Overcoming MET-Dependent Resistance to Selective RET Inhibition in Patients with RET Fusion–Positive Lung Cancer by Combining Selpercatinib with Crizotinib a C Ezra Y

Home , Emibetuzumab, RET inhibitor, Selpercatinib

Published OnlineFirst October 20, 2020; DOI: 10.1158/1078-0432.CCR-20-2278

CLINICAL CANCER RESEARCH | RESEARCH BRIEFS: CLINICAL TRIAL BRIEF REPORT

Overcoming MET-Dependent Resistance to Selective RET Inhibition in Patients with RET Fusion–Positive Lung Cancer by Combining Selpercatinib with Crizotinib A C Ezra Y. Rosen1, Melissa L. Johnson2, Sarah E. Clifford3, Romel Somwