cancer, major medical societies recommend that women with ovarian cancer undergo genetic testing, regardless of family history. . In women with pathologically confirmed advanced epithelial ovarian cancer, the NCCN Guidelines currently recommend patients have germline and somatic genetic testing.36 In the absence of a BRCA1/2 mutation, HRD status may provide information on the magnitude of benefit of PARP inhibitor therapy (category 2B).36 For patients with recurrent disease, tumor molecular testing is recommended before initiation of therapy; if not previously done, the recommended testing should include BRCA, and MSI or dMMR.36 NCCN also recommends that HRD and additional somatic testing can be considered.36 . SGO also recommends that all women, regardless of age, with ovarian, tubal, and peritoneal cancer undergo genetic testing even if they do not have a family history of the disease.283 SGO does not make a recommendation on the use of HRD and dMMR testing.283 Although genetic testing in all women with ovarian cancer is recommended by most guidelines, as previously described, only approximately 10% to 30% of women with ovarian cancer undergo genetic testing.255,285 Many factors have been cited as contributing to these low rates, including lack of access to testing, lack of physician knowledge, patient lack of knowledge about the value of testing, and/or anxiety around undergoing genetic testing.286 Moreover, undertesting in ovarian cancer may reflect a relatively low public awareness of and advocacy for ovarian cancer.255

Place of CGP in Ovarian Cancer For patients with ovarian cancer, regardless of stage at diagnosis, surgery is typically recommended as the first treatment option.36 Based on this treatment pattern, most patients have tumor tissue available or banked for molecular testing. Currently, clinical guidelines from NCCN, SGO, and ASCO recommend using tumor tissue for genetic testing of patients with ovarian cancer.36,283,287 Table 5-22 reviews the recommended biomarker testing to be conducted in ovarian cancer and the ability of the FoundationOne CDx Assay to test for these recommended biomarkers.

CONFIDENTIAL PAGE 88 US-FDX-2000073 Last update: January 22, 2021

Table 5-22. Review of Guideline-Recommended Biomarker Testing and FoundationOne CDx Assay Capabilities in Ovarian Cancer Applicable guidelines Recommended Biomarker included for tumor profiling biomarker testing in FoundationOne CDx assay

NCCN Guidelines for BRCA1/2 variants (for histologically confirmed disease) Yes Ovarian Cancer V.1.202036 dMMR/MSI-H (if recurrent or refractory disease) Yes

ASCO287 BRCA1/2 Yes dMMRa Yes a Clear cell, endometrioid, or mucinous ovarian cancer be offered somatic tumor testing for dMMR. BRCA, breast cancer gene; dMMR, DNA mismatch repair; HRD, homologous recombination deficiency; MSI-H, microsatellite instability-high; NCCN, National Comprehensive Cancer Network.

Clinical Utility and Validity of FoundationOne CDx in Ovarian Cancer FoundationOne CDx is an in vitro diagnostic device that targets 324 genes. The test also detects the genomic signatures TMB and MSI-H. FoundationOne CDx utilizes DNA isolated from FFPE from the tumor tissue of cancer patients. The test is intended to be used as a companion diagnostic to identify patients with ovarian cancer who may benefit from treatment with the targeted therapies listed in Table 5-23 in accordance with the approved therapeutic product labeling. Additionally, FoundationOne CDx is intended to provide tumor mutation profiling to be used by qualified healthcare professionals in accordance with professional guidelines in oncology for patients with malignant neoplasms, including ovarian cancer.1 As described in the Pan-tumor section, other biomarkers (NTRK, TMB) are relevant in all solid tumors.

Table 5-23. Companion Diagnostic Indications Pertinent for Patients With Ovarian Cancer Tumor type Biomarker(s) detected Therapy Lynparza® (olaparib) Ovarian BRCA1 and BRCA2 alterations Rubraca® (rucaparib) BRCA, breast cancer susceptibility gene. Source: FoundationOne CDx Label1

BRCA1/BRCA2 FoundationOne CDx is an FDA-approved companion diagnostic to determine treatment with olaparib or rucaparib in patients with ovarian cancer.1 FoundationOne CDx demonstrated equivalence to FoundationFocus CDxBRCA, the original companion diagnostic for detection of BRCA1/2 mutations in formalin-fixed paraffin-embedded (FFPE) ovarian cancer tumor specimens to select for rucaparib eligibility.1,108 FoundationOne CDx replaced FoundationFocus in the market in 2018. For olaparib in ovarian cancer, the FoundationOne CDx CTA was used to identify somatic BRCA1/2 on prospectively collected tumor samples for patients enrolled in the clinical trial (SOLO-1) based on local (germline or somatic) or central (germline) testing.1,109 Both olaparib and rucaparib showed improved clinical

CONFIDENTIAL PAGE 89 US-FDX-2000073 Last update: January 22, 2021

outcomes compared with placebo in those patients with BRCA1/2 mutations as determined by FoundationOne CDx.1

The clinical performance of FoundationOne CDx for BRCA1/2 classification was established based on available tumor analysis using the FoundationOne CDx in the clinical study D0818C00001 (SOLO1).109 SOLO1 was a Phase III, randomized, double-blind, placebo-controlled, multicenter trial, that compared the efficacy of olaparib with placebo in patients with advanced ovarian, fallopian tube, or primary peritoneal cancer with BRCA mutation (documented mutation in BRCA1 or BRCA2) following first-line platinum-based chemotherapy.109 A total of 391 patients were randomized (2:1) to receive olaparib tablets 300 mg orally twice daily (n=260) or placebo (n=131). Patients were required to have a documented mutation in BRCA1 or BRCA2 that were known or predicted to be a loss of function mutation.1 . The median PFS in both the FoundationOne CDx-positive patients and the full analysis set from the clinical trial based on the full analysis set were not reached at the time of analysis for patients treated with olaparib, as compared with 11.9 months and 13.8 months for placebo-treated patients, respectively. The percent of progression events was similar between the full analysis set and the FoundationOne CDx determined set (39% in each group). When comparing olaparib to placebo for progression or death in this patient population, the HR was similar for those with BRCA mutation status as determined in the full analysis set (HR: 0.3; 95% CI; 0.23, 0.41) and those determined by FoundationOne CDx (HR: 0.28; 95% CI: 0.20, 0.38) (Table 5-24).1 . In the SOLO-1 trial, after a median follow-up of 4.8 years in patients treated with maintenance olaparib and 5 years in placebo-treated patients, the median PFS was 56.0 months vs 13.8 months for patients treated with olaparib vs placebo, respectively (HR: 0.33; 95% CI: 0.25, 0.43).110

FoundationOne CDx and FoundationFocus CDxBRCA LOH assays are equivalent with the exception of an updated analysis pipeline in use for FoundationOne CDx and reporting software that allow for comprehensive reporting of all relevant alterations detected by the FoundationOne CDx platform. Comprehensive validation of the analysis pipeline which included robust regression testing and reanalysis of FoundationFocus CDx BRCA LOH clinical bridging sample data was performed. The assays were determined to be concordant for determining HRD status for patients who may benefit from rucaparib treatment.1 . The median PFS of patients treated with rucaparib who were determined to be BRCA1/2-positive by FoundationOne CDx was 16.6 months as compared with 5.4 months for patients treated with placebo (HR: 0.24; 95% CI: 0.16-0.36) (Table 5-24).1

Table 5-24. Clinical Utility of FoundationOne CDx for Ovarian Cancer Companion Diagnostic Claims Drug and alteration Clinical endpoint FoundationOne CDx Full analysis set results results Olaparib PFSa, months N=206 N=260 b BRCA1/2 HR (95% CI) Not reached Not reached 0.28 (0.20, 0.38) 0.30 (0.23, 0.41)

CONFIDENTIAL PAGE 90 US-FDX-2000073 Last update: January 22, 2021

Rucaparib PFSc, months N=124 NR b BRCA1/2 HR (95% CI) 16.6 NR 0.24 (0.16, 0.36) a Investigator assessed median PFS evaluated according to RECIST v1.1. a HR compares olaparib or rucaparib to placebo. c Investigator assessed median PFS. CI, confidence interval; HR, hazard ratio; NR, not reported; PFS, progression-free survival; RECIST, Response Evaluation Criteria in Solid Tumors. Source: FoundationOne CDx Label1

LOH HRD is the state that results in abnormal functioning of the homologous recombination repair pathway, a DNA repair mechanism.46 This state may be caused by the accumulation of DNA errors, which can be detected by measuring genomic LOH.288 Identifying patients’ HRD status may help to identify patients who are more likely to benefit from agents that target the homologous recombination pathway, such as poly (ADP-ribose) polymerase (PARP) inhibitors. FoundationOne CDx also received FDA approval for detecting HRD status in ovarian cancer patients, where testing positive for HRD is defined as tBRCA- positive and/or LOH high. The approval of FoundationOne CDx is based on the approval of FoundationFocus CDx; FoundationOne CDx and FoundationFocus CDx assays are equivalent with the exception of an updated analysis pipeline in use for FoundationOne CDx and reporting software that allow for comprehensive reporting of all relevant alterations detected by the FoundationOne CDx platform.1 . In a study of 564 patients with high-grade, platinum-sensitive ovarian cancer, patients with HRD tumors (defined as having BRCA mutations and/or genomic LOH >16% as assessed by NGS testing at Foundation Medicine, using FoundationFocus CDx [somatic] and Myriad Genetics [germline]) experienced improved median PFS from treatment with the PARP inhibitor rucaparib (13.6 months; 95% CI: 10.9–16.2) compared with placebo (5.4 months; 95% CI: 5.1–5.6), with a HR of 0.32 (95% CI: 0.25–0.42; P<0.0001).289 Reanalysis of the clinical efficacy data demonstrated that FoundationOne CDx and FoundationFocus CDx have similar performance in identifying HRD+ patients who may benefit from rucaparib treatment.1 . In an earlier study in platinum-sensitive, BRCA-wildtype ovarian cancer, rucaparib in the treatment setting elicited improved median PFS (7.2 vs 5.0 months; HR: 0.51) and objective RR (33.3% vs 9.6%; P=0.0003) in patients with a high genomic LOH score (≥16%, as assessed by NGS testing) compared with patients with a low genomic LOH score.108,290

CONFIDENTIAL PAGE 91 US-FDX-2000073 Last update: January 22, 2021

CGP in Prostate Cancer

. Prostate cancer is the most commonly diagnosed cancer in men, accounting for an estimated 20% of new cancer diagnoses and up to 10% of cancer-related deaths in the US.247 . Around 5% of patients present with distant or metastatic disease at diagnosis; additionally, up to 20% of patients with localized disease at diagnosis will progress to castrate-resistant prostate cancer (CRPC), with the majority of these patients having metastatic disease within 5 years of diagnosis.4,247,291 291 o Survival from diagnosis of CRPC is 14 months. . Germline and somatic mutations of genes involved in DNA damage repair (DDR) pathways occur in 15% to 30% of patients with metastatic prostate cancer; up to 27% may be due to alterations, specifically, in BRCA, ATM, or CHEK2.145,292 o More recent trials of the efficacy of agents targeted against alterations in the DDR pathway vs standard of care in patients with metastatic castration-resistant prostate cancer (mCRPC) have shown improved radiographic PFS, with a favorable trend for OS.293 . The NCCN Guidelines for Prostate Cancer recommend tumor testing for HRR gene mutations in all men diagnosed with metastatic prostate cancer, including BRCA1, BRCA2, ATM, PALB2, FANCA, RAD51D, CHEK2, CDK12, and consider testing in men diagnosed with regional prostate cancer. Somatic testing may require repetition when prostate cancer progresses after treatment. Tumor testing for MSI or dMMR is recommended for all men with metastatic CRPC and can be considered for men with castrate-naïve metastatic or regional prostate cancer.38 . FoundationOne CDx is an FDA-approved companion diagnostic for the detection of HRR (BRCA1, BRCA2, ATM, BARD1, BRIP1, CDK12, CHEK1, CHEK2, FANCL, PALB2, RAD51B, RAD51C, RAD51D and RAD54L) alterations for treatment with olaparib in patients with prostate cancer.1

Unmet Need for Molecular Testing in Prostate Cancer Despite a decreased trend in mortality seen for patients diagnosed with prostate cancer, mCRPC remains a treatment challenge, with 5-year survival rates of approximately 31% and a median OS of less than 2 years with life-prolonging therapy.247,294 More recently, different DDR pathways have been recognized to be frequently altered in the advanced stages of prostate cancer, with mutations in the DDR genes detected in 15% to 30% of patients with mCRPC (Table 5-25); these can be both inherited (ie, germline) or acquired (ie, somatic).144,145 The DDR pathways have different constituent genes that, when mutated, contribute to a deficiency in the DNA repair capability. For example, DNA double-strand breaks are typically repaired by the HRR gene; HRR genes include recombination DNA repair genes such as BRCA1/2, ATM, CHEK1, CHEK2, WES, BARD1, BRIP1, FAM175A, MRE11A, NBN, PALB2, RAD51C, and RAD51D.145,295 The MMR pathway is charged with repairing DNA bases that are mispaired during DNA replication; the majority of genes involved in this pathway belong to the MSH and MLH family, including several sensor molecules such as MSH2 and MSH6.144

CONFIDENTIAL PAGE 92 US-FDX-2000073 Last update: January 22, 2021

Table 5-25. DDR Genes Frequently Altered in Prostate Cancer DDR gene Common aberration Tumor site References types Primary Metastatic BRCA2 Deletion, mutation 3% 13.3% Abeshouse A, et al. Cell. 2015;163:1011-1125. Robinson D, et al. Cell. 2015;161:1215-1228. ATM Deletion, mutation 4% 7.3% Wu YM, et al. Cell. 2018;173:1770-1782.e14. Shelley MD, et al. Evidence-Based Urol. 2010:293-303. CHEK2 Germline mutation 0% 1.87% Pritchard CC, et al. N Engl J Med. 2016;375:443-453. CDK12 Mutation 1% 6.9% Wu YM, et al. Cell. 2018;173:1770-1782.e14. BRCA1 Mutation 1% 0.7% Robinson D, et al. Cell. 2015;161:1215-1228. FANCD2 Copy loss 6% – Robinson D, et al. Cell. 2015;161:1215-1228. RAD51C Copy loss 3% – Robinson D, et al. Cell. 2015;161:1215-1228. RAD51D Germline mutation – 0.43% Pritchard CC, et al. N Engl J Med. 2016;375: 443-453. MSH2 Copy loss, mutation, 0.3% 2% Pritchard CC, et al. rearrangements Nat Commun. 2014;5:1-6. MSH6 Mutation 1.5% 2% Pritchard CC, et al. Nat Commun. 2014;5:1-6. Antonarakis ES, et al. Eur Urol. 2018:1-5. MLH1 Copy loss, epigenetic 0.3% 0.7% Robinson D, et al. silencing Cell. 2015;161:1215-1228. ATM, ataxia telangiectasia mutated; BRCA, breast cancer susceptibility gene; CHEK2, checkpoint kinase 2; CKD12, cyclin- dependent kinase 12; FANCD2, Fanconi anemia group D2; MLH1, MutL homolog 1; MSH, mismatch repair protein involved in the DNA mismatch repair system; RAD, genes that encode for members of the RAD51 protein family that are known to be involved in homologous recombination and repair of DNA. Source: Athie 2018144; references cited are as cited by this publication.

Inhibition of PARP is currently an active area of investigation in the development of new agents for mCRPC; PARP is involved in the repair of single-stranded DNA breaks, and its inhibition is circumvented by proteins of the HRR pathway.295 In patients with mutations of the genes involved in the HRR pathway, PARP may effectively block DNA repair, leading to cell death.295 PARP inhibitors have shown improvement in outcomes in patients with mCRPC with genomic alterations in the HRR

CONFIDENTIAL PAGE 93 US-FDX-2000073 Last update: January 22, 2021

pathway.144,295 Further, MSI-H/dMMR are uncommon yet therapeutically meaningful in patients with metastatic prostate cancer.296 . In a phase 3 trial evaluating the efficacy of olaparib vs physician choice of a standard of care hormonal agent (pcHA) (ie, enzalutamide or abiraterone) in patients with mCRPC and progression on a previous treatment with enzalutamide or abiraterone and with alterations in any of 15 predefined genes involved in the HRR pathway, olaparib improved radiographic PFS by approximately 4 months, with a favorable trend for OS despite >80% crossover to olaparib.293 . In another study, evaluating rucaparib in the PARP inhibitor rucaparib in patients who have progressed on an androgen receptor (AR)-directed therapy and chemotherapy who harbor an alteration BRCA1, BRCA2, ATM, or other prespecified DDR gene, revealed an ORR of 43.9% (95% CI: 30.7, 57.6) and a median time to prostate-specific antigen progression of 6.5 months (95% CI: 5.7, 7.5) in patients with a BRCA1 or BRCA2 mutation.297 . In a small study of mCRPC patients who were MSI-H and treated with immunotherapy, 45% of patients had a durable clinical benefit, which the authors concluded was in line with other neoplasms with this same genomic alteration.296 Despite the improved outcomes with targeted therapies in men with mCRPC, many patients who should be tested pursuant to guidance by the NCCN go untested; and while there is a paucity of data for testing unselected prostate cancer patients, data for germline testing show that guideline adherence is problematic. . A survey conducted by the Germline Genetics Working Group of the Prostate Cancer Clinical Trials Consortium (PCCTC) that was administered to medical oncologists who see patients with prostate cancer (n=26) from PCCTC affiliate sites revealed that 62% of oncologists surveyed would consider germline genomic testing in all metastatic prostate cancer patients; the remainder would only consider testing for patients with a family history and/or for clinical trial eligibility (27%) or for patients with a family history of genomic testing (12%).298

Place of CGP in Prostate Cancer The NCCN Guidelines for Prostate Cancer recommend molecular testing in all men with metastatic prostate cancer.38 Table 5-26 reviews the recommended biomarker testing to be conducted in prostate cancer, as well as the ability of the FoundationOne CDx assay to test for these recommended biomarkers.

Table 5-26. Review of Guideline-Recommended Biomarker Testing for Patients With Advanced (Regional and Metastatic) Prostate Cancer and FoundationOne CDx Assay Capabilities Applicable guidelines for Recommended Biomarker included in tumor profiling biomarker testing FoundationOne CDx assay NCCN Guidelines for Prostate Homologous recombination gene Yes Cancer V.3.202038a mutationsb (BRCA1, BRCA2, ATM, PALB2, FANCA, RAD51D, CDK12, CHEK2) dMMR (ie, MLH1, MSH2, MSH6, Yes and PMS2) or MSI-Hc a Category 2A: Based upon lower-level evidence, there is uniform NCCN consensus that the intervention is appropriate.

CONFIDENTIAL PAGE 94 US-FDX-2000073 Last update: January 22, 2021

All NCCN recommendations are category 2A unless otherwise indicated. b Tumor testing is recommended in patients with metastatic prostate cancer and can be considered in patients with regional prostate cancer. c Tumor testing can be considered in patients with regional or castration-naive metastatic prostate cancer and is recommended in patients with mCRPC. BRCA, breast cancer susceptibility gene; CHEK2, checkpoint kinase 2; CKD12, cyclin-dependent kinase 12; dMMR, mismatch repair deficient; MLH1, MutL homolog 1; MSH, mismatch repair protein involved in the DNA mismatch repair system; MSI-H, microsatellite instability–high; NCCN, National Comprehensive Cancer Network; RAD, genes that encode for members of the RAD51 protein family that are known to be involved in homologous recombination and repair of DNA.

Clinical Utility and Validity of FoundationOne CDx in Prostate Cancer FoundationOne CDx is an in vitro diagnostic device that targets 324 genes. The test also detects the genomic signatures TMB and MSI-H. FoundationOne CDx utilizes DNA isolated from FFPE from the tumor tissue of cancer patients. The test is intended to be used as a companion diagnostic to identify patients with prostate cancer who may benefit from treatment with the targeted therapies listed in Table 5-27 in accordance with the approved therapeutic product labeling. Additionally, FoundationOne CDx is intended to provide tumor mutation profiling to be used by qualified healthcare professionals in accordance with professional guidelines in oncology for patients with malignant neoplasms, including prostate cancer.1 As described in the Pan-tumor section, other biomarkers (NTRK, TMB) are relevant in all solid tumors.

Table 5-27. Companion Diagnostic Indications Pertinent for Patients With Prostate Cancer Tumor type Biomarker(s) detected Therapy HRR (BRCA1, BRCA2, ATM, BARD1, BRIP1, CDK12, Lynparza® (olaparib) Prostate cancer CHEK1, CHEK2, FANCL, PALB2, RAD51B, RAD51C, RAD51D and RAD54L) alterations BRCA, breast cancer susceptibility gene. Source: FoundationOne CDx Label1

HRR FoundationOne CDx is an FDA-approved companion diagnostic to determine treatment with olaparib in patients with prostate cancer. FoundationOne CDx was the registrational assay used in the PROfound trial.111 PROfound was a Phase III, randomised, open-label, multicentre trial to assess the efficacy and safety of olaparib monotherapy in patients with metastatic castration-resistant prostate cancer (mCRPC) that have qualifying HRR gene mutations that were predicted to be deleterious or suspected deleterious (known or predicted to be detrimental/lead to loss of function) who have failed prior treatment with a new hormonal agent (NHA) (investigators choice of NHA with either enzalutamide 160 mg orally once daily or abiraterone acetate 1000 mg orally once daily with prednisone 5 mg orally twice daily [prednisolone was permitted for use instead of prednisone, if necessary]).111 Eligible patients were those with HRRm mCRPC, who had progressed following prior treatment with an NHA.111 All patients must have had a qualifying HRR mutation assessed via the Foundation Medicine CLIA HRR CTA to be randomised.111 Qualifying HRR gene mutations were BRCA1, BRCA2 and ATM for Cohort A; and BARD1, BRIP1,

CONFIDENTIAL PAGE 95 US-FDX-2000073 Last update: January 22, 2021

CDK12, CHEK1, CHEK2, FANCL, PALB2, PPP2R2A, RAD51B, RAD51C, RAD51D and RAD54L for Cohort B.1,111 . Within cohort A, the ORR for the full analysis set was similar to that of the FoundationOne CDx determined HRRm patients (33.3% vs 33.8%) for olaparib treated patients. Further, the median radiological PFS (rPFS) was similar within this cohort for the full analysis (7.4 months; 95% CI: 6.24, 9.33) set and those as determined by FoundationOne CDx (7.4 months; 95% CI: 6.87, 9.33) for patients treated with olaparib, as compared to 3.6 months for those treated with NHA in both groups. The HR for the comparison of radiological progression or death for olaparib to placebo was also similar for the full analysis set (HR: 0.34; 95% CI: 0.25, 0.47) and those determined by FoundationOne CDx (HR: 0.33; 95% CI: 0.24, 0.46).1 . There was a statistically significant improvement in rPFS as assessed by the blinded independent review committee for olaparib-treated patients compared with investigators choice of NHA- treated patients in Cohort A+B, with a 51% reduction in the risk of radiological disease progression or death and a prolongation of median progression-free interval of 2.3 months with olaparib vs investigator’s choice of NHA (HR=0.49; 95% CI 0.38, 0.63; P<0.0001; median rPFS 5.8 months for the full analysis set and 6.2 months for the confirmed FoundationOne CDx subgroup vs 3.5 months for investigator’s choice of NHA) (Table 5-28).1 . In the PROfound study, the median OS for olaparib was 17.4 months vs 13.6 months for placebo in patients with HRR mutations (BRCA1, BRCA2, ATM, BARD1, BRIP1, CDK12, CHEK1, CHEK2, FANCL, PALB2, RAD51B, RAD51C, RAD51D and RAD54L) (HR: 0.76, 95% CI: 0.58, 1.00).111

Table 5-28. Clinical Utility of FoundationOne CDx for Prostate Cancer Companion Diagnostic Claims Drug and alteration Clinical endpoint FoundationOne CDx Full analysis set results results N=256 N=248 Olaparib HRR (BRCA1, BRCA2, ATM, BARD1, BRIP1, rPFSa, months 6.2 5.8 CDK12, CHEK1, HRb (95% CI) 0.49 (0.38, 0.63) 0.49 (0.38, 0.63) CHEK2, FANCL, PALB2, RAD51B, RAD51C, RAD51D and RAD54L) a rPFS based on BICR using RECIST v1.1 and/or PCWG3, or death (by any cause in the absence of progression) regardless of whether the patient withdrew from randomized therapy or received another anticancer therapy prior to progression. b HR for both FoundationOne CDx and full analysis set compares olaparib to investigator’s choice of therapy (either enzalutamide 160 mg orally once daily or abiraterone acetate 1000 mg orally once daily with prednisone 5 mg orally twice daily [prednisolone was permitted for use instead of prednisone, if necessary]) for radiological disease progression or death. BICR, blinded independent central review; CI, confidence interval; HR, hazard ratio; HRR, homologous recombination repair; ORR, overall response rate; PCWG3, Prostate Cancer Working Group 3; RECIST, Response Evaluation Criteria in Solid Tumors; rPFS, radiological progression-free survival. Source: FoundationOne CDx Label1

CONFIDENTIAL PAGE 96 US-FDX-2000073 Last update: January 22, 2021

CGP in Melanoma

. Melanoma skin cancer accounts for an estimated 5.6% of new cancer diagnoses in the United States.299 . Survival for metastatic melanoma at 5 years is 27.3%, with approximately 4% diagnosed at this stage; however, 5-year survival is still only 66.2% in the 9% of patients initially diagnosed with regional (ie, lymph node involvement) disease due to metastatic recurrence.299 . Historically, melanoma patients with inoperable metastatic disease had median survival times of 8 to 12 months with chemotherapy with dacarbazine as the standard of care.300 o Additionally, melanoma was not broadly treated with adjuvant therapy (therapy after resection of the primary lesion) largely due to an absence of efficacious systemic therapy options.301 . More recently, selective inhibitors of intracellular proteins BRAF and MEK have shown improved outcomes in patients with BRAF V600e mutated melanoma, as have immune checkpoint inhibitors (ie, both blockers of cytotoxic T-lymphocyte-associated antigen-4 [CTLA-4] and of the programmed death-1 [PD-1] receptor).300 o Employing either or both of these modalities within the melanoma treatment landscape has increased the median survival to greater than 2 years.300 o Despite these significant advances, there remain populations of patients refractory to currently available therapies for whom the development of new and adjunct targeted therapies has been a focus of major research.302 . NCCN Guidelines recommendations for the provision of clinical care for patients with cutaneous melanoma note the following: a number of somatic genetic alterations have been identified in cutaneous melanoma, a few of which are targetable driver mutations that have proven useful to guide treatment decisions and/or clinical trial eligibility.31 Specifically for metastatic disease: o Tissue should be obtained to ascertain alterations in BRAF and in the appropriate clinical setting, KIT, from either biopsy of the metastasis (preferred) or archival material if the patient is being considered for targeted therapy. o Consider broader genomic profiling (eg, larger NGS panels, BRAF non-V600 mutations) if the test results might guide future treatment decisions or eligibility for participation in a clinical trial o Additionally, BRAF mutation testing is recommended for patients with stage III for whom future BRAF-directed therapy may be an option . FoundationOne CDx is an FDA-approved companion diagnostic for detection of BRAF V600E and V600K mutations for selection of therapy with vemurafenib or dabrafenib (V600E mutations) or trametinib or cobimetinib in combination with vemurafenib (V600E or V600K mutations).1

Unmet Need for Molecular Testing in Melanoma The treatment of melanoma is evolving rapidly with the advent of targeted agents and immunotherapies; the two main classes of agents currently are targeted agents against BRAF mutations and immune checkpoint inhibitors.302 Approximately ~40% of patients with melanoma harbor a BRAF mutation, making them eligible for treatment with BRAF/MEK inhibitor therapy.302 Both immunotherapies and BRAF/MEK inhibitor therapies have improved OS as compared with historically used chemotherapy.300

CONFIDENTIAL PAGE 97 US-FDX-2000073 Last update: January 22, 2021

. In a meta-analysis, use of a BRAF + MEK inhibitor or PD-1 ± CTLA-4 inhibition improved mean OS versus chemotherapy (Figure 5-9); notably, this was consistent in both first and in subsequent lines of therapy – an important point as patients will inevitably progress and for those patients with a targetable BRAF mutation, this gives them the opportunity to see both of these modalities at some point during their patient journey.300

Figure 5-9. Mean OS Curves, Created by Weighted Averaging of Digitized KM Survival Curves From Selected Clinical Trials of Melanoma Patients

CTLA-4, cytotoxic T lymphocyte antigen-4; PD-1, programmed cell death protein-1. Source: Ugurel 2017.300

Additionally, with the success with these agents seen in the metastatic setting, both immunotherapy and the BRAF + MEK combination of dabrafenib + trametinib have moved into the adjuvant treatment landscape and are being investigated as neo-adjuvant treatment for stage III disease.303

Place of CGP in Melanoma Given the number of genomic alterations with a targetable/actionable agent available for the management of mCRC and the ability to detect other infrequent alterations, approaches using broad panel or CGP represent the most complete method for determining genomic alterations for patients with advanced cutaneous melanoma. NCCN Guidelines for Cutaneous Melanoma note to consider broader genomic

CONFIDENTIAL PAGE 98 US-FDX-2000073 Last update: January 22, 2021

profiling (eg, larger NGS panels, BRAF non V600 mutations) if the test results might guide future treatment decisions or eligibility for participation in a clinical trial. Further, if BRAF single-gene testing was the initial test performed and is negative, clinicians should strongly consider larger NGS panels to identify other potential genetic targets (eg, KIT, BRAF non-V600).31

Table 5-29 reviews the recommended biomarker testing to be conducted in cutaneous melanoma, as well as the ability of the FoundationOne Tissue CDx assay to test for these recommended biomarkers.

Table 5-29. Review of Guideline-Recommended Biomarker Testing for Patients With Cutaneous Melanoma and FoundationOne Tissue CDx Assay Capabilities Biomarker included in Applicable guidelines for Recommended FoundationOne Tissue CDx tumor profiling biomarker testinga assay NCCN Guidelines for BRAF mutationsb Yes Cutaneous Melanoma c V.1.202131 KIT Yes a Category 1: Based upon high-level evidence, there is uniform NCCN consensus that the intervention is appropriate. Category 2A: Based upon lower-level evidence, there is uniform NCCN consensus that the intervention is appropriate. All NCCN recommendations are category 2A unless otherwise indicated. b Recommended for stage IV disease or for clinical recurrence and for stage III disease at high risk for recurrence for whom future BRAF-directed therapy may be an option. c Recommended in the appropriate clinical setting from either biopsy of the metastasis (preferred) or archival material if the patient is being considered for targeted therapy. ALK, anaplastic lymphoma kinase; BRAF, serine/threonine-protein kinase B-raf; NCCN, National Comprehensive Cancer Network; NTRK, neurotrophic receptor tyrosine kinase; RAS, rat sarcoma virus gene family [includes KRAS and NRAS]; ROS1, oncogene that encodes the ROS1 tyrosine kinase receptor.

Clinical Utility and Validity of FoundationOne CDx in Melanoma FoundationOne CDx is an in vitro diagnostic device that targets 324 genes. The test also detects the genomic signatures TMB and MSI-H. FoundationOne CDx utilizes DNA isolated from FFPE from the tumor tissue of cancer patients. The test is intended to be used as a companion diagnostic to identify patients with prostate cancer who may benefit from treatment with the targeted therapies listed in Table 5-30 in accordance with the approved therapeutic product labeling. Additionally, FoundationOne CDx is intended to provide tumor mutation profiling to be used by qualified healthcare professionals in accordance with professional guidelines in oncology for patients with malignant neoplasms, including melanoma.1 As described in the Pan-tumor section, other biomarkers (NTRK, TMB) are relevant in all solid tumors.

Table 5-30. Companion Diagnostic Indications Pertinent for Patients With Melanoma Tumor type Biomarker(s) detected Therapy BRAF V600E Tafinlar® (dabrafenib) Melanoma Zelboraf® (vemurafenib)

CONFIDENTIAL PAGE 99 US-FDX-2000073 Last update: January 22, 2021

Tumor type Biomarker(s) detected Therapy BRAF V600E or V600K Mekinist® (trametinib) or Cotellic® (cobimetinib), in combination with Zelboraf® (vemurafenib) BRAF, serine/threonine-protein kinase B-raf. Source: FoundationOne CDx Label1

BRAF V600 Clinical validity of the FoundationOne CDx assay was evaluated as a companion diagnostic in identification of patients with BRAF V600 alterations.1 . Based on the concordance results described in Table 5-31, FoundationOne CDx has been demonstrated to be concordant to the cobas BRAF V600 Mutation Test and the THxID BRAF kit for the detection of BRAF V600 genomic alterations.1

Table 5-31. Clinical Validity of FoundationOne CDx for Melanoma Companion Diagnostic Claims Biomarker PPA, % (n/N) NPA, % (n/N) Comparator assay BRAF V600 99.4 (166/167) 89.6 (121/135)a cobas BRAF V600 Mutation Test BRAF V600E 99.3 (149/150) 99.2 (121/122) BRAF V600 dinucleotide 96.3 (26/27) 100 (24/24) THxID™ BRAF kit a The reported difference in NPA values for BRAF V600 and BRAF V600E are likely attributed to known sensitivity differences in the cobas test, which has lower sensitivity for detection of dinucleotide V600 alterations than for the single nucleotide V600E c.1799T>A alteration, especially for samples in which FoundationOne CDx detected the nucleotides to be of lower than 40% mutational allele frequency, leading to low NPA values. NPA, negative percent agreement; PPA, positive percent agreement. Source: FoundationOne CDx Label1

CONFIDENTIAL PAGE 100 US-FDX-2000073 Last update: January 22, 2021

CGP in Cholangiocarcinoma

. Cholangiocarcinomas are a part of a group of biliary tract cancers that arise from the epithelial cells of the biliary tree and include both intra- and extra-hepatic sites of origin; of note, gallbladder cancer is also included in this grouping.304 305 o Globally, biliary tract cancers represent 3-5% of cancer diagnoses. . The mortality rate for patients with biliary tract cancers, in general, is high with most patients diagnosed in advanced stages; the 5-year overall survival with for intra- and extra-hepatic cholangiocarcinoma, specifically, is 8% to 10%.305,306 . Prior to the chemotherapy regimen of cisplatin combined with gemcitabine, there was no standard of care for advanced or metastatic disease; and although cisplatin and gemcitabine have improved survival, the median overall survival with this regimen is still <1 year (11.7 months).307 o Further, after progression on cisplatin/gemcitabine, with chemotherapy (oxaliplatin and fluorouracil, ie, FOLFOX), median overall survival was increased by <1 month versus best supportive care, alone (6.2 vs 5.3 months).306 . The NCCN Guidelines for the provision of clinical care for patients with advanced biliary tract cancers note the following: a biopsy for molecular testing is recommended to potentially guide targeted treatment.34 Specifically these guidelines note that: o There is a potentially increasing role for molecular profiling of cholangiocarcinomas, with IDH1/2 mutations occurring in 10% to 23% and FGFR mutations occurring in 8% to 14% of intrahepatic cholangiocarcinomas.34 . FoundationOne CDx is an FDA-approved companion diagnostic for detection FGFR fusions or select rearrangements for selection of therapy with pemigatinib.1

Unmet Need for Molecular Testing in Cholangiocarcinoma Advanced biliary tract carcinomas such as cholangiocarcinoma carry a poor prognosis owing to a lack of chemosensitivity.306 Best supportive care has been associated with a median overall survival of ~2.5 months, and while survival has been improved with chemotherapy (specifically with a combination of cisplatin and gemcitabine), the median overall survival is still <12 months.306,307 More recently, however, molecular characterization has demonstrated that the mutations in genes such as IDH1/2 genes as well as FGFR2 fusions, which are mainly associated mostly with cholangiocarcinoma, may have therapeutic impact with the availability of agents that target these anomalies).112,305,308 Specifically, in the second-line setting, chemotherapy alone in an unselected (ie, non-CGP tested) population of patients with advanced or metastatic biliary tract cancer, increased median overall survival by <1 month versus best supportive care, alone (6.2 vs 5.3 months).306,309 However, in patients genomically matched to targeted therapies in clinical trials: . Patients with advanced or metastatic cholangiocarcinoma who had progressed after 1 prior therapy, and who harbored an FGFR2 fusion or rearrangement (n=107) had a median overall survival of 21.1 months (95% CI: 14.8, not evaluable) when treated with the FGFR-inhibitor, pemigatinib.112 . Patients with up to 2 previous treatments and with an IDH1-mutated, advanced or metastatic cholangiocarcinoma, median overall survival was 10.8 months (95% CI: 7.7, 17.6) for patients treated with ivosidenib vs 9.7 months (95% CI: 4.8, 12.1) with placebo; notably, however, cross-

CONFIDENTIAL PAGE 101 US-FDX-2000073 Last update: January 22, 2021

over to ivosidenib at progression was allowed in the placebo group.308 o The median pre-specified adjusted overall survival (adjusted utilizing a rank-preserving structural failure time method to reconstruct the survival curve for patients receiving placebo as if crossover had never occurred) revealed a median overall survival of 6.0 months (95% CI: 3.6, 6.3) in the placebo group.

Place of CGP in Cholangiocarcinoma Future clinical trial design, in order to improve patient outcomes over one-size-fits-all chemotherapy, need to be increasingly geared toward characterization of histologic and molecular subtypes of biliary tract cancers in an effort to the select patients for the appropriately targeted agent utilizing biomarkers that could predict treatment response.304 Further, CGP (in the form of whole-genome and transcriptome sequencing) recently highlighted that in a large cohort of patients with biliary tract cancers, including intra-hepatic cholangiocarcinoma (n=145), perihilar or distal cholangiocarcinoma (n=86) or gallbladder cancer (n=29), a potentially targetable alteration was seen in up to 39% of patients.306,310 NCCN Guidelines for Hepatobiliary Cancers note to consider additional molecular testing (ie, in addition to MSI/dMMR testing) to potentially guide targeted treatment.34 Table 5-32 reviews the recommended biomarker testing to be conducted in biliary tract cancers, as well as the ability of the FoundationOne Tissue CDx assay to test for these recommended biomarkers.

Table 5-32. Review of Guideline-Recommended Biomarker Testing for Patients With Hepatobiliary Cancers and FoundationOne Tissue CDx Assay Capabilities Applicable guidelines for Recommended Biomarker included in tumor profiling biomarker testinga FoundationOne Tissue CDx assay NCCN Guidelines for Gallbladder cancer, unresectable or metastatic Hepatobiliary Cancers NTRK fusions Yes V.5.202034 MSI/dMMR Yes Intra- and extra-hepatic cholangiocarcinoma, unresectable or metastatic NTRK fusions Yes MSI/dMMR Yes FGFR2 fusions Yes IDH1 mutations Yes a Category 1: Based upon high-level evidence, there is uniform NCCN consensus that the intervention is appropriate. Category 2A: Based upon lower-level evidence, there is uniform NCCN consensus that the intervention is appropriate. All NCCN recommendations are category 2A unless otherwise indicated. FGFR2, fibroblast growth factor receptor; IDH, isocitrate dehydrogenase; MSI – microsatellite instability; NTRK, neurotrophic receptor tyrosine kinase.

Clinical Utility and Validity of FoundationOne CDx in Cholangiocarcinoma FoundationOne CDx is an in vitro diagnostic device that targets 324 genes. The test also detects the genomic signatures TMB and MSI-H. FoundationOne CDx utilizes DNA isolated from FFPE from the

CONFIDENTIAL PAGE 102 US-FDX-2000073 Last update: January 22, 2021

tumor tissue of cancer patients. The test is intended to be used as a companion diagnostic to identify patients with cholangiocarcinoma who may benefit from treatment with the targeted therapies listed in Table 5-33 in accordance with the approved therapeutic product labeling. Additionally, FoundationOne CDx is intended to provide tumor mutation profiling to be used by qualified healthcare professionals in accordance with professional guidelines in oncology for patients with malignant neoplasms, including cholangiocarcinoma.1 As described in the Pan-tumor section, other biomarkers (NTRK, TMB) are relevant in all solid tumors.

Table 5-33. Companion Diagnostic Indications Pertinent for Patients With Cholangiocarcinoma Tumor type Biomarker(s) detected Therapy Cholangiocarcinoma FGFR2 rearrangements Pemazyre™ (pemigatinib)

FGFR2, fibroblast growth factor receptor. Source: FoundationOne CDx Label1

FGFR2 FIGHT-202 was a prospective, multicenter, open-label phase 2 study to assess the efficacy of pemigatinib in patient with previously treated, advanced cholangiocarcinoma.112 Patients were assigned to 1 of 3 cohorts (patients with FGFR2 fusions and rearrangements, patients with other FGF/FGFR alterations, or patients with no FGF/FGFR alterations) based on FoundationOne CTA test results.112 Retrospective testing with FoundationOne CDx yielded 181 CDx-evaluable results (84 positive and 97 negative). Of the 107 patients in the FGFR2 fusions or rearrangements cohort in FIGHT-202, banked DNA samples were available for 80 patients for the FoundationOne CDx bridging study.1 . Agreement between FoundationOne CDx and the FoundationOne CTA was demonstrated with PPA, NPA, OPA, adjusted PPV, and adjusted NPV all exhibiting 100% agreement for the detection of FGFR2 fusions or rearrangements (Table 5-34).1 o Based on these results, FoundationOne CDx has been demonstrated to be concordant to the FoundationOne CTA for the detection of FGFR2 fusions or rearrangements.1 . The clinical utility of FoundationOne CDx was established as the ORR of the pemigatinib treated patients as determined by the CTA or by FoundationOne CDx was similar (35.51%, 95% CI: 26.50, 45.35 and 37.50%, 95% CI: 26.92, 49.04, respectively) (Table 5-34).1

Table 5-34. Clinical Validity and Utility of FoundationOne CDx for Companion Diagnostic Claims Drug and Concordance Clinical efficacy alteration ORRa, % (95% CI) PPA, % NPA, % Comparator FoundationOne CTA results (n/N) (n/N) assay CDx results N=107 N=80 Pemigatinib 100 (84/84) 100 (97/97) FoundationOne 37.50 35.51 FGFR2 CTA (26.92, 49.04) (26.50, 45.35) rearrangements a ORR per central review per RECIST v1.1.

CONFIDENTIAL PAGE 103 US-FDX-2000073 Last update: January 22, 2021

CI, confidence interval; CTA, clinical trial assay; NPA, negative percent agreement; ORR, overall response rate; PPA, positive percent agreement; RECIST, Response Evaluation Criteria in Solid Tumors. Source: FoundationOne CDx Label1

CONFIDENTIAL PAGE 104 US-FDX-2000073 Last update: January 22, 2021

6 ECONOMIC VALUE OF FOUNDATIONONE CDX

Economic Benefits Associated With CGP in Advanced Cancer

. The ability of CGP to identify patients for clinical trial enrollment could potentially lead to cost offsets because of the diversion of anticancer drug costs to the study sponsor. . Published studies assessing the economic value of CGP show that precision medicine-based approaches in oncology lead to an increase in total medical costs to the health plan; however, these increases are modest and occur mainly because patients are treated for longer periods because of prolonged life and/or slower disease progression. . A customized budget impact model was developed based on published evidence; when an 8% increase in use of CGP was modeled, the budget impact (including cost of testing) was modest.

There are few published economic analyses of FoundationOne CDx or other CGP tests that evaluate the impact on health plans. There is the opportunity for additional cost savings as testing rates increase due to the ability of CGP to identify patients who are eligible for clinical trials of investigational agents, thus diverting a proportion of drug costs to study sponsors.59 However, the use of a precision medicine-based approach may lead to an increase in total medical costs primarily because it achieves the ultimate goal in oncology of prolonging life and delaying disease progression.202,311 In particular, patient costs are similar per week between those receiving precision medicine compared with chemotherapy or best supportive care, and thus the observed cost increases occur primarily because patients receive treatment for a longer period.11 Although further research is needed to determine whether the main drivers of patient costs are shifting, patients who receive precision medicine-based therapy may incur higher drug costs but may simultaneously incur lower costs associated with the impacts of treatment toxicity.11,311 Ultimately, the evidence shows that patients treated with precision medicine-based approaches may live longer and have fewer treatment-related complications, which is accompanied by a manageable increase in overall cost.202,312

For the most up-to-date information related to the economic value associated with FoundationOne CDx, please contact Foundation Medicine Payer Relations.

Potential savings associated with clinical trial enrollment Clinical trials provide patients who have advanced cancer and few remaining treatment options with access to investigational agents; often, these are targeted to genomic alterations that are identified by CGP and for which no approved therapies are currently available. Physicians who were surveyed from the TAPUR™ Study, a multi-basket study of agents targeting tumor genomics that also examined attitudes, perceptions, and use of tumor genomic testing reported most often utilizing tumor genomic testing for pretreated advanced cancer patients without other options (96%).313 In addition to the opportunity to improve outcomes among these patients, enrollment into clinical trials may also lead to economic benefits arising from the diversion of anticancer drug costs to the study sponsor. Enrollment rates into clinical trials were assessed in a single cohort of patients tested with CGP.

CONFIDENTIAL PAGE 105 US-FDX-2000073 Last update: January 22, 2021

Clinical trial enrollment among patients with solid tumors A prospective, observational study of 120 enrolled patients with incurable, solid tumor malignancies who had progressed on 1 or more prior lines of therapy or who had no standard first-line treatment options available was undertaken to assess the feasibility of CGP with FoundationOne in a community practice setting using a centralized molecular tumor board approach.59 Of 63 patients for whom a treatment was recommended, CGP led to genomic-matched therapy in 39 (62%), with 16 of these patients enrolling in a clinical trial. Historically, clinical trial enrollment rates have been low, particularly in the community oncology setting, and identifying trials specific for each patient was burdensome. However, with the advancing knowledge of genomic biomarkers and emerging treatment options, as well as the use of CGP testing to increase the number of eligible patients who are identified and ultimately enrolled in clinical trials, trial enrollment remains a potential source of cost diversion back to study sponsors among patients with advanced cancer. A retrospective single-center analysis of oncology patients who underwent CGP estimated the potential cost offset associated with deferred drug costs of 20 patients who enrolled in a clinical trial following CGP. Assuming a treatment duration of 3.23 months and an avoidance of the cost of the regimen most likely to be chosen by a provider off-trial, the payer may have accrued an annual cost benefit of $25,000 per patient enrolled in a clinical trial.61

Matched compared with unmatched therapy In total, 3 studies comparing matched with unmatched therapies were identified. These studies provide additional clinical evidence supporting the fact that matched therapy may be superior to unmatched therapy and also show that the additional costs of matched treatment are disproportionally related to improved clinical outcomes rather than to incremental increases in monthly drug costs. Furthermore, the low percentages of patients currently receiving molecular diagnostic testing is highlighted. CGP with FoundationOne in patients with diverse refractory cancers In an analysis that estimated the cost of anticancer therapy directed by CGP,311 costs were estimated using complete data from a prospective, nonrandomized, phase 1 oncology center study that investigated patients (N=188) with diverse refractory cancers who underwent CGP with FoundationOne and who were treated with matched (n=122) or unmatched (n=66) therapy.67 A drug was considered matched if the half maximal inhibitory concentration affected the target at low nanomolar range (for small molecule inhibitors) or if the target was the primary one recognized by an antibody. Further details on matching definitions are available in the primary clinical study manuscript.67 Patients who received matched therapy had a longer TTF than patients who received unmatched therapy (difference, 1.5 months; P<0.002) and had a longer observed OS (difference, 2.4 months; P<0.002). Total drug treatment costs were higher among patients treated with matched therapy than among those who were unmatched ($68,729 vs $30,664; P=0.003; Figure 6-1). However, most of the increased costs were attributable to a longer duration of therapy (ie, a longer TTF) rather than higher monthly drug costs (66.3% vs 33.7%, respectively; Figure 6-1).

CONFIDENTIAL PAGE 106 US-FDX-2000073 Last update: January 22, 2021

Figure 6-1. Comparison of Total Drug Treatment Costs Between Matched and Unmatched Therapy Among Patients Who Received CGP

CGP, comprehensive genomic profiling. Source: Signorovitch et al. (2017).311

In summary, this analysis of patients who received treatment following CGP in a phase 1 clinical trial showed that treatment with matched therapy was associated with longer treatment durations, improved survival, and manageable incremental costs compared with unmatched therapy. Real-world matched cohort study of patients with metastatic cancers In a matched cohort study of 72 patients with metastatic cancers who were treated in the precision cancer medicine program within the Intermountain Healthcare delivery system between 2013 and 2015, outcomes of a cohort of patients who underwent genomic testing and received targeted treatment (n=36) were compared with those of a historical control cohort of patients who received standard chemotherapy (n=29) or best supportive care (n=7) between 2010 and 2015.11 Total medical and drug costs were both significantly increased among patients receiving precision medicine-based treatment compared with historical controls (Table 6-1). The difference was somewhat attenuated, however, when the costs were averaged by PFS week ($4,665 vs $5,000; P=0.126).

Table 6-1. Healthcare-Associated Cost Outcomes With Precision Medicine-Based Treatment Compared With Standard Chemotherapy or Best Supportive Care Precision medicine group Historical controls (n=22) (n=22) P-Value Total costs per patient 91,790 (85,070) 40,782 (42,267) 0.002 Total drug costs per patient 59,259 (51,425) 20,189 (34,299) <0.001 Cost per patient per PFS week 4,665 (3,041) 5,000 (6,509) 0.126 Data are presented as mean (standard deviation) US dollars unless otherwise stated. PFS, progression-free survival. Source: Haslem et al. (2017).11

In summary, this study demonstrated that the survival benefit of precision medicine-based treatment over traditional chemotherapy appears to be the driving factor behind increased total costs.

CONFIDENTIAL PAGE 107 US-FDX-2000073 Last update: January 22, 2021

Real-world utilization of molecular diagnostic testing and matched drug therapies A US healthcare claims analysis of 8,193 patients with 6 metastatic cancer types (breast, NSCLC, CRC, head and neck, ovarian, and uterine) investigated the rates of utilization and the average costs of molecular diagnostic testing.312 Across the patient sample, genomically matched targeted therapy was used by 6% to 11% of patients with metastatic breast cancer, NSCLC, and CRC; no patients with metastatic head and neck, ovarian, or uterine cancers used this therapy. Unmatched targeted therapy was utilized by 1% to 21% in patients across the different cancer types (Table 6-2). The cost of genomically matched therapy compared with cytotoxic chemotherapy on a per-patient-per- month (PPPM) basis was $349 vs $293 in patients with breast cancer; $255 vs $425 in patients with NSCLC; and $164 vs $701 in patients with CRC. Unmatched targeted therapy was less costly than cytotoxic chemotherapy for all cancer types with the exception of head and neck cancer ($77 vs $45). Total medical costs excluding anticancer drug costs ranged from $6,618 to $9,940 PPPM, driven primarily by outpatient visits and hospitalizations (Table 6-2). Overall, this study demonstrated that molecular testing in general is highly underutilized despite its relatively low cost of use, and there is no clear trend in higher cost associated with using matched targeted therapy over chemotherapy.

Table 6-2. Utilization of Molecular Testing and Healthcare Costs by Cancer-Specific Cohort (Metastatic Cancer) Breast NSCLC CRC Head Ovarian Uterine (n=3,414) (n=2,231) (n=1,611) and neck (n=275) (n=151) (n=511) Utilization, % Biopsy procedures 23.2 31.1 30.9 33.3 73.1 6.6 Molecular diagnostic 52.4 41.9 37.4 34.1 40.7 42.4 tests Targeted therapy Genomically 11 9 6 0 0 0 matched Endocrine 60 0 0 0 14 16 Unmatched 3 7 21 5 9 1 Cytotoxic 41 39 52 26 47 21 chemotherapy Mean (SD) PPPM costs, $ Biopsy procedures 14 (180) 37 (531) 12 (142) 16 (80) 42 (84) 3 (31) Molecular diagnostic 40 (127) 35 (132) 29 (155) 7 (24) 106 (780) 32 (135) tests

CONFIDENTIAL PAGE 108 US-FDX-2000073 Last update: January 22, 2021

Targeted therapy Genomically 349 (1,464) 255 (1,152) 164 (1,005) NA NA NA matched Endocrine 60 (238) – – – 4 (31) 3 (14) Unmatched 84 (697) 240 (1,420) 545 (1,685) 77 (690) 176 (845) NA Cytotoxic 293 (823) 425 (1,528) 701 (1,709) 45 (264) 300 (881) 61 (299) chemotherapy Total medical costs, 6,667 8,405 8,521 6,618 9,940 7,823 excluding anticancer (11,011) (16,642) (14,503) (9,969) (15,043) (15,176) drugs Hospice/palliative 17 (152) 114 (605) 61 (351) 49 (403) 80 (412) 35 (304) care Emergency 93 (359) 244 (715) 161 (472) 91 (288) 223 (559) 153 (375) department visits Hospitalizations 1,484 3,582 3,842 1,894 5,306 3,207 (8,181) (12,965) (11,847) (5,407) (13,243) (12,930) Outpatient visits 4,946 4,322 4,172 4,452 3,990 4,258 (6,642) (7,831) (6,991) (7,113) (5,166) (7,087) Other visits 128 (375) 142 (614) 284 (1,322) 132 (430) 341 170 (830) (1,356) CRC, colorectal cancer; NA, not applicable; NSCLC, non-small cell lung cancer; PPPM, per-patient-per-month; SD, standard deviation; USD, United States dollars. Utilization is the proportion of patients with any use. Costs are in 2015 USD. Source: Chawla et al. (2018).312

Comprehensive genomic profiling compared with conventional testing Offering CGP over conventional molecular testing may allow a greater number of patients to benefit from matched therapies without the need for sequential testing for individual genomic alterations. Although it has been suggested that a higher proportion of matched therapy use leads to an increase in cost, this increase been shown to occur primarily because of improved patient survival, which is the ultimate goal of CGP.11,311 The budget impact of increased CGP over conventional testing use was estimated in 1 identified economic model. Budget impact and clinical outcomes of patients with NSCLC This study modeled the incremental increase in a healthcare system budget associated with an increased use of CGP compared with conventional molecular diagnostic testing among patients with advanced NSCLC.314 In particular, it estimated the incremental benefits and costs of CGP using FoundationOne compared with conventional molecular diagnostic testing using a decision-analytic model to compare a 2% rate of CGP use vs an increase to 10% among 2 million covered lives. Clinical outcomes, budget impact, and the number needed to test (NNT) in order to save 1 life-year were estimated.

CONFIDENTIAL PAGE 109 US-FDX-2000073 Last update: January 22, 2021

In the model population, 532 patients were estimated to have advanced NSCLC, and 266 were estimated to be tested with molecular diagnostics (Table 6-3). An increase in CGP among those tested from 2% to 10% (the equivalent of 21 additional patients tested with CGP) was associated with an incremental increase of $1,600 in cost and 1.9 additional life-years, resulting in a budget impact of $0.02 per-member per-month (PMPM). Most of the budget impact was attributable to changes in drug treatment, longer treatment, and longer survival ($0.013 PMPM); the remainder was due to incremental costs of FoundationOne vs conventional molecular diagnostic tests ($0.005 PMPM). Total per-patient drug costs and total costs with 2% CGP utilization were $45,305 and $106,119; with 10% CGP utilization, these costs increased only slightly to $45,946 and $107,720, respectively. This analysis also found that, among those tested, approximately 12 patients would need to be tested with FoundationOne compared with conventional molecular diagnostic testing to gain 1 life-year (Table 6-3). Sensitivity analyses found that the model was most sensitive to the line of therapy at which CGP was used, the proportion of patients using CGP, and OS and treatment duration. Adjusting the line of therapy to 1 for all patients increased the budget impact to slightly more than $0.03 PMPM and reduced the NNT to 5. Overall, this study demonstrated that an 8% absolute increase in the use of CGP in patients with NSCLC led to a modest budget impact. As shown in other economic studies, a longer duration of treatment and longer survival were notable drivers of the cost increase.

Table 6-3. Incremental OS and Budget Impact With Increase in CGP Base case Increase in CGP (FoundationOne (FoundationOne testing rate, 2%) testing rate, 10%) Difference Testing patterns, n Advanced NSCLC 532 532 NA Molecular diagnostic testing 266 266 NA Undergoing CGP 5 27 21 Costs/patient, $a Total 106,119 107,720 1,600 Diagnostic testing 1,026 1,415 390 Biopsy 411 396 −15 Medical 59,377 59,963 585 Drug 45,305 45,946 640 Matched 10,679 11,914 1,235 Nonmatched 34,626 34,031 −595 Total life-years 188.5 190.4 1.9 Budget impact, $PMPM 0.02

CONFIDENTIAL PAGE 110 US-FDX-2000073 Last update: January 22, 2021

NNT 12 CGP, comprehensive genomic profiling; NA, not applicable; NNT, number needed to test; NSCLC, non-small cell lung cancer; OS, overall survival; PMPM, per-member per-month; USD, United States dollars. a Costs are in USD. Source: Signorovitch et al. (2019).314

CONFIDENTIAL PAGE 111 US-FDX-2000073 Last update: January 22, 2021

7 APPENDICES

Tables Table 1-1. Detection of More Patients With Actionable Genomic Alterations With CGP in Specific Tumor Types ...... 6 Table 1-2. NCCN Guidelines Recommendations for Broad Molecular Testing/NGS ...... 7 Table 1-3. NCCN Guidelines Recommended Biomarkers for Molecular Testing in Select Tumor Types .. 8 Table 1-4. Selected Outcomesa in Studies Showing the Clinical Utility of Foundation Medicine CGP .... 10 Table 1-5. FoundationOne CDx Companion Diagnostic Claims ...... 11 Table 1-6. Clinical Validity of FoundationOne CDx for Pan-tumor Companion Diagnostic Claims ...... 12 Table 1-7. Clinical Utility of FoundationOne CDx for Pan-tumor Companion Diagnostic Claims ...... 13 Table 1-8. Clinical Validity of FoundationOne CDx for NSCLC Companion Diagnostic Claims ...... 13 Table 1-9. Clinical Utility of FoundationOne CDx for NSCLC Companion Diagnostic Claims ...... 14 Table 1-10. Clinical Validity of FoundationOne CDx for Breast Cancer Companion Diagnostic Claims 15 Table 1-11. Clinical Utility of FoundationOne CDx for Breast Cancer Companion Diagnostic Claims ... 15 Table 1-12. Clinical Validity of FoundationOne CDx for CRC Companion Diagnostic Claims ...... 16 Table 1-13. Clinical Utility of FoundationOne CDx for Ovarian Cancer Companion Diagnostic Claims 16 Table 1-14. Clinical Utility of FoundationOne CDx for Prostate Cancer Companion Diagnostic Claims 17 Table 1-15. Clinical Validity of FoundationOne CDx for Melanoma Companion Diagnostic Claims ...... 18 Table 1-16. Clinical Validity of FoundationOne CDx for Companion Diagnostic Claims ...... 18 Table 1-17. Clinical Utility of FoundationOne CDx for Companion Diagnostic Claims ...... 18 Table 2-1. Molecular Testing Recommendations in Select Tumor Types per NCCN Guidelines ...... 27 Table 2-2. Improved Clinical Outcomes With Genomically Matched Targeted Therapy, Pan-Tumor Studies ...... 29 Table 2-3. Improved Clinical Outcomes With Immunotherapy Matched to TMB or MSI-H/dMMR ...... 31 Table 2-4. Biomarkers and Real-World Testing Patterns in Select Advanced Cancers ...... 34 Table 2-5. Improved Detection of Genomic Alterations With CGP Testing ...... 36 Table 3-1. FoundationOne CDx Companion Diagnostic Indicationsa ...... 39 Table 3-2. Sample Validation for FoundationOne CDx Companion Diagnostic Claims ...... 44 Table 3-3. Sample Validation for FoundationOne CDx Companion Diagnostic Claims ...... 44 Table 3-4. Sample Validation for FoundationOne CDx ...... 45 Table 3-5. Concordance Summary for Genomic Alterations Between FoundationOne CDx and FoundationOne ...... 46

CONFIDENTIAL PAGE 112 US-FDX-2000073 Last update: January 22, 2021

Table 3-6. Concordance data for FoundationOne CDx Claims ...... 47 Table 4-1. Clinical Utility of CGP ...... 52 Table 4-2. NCCN Guidelines Recommendations for Broad Molecular Testing/NGS ...... 55 Table 5-1. FDA-approved Tumor Agnostic Therapies ...... 60 Table 5-2. Select Tumor Agnostic Therapies in Development ...... 61 Table 5-3. Clinical Utility of FoundationOne CDx for Determining TMB-H Status (≥10 mut/Mb) in Solid Tumors in the Efficacy Analysis Population ...... 67 Table 5-4. Clinical Validity and Utility of FoundationOne CDx for Pan-tumor NTRK Gene Fusion Companion Diagnostic Claim ...... 68 Table 5-5. Overall Survival by Receipt vs No Receipt of Genotype-Directed Therapy in Metastatic NSCLC Patients ...... 71 Table 5-6. Review of Guideline-Recommended Biomarker Testing and FoundationOne CDx Assay Capabilities ...... 71 Table 5-7. Companion Diagnostic Indications Pertinent for Patients With NSCLC ...... 73 Table 5-8. Clinical Validity of FoundationOne CDx for NSCLC EGFR Mutation Companion Diagnostic Claims ...... 74 Table 5-9. Clinical Validity of FoundationOne CDx for NSCLC ALK Rearrangement Companion Diagnostic Claims ...... 74 Table 5-10. Clinical Validity of FoundationOne CDx for NSCLC BRAF V600E Mutation Companion Diagnostic Claim ...... 75 Table 5-11. Clinical Validity and Utility of FoundationOne CDx for NSCLC MET Exon 14 Skipping Companion Diagnostic Claim ...... 76 Table 5-12. Genomic Alterations With Targeted Therapies Available or in Clinical Trials ...... 77 Table 5-13. BRCA1/2 Testing by HR Status and Known Family History of Breast or Ovarian Cancer .... 79 Table 5-14. Review of Guideline-Recommended Biomarker Testing and FoundationOne CDx Assay Capabilities in Breast Cancer ...... 80 Table 5-15. FoundationOne CDx Companion Diagnostic Indications Pertinent for Patients With Breast Cancer ...... 80 Table 5-16. Clinical Validity of FoundationOne CDx for Breast Cancer HER2 Amplifications Companion Diagnostic Claims ...... 81 Table 5-17. Clinical Validity and Utility of FoundationOne CDx for Breast Cancer PIK3CA Mutation Companion Diagnostic Claim ...... 82 Table 5-18. Review of Guideline-Recommended Biomarker Testing for Patients With Metastatic Colon and Rectal Cancers and FoundationOne Tissue CDx Assay Capabilities ...... 85 Table 5-19. Companion Diagnostic Indications Pertinent for Patients With Colorectal Cancer ...... 86

CONFIDENTIAL PAGE 113 US-FDX-2000073 Last update: January 22, 2021

Table 5-20. Clinical Validity of FoundationOne CDx for CRC Companion Diagnostic Claims ...... 86 Table 5-21. Genomic Alterations With Targeted Therapies Available ...... 87 Table 5-22. Review of Guideline-Recommended Biomarker Testing and FoundationOne CDx Assay Capabilities in Ovarian Cancer ...... 89 Table 5-23. Companion Diagnostic Indications Pertinent for Patients With Ovarian Cancer ...... 89 Table 5-24. Clinical Utility of FoundationOne CDx for Ovarian Cancer Companion Diagnostic Claims 90 Table 5-25. DDR Genes Frequently Altered in Prostate Cancer ...... 93 Table 5-26. Review of Guideline-Recommended Biomarker Testing for Patients With Advanced (Regional and Metastatic) Prostate Cancer and FoundationOne CDx Assay Capabilities ...... 94 Table 5-27. Companion Diagnostic Indications Pertinent for Patients With Prostate Cancer ...... 95 Table 5-28. Clinical Utility of FoundationOne CDx for Prostate Cancer Companion Diagnostic Claims 96 Table 5-29. Review of Guideline-Recommended Biomarker Testing for Patients With Cutaneous Melanoma and FoundationOne Tissue CDx Assay Capabilities ...... 99 Table 5-30. Companion Diagnostic Indications Pertinent for Patients With Melanoma ...... 99 Table 5-31. Clinical Validity of FoundationOne CDx for Melanoma Companion Diagnostic Claims .... 100 Table 5-32. Review of Guideline-Recommended Biomarker Testing for Patients With Hepatobiliary Cancers and FoundationOne Tissue CDx Assay Capabilities ...... 102 Table 5-33. Companion Diagnostic Indications Pertinent for Patients With Cholangiocarcinoma ...... 103 Table 5-34. Clinical Validity and Utility of FoundationOne CDx for Companion Diagnostic Claims .... 103 Table 6-1. Healthcare-Associated Cost Outcomes With Precision Medicine-Based Treatment Compared With Standard Chemotherapy or Best Supportive Care ...... 107 Table 6-2. Utilization of Molecular Testing and Healthcare Costs by Cancer-Specific Cohort (Metastatic Cancer) ...... 108 Table 6-3. Incremental OS and Budget Impact With Increase in CGP ...... 110 Table 7-1. NCCN Guidelines Recommendations for Molecular Testing Within Select Solid Tumor Cancer Types and Relevant Foundation Medicine Testing ...... 126 Table 7-2. Biomarker-based Targeted Therapies and Immunotherapies Recommended in NCCN Guidelinesa in Selected Cancers along with Companion Diagnostics ...... 133 Table 7-3. FoundationOne Portfolio ...... 145 Table 7-4. List of Genes With Full Coding Exonic Regions Included in the FoundationOne CDx Assay for the Detection of Substitutions, Indels, and CNAs ...... 146 Table 7-5. List of Genes With Select Intronic Regions Included in the FoundationOne CDx assay for the Detection of Gene Rearrangements, a Promoter Region, and an ncRNA Gene ...... 148 Table 7-6. Concordance Summary for Short Variants (Including VUS) Between FoundationOne CDx and UW OncoPlex ...... 153

CONFIDENTIAL PAGE 114 US-FDX-2000073 Last update: January 22, 2021

Table 7-7. Clinical Validity of FoundationOne CDx: Pan-Tumor Studies of FoundationOne ...... 156 Table 7-8. Clinical Validity of FoundationOne CDx: Tumor-Specific Studies of FoundationOne ...... 159 Table 7-9. Clinical Utility of FoundationOne CDx: Pan-Tumor Studies of FoundationOne ...... 163 Table 7-10. Clinical Utility of FoundationOne CDx: Tumor-Specific Studies of FoundationOne ...... 168 Table 7-11. Clinical Utility of NGS: Pan-Tumor Meta-Analyses and Studies of Non-FoundationOne Tests ...... 176

Figures Figure 1-1. Foundation Medicine Portfolio ...... 2 Figure 1-2. Estimated 2020 Distribution of Stage of Disease Within the Select Solid Tumor Cancers ...... 3 Figure 1-3. Response Rates (A) and Survival (B) With Genomically Matched Therapy vs Nonmatched Therapy Across Tumor Types...... 4 Figure 2-1. Estimated 2020 Distribution of Stage of Disease Within the Select Solid Tumor Cancers ..... 24 Figure 2-2. 5-Year Survival Rates in Select Metastatic Solid Tumorsa ...... 25 Figure 3-1. FoundationOne CDx Report Guide ...... 42 Figure 5-1. Spectrum of Genomic Alterations Across Tumor Types ...... 59 Figure 5-2. Prevalence of NTRK Gene Fusions by Tumor Type ...... 60 Figure 5-3. Overlap of Immunotherapy Biomarkers Among PD-L1 Tested Patients Across Tumor Types ...... 63 Figure 5-4. Overall Response Rate with Pembrolizumab in KEYNOTE-158 ...... 63 Figure 5-5. Correlation of Foundation Medicine TMB and WES (A), Reproducibility of Foundation Medicine TMB (B), and Correlation of Panel TMB and WES by DNA Sequence Amount (C) ...... 66 Figure 5-6. Prevalence of Actionable Genomic Alterations in NSCLC Patientsa233-239 ...... 70 Figure 5-7. Prevalence of Actionable Genomic Alterations in mCRC Patients ...... 83 Figure 5-8. Anti-EGFR Treatment Benefit for Tumors Without any RAS Mutations (All RAS Wild Type) ...... 84 Figure 5-9. Mean OS Curves, Created by Weighted Averaging of Digitized KM Survival Curves From Selected Clinical Trials of Melanoma Patients ...... 98 Figure 6-1. Comparison of Total Drug Treatment Costs Between Matched and Unmatched Therapy Among Patients Who Received CGP...... 107 Figure 7-1. FoundationOne CDx Sample Report ...... 149

CONFIDENTIAL PAGE 115 US-FDX-2000073 Last update: January 22, 2021

Terms and Definitions Actionable alteration A variation in DNA that is predicted to affect a patient’s response to treatment and, therefore, guides selection of therapies. Also known as actionable mutations. Advanced cancer Cancer that is classified as locally advanced or metastatic, which corresponds to the SEER classifications of regional and distant cancer, respectively. Cancer is classified as regional when it has spread beyond the primary site to nearby lymph nodes or organs and tissues. Cancer is classified as distant when it has spread from the primary site to distant organs or distant lymph nodes. Distant cancer is also described as metastatic or stage IV cancer. Allele frequency The incidence of a gene variant in a population. An allele frequency is calculated by dividing the number of times the allele of interest is observed in a population by the total number of copies of all the alleles at that particular genetic locus in the population. Alteration See “Genomic alteration.” Base pair Molecules called nucleotides, on opposite strands of the DNA double helix, which form chemical bonds with one another. These chemical bonds act like rungs in a ladder and hold the strands of DNA together. There are 4 nucleotides, or bases, of DNA: adenine (A), cytosine (C), guanine (G), and thymine (T). These bases form specific pairs (A with T, and G with C). Base pair substitution An alteration that exchanges one single base for another; the result could be a change in DNA sequence that substitutes one amino acid for another and may alter the resulting protein, no change in the amino acid sequence and thus no effect on the resulting protein, or a termination of the coding region resulting in a truncated protein. Biomarker A defined characteristic that is measured as an indicator of normal biological processes, pathogenic processes, or responses to an exposure or intervention. Molecular, histologic, radiographic, or physiologic characteristics are all types of biomarkers. Cancer of unknown A case in which cancer cells are found in the body, but the location primary where the cells first started growing (the origin or primary site) cannot be determined. Also called carcinoma of unknown primary and CUP. Companion diagnostic An FDA-approved medical device or test that provides information to support the safe and effective use of a corresponding drug or biological product. Complete response The disappearance of all signs of cancer in response to treatment (does not necessarily indicate cure).

CONFIDENTIAL PAGE 116 US-FDX-2000073 Last update: January 22, 2021

Comprehensive genomic A hybrid capture-based NGS platform that has been optimized to profiling identify all types of molecular alterations (single nucleotide variants, small and large indels, CNAs, and structural variations) in cancer-related genes in a single test using complex and often proprietary bioinformatics. CGP may also include testing for MSI and TMB. Concordance Agreement; in the context of FoundationOne CDx, concordance represents agreement between the results of FoundationOne CDx and other NGS-based tests or selected FDA-approved non-NGS companion diagnostic assays. Copy number alteration An alteration that results in a gain or loss in copies of sections of DNA. Distant cancer Cancer that has spread from the primary site to distant organs or distant lymph nodes. Distant cancer is also described as metastatic cancer or stage IV cancer. DNA sequencing A laboratory process used to determine the exact sequence (order) of the 4 building blocks, or bases, that make up DNA (identified by the letters A, C, G, and T; see “Base pair” for additional information). DNA sequencing can be used to find genomic alterations. Duration of response The time between the initial response to therapy and subsequent disease progression or relapse. First-line therapy The first therapeutic intervention for a disease. When used by itself, first-line therapy represents the accepted best treatment. Also called induction therapy, primary therapy, and primary treatment. Fluorescence in situ A laboratory technique used to look at genes or chromosomes in cells hybridization and tissues. Pieces of DNA that contain a fluorescent dye are made in the laboratory and added to cells or tissues on a glass slide. When these pieces of DNA bind to specific genes or areas of chromosomes on the slide, they are visible when viewed under a microscope sensitive to fluorescent light. Also called FISH. Formalin-fixed paraffin- Sample (tumor) tissue preserved and archived for future analysis; embedded tissue analysis may include RNA and DNA extraction. Gene rearrangement A large alteration of a chromosome or large chromosomal regions that can take the form of deletions, duplications, insertions, inversions, or translocations. Genomic alteration A change in DNA sequence; examples include base pair substitutions, insertion and deletion alterations, CNAs, and gene rearrangements. Genomic alterations can lead to proteins with abnormal levels of expression and/or function.

CONFIDENTIAL PAGE 117 US-FDX-2000073 Last update: January 22, 2021

Genomic loss of A profile of the percentage of the tumor genome that is under focal loss heterozygosity score of one allele; focal LOH events accumulate as genomic “scars” because of incorrect DNA double-strand break repair when the homologous recombination pathway is deficient. Hazard ratio A measure of how often a particular event happens in one group compared with how often it happens in another group, over time. In cancer research, hazard ratios are often used in clinical trials to compare survival and other dichotomous outcomes at any point in time between an experimental group of patients who have been assigned a specific treatment and a control group assigned a different treatment or placebo. A hazard ratio of 1.0 indicates no difference in outcomes between the groups. A hazard ratio >1 or <1 may indicate that the outcome was better in one of the groups. The confidence interval is used to measure the precision of the hazard ratio; if the confidence interval includes 1, then the hazard ratio is not significant. Homologous repair Abnormal functioning of the homologous DNA repair pathway; the deficiency FoundationOne CDx test defines HRD as tBRCA-positive and/or LOH high. Hotspot analysis In cancer, a hotspot analysis assesses specific alterations in prespecified regions of the gene known to be associated with the diagnosis, prognosis, or treatment of cancer (as opposed to sequencing the entire gene of interest). Hybridization capture A signal amplification method where an RNA probe is annealed to target DNA. Subsequently, a captured antibody binds the DNA-RNA hybrid to a solid surface. Immunohistochemistry A laboratory analysis that uses antibodies to test for certain antigens (markers) in a cell or tissue sample. The antibodies are usually linked to an enzyme or a fluorescent dye. When the antibodies bind to the antigen in the tissue sample, the enzyme or dye is activated, and the antigen can be visualized under a microscope. Immunohistochemistry is used to help diagnose diseases, such as cancer. Immunotherapy A type of treatment used frequently in patients with cancer that uses substances to stimulate or suppress the immune system. Includes cytokines, vaccines, and some monoclonal antibodies; therapies often target certain cells of the immune system although others affect the immune system more broadly. Insertion or deletion Alterations in which extra base pairs are inserted into a new place in the alteration (indel) DNA (insertions) or in which a section of DNA is removed (deletions).

CONFIDENTIAL PAGE 118 US-FDX-2000073 Last update: January 22, 2021

Laboratory-developed A medical procedure that involves testing a sample of blood, urine, or test other substance from the body. Laboratory tests can help to determine a diagnosis, plan treatment, monitor treatment efficacy, and/or monitor disease progression over time. These must be CLIA-certified but are not FDA-approved. Library construction The process through which DNA fragments of interest are cloned into vectors in order to be isolated for further study. Matched therapy Treatment matched to a patient’s genomic alteration(s) and/or profile. Metastatic cancer Cancer that has spread from the primary site (where the cancer started) to other places in the body. Microsatellite instability A change that occurs in the DNA of certain cells (such as tumor cells) in which the number of repeats of microsatellites (short, repeated sequences of DNA) is different than the number of repeats that was in the DNA when it was inherited. The cause of microsatellite instability may be a defect in the ability to repair mistakes made when DNA is copied in the cell. Mismatch repair Replacement of mismatched DNA base pairs by the enzyme DNA polymerase. Involves the removal of the incorrect base and replacement with the correct base. Negative percent The ability of a DNA test to correctly identify wild-type bases (ie, the agreement probability that the test will not identify a variant that is not present). Negative percent agreement reflects the frequency of false positives. Next-generation A DNA sequencing technology that allows hundreds of genes to be sequencing sequenced at the same time. In the oncology space, this technology is used to interrogate clinically relevant genes to identify 4 classes of actionable alterations: base substitutions, short indels, CNAs, and gene fusions. (For example, FoundationOne CDx utilizes hybrid capture- based NGS to detect alterations in a total of 324 genes and report genomic signatures including MSI and TMB). Objective response rate Objective response rate is defined as the proportion of patients with tumor size reduction of a predefined amount and for a minimum time period (ie, until documented tumor progression). Objective response equals the sum of confirmed partial responses and complete responses.

CONFIDENTIAL PAGE 119 US-FDX-2000073 Last update: January 22, 2021

Odds ratio A measure of the odds of an event happening in one group exposed to a potential risk factor compared to the odds of the same event happening in another group that has not been exposed to the potential risk factor. In cancer research, odds ratios are most often used in case-control (backward-looking) studies to determine if exposure to a potential risk factor increases the risk of cancer. An odds ratio of 1.0 means that both groups had the same odds of developing cancer regardless of their exposure to the potential risk factor. An odds ratio >1 may indicate that exposure to a risk factor could increase the odds of developing cancer whereas an odds ratio <1 may indicate that exposure could reduce the risk of cancer. The confidence interval is used to estimate the precision of the odds ratio; a large confidence interval indicates a low level of precision whereas a small confidence interval indicates a higher level of precision. The confidence interval does not report a measure’s statistical significance. Odds ratios are also known as relative odds. Overall survival The time from randomization until death from any cause. Overall survival rate The percentage of people in a study or treatment group who are still alive for a certain period of time after they were diagnosed with or started treatment for a disease, such as cancer. The OS rate is often stated as a 5-year survival rate, which is the percentage of people in a study or treatment group who are alive 5 years after their diagnosis or the start of treatment. Partial response A decrease in the size of a tumor, or the extent of cancer in the body, in response to treatment. Also called partial remission. Polymerase chain A laboratory method used to make many copies of a specific fragment of reaction DNA from a sample that contains very small amounts of that DNA. The method allows DNA to be amplified sufficiently to detect certain changes in a gene, such as a genomic alteration. Positive percent The proportion of non-reference standard positive subjects in whom the agreement new test is positive. Positive percent agreement reflects the frequency of false negatives. Precision medicine A form of medicine that uses information about a person’s genes, proteins, and environment to prevent, diagnose, and treat disease. In cancer, precision medicine can identify specific information from tumors to help diagnose patients, plan treatment, monitor treatment efficacy, and/or determine disease prognosis. Examples include using targeted therapies to treat specific types of cancer cells, such as HER2-positive breast cancer cells, or using tumor marker testing to help diagnose cancer. Predictive genomic Alterations that can directly match patients to a targeted treatment alterations pathway or clinical trial.

CONFIDENTIAL PAGE 120 US-FDX-2000073 Last update: January 22, 2021

Prognostic genomic Alterations that provide information on the anticipated course of disease, alterations including patient survival. Progression-free survival The time from randomization until objective tumor progression or death. The precise definition of tumor progression is important and should be carefully detailed in the protocol. Progressive disease Cancer that is growing, spreading, or getting worse. According to RECIST v1.1. criteria, disease progression occurs when there is a 20% increase in the sum of the longest diameter from nadir, 20% increase in the sum of diameters, and at least a 5 millimeter increase from nadir. Recurrence-free survival The length of time that the patient survives without any signs or symptoms of that cancer after treatment ends. Also known as disease- free survival. Recurrence rate The rate at which disease or medical condition recurs or returns (after treatment). Recurrent cancer Cancer that has recurred (come back), usually after a period of time during which the cancer could not be detected. The cancer may come back to the same place as the original (primary) tumor or to another place in the body. Regional cancer Cancer that has spread beyond the primary site to nearby lymph nodes or organs and tissues. Relative survival rate A method of comparing the survival of people who have a specific disease with those who do not over a certain period of time. This is often measured for 5 years from the date of diagnosis or the start of treatment and is calculated by dividing the percentage of patients with the disease who are still alive at the end of the period of time by the percentage of people in the general population of the same sex and age who are alive at the end of the same time period. The relative survival rate shows whether the disease shortens life. Response rate The percentage of patients whose cancer shrinks or disappears after treatment. Solid tumor An abnormal mass of tissue that usually does not contain cysts or liquid areas. Somatic mutation An alteration in DNA that occurs after conception. Somatic mutations can occur in any of the cells of the body except the germ cells (sperm and egg) and therefore are not passed on to children. These alterations can (but do not always) cause cancer or other diseases. Stable disease Cancer that is neither decreasing nor increasing in extent or severity. Stage IV cancer Cancer that has spread to distant parts of the body at the time that the patient is initially diagnosed with cancer.

CONFIDENTIAL PAGE 121 US-FDX-2000073 Last update: January 22, 2021

Targeted therapy In cancer, a method of treatment that uses drugs or other substances to identify and attack specific types of cancer cells while causing less harm to noncancer cells. Some targeted therapies block the action of certain enzymes, proteins, or other molecules involved in the growth and spread of cancer cells. Other types of targeted therapies help the immune system kill cancer cells or deliver toxic substances directly to cancer cells and kill them. Targeted therapy may have fewer adverse effects than other types of cancer treatment. Most targeted therapies are composed of either small-molecule drugs or monoclonal antibodies. Time to treatment failure The time from randomization to treatment discontinuation for any reason, including disease progression, treatment toxicity, patient preference, or death. Tumor mutational Calculated using the number of somatic base substitution or burden insertion/deletion alterations per megabase of the coding region after filtering to remove known somatic and deleterious mutations and by subsequently extrapolating that value to the exome or genome as a whole. Tumor mutational Often corresponds to a TMB ≥20 mutations per megabase. burden-high Tumor mutational Often corresponds to a TMB of 6–19 mutations per megabase. burden-intermediate Tumor mutational Often corresponds to a TMB of ≤5 mutations per megabase. burden-low Unmatched therapy In cancer, general treatment offered to a patient that is not based on genomic alterations, such as chemotherapy. Variant An alteration in the most common DNA nucleotide sequence. The term variant can be used to describe an alteration that may be benign, pathogenic, or of unknown significance.

CONFIDENTIAL PAGE 122 US-FDX-2000073 Last update: January 22, 2021

List of Abbreviations 1L first line 2L+ second line or higher AKT protein kinase B ALK anaplastic lymphoma kinase APC adenomatous polyposis coli ARID1A AT-rich interactive domain-containing protein BICR blinded independent central review BRCA breast cancer gene CB clinical benefit CDK cyclin-dependent kinase CDKN2A/B cyclin-dependent kinase inhibitor 2A/2B CGP comprehensive genomic profiling CI confidence interval CLIA Clinical Laboratory Improvement Amendments CMS Centers for Medicare & Medicaid Services CNA copy number alteration CNS central nervous system CPS combined positive score CR complete response CRC colorectal cancer CRPC castration-resistant prostate cancer CT computed tomography CTA clinical trial assay CUP cancer of unknown primary dMMR mismatch repair deficient DNA deoxyribonucleic acid DOR duration of response EGFR epidermal growth factor receptor ER estrogen receptor FAS full analysis set F1 FoundationOne F1CDx FoundationOne CDx FDA Food and Drug Administration FFPE formalin-fixed paraffin embedded FGF fibroblast growth factor FGFR fibroblast growth factor receptor FISH fluorescence in situ hybridization FLT1 FMS-related tyrosine kinase 1 FRS2 fibroblast growth factor receptor substrate 2 GBC gallbladder carcinoma

CONFIDENTIAL PAGE 123 US-FDX-2000073 Last update: January 22, 2021

GFR growth factor receptor HBOC hereditary breast and ovarian cancer HER2 human epidermal growth factor receptor 2 HR hazard ratio HRD homologous recombination deficient HRR homologous recombination repair IHC immunohistochemistry indel insertion and deletion alteration IQR interquartile range JAK Janus kinase KRAS V-Ki-ras2 Kirsten rat sarcoma LDT laboratory-developed test LOH loss of heterozygosity MAPK mitogen-activated protein kinase Mb megabase MLL2 lysine methyltransferase 2D MMR mismatch repair MSI microsatellite instability MSI-H microsatellite instability-high MTB molecular tumor board mTOR molecular target of rapamycin mut mutation NA not applicable NCCN National Comprehensive Cancer Network NCD national coverage determination NF1 neurofibromin 1 NGS next-generation sequencing NNT number needed to test NOS not otherwise specified NPA negative percent agreement NPV negative predictive value NR not reported NSCLC non-small cell lung cancer NTRK neurotrophic receptor tyrosine kinase OPA overall percent agreement OR odds ratio ORR objective response rate OS overall survival PARP poly adenosine diphosphate ribose polymerase PCR polymerase chain reaction PD progressive disease

CONFIDENTIAL PAGE 124 US-FDX-2000073 Last update: January 22, 2021

PD-1 programmed cell death 1 PD-L1 programmed death-ligand 1 PFS progression-free survival PI3K phosphatidylinositol-4,5-bisphosphate 3-kinase PIK3CA phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha PMPM per-member per-month PPA positive percent agreement PPPM per-patient per-month PPV positive predictive value PR partial response PTEN phosphatase and tensin homolog RB1 retinoblastoma-1 RECIST Response Evaluation Criteria in Solid Tumors RFS recurrence-free survival RR response rate SD stable disease SEER Surveillance, Epidemiology, and End Results SETD2 SET domain containing 2 SNV single nucleotide variant TAPUR Targeted Agent and Profiling Utilization Registry tBRCA tumor BRCA TERT telomerase reverse transcriptase TMB tumor mutational burden TMD transmembrane domain TNBC Triple-negative breast cancer TP53 tumor protein P53 TSG tumor suppressor gene TTF time to treatment failure USD United States dollars VEGF vascular endothelial growth factor VUS variants of unknown significance WT wild-type

CONFIDENTIAL PAGE 125 US-FDX-2000073 Last update: January 22, 2021

NCCN Guidelines Recommendations for Molecular Testing Table 7-1 reviews the NCCN Guidelines recommendations pertaining to molecular testing across solid tumor types. Table 7-2 provides an overview of the NCCN Guidelines recommendations pertaining to biomarker-recommended therapies and which of these therapies require a companion diagnostic; this table also provides alignment for which of these biomarker-directed therapies FoundationOne CDx is the companion diagnostic.

Table 7-1. NCCN Guidelines Recommendations for Molecular Testing Within Select Solid Tumor Cancer Types and Relevant Foundation Medicine Testing Tumor type NCCN recommendationsa FoundationOne CDx and applicable alignment NCCN Guidelines for tumor profiling Tumors for which FoundationOne CDx has an FDA-approved companion diagnostic claim Breast cancerb HER2 testing at diagnosis and of a metastatic site at FDA-approved to report NCCN progression ERBB2/HER2 amplification and Guidelines for Asses for BRCA1/2 germline mutations in all patients as a companion Breast Cancer with recurrent or metastatic breast cancer to identify diagnostic for trastuzumab, ado- V.6.202027 candidates for PARP inhibitor therapy. While olaparib trastuzumab-emtansine, or and talazoparib are FDA-indicated in HER2 negative pertuzumab disease, the panel supports use in any breast cancer FDA-approved to report subtype associated with a germline BRCA1 or BRCA2 alterations in PIK3CA and as a mutation companion diagnostic for For stage IV or recurrent TNBC, assess PD-L1 biomarker alpelisib status on tumor-infiltrating immune cells to identify FDA-approved for detection of patients most likely to benefit from atezolizumab plus NTRK1/2/3 gene fusions for use albumin-bound paclitaxel as a companion diagnostic for For stage IV or recurrent breast cancer, assess for larotrectinib across solid tumors PIK3CA mutation with tumor or liquid biopsy if hormone For patients with breast cancer, receptor-positive/HER2-negative and if considering PD-L1 testing is performed with therapy with alpelisib. PIK3CA mutation testing can be the Ventana PD-L1 (SP142) done on tumor tissue or ctDNA in peripheral blood companion diagnostic IHC (liquid biopsy). If liquid biopsy is negative, tumor tissue assay. Ventana PD-L1 (SP142) testing is recommended is FDA-approved as a For stage IV or recurrent breast cancer, larotrectinib and companion diagnostic to identify entrectinib are FDA-approved therapies useful in certain advanced, metastatic TNBC circumstances for NTRK gene fusion positive patients patients who may respond to the immune checkpoint inhibitor For stage IV or recurrent breast cancer, pembrolizumab is therapy (eg, atezolizumab), and useful in certain circumstances for MSI-H/dMMR tumors can be ordered through Systemic Therapy for ER- and/or PR-Positive Recurrent Foundation Medicine along with or Stage IV (M1) Disease preferred option for HER2- CGP test negative and postmenopausal patients or premenopausal patients receiving ovarian ablation or suppression following progression on or after an endocrine-based regimen: Fulvestrant + alpelisib for PIK3CA-mutated

CONFIDENTIAL PAGE 126 US-FDX-2000073 Last update: January 22, 2021

tumors as a second-line or subsequent treatment (Category 1) CRC All patients with metastatic CRC should have tumor FDA-approved to report NCCN tissue genotyped for RAS (KRAS and NRAS) and BRAF alterations in KRAS and NRAS Guidelines for mutations, and HER2 amplifications individually or as and as companion Colon Cancer part of an NGS panel. If known RAS/RAF mutation, diagnostic for cetuximab or V.1.202130 HER2 testing is not indicated. NGS panels have the panitumumab ability to pick up rare and actionable mutations and NCCN FDA-approved for detection of fusions. All patients with CRC should be tested for MMR Guidelines for NTRK1/2/3 gene fusions for use or MSI at diagnosis. MSI may be accomplished with a Rectal Cancer as a companion diagnostic for validated NGS panel, especially in patients with V.1.202139 larotrectinib across solid tumors metastatic disease who require genotyping FoundationOne CDx is also able of RAS and BRAF. to detect alterations in BRAF, NTRK fusions are extremely rare in CRC and data HER2 amplification, and support limiting the subpopulation of CRC that should be genomic signatures including tested for NTRK fusions to those with wild-type KRAS, MSI NRAS, BRAF, and arguably to those that are dMMR/MSI-H. If testing is done, selection of the appropriate assay for NTRK fusion detection depends on tumor type and genes involved, as well as consideration of other factors such as available material, accessibility of various clinical assays, and whether comprehensive genomic testing is needed concurrently. Endometrial All endometrial cancers should be tested for DNA FDA-approved for detection of cancerb mismatch repair or MSI NTRK1/2/3 gene fusions for use NCCN Clinically significant molecular subgroups with differing as a companion diagnostic for Guidelines for prognoses include POLE mutations, MSI-H, copy larotrectinib across solid tumors Uterine number low, and copy number high. Ancillary studies for FDA-approved for the detection Neoplasms POLE mutations, MMR/MSI, and aberrant p53 of the TMB-H (≥10 mut/Mb) for V.1.202140 expression are encouraged to complement morphologic use of pembrolizumab across assessment of histologic tumor type solid tumors Consider NTRK gene fusion testing for metastatic or FoundationOne CDx is able to recurrent endometrial carcinoma detect ERBB2/HER2 Consider TMB testing through a validated and/or FDA- amplification and is FDA- approved assay approved as a companion diagnostic for such in breast HER2 IHC testing (with reflex to FISH for equivocal cancer; it is also able to detect IHC) is recommended for possible treatment of advanced alterations in POLE and genomic stage or recurrent serous endometrial carcinoma signatures such as MSI The FoundationOne CDx clinical report lists clinical trials for which the patient may be eligible based on the results of testing. Foundation Medicine proactively contacts healthcare providers when a patient qualifies for NCI-MATCH based on the genomic alteration(s) identified

CONFIDENTIAL PAGE 127 US-FDX-2000073 Last update: January 22, 2021

Gastric, If patients with unresectable locally advanced, locally FoundationOne CDx is an FDA- esophageal, and recurrent, or metastatic gastric, esophageal, or approved companion diagnostic esophagogastric esophagogastric cancers it is recommended to perform for detection of NTRK1/2/3 gene junction cancers HER2, PD-L1, MSI by PCR/MMR by IHC (if not fusions for use of larotrectinib NCCN previously done) if metastatic adenocarcinoma is across solid tumors. Guidelines for documented or suspected It is able to detect ERBB2/HER2 Gastric Cancer NGS: At present, 3 targeted therapeutic agents, amplification and is FDA- V.4.202033 trastuzumab, ramucirumab, and pembrolizumab, have approved as a companion NCCN been approved by the FDA for use in gastric, esophageal, diagnostic for such in breast Guidelines for and esophagogastric junction cancers. Trastuzumab is cancer, and it is able to detect Esophageal and based on testing for HER2 positivity. Pembrolizumab is genomic signatures such as MSI Esophagogastric based on MSI by PCR and MMR by IHC, and PD-L1 The FoundationOne CDx clinical Junction expression by CPS. The FDA granted approval for the report lists clinical trials for use of select TRK inhibitors for NTRK gene fusion- Cancers which the patient may be eligible V.5.202032 positive solid tumors. When limited tissue is available for testing, sequential testing of single biomarkers or use of based on the results of testing. limited molecular diagnostic panels may quickly exhaust Foundation Medicine proactively the sample. In these scenarios, comprehensive genomic contacts healthcare providers profiling via a validated NGS assay performed in a when a patient qualifies for NCI- CLIA-approved laboratory may be used for the MATCH based on the genomic identification of HER2 amplification, MSI, and NTRK alteration(s) identified gene fusions. It should be noted that NGS has several inherent limitations and thus whenever possible, the use of gold-standard assays (IHC/FISH/targeted PCR) should be performed For patients who are unable to undergo a traditional biopsy, liquid biopsy testing using a validated NGS- based comprehensive genomic profiling assay performed in a CLIA-approved laboratory may be considered. A negative result should be interpreted with caution as this does not exclude the presence of tumor mutations or amplifications Hepatobiliary Unresectable or metastatic gallbladder cancer: FoundationOne CDx is able to cancer For unresectable or metastatic gallbladder cancer, detect FGFR2 fusions and select NCCN dMMR/MSI testing is recommended. For patients with rearrangements and is FDA- Guidelines for dMMR/MSI-H tumors or a family history suggestive of approved as a companion Hepatobiliary BRCA1/2 mutations, consider germline testing diagnostic for pemigatinib Cancer Consider additional molecular testing in patients with FDA-approved for detection of V.5.202034 unresectable or metastatic gallbladder cancer; this testing NTRK1/2/3 gene fusions as a may include NTRK gene fusion testing companion diagnostic for Unresectable or metastatic intra- or extra-hepatic larotrectinib across solid tumors cholangiocarcinoma: FoundationOne CDx is also able For unresectable or metastatic intrahepatic to detect IDH1 mutations cholangiocarcinoma or extrahepatic cholangiocarcinoma, The FoundationOne CDx clinical dMMR/MSI testing is recommended. For patients with report lists clinical trials for dMMR/MSI-H tumors or a family history suggestive of which the patient may be eligible BRCA1/2 mutations, consider germline testing based on the results of Consider additional molecular testing in patients with testing. Foundation Medicine unresectable or metastatic intrahepatic proactively contacts healthcare providers when a patient qualifies for NCI-MATCH based

CONFIDENTIAL PAGE 128 US-FDX-2000073 Last update: January 22, 2021

cholangiocarcinoma or extrahepatic cholangiocarcinoma; on the genomic alteration(s) this testing may include NTRK gene fusion testing identified Testing for FGFR2 fusions or rearrangements and IDH1 mutations to determine targeted treatment for patients with progression on a preferred first-line regimen For patients with unresectable or metastatic biliary tract cancers the following therapies are useful in certain circumstances: larotrectinib or entrectinib for NTRK gene fusion positive; pembrolizumab for MSI-H/dMMR; pemigatinib for FGFR2 fusions or rearrangements; ivosidenib or IDH1 mutations Hepatocellular carcinoma: Larotrectinib and entrectinib are treatment options for patients with hepatocellular carcinoma that is NTRK gene fusion positive Cutaneous Stage III: FDA-approved to detect melanoma BRAF mutation testing is recommended for patients with alterations in BRAF, as a NCCN stage III for whom future BRAF-directed therapy may be companion diagnostic in Guidelines for an option melanoma for dabrafenib or vemurafenib or trametinib Melanoma: Stage IV or clinical recurrence: Cutaneous or cobimetinib in combination Obtain tissue to ascertain alterations in BRAF and in the V.1.202131 with vemurafenib appropriate clinical setting, KIT, from either biopsy of FDA-approved for detection of the metastasis (preferred) or archival material if the NTRK1/2/3 gene fusions as a patient is being considered for targeted therapy. Consider companion diagnostic for broader genomic profiling (eg, larger NGS panels, BRAF larotrectinib across solid tumors non-V600 mutations) if the test results might guide future treatment decisions or eligibility for participation in a FoundationOne CDx is also able clinical trial to detect alterations in NRAS and KIT and rearrangements ALK, If BRAF single-gene testing was the initial test performed and ROS1, across 324 genes and is negative, clinicians should strongly consider larger NGS panels to identify other potential genetic targets (eg, The FoundationOne CDx clinical KIT, BRAF non-V600) report lists clinical trials for which the patient may be eligible The NCCN Guidelines describe the specific implications based on the results of of testing for BRAF, NRAS, and KIT mutations and less testing. Foundation Medicine common mutations such as fusions in NTRK1, NTRK2, proactively contacts healthcare NTRK3, ALK, and ROS1, as well as the emerging role of providers when a patient tumor mutation burden. Fusions in NTRK1, NTRK2, and qualifies for NCI-MATCH based NTRK3 correspond to a high response rate to TRK on the genomic alteration(s) inhibitors larotrectinib or entrectinib.102,315 Fusions in identified ALK and ROS1 may predispose to activity from inhibitors of these genes (eg, crizotinib, entrectinib)315 NSCLC The NCCN NSCLC Guidelines Panel recommends FDA-approved to report NCCN molecular testing and strongly advises broad molecular alterations in EGFR, ALK, MET Guidelines for profiling in eligible patients with metastatic NSCLC with exon 14 skipping, and BRAF and NSCLC the goal of identifying rare driver mutations for which as a companion diagnostic V.2.202135 effective drugs may already be available or to for afatinib, gefitinib, erlotinib, appropriately counsel patients regarding the availability osimertinib, alectinib, crizotinib, of clinical trials.c Broad molecular profiling is a key ceritinib, capmatinib, component of the improvement of care of patients with or dabrafenib in combination NSCLC Testing should include EGFR (Category 1), ALK with trametinib in NSCLC

CONFIDENTIAL PAGE 129 US-FDX-2000073 Last update: January 22, 2021

(Category 1), ROS1, BRAF, NTRK1/2/3 gene fusions, FDA-approved for detection of MET exon 14 skipping mutations, and RET. Emerging NTRK1/2/3 gene fusions as a biomarkers to identify novel therapies include high-level companion diagnostic for MET amplification and ERBB2. KRAS testing may larotrectinib across solid tumors identify patient who are unlikely to benefit from further FoundationOne CDx is also able molecular testing to detect alterations in ROS1, To minimize tissue use and potential wastage, NCCN KRAS, MET, RET, and ERBB2 recommends broad molecular profiling which includes a across 324 genes minimum of the following: EGFR, ALK, ROS1, BRAF, The FoundationOne CDx clinical MET exon 14-skipping mutation, NTRK1/2/3, and RET report lists clinical trials for rearrangements which the patient may be eligible If there is insufficient tissue to allow testing for all of based on the results of EGFR, ALK, ROS1, BRAF, MET exon 14 skipping, testing. Foundation Medicine NTRK1/2/3, and RET repeat biopsy and/or plasma testing proactively contacts healthcare should be done providers when a patient The use of cell-free/circulating tumor DNA can be qualifies for NCI-MATCH based considered in specific clinical circumstances, most on the genomic alteration(s) notably if a patient is medically unfit for invasive tissue identified sampling; or, if following pathologic confirmation of a NSCLC diagnosis, there is insufficient material for molecular analysis Ovarian cancerb Both somatic and germline BRCA1/BRCA2 testing is FDA-approved to detect NCCN recommended at diagnosis for patients with alterations in BRCA1/2 and as a Guidelines for pathologically confirmed epithelial ovarian companion diagnostic in ovarian Ovarian Cancer cancer/fallopian tube cancer/primary peritoneal cancer. cancer for rucaparib and olaparib V.1.202036 Germline and/or somatic status informs maintenance FDA-approved to report therapy genomic LOH from FFPE Tumor molecular testing is recommended prior to ovarian tumor tissue. Positive initiation of therapy for persistent/recurrent disease. HRD status (defined as tBRCA Validated molecular testing should be performed in a positive and/or LOH high) in CLIA-approved facility using the most recent available ovarian cancer patients is tumor tissue. Testing recommended to include at least: associated with improved PFS BRCA1/2 (via NGS) and MSI or DNA mismatch repair if from rucaparib maintenance not previously done. Evaluation of homologous therapy in accordance with the recombination deficiency can be considered. Additional Rubraca product label somatic tumor testing can be considered at the FoundationOne CDx is also able physician’s discretion to identify genetic alterations for to detect alterations in other which FDA-approved tumor-specific or tumor-agnostic homologous recombination targeted therapy options exist pathway genes (eg, ATM, BRIP1, Tumor molecular testing can be considered if not CHEK2, FANCA, FANCL, previously done at diagnosis of less common ovarian FANCM, NBN, RAD51C, cancer (LCOC) histologies. Due to emerging therapeutics RAD51D, and RAD54L) for less common ovarian cancers, there is value in The FoundationOne CDx clinical identifying potential pathways for rare cancers and it may report lists clinical trials for be useful for clinical trial recruitment. There are limited which the patient may be eligible data in these cancers given their infrequency and it will based on the results of be difficult to acquire prospective data. Individualized testing. Foundation Medicine treatment may be the best treatment for these rare tumors proactively contacts healthcare providers when a patient qualifies for NCI-MATCH based

CONFIDENTIAL PAGE 130 US-FDX-2000073 Last update: January 22, 2021

on the genomic alteration(s) identified Prostate cancer Tumor testing is recommended for HRR gene mutations FDA-approved to detect HRR NCCN (HRRm) in all men diagnosed with metastatic prostate gene (BRCA1, BRCA2, ATM, Guidelines for cancer, including BRCA1, BRCA2, ATM, PALB2, BARD1, BRIP1, CDK12, Prostate Cancer FANCA, RAD51D, CHEK2, CDK12 and consider testing CHEK1, CHEK2, FANCL, V.3.202038 for HRRm in men diagnosed with regional prostate PALB2, RAD51B, RAD51C, cancer. At present, this information may be used for RAD51D and RAD54L) genetic counseling, early use of platinum chemotherapy, olaparib, and/or eligibility for clinical trials (e.g., PARP alterations and as a companion inhibitors) diagnostic for olaparib Clinical trials may include additional candidate DNA FDA-approved to report MSI, repair genes under investigation as molecular biomarkers BRCA1, BRCA2, ATM, PALB2, If mutations in BRCA2, BRCA1, ATM, CHEK2, or FANCA, RAD51D, and CHEK2 PALB2 are found and/or there is a strong family history The FoundationOne CDx clinical of cancer, refer to genetic counseling to assess for the report lists clinical trials for possibility of HBOC which the patient may be eligible Somatic testing may require repetition when prostate based on the results of cancer progresses after treatment testing. Foundation Medicine Tumor testing for MSI or dMMR is recommended for all proactively contacts healthcare men with metastatic CRPC and can be considered for providers when a patient men with castrate-naïve metastatic or regional prostate qualifies for NCI-MATCH based cancer. DNA analysis for MSI and IHC for MMR are on the genomic alteration(s) different assays measuring the same biological effect. If identified MSI is used, testing using an NGS assay validated for prostate cancer is preferred. If MSI-H or dMMR is found, refer to genetic counseling to assess for the possibility of Lynch syndrome. MSI-H or dMMR indicate eligibility for pembrolizumab in later lines of treatment for CRPC Olaparib is a treatment option for patients with mCRPC and a pathogenic mutation (germline and/or somatic) in a homologous recombination repair gene (BRCA1, BRCA2, ATM, BARD1, BRIP1, CDK12, CHEK1, CHEK2, FANCL, PALB2, RAD51B, RAD51C, RAD51D, or RAD54L), who have been treated with androgen receptor-directed therapy. Patients with PPP2R2A mutations in the PROfound trial experienced an unfavorable risk-benefit profile. Therefore, olaparib is not recommended in patients with a PPP2R2A mutations Rucaparib is a treatment option for patients with mCRPC and a pathogenic BRCA1 or BRCA2 mutation (germline and/or somatic) who have been treated with androgen receptor-directed therapy and a taxane-based chemotherapy. If the patient is not fit for chemotherapy, rucaparib can be considered even if taxane-based therapy has not been given Vulvar cancer In patients with advanced or recurrent/metastatic disease, FDA-approved for detection of NCCN biomarker therapy for second-line treatment is NTRK1/2/3 gene fusions as a Guidelines for recommended for patients with TMB-H tumors, PD-L1 companion diagnostic for Vulvar Cancer positive tumors, or MSI-H/dMMR tumors larotrectinib across solid tumors

CONFIDENTIAL PAGE 131 US-FDX-2000073 Last update: January 22, 2021

V.2.2021 (pembrolizumab) or patients with NTRK gene fusion FDA-approved for the detection positive tumors (larotrectinib or entrectinib; category 2B) of the TMB-H (≥10 mut/Mb) for use of pembrolizumab across solid tumors FoundationOne CDx is also able to detect the genomic signature MSI-H The FoundationOne CDx clinical report lists clinical trials for which the patient may be eligible based on the results of testing. Foundation Medicine proactively contacts healthcare providers when a patient qualifies for NCI-MATCH based on the genomic alteration(s) identified Tumors for which FoundationOne CDx is an approved test without a companion diagnostic claim Bladder cancer The panel recommends that molecular/genomic testing FoundationOne CDx is able to NCCN be performed for stages IVA and IVB bladder cancer and detect alterations in FGFR2 and Guidelines for may be considered for stage IIIB. Testing should be FGFR3 Bladder Cancer carried out early, ideally at diagnosis of advanced bladder The FoundationOne CDx clinical V.6.202028 cancer, in order to facilitate treatment decision-making report lists clinical trials for and to prevent delays in administering later lines of which the patient may be eligible therapy. In addition to determining eligibility for FDA- based on the results of testing. approved therapies, molecular/genomic testing may be Foundation Medicine proactively used to screen for clinical eligibility contacts healthcare providers Molecular/genomic testing (including testing for FGFR2 when a patient qualifies for NCI- or FGFR3 alterations) is recommended for stage IVA MATCH based on the genomic and IVB and should be considered for stage IIIB. alteration(s) identified Glioblastoma There are no identified targeted agents with demonstrated The FoundationOne CDx clinical NCCN efficacy in glioblastoma. However, molecular testing of report lists driver mutations and Guidelines for tumors is encouraged because if a driver mutation is clinical trials for which the CNS Cancers detected, it may be reasonable to treat with a targeted patient may be eligible based on V.3.202029 therapy on a compassionate use basis and/or the patient the results of testing. Foundation may have more treatment options in the context of a Medicine proactively contacts clinical trial. Molecular testing also has a valuable role in healthcare providers when a improving diagnostic accuracy and prognostic patient qualifies for NCI- stratification that may inform treatment selection MATCH based on the genomic alteration(s) identified Pancreatic Tumor/somatic gene profiling is recommended for FoundationOne CDx is able to cancer patients with locally advanced/metastatic disease who are detect alterations in ALK, NTRK, NCCN candidates for anti-cancer therapy to identify uncommon ROS1 gene fusions; mutations in Guidelines for mutations. Consider specifically testing for actionable BRAF, BRCA1/2, HER2, KRAS, Pancreatic somatic findings including, but not limited to fusions PALB2, and MSI Cancer (ALK, NRG1, NTRK, ROS1), mutations (BRAF, The FoundationOne CDx clinical V.1.202137 BRCA1/2, HER2, KRAS, PALB2), and MMR deficiency report lists driver mutations and (detected by tumor IHC, PCR, or NGS) clinical trials for which the patient may be eligible based on the results of testing. Foundation

CONFIDENTIAL PAGE 132 US-FDX-2000073 Last update: January 22, 2021

Testing on tumor tissue is preferred; however, cell-free Medicine proactively contacts DNA testing can be considered if tumor tissue testing is healthcare providers when a not feasible patient qualifies for NCI- MATCH based on the genomic alteration(s) identified CLIA, Clinical Laboratory Improvement Amendments; CNS, central nervous system; CPS, combined positive score; CRC, colorectal cancer; CRPC, castration-resistant prostate cancer; DNA, deoxyribonucleic acid; ER, estrogen receptor; FDA, Food and Drug Administration; FFPE, formalin-fixed, paraffin-embedded; HBOC, hereditary breast and ovarian cancer; HRD, homologous recombination deficiency; IHC, immunohistochemistry; LOH, loss of heterozygosity; MSI, microsatellite instability; MMR, mismatch repair; NCCN, National Comprehensive Cancer Network; NGS, next-generation sequencing; NSCLC, non-small cell lung cancer; NTRK, neurotrophic receptor tyrosine kinase; PFS, progression-free survival; tBRCA, tumor BRCA; TMB, tumor mutational burden; TNBC, triple-negative breast cancer. a All NCCN recommendations are category 2A unless otherwise indicated. Additionally, NCCN states that the best management of any patient with cancer is in a clinical trial. b Data are specific to women. Sources: Foundation Medicine, Inc. (2020)1, Hempelman et al. (2017)316, Swisher et al. (2017)108.

Table 7-2. Biomarker-based Targeted Therapies and Immunotherapies Recommended in NCCN Guidelinesa in Selected Cancers along with Companion Diagnostics Does Is a CDx Foundatio required n Medicine per FDA have a CDx approved claimc for Genomic Drug recommended NCCN drug this Cancer Type alteration by NCCN recommendationb label? alteration? Metastatic NSCLCc ALK Alectinib Category 1; preferred Y Y (1L) Ceritinib Category 1 (1L) Y Y Brigatinib Category 1; preferred Nd --- (1L) Crizotinib Category 1 (1L) Y Y Lorlatinib Category 1; preferred Nd -- (1L) Lorlatinib Category 2A (2L or Nd --- 3L)e BRAF Dabrafenib + Category 2A; Y/Y Y f V600E trametinib preferred (1L or 2L) Vemurafenib Category 2A (1L) NAg --- EGFR Osimertinib Category 1; preferred Y Y (1L)

CONFIDENTIAL PAGE 133 US-FDX-2000073 Last update: January 22, 2021

Does Is a CDx Foundatio required n Medicine per FDA have a CDx approved claimc for Genomic Drug recommended NCCN drug this Cancer Type alteration by NCCN recommendationb label? alteration? Erlotinib Category 1 (1L) Y Y Afatinib Category 1 (1L) Y Y Dacomitinib Category 1 (1L) Y N Gefitinib Category 1 (1L) Y Y EGFR T790M Osimertinib Category 1 (2L)e Y Y ROS-1 Crizotinib Category 2A; Y N preferred (1L) Entrectinib Category 2A; Nd --- preferred (1L) Entrectinib Category 2A (2L) Nd --- Ceritinib Category 2A (1L) NAg --- Lorlatinib Category 2A (2L) NAg --- NTRK1/2/3 Larotrectinib Preferred (1L); Nd Y Category 2A (2L)h Entrectinib Preferred (1L); Nd --- Category 2A (2L)h MET exon 14 Capmatinib Category 2A; Y Y skipping preferred (1L or 2L)i Crizotinib Category 2A (1L or NAg --- 2L)i RET Selpercatinib Category 2A; Nd --- preferred (1L or 2L)j Pralsetinib Category 2A; Y N preferred (1L or 2L)j Cabozantinib Category 2A (1L or NAg --- 2L)j Vandetanib Category 2B (1L) NAg --- PD-L1 ≥50% Pembrolizumab ± Category 1; preferred Y (PD- N and EGFR, platinum-based (1L) L1)k ALK, ROS1, chemotherapy

CONFIDENTIAL PAGE 134 US-FDX-2000073 Last update: January 22, 2021

Does Is a CDx Foundatio required n Medicine per FDA have a CDx approved claimc for Genomic Drug recommended NCCN drug this Cancer Type alteration by NCCN recommendationb label? alteration? BRAF, MET Atezolizumab Category 1; preferred Y (PD- N exon 14 (1L) L1) skipping l m mutation, Atezolizumab + Category 1 (1L) N --- NTRK1/2/3, platinum-based and RET chemotherapy + VEGF negative inhibitor Atezolizumab + Category 2A (1L) Nm --- platinum-based chemotherapy Nivolumab + Category 1 (1L) Y (PD- N ipilimumab L1) Nivolumab + Category 2A (1L) Y (PD- N ipilimumab + L1) platinum-based chemotherapy PD-L1 Pembrolizumab Category 2B (1L) Y (PD- N ≥1%─49% and L1)k EGFR, ALK, Pembrolizumab + Category 1 (1L); Nk --- ROS1, BRAF, platinum-based preferred MET exon 14 chemotherapy skipping mutation, Atezolizumab + Category 1l (1L) Nm --- NTRK1/2/3, platinum-based and RET chemotherapy + VEGF negative inhibitor Atezolizumab + Category 2A (1L) Nm --- platinum-based chemotherapy Nivolumab + Category 1 (1L) Y (PD- N ipilimumab L1) Nivolumab + Category 2A (1L) Y (PD- N ipilimumab + L1) platinum-based chemotherapy

CONFIDENTIAL PAGE 135 US-FDX-2000073 Last update: January 22, 2021

Does Is a CDx Foundatio required n Medicine per FDA have a CDx approved claimc for Genomic Drug recommended NCCN drug this Cancer Type alteration by NCCN recommendationb label? alteration? PD-L1 <1% Pembrolizumab + Category 1 (1L); N --- and EGFR, platinum-based preferred ALK, ROS1, chemotherapy BRAF, MET Atezolizumab + Category 1l (1L) Nm --- exon 14 platinum-based skipping chemotherapy + VEGF mutation, inhibitor NTRK1/2/3, and RET Atezolizumab + Category 2A (1L) N --- negative platinum-based chemotherapy Nivolumab + Category 2A (1L) Nn --- ipilimumab ± platinum-based chemotherapy Emerging Biomarkers High-level Crizotinib Category 2A NAg --- MET Capmatinib Category 2A NAg --- amplification ERBB2 Ado-trastuzumab Category 2A NAg --- (HER2) emtansine Fam-trastuzumab Category 2A NAg --- deruxtecan-nxki Breast Cancer ERBB2 Pertuzumab + Category 1 (with Yo Y (HER2) trastuzumab + taxane docetaxel), preferred; Category 2A (with paclitaxel), preferred (1L+) Tucatinib + Category 1 (2L+) N --- trastuzumab + capecitabine Ado-trastuzumab Category 2A (1L+) Y Y emtansine (TDM-1) Fam-trastuzumab Category 2A (3L+) Nd --- deruxtecan-nxki

CONFIDENTIAL PAGE 136 US-FDX-2000073 Last update: January 22, 2021

Does Is a CDx Foundatio required n Medicine per FDA have a CDx approved claimc for Genomic Drug recommended NCCN drug this Cancer Type alteration by NCCN recommendationb label? alteration? Trastuzumab + Category 2A (1L+) Y Y chemotherapyp Lapatinib + Category 2A (1L+) N --- capecitabine Trastuzumab + Category 2A (1L+) Yo Y lapatinib (without cytotoxic therapy) Neratinib + Category 2A (1L+) N --- capecitabine BRCA1/BRCA2 Olaparib Category 1; preferred Y Y (1L+) Talazoparib Category 1; preferred Y Y (1L+) PIK3CA Alpelisib + fulvestrant Category 1; preferred Y Y (HR- 2L therapy (2L+) positive/HER-2 negative) PD-L1 ≥1% on Atezolizumab + Category 2A; Y N tumor albumin-bound preferred (1L+) infiltrating paclitaxel immune cells (HR- negative/HER- 2 negative) NTRK Larotrectinib Category 2A (2L+) Nd Y Entrectinib Category 2A (2L+) Nd --- MSI-H/dMMR Pembrolizumab Category 2A (2L+) Nd --- CRC KRAS/NRAS/ Cetuximab + Category 2A (1L) Y Y BRAF WT and (FOLFOX or left-sided FOLFIRIr) tumor onlyq Panitumumab + Category 2A (1L) Y Y (FOLFOX or FOLFIRIr )

CONFIDENTIAL PAGE 137 US-FDX-2000073 Last update: January 22, 2021

Does Is a CDx Foundatio required n Medicine per FDA have a CDx approved claimc for Genomic Drug recommended NCCN drug this Cancer Type alteration by NCCN recommendationb label? alteration? Cetuximab or Category 2B (1L; Y Y panitumumab only recommended in patients not appropriate for intensive therapy) KRAS/NRAS/ Cetuximab + Category 2A (2L+) Y Y BRAF WTq (FOLFOX or FOLFIRIr ) Panitumumab + Category 2A (2L+) Y Y (FOLFOX or FOLFIRIr ) Irinotecanr + Category 2A (2L+) Y Y cetuximab or panitumumabf,s BRAF V600E Encorafenib + Category 2A (2L) Y/NAg N (Cetuximab or Panitumumab)t HER2- Trastuzumabu + Category 2A (1L NAg/NAg N amplified and (Pertuzumab or [only recommended RAS and BRAF Lapatinib)v or fam- in patients not WT trastuzumab appropriate for deruxtecan-nkxiw intensive therapy] and subsequentx therapy) dMMR/MSI-H Pembrolizumaby,z Category 2A (1L and Nd N only subsequentx therapy) Nivolumaby,z Category 2A (1L Nd N [only recommended in patients not appropriate for intensive therapy] and subsequentx therapy) Nivolumab + Category 2B (1L Nd/Nd N ipilimumaby,z [only recommended in patients not

CONFIDENTIAL PAGE 138 US-FDX-2000073 Last update: January 22, 2021

Does Is a CDx Foundatio required n Medicine per FDA have a CDx approved claimc for Genomic Drug recommended NCCN drug this Cancer Type alteration by NCCN recommendationb label? alteration? appropriate for intensive therapy] and subsequentx therapy) NTRK gene Larotrectinib Category 2A (2L+) Nd Y fusion positive Entrectinib Category 2A (2L+) Nd --- Ovarian Canceraa HRDbb Niraparib Category 2A (4L+) Y Y Deleterious BRCA1/2 Olaparib Category 2A (3L+) Y Y (germline) mutation Deleterious BRCA1/2 (germline Rucaparib Category 2A (3L+) Y Y and/or somatic) mutation NTRK gene Entrectinib Category 2A Nd --- fusion positive Larotrectinib Category 2A Nd Y tumors MSI-H/dMMR Pembrolizumab Category 2A Nd --- solid tumors Prostate Cancercc BRCA1/2mdd Rucaparib Category 2A (2L and Y Y subsequent therapy) HRRmee Olaparib Category 1/2Bff (2L Y Y and subsequent therapy) MSI-H/dMMR Pembrolizumab Category 2B (2L and Nd --- subsequent therapy) Melanoma: BRAF V600- Dabrafenib + Category 1 (1L Y/Y Y Cutaneousgg activating trametinib systemic therapy); hh,ii,jj mutation Category 2A (2L or subsequentkk systemic therapy)

CONFIDENTIAL PAGE 139 US-FDX-2000073 Last update: January 22, 2021

Does Is a CDx Foundatio required n Medicine per FDA have a CDx approved claimc for Genomic Drug recommended NCCN drug this Cancer Type alteration by NCCN recommendationb label? alteration? Vemurafenib + Category 1 (1L Y/N Y cobimetinib systemic therapy); Category 2A (2L or subsequentkk systemic therapy) Encorafenib + Category 1 (1L Y/Y N binimetinib systemic therapy); Category 2A (2L or subsequentkk systemic therapy)

BRAF V600- Vemurafenib + Category 2A (1L Y/N/NAg Y activating cobimetinib + systemic therapy) mutation atezolizumabll KIT-activating Imatinib Category 2A (2L or NAg --- mutation in subsequent systemic tumor therapy)kk NTRK gene Larotrectinib Category 2A (2L or Nd Y fusion-positive subsequent systemic tumors therapy)kk Entrectinib Category 2A (2L or Nd --- subsequent systemic therapy)kk NRAS-mutated Binimetinibmm,nn Category 2B (2L or NAg N tumors subsequent systemic therapy)kk Cholangiocarcinomao FGFR2 fusions Pemigatinib Category 2A (2L) Y Y o or select rearrangements IDH1 Ivosidenib Category 2A (2L) NAg N NTRK Entrectinib Category 2A (1L and Nd --- subsequent therapy) Larotrectinib Category 2A (1L and Nd Y subsequent therapy)

CONFIDENTIAL PAGE 140 US-FDX-2000073 Last update: January 22, 2021

Does Is a CDx Foundatio required n Medicine per FDA have a CDx approved claimc for Genomic Drug recommended NCCN drug this Cancer Type alteration by NCCN recommendationb label? alteration? MSI-H/dMMR Pembrolizumabpp Category 2A (1L and Nd --- subsequent therapy) a Individual guidelines contain differing recommendations for extent of molecular testing; please refer to the individual guidelines at NCCN.org for information on individual cancers by site. b Category 1: Based upon high-level evidence (eg, phase 3 randomized trials), there is uniform NCCN consensus (>85%) that the intervention is appropriate. Category 2A: Based upon lower-level evidence (eg, phase 2 trials), there is uniform NCCN consensus that the intervention is appropriate. Category 2B: Based upon lower-level evidence, there is NCCN consensus (50%-85%) that the intervention is appropriate. All NCCN recommendations are category 2A unless otherwise indicated. c The NCCN Guidelines recommend biomarker testing in eligible patients with metastatic NSCLC and strongly advise broad molecular profiling to identify actionable biomarkers, including rare oncogenic driver variants, for which effective therapy may be available. The NCCN Guidelines for NSCLC provide recommendations for individual biomarkers that should be tested and recommend testing techniques but do not endorse any specific commercially available biomarker assays. d This indication is approved under accelerated approval based on surrogate endpoint(s) (eg, event-free survival, objective response rate, complete response, PFS or time to progression). Continued approval for this indication may be contingent upon verification and description of clinical benefit in the confirmatory trials. e Depending on 1L and 2L therapy and type of progression. f If not previously given. g Not FDA-approved for this indication. h If NTRK1/2/3 inhibitors not used 1L. i If MET exon 14 skipping mutation inhibitors not used 1L. j If RET inhibitors not given 1L. k Pembrolizumab is approved for the treatment of NSCLC in combination with platinum-based chemotherapy in patients whose tumors do not harbor EGFR or ALK genomic alterations; no testing is required for PD-L1 for pembrolizumab in combination with platinum-based chemotherapy. Pembrolizumab is also approved as a single agent for tumors expressing PD-L1 (tumor proportion score ≥1%) as determined by an FDA-approved test, with no EGFR or ALK genomic alterations or disease progression of FDA-approved therapies for these aberrations prior to pembrolizumab. l Atezolizumab in combination with platinum-based chemotherapy ± VEGF inhibitor is only recommended for patients with adenocarcinoma, large cell, or NSCLC NOS. Carboplatin + paclitaxel + bevacizumab + atezolizumab is Category 1; carboplatin + albumin-bound paclitaxel + atezolizumab is Category 2A. m Atezolizumab is approved for the treatment of NSCLC in combination with platinum-based chemotherapy in patients whose tumors do not harbor EGFR or ALK genomic alterations; no testing is required for PD-L1 for atezolizumab in combination with platinum-based chemotherapy. Atezolizumab is also approved as a single agent for tumors with a high PD-L1 expression (PD-L1 stained ≥50% of tumor cells [TC ≥50%] or PD-L1 stained tumor-infiltrating immune cells [IC] covering ≥10% of the tumor area [IC ≥10%]) as determined by an FDA-approved test, with no EGFR or ALK genomic alterations or disease progression of FDA- approved therapies for these aberrations prior to atezolizumab. n Nivolumab + ipilimumab is approved regardless of PD-L1 expression in adult patients with metastatic or recurrent NSCLC with no EGFR or ALK genomic tumor aberrations as first-line treatment in combination with 2 cycles of platinum-doublet chemotherapy.

CONFIDENTIAL PAGE 141 US-FDX-2000073 Last update: January 22, 2021 o Within this regimen, trastuzumab is the only therapy that requires an FDA approved companion diagnostic test. p Chemotherapy to be used in combination with trastuzumab includes paclitaxel ± carboplatin, docetaxel, vinorelbine, and capecitabine. Trastuzumab given in combination with an anthracycline is associated with significant cardiac toxicity. Concurrent use of trastuzumab and pertuzumab with an anthracycline should be avoided. Trastuzumab may be safely combined with all non- anthracycline containing preferred and other single agents for recurrent or metastatic breast cancer. q The panel defines the left side of the colon as splenic flexure to rectum. Evidence suggests that patients with tumors originating on the right side of the colon (hepatic flexure through cecum) are unlikely to respond to cetuximab and panitumumab in first-line therapy for metastatic disease. Data on the response to cetuximab and panitumumab in patients with primary tumors originating in the transverse colon (hepatic flexure to splenic flexure) are lacking. r Irinotecan should be used with caution in patients with Gilbert’s disease or elevated serum bilirubin. There is a commercially available test for UGT1A1. Guidelines for use in clinical practice have not been established. s Cetuximab or panitumumab are recommended in combination with irinotecan-based therapy or as a single-agent therapy for patients who cannot tolerate irinotecan. t In the second-line setting for BRAF V600E mutation positive tumors, there is phase 3 evidence for better efficacy with targeted therapies over FOLFIRI. u An FDA-approved biosimilar is an appropriate substitute for trastuzumab. v If no previous treatment with HER2 inhibitor. w Some activity was seen after a previous HER2-targeted regimen. May not be indicated in patients with underlying lung issues due to lung toxicity (2.6% report of deaths from interstitial lung disease). x If no previous treatment with a checkpoint inhibitor. y These therapies are FDA approved for colorectal cancer that has progressed following treatment with fluoropyrimidine, oxaliplatin, and irinotecan. However, a number of patients in the clinical trials had not received all three prior systemic therapies. Thirty-seven percent of patients received nivolumab monotherapy and 24% received ipilimumab/nivolumab combination therapy in first- or second-line, and 28% and 31% of patients had not received all three prior therapies before treatment with nivolumab or ipilimumab/nivolumab, respectively. z If disease response, consider discontinuing checkpoint inhibitor after 2 years of treatment. aa This includes both platinum-sensitive recurrent ovarian cancer and platinum-resistant recurrent ovarian cancer. Biomarker- based recommendations in other settings are not included in this table. bb HRD defined by either: 1) a deleterious or suspected deleterious BRCA mutation; or 2) genomic instability and progression >6 months after response to the last platinum-based chemotherapy. cc All systemic therapies for subsequent treatment are Category 2B if visceral metastases are present. dd BRCA1/2 mutations and HRRm refer to germline and/or somatic pathogenic mutations. ee HRRm in prostate cancer includes germline and/or somatic pathogenic mutations in the following genes: BRCA1, BRCA2, ATM, BARD1, BRIP1, CDK12, CHEK1, CHEK2, FANCL, PALB2, RAD51B, RAD51C, RAD51D, and RAD54L. ff Olaparib is Category 1 recommendation for second-line treatment following first-line abiraterone/enzalutamide in patients with HRRm tumors; olaparib is Category 2B for second-line treatment following first-line docetaxel in patients with HRRm tumors; olaparib is Category 1 for subsequent treatment. gg This table includes only biomarker-based options for unresectable or metastatic disease; biomarker-based options in other settings are omitted. hh Positive VE1 IHC results are sufficient for starting targeted therapy in patients who are symptomatic or have rapidly progressing disease. Confirmatory BRAF molecular testing is encouraged. ii In previously untreated patients with unresectable AJCC 7th edition stage IIIC or stage IV disease, BRAF/MEK inhibitor combination therapy was associated with improved response rate, PFS, and OS compared to BRAF inhibitor monotherapy. jj If BRAF/MEK inhibitor combination therapy is contraindicated, BRAF-inhibitor monotherapy is an option, especially in patients who are not appropriate candidates for checkpoint immunotherapy\ kk For patients who experience progression of melanoma during or shortly after first-line therapy, consider second-line agents if not used first-line and if from a different class. For patients who progressed on single agent anti-PD-1 checkpoint immunotherapy, anti- PD-1/ipilimumab combination therapy or ipilimumab monotherapy is a reasonable treatment option. For patients who experience

CONFIDENTIAL PAGE 142 US-FDX-2000073 Last update: January 22, 2021 disease control (CR, PR, SD) and have no residual toxicity, but subsequently experience disease progression/relapse >3 months after treatment discontinuation, re-induction with the same agent or same class of agent may be considered. ll In a randomized, double-blind, placebo-controlled, phase 3 trial, the addition of atezolizumab to vemurafenib and cobimetinib was associated with longer median PFS and longer duration of response; however, the triplet induced more toxicity than the vemurafenib/cobimetinib doublet. Until mature OS data are published, it is not clear that the triplet regimen is preferred over sequential BRAF/MEK inhibitor therapy followed by PD-L1 or PD-1 inhibition. mm Binimetinib for NRAS-mutated tumors that have progressed after prior immune checkpoint inhibitor therapy. nn In patients who were previously untreated or who had prior failure of immunotherapy, binimetinib was associated with a response rate of 15%, and demonstrated a modest improvement in PFS with no improvement in OS compared with single-agent dacarbazine. oo For second-line therapies, treatment selection depends on clinical factors including previous treatment regimen/agent and extent of liver function pp There are limited clinical trial data to support pembrolizumab in this setting. Sicklick JK, Leyland-Jones B, Kato S, et al. Personalized, molecularly matched combination therapies for treatment-naïve, lethal malignancies: the I-PREDICT Study. J Clin Oncol 2017;35:2512. ALK, anaplastic lymphoma kinase; BRAF, v-raf murine sarcoma viral oncogene homolog B1; BRCA, breast cancer susceptibility gene; CDx, companion diagnostic; CR, complete response; dMMR, DNA mismatch repair; EGFR, epidermal growth factor receptor; ERBB2, Erb-B2 receptor tyrosine kinase 2; FDA, Food and Drug Administration; FGFR, fibroblast growth factor receptor; HER, human epidermal growth factor; HER2, human epidermal growth factor receptor 2; HR, hormone receptor; HRR, homologous recombination repair; HRRm, homologous recombination repair mutation; IDH, Isocitrate dehydrogenase; KIT, v-kit Hardy-Zuckerman 4 feline sarcoma viral oncogene homolog; KRAS, Kirsten rat sarcoma viral oncogene homolog; MET, mesenchymal epithelial transition factor receptor; MSI-H, microsatellite instability-high; NA, not applicable; NCCN, National Comprehensive Cancer Network; N, no; NRAS, neuroblastoma ras viral oncogene homolog; NTRK, neurotrophic tyrosine kinase receptor; PD-L1, programmed death-ligand 1; PIK3CA, phosphatidylinositol-4,5-bisphosphate 3- kinase catalytic subunit alpha; PR, partial response; RAS, rat sarcoma; RET, ret proto-oncogene; ROS-1, c-ros oncogene 1; SD, stable disease; TMB, tumor mutational burden; WT, wild type; Y, yes. Referenced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Non-Small Cell Lung Cancer V.2.2021. © National Comprehensive Cancer Network, Inc. 2020. All rights reserved. Accessed December 22, 2020. To view the most recent and complete version of the NCCN Guidelines, go online to NCCN.org. Referenced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Breast Cancer V.6.2020. © National Comprehensive Cancer Network, Inc. 2020. All rights reserved. Accessed December 22, 2020. To view the most recent and complete version of the NCCN Guidelines, go online to NCCN.org. Referenced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Colon Cancer V.4.2020. © National Comprehensive Cancer Network, Inc. 2020. All rights reserved. Accessed December 22, 2020. To view the most recent and complete version of the NCCN Guidelines, go online to NCCN.org Referenced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Rectal Cancer V.6.2020. © National Comprehensive Cancer Network, Inc. 2020. All rights reserved. Accessed December 22, 2020. To view the most recent and complete version of the NCCN Guidelines, go online to NCCN.org Referenced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Ovarian Cancer V.1.2020. © National Comprehensive Cancer Network, Inc. 2020. All rights reserved. Accessed December 22, 2020. To view the most recent and complete version of the NCCN Guidelines, go online to NCCN.org. Referenced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Prostate Cancer V.3.2020. © National Comprehensive Cancer Network, Inc. 2020. All rights reserved. Accessed December 22, 2020. To view the most recent and complete version of the NCCN Guidelines, go online to NCCN.org Referenced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Melanoma: Cutaneous V.1.2021. © National Comprehensive Cancer Network, Inc. 2020. All rights reserved. Accessed December 22, 2020. To view the most recent and complete version of the NCCN Guidelines, go online to NCCN.org Referenced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Hepatobiliary Cancers V.5.2020. © 2020 National Comprehensive Cancer Network, Inc. All rights reserved. The NCCN Guidelines® and illustrations herein may not be reproduced in any form for any purpose without the express written permission of the NCCN. To view the most recent and complete version of the NCCN Guidelines, go online to NCCN.org

CONFIDENTIAL PAGE 143 US-FDX-2000073 Last update: January 22, 2021

Sources: Alecensa PI 201980; Alunbrig PI 2019317; Braftovi PI 2018318; Cabometyx PI 2020319; Caprelsa PI 2018320; Cotellic PI 201881; Enhertu PI 2019321; Erbitux PI 201982; Gilotrif PI 201983; Gleevec PI 2020322; Herceptin PI 201884; Iressa PI 201985; Kadcyla PI 2013323; Keytruda PI 202043; Lorbrena PI 2019324; Lynparza PI 2020325; Mekinist PI 202086; Mektovi PI 2020326; NCCN Breast Cancer V.6.202027; NCCN Colon Cancer V.1.202130; NCCN Rectal Cancer V.1.202139; NCCN Melanoma: Cutaneous V.1.202131; NCCN Hepatobiliary V.5.202034; NCCN NSCLC V.2.202135; NCCN Ovarian Cancer V.1.202036; NCCN Prostate Cancer V.3.202038; Nerlynx PI 2020327; Opdivo PI 2020328; Pemazyre PI 2020329; Perjeta PI 201287; Piqray PI 201988; Retevmo PI 2020330; Rozlytrek PI 2019190; Rubraca PI 2020331;Tabrecta PI 2020332; Tafinlar PI 202089; Tagrisso PI 201990; Talzenna PI 2018333; Tarceva PI 201991; Tecentriq PI 2020334; Tibsovo PI 2020335; Tukysa PI 2020336; Tykerb PI 2018337; Vectibix PI 201792; Vitrakvi PI 2019189;Vizimpro PI 2018338; Xalkori PI 201993; Yervoy PI 2020339; Zejula PI 2020340; Zelboraf PI 202094; Zykadia PI 201995.

CONFIDENTIAL PAGE 144 US-FDX-2000073 Last update: January 22, 2021

FoundationOne Portfolio Description of Decision Support Services As shown in Table 7-3, Foundation Medicine provides a portfolio of CGP tests and services to ensure patient access to genomic insights regardless of cancer type or specimen type.

Table 7-3. FoundationOne Portfolio

FoundationOne® FoundationOne® FoundationOne® IHC CDx Liquid CDx Heme FDA- FDA-approved CDx FDA-approved CDx – FDA-approved approved for 23 targeted for 7 targeted therapies CDx for 2 CDx claims therapies immunotherapies Target All solid tumors Liquid biopsy Hematologic Specific solid tumor types (ctDNA): all solid malignancies, tumors tumors sarcomas (soft tissue + bone) Number of 324 (DNA) 324 (DNA)a 406 (DNA) - genes 265 (RNA) analyzed Genomic • TMB • bTMB • TMB • PD-L1 b signatures/ • MSI • MSI-H • MSI biomarkers • Tumor fraction FFPE tissue Peripheral whole • FFPE tissue FFPE tissue blood • Bone marrow aspirate • Peripheral whole blood 16 USS + 1 H&E slide

c or Specimen 1 FFPE block 10 USS or 2 tubes (8.5mL each) or 2.5 mL bone marrow of peripheral whole 4 USS 1 blockd + 1 H&E aspirate blood slide or 1 filled EDTA tube + 2.5 mL Paxgene tube peripheral whole blood Report Point mutations, Point mutations, Point mutations, TPS and/or features insertions/deletions, insertions/deletions, insertions/deletions, CPS for copy number copy number copy number approved/validated alterations, select alterations, select alterations, tumor types rearrangements rearrangements rearrangements Typical <2 weeks About 9 days 2 weeks 5 days turnaround timee

CONFIDENTIAL PAGE 145 US-FDX-2000073 Last update: January 22, 2021

a FoundationOne Liquid CDx is FDA-approved to report substitutions and indels in 311 genes, including rearrangements in ALK and BRCA1/2 and copy number alterations in BRCA1/2 and ERBB2 (HER2). Comprehensive results across all 324 genes are reported as a laboratory professional service which is not reviewed or approved by the FDA. bTMB, MSI-H status, and tumor fraction are reported as a laboratory professional service which is not reviewed or approved by the FDA. b MSI status will be reported for samples determined to have high microsatellite instability. c For full details, refer to specimen instructions at www.foundationmedicine.com. d FFPE is preferred. e Based on typical turnaround time from receipt of specimen. bTMB, blood tumor mutational burden; CDx, companion diagnostic; CPS, combined positive score; ctDNA, circulating tumor DNA; DNA, deoxyribonucleic acid; EDTA, ethylenediaminetetraacetic acid; FFPE, formalin-fixed paraffin embedded; H&E, hematoxylin and eosin; mL, milliliter; MSI, microsatellite instability; MSI-H – microsatellite instability-high; PD-L1, programmed death ligand-1; TMB, tumor mutational burden; TPS, tumor proportion score. Source: Foundation Medicine, Inc.

Foundation Medicine’s Services go “beyond the test” by providing a clear, in-depth report that supports clinical decision making by: . Providing insights on the patient’s genomic profile and associated targeted therapies, immunotherapies, and relevant clinical trials . Highlighting important disease-relevant genes with no reportable alterations identified . Highlighting genomic alterations associated with potential resistance to therapy to help rule out potentially ineffective treatment. Please refer to the sample report (Figure 7-1) for an example of this clinical decision support. In addition to the in-depth report, Foundation Medicine helps offers decision support services and technology solutions to help streamline patient care, including: . FoundationINSIGHTS™ . Clinical trial matching . Digital access . Electronic medical record (EMR) interfacing . MD Case Consultant Program . Molecular tumor boards (>30 nationwide)

FoundationOne CDx Product Description FoundationOne CDx employs a single DNA extraction method from routine FFPE biopsy or surgical resection specimens, 50-1000 ng of which will undergo whole-genome shotgun library construction and hybridization-based capture of all coding exons from 309 cancer-related genes, one promoter region, one non-coding (ncRNA), and select intronic regions from 34 commonly rearranged genes, 21 of which also include the coding exons (refer to Table 7-4 and Table 7-5 for the complete list of genes included in FoundationOne CDx).

Table 7-4. List of Genes With Full Coding Exonic Regions Included in the FoundationOne CDx Assay for the Detection of Substitutions, Indels, and CNAs ABL1 BRAF CDKN1A EPHA3 FGFR4 IKZF1 MCL1 NKX2-1 PMS2 RNF43 TET2 ACVR1B BRCA1 CDKN1B EPHB1 FH INPP4B MDM2 NOTCH1 POLD1 ROS1 TGFBR2 AKT1 BRCA2 CDKN2A EPHB4 FLCN IRF2 MDM4 NOTCH2 POLE RPTOR TIPARP

CONFIDENTIAL PAGE 146 US-FDX-2000073 Last update: January 22, 2021

AKT2 BRD4 CDKN2B ERBB2 FLT1 IRF4 MED12 NOTCH3 PPARG SDHA TNFAIP3 AKT3 BRIP1 CDKN2C ERBB3 FLT3 IRS2 MEF2B NPM1 PPP2R1A SDHB TNFRSF14 ALK BTG1 CEBPA ERBB4 FOXL2 JAK1 MEN1 NRAS PPP2R2A SDHC TP53 ALOX12B BTG2 CHEK1 ERCC4 FUBP1 JAK2 MERTK NT5C2 PRDM1 SDHD TSC1 AMER1 BTK CHEK2 ERG GABRA6 JAK3 MET NTRK1 PRKAR1A SETD2 TSC2 APC C11orf30 CIC ERRFI1 GATA3 JUN MITF NTRK2 PRKCI SF3B1 TYRO3 AR CALR CREBBP ESR1 GATA4 KDM5A MKNK1 NTRK3 PTCH1 SGK1 U2AF1 ARAF CARD11 CRKL EZH2 GATA6 KDM5C MLH1 P2RY8 PTEN SMAD2 VEGFA ARFRP1 CASP8 CSF1R FAM46C GID4 KDM6A MPL PALB2 PTPN11 SMAD4 VHL (C17orf39) ARID1A CBFB CSF3R FANCA GNA11 KDR MRE11A PARK2 PTPRO SMARCA4 WHSC1 ASXL1 CBL CTCF FANCC GNA13 KEAP1 MSH2 PARP1 QKI SMARCB1 WHSC1L1 ATM CCND1 CTNNA1 FANCG GNAQ KEL MSH3 PARP2 RAC1 SMO WT1 ATR CCND2 CTNNB1 FANCL GNAS KIT MSH6 PARP3 RAD21 SNCAIP XPO1 ATRX CCND3 CUL3 FAS GRM3 KLHL6 MST1R PAX5 RAD51 SOCS1 XRCC2 AURKA CCNE1 CUL4A FBXW7 GSK3B KMT2A MTAP PBRM1 RAD51B SOX2 ZNF217 (MLL) AURKB CD22 CXCR4 FGF10 H3F3A KMT2D MTOR PDCD1 RAD51C SOX9 ZNF703 (MLL2) AXIN1 CD274 CYP17A1 FGF12 HDAC1 KRAS MUTYH PDCD1LG2 RAD51D SPEN AXL CD70 DAXX FGF14 HGF LTK MYC PDGFRA RAD52 SPOP BAP1 CD79A DDR1 FGF19 HNF1A LYN MYCL PDGFRB RAD54L SRC BARD1 CD79B DDR2 FGF23 HRAS MAF MYCN PDK1 RAF1 STAG2 BCL2 CDC73 DIS3 FGF3 HSD3B1 MAP2K1 MYD88 PIK3C2B RARA STAT3 BCL2L1 CDH1 DNMT3A FGF4 ID3 MAP2K2 NBN PIK3C2G RB1 STK11 BCL2L2 CDK12 DOT1L FGF6 IDH1 MAP2K4 NF1 PIK3CA RBM10 SUFU BCL6 CDK4 EED FGFR1 IDH2 MAP3K1 NF2 PIK3CB REL SYK BCOR CDK6 EGFR FGFR2 IGF1R MAP3K13 NFE2L2 PIK3R1 RET TBX3 BCORL1 CDK8 EP300 FGFR3 IKBKE MAPK1 NFKBIA PIM1 RICTOR TEK CNA, copy number alteration. Source: FoundationOne CDx Label1

CONFIDENTIAL PAGE 147 US-FDX-2000073 Last update: January 22, 2021

Table 7-5. List of Genes With Select Intronic Regions Included in the FoundationOne CDx assay for the Detection of Gene Rearrangements, a Promoter Region, and an ncRNA Gene ALK BRCA1 ETV4 EZR KIT MYC NUTM1 RET SLC34A2 introns introns 2, intron 8 introns intron 16 intron 1 intron 1 introns intron 4 18, 19 7, 8, 12, 9–11 7–11 16, 19, 20 BCL2 BRCA2 ETV5 FGFR1 KMT2A NOTCH2 PDGFRA ROS1 TERCb 3’UTR intron 2 introns 6, introns 1, (MLL) intron 26 introns 7, introns 7 5, 17 introns 9, 11 31–35 6–11 BCR CD74 ETV6 FGFR2 MSH2 NTRK1 RAF1 RSPO2 TERTc introns introns introns 5, introns 1, intron 5 introns 8– introns 4– intron 1 8, 13, 14 6–8 6a 17 11 8 BRAF EGFR EWSR1 FGFR3 MYB NTRK2 RARA SDC4 TMPRSS2 introns introns 7, introns intron 17 intron 14 intron 12 intron 2 intron 2 introns 1–3 7–10 15, 24– 7–13 27 a ETV6 is common rearrangement partner for NTRK3. b TERC is a ncRNA. c The promoter region of TERT interrogated. ncRNA, non-coding ribonucleic acid. Source: FoundationOne CDx Label1

CONFIDENTIAL PAGE 148 US-FDX-2000073 Last update: January 22, 2021

FoundationOne CDx Sample Report

Figure 7-1. FoundationOne CDx Sample Report

CONFIDENTIAL PAGE 149 US-FDX-2000073 Last update: January 22, 2021

CONFIDENTIAL PAGE 150 US-FDX-2000073 Last update: January 22, 2021

CONFIDENTIAL PAGE 151 US-FDX-2000073 Last update: January 22, 2021

CONFIDENTIAL PAGE 152 US-FDX-2000073 Last update: January 22, 2021

Analytic Validity of FoundationOne CDx In addition to the large concordance study described in Analytic Validity of FoundationOne CDx, summaries of additional concordance studies between FoundationOne CDx and FDA-approved non-NGS companion diagnostic assays, as well as an externally validated NGS laboratory-developed test (LDT) assay, are summarized below.

Concordance of FoundationOne CDx to Externally Validated NGS Assays A concordance study between FoundationOne CDx and an externally validated NGS LDT assay (UW OncoPlex™) was also conducted. Genomic alterations across 157 overlapping genes detected by FoundationOne CDx and UW OncoPlex were compared across 188 samples covering 46 different tumor types. UW OncoPlex was set as the reference test and PPA, positive predictive value (PPV), NPA, and negative predictive value (NPV) between the 2 tests were calculated. FoundationOne CDx consistently showed high PPA, NPA, PPV, and NPV with UW OncoPlex for short variants, base substitutions, and indels. Results are summarized in Table 7-6.

Table 7-6. Concordance Summary for Short Variants (Including VUS) Between FoundationOne CDx and UW OncoPlex Genomic F1CDx+/ F1CDx- F1CDx+ F1CDx- PPA, % NPA, % alteration UW+, n /UW+, n /UW-, n /UW-, n (95% CI) (95% CI) All short variants 1,282 73 375 284,218 94.6 99.9 (93.3–95.8) (99.9–99.9) Substitutions 1,111 39 334 242,540 96.6 99.9 (95.4–97.6) (99.8–99.9) Indels 171 34 41 41,678 83.4 99.9 (77.6–88.2) (99.9–99.9) PIK3CA 53 0 0 48 100.0 100.0 substitutions in (93.3–100.0) (92.6–100.0) breast cancer NTRK1, NTRK2, 89 3 14 520 98.9 99.9 NTRK3 fusions (97.8–100.0) (99.9–99.9) CI, confidence interval; F1CDx, FoundationOne CDx; indel, insertion/deletion alteration; NPA, negative percent agreement; NPV, negative predictive value; NTRK, neurotrophic receptor tyrosine kinase; PIK3CA, phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha; PPA, positive percent agreement; PPV, positive predictive value; VUS, variants of unknown significance; UW, UW OncoPlex.

CONFIDENTIAL PAGE 153 US-FDX-2000073 Last update: January 22, 2021

Clinical Validity of FoundationOne CDx FoundationOne CDx as a companion diagnostic test for 23 drug therapies in 7 tumor types; further, FoundationOne CDx is a companion diagnostic for TMB high across all solid tumors. The evidence supporting the clinical validity of FoundationOne CDx for these companion diagnostic indications can be found in Evidence of Improved Clinical Outcomes With FoundationOne CDx.

Additional Studies Supporting Clinical Validity of FoundationOne CDx by Tumor Type

NSCLC . Ali SM, Hensing T, Schrock AB, et al. Comprehensive genomic profiling identifies a subset of crizotinib-responsive ALK-rearranged non-small cell lung cancer not detected by fluorescence in situ hybridization. Oncologist. 2016;21(6):762-770 . Lin JJ, Zhu VW, Yoda S, et al. Impact of EML4-ALK variant on resistance mechanisms and clinical outcomes in ALK-positive lung cancer. J Clin Oncol. 2018;36(12):1199-1206. . Ou SI, Schrock AB, Bocharov EV, et al. HER2 transmembrane domain (TMD) mutations (V659/G660) that stabilize homo- and heterodimerization are rare oncogenic drivers in lung adenocarcinoma that respond to afatinib. J Thorac Oncol. 2017;12(3):446-457. . Pekar-Zlotin M, Hirsch FR, Soussan-Gutman L, et al. Fluorescence in situ hybridization, immunohistochemistry, and next-generation sequencing for detection of EML4-ALK rearrangement in lung cancer. Oncologist. 2015;20(3):316-322. . Qin A, Johnson A, Ross JS, et al. Detection of known and novel FGFR fusions in non-small cell lung cancer by comprehensive genomic profiling. J Thorac Oncol. 2019;14(1):54-62. . Riess JW, Gandara DR, Frampton GM, et al. Diverse EGFR exon 20 insertions and co-occurring molecular alterations identified by comprehensive genomic profiling of NSCLC. J Thorac Oncol. 2018;13(10):1560-1568. . Schrock AB, Frampton GM, Herndon D, et al. Comprehensive genomic profiling identifies frequent drug-sensitive EGFR exon 19 deletions in NSCLC not identified by prior molecular testing. Clin Cancer Res. 2016;22(13):3281-3285. . Schrock AB, Frampton GM, Suh J, et al. Characterization of 298 patients with lung cancer harboring MET exon 14 skipping alterations. J Thorac Oncol. 2016;11(9):1493-1502. . Schrock AB, Li SD, Frampton GM, et al. Pulmonary sarcomatoid carcinomas commonly harbor either potentially targetable genomic alterations or high tumor mutational burden as observed by comprehensive genomic profiling. J Thorac Oncol. 2017;12(6):932-942 . Schrock AB, Zhu VW, Hsieh WS, et al. Receptor tyrosine kinase fusions and BRAF kinase fusions are rare but actionable resistance mechanisms to EGFR tyrosine kinase inhibitors. J Thorac Oncol. 2018;13(9):1312-1323. . Suh JH, Schrock AB, Johnson A, et al. Hybrid capture-based comprehensive genomic profiling identifies lung cancer patients with well-characterized sensitizing epidermal growth factor receptor point mutations that were not detected by standard of care testing. Oncologist. 2018;23(7):776-781. . Vaishnavi A, Capelletti M, Le AT, et al. Oncogenic and drug-sensitive NTRK1 rearrangements in lung cancer. Nat Med. 2013;19(11):1469-1472. Breast cancer . Levin MK, Wang K, Yelensky R, et al. Genomic alterations in DNA repair and chromatin remodeling genes in estrogen receptor-positive metastatic breast cancer patients with exceptional responses to capecitabine. Cancer Med. 2015;4(8):1289-1293.

CONFIDENTIAL PAGE 154 US-FDX-2000073 Last update: January 22, 2021

. Paratala BS, Chung JH, Williams CB, et al. RET rearrangements are actionable alterations in breast cancer. Nat Commun. 2018;9(1):4821. Colorectal cancer . Fabrizio DA, George TJ, Dunne RF, et al. Beyond microsatellite testing: assessment of tumor mutational burden identifies subsets of colorectal cancer who may respond to immune checkpoint inhibition. J Gastrointest Oncol. 2018;9(4):610-617. . Pietrantonio F, Di nicolantonio F, Schrock AB, et al. ALK, ROS1, and NTRK rearrangements in metastatic colorectal cancer. J Natl Cancer Inst. 2017;109(12). . Pietrantonio F, Di nicolantonio F, Schrock AB, et al. RET fusions in a small subset of advanced colorectal cancers at risk of being neglected. Ann Oncol. 2018;29(6):1394-1401. . Ross JS, Fakih M, Ali SM, et al. Targeting HER2 in colorectal cancer: the landscape of amplification and short variant mutations in ERBB2 and ERBB3. Cancer. 2018;124(7):1358- 1373. Melanoma . Boussemart L, Nelson A, Wong M, et al. Hybrid capture-based genomic profiling identifies BRAF V600 and Non-V600 alterations in melanoma samples negative by prior testing. Oncologist. 2019;24(5):657-663. . Johnson DB, Frampton GM, Rioth MJ, et al. Targeted next generation sequencing identifies markers of response to PD-1 blockade. Cancer Immunol Res. 2016;4(11):959-967. Ovarian cancer . Elvin J, He Y, Odunsi K, et al. Comprehensive genomic profiling (CGP) with loss of heterozygosity (LOH) to identify therapeutically relevant subsets of ovarian cancer (OC) [abstract 5512]. J Clin Oncol. 2017;35(suppl). . Elvin JA, Chura J, Gay LM, Markman M. Comprehensive genomic profiling (CGP) of ovarian clear cell carcinomas (OCCC) identifies clinically relevant genomic alterations (CRGA) and targeted therapy options. Gynecol Oncol Rep. 2017;20:62-66. . Kondrashova O, Nguyen M, Shield-Artin K, et al. Secondary somatic mutations restoring RAD51C and RAD51D associated with acquired resistance to the PARP inhibitor rucaparib in high-grade ovarian carcinoma. Cancer Discov. 2017;7(9):984-998. . Ross JS, Ali SM, Wang K, et al. Comprehensive genomic profiling of epithelial ovarian cancer by next generation sequencing-based diagnostic assay reveals new routes to targeted therapies. Gynecologic Oncol. 2013;130(3):554-559.

CONFIDENTIAL PAGE 155 US-FDX-2000073 Last update: January 22, 2021

Clinical Validity of Previous Versions of Foundation Medicine Tissue Assay The clinical validity of study results associated with FoundationOne are presented in Table 7-7 for pan-tumor studies and Table 7-8 for tumor- specific studies.

Table 7-7. Clinical Validity of FoundationOne CDx: Pan-Tumor Studies of FoundationOne Frequent genomic alterations Reference Study design Tumor type identified Conclusions Ross et al. Retrospective analysis of Non-NSCLC, n=97,678 ALK: n=876 • CGP may have increased sensitivity in 341 (2017) patients (N=114,200) who NSCLC, n=21,522 EML4: n=631 detecting ALK fusions for different underwent CGP with tumor types FoundationOne Hirshfield et Prospective trial of patients Gynecological, n=16 ≥1 genomic alteration: 88 (96%) • Use of targeted NGS in the setting of 62 al. (2016) with rare or refractory tumors Breast, n=14 Commonly altered genes: an institutional molecular tumor board (N=100) who underwent led to implementable clinical action in Pancreatic, n=10 • TP53: 41% CGP with FoundationOne over one-third of patients who had • CDKN2A/B: 22% Sarcoma/gastrointestinal rare or refractory tumors • KRAS: 16% stromal tumor, n=10 • Of 8 patients assessed for response to • PIK3CA: 15% Colorectal, n=9 targeted therapy, 3 achieved SD, 2 had • PTEN: 14% PD, 1 had an ongoing response, and 1 Genitourinary, n=8 • BRAF: 7% had a significant response Thyroid, n=5 Started targeted therapy: 31 (35%) Upper gastrointestinal, Eligible for clinical trial: 2 n=5 Hepatobiliary, n=4 Neuroendocrine, n=4 Skin, n=4 CUP, n=3 Thoracic, n=3 Other, n=3 Ross et al. Retrospective study where Patients who harbored BRAF fusions with intact BRAF • Targeted NGS supports the (2016)342 data from consecutive BRAF fusions: kinase domains: 55/20,573 (0.3%) identification of rare driver genomic alterations that occur across a broad patients with Glioma, n=15 range of malignancies

CONFIDENTIAL PAGE 156 US-FDX-2000073 Last update: January 22, 2021

Frequent genomic alterations Reference Study design Tumor type identified Conclusions relapsed/refractory solid Melanoma, n=14 Common alterations in subset of • In 2 case studies assessing response to tumor or hematologic Non-small cell lung, n=8 samples with BRAF fusions: therapy, Patient 1 had a response that malignancy samples enabled resection of previously Colorectal, n=4 • CDKN2A/B: 29% (N=20,573) submitted for • TP53: 22% unresectable tumor, and Patient 2 had Pancreatic, n=3 CGP with FoundationOne or • PTEN: 11% an antitumor response a FoundationOne Heme were Thyroid, n=3 • PIK3CA: 9% analyzed Breast, n=2 • PBRM1: 7% Cancer of unknown • APC: 7% primary, n=2 • EGFR: 7% Esophageal, head and neck, prostate, soft tissue, n=1 each Klempner et Retrospective study of Not reported BRAF kinase domain duplication: • CGP can identify BRAF kinase al. (2016)343 patients with BRAF kinase 9 cases domain duplications across multiple domain duplication in tumor types as a clinically important multiple tumor types genomic alteration and therapeutic (N=50,000) whose samples target were interrogated using • In a single patient assessed for FoundationOne; clinical response to targeted therapy, PR was outcomes in a case study achieved and persisted for >12 months reported Frampton et Retrospective study of Patients who harbored MET exon 14 splice site • CGP successfully identified the al. (2015)344 patients with various MET exon 14 splice site alterations: 221/38,028 (0.6%) challenging MET exon 14 alterations advanced cancers alteration: in multiple tumor types and conferred (N=38,028) who underwent Lung adenocarcinoma, clinical sensitivity to targeted CGP with FoundationOne n=131 therapies ordered as part of routine • In 3 patients assessed for response to Other lung neoplasms, clinical practice targeted therapy, all 3 achieved a PR n=62

Unknown primary origin, n=15 Other, n=7 Brain gliomas, n=6

CONFIDENTIAL PAGE 157 US-FDX-2000073 Last update: January 22, 2021

Frequent genomic alterations Reference Study design Tumor type identified Conclusions

Schwaederle Retrospective US chart Gastrointestinal, n=110 Most common gene abnormalities: • CGP identified theoretically 65 actionable alterations in 90% of et al. (2015) review of patients with Breast, n=83 • TP53: 44% diverse cancers (N=439) who patients with diverse cancers Brain, n=62 • KRAS: 16% underwent CGP with • PIK3CA: 12% FoundationOne Gynecologic, n=37

Hematologic, n=36 ≥1 potentially actionable alteration Head and neck, n=34 identified: 393 (90%) Melanoma, n=32 • On-label: 89 (20%) Lung, n=27 • Off-label: 296 (67%) Sarcoma, n=8 • Clinical trial: 393 (90%) Carcinoma, n=5 Unknown origin, n=2 Carcinoid tumor, nerve sheath tumor, sarcomatoid neoplasm, n=1 each Johnson et al. Retrospective study of Solid, n=101 Potentially actionable mutations: • CGP can uncover additional genotype- 53 (2014) patients with histologically Hematologic, n=2 86 (83%) directed treatment options for patients confirmed malignancy • 26% had alterations that • In 3 case studies of patients assessed (n=103) who received CGP predicted sensitivity to targeted for response following genotype- with FoundationOne agents already approved for the directed treatment, all 3 achieved tumor type assessed clinically meaningful responses • 17% had alterations that could be targeted by agents approved for another cancer type • 51% were potential candidates for clinical trials of therapies that have demonstrated at least early activity in the clinical setting • 6% were eligible for clinical trials with only a preclinical rationale for their use

CONFIDENTIAL PAGE 158 US-FDX-2000073 Last update: January 22, 2021

Frequent genomic alterations Reference Study design Tumor type identified Conclusions • 44% involved in cell-cycle regulation • 31% involved in PI3K/AKT pathway • 19% involved in MAPK pathway Received genotype-directed therapy: 18 (21%) • 7 patients received clinically available agents • 11 patients enrolled in clinical trials ALK, anaplastic lymphoma kinase; AKT, protein kinase B; APC, adenomatous polyposis coli; CDKN2A/B, cyclin-dependent kinase inhibitor 2A/2B; CGP, comprehensive genomic profiling; CUP, cancer of unknown primary; EGFR, epidermal growth factor receptor; EML4, echinoderm microtubule-associated protein-like 4; KRAS, Ki-ras2 Kirsten rat sarcoma viral oncogene homolog; MAPK, mitogen-activated protein kinase; MET, mesenchymal-epithelial transition factor; NGS, next-generation sequencing; PD, progressive disease; PI3K, phosphatidylinositol-4,5-bisphosphate 3-kinase; PIK3CA, phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha phosphatidylinositol-4,5- bisphosphate 3-kinase; PR, partial response; PTEN, phosphatase and tensin homolog; SD, stable disease; TP53, tumor protein P53. a FoundationOne Heme is a CGP assay for hematologic malignancies and sarcomas used to provide information about clinically significant alterations, potential targeted therapies, available clinical trials, and quantitative markers that may support immunotherapy clinical trial enrollment.

Table 7-8. Clinical Validity of FoundationOne CDx: Tumor-Specific Studies of FoundationOne Frequent genomic alterations Reference Study design identified Conclusions Cancer of unknown primary Gay et al. (2017)345 Retrospective cohort study Genomic alterations not reported • CGP identified a significant number of patients with CUP who had TMB-high (≥20 mutations per (N=6,166) in patients with TMB-high (%): tumors for which the megabase), which is a biomarker suggestive of • Squamous cell carcinoma: 23% primary site was unknown sensitivity to select immunotherapies • Malignant neoplasm NOS: 15% who underwent CGP with • TMB-high found in urothelial tumors, SCC tumors • Urothelial: 13% FoundationOne (23%), and tumors difficult to define histologically • NOS (CUP): 11% (15%) • Small cell: 9% • For adenocarcinoma CUP or CUP, the most common • Adenocarcinoma CUP: 8% tumors, 8%–11% have TMB-high • Neuroendocrine: 6%

CONFIDENTIAL PAGE 159 US-FDX-2000073 Last update: January 22, 2021

Frequent genomic alterations Reference Study design identified Conclusions Ross et al. (2015)346 Retrospective clinical • KRAS substitution, n=40 • Given the poor prognosis of CUP treated by study using samples • CDKN2A loss, substitution, truncation, nontargeted conventional therapies, CGP shows the submitted from patients n=37 potential to identify targeted therapies to improve diagnosed with CUP • MCL1 amplification, n=19 patient outcomes while potentially reducing the often (adenocarcinoma • PIK3CA substitution, amplification, n=17 costly and time-consuming search for the anatomic site histology, N=125; other • PTEN loss, substitution, truncation, n=14 of origin histologies, N=75) who • STK11 truncation, n=13 • A total of 26 alterations are associated with targeted underwent CGP with • RICTOR amplification, n=12 therapies approved for use in patients with a known FoundationOne • BRAF substitution, n=11 primary tumor type • BRCA2 substitution, truncation, n=11 • A total of 14 cases harbored off-target genomic • ERBB2 substitution, n=9 alterations • ATM substitution, truncation, n=7 • EGFR substitution, n=6 • Remaining actionable genomic alterations were linked • ERBB2 amplification, n=6 to registered clinic trials • NF1 loss, truncation, n=6 • MET amplification, n=5 • NOTCH1 substitution, truncation, n=5 • CCND2 amplification, n=4 • FGFR1 substitution, amplification, fusion, n=4 • FGFR2 substitution, fusion, n=4 • ALK fusion, n=2 • RET fusion/substitution, n=1 • ROS1 fusion, n=1 Chung et al. Case report of a patient EML4-ALK rearrangement • CGP resulted in the front-line use of targeted therapy (2014)347 with a malignant neoplasm and a significant sustained antitumor response of unknown histology • Patient was asymptomatic at 5 months follow-up (N=1) who underwent CGP with FoundationOne Palma et al. Case report of a patient KRAS G12V mutation while lacking • CGP led to the detection of an unexpected genomic (2014)348 diagnosed with CUP alterations in EGFR or ALK and a MET alteration and the successful institution of an available (N=1) who underwent amplification targeted therapy for the patient with previously treated real-time PCR, break-apart CUP

CONFIDENTIAL PAGE 160 US-FDX-2000073 Last update: January 22, 2021

Frequent genomic alterations Reference Study design identified Conclusions FISH, followed by CGP • Patient experienced a complete metabolic response that with FoundationOne persisted for more than 19 months Pancreatic cancer Singhi et al. Retrospective study on • KRAS: n=3,180 (88%) • CGP can identify potentially targetable and predictive (2019)349 samples from patients with • TP53: n=2,643 (74%) biomarkers in patients with pancreatic ductal adenocarcinoma pancreatic ductal • CDKN2A: n=1,572 (44%) • Biomarker identification can help inform early adenocarcinomas • SMAD4: n=794 (22%) detection strategies for pancreatic ductal (N=3,594) who underwent • BRCA/FANC gene family: n=489 (14%) adenocarcinomas CGP with FoundationOne • Potentially actionable kinase fusions in 12% (51/445) KRAS-WT cases Singhi et al. Prospective study of ALK rearrangements • ALK fusions are rare but do occur in pancreatic ductal (2017)350 patients with pancreatic adenocarcinoma ductal adenocarcinoma (N=3,170) who harbored an ALK fusion gene (n=5) and received matched therapy following CGP with FoundationOne; outcomes are reported in a case series (n=4) Chmielecki et al. Retrospective US-based BRAF and RAF1: approximately 23% of • CGP identified potentially actionable genomic (2014)351 study on samples from cases, the most common being an SND1- alterations in patients with pancreatic acinar cell patients with pancreatic BRAF fusion (n=5) carcinoma acinar cell carcinoma (N=44) who underwent CGP with FoundationOne Prostate cancer Chung et al. Prospective study on • TP53: 43.5% • 57% of cases were identified as genomic alterations (2019)352 samples from patients • ETS fusion: 35.5% with an available targeted therapy (N=3,476) with prostate • PTEN: 32.2% • In 2,624 cases where genomic LOH could be assessed, cancer who underwent • TMPRSS2-ERG: 31.2% the median genomic LOH score was 8.5% (IQR: 5.8– CGP with FoundationOne • AR: 22.5% 12.2)

CONFIDENTIAL PAGE 161 US-FDX-2000073 Last update: January 22, 2021

Frequent genomic alterations Reference Study design identified Conclusions • MYC: 12.3% o 447/2,624 (17%) were considered to have high • BRCA2: 9.8% genomic LOH (>14%) • RB1: 9.7% • APC: 9.3% • MLL3/KMT2C: 7.8% • SPOP: 7.7% • PIK3CA: 6.0% • CDK12: 5.6% • BRAF/RAF1 fusion/rearrangement: 1.2% Ikeda et al. (2018)353 Retrospective study on • TP53: 55.2% • Use of NGS in prostate cancer is low compared to samples from patients • PTEN: 29.9% other tumor types; however, patients with advanced (N=67) with prostate • TMPRSS2 fusion: 28.4% prostate cancer have diverse genomic alterations cancer who underwent • MYC: 17.9% • Potentially targetable alterations were identified in CGP by FoundationOne • PIK3CA: 13.4% 84% of cases • APC: 9.0% • Identification of genomic alterations in advanced • BRCA2: 9.0% prostate cancer can be helpful in determining treatment • CCND1: 9.0% but more genomically based studies in prostate cancer • RB1: 9.0% are needed APC, adenomatous polyposis coli gene; BRCA, breast cancer susceptibility gene; CCND1, cyclin D1; CDK, cyclin-dependent kinase inhibitor; CGP, comprehensive genomic profiling; ETS, erythroblast transformation-specific; IQR, interquartile range; LOH, loss of heterozygosity; MLL3/KMT2C, myeloid/lymphoid or mixed-lineage leukemia protein 3/lysine N-methyltransferase 2C; MYC, Myc proto-oncogene protein; NGS, next-generation sequencing; PIK3CA, phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha phosphatidylinositol-4,5-bisphosphate 3-kinase; PTEN, phosphatase and tensin homolog; RB, retinoblastoma protein; SPOP, speckle-type POZ gene; TMPRSS2-ERG, transmembrane protease serine 2:v‑ets erythroblastosis virus E26 oncogene homolog; TP53, tumor protein P53.

Clinical Utility of Previous Versions of Foundation Medicine Tissue Assay The clinical validity of study results associated with FoundationOne are presented in Table 7-9 for pan-tumor studies and Table 7-10 for tumor- specific studies.

CONFIDENTIAL PAGE 162 US-FDX-2000073 Last update: January 22, 2021

Table 7-9. Clinical Utility of FoundationOne CDx: Pan-Tumor Studies of FoundationOne Clinical outcomes Frequent genomic No. Reference Study design Tumor type alterations identified patients Outcome Reitsma et Retrospective single- NSCLC, n=36 86/96 (90%) patients had a 24a CGP testing and treatment patterns, n: 61 al. (2019) center US-based Colorectal, n=13 genomic alteration that • Patients with clinically relevant genomic study of patients could be predictive of Breast, n=8 alterations detected matching to FDA- (N=96) who sensitivity or resistance to approved therapy and clinical trials: 86 Urothelial/bladder, underwent CGP either FDA-approved or • Patients treated following CGP: 70/86 testing with either n=7 investigational therapies o Patients treated with CGP-matched FoundationOne or Unknown primary on the basis of clinical or targeted therapy: 15/70 FoundationOne Heme origin, n=5 preclinical evidence o Patients who enrolled in CGP-matched Sarcoma, n=5 clinical trials: 6/70 Hematological • Patients with previous conventional molecular malignancy, n=4 diagnostic testing: 32 Patients where CGP detected a previously Melanoma, n=4 o unidentified actionable genomic Pancreatobiliary, n=4 alteration: 27/32 Others, n=10 o Patients who received treatment informed by an actionable alteration not identified by previous testing: 6/27 o Patients who experienced clinical benefit from CGP-detected treatment: 4/6 CGP testing treatment outcomes, median OS: • All patients (N=95): 4.8 months (range: 0–31) o Patients treated with a CGP-matched targeted therapy or immunotherapy: (n=15): 9.5 months (range: 1.1–24.2) o Patients enrolled in a clinical trial (n=9): 4.5 months (range: 2.3–20.6) • Patients not enrolled in a clinical trial and did not receive a CGP-matched targeted therapy or immunotherapy (n=71): 4.6 months (range: 0– 30.9)

CONFIDENTIAL PAGE 163 US-FDX-2000073 Last update: January 22, 2021

Clinical outcomes Frequent genomic No. Reference Study design Tumor type alterations identified patients Outcome Kato et al. Prospective study of Castleman’s disease, Most common alterations 21a All patients who underwent matched targeted (2018)63 patients with rare n=4 in 33 patients were, n (%): therapy, N: tumors, presenting at Erdheim-Chester • TP53: 15 (45.5) • SD ≥6 months: 3/21 (14.3%) the Rare Tumor disease, n=5 • CDKN2A/B, FRS2, • PR: 6/21 (28.6%) Clinic (N=40) who High-grade serous MDM2, RB1, KRAS: 4 • CR: 2/21 (9.5%) underwent CGP ovarian cancer, n=4 (12.1) each • Median PFS (range): 19.6 months (0.99+ to (n=33/40), the • MLL2, PIK3CA, 26.1+) Basal cell carcinoma, majority (n=31) by PIK3R1, PTEN, SETD2, CUP, metaplastic Matched patients who were assessed for FoundationOne TERT: 3 (9.1) each carcinoma of breast, matched vs last prior unmatched therapy: • AKT1, APC, ARID1A, papillary serous • Median PFS (n=12/21): 19.7 vs 3.5 months ASXL1, BCOR, BRCA1, carcinoma of ovary; (HR, 0.26; 95% CI: 0.10–0.71; P=0.008) CCNE1, CDK4, n=2 each • PFS ratiob >1.3, n/N: 8/12 CDKN2A, EGFR, Adenoid cystic EWSR1, FANCC, FAT1, carcinoma, FBXW7, JAK1, MYC, ameloblastoma, NOTCH1, PTCH1, ampullary carcinoma, SMARCA4, STK11, anal squamous cell TERC; 2 (6.1) each carcinoma, angiosarcoma of breast, chondrosarcoma, desmoid tumor sarcoma, endometrial stroma sarcoma, fallopian cancer, fibromyxoid sarcoma, glioblastoma multiforme, liposarcoma, Merkel cell, myoepithelial carcinoma, myxofibrosarcoma,

CONFIDENTIAL PAGE 164 US-FDX-2000073 Last update: January 22, 2021

Clinical outcomes Frequent genomic No. Reference Study design Tumor type alterations identified patients Outcome neuroendocrine tumor of the uterine cervix, ocular melanoma, thymoma, yolk sac tumor; n=1 each Sadaps et al. Retrospective single- Lung: n=122 (20.3%) 310 patients (51.7%) were 313a Matched patients, median OS: 60 (2018) center study (N=600) Breast: n=99 (16.5%) recommended targeted • Patients receiving no treatment after tumor of patients with therapy on the basis of Colorectal: n=80 genomic profiling: 5.5 months incurable solid tumor sequencing results (13.3%) • Patients receiving non-genomics-driven malignancies who Most common therapy after GTB recommendations (n=218, Head and neck: n=57 received alternations, n/N: 70%): 14.4 months (P<0.001) FoundationOne (9.5%) • PIK3CA/PIK3R/PTEN • Patients receiving genomics-driven therapy testing Pancreatobiliary: (n=78) after GTB recommendations (n=95, 15.8%): n=52 (8.7%) • KRAS (n=65) 18 months (P<0.001) Melanoma: n=28 • FGF/FGFR (n=49) o On-label therapy: 25 months (n=22) (4.7%) • CDK (n=32) o Off-label therapy: 16.4 months (n=45) Prostate: n=26 (4.3%) • HER2 (n=28) o Clinical trial: 18.3 months (n=28) • Genomics vs nongenomic comparison: HR, Bladder: n=19 (3.2%) • BRAF (n=25) 0.9, 95% CI: 0.64, 1.25, P=0.52 Glioblastoma: n=13 (2.2%) Mesothelioma: n=10 (1.7%) Renal: n=8 (1.3%) Esophageal: n=7 (1.2%) Sarcoma: n=6 (1.0%) Glioma: n=6 (1.0%) Meningioma: n=4 (0.7%) Adrenal: n=3 (0.5%)

CONFIDENTIAL PAGE 165 US-FDX-2000073 Last update: January 22, 2021

Clinical outcomes Frequent genomic No. Reference Study design Tumor type alterations identified patients Outcome Liver: n=3 (0.5%) Haslem et Retrospective study NSCLC, n=22 On-label, n=3 72 Precision medicine vs controls: 11 al. (2017) of patients with 10 Colon, n=16 Off-label, n=33 Average PFS: 22.9 vs 12.0 weeks (HR: 0.47; metastatic cancers of Breast, n=10 95% CI: 0.29–0.75; P=0.002) diverse subtypes (N=36) and a Head and neck, n=8 matched cohort of Bladder, n=4 historical controls Pancreas, n=4 (N=36) who Cholangiocarcinoma, underwent NGS n=2 testing with a range of tests, including Gastric, n=2 FoundationOne Melanoma, n=2 Ovary, n=2 PREDICT- Retrospective study Breast, n=60 • CDKN2A: 34 (18.9%) 180a Matched vs unmatched therapy: 74 USCD of patients with Gastrointestinal, n=35 • TP53: 87 (48.3%) • CR/PR or SD ≥6 months: 34.5% vs 16.1% advanced solid • PTEN: 19 (10.6%) Genitourinary, n=22 (P≤0.020) malignancies • PFS ratiob >1.3: 45.3% vs 19.3% (P=0.004 Head and neck, n=16 (N=347) who univariable and P=0.057 multivariable/ underwent CGP with Lung, n=16 propensity score analysis) FoundationOne Brain, n=15 • Median PFS: 4.0 vs 3.0 months (P=0.039) Skin/melanoma, n=11 • Median OS: 15.7 vs 10.6 months for matching score >0.2 vs ≤0.2 (P=0.040) Rodriguez- Prospective, Ovarian, n=41 93% with actionable 64a Matched patients: Rodriguez et single-arm study of Uterine, n=25 genomic alterations • 64/69 (93%) patients received MTB-based al. (2016)64 patients with rare or Vaginal, n=2 Most common alterations, therapy recommendations refractory n/N: • 25/64 (39%) proceeded to MTB-recommended gynecologic cancers Cervical, n=1 • TP53: 45/61 therapy (N=69) who received • PIK3CA: 13/61 • 16/25 (64%) of patients who received MTB- FoundationOne • MYC: 8/61 based therapy recommendations had a CR, PR, testing and MTB • KRAS: 8/61 SD, or CB

CONFIDENTIAL PAGE 166 US-FDX-2000073 Last update: January 22, 2021

Clinical outcomes Frequent genomic No. Reference Study design Tumor type alterations identified patients Outcome recommendations for • ARID1A: 8/61 • Outcomes: CR, n=1; PR, n=9; SD, n=4; CB, treatment • PTEN: 7/61 n=2; PD, n=3 • Unknown or no tumor data: n=6 • PFS duration similar to or longer than their PFS duration with a prior therapy: n=6/23 (26%) Wheler et al. Prospective study Ovarian: 18% 93.5% with ≥1 actionable 188a Matched vs unmatched therapy: 67 (2016) (N=500) of patients Breast: 16% genomic alteration • CR/PR or SD ≥6 months: 19% vs 8% with advanced Sarcoma: 13% (P=0.061) malignancies • Median TTF: 2.8 vs 1.9 months (P=0.001) Renal: 7% evaluated with CGP • Median OS: 9.3 vs 7.2 months (P=0.087) by FoundationOne Wheler et al. (including Gastrointestinal, n=34 106a Patients with TP53 alterations (n=106) treated 163 (2016) subanalyses) Ovarian, n=33 with vs without anti-VEGF therapy: Breast, n=31 • CR/PR or SD ≥6 months: 29% vs 9% (P=0.020) Sarcoma, n=24 • Median TTF: 3.2 vs 2.1 months (P=0.021) Kidney, n=13 • Median OS: 8.0 vs 7.5 months (P=0.163) Head and neck, n=12 Uterus, n=9 Gynecological (non-uterine), n=6 Lung, n=6 Melanoma, n=6 Neuroendocrine, n=6 Adenoid cystic, n=2 Prostate, n=2 CUP, liver, thymic, thyroid; n=1 each a Evaluable patients are those who have received at least 1 dose of the targeted therapy for whom outcomes data are available.

CONFIDENTIAL PAGE 167 US-FDX-2000073 Last update: January 22, 2021 b PFS ratio refers to the PFS while receiving the current therapy line vs the previous therapy line. AKT, protein kinase B; ARID1A, AT-rich interactive domain-containing protein; CB, clinical benefit; CDKN2A/B, cyclin-dependent kinase inhibitor 2A/2B; CGP, comprehensive genomic profiling; CI, confidence interval; CR, complete response; CUP, cancer of unknown primary; EGFR, epidermal growth factor receptor; EHCCA, extrahepatic cholangiocarcinoma; ER, estrogen receptor; FGFR, fibroblast growth factor receptor; FLT1, FMS-related tyrosine kinase 1; FRS2, fibroblast growth factor receptor substrate 2; GBC, gallbladder carcinoma; HER2, human epidermal growth factor receptor 2; HR, hazard ratio; IDH1, isocitrate dehydrogenase 1; IHCCA, intrahepatic cholangiocarcinoma; KRAS, Ki-ras2 Kirsten rat sarcoma viral oncogene homolog; Mb, megabase; MTB, molecular tumor board; NF1, neurofibromin 1; NSCLC, non-small cell lung cancer; OR, odds ratio; OS, overall survival; PD, progressive disease; PFS, progression-free survival; PIK3CA, phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha phosphatidylinositol-4,5-bisphosphate 3-kinase; PR, partial response; PTEN, phosphatase and tensin homolog; RB1, retinoblastoma-1; RET, ret proto-oncogene; RR, response rate; SD, stable disease; TMB, tumor mutational burden; TP53, tumor protein P53; TTF, time to treatment failure; VEGF, vascular endothelial growth factor.

Table 7-10. Clinical Utility of FoundationOne CDx: Tumor-Specific Studies of FoundationOne Clinical outcomes Frequent genomic a Reference Study design alterations identified No. patients Outcome NSCLC Singal et al. Retrospective study of All patients (n=4,064) 575 Advanced diagnosis patients with a driver 79 (2019) NSCLC patients who • EGFR: n=701 (17.2%) alteration (n=1,260): underwent CGP testing • ALK: n=128 (3.1%) • 609/1,260 (48.3%) received a guideline- with FoundationOne • ROS: n=42 (1.0%) recommended targeted-therapy, including: Advanced disease (n=3,522) o 405/630 (64.3%) of those with EGFR mutations • EGFR: n=630 (17.9%) 75/107 (70.1%) of those with ALK mutations • ALK: n=107 (3.0%) o

• ROS: n=not reported Survival outcomes for evaluable patients

(n=1,135) with advanced diagnosis and with an altered oncogenic driver gene (EGFR, ALK, ROS1, MET, BRAF, RET, or ERBB2): • Patients receiving a guideline-recommended- targeted therapy (n=575), median OS: 18.6 months • Patients who did not receive a guideline-directed targeted therapy (n=560), median OS: 11.4 months • Difference: 7.1 months (95% CI: 3.5–10.1; P<0.001) Skoulidis et al. Retrospective study of All patients were KRAS- 174 STK11/LKB1 + KRAS-mutant: 354 NSCLC patients (n=174) (2018) mutant • ORR to PD-1 inhibition: 7.4%

CONFIDENTIAL PAGE 168 US-FDX-2000073 Last update: January 22, 2021

Clinical outcomes Frequent genomic a Reference Study design alterations identified No. patients Outcome who received PD-1 • STK11/LKB1: 31% • Median PFS: 1.8 months inhibitor monotherapy • TP53: 32% • Median OS: 6.4 months (165/174, 95%) or in • KRAS-mutant only: 37% TP53 and KRAS-mutant: combination with CTLA4 • ORR to PD-1 inhibition: 35.7% blockade (9/174, 5%) who • PFS: median 3.0 months underwent CGP with • OS: 16.0 months FoundationOne KRAS-mutant only: • ORR to PD-1 inhibition: 28.6% • PFS: median 2.7 months • OS: median 16.1 months Hazard ratio for risk of progression: • STK11/LKB1 + KRAS-mutant vs TP53 and KRAS- mutant, HR: 1.77; 95% CI: 1.16–2.69; P=0.0072 • STK11/LKB1 + KRAS-mutant vs KRAS-mutant only, HR: 1.98; 95% CI: 1.33–2.94; P<0.001 Hazard ratio for risk of death: • STK11/LKB1 + KRAS-mutant vs KRAS-mutant only, HR: 1.99; 95% CI, 1.29–3.06; P=0.0015 Rozenblum et al. Retrospective study of Common genes with 37 Response to matched therapy: 71 (2017) patients with advanced lung mutational or structural • ORR: 64.7% adenocarcinoma (N=101) change (level 1 analysis) • CR: 5 (14.7%) who underwent NGS involved in 121 genomic • PR: 17 (50%) testing with FoundationOne alterations: • SD: 9 (26.5%) or Guardant360 • KRAS: 18.2% • PD: 3 (8.8%) • EGFR: 16.5% • RET: 7.4% • STK11: 7.4% • ALK: 6.6% • ERBB2: 5.8% • MET: 5.8%

CONFIDENTIAL PAGE 169 US-FDX-2000073 Last update: January 22, 2021

Clinical outcomes Frequent genomic a Reference Study design alterations identified No. patients Outcome Most common actionable genomic alterations (level 2 analysis): • EGFR mutations: 15% • RET rearrangements: 9% • ALK rearrangements: 8% • MET amplifications and/or exon 14 mutations: 6% BATTLE-2355 Prospective randomized KRAS mutation: 27% 186 With vs without erlotinib: control trial of patients with KRAS-mutant tumors advanced refractory • PFS: 2.5 vs 1.8 months (P=0.04) NSCLC (N=200) who underwent CGP testing Wild-type KRAS (n=140 with • OS: 9.0 vs 5.1 months (P=0.03) FoundationOne) Breast cancer Wheler et al. Retrospective study of FGF/FGFR amplifications: 49 FGFR inhibitor: 356 (2016) patients with metastatic 24/112 (21%) • Response: 0/3 breast cancer (N=112) in a • FGF: 13 PI3K/AKT/mTOR inhibitor: phase 1 program who • FGFR: 10 • CR/PR or SD ≥6 months: 1/7 (14%) underwent CGP testing • FGFR and FGF: 1 with FoundationOne PI3K/AKT/mTOR-targeted therapy: 15/24 (63%) also had a co- occurring alteration in the Patients with FGF/FGFR amplification and PI3K/AKT/mTOR pathway PI3K/AKT/mTOR alteration vs FGF/FGFR only • CR/PR or SD ≥6 months: 8/11 (73%) vs 12/35 (34%; P=0.0376) • TTF: 6.8 (95% CI: 2.413–11.187) vs 3.7 (95% CI: 2.39–5.01) months (P=0.053) Balko et al. Retrospective study of Actionable genomic 74 Poor RFS associated with: 357 (2014) patients with triple-negative alterations: 90% of tumors • JAK2 amplification: HR, 3.36 (P=0.006) breast cancer and residual • High MEK activation score: HR, 1.758 (P=0.059) disease after neoadjuvant

CONFIDENTIAL PAGE 170 US-FDX-2000073 Last update: January 22, 2021

Clinical outcomes Frequent genomic a Reference Study design alterations identified No. patients Outcome chemotherapy (N=74) who Alteration categories Poor OS associated with: underwent CGP testing identified: • BRCA1 truncations/mutations: HR, 2.5 (P=0.041) with FoundationOne • Cell cycle genomic • JAK2 amplification: HR, 4.16 (P=0.002) alterations • PTEN alterations: HR, 0.14 (P=0.03) • PI3K/mTOR alterations • High MEK activation score: HR, 2.264 (P=0.013) • GFR amplificationsb • RAS/MAPK alterations • Mutations of BRCA1/2 or ATM Ganesan et al. Retrospective study of Most common alterations 98 Chemotherapy alone (evaluable n=7/8) 77 (2014) consecutive patients with (n=9 patients): • Median PFS: 2.1 months (95% CI: 0.9–3.3; advanced or metastatic • TP53 mutation: 8 P<0.0001) triple-negative metastatic • MYC amplification: 4 Chemotherapy and targeted agent (evaluable breast cancer (N=106) • PIK3R1 mutation: 2 n=57/63) treated in a phase 1 clinic • FGFR2, MCL1, and • CR: 1 who underwent CGP testing CCND1 amplification: 2 • PR: 6 with FoundationOne each • SD ≥6 months: 4 • NF2, PTEN, KDM6A, and • Median PFS: 3.0 months (95% CI: 1.9–4.1; RB1: 1 each P<0.0001) • FGFR1, PIK3CA, CDK8, MAP2K2, KRAS Single-agent targeted drug (evaluable n=15/15) amplification: 1 each • Median PFS: 1.1 months (95% CI: 0.7–1.4; • PTEN deletion: 1 P<0.0001) ≥2 targeted agents (evaluable n=19/20) • PR: 1 • Median PFS: 1.9 months (95% CI: 1.4–2.4; P<0.0001) Overall matched vs unmatched therapy: • CR/PR or SD ≥6 months: 33% vs 8% (P=0.018) • Median PFS: 6.4 vs 1.9 months (P=0.001) Colorectal cancer

CONFIDENTIAL PAGE 171 US-FDX-2000073 Last update: January 22, 2021

Clinical outcomes Frequent genomic a Reference Study design alterations identified No. patients Outcome Schrock et al. Multicenter, retrospective TMB ≥37 mutation/Mb in ~65% 22 TMB-high (≥37 mutations/Mb) vs TMB-low (<37 (2019)358 study in patients with of cases mutations/Mb): metastatic CRC (N=22) who • OR: 13/13 (100%) vs 3/9 (33%) (P<0.001) had confirmed MSI-H by • Median PFS: not reached (median follow-up >18 NGS by Foundation months) vs 2 months (P=0.00087) Medicine and were treated with PD-1 or PD-L1 inhibitors Jones et al. Multicenter, retrospective, non-V600BRAF: 208 590 V600EBRAF vs non-V600BRAF 359 (2017) US-based cohort study in V600EBRAF: 133 • Median OS: 11.4 vs 60.7 months (P<0.001) patients with metastatic WTBRAF: 249 V600EBRAF vs WTBRAF CRC (N=9,643) who were identified from NGS • Median OS: 11.4 vs 43.0 months (P<0.001) databases at 3 molecular In a multivariable analysis, non‑V600BRAF was genetics reference independently associated with improved OS (HR, laboratories and underwent 0.18; P<0.001) NGS testing with Ampliseq Cancer Hotspot Panel v2 or FoundationOne Melanoma Boussemart et al. Retrospective study (N=12) BRAF mutations: 0% (all were 12 Responses to immunotherapy: 360 (2018) conducted in samples from BRAF wild-type) • 9/12 patients had clinical treatment history patients with desmoplastic Other mutations included: available melanoma who underwent • TP53: 9/12 • 6/9 patients received a PD-1 inhibitor CGP to assess TMB and • NF1: 6/12 • SD or better: 5/5 (based on those who had response to PD-1 inhibitors • NF2: 2/12 available response data); PR: 3/5 • CDKN2A: 6/12 • ARID2: 5/12 Wheler et al. Retrospective US-based BRAF mutations: 100% 10 Response to matched therapy: 361 (2015) cohort study (N=10) in Other mutations included: • CR: 4 (40%) patients with advanced • CDKN2A deletions: 4 • PR: 6 (60%) BRAF mutation-positive

CONFIDENTIAL PAGE 172 US-FDX-2000073 Last update: January 22, 2021

Clinical outcomes Frequent genomic a Reference Study design alterations identified No. patients Outcome melanoma who underwent • CDKN2B deletions: 3 • Individual TTFs among patients with CR: 5.6, NGS with genomic • BRAF amplifications: 2 23.6+, 27.4+, and 28.7+ months libraries at The University • PAX5: 2 • Individual TTFs among patients with PR: 3.0, 4.2, of Texas MD Anderson • NRAS: 1 5.7, 7.0, 7.9, and 11.2 months Cancer Center or • APC: 1 Foundation Medicinec • NF1: 1 • Aurora kinase A: 1 • MYC: 1 • MITF: 1 • MET: 1 • RB1 alterations: 1 • ATM: 1 Ovarian cancer SOLO-1 Randomized, double-blind, tBRCA-mutant: n=313 Full analysis Olaparib vs placebo in full analysis set: Moore et al. multi-center, placebo- set: N=391 • Investigator-assessed PFS: HR, 0.30 (95% CI: (2018)109 and controlled, phase 3 trial to F1CDx- 0.23–0.41); P<0.001 assess the efficacy and Foundation identfied Olaparib vs placebo in tBRCA mutation subset: Medicine safety of olaparib vs tBRCA • Investigator-assessed PFS: HR, 0.28 (95% CI: (2019)1 placebo in patients with mutation: newly diagnosed advanced 0.20–0.38); P<0.0001 ovarian carcinoma with BRCA1 and/or BRCA2 mutation who had complete or partial response after platinum-based chemotherapy Coleman et al. Randomized, double-blind, BRCA-mutant: n=196 564 in nested BRCA-mutant carcinoma subcohort: 289 (2017) multi-center, placebo- • Germline: n=130 (over-lapping • Rucaparib group (n=130) median PFS: 16.6 controlled, phase 3 trial • Unknown: n=10 cohorts) months (95% CI: 13.4–22.9) Patient tumor samples were • Somatic: n=56 • Placebo group (n=66) median PFS: 5.4 months assessed for LOH with HRD cohortd: n=354 (includes ITT=564 (95% CI: 3.4–6.7) Foundation Medicine’s T5 • HR: 0.23 (95% CI: 0.16–0.34; P<0.0001) n=196 who were BRCA-mutant BRCA-mutant: NGS assay

CONFIDENTIAL PAGE 173 US-FDX-2000073 Last update: January 22, 2021

Clinical outcomes Frequent genomic a Reference Study design alterations identified No. patients Outcome plus n=158 with LOH detected n=196 HRD (LOH ≥16%, ± BRCA mutation) subcohort: but BRCA wild-type) HRD: n=354 • Rucaparib group (n=236) median PFS: 13.6 months (95% CI: 10.9–16.2) • Placebo group (n=118) median PFS: 5.4 months (95% CI: 5.1–5.6) • HR: 0.32 (95% CI: 0.24–0.42; P<0.0001) ITT population: • Rucaparib group (n=375) median PFS: 10.8 months (95% CI: 8.3–11.4) • Placebo group (n=189) median PFS: 5.4 months (95% CI: 5.3–5.5) • HR: 0.36 (95% CI: 0.30–0.45; P<0.0001) Swisher et al. Prospective, multicenter, BRCA-mutant, n=40 204 BRCA-mutant (n=40): 108 (2017) phase 2 open-label study LOH high (≥14%), n=82 • Median PFS: 12.8 months (95% CI: 9.0–14.7) conducted in patients with LOH low (<14%), n=70 • HR: 0.27 (95% CI: 0.16–0.44; P<0.0001) platinum-sensitive • Median DOR: 9.2 months (95% CI: 6.4–12.9;

recurrent high-grade P=0.013) ovarian carcinoma to LOH high (≥14%) cohort (n=82): identify molecular predictors of rucaparib • Median PFS: 5.7 months (95% CI: 5.3–7.6) sensitivity • HR: 0.62 (95% CI: 0.42–0.90; P=0.011) • Median DOR: 10.8 months (95% CI: 5.7–not reached; P=0.022) LOH low (<14%) cohort (n=70): • Median PFS: 5.2 months (95% CI: 3.6–5.5) • Median DOR: 5.6 months (95% CI: 4.6–8.5) Pancreatic cancer Pishvaian et al. Prospective tumor sample • KRAS: 87% 126 Types of treatments selected by patients who (2018)75 analysis (N=640) in • CDKN2A: 45% chose treatment options after molecular testing: pancreatic cancer patients • CDK4/6 or CCND1/2/3: 8.1% • Standard of care: 80/126 as part of the Know Your • STK11: 4.7% • Off-label, molecular targeted therapy: 20/126 Tumor program • ATM: 4.5% • Clinical trials: 26/126

CONFIDENTIAL PAGE 174 US-FDX-2000073 Last update: January 22, 2021

Clinical outcomes Frequent genomic a Reference Study design alterations identified No. patients Outcome • PI3K: 3.7% • Patients with highly actionable molecular • BRCA1/2: 2.9% alterations receiving molecular targeted therapies • AKT: 2.8% that matched biomarkers: 17/126 • ERBB2: 2.8% • Patients with highly actionable molecular alterations receiving standard of care or molecularly targeted agents that did not match biomarkers: 18/126

Patients receiving genomically matched therapy vs those receiving non-genomically matched therapy: • Median PFS: 4.1 vs 1.9 months (HR: 0.47; P=0.03) • Median OS: 1.5 vs 0.9 years (statistical significance not reached) a Evaluable patients are those who have received at least 1 dose of the targeted therapy for whom outcomes data are available. b GFR amplifications include: EGFR, MET, KIT, FGFR1-2 and -4, or IGF1R. c Genomic libraries were captured for 3,230 exons in 182 cancer-related genes plus 37 introns from 14 genes often rearranged in cancer; they were sequenced to an average median depth of 734X, with 99% of bases covered >100X. d The HRD cohort was composed of patients with BRCA-mutated carcinoma or BRCA wild-type and high genomic loss of heterozygosity carcinomas. AKT, protein kinase B; ALK, anaplastic lymphoma kinase; APC, adenomatous polyposis coli; ATM, ataxia-telangiectasia mutated gene; BRCA, breast cancer; CDKN2A/B, cyclin-dependent kinase inhibitor 2A/B; CGP, comprehensive genomic profiling; CI, confidence interval; CR, complete response; CRC, colorectal cancer; EGFR, epidermal growth factor receptor; FGF, fibroblast growth factor; FGFR, fibroblast growth factor receptor; GFR, growth factor receptor; HR, hazard ratio; HRD, homologous recombinant deficient; ITT, intention to treat; JAK, Janus kinase; KRAS, V-Ki-ras2 Kirsten rat sarcoma; LOH, loss of heterozygosity; MAPK, mitogen-activated protein kinase; MET, mesenchymal-epithelial transition factor; mTOR, mammalian target of rapamycin; MSI-H, microsatellite-high; MTIF, microphthalmia-associated transcription factor; NF1, neurofibromin 1; NF2, neurofibromin 2; NGS, next-generation sequencing; NSCLC, non-small cell lung cancer; ORR, objective response rate; OS, overall survival; PAX5, paired box gene 5; PD, progressive disease; PFS, progression-free survival; PI3K, phosphoinositide 3-kinase; PR, partial response; PTEN, phosphatase and tensin homolog; RB1, retinoblastoma-1; RET, ret proto-oncogene; RFS, recurrence-free survival; SD, stable disease; STK11, serine/threonine kinase 11 gene; TP53, tumor protein P53; TTF, time to treatment failure; TMB, tumor mutational burden; WT, wild-type.

CONFIDENTIAL PAGE 175 US-FDX-2000073 Last update: January 22, 2021

Clinical Utility of NGS The clinical utility of NGS is described in more detail in Table 7-11.

Table 7-11. Clinical Utility of NGS: Pan-Tumor Meta-Analyses and Studies of Non-FoundationOne Tests Potentially actionable Reference Study design Tumor type alterations Clinical outcomes Drilon et al. Prospective basket study of Salivary gland tumor, NTRK1, n=25 Response rates for all patients: 102 (2018) patients with TRK n=12 NTRK2, n=1 • ORR per independent radiologic review: 75% fusion-positive cancer Other soft tissue sarcoma, NTRK3, n=29 (95% CI: 61–85) (N=55) as identified with n=11 CR: 13% NGS o Infantile fibrosarcoma, o PR: 62% n=7 o SD: 13% PD: 9% Thyroid, n=5 o o Unable to be evaluated: 4% Colon, n=4 • ORR per investigator assessment: 80% (95% Lung, n=4 CI: 67–90) Melanoma, n=4 o CR: 16% PR: 64% GIST, n=3 o o SD: 9% Cholangiocarcinoma, n=2 o PD: 11% Appendix, n=1 • Median DOR: not reached after a median Breast, n=1 follow-up of 8.3 months (range: 0.03–24.9) • Median PFS: not reached after a median Pancreatic, n=1 follow-up of 9.9 months (range: 0.7, 25.9) Kopetz et al. Prospective study of Colorectal, n=78 Patients with ≥1 Matched therapy among patients with an 13 (2018) molecular profiling with a Sarcoma, n=71 potentially actionable alteration: 409-gene panel among actionable alteration Head and neck, n=43 • OS vs unmatched therapy: HR, 0.47; 95% CI: patients with refractory solid not previously 0.25–0.89; P=0.017 Ovarian, n=43 tumor cancers (N=521) who identified, n=214 • OS vs remainder of cohort: P=0.00027 previously underwent Breast, n=42 (41%) molecular testing with a 46- Lung, n=34 or 50-gene panel All others, n=210

CONFIDENTIAL PAGE 176 US-FDX-2000073 Last update: January 22, 2021

Potentially actionable Reference Study design Tumor type alterations Clinical outcomes Patients matched to targeted therapy, n=40/214 (19%) Radovich et al. Prospective study of Soft tissue sarcoma, n=21 Patients with a PFS Genomically matched vs non-genomically 98 (2016) patients with 29 metastatic Breast, n=16 ratio >1.3: matched treatment groups: solid tumors (N=168) who a Colorectal, n=12 On-label, n=1 • PFS ratio >1.3: 43.2% vs 5.3% (P<0.0001) underwent CGP with PCDx • PFS ratioa >1: 50% vs 19.3% (P=0.0011) Pancreas, n=8 Off-label, n=13 assay • Mean PFS ratioa: 1.34 vs 0.8 (P=0.05) Investigational, n=5 Bladder/urothelial, n=6 • Median PFS: 86 vs 49 days (HR: 0.55; Non-small cell lung 95% CI: 0.37–0.84; P=0.005) cancer, adeno, n=4 Melanoma, n=3 Neuroendocrine, n=3 Ovarian, n=3 Cholangiocarcinoma, esophageal, head and neck, non-small cell lung cancer, squamous cell, small cell lung cancer, n=2 each Adrenal cortical carcinoma, ampulla of Vater, cervical, chondrosarcoma, endometrial, Ewing, glioblastoma multiforme, hepatocellular, myoepithelial, prostate, renal cell, small bowel, squamous cell carcinoma NOS, thyroid anaplastic,

CONFIDENTIAL PAGE 177 US-FDX-2000073 Last update: January 22, 2021

Potentially actionable Reference Study design Tumor type alterations Clinical outcomes unknown primary; n=1 each Schwaederle et al. Meta-analysis of 346 Solid, n=9,946 Not available Matched vs nonmatched therapyc: 12 (2016) phase 1 studies of Hematologic, n=3,257 • RR: 31% vs 5% (P<0.001) hematologic and solid • Median RRs were significantly higher if the tumors receiving matched biomarker was a genomic alteration vs a therapies (N=13,203) protein biomarker (42% vs 22.4%; P=0.001) published in a 3-year time • Median PFS: 5.70 vs 2.95 months (P<0.001) period IMPACT/ Prospective study of Tumor type ≥1 somatic mutation, All patients, after a median follow-up of 18 COMPACT362 patients with 8 advanced profiled/enrolled: n=938/1640 (57%) months solid tumor types (N=1,893) Gynecological, n=405/430 Enrolled in • 245 patients who were tested were whose tissue was profiled Breast, n=310/341 genotype-matched subsequently treated on 277 therapeutic with a targeted NGS panel: trials, n=84/1640 clinical trials MALDI-TOF MS, TruSeq Colorectal, n=299/326 (5%) Matched vs unmatched therapy: Amplicon Cancer Panel, and Lung, n=256/339 • ORR: 19% vs 9% (P=0.026) Ion AmpliSeq Cancer Panel Pancreatobiliary, • Median OS: 16 vs 13 months (P=0.10) n=104/151 Upper aerodigestive, n=102/115 Other, n=81/99 Genitourinary, n=74/92 Jardim et al. Meta-analysis of 112 trials Solid, n=67 Targeted agents, Matched vs nonmatched therapye: 119 (2015) assessing 58 approved Hematologic, n=45 n=43 (74%) • RR: 48% vs 23% (P<0.001) cancer therapies (multiple • Median PFS: 8.3 vs 5.5 months (adjusted cancer diagnoses) in P=0.002) hematologic and solid tumor • Median OS: 19.3 vs 13.5 months (adjusted types P=0.04)

CONFIDENTIAL PAGE 178 US-FDX-2000073 Last update: January 22, 2021

Potentially actionable Reference Study design Tumor type alterations Clinical outcomes NEXT-1363 Prospective study of Gastric adenocarcinoma, Ongoing clinical trial Response rates, genome-matched (n=103) vs patients with 25 metastatic n=133 of a molecularly non-matched treatment (n=226): solid tumors (N=428) who NSCLC, n=94 targeted agent, n=31 • Gastrointestinal/hepatobiliary/rare tumors underwent broad molecular Colorectal (n=231): 42.6% vs 24.3% (P=0.009) profiling with Ion AmpliSeq adenocarcinoma, n=60 Molecularly targeted • Lung cancer cohort (n=98): 61.2% vs 28.6% Cancer Hotspot Panel v2 agents in the context (P=0.001) and nCounter Copy Number Soft tissue sarcoma, n=25 of clinical practice, • PR: 51.5% vs 25.2% Variation Assayb Hepatocellular, n=21 n=72 • SD: 30.1% vs 40.3% Gastroenteropancreatic • PD: 13.6% vs 22.6% • On-label, n=42 neuroendocrine tumor, • Not available: 4.9% vs 18.5% • Off-label, n=30 n=15 Targeted agents but Cholangiocarcinoma, did not receive the n=14 treatment due to Melanoma, n=12 deterioration of Pancreatic disease, n=3 adenocarcinoma, n=9 Gallbladder, n=5 Small cell lung, n=5 Gastrointestinal stromal tumors, n=4 Metastatic carcinoma of unknown origin, n=4 Thymic carcinoma, n=4 Esophageal squamous carcinoma, n=3 Non-melanoma skin, n=3 Ovarian, n=3 Renal cell, n=3 Ampulla of Vater cancer, n=3

CONFIDENTIAL PAGE 179 US-FDX-2000073 Last update: January 22, 2021

Potentially actionable Reference Study design Tumor type alterations Clinical outcomes Bladder, n=2 Uterine/cervix, n=2 Adrenocortical carcinoma, duodenal adenocarcinoma, glottic, peritoneal mesothelioma; n=1 each Schwaederle et al. Meta-analysis of 570 Lung, n=133 Not available Matched vs nonmatched therapyc: 48 (2015) phase 2 studies of 641 Hematologic, n=104 • RR: 31% vs 10.5% (P<0.001) single agents (N=32,149) Gastrointestinal, n=92 • Median PFS: 5.9 vs 2.7 months (P<0.001) published in a 3-year time • Median OS: 13.7 vs 8.9 months (P<0.001) Urinary, n=73 period • Treatment-related death rate: 1.5% vs 2.3% Breast, n=54 (P<0.001) Gynecologic, n=53 • Combined: Matched arms using a genomic Skin/melanoma, n=38 biomarker had higher median RR and prolonged median PFS and OS (all P≤0.05) Brain, n=36 compared with matched arms using a protein Other, n=34 biomarker Head and neck, n=24 Spetzler et al. Prospective cohort study of Ovary, n=308 Not available Matched vs unmatched therapy: 364 (2015) patients with solid tumors Breast, n=149 • Median OS, from time of profiling: 1,068 vs (N=1,180) Female genital tract 646 days (HR: 0.68; P=0.001) malignancy, n=123 • Median OS, from diagnosis: 978 vs 580 days (HR: 0.714; P=0.0003) Colorectal, n=120 Non-small cell lung cancer, n=105 Urinary tract, n=30 Neuroendocrine tumors, n=25 Gastroesophageal, n=24

CONFIDENTIAL PAGE 180 US-FDX-2000073 Last update: January 22, 2021

Potentially actionable Reference Study design Tumor type alterations Clinical outcomes Pancreatic adenocarcinoma, n=22 Soft tissue tumors, n=22 Leiomyosarcoma, n=17 Head and neck squamous carcinoma, n=14 Cancer of unknown primary, n=9 Unknown, n=9 Melanoma, n=7 Cholangiocarcinoma, n=6 Anal, n=5 Glioblastoma, n=5 Prostatic adenocarcinoma, n=5 Liver hepatocellular carcinoma, n=4 Non-epithelial ovarian cancer, n=4 Epithelial skin cancer, n=2 Lymphoma, n=2 Major and minor salivary glands, n=2 Adrenal cortical carcinoma, lung bronchioloalveolar carcinoma, mesothelioma, neuroblastoma, paragangliomas, small intestinal malignancies,

CONFIDENTIAL PAGE 181 US-FDX-2000073 Last update: January 22, 2021

Potentially actionable Reference Study design Tumor type alterations Clinical outcomes thyroid carcinoma, uveal melanoma; n=1 each Tsimberidou et al. Prospective study of Melanoma, n=291 Not available Matched vs unmatched therapyd: 365 (2012) patients with advanced or Colorectal, n=62 • ORR: 27% (CR: 2%; PR: 25%) vs 5% (all metastatic cancer (N=1,144) Thyroid, n=34 PRs; P<0.0001) with ≥1 molecular • SD ≥6 months: 23% vs 10% Other, n=23 aberration who received • Median TTF: 5.2 vs 2.2 months (P<0.0001) matched or unmatched Lung, n=22 • Median OS: 13.4 vs 9.0 months (P=0.017) therapy Breast, n=16 Ovarian, n=12 Genitourinary, other, n=10 Endometrial, n=9 Gastrointestinal, other, n=8 Pancreatic, n=8 Head and neck, n=7 CGP, comprehensive genomic profiling; CI, confidence interval; CR, complete response; HR, hazard ratio; NGS, next-generation sequencing; NOS, not otherwise specified; ORR, objective response rate; OR, odds ratio; OS, overall survival; PD, progressive disease; PFS, progression-free survival; PR, partial response; RR, response rate; SD, stable disease; TP53, tumor protein P53; TTF, time to treatment failure; VEGF, vascular endothelial growth factor. a PFS ratio refers to the PFS while receiving the current therapy line vs the previous therapy line. b Test can be considered broad molecular profiling, which is based on NGS technology. c In both studies, matched therapy was defined as cases in which a cognate biomarker was used for treatment indication, or no cognate biomarker was used but ≥50% of patients were known to harbor the cognate biomarker. d Matched therapy included available genome-matched trials and genome-matched treatments in practice. e Personalized therapy was considered to be any biomarker-matched targeted therapy.

CONFIDENTIAL PAGE 182 US-FDX-2000073 Last update: January 22, 2021

References 1. Foundation Medicine Inc. FoundationOne CDx™ Technical Information. 2020; May 26, 2020. Available at: www.f1cdxlabel.com. 2. Kakushadze ZR, Rakesh; Yu, Willie. Estimating Cost Savings from Early Cancer Diagnosis. Data. 2017;2(30):1-16. https://www.mdpi.com/2306-5729/2/3/30. 3. Kantar Health. CancerMPact Epidemiology, US. April 23, 2020 2020. 4. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2019. CA Cancer J Clin. 2019;69(1):7-34. https://pubmed.ncbi.nlm.nih.gov/30620402/. 5. Lux MP, Lewis K, Rider A, Niyazov A. Abstract P6-08-09: BRCA1/2 testing in adult women with HER2- advanced breast cancer (ABC): Results from a US real world study. Cancer research. 2020;80(4 Supplement):P6-08-09-P06-08-09. https://cancerres.aacrjournals.org/content/80/4_Supplement/P6-08-09. 6. Wong CI, Zerillo JA, Stuver SO, Siegel J, Jacobson JO, McNiff KK. Adverse events as a cause of unscheduled hospitalizations for solid tumor oncology patients receiving chemotherapy. Journal of Clinical Oncology. 2018;36(15_suppl):e18815-e18815. https://ascopubs.org/doi/abs/10.1200/JCO.2018.36.15_suppl.e18815. 7. White Al-Habeeb N, Kulasingam V, Diamandis EP, et al. The Use of Targeted Therapies for Precision Medicine in Oncology. Clinical chemistry. 2016;62(12):1556-1564. https://pubmed.ncbi.nlm.nih.gov/27679436/. 8. Hassett MJ, O'Malley AJ, Pakes JR, Newhouse JP, Earle CC. Frequency and cost of chemotherapy-related serious adverse effects in a population sample of women with breast cancer. Journal of the National Cancer Institute. 2006;98(16):1108-1117. https://pubmed.ncbi.nlm.nih.gov/16912263/. 9. Ashdown ML, Robinson AP, Yatomi-Clarke SL, et al. Chemotherapy for late-stage cancer patients: Meta-analysis of complete response rates. F1000Research. 2015;4:232. https://pubmed.ncbi.nlm.nih.gov/26834979/. 10. Massard C, Michiels S, Ferte C, et al. High-Throughput Genomics and Clinical Outcome in Hard- to-Treat Advanced Cancers: Results of the MOSCATO 01 Trial. Cancer Discov. 2017;7(6):586- 595. https://pubmed.ncbi.nlm.nih.gov/28365644/. 11. Haslem DS, Van Norman SB, Fulde G, et al. A retrospective analysis of precision medicine outcomes in patients with advanced cancer reveals improved progression-free survival without increased health care costs. Journal of oncology practice. 2017;13(2):e108-e119. https://pubmed.ncbi.nlm.nih.gov/27601506/. 12. Schwaederle M, Zhao M, Lee JJ, et al. Association of biomarker-based treatment strategies with response rates and progression-free survival in refractory malignant neoplasms: A meta-analysis. JAMA oncology. 2016;2(11):1452-1459. https://pubmed.ncbi.nlm.nih.gov/27273579/. 13. Kopetz S, Shaw KRM, Lee JJ, et al. Use of a Targeted Exome Next-Generation Sequencing Panel Offers Therapeutic Opportunity and Clinical Benefit in a Subset of Patients With Advanced Cancers. JCO Precision Oncology. 2019(3):1-14. https://ascopubs.org/doi/abs/10.1200/PO.18.00213.

CONFIDENTIAL PAGE 183 US-FDX-2000073 Last update: January 22, 2021

14. Washington State Health Care Authority. Scott Ramsey Summit Presentation. 2020; April 10, 2020. Available at: https://www.hca.wa.gov/assets/program/scott-ramsey-summit- presentation.pdf. 15. Hutchinson Institute for Cancer Outcomes Research. Community Cancer Care in Washington State: Quality and Cost Report 2019. 2019; April 10, 2020. Available at: https://www.fredhutch.org/content/dam/www/research/institute-networks-ircs/hicor/HICOR- Community-Cancer-Care-Report-2019.pdf. 16. Kaye DR, Min HS, Herrel LA, Dupree JM, Ellimoottil C, Miller DC. Costs of Cancer Care Across the Disease Continuum. The oncologist. 2018;23(7):798-805. https://theoncologist.onlinelibrary.wiley.com/doi/full/10.1634/theoncologist.2017-0481. 17. Sun L, Legood R, Dos-Santos-Silva I, Gaiha SM, Sadique Z. Global treatment costs of breast cancer by stage: A systematic review. PloS one. 2018;13(11):e0207993. https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0207993. 18. Institute of Medicine and National Research Council. Fulfilling the Potential of Cancer Prevention and Early Detection. 2003; April 10, 2020. Available at: https://www.nap.edu/catalog/10263/fulfilling-the-potential-of-cancer-prevention-and-early- detection. 19. Buffery D. Innovation Tops Current Trends in the 2016 Oncology Drug Pipeline. American health & drug benefits. 2016;9(4):233-238. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5004820/. 20. IQVIA. Global Oncology Trends 2019. Institute Report. 2019; April 21, 2020. Available at: https://www.iqvia.com/insights/the-iqvia-institute/reports/global-oncology-trends-2019. 21. Chawla A, Peeples M, Li N, Anhorn R, Ryan J, Signorovitch J. Real-world utilization of molecular diagnostic testing and matched drug therapies in the treatment of metastatic cancers. J Med Econ. 2018;21(6):543-552. https://www.ncbi.nlm.nih.gov/pubmed/29295635. 22. Mason C, Ellis PG, Lokay K, et al. Patterns of Biomarker Testing Rates and Appropriate Use of Targeted Therapy in the First-Line, Metastatic Non-Small Cell Lung Cancer Treatment Setting. Journal of clinical pathways : the foundation of value-based care. 2018;4(1):49-54. https://www.journalofclinicalpathways.com/article/patterns-biomarker-testing-rates-and- appropriate-use-targeted-therapy-first-line-metastatic. 23. Gutierrez ME, Price KS, Lanman RB, et al. Genomic Profiling for KRAS, NRAS, BRAF, Microsatellite Instability, and Mismatch Repair Deficiency Among Patients With Metastatic Colon Cancer. JCO Precision Oncology. 2019(3):1-9. https://ascopubs.org/doi/abs/10.1200/PO.19.00274. 24. Gutierrez ME, Choi K, Lanman RB, et al. Genomic profiling of advanced non-small cell lung cancer in community settings: Gaps and opportunities. Clinical lung cancer. 2017;18(6):651-659. https://www.sciencedirect.com/science/article/pii/S1525730417301092?via%3Dihub. 25. Keeling P. Transforming the pharm business model to prevent lost treatment opportunities and optimize return on investment. Diaceutics White Paper. 2017. Available at: https://www.diaceutics.com/docs/ft-2017-white-paper.pdf. 26. Audibert CM SM, Glass DJ, et al. Trends in the Molecular Diagnosis of Lung Cancer: Results from an Online Market Research Survey Friends of Cancer Research. 2016:1-19.

CONFIDENTIAL PAGE 184 US-FDX-2000073 Last update: January 22, 2021

https://www.focr.org/sites/default/files/pdf/FINAL%202017%20Friends%20NSCLC%20White% 20Paper.pdf. 27. Referenced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Breast Cancer V.6.2020. © National Comprehensive Cancer Network, Inc.; 2020. All rights reserved. Accessed November 19, 2020. To view the most recent and complete version of the guideline, go online to NCCN.org. 28. Referenced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Bladder Cancer V.6.2020. © National Comprehensive Cancer Network, Inc; 2020. All rights reserved. Accessed September 15, 2020. To view the most recent and complete version of the guideline, go online to NCCN.org. 29. Referenced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Central Nervous System Cancers V.3.2020. © National Comprehensive Cancer Network, Inc; 2020. All rights reserved. Accessed September 15, 2020. To view the most recent and complete version of the guideline, go online to NCCN.org. 30. Referenced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Colon Cancer V.1.2021. © National Comprehensive Cancer Network, Inc; 2020. All rights reserved. Accessed December 28, 2020. To view the most recent and complete version of the guideline, go online to NCCN.org. 31. Referenced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Melanoma: Cutaneous V.1.2021. © National Comprehensive Cancer Network, Inc; 2020. All rights reserved. Accessed December 7, 2020. To view the most recent and complete version of the guideline, go online to NCCN.org. 32. Referenced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Esophageal and Esophagogastric Junction Cancers V.5.2020. © National Comprehensive Cancer Network, Inc; 2020. All rights reserved. Accessed December 28, 2020. To view the most recent and complete version of the guideline, go online to NCCN.org. 33. Referenced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Gastric Cancer V.4.2020. © National Comprehensive Cancer Network, Inc; 2020. All rights reserved. Accessed December 28, 2020. To view the most recent and complete version of the guideline, go online to NCCN.org. 34. Referenced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Hepatobiliary Cancers V.5.2020. © National Comprehensive Cancer Network, Inc; 2020. All rights reserved. Accessed September 9, 2020. To view the most recent and complete version of the guideline, go online to NCCN.org. 35. Referenced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Non-Small Cell Lung Cancer V.2.2021. © National Comprehensive Cancer Network, Inc; 2020. All rights reserved. Accessed December 28, 2020. To view the most recent and complete version of the guideline, go online to NCCN.org. 36. Referenced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Ovarian Cancer V.1.2020. © National Comprehensive Cancer Network, Inc; 2020. All rights reserved. Accessed March 29, 2020. To view the most recent and complete version of the guideline, go online to NCCN.org. 37. Referenced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Pancreatic Adenocarcinoma V.1.2021. © National Comprehensive Cancer

CONFIDENTIAL PAGE 185 US-FDX-2000073 Last update: January 22, 2021

Network, Inc; 2020. All rights reserved. Accessed December 7, 2020. To view the most recent and complete version of the guideline, go online to NCCN.org. 38. Referenced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Prostate Cancer V.3.2020. © National Comprehensive Cancer Network, Inc; 2020. All rights reserved. Accessed November 19, 2020. To view the most recent and complete version of the guideline, go online to NCCN.org. 39. Referenced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Rectal Cancer V.1.2021. © National Comprehensive Cancer Network, Inc; 2020. All rights reserved. Accessed December 28, 2020. To view the most recent and complete version of the guideline, go online to NCCN.org. 40. Referenced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Uterine Neoplasms V.1.2021. © National Comprehensive Cancer Network, Inc; 2020. All rights reserved. Accessed December 7, 2020. To view the most recent and complete version of the guideline, go online to NCCN.org. 41. Referenced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Vulvar Cancer V.2.2021. © National Comprehensive Cancer Network, Inc; 2020. All rights reserved. Accessed December 7, 2020. To view the most recent and complete version of the guideline, go online to NCCN.org. 42. Jarvis L. Cancer, redefined. 2017; September 10, 2020. Available at: https://cen.acs.org/articles/95/i27/Cancer-redefined.html. 43. Keytruda (pembrolizumab) [prescribing information]. Merck Sharp & Dohme Corp. June 2020; April 23, 2020. Available at: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=9333c79b-d487-4538-a9f0- 71b91a02b287. 44. US Food and Drug Administration. FDA approves larotrectinib for solid tumors with NTRK gene fusions. 2018; July 23, 2019. Available at: https://www.fda.gov/drugs/fda-approves-larotrectinib- solid-tumors-ntrk-gene-fusions. 45. US Food and Drug Administration. FDA approves entrectinib for NTRK solid tumors and ROS-1 NSCLC. 2019; September 10, 2019. Available at: https://www.fda.gov/drugs/resources- information-approved-drugs/fda-approves-entrectinib-ntrk-solid-tumors-and-ros-1-nsclc. 46. El-Deiry WS, Goldberg RM, Lenz HJ, et al. The current state of molecular testing in the treatment of patients with solid tumors, 2019. CA Cancer J Clin. 2019. http://www.ncbi.nlm.nih.gov/pubmed/31116423. 47. Wongchenko M, Dent R, Kim S, et al. Cell-free (CF)DNA analysis identifies PIK3CA/AKT1 mutations associated with greater PFS improvement from the addition of ipatasertib (IPAT) to paclitaxel (P) in triple-negative breast cancer (TNBC). Annals of oncology : official journal of the European Society for Medical Oncology. 2017;28(5 Suppl):v22-v42. https://oncologypro.esmo.org/meeting-resources/esmo-2017-congress/Cell-free-cf-DNA-analysis- identifies-PIK3CA-AKT1-mutations-associated-with-greater-PFS-improvement-from-the- addition-of-ipatasertib-IPAT-to-paclitaxel-P-in-triple-negative-breast-cancer-TNBC. 48. Schwaederle M, Zhao M, Lee JJ, et al. Impact of precision medicine in diverse cancers: A meta- analysis of phase II clinical trials. J Clin Oncol. 2015;33(32):3817-3825. https://ascopubs.org/doi/10.1200/JCO.2015.61.5997.

CONFIDENTIAL PAGE 186 US-FDX-2000073 Last update: January 22, 2021

49. Trabucco SE, Gowen K, Maund SL, et al. A Novel Next-Generation Sequencing Approach to Detecting Microsatellite Instability and Pan-Tumor Characterization of 1000 Microsatellite Instability-High Cases in 67,000 Patient Samples. The Journal of molecular diagnostics : JMD. 2019;21(6):1053-1066. https://www.sciencedirect.com/science/article/pii/S1525157819303587?via%3Dihub. 50. Gandara DR, Paul SM, Kowanetz M, et al. Blood-based tumor mutational burden as a predictor of clinical benefit in non-small-cell lung cancer patients treated with atezolizumab. Nat Med. 2018;24(9):1441-1448. https://www.nature.com/articles/s41591-018-0134-3. 51. Chalmers ZR, Connelly CF, Fabrizio D, et al. Analysis of 100,000 human cancer genomes reveals the landscape of tumor mutational burden. Genome medicine. 2017;9(1):34. https://genomemedicine.biomedcentral.com/articles/10.1186/s13073-017-0424-2. 52. Hellmann MD, Ciuleanu TE, Pluzanski A, et al. Nivolumab plus ipilimumab in lung cancer with a high tumor mutational burden. New Engl J Med. 2018;378(22):2093-2104. https://www.nejm.org/doi/full/10.1056/NEJMoa1801946. 53. Johnson DB, Dahlman KH, Knol J, et al. Enabling a genetically informed approach to cancer medicine: a retrospective evaluation of the impact of comprehensive tumor profiling using a targeted next-generation sequencing panel. The oncologist. 2014;19(6):616-622. https://theoncologist.onlinelibrary.wiley.com/doi/full/10.1634/theoncologist.2014-0011. 54. Colombo N, Sessa C, du Bois A, et al. ESMO-ESGO consensus conference recommendations on ovarian cancer: pathology and molecular biology, early and advanced stages, borderline tumours and recurrent disease†. Annals of oncology : official journal of the European Society for Medical Oncology. 2019;30(5):672-705. https://www.sciencedirect.com/science/article/pii/S0923753419311627?via%3Dihub. 55. Marquart J, Chen EY, Prasad V. Estimation of the Percentage of US Patients With Cancer Who Benefit From Genome-Driven Oncology. JAMA oncology. 2018;4(8):1093-1098. https://jamanetwork.com/journals/jamaoncology/fullarticle/2678901. 56. Parsons DW, Roy A, Yang Y, et al. Diagnostic Yield of Clinical Tumor and Germline Whole- Exome Sequencing for Children With Solid Tumors. JAMA oncology. 2016;2(5):616-624. https://jamanetwork.com/journals/jamaoncology/fullarticle/2484356. 57. Harris MH, DuBois SG, Glade Bender JL, et al. Multicenter Feasibility Study of Tumor Molecular Profiling to Inform Therapeutic Decisions in Advanced Pediatric Solid Tumors: The Individualized Cancer Therapy (iCat) Study. JAMA oncology. 2016;2(5):608-615. https://jamanetwork.com/journals/jamaoncology/fullarticle/2484355. 58. Boland GM, Piha-Paul SA, Subbiah V, et al. Clinical next generation sequencing to identify actionable aberrations in a phase I program. Oncotarget. 2015;6(24):20099-20110. https://www.oncotarget.com/article/4040/text/. 59. Powell SF, Dib EG, Bleeker JS, et al. Delivering precision oncology in a community cancer program: Results from a prospective observational study. JCO Precis Oncol. 2018(2):1-12. http://ascopubs.org/doi/abs/10.1200/PO.17.00220. 60. Sadaps M, Funchain P, Mahdi H, et al. Precision oncology in solid tumors: A longitudinal tertiary care center experience. JCO Precis Oncol. 2018;doi:10.1200/PO.18.00186. http://ascopubs.org/doi/abs/10.1200/PO.18.00186.

CONFIDENTIAL PAGE 187 US-FDX-2000073 Last update: January 22, 2021

61. Reitsma M, Fox J, Vanden Borre P, et al. Impact of a collaboration between a health plan, oncology practice, and comprehensive genomic profiling company from the payer perspective. J Manag Care Spec Pharm. 2019;doi:10.18553/jmcp.2019.18309. https://www.jmcp.org/doi/10.18553/jmcp.2019.18309. 62. Hirshfield KM, Tolkunov D, Zhong H, et al. Clinical actionability of comprehensive genomic profiling for management of rare or refractory cancers. The oncologist. 2016;21(11):1315-1325. https://theoncologist.onlinelibrary.wiley.com/doi/full/10.1634/theoncologist.2016-0049. 63. Kato S, Kurasaki K, Ikeda S, Kurzrock R. Rare Tumor Clinic: The University of California San Diego Moores Cancer Center experience with a precision therapy approach. The oncologist. 2018;23(2):171-178. https://theoncologist.onlinelibrary.wiley.com/doi/full/10.1634/theoncologist.2017-0199. 64. Rodriguez-Rodriguez L, Hirshfield KM, Rojas V, et al. Use of comprehensive genomic profiling to direct point-of-care management of patients with gynecologic cancers. Gynecologic oncology. 2016;141(1):2-9. https://www.sciencedirect.com/science/article/pii/S0090825816300452?via%3Dihub. 65. Schwaederle M, Daniels GA, Piccioni DE, et al. On the road to precision cancer medicine: Analysis of genomic biomarker actionability in 439 patients. Molecular cancer therapeutics. 2015;14(6):1488-1494. https://mct.aacrjournals.org/content/14/6/1488. 66. Sicklick JK, Kato S, Okamura R, et al. Molecular profiling of cancer patients enables personalized combination therapy: the I-PREDICT study. Nat Med. 2019;25(5):744-750. http://www.ncbi.nlm.nih.gov/pubmed/31011206. 67. Wheler JJ, Janku F, Naing A, et al. Cancer therapy directed by comprehensive genomic profiling: A single center study. Cancer research. 2016;76(13):3690-3701. https://cancerres.aacrjournals.org/content/76/13/3690. 68. Boussemart L, Nelson A, Wong M, et al. Hybrid Capture-Based Genomic Profiling Identifies BRAF V600 and Non-V600 Alterations in Melanoma Samples Negative by Prior Testing. The oncologist. 2019;24(5):657-663. https://theoncologist.onlinelibrary.wiley.com/doi/full/10.1634/theoncologist.2018-0271 69. Rankin A, Klempner SJ, Erlich R, et al. Broad Detection of Alterations Predicted to Confer Lack of Benefit From EGFR Antibodies or Sensitivity to Targeted Therapy in Advanced Colorectal Cancer. The oncologist. 2016;21(11):1306-1314. https://theoncologist.onlinelibrary.wiley.com/doi/full/10.1634/theoncologist.2016-0148. 70. Vasan N, Razavi P, Johnson JL, et al. Double PIK3CA mutations in cis increase oncogenicity and sensitivity to PI3Kα inhibitors. Science (New York, N.Y.). 2019;366(6466):714-723. https://science.sciencemag.org/content/366/6466/714. 71. Rozenblum AB, Ilouze M, Dudnik E, et al. Clinical impact of hybrid capture-based next- generation sequencing on changes in treatment decisions in lung cancer. Journal of thoracic oncology : official publication of the International Association for the Study of Lung Cancer. 2017;12(2):258-268. https://www.sciencedirect.com/science/article/pii/S1556086416312412?via%3Dihub. 72. Deverka P, Messner DA, McCormack R, et al. Generating and evaluating evidence of the clinical utility of molecular diagnostic tests in oncology. Genetics in medicine : official journal of the American College of Medical Genetics. 2016;18(8):780-787. https://www.nature.com/articles/gim2015162.

CONFIDENTIAL PAGE 188 US-FDX-2000073 Last update: January 22, 2021

73. Freedman AN, Klabunde CN, Wiant K, et al. Use of Next-Generation Sequencing Tests to Guide Cancer Treatment: Results From a Nationally Representative Survey of Oncologists in the United States. JCO Precision Oncology. 2018(2):1-13. https://ascopubs.org/doi/abs/10.1200/PO.18.00169. 74. Schwaederle M, Parker BA, Schwab RB, et al. Precision oncology: The UC San Diego Moores Cancer Center PREDICT experience. Molecular cancer therapeutics. 2016;15(4):743-752. https://mct.aacrjournals.org/content/15/4/743. 75. Pishvaian MJ, Bender RJ, Halverson D, et al. Molecular profiling of patients with pancreatic cancer: Initial results from the know your tumor initiative. Clin Cancer Res. 2018;24(20):5018- 5027. https://clincancerres.aacrjournals.org/content/24/20/5018. 76. American Cancer Society. Barriers to patient enrollment in therapeutic clinical trials for cancer: A landscape report. 2018; January 14, 2018. Available at: https://www.fightcancer.org/sites/default/files/National%20Documents/Clinical-Trials-Executive- Summary.pdf. 77. Ganesan P, Moulder S, Lee JJ, et al. Triple-negative breast cancer patients treated at MD Anderson Cancer Center in phase I trials: improved outcomes with combination chemotherapy and targeted agents. Molecular cancer therapeutics. 2014;13(12):3175-3184. https://pubmed.ncbi.nlm.nih.gov/25253784/. 78. Pishvaian MJ, Blais EM, Brody JR, et al. Overall survival in patients with pancreatic cancer receiving matched therapies following molecular profiling: a retrospective analysis of the Know Your Tumor registry trial. The Lancet. Oncology. 2020;21(4):508-518. https://www.sciencedirect.com/science/article/abs/pii/S1470204520300747?via%3Dihub. 79. Singal G, Miller PG, Agarwala V, et al. Association of Patient Characteristics and Tumor Genomics With Clinical Outcomes Among Patients With Non-Small Cell Lung Cancer Using a Clinicogenomic Database. Jama. 2019;321(14):1391-1399. https://jamanetwork.com/journals/jama/fullarticle/2730114. 80. Alecensa (alectinib) [prescribing information]. Genentech, Inc. June 2018; April 23, 2020. Available at: https://www.gene.com/download/pdf/alecensa_prescribing.pdf. 81. Cotellic (cobimetinib) [prescribing information]. Genentech, Inc. January 2018; April 23, 2020. Available at: https://www.gene.com/download/pdf/cotellic_prescribing.pdf. 82. Erbitux (cetuximab) [prescribing information]. ImClone LLC. April 2019; April 23, 2020. Available at: https://uspl.lilly.com/erbitux/erbitux.html. 83. Gilotrif (afatinib) [prescribing information]. Boehringer Ingelheim Pharmaceuticals, Inc. October 2019; April 23, 2020. Available at: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=fd638e5e-8032-e7ca-0179- 95e96ab5d387. 84. Herceptin (trastuzumab) [prescribing information]. Genentech, Inc. November 2018; April 23, 2020. Available at: hhttps://www.gene.com/download/pdf/herceptin_prescribing.pdf. 85. Iressa (gefitinib) [prescribing information]. AstraZeneca Pharmaceuticals LP. May 2019; April 23, 2020. Available at: https://www.azpicentral.com/iressa/iressa.pdf#page=1. 86. Mekinist (trametinib) [prescribing information]. Novartis Pharmaceuticals Corporation. June 2020; August 11, 2020. Available at:

CONFIDENTIAL PAGE 189 US-FDX-2000073 Last update: January 22, 2021

https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=0002ad27-779d-42ab-83b5- bc65453412a1. 87. Perjeta (pertuzumab) [prescribing information]. Genentech, Inc. January 2020; April 23, 2020. Available at: https://www.gene.com/download/pdf/perjeta_prescribing.pdf. 88. Piqray (alpelisib) [prescribing information]. Novartis Pharmaceuticals Corporation. May 2019; April 23, 2020. Available at: https://www.novartis.us/sites/www.novartis.us/files/piqray.pdf. 89. Tafinlar (dabrafenib) [prescribing information]. Novartis Pharmaceuticals Corporation. April 2020; April 23, 2020. Available at: https://www.novartis.us/sites/www.novartis.us/files/tafinlar.pdf. 90. Tagrisso (osimertinib) [prescribing information]. AstraZeneca Pharmaceuticals LP. June 2020; August 11, 2020. Available at: https://www.azpicentral.com/tagrisso/tagrisso.pdf#page=1. 91. Tarceva (erlotinib) [prescribing information]. Genentech, Inc. October 2016; April 23, 2020. Available at: https://www.gene.com/download/pdf/tarceva_prescribing.pdf. 92. Vectibix (panitumumab) [prescribing information]. Incyte Corporation. June 2017; September 15, 2020. Available at: https://www.pi.amgen.com/~/media/amgen/repositorysites/pi-amgen- com/vectibix/vectibix_pi.pdf. 93. Xalkori (crizotinib) [prescribing information]. Pfizer Laboratories Div Pfizer Inc. June 2019; April 23, 2020. Available at: http://labeling.pfizer.com/showlabeling.aspx?id=676. 94. Zelboraf (vemurafenib) [prescribing information]. Genentech, Inc. May 2020; August 11, 2020. Available at: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=38eea320-7e0c-485a- bc30-98c3c45e2763. 95. Zykadia (ceritinib) [prescribing information]. Novartis Pharmaceuticals Corporation. March 2019; April 23, 2020. Available at: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=fff5d805-4ffd-4e8e-8e63- 6f129697563e. 96. Keytruda® [pembrolizumab] FDA Package Insert. Merck Sharp and Dohme. 2014; March 30, 2018. Available at: https://www.accessdata.fda.gov/drugsatfda_docs/label/2017/125514s031lbl.pdf. 97. Le Tourneau C, Delord JP, Goncalves A, et al. Molecularly targeted therapy based on tumour molecular profiling versus conventional therapy for advanced cancer (SHIVA): A multicentre, open-label, proof-of-concept, randomised, controlled phase 2 trial. The Lancet. Oncology. 2015;16(13):1324-1334. https://www.sciencedirect.com/science/article/abs/pii/S1470204515001886?via%3Dihub. 98. Radovich M, Kiel PJ, Nance SM, et al. Clinical benefit of a precision medicine based approach for guiding treatment of refractory cancers. Oncotarget. 2016;7(35):56491-56500. https://www.oncotarget.com/article/10606/text/. 99. Madison R, Schrock AB, Castellanos E, et al. Retrospective analysis of real-world data to determine clinical outcomes of patients with advanced non-small cell lung cancer following cell- free circulating tumor DNA genomic profiling. Lung cancer (Amsterdam, Netherlands). 2020;148:69-78. https://www.sciencedirect.com/science/article/pii/S0169500220305572?via%3Dihub.

CONFIDENTIAL PAGE 190 US-FDX-2000073 Last update: January 22, 2021

100. US Food and Drug Administration. List of Cleared or Approved Companion Diagnostic Devices (In Vitro and Imaging Tools). 2020; December 23, 2020. Available at: https://www.fda.gov/medical-devices/vitro-diagnostics/list-cleared-or-approved-companion- diagnostic-devices-vitro-and-imaging-tools. 101. Hong DS, DuBois SG, Kummar S, et al. Larotrectinib in patients with TRK fusion-positive solid tumours: a pooled analysis of three phase 1/2 clinical trials. The Lancet. Oncology. 2020;21(4):531-540. https://www.sciencedirect.com/science/article/abs/pii/S1470204519308563?via%3Dihub. 102. Drilon A, Laetsch TW, Kummar S, et al. Efficacy of Larotrectinib in TRK Fusion-Positive Cancers in Adults and Children. The New England journal of medicine. 2018;378(8):731-739. http://www.ncbi.nlm.nih.gov/pubmed/29466156. 103. Foundation Medicine Inc. FoundationOne CDx™ Summary of Safety and Effectiveness. S016. 2020; May 26, 2020. Available at: https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpma/pma.cfm?id=P170019S016. 104. Foundation Medicine Inc. FoundationOne CDx™ Summary of Safety and Effectiveness. S017. 2020; May 26, 2020. Available at: https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpma/pma.cfm?id=P170019S017. 105. Wolf J, Seto T, Han JY, et al. Capmatinib in MET Exon 14-Mutated or MET-Amplified Non- Small-Cell Lung Cancer. The New England journal of medicine. 2020;383(10):944-957. https://www.nejm.org/doi/10.1056/NEJMoa2002787. 106. Foundation Medicine Inc. FoundationOne CDx™ Summary of Safety and Effectiveness. S011. 2020; May 26, 2020. Available at: https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpma/pma.cfm?id=P170019S011. 107. Andre F, Ciruelos E, Rubovszky G, et al. Alpelisib for PIK3CA-Mutated, Hormone Receptor- Positive Advanced Breast Cancer. The New England journal of medicine. 2019;380(20):1929- 1940. https://www.nejm.org/doi/10.1056/NEJMoa1813904. 108. Swisher EM, Lin KK, Oza AM, et al. Rucaparib in relapsed, platinum-sensitive high-grade ovarian carcinoma (ARIEL2 Part 1): an international, multicentre, open-label, phase 2 trial. The Lancet. Oncology. 2017;18(1):75-87. http://www.ncbi.nlm.nih.gov/pubmed/27908594. 109. Moore K, Colombo N, Scambia G, et al. Maintenance Olaparib in Patients with Newly Diagnosed Advanced Ovarian Cancer. The New England journal of medicine. 2018;379(26):2495-2505. https://www.nejm.org/doi/10.1056/NEJMoa1810858. 110. Banerjee S, Moore KN, Colombo N, et al. 811MO Maintenance olaparib for patients (pts) with newly diagnosed, advanced ovarian cancer (OC) and a BRCA mutation (BRCAm): 5-year (y) follow-up (f/u) from SOLO1. Annals of Oncology. 2020;31:S613. http://www.sciencedirect.com/science/article/pii/S0923753420409469. 111. Hussain M, Mateo J, Fizazi K, et al. Survival with Olaparib in Metastatic Castration-Resistant Prostate Cancer. The New England journal of medicine. 2020. https://www.nejm.org/doi/10.1056/NEJMoa2022485. 112. Abou-Alfa GK, Sahai V, Hollebecque A, et al. Pemigatinib for previously treated, locally advanced or metastatic cholangiocarcinoma: a multicentre, open-label, phase 2 study. The Lancet. Oncology. 2020;21(5):671-684. https://www.sciencedirect.com/science/article/abs/pii/S1470204520301091?via%3Dihub.

CONFIDENTIAL PAGE 191 US-FDX-2000073 Last update: January 22, 2021

113. Chawla A, Janku F, Wheler JJ, et al. Estimated cost of anticancer therapy directed by comprehensive genomic profiling in a single-center study. JCO Precis Oncol. 2018;doi:10.1200/PO.18.00074. http://ascopubs.org/doi/abs/10.1200/PO.18.00074. 114. Signorovitch J, Zhou Z, Ryan J, Chawla A. Budget impact analysis of comprehensive genomic profiling in patients with advanced non-small cell lung cancer. J Med Econ. 2018;doi:10.1080/13696998.2018.1549056. https://www.tandfonline.com/doi/abs/10.1080/13696998.2018.1549056?journalCode=ijme20. 115. Siravegna G, Mussolin B, Venesio T, et al. How liquid biopsies can change clinical practice in oncology. Annals of oncology : official journal of the European Society for Medical Oncology. 2019;30(10):1580-1590. https://www.sciencedirect.com/science/article/pii/S0923753419609725?via%3Dihub. 116. National Cancer Institute. Cancer Stat Facts. 2020; April 21, 2020. Available at: https://seer.cancer.gov/statfacts/. 117. American Cancer Society. How is Chemotherapy Used to Treat Cancer? 2018; March 29, 2020. Available at: https://www.cancer.org/treatment/treatments-and-side-effects/treatment- types/chemotherapy/how-is-chemotherapy-used-to-treat-cancer.html. 118. Tsimberidou AM, Hong DS, Wheler JJ, et al. Precision medicine: Clinical outcomes including long-term survival according to the pathway targeted and treatment period–The IMPACT study. J Clin Oncol. 2018;36(18 Suppl):LBA2553. https://ascopubs.org/doi/abs/10.1200/JCO.2018.36.18_suppl.LBA2553. 119. Jardim DL, Schwaederle M, Wei C, et al. Impact of a biomarker-based strategy on oncology drug development: A meta-analysis of clinical trials leading to FDA approval. Journal of the National Cancer Institute. 2015;107(11):djv253. https://academic.oup.com/jnci/article/107/11/djv253/2457697. 120. Fred Hutchinson Cancer Research Center. Living with Stage 4: The breast cancer no one understands. 2016; April 21, 2020. Available at: https://www.fredhutch.org/en/news/center- news/2014/10/stage-4-metastatic-misunderstood-breast-cancer.html. 121. Pascale M, Azinwi CN, Marongiu B, Pesce G, Stoffel F, Roggero E. The outcome of prostate cancer patients treated with curative intent strongly depends on survival after metastatic progression. BMC cancer. 2017;17(1):651. https://bmccancer.biomedcentral.com/articles/10.1186/s12885-017-3617-6. 122. Berger KN, Pu JJ. PD-1 pathway and its clinical application: A 20-year journey after discovery of the complete human PD-1 gene. Gene. 2018;638:20-25. https://www.sciencedirect.com/science/article/abs/pii/S0378111917307825?via%3Dihub. 123. Campbell BB, Light N, Fabrizio D, et al. Comprehensive analysis of hypermutation in human cancer. Cell. 2017;171(5):1042-1056.e1010. https://www.sciencedirect.com/science/article/pii/S009286741731142X?via%3Dihub. 124. Felsenstein KM, Theodorescu D. Precision medicine for urothelial bladder cancer: update on tumour genomics and immunotherapy. Nature reviews. Urology. 2018;15(2):92-111. https://www.nature.com/articles/nrurol.2017.179. 125. Kou T, Kanai M, Matsumoto S, Okuno Y, Muto M. The possibility of clinical sequencing in the management of cancer. Japanese journal of clinical oncology. 2016;46(5):399-406. https://academic.oup.com/jjco/article/46/5/399/2218999.

CONFIDENTIAL PAGE 192 US-FDX-2000073 Last update: January 22, 2021

126. Pai SG, Carneiro BA, Chae YK, et al. Correlation of tumor mutational burden and treatment outcomes in patients with colorectal cancer. Journal of gastrointestinal oncology. 2017;8(5):858- 866. https://jgo.amegroups.com/article/view/14866/13046. 127. Yates LR, Seoane J, Le Tourneau C, et al. The European Society for Medical Oncology (ESMO) Precision Medicine Glossary. Annals of oncology : official journal of the European Society for Medical Oncology. 2018;29(1):30-35. https://www.sciencedirect.com/science/article/pii/S0923753419350112?via%3Dihub. 128. Precision Oncoloy News. Survey of Precision Oncology Programs Finds Agreement on Testing Divergence in Care Delivery. 2019; March 29, 2020. Available at: https://www.precisiononcologynews.com/cancer/survey-precision-oncology-programs-finds- agreement-testing-divergence-care-delivery#.XoE9CohKiUl. 129. Hoesing H. Clinical Practice Guidelines: Closing the Gap Betwen Theory and Practice. 2016; March 29, 2020. Available at: https://www.elsevier.com/__data/assets/pdf_file/0007/190177/JCI- Whitepaper_cpgs-closing-the-gap.pdf. 130. Katz SJ, Jagsi R, Morrow M. Reducing Overtreatment of Cancer With Precision Medicine: Just What the Doctor Ordered. Jama. 2018;319(11):1091-1092. https://jamanetwork.com/journals/jama/article-abstract/2673741. 131. Baudino TA. Targeted cancer therapy: The next generation of cancer treatment. Current drug discovery technologies. 2015;12(1):3-20. https://www.eurekaselect.com/131876/article. 132. Pardoll DM. The blockade of immune checkpoints in cancer immunotherapy. Nature reviews. Cancer. 2012;12(4):252-264. https://www.nature.com/articles/nrc3239. 133. Harrington KJ, Ferris RL, Blumenschein G, Jr., et al. Nivolumab versus standard, single-agent therapy of investigator's choice in recurrent or metastatic squamous cell carcinoma of the head and neck (CheckMate 141): health-related quality-of-life results from a randomised, phase 3 trial. The Lancet. Oncology. 2017;18(8):1104-1115. https://www.sciencedirect.com/science/article/abs/pii/S1470204517304217?via%3Dihub. 134. Reck M, Taylor F, Penrod JR, et al. Impact of nivolumab versus docetaxel on health-related quality of life and symptoms in patients with advanced squamous non-small cell lung cancer: Results from the CheckMate 017 study. Journal of thoracic oncology : official publication of the International Association for the Study of Lung Cancer. 2018;13(2):194-204. https://www.sciencedirect.com/science/article/pii/S1556086417328782?via%3Dihub. 135. Brahmer JR, Rodriguez-Abreu D, Robinson AG, et al. Health-related quality-of-life results for pembrolizumab versus chemotherapy in advanced, PD-L1-positive NSCLC (KEYNOTE-024): A multicentre, international, randomised, open-label phase 3 trial. The Lancet. Oncology. 2017;18(12):1600-1609. https://www.sciencedirect.com/science/article/abs/pii/S1470204517306903?via%3Dihub. 136. Schadendorf D, Dummer R, Hauschild A, et al. Health-related quality of life in the randomised KEYNOTE-002 study of pembrolizumab versus chemotherapy in patients with ipilimumab- refractory melanoma. Eur J Cancer. 2016;67:46-54. https://www.sciencedirect.com/science/article/pii/S0959804916323437?via%3Dihub. 137. Goodman AM, Kato S, Bazhenova L, et al. Tumor mutational burden as an independent predictor of response to immunotherapy in diverse cancers. Molecular cancer therapeutics. 2017;16(11):2598-2608. https://mct.aacrjournals.org/content/16/11/2598.

CONFIDENTIAL PAGE 193 US-FDX-2000073 Last update: January 22, 2021

138. Goodman AM, Sokol ES, Frampton GM, Lippman SM, Kurzrock R. Microsatellite-Stable Tumors with High Mutational Burden Benefit from Immunotherapy. Cancer Immunol Res. 2019;7(10):1570-1573. https://cancerimmunolres.aacrjournals.org/content/7/10/1570. 139. Le DT, Uram JN, Wang H, et al. PD-1 blockade in tumors with mismatch-repair deficiency. The New England journal of medicine. 2015;372(26):2509-2520. https://www.nejm.org/doi/10.1056/NEJMoa1500596. 140. Hou H, Liu D, Zhou N, et al. Broad, hybrid capture-based next generation sequencing identified actionable genomic alterations in HER2-negative breast cancer. Journal of Clinical Oncology. 2016;34(15_suppl):e12544-e12544. https://ascopubs.org/doi/abs/10.1200/JCO.2016.34.15_suppl.e12544. 141. Wang L, Page DR, Shewade A, et al. Utilization and timing of Foundation Medicine Inc. (FMI) testing in U.S. advanced non-small cell lung cancer (aNSCLC) patients. 18th IASLC World Conference on Lung Cancer. 2017. https://www.jto.org/article/S1556-0864(17)31420-X/fulltext. 142. Jordan EJ, Kim HR, Arcila ME, et al. Prospective Comprehensive Molecular Characterization of Lung Adenocarcinomas for Efficient Patient Matching to Approved and Emerging Therapies. Cancer Discov. 2017;7(6):596-609. https://cancerdiscovery.aacrjournals.org/content/7/6/596. 143. Shepherd FA. Targeted Therapy: The New Frontier. Presented at the 2019 American Society of Clinical Oncology (ASCO) Annual Meeting; May 31 - June 1, 2019; Chicago, IL. 2019. https://meetinglibrary.asco.org/record/168046/video. 144. Athie A, Arce-Gallego S, Gonzalez M, et al. Targeting DNA Repair Defects for Precision Medicine in Prostate Cancer. Current Oncology Reports. 2019;21(5):42. https://doi.org/10.1007/s11912-019-0790-6. 145. Warner EW, Yip SM, Chi KN, Wyatt AW. DNA repair defects in prostate cancer: impact for screening, prognostication and treatment. BJU international. 2019;123(5):769-776. https://bjui- journals.onlinelibrary.wiley.com/doi/abs/10.1111/bju.14576. 146. Gierman HJ, S. G, M. L, et al. Genomic testing and treatment landscape in patients with advanced non-small cell lung cancer (aNSCLC) using real-world data from community oncology practices. American Society of Clinical Oncology Annual Meeting; June 3, 2019; Chicago, IL. 2019. https://ascopubs.org/doi/abs/10.1200/JCO.2019.37.15_suppl.1585. 147. Kratz J, Burkard M, O'Meara T, Pusztai L, Veitch Z, Bedard PL. Incorporating Genomics Into the Care of Patients With Advanced Breast Cancer. American Society of Clinical Oncology educational book. American Society of Clinical Oncology. Annual Meeting. 2018;38:56-64. https://ascopubs.org/doi/10.1200/EDBK_200731. 148. Pauletti G, Dandekar S, Rong H, et al. Assessment of Methods for Tissue-Based Detection of the HER-2/neu Alteration in Human Breast Cancer: A Direct Comparison of Fluorescence In Situ Hybridization and Immunohistochemistry. Journal of Clinical Oncology. 2000;18(21):3651- 3664. https://ascopubs.org/doi/abs/10.1200/JCO.2000.18.21.3651. 149. Yaziji H, Goldstein LC, Barry TS, et al. HER-2 testing in breast cancer using parallel tissue- based methods. Jama. 2004;291(16):1972-1977. https://jamanetwork.com/journals/jama/fullarticle/198629. 150. Konstantinopoulos PA, Ceccaldi R, Shapiro GI, D'Andrea AD. Homologous Recombination Deficiency: Exploiting the Fundamental Vulnerability of Ovarian Cancer. Cancer Discov. 2015;5(11):1137-1154. https://cancerdiscovery.aacrjournals.org/content/5/11/1137.

CONFIDENTIAL PAGE 194 US-FDX-2000073 Last update: January 22, 2021

151. da Cunha Colombo Bonadio RR, Fogace RN, Miranda VC, Diz M. Homologous recombination deficiency in ovarian cancer: a review of its epidemiology and management. Clinics (Sao Paulo, Brazil). 2018;73(suppl 1):e450s. http://www.scielo.br/scielo.php?script=sci_arttext&pid=S1807- 59322018000200308&lng=en&nrm=iso&tlng=en. 152. Gee ME, Faraahi Z, McCormick A, Edmondson RJ. DNA damage repair in ovarian cancer: unlocking the heterogeneity. Journal of ovarian research. 2018;11(1):50. https://ovarianresearch.biomedcentral.com/articles/10.1186/s13048-018-0424-x. 153. Cohen R, Pudlarz T, Delattre JF, Colle R, André T. Molecular Targets for the Treatment of Metastatic Colorectal Cancer. Cancers. 2020;12(9). https://www.mdpi.com/2072- 6694/12/9/2350. 154. Rico A, Pollack LA, Thompson TD, et al. KRAS testing and first-line treatment among patients diagnosed with metastatic colorectal cancer using population data from ten National Program of Cancer Registries in the United States. Journal of cancer research & therapy. 2016;5(2):7-13. https://nobleresearch.org/Doi/10.14312/2052-4994.2017-2. 155. Landsman-Blumberg PB, Carter GC, Johnson BH, et al. Metastatic colorectal cancer treatment patterns according to kirsten rat sarcoma viral oncogene homolog genotype in U.S. Community- based oncology practices. Clinical colorectal cancer. 2014;13(3):178-184. https://www.sciencedirect.com/science/article/abs/pii/S1533002814000577?via%3Dihub. 156. Drilon A, Wang L, Arcila ME, et al. Broad, hybrid capture-based next-generation sequencing identifies actionable genomic alterations in lung adenocarcinomas otherwise negative for such alterations by other genomic testing approaches. Clin Cancer Res. 2015;21(16):3631-3639. https://clincancerres.aacrjournals.org/content/21/16/3631. 157. Gray PN, Dunlop CL, Elliott AM. Not All Next Generation Sequencing Diagnostics are Created Equal: Understanding the Nuances of Solid Tumor Assay Design for Somatic Mutation Detection. Cancers. 2015;7(3):1313-1332. https://www.mdpi.com/2072-6694/7/3/837. 158. Ali SM, Hensing T, Schrock AB, et al. Comprehensive genomic profiling identifies a subset of crizotinib-responsive alk-rearranged non-small cell lung cancer not detected by fluorescence in situ hybridization. The oncologist. 2016;21(6):762-770. https://theoncologist.onlinelibrary.wiley.com/doi/full/10.1634/theoncologist.2015-0497. 159. Chmielecki J, Ross JS, Wang K, et al. Oncogenic alterations in ERBB2/HER2 represent potential therapeutic targets across tumors from diverse anatomic sites of origin. The oncologist. 2015;20(1):7-12. https://theoncologist.onlinelibrary.wiley.com/doi/full/10.1634/theoncologist.2014-0234. 160. Ross JS, Gay LM, Wang K, et al. Nonamplification ERBB2 genomic alterations in 5605 cases of recurrent and metastatic breast cancer: An emerging opportunity for anti-HER2 targeted therapies. Cancer. 2016;122(17):2654-2662. https://acsjournals.onlinelibrary.wiley.com/doi/full/10.1002/cncr.30102. 161. Suh JH, Johnson A, Albacker L, et al. Comprehensive genomic profiling facilitates implementation of the National Comprehensive Cancer Network guidelines for lung cancer biomarker testing and identifies patients who may benefit from enrollment in mechanism-driven clinical trials. The oncologist. 2016;21(6):684-691. https://theoncologist.onlinelibrary.wiley.com/doi/full/10.1634/theoncologist.2016-0030.

CONFIDENTIAL PAGE 195 US-FDX-2000073 Last update: January 22, 2021

162. Frampton GM, Fichtenholtz A, Otto GA, et al. Development and validation of a clinical cancer genomic profiling test based on massively parallel DNA sequencing. Nature biotechnology. 2013;31(11):1023-1031. https://www.nature.com/articles/nbt.2696. 163. Wheler JJ, Janku F, Naing A, et al. TP53 alterations correlate with response to VEGF/VEGFR inhibitors: implications for targeted therapeutics. Molecular cancer therapeutics. 2016;15(10):2475-2485. https://mct.aacrjournals.org/content/15/10/2475. 164. US Food and Drug Administration. Hematology/oncology (cancer) approvals & safety notifications. 2018; July 23, 2019. Available at: https://www.fda.gov/drugs/resources- information-approved-drugs/hematologyoncology-cancer-approvals-safety-notifications. 165. Foundation Medicine Inc. FoundationOne Liquid CDx™ Technical Information. 2020; May 26, 2020. Available at: www.f1lcdxlabel.com. 166. Rizvi H, Sanchez-Vega F, La K, et al. Molecular determinants of response to anti-programmed cell death (PD)-1 and anti-programmed death-ligand 1 (PD-L1) blockade in patients with non- small-cell lung cancer profiled with targeted next-generation sequencing. J Clin Oncol. 2018;36(7):633-641. https://ascopubs.org/doi/10.1200/JCO.2017.75.3384. 167. Merino DM, McShane LM, Fabrizio D, et al. Establishing guidelines to harmonize tumor mutational burden (TMB): in silico assessment of variation in TMB quantification across diagnostic platforms: phase I of the Friends of Cancer Research TMB Harmonization Project. J Immunother Cancer. 2020;8(1). https://jitc.bmj.com/content/8/1/e000147. 168. Kocarnik JM, Shiovitz S, Phipps AI. Molecular phenotypes of colorectal cancer and potential clinical applications. Gastroenterology report. 2015;3(4):269-276. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4650976. 169. Abkevich V, Timms KM, Hennessy BT, et al. Patterns of genomic loss of heterozygosity predict homologous recombination repair defects in epithelial ovarian cancer. Br J Cancer. 2012;107(10):1776-1782. http://www.ncbi.nlm.nih.gov/pubmed/23047548. 170. Wang ZC, Birkbak NJ, Culhane AC, et al. Profiles of genomic instability in high-grade serous ovarian cancer predict treatment outcome. Clin Cancer Res. 2012;18(20):5806-5815. http://www.ncbi.nlm.nih.gov/pubmed/22912389. 171. Watkins JA, Irshad S, Grigoriadis A, Tutt AN. Genomic scars as biomarkers of homologous recombination deficiency and drug response in breast and ovarian cancers. Breast Cancer Res. 2014;16(3):211. http://www.ncbi.nlm.nih.gov/pubmed/25093514. 172. Vanderstichele A, Busschaert P, Olbrecht S, Lambrechts D, Vergote I. Genomic signatures as predictive biomarkers of homologous recombination deficiency in ovarian cancer. Eur J Cancer. 2017;86:5-14. http://www.ncbi.nlm.nih.gov/pubmed/28950147. 173. Cancer Genome Atlas Research N. Integrated genomic analyses of ovarian carcinoma. Nature. 2011;474(7353):609-615. http://www.ncbi.nlm.nih.gov/pubmed/21720365. 174. Venkitaraman AR. A growing network of cancer-susceptibility genes. The New England journal of medicine. 2003;348(19):1917-1919. http://www.ncbi.nlm.nih.gov/pubmed/12736286. 175. Centers for Medicare & Medicaid Services. Decision Memo for Next Generation Sequencing (NGS) for Medicare Beneficiaries with Advanced Cancer (CAG-00450N). 2018; March 16, 2018. Available at: https://www.cms.gov/medicare-coverage-database/details/nca-decision- memo.aspx?NCAId=290&bc=AAAAAAAAACAA&.

CONFIDENTIAL PAGE 196 US-FDX-2000073 Last update: January 22, 2021

176. Gadgeel SM, Mok TSK, Peters S, et al. LBA81_PR - Phase II/III blood first assay screening trial (BFAST) in patients (pts) with treatment-naïve NSCLC: Initial results from the ALK+ cohort. Ann. Oncol. 2019;30:v918. http://www.sciencedirect.com/science/article/pii/S0923753419604394. 177. Goodman AM, Castro A, Pyke RM, et al. MHC-I genotype and tumor mutational burden predict response to immunotherapy. Genome Med. 2020;12(1):45. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7236948/pdf/13073_2020_Article_743.pdf 178. Stransky N, Cerami E, Schalm S, Kim JL, Lengauer C. The landscape of kinase fusions in cancer. Nature communications. 2014;5:4846. https://www.nature.com/articles/ncomms5846. 179. Rosen EY, Goldman DA, Hechtman JF, et al. TRK Fusions Are Enriched in Cancers with Uncommon Histologies and the Absence of Canonical Driver Mutations. Clin Cancer Res. 2019. https://clincancerres.aacrjournals.org/content/26/7/1624. 180. Hyman DM vTC, Albert CM, et al. Durability of response with larotrectinib in adult and pediatric patients with TRK fusion cancer. Ann. Oncol. 2019;30(suppl_5):v159-v193. https://oncologypro.esmo.org/meeting-resources/esmo-2019-congress/Durability-of-response- with-larotrectinib-in-adult-and-pediatric-patients-with-TRK-fusion-cancer. 181. Kumar-Sinha C, Chinnaiyan AM. Precision oncology in the age of integrative genomics. Nature biotechnology. 2018;36(1):46-60. https://www.nature.com/articles/nbt.4017. 182. Vogelstein B, Papadopoulos N, Velculescu VE, Zhou S, Diaz LA, Jr., Kinzler KW. Cancer genome landscapes. Science (New York, N.Y.). 2013;339(6127):1546-1558. https://science.sciencemag.org/content/339/6127/1546. 183. Horak P, Fröhling S, Glimm H. Integrating next-generation sequencing into clinical oncology: strategies, promises and pitfalls. ESMO open. 2016;1(5):e000094. https://www.sciencedirect.com/science/article/pii/S2059702920325217?via%3Dihub. 184. Amatu A, Sartore-Bianchi A, Siena S. NTRK gene fusions as novel targets of cancer therapy across multiple tumour types. ESMO open. 2016;1(2):e000023. https://www.sciencedirect.com/science/article/pii/S2059702920325795?via%3Dihub. 185. Redig AJ, Jänne PA. Basket trials and the evolution of clinical trial design in an era of genomic medicine. J Clin Oncol. 2015;33(9):975-977. https://ascopubs.org/doi/10.1200/JCO.2014.59.8433. 186. Rahal Z, Abdulhai F, Kadara H, Saab R. Genomics of adult and pediatric solid tumors. American journal of cancer research. 2018;8(8):1356-1386. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6129500/. 187. Solomon JP, Linkov I, Rosado A, et al. NTRK fusion detection across multiple assays and 33,997 cases: diagnostic implications and pitfalls. Modern pathology : an official journal of the United States and Canadian Academy of Pathology, Inc. 2020;33(1):38-46. https://www.nature.com/articles/s41379-019-0324-7. 188. Hyman DM, Taylor BS, Baselga J. Implementing Genome-Driven Oncology. Cell. 2017;168(4):584-599. https://www.sciencedirect.com/science/article/pii/S0092867416317299?via%3Dihub. 189. Vitrakvi (larotrectinib) [prescribing information]. Bayer HealthCare Pharmaceuticals Inc. July 2019; April 23, 2020. Available at: http://labeling.bayerhealthcare.com/html/products/pi/vitrakvi_PI.pdf.

CONFIDENTIAL PAGE 197 US-FDX-2000073 Last update: January 22, 2021

190. Rozlytrek (entrectinib) [prescribing information]. Genentech Inc. August 2019; April 23, 2020. Available at: https://www.gene.com/download/pdf/rozlytrek_prescribing.pdf. 191. Looney AM, Nawaz K, Webster RM. Tumour-agnostic therapies. Nature reviews. Drug discovery. 2020;19(6):383-384. https://www.nature.com/articles/d41573-020-00015-1. 192. Garber K. Tissue-agnostic cancer drug pipeline grows, despite doubts. Nature reviews. Drug discovery. 2018;17(4):227-229. https://www.nature.com/articles/nrd.2018.6. 193. ClinicalTrials.gov. 2020; September 15, 2020. Available at: https://clinicaltrials.gov/. 194. Du X, Shao Y, Qin HF, Tai YH, Gao HJ. ALK-rearrangement in non-small-cell lung cancer (NSCLC). Thoracic cancer. 2018;9(4):423-430. https://onlinelibrary.wiley.com/doi/full/10.1111/1759-7714.12613. 195. Fink J. More Actionable Targets Improve Therapy in NSCLC. 2020; March 29, 2020. Available at: http://www.targetedonc.com/publications/targeted-therapy-news/2019/july2019/more- actionable-targets-improve-therapy-in-nsclc. 196. Morris TA, Khoo C, Solomon BJ. Targeting ROS1 Rearrangements in Non-small Cell Lung Cancer: Crizotinib and Newer Generation Tyrosine Kinase Inhibitors. Drugs. 2019;79(12):1277- 1286. https://link.springer.com/article/10.1007/s40265-019-01164-3. 197. Guo Y, Cao R, Zhang X, et al. Recent Progress in Rare Oncogenic Drivers and Targeted Therapy For Non-Small Cell Lung Cancer. OncoTargets and therapy. 2019;12:10343-10360. https://www.dovepress.com/recent-progress-in-rare-oncogenic-drivers-and-targeted-therapy-for- non-peer-reviewed-article-OTT. 198. Klempner SJ, Fabrizio D, Bane S, et al. Tumor Mutational Burden as a Predictive Biomarker for Response to Immune Checkpoint Inhibitors: A Review of Current Evidence. Oncologist. 2020;25(1):e147-e159. https://www.ncbi.nlm.nih.gov/pubmed/31578273 199. Goodman AM, Piccioni D, Kato S, et al. Prevalence of PDL1 Amplification and Preliminary Response to Immune Checkpoint Blockade in Solid Tumors. JAMA Oncol. 2018;4(9):1237-1244. https://jamanetwork.com/journals/jamaoncology/articlepdf/2684636/jamaoncology_goodman_20 18_oi_180034.pdf. 200. Yarchoan M, Hopkins A, Jaffee EM. Tumor Mutational Burden and Response Rate to PD-1 Inhibition. N Engl J Med. 2017;377(25):2500-2501. https://www.nejm.org/doi/pdf/10.1056/NEJMc1713444?articleTools=true 201. Fabrizio DA, George TJ, Jr., Dunne RF, et al. Beyond microsatellite testing: assessment of tumor mutational burden identifies subsets of colorectal cancer who may respond to immune checkpoint inhibition. J Gastrointest Oncol. 2018;9(4):610-617. https://www.ncbi.nlm.nih.gov/pubmed/30151257 202. Signorovitch J, Zhou Z, Ryan J, Chawla A. Comprehensive genomic profiling versus conventional molecular testing of patients with advanced non-small cell lung cancer: overall survival and cost in a US health plan population [abstract 6599]. J Clin Oncol. 2017;35(15_suppl):6599. https://ascopubs.org/doi/abs/10.1200/JCO.2017.35.15_suppl.6599. 203. Buchhalter I, Rempel E, Endris V, et al. Size matters: Dissecting key parameters for panel-based tumor mutational burden analysis. Int J Cancer. 2019;144(4):848-858. https://www.ncbi.nlm.nih.gov/pubmed/30238975

CONFIDENTIAL PAGE 198 US-FDX-2000073 Last update: January 22, 2021

204. Referenced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Neuroendocrine and Adrenal Tumors V.2.2020. © National Comprehensive Cancer Network, Inc; 2020. All rights reserved. Accessed December 28, 2020. To view the most recent and complete version of the guideline, go online to NCCN.org. 205. Referenced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Bone Cancer V.1.2021. © National Comprehensive Cancer Network, Inc; 2020. All rights reserved. Accessed December 28, 2020. To view the most recent and complete version of the guideline, go online to NCCN.org. 206. Referenced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Cervical Cancer V.1.2021. © National Comprehensive Cancer Network, Inc; 2020. All rights reserved. Accessed December 28, 2020. To view the most recent and complete version of the guideline, go online to NCCN.org. 207. Referenced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Thyroid Carcinoma V.2.2020. © National Comprehensive Cancer Network, Inc; 2020. All rights reserved. Accessed December 28, 2020. To view the most recent and complete version of the guideline, go online to NCCN.org. 208. Referenced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Occult Primary V.1.2021. © National Comprehensive Cancer Network, Inc; 2020. All rights reserved. Accessed December 28, 2020. To view the most recent and complete version of the guideline, go online to NCCN.org. 209. Lange AM, Lo HW. Inhibiting TRK Proteins in Clinical Cancer Therapy. Cancers. 2018;10(4). https://www.mdpi.com/2072-6694/10/4/105/htm. 210. Doebele RC, Drilon A, Paz-Ares L, et al. Entrectinib in patients with advanced or metastatic NTRK fusion-positive solid tumours: integrated analysis of three phase 1-2 trials. The Lancet. Oncology. 2020;21(2):271-282. https://www.sciencedirect.com/science/article/abs/pii/S1470204519306916?via%3Dihub. 211. Referenced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Gastrointestinal Stromal Tumors (GISTs) V.1.2021. © National Comprehensive Cancer Network, Inc; 2020. All rights reserved. Accessed December 28, 2020. To view the most recent and complete version of the guideline, go online to NCCN.org. 212. Referenced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Head and Neck Cancers V.1.2021. © National Comprehensive Cancer Network, Inc; 2020. All rights reserved. Accessed December 28, 2020. To view the most recent and complete version of the guideline, go online to NCCN.org. 213. US Food and Drug Administration. FDA grants accelerated approval to pembrolizumab for first tissue/site agnostic indication. 2017; March 30, 2018. Available at: https://www.fda.gov/Drugs/InformationOnDrugs/ApprovedDrugs/ucm560040.htm. 214. Le DT, Durham JN, Smith KN, et al. Mismatch repair deficiency predicts response of solid tumors to PD-1 blockade. Science (New York, N.Y.). 2017;357(6349):409-413. https://science.sciencemag.org/content/357/6349/409. 215. Referenced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Testicular Cancer V.1.2021. © National Comprehensive Cancer Network, Inc; 2020. All rights reserved. Accessed December 28, 2020. To view the most recent and complete version of the guideline, go online to NCCN.org.

CONFIDENTIAL PAGE 199 US-FDX-2000073 Last update: January 22, 2021

216. Referenced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Small Bowel Adenocarcinoma V.2.2020. © National Comprehensive Cancer Network, Inc; 2020. All rights reserved. Accessed December 28, 2020. To view the most recent and complete version of the guideline, go online to NCCN.org. 217. Stenzinger A, Endris V, Budczies J, et al. Harmonization and Standardization of Panel-Based Tumor Mutational Burden Measurement: Real-World Results and Recommendations of the Quality in Pathology Study. J Thorac Oncol. 2020. https://www.ncbi.nlm.nih.gov/pubmed/32119917 218. Mankor JM, Paats MS, Groenendijk FH, et al. Impact of panel design and cut-off on tumour mutational burden assessment in metastatic solid tumour samples. Br J Cancer. 2020;122(7):953- 956. https://www.ncbi.nlm.nih.gov/pubmed/32094484 219. Budczies J, Allgauer M, Litchfield K, et al. Optimizing panel-based tumor mutational burden (TMB) measurement. Ann Oncol. 2019;30(9):1496-1506. https://www.ncbi.nlm.nih.gov/pubmed/31268125 220. Merino DM, McShane LM, Fabrizio D, et al. Establishing guidelines to harmonize tumor mutational burden (TMB): in silico assessment of variation in TMB quantification across diagnostic platforms: phase I of the Friends of Cancer Research TMB Harmonization Project. J Immunother Cancer. 2020;8(1). https://www.ncbi.nlm.nih.gov/pubmed/32217756 221. Wu HX, Wang ZX, Zhao Q, Wang F, Xu RH. Designing gene panels for tumor mutational burden estimation: the need to shift from 'correlation' to 'accuracy'. J Immunother Cancer. 2019;7(1):206. https://www.ncbi.nlm.nih.gov/pubmed/31387639 222. Truesdell J, Miller VA, Fabrizio D. Approach to evaluating tumor mutational burden in routine clinical practice. Translational lung cancer research. 2018;7(6):678-681. https://tlcr.amegroups.com/article/view/25235/18843. 223. Fancello L, Gandini S, Pelicci PG, Mazzarella L. Tumor mutational burden quantification from targeted gene panels: major advancements and challenges. J Immunother Cancer. 2019;7(1):183. https://jitc.bmj.com/content/7/1/183. 224. Blackhall F, Kim DW, Besse B, et al. Patient-reported outcomes and quality of life in PROFILE 1007: a randomized trial of crizotinib compared with chemotherapy in previously treated patients with ALK-positive advanced non-small-cell lung cancer. Journal of thoracic oncology : official publication of the International Association for the Study of Lung Cancer. 2014;9(11):1625-1633. https://www.sciencedirect.com/science/article/pii/S1556086415307292?via%3Dihub. 225. Bevins N, Sun S, Gaieb Z, Thorson JA, Murray SS. Comparison of commonly used solid tumor targeted gene sequencing panels for estimating tumor mutation burden shows analytical and prognostic concordance within the cancer genome atlas cohort. J Immunother Cancer. 2020;8(1). https://www.ncbi.nlm.nih.gov/pubmed/32217764 226. Marabelle A, Fakih MG, Lopez J, et al. Association of tumour mutational burden with outcomes in patients with select advanced solid tumours treated with pembrolizumab in KEYNOTE-158. Annals of Oncology. 2019;30:v477-v478. https://doi.org/10.1093/annonc/mdz253.018. 227. Boland CR, Thibodeau SN, Hamilton SR, et al. A National Cancer Institute Workshop on Microsatellite Instability for cancer detection and familial predisposition: development of international criteria for the determination of microsatellite instability in colorectal cancer. Cancer research. 1998;58(22):5248-5257. https://pubmed.ncbi.nlm.nih.gov/9823339/

CONFIDENTIAL PAGE 200 US-FDX-2000073 Last update: January 22, 2021

228. Murphy KM, Zhang S, Geiger T, et al. Comparison of the microsatellite instability analysis system and the Bethesda panel for the determination of microsatellite instability in colorectal cancers. The Journal of molecular diagnostics : JMD. 2006;8(3):305-311. https://www.sciencedirect.com/science/article/pii/S1525157810603078?via%3Dihub. 229. The Surveillance, Epidemiology, and End Results (SEER) Program. SEER Cancer Stat Facts: Lung and Bronchus Cancer. National Cancer Institute. Bethesda, MD, . The National Cancer Institute. 2020. https://seer.cancer.gov/statfacts/html/lungb.html. 230. Zheng M. Classification and Pathology of Lung Cancer. Surgical oncology clinics of North America. 2016;25(3):447-468. https://www.ncbi.nlm.nih.gov/pubmed/?term=10.1016%2Fj.soc.2016.02.003. 231. Owonikoko TK, Ragin C, Chen Z, et al. Real-world effectiveness of systemic agents approved for advanced non-small cell lung cancer: a SEER-Medicare analysis. The oncologist. 2013;18(5):600-610. https://www.ncbi.nlm.nih.gov/pubmed/?term=The+Oncologist+2013%3B18%3A600%E2%80% 93610. 232. Kris MG, Johnson BE, Berry LD, et al. Using multiplexed assays of oncogenic drivers in lung cancers to select targeted drugs. Jama. 2014;311(19):1998-2006. https://jamanetwork.com/journals/jama/fullarticle/1872815. 233. Gainor JF, Shaw AT. Novel targets in non-small cell lung cancer: ROS1 and RET fusions. The oncologist. 2013;18(7):865-875. https://www.ncbi.nlm.nih.gov/pubmed/?term=Oncologist.+2013%3B18(7)%3A865-875. 234. Bubendorf L, Buttner R, Al-Dayel F, et al. Testing for ROS1 in non-small cell lung cancer: a review with recommendations. Virchows Archiv : an international journal of pathology. 2016;469(5):489-503. https://link.springer.com/article/10.1007%2Fs00428-016-2000-3. 235. Baik CS, Myall NJ, Wakelee HA. Targeting BRAF-Mutant Non-Small Cell Lung Cancer: From Molecular Profiling to Rationally Designed Therapy. The oncologist. 2017;22(7):786-796. https://www.ncbi.nlm.nih.gov/pubmed/?term=Oncologist.+2017%3B22(7)%3A786-796. 236. Chu QS. Targeting non-small cell lung cancer: driver mutation beyond epidermal growth factor mutation and anaplastic lymphoma kinase fusion. Therapeutic advances in medical oncology. 2020;12:1758835919895756. https://www.ncbi.nlm.nih.gov/pubmed/?term=10.1177%2F1758835919895756. 237. Golding B, Luu A, Jones R, Viloria-Petit AM. The function and therapeutic targeting of anaplastic lymphoma kinase (ALK) in non-small cell lung cancer (NSCLC). Molecular cancer. 2018;17(1):52. https://www.ncbi.nlm.nih.gov/pubmed/?term=Mol+Cancer.+2018%3B17(1)%3A52.doi%3A10.1 186%2Fs12943-018-0810-4. 238. Hofman P. ALK in Non-Small Cell Lung Cancer (NSCLC) Pathobiology, Epidemiology, Detection from Tumor Tissue and Algorithm Diagnosis in a Daily Practice. Cancers. 2017;9(8). https://www.ncbi.nlm.nih.gov/pubmed/?term=10.3390%2Fcancers9080107. 239. Aran V, Omerovic J. Current Approaches in NSCLC Targeting K-RAS and EGFR. International journal of molecular sciences. 2019;20(22). https://www.ncbi.nlm.nih.gov/pubmed/?term=10.3390%2Fijms20225701.

CONFIDENTIAL PAGE 201 US-FDX-2000073 Last update: January 22, 2021

240. Mok TSK, Wu YL, Kudaba I, et al. Pembrolizumab versus chemotherapy for previously untreated, PD-L1-expressing, locally advanced or metastatic non-small-cell lung cancer (KEYNOTE-042): a randomised, open-label, controlled, phase 3 trial. Lancet. 2019;393(10183):1819-1830. https://www.ncbi.nlm.nih.gov/pubmed/?term=Lancet.+2019%3B393(10183)%3A1819-1830. 241. Hsu PC, Jablons DM, Yang CT, You L. Epidermal Growth Factor Receptor (EGFR) Pathway, Yes-Associated Protein (YAP) and the Regulation of Programmed Death-Ligand 1 (PD-L1) in Non-Small Cell Lung Cancer (NSCLC). International journal of molecular sciences. 2019;20(15). https://www.ncbi.nlm.nih.gov/pubmed/?term=10.3390%2Fijms20153821. 242. Lindeman NI, Cagle PT, Aisner DL, et al. Updated molecular testing guideline for the selection of lung cancer patients for treatment with targeted tyrosine kinase inhibitors: Guideline from the College of American Pathologists, the International Association for the Study of Lung Cancer, and the Association for Molecular Pathology. Archives of pathology & laboratory medicine. 2018;142(3):321-346. https://meridian.allenpress.com/aplm/article/142/3/321/103064/Updated- Molecular-Testing-Guideline-for-the. 243. Kalemkerian GP, Narula N, Kennedy EB. Molecular Testing Guideline for the Selection of Lung Cancer Patients for Treatment With Targeted Tyrosine Kinase Inhibitors: American Society of Clinical Oncology Endorsement Summary of the College of American Pathologists/International Association for the Study of Lung Cancer/Association for Molecular Pathology Clinical Practice Guideline Update. Journal of oncology practice. 2018;14(5):323-327. https://ascopubs.org/doi/10.1200/JOP.18.00035. 244. Rolfo C, Mack PC, Scagliotti GV, et al. Liquid biopsy for advanced non-small cell lung cancer (NSCLC): A statement paper from the IASLC. Journal of thoracic oncology : official publication of the International Association for the Study of Lung Cancer. 2018;13(9):1248-1268. https://www.sciencedirect.com/science/article/pii/S1556086418306804?via%3Dihub. 245. Wolf J, Han J, Nishio M, et al. GEOMETRY mono-1: Phase II, multicenter study of MET inhibitor Capmatinib (INC280) in EGFR wt, metdysregulated advanced NSCLC. Journal of Thoracic Oncology. 2017;12(11):S1578-S1579. http://www.embase.com/search/results?subaction=viewrecord&from=export&id=L619317598. 246. Wolf J, Seto T, Han JY, et al. Results of the GEOMETRY mono-1 phase II study for evaluation of the MET inhibitor capmatinib (INC280) in patients (PTS) with METDex14 mutated advanced non-small cell lung cancer (NSCLC). Annals of Oncology. 2018;29:viii741-viii742. http://www.embase.com/search/results?subaction=viewrecord&from=export&id=L628560288. 247. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2020. CA Cancer J Clin. 2020;70(1):7-30. https://acsjournals.onlinelibrary.wiley.com/doi/full/10.3322/caac.21590. 248. Yuan Y, Yost SE, Yim J, et al. Genomic mutation-driven metastatic breast cancer therapy: a single center experience. Oncotarget. 2017;8(16):26414-26423. https://www.oncotarget.com/article/14476/text/. 249. Marhold M, Bartsch R, Zielinski C. Recent developments and translational aspects in targeted therapy for metastatic breast cancer. ESMO open. 2016;1(3):e000036. https://www.sciencedirect.com/science/article/pii/S2059702920325606?via%3Dihub. 250. Breat Cancer Metastasis and Drug Resistance: Challenges and Progress. Second ed. Advances in Experimental Medicine and Biology. 2019.

CONFIDENTIAL PAGE 202 US-FDX-2000073 Last update: January 22, 2021

251. Litton JK, Burstein HJ, Turner NC. Molecular Testing in Breast Cancer. American Society of Clinical Oncology educational book. American Society of Clinical Oncology. Annual Meeting. 2019;39:e1-e7. https://ascopubs.org/doi/10.1200/EDBK_237715. 252. Robson M, Im SA, Senkus E, et al. Olaparib for Metastatic Breast Cancer in Patients with a Germline BRCA Mutation. The New England journal of medicine. 2017;377(6):523-533. https://www.nejm.org/doi/10.1056/NEJMoa1706450. 253. Litton JK, Rugo HS, Ettl J, et al. Talazoparib in Patients with Advanced Breast Cancer and a Germline BRCA Mutation. The New England journal of medicine. 2018;379(8):753-763. https://www.nejm.org/doi/10.1056/NEJMoa1802905. 254. Schmid P, Adams S, Rugo HS, et al. Atezolizumab and Nab-Paclitaxel in Advanced Triple- Negative Breast Cancer. The New England journal of medicine. 2018;379(22):2108-2121. https://www.nejm.org/doi/10.1056/NEJMoa1809615. 255. Kurian AW, Ward KC, Howlader N, et al. Genetic Testing and Results in a Population-Based Cohort of Breast Cancer Patients and Ovarian Cancer Patients. J Clin Oncol. 2019;37(15):1305- 1315. https://ascopubs.org/doi/10.1200/JCO.18.01854. 256. Surveillance, Epidemiology, and End Results Program. SEER Cancer Stat Facts: Colorectal Cancer. 2020. Available at: https://seer.cancer.gov/statfacts/html/colorect.html 257. Tournigand C, André T, Achille E, et al. FOLFIRI followed by FOLFOX6 or the reverse sequence in advanced colorectal cancer: a randomized GERCOR study. J Clin Oncol. 2004;22(2):229-237. https://ascopubs.org/doi/10.1200/JCO.2004.05.113 258. Tejpar S, Stintzing S, Ciardiello F, et al. Prognostic and Predictive Relevance of Primary Tumor Location in Patients With RAS Wild-Type Metastatic Colorectal Cancer: Retrospective Analyses of the CRYSTAL and FIRE-3 Trials. JAMA oncology. 2017;3(2):194-201. https://jamanetwork.com/journals/jamaoncology/fullarticle/2565703. 259. Weinberg BA, Marshall JL, Hartley M, Salem ME. A paradigm shift from one-size-fits-all to tailor-made therapy for metastatic colorectal cancer. Clinical advances in hematology & oncology : H&O. 2016;14(2):116-128. https://pubmed.ncbi.nlm.nih.gov/27057810/. 260. Lieu CH, Corcoran RB, Overman MJ. Integrating Biomarkers and Targeted Therapy Into Colorectal Cancer Management. American Society of Clinical Oncology educational book. American Society of Clinical Oncology. Annual Meeting. 2019;39:207-215. https://ascopubs.org/doi/10.1200/EDBK_240839. 261. Douillard JY, Oliner KS, Siena S, et al. Panitumumab-FOLFOX4 treatment and RAS mutations in colorectal cancer. The New England journal of medicine. 2013;369(11):1023-1034. https://www.nejm.org/doi/10.1056/NEJMoa1305275. 262. Van Cutsem E, Lenz HJ, Köhne CH, et al. Fluorouracil, leucovorin, and irinotecan plus cetuximab treatment and RAS mutations in colorectal cancer. J Clin Oncol. 2015;33(7):692-700. https://ascopubs.org/doi/10.1200/JCO.2014.59.4812. 263. Sorich MJ, Wiese MD, Rowland A, Kichenadasse G, McKinnon RA, Karapetis CS. Extended RAS mutations and anti-EGFR monoclonal antibody survival benefit in metastatic colorectal cancer: a meta-analysis of randomized, controlled trials. Annals of oncology : official journal of the European Society for Medical Oncology. 2015;26(1):13-21. https://www.sciencedirect.com/science/article/pii/S0923753419313158?via%3Dihub.

CONFIDENTIAL PAGE 203 US-FDX-2000073 Last update: January 22, 2021

264. Kopetz S, Grothey A, Yaeger R, et al. Encorafenib, Binimetinib, and Cetuximab in BRAF V600E-Mutated Colorectal Cancer. The New England journal of medicine. 2019;381(17):1632- 1643. https://www.nejm.org/doi/10.1056/NEJMoa1908075. 265. Overman MJ, Vauthey J, Aloia TA, et al. Circulating tumor DNA (ctDNA) utilizing a high- sensitivity panel to detect minimal residual disease post liver hepatectomy and predict disease recurrence. J Clin Oncol. 2017;35(15 Suppl):3522. http://ascopubs.org/doi/abs/10.1200/JCO.2017.35.15_suppl.3522. 266. Le DT, Kim TW, Van Cutsem E, et al. Phase II Open-Label Study of Pembrolizumab in Treatment-Refractory, Microsatellite Instability-High/Mismatch Repair-Deficient Metastatic Colorectal Cancer: KEYNOTE-164. J Clin Oncol. 2020;38(1):11-19. https://ascopubs.org/doi/10.1200/JCO.19.02107. 267. Sartore-Bianchi A, Trusolino L, Martino C, et al. Dual-targeted therapy with trastuzumab and lapatinib in treatment-refractory, KRAS codon 12/13 wild-type, HER2-positive metastatic colorectal cancer (HERACLES): a proof-of-concept, multicentre, open-label, phase 2 trial. The Lancet. Oncology. 2016;17(6):738-746. https://www.sciencedirect.com/science/article/abs/pii/S1470204516001509?via%3Dihub. 268. Meric-Bernstam F, Hurwitz H, Raghav KPS, et al. Pertuzumab plus trastuzumab for HER2- amplified metastatic colorectal cancer (MyPathway): an updated report from a multicentre, open- label, phase 2a, multiple basket study. The Lancet. Oncology. 2019;20(4):518-530. https://www.sciencedirect.com/science/article/abs/pii/S1470204518309045?via%3Dihub. 269. National Cancer Institute. SEER*Explorer: An interactive website for SEER cancer statistics 2019; January 14, 2019. Available at: https://seer.cancer.gov/explorer. 270. Tsibulak I, Zeimet AG, Marth C. Hopes and failures in front-line ovarian cancer therapy. Critical reviews in oncology/hematology. 2019;143:14-19. https://www.sciencedirect.com/science/article/pii/S1040842819301660?via%3Dihub. 271. Arend R, Westin SN, Coleman RL. Decision analysis for secondline maintenance treatment of platinum sensitive recurrent ovarian cancer: a review. International journal of gynecological cancer : official journal of the International Gynecological Cancer Society. 2020;30(5):684-694. https://ijgc.bmj.com/content/30/5/684. 272. Pignata S, Pisano C, Di Napoli M, Cecere SC, Tambaro R, Attademo L. Treatment of recurrent epithelial ovarian cancer. Cancer. 2019;125 Suppl 24:4609-4615. https://acsjournals.onlinelibrary.wiley.com/doi/full/10.1002/cncr.32500. 273. Giornelli GH. Management of relapsed ovarian cancer: a review. SpringerPlus. 2016;5(1):1197. https://springerplus.springeropen.com/articles/10.1186/s40064-016-2660-0. 274. Pujade-Lauraine E, Hilpert F, Weber B, et al. Bevacizumab combined with chemotherapy for platinum-resistant recurrent ovarian cancer: The AURELIA open-label randomized phase III trial. J Clin Oncol. 2014;32(13):1302-1308. https://ascopubs.org/doi/10.1200/JCO.2013.51.4489. 275. Lheureux S, Braunstein M, Oza AM. Epithelial ovarian cancer: Evolution of management in the era of precision medicine. CA Cancer J Clin. 2019;69(4):280-304. https://acsjournals.onlinelibrary.wiley.com/doi/full/10.3322/caac.21559. 276. Pujade-Lauraine E, Ledermann JA, Selle F, et al. Olaparib tablets as maintenance therapy in patients with platinum-sensitive, relapsed ovarian cancer and a BRCA1/2 mutation (SOLO2/ENGOT-Ov21): a double-blind, randomised, placebo-controlled, phase 3 trial. The

CONFIDENTIAL PAGE 204 US-FDX-2000073 Last update: January 22, 2021

Lancet. Oncology. 2017;18(9):1274-1284. https://www.sciencedirect.com/science/article/abs/pii/S1470204517304692?via%3Dihub. 277. Domchek SM, Aghajanian C, Shapira-Frommer R, et al. Efficacy and safety of olaparib monotherapy in germline BRCA1/2 mutation carriers with advanced ovarian cancer and three or more lines of prior therapy. Gynecologic oncology. 2016;140(2):199-203. https://www.sciencedirect.com/science/article/abs/pii/S0090825815302237?via%3Dihub. 278. Mirza MR, Monk BJ, Herrstedt J, et al. Niraparib Maintenance Therapy in Platinum-Sensitive, Recurrent Ovarian Cancer. The New England journal of medicine. 2016;375(22):2154-2164. https://www.nejm.org/doi/10.1056/NEJMoa1611310. 279. Moore KN, Secord AA, Geller MA, et al. Niraparib monotherapy for late-line treatment of ovarian cancer (QUADRA): a multicentre, open-label, single-arm, phase 2 trial. The Lancet. Oncology. 2019;20(5):636-648. https://www.sciencedirect.com/science/article/abs/pii/S1470204519300294?via%3Dihub. 280. Coleman RL, Oza AM, Lorusso D, et al. Rucaparib maintenance treatment for recurrent ovarian carcinoma after response to platinum therapy (ARIEL3): a randomised, double-blind, placebo- controlled, phase 3 trial. Lancet. 2017;390(10106):1949-1961. https://www.sciencedirect.com/science/article/abs/pii/S0140673617324406?via%3Dihub. 281. Marabelle A, Le DT, Ascierto PA, et al. Efficacy of Pembrolizumab in Patients With Noncolorectal High Microsatellite Instability/Mismatch Repair-Deficient Cancer: Results From the Phase II KEYNOTE-158 Study. J Clin Oncol. 2020;38(1):1-10. https://ascopubs.org/doi/10.1200/JCO.19.02105. 282. Norquist BM, Harrell MI, Brady MF, et al. Inherited Mutations in Women With Ovarian Carcinoma. JAMA oncology. 2016;2(4):482-490. https://jamanetwork.com/journals/jamaoncology/fullarticle/2479125. 283. Lancaster JM, Powell CB, Chen LM, Richardson DL. Society of Gynecologic Oncology statement on risk assessment for inherited gynecologic cancer predispositions. Gynecologic oncology. 2015;136(1):3-7. https://www.sciencedirect.com/science/article/abs/pii/S0090825814013122?via%3Dihub. 284. Hoppe MM, Sundar R, Tan DSP, Jeyasekharan AD. Biomarkers for Homologous Recombination Deficiency in Cancer. Journal of the National Cancer Institute. 2018;110(7):704-713. https://academic.oup.com/jnci/article/110/7/704/4999673. 285. Childers CP, Childers KK, Maggard-Gibbons M, Macinko J. National Estimates of Genetic Testing in Women With a History of Breast or Ovarian Cancer. J Clin Oncol. 2017;35(34):3800- 3806. https://ascopubs.org/doi/10.1200/JCO.2017.73.6314. 286. Hoskins PJ, Gotlieb WH. Missed therapeutic and prevention opportunities in women with BRCA-mutated epithelial ovarian cancer and their families due to low referral rates for genetic counseling and BRCA testing: A review of the literature. CA Cancer J Clin. 2017;67(6):493-506. https://acsjournals.onlinelibrary.wiley.com/doi/full/10.3322/caac.21408. 287. Konstantinopoulos PA, Norquist B, Lacchetti C, et al. Germline and Somatic Tumor Testing in Epithelial Ovarian Cancer: ASCO Guideline. J Clin Oncol. 2020:Jco1902960. https://ascopubs.org/doi/10.1200/JCO.19.02960.

CONFIDENTIAL PAGE 205 US-FDX-2000073 Last update: January 22, 2021

288. Pawlyn C, Loehr A, Ashby C, et al. Loss of heterozygosity as a marker of homologous repair deficiency in multiple myeloma: a role for PARP inhibition? Leukemia. 2018;32(7):1561-1566. http://www.ncbi.nlm.nih.gov/pubmed/29467487. 289. Coleman RL, Oza AM, Lorusso D, et al. Rucaparib maintenance treatment for recurrent ovarian carcinoma after response to platinum therapy (ARIEL3): a randomised, double-blind, placebo- controlled, phase 3 trial. Lancet. 2017;390(10106):1949-1961. http://www.ncbi.nlm.nih.gov/pubmed/28916367. 290. Coleman RL, Swisher EM, Oza AM, et al. Refinement of prespecified cutoff for genomic loss of heterozygosity (LOH) in ARIEL2 part 1: A phase II study of rucaparib in patients (pts) with high grade ovarian carcinoma (HGOC). Journal of Clinical Oncology. 2016;34(15_suppl):5540-5540. https://ascopubs.org/doi/abs/10.1200/JCO.2016.34.15_suppl.5540. 291. Kirby M, Hirst C, Crawford ED. Characterising the castration-resistant prostate cancer population: a systematic review. International journal of clinical practice. 2011;65(11):1180- 1192. https://www.ncbi.nlm.nih.gov/pubmed/?term=Int+J+Clin+Pract.+2011%3B65(11)%3A1180- 1192. 292. Abida W, Armenia J, Gopalan A, et al. Prospective Genomic Profiling of Prostate Cancer Across Disease States Reveals Germline and Somatic Alterations That May Affect Clinical Decision Making. JCO Precis Oncol. 2017;2017. https://pubmed.ncbi.nlm.nih.gov/28825054/. 293. Sandhu SK, Hussain M, Mateo J, et al. LBA8 - PROfound: Phase III study of olaparib versus enzalutamide or abiraterone for metastatic castration-resistant prostate cancer (mCRPC) with homologous recombination repair (HRR) gene alterations. Annals of Oncology. 2019;30:ix188- ix189. http://www.sciencedirect.com/science/article/pii/S0923753419582126. 294. Higano C, Sternberg CN, Saad F, et al. Treatment patterns and outcomes for metastatic castration-resistant prostate cancer (mCRPC) in a real-world setting: a retrospective study of more than 2500 patients. Presented at the American Society of Clinical Oncology-Genitourinary Cancers Symposium; San Francisco, CA; February 14-16. 2019. https://ascopubs.org/doi/abs/10.1200/JCO.2019.37.7_suppl.256. 295. Saad F, Shore N, Zhang T, Sharma S, Cho HK, Jacobs IA. Emerging therapeutic targets for patients with advanced prostate cancer. Cancer treatment reviews. 2019;76:1-9. https://www.sciencedirect.com/science/article/abs/pii/S0305737219300507?via%3Dihub. 296. Abida W, Cheng ML, Armenia J, et al. Analysis of the Prevalence of Microsatellite Instability in Prostate Cancer and Response to Immune Checkpoint Blockade. JAMA oncology. 2019;5(4):471- 478. https://www.ncbi.nlm.nih.gov/pubmed/?term=JAMA+Oncology+5%2C+471%E2%80%93478+( 2019). 297. Wassim Abida, David Campbell, Akash Patnaik, Brieuc Sautois, Jeremy Shapiro, al. E. Preliminary Results from the TRITON2 Study of Rucaparib in Patients with DNA Damage Repair (DDR)-Deficient Metastatic Castration-Resistant Prostate Cancer (mCRPC): Updated Analyses. ESMO Congress | 27 September–1 October | Barcelona, Spain. 2019. https://www.clovisoncology.com/media/1160/esmo2019_wabida_poster.pdf. 298. Paller CJ, Antonarakis ES, Beer TM, et al. Germline Genetic Testing in Advanced Prostate Cancer; Practices and Barriers: Survey Results from the Germline Genetics Working Group of the Prostate Cancer Clinical Trials Consortium. Clinical genitourinary cancer. 2019;17(4):275-

CONFIDENTIAL PAGE 206 US-FDX-2000073 Last update: January 22, 2021

282.e271. https://www.sciencedirect.com/science/article/abs/pii/S1558767319301399?via%3Dihub. 299. Surveillance, Epidemiology, and End Results Program. SEER Cancer Stat Facts: Melanoma of the Skin. . 2020, 2020. Available at: https://seer.cancer.gov/statfacts/html/melan.html. 300. Ugurel S, Röhmel J, Ascierto PA, et al. Survival of patients with advanced metastatic melanoma: the impact of novel therapies-update 2017. Eur J Cancer. 2017;83:247-257. https://www.sciencedirect.com/science/article/abs/pii/S0959804917310791?via%3Dihub. 301. Testori AAE, Chiellino S, van Akkooi ACJ. Adjuvant Therapy for Melanoma: Past, Current, and Future Developments. Cancers. 2020;12(7). https://www.mdpi.com/2072-6694/12/7/1994. 302. Mason R, Au L, Ingles Garces A, Larkin J. Current and emerging systemic therapies for cutaneous metastatic melanoma. Expert opinion on pharmacotherapy. 2019;20(9):1135-1152. https://www.tandfonline.com/doi/abs/10.1080/14656566.2019.1601700?journalCode=ieop20. 303. Eggermont AMM, Dummer R. The 2017 complete overhaul of adjuvant therapies for high-risk melanoma and its consequences for staging and management of melanoma patients. Eur J Cancer. 2017;86:101-105. https://www.sciencedirect.com/science/article/abs/pii/S0959804917312959?via%3Dihub. 304. Zhao DY, Lim KH. Current biologics for treatment of biliary tract cancers. Journal of gastrointestinal oncology. 2017;8(3):430-440. https://jgo.amegroups.com/article/view/13912/11520. 305. Malenica I, Donadon M, Lleo A. Molecular and Immunological Characterization of Biliary Tract Cancers: A Paradigm Shift Towards a Personalized Medicine. Cancers. 2020;12(8). https://www.mdpi.com/2072-6694/12/8/2190. 306. Palmieri LJ, Lavolé J, Dermine S, et al. The choice for the optimal therapy in advanced biliary tract cancers: Chemotherapy, targeted therapies or immunotherapy. Pharmacology & therapeutics. 2020;210:107517. https://www.sciencedirect.com/science/article/abs/pii/S0163725820300450?via%3Dihub. 307. Valle J, Wasan H, Palmer DH, et al. Cisplatin plus gemcitabine versus gemcitabine for biliary tract cancer. The New England journal of medicine. 2010;362(14):1273-1281. https://www.nejm.org/doi/full/10.1056/NEJMoa0908721. 308. Abou-Alfa GK, Macarulla T, Javle MM, et al. Ivosidenib in IDH1-mutant, chemotherapy- refractory cholangiocarcinoma (ClarIDHy): a multicentre, randomised, double-blind, placebo- controlled, phase 3 study. The Lancet. Oncology. 2020;21(6):796-807. https://www.sciencedirect.com/science/article/abs/pii/S1470204520301571?via%3Dihub. 309. Lamarca A, Palmer DH, Wasan HS, et al. ABC-06 | A randomised phase III, multi-centre, open- label study of active symptom control (ASC) alone or ASC with oxaliplatin / 5-FU chemotherapy (ASC+mFOLFOX) for patients (pts) with locally advanced / metastatic biliary tract cancers (ABC) previously-treated with cisplatin/gemcitabine (CisGem) chemotherapy. Journal of Clinical Oncology. 2019;37(15_suppl):4003-4003. https://ascopubs.org/doi/abs/10.1200/JCO.2019.37.15_suppl.4003. 310. Nakamura H, Arai Y, Totoki Y, et al. Genomic spectra of biliary tract cancer. Nature genetics. 2015;47(9):1003-1010. https://www.nature.com/articles/ng.3375.

CONFIDENTIAL PAGE 207 US-FDX-2000073 Last update: January 22, 2021

311. Signorovitch J, Wheler J, Miller VA, Ryan J, Zhou Z, Chawla A. Estimated cost of anti-cancer therapy directed by comprehensive genomic profiling in a single-center study [abstract 6605]. J Clin Oncol. 2017;35(15_suppl):6605. https://ascopubs.org/doi/full/10.1200/PO.18.00074. 312. Chawla A, Peeples M, Li N, Anhorn R, Ryan J, Signorovitch J. Real-world utilization of molecular diagnostic testing and matched drug therapies in the treatment of metastatic cancers. J Med Econ. 2018:1-10. https://www.ncbi.nlm.nih.gov/pubmed/29295635. 313. Bruinooge SS, Dueck AC, Gray SW, et al. Use, attitudes, and perceptions of tumor genomic testing: Survey of TAPUR physicians. Journal of Clinical Oncology. 2019;37(15_suppl):6531- 6531. https://ascopubs.org/doi/abs/10.1200/JCO.2019.37.15_suppl.6531. 314. Signorovitch J, Zhou Z, Ryan J, Anhorn R, Chawla A. Budget impact analysis of comprehensive genomic profiling in patients with advanced non-small cell lung cancer. J Med Econ. 2019;22(2):140-150. http://www.ncbi.nlm.nih.gov/pubmed/30430885. 315. Drilon A, Siena S, Ou SI, et al. Safety and Antitumor Activity of the Multitargeted Pan-TRK, ROS1, and ALK Inhibitor Entrectinib: Combined Results from Two Phase I Trials (ALKA-372- 001 and STARTRK-1). Cancer Discov. 2017;7(4):400-409. https://www.ncbi.nlm.nih.gov/pubmed/28183697. 316. Hempelmann JA, Lockwood CM, Konnick EQ, et al. Microsatellite instability in prostate cancer by PCR or next-generation sequencing. J Immunother Cancer. 2018;6(1):29. http://www.ncbi.nlm.nih.gov/pubmed/29665853. 317. Alunbrig (brigatinib) [prescribing information]. Millennium Pharmaceuticals, Inc. May 2020; August 11, 2020. Available at: https://www.alunbrig.com/assets/pi.pdf. 318. Braftovi (encorafenib) [prescribing information]. Array BioPharma Inc. April 2020; November 19, 2020. Available at: http://labeling.pfizer.com/ShowLabeling.aspx?id=12990. 319. Cabometyx (cabozantinib) [prescribing information]. Exelixis, Inc. July 2020; August 11, 2020. Available at: https://www.cabometyxhcp.com/pi/. 320. Caprelsa (vandetanib) [prescribing information]. Genzyme Corporation. June 2020; June 24, 2020. Available at: http://www.caprelsa.com/files/caprelsa-pi.pdf. 321. Enhertu (fam-trastuzumab deruxtecan-nxki) [prescribing information]. Daiichi Sankyo, Inc. December 2019; August 11, 2020. Available at: https://dsi.com/prescribing-information- portlet/getPIContent?productName=Enhertu&inline=true. 322. Gleevec (imatinib) [prescribing information]. Novartis Pharmaceuticals Corporation. August 2020; September 15, 2020. Available at: https://www.novartis.us/sites/www.novartis.us/files/gleevec_tabs.pdf. 323. Kadcyla (ado-trastuzumab emtansine) [prescribing information]. May 2019; Augsut 11, 2020. Available at: https://www.gene.com/download/pdf/kadcyla_prescribing.pdf. 324. Lorbrena (lorlatinib) [prescribing information]. Pfizer Laboratories Div Pfizer Inc. May 2020; August 11, 2020. Available at: http://labeling.pfizer.com/ShowLabeling.aspx?id=11140. 325. Lynparza (olaparib) [prescribing information]. AstraZeneca Pharmaceuticals LP. May 2020; August 11, 2020. Available at: https://www.azpicentral.com/lynparza_tb/lynparza_tb.pdf#page=1. 326. Mektovi (binimetinib) [prescribing information]. Array BioPharma Inc. August 2020; September 15, 2020. Available at: http://labeling.pfizer.com/ShowLabeling.aspx?id=12988.

CONFIDENTIAL PAGE 208 US-FDX-2000073 Last update: January 22, 2021

327. Nerlynx (neratinib) [prescribing information]. Puma Biotechnology, Inc. July 2020; August 11, 2020. Available at: https://nerlynx.com/pdf/full-prescribing-information.pdf. 328. Opdivo (nivolumab) [prescribing information]. Bristol-Myers Squibb Company. June 2020; September 15, 2020. Available at: https://packageinserts.bms.com/pi/pi_opdivo.pdf. 329. Pemazyre (pemigatinib) [prescribing information]. Incyte Corporation. April 2020; September 15, 2020. Available at: https://www.pemazyre.com/pdf/prescribing-information.pdf. 330. Retevmo (selpercatinib) [prescribing information]. Eli Lilly and Company. May 2020; August 12, 2020. Available at: https://uspl.lilly.com/retevmo/retevmo.html#pi. 331. Rubraca (rucaparib) [prescribing information]. Clovis Oncology, Inc. May 2020; August 11, 2020. Available at: https://clovisoncology.com/pdfs/RubracaUSPI.pdf. 332. Tabrecta [capmatinib] (prescribing information]. Novartis Pharmaceuticals Corporation. May 2020; June 24, 2020. Available at: https://www.novartis.us/sites/www.novartis.us/files/tabrecta.pdf. 333. Talzenna (talazoparib) [prescribing information]. Pfizer Laboratories Div Pfizer Inc. March 2020; August 11, 2020. Available at: http://labeling.pfizer.com/ShowLabeling.aspx?id=11046. 334. Tecentriq (atezolizumab) [prescribing information]. Genentech, Inc. July 2020; August 11, 2020. Available at: https://www.gene.com/download/pdf/tecentriq_prescribing.pdf. 335. Tibsovo (ivosidenib) [prescribing information]. Incyte Corporation. May 2019; September 15, 2020. Available at: https://www.tibsovopro.com/pdf/prescribinginformation.pdf. 336. Tukysa (tucatinib) [prescribing information]. Seattle Genetics, Inc. April 2020; August 11, 2020. Available at: https://seagendocs.com/TUKYSA_Full_Ltr_Master.pdf. 337. Tykerb (lapatinib) [prescribing information]. Novartis Pharmaceuticals Corporation. December 2018; August 11, 2020. Available at: https://www.novartis.us/sites/www.novartis.us/files/tykerb.pdf. 338. Vizimpro (dacomitinib) [prescribing information]. Pfizer Laboratories Div Pfizer Inc. September 2018; August 11, 2020. Available at: http://labeling.pfizer.com/ShowLabeling.aspx?id=11019. 339. Yervoy (ipilimumab) [prescribing information]. Bristol-Myers Squibb Company. August 2020; September 15, 2020. Available at: https://packageinserts.bms.com/pi/pi_yervoy.pdf. 340. Zejula (niraparib) [prescribing information]. GlaxoSmithKline. April 2020; August 11, 2020. Available at: https://www.gsksource.com/pharma/content/dam/GlaxoSmithKline/US/en/Prescribing_Informati on/Zejula/pdf/ZEJULA-PI-PIL.PDF. 341. Ross JS, Ali SM, Fasan O, et al. ALK Fusions in a Wide Variety of Tumor Types Respond to Anti-ALK Targeted Therapy. The oncologist. 2017;22(12):1444-1450. http://www.ncbi.nlm.nih.gov/pubmed/29079636. 342. Ross JS, Wang K, Chmielecki J, et al. The distribution of BRAF gene fusions in solid tumors and response to targeted therapy. International journal of cancer. 2016;138(4):881-890. https://onlinelibrary.wiley.com/doi/10.1002/ijc.29825. 343. Klempner SJ, Bordoni R, Gowen K, et al. Identification of BRAF kinase domain duplications across multiple tumor types and response to RAF inhibitor therapy. JAMA oncology. 2016;2(2):272-274. https://jamanetwork.com/journals/jamaoncology/fullarticle/2469337.

CONFIDENTIAL PAGE 209 US-FDX-2000073 Last update: January 22, 2021

344. Frampton GM, Ali SM, Rosenzweig M, et al. Activation of MET via diverse exon 14 splicing alterations occurs in multiple tumor types and confers clinical sensitivity to MET inhibitors. Cancer Discov. 2015;5(8):850-859. https://cancerdiscovery.aacrjournals.org/content/5/8/850. 345. Gay L, Fabrizio D, Frampton G, et al. Mutational burden of tumors with primary site unknown [abstract 3039]. J Clin Oncol. 2017;35(15_suppl):3039. https://ascopubs.org/doi/abs/10.1200/JCO.2017.35.15_suppl.3039. 346. Ross JS, Wang K, Gay L, et al. Comprehensive genomic profiling of carcinoma of unknown primary site: New routes to targeted therapies. JAMA oncology. 2015;1(1):40-49. https://jamanetwork.com/journals/jamaoncology/fullarticle/2108853. 347. Chung JH, Ali SM, Davis J, et al. A poorly differentiated malignant neoplasm lacking lung markers harbors an EML4-ALK rearrangement and responds to crizotinib. Case Rep Oncol. 2014;7(3):628-632. https://www.karger.com/Article/FullText/367780. 348. Palma NA, Ali SM, O'Connor J, et al. Durable response to crizotinib in a MET-amplified, KRAS- mutated carcinoma of unknown primary. Case reports in oncology. 2014;7(2):503-508. https://www.karger.com/Article/FullText/365326. 349. Singhi AD, George B, Greenbowe JR, et al. Real-Time Targeted Genome Profile Analysis of Pancreatic Ductal Adenocarcinomas Identifies Genetic Alterations That Might Be Targeted With Existing Drugs or Used as Biomarkers. Gastroenterology. 2019;156(8):2242-2253 e2244. http://www.ncbi.nlm.nih.gov/pubmed/30836094. 350. Singhi AD, Ali SM, Lacy J, et al. Identification of targetable ALK rearrangements in pancreatic ductal adenocarcinoma. J Natl Compr Canc Netw. 2017;15(5):555-562. http://www.ncbi.nlm.nih.gov/pubmed/30836094. 351. Chmielecki J, Hutchinson KE, Frampton GM, et al. Comprehensive genomic profiling of pancreatic acinar cell carcinomas identifies recurrent RAF fusions and frequent inactivation of DNA repair genes. Cancer Discov. 2014;4(12):1398-1405. https://cancerdiscovery.aacrjournals.org/content/4/12/1398. 352. Chung JH, Dewal N, Sokol E, et al. Prospective Comprehensive Genomic Profiling of Primary and Metastatic Prostate Tumors. JCO Precis Oncol. 2019;3. http://www.ncbi.nlm.nih.gov/pubmed/31218271. 353. Ikeda S, Elkin SK, Tomson BN, Carter JL, Kurzrock R. Next-generation sequencing of prostate cancer: genomic and pathway alterations, potential actionability patterns, and relative rate of use of clinical-grade testing. Cancer Biol Ther. 2018:1-8. http://www.ncbi.nlm.nih.gov/pubmed/30339521. 354. Skoulidis F, Goldberg ME, Greenawalt DM, et al. STK11/LKB1 Mutations and PD-1 Inhibitor Resistance in KRAS-Mutant Lung Adenocarcinoma. Cancer Discov. 2018;8(7):822-835. http://www.ncbi.nlm.nih.gov/pubmed/29773717. 355. Papadimitrakopoulou V, Lee JJ, Wistuba, II, et al. The BATTLE-2 study: A biomarker-integrated targeted therapy study in previously treated patients with advanced non-small-cell lung cancer. J Clin Oncol. 2016;34(30):3638-3647. https://ascopubs.org/doi/10.1200/JCO.2015.66.0084. 356. Wheler JJ, Atkins JT, Janku F, et al. Presence of both alterations in FGFR/FGF and PI3K/AKT/mTOR confer improved outcomes for patients with metastatic breast cancer treated with PI3K/AKT/mTOR inhibitors. Oncoscience. 2016;3(5-6):164-172. https://www.oncoscience.us/article/307/text/.

CONFIDENTIAL PAGE 210 US-FDX-2000073 Last update: January 22, 2021

357. Balko JM, Giltnane JM, Wang K, et al. Molecular profiling of the residual disease of triple- negative breast cancers after neoadjuvant chemotherapy identifies actionable therapeutic targets. Cancer Discov. 2014;4(2):232-245. https://cancerdiscovery.aacrjournals.org/content/4/2/23. 358. Schrock AB, Ouyang C, Sandhu J, et al. Tumor mutational burden is predictive of response to immune checkpoint inhibitors in MSI-high metastatic colorectal cancer. Annals of oncology : official journal of the European Society for Medical Oncology. 2019. http://www.ncbi.nlm.nih.gov/pubmed/31038663. 359. Jones JC, Renfro LA, Al-Shamsi HO, et al. (Non-V600) BRAF Mutations Define a Clinically Distinct Molecular Subtype of Metastatic Colorectal Cancer. J Clin Oncol. 2017;35(23):2624- 2630. https://ascopubs.org/doi/10.1200/JCO.2016.71.4394. 360. Boussemart L, Johnson A, Schrock AB, et al. Tumor mutational burden and response to programmed cell death protein 1 inhibitors in a case series of patients with metastatic desmoplastic melanoma. J Am Acad Dermatol. 2019;80(6):1780-1782. http://www.ncbi.nlm.nih.gov/pubmed/30576761. 361. Wheler J, Yelensky R, Falchook G, et al. Next generation sequencing of exceptional responders with BRAF-mutant melanoma: implications for sensitivity and resistance. BMC cancer. 2015;15:61. https://bmccancer.biomedcentral.com/articles/10.1186/s12885-015-1029-z. 362. Stockley TL, Oza AM, Berman HK, et al. Molecular profiling of advanced solid tumors and patient outcomes with genotype-matched clinical trials: the Princess Margaret IMPACT/COMPACT trial. Genome medicine. 2016;8(1):109. https://genomemedicine.biomedcentral.com/articles/10.1186/s13073-016-0364-2. 363. Kim ST, Lee J, Hong M, et al. The NEXT-1 (Next generation pErsonalized tX with mulTi-omics and preclinical model) trial: prospective molecular screening trial of metastatic solid cancer patients, a feasibility analysis. Oncotarget. 2015;6(32):33358-33368. https://www.oncotarget.com/article/5188/text/. 364. Spetzler D, Xiao N, Burnett K, et al. Multi-platform molecular profiling of 1,180 patients increases median overall survival and influences treatment decision in 53% of cases [abstract]. Eur. J. Cancer. 2015;51:S44. http://dx.doi.org/10.1016/S0959-8049(16)30140-X. 365. Tsimberidou AM, Iskander NG, Hong DS, et al. Personalized medicine in a phase I clinical trials program: the MD Anderson Cancer Center initiative. Clin Cancer Res. 2012;18(22):6373-6383. https://clincancerres.aacrjournals.org/content/18/22/6373.

CONFIDENTIAL PAGE 211 US-FDX-2000073 © 2021 Foundation Medicine, Inc. | FoundationOne® is a registered trademark of Foundation Medicine, Inc. | www.foundationmedicine.com | Tel. 888.988.3639 | Fax 617.418.2290 | US-FDX-2000073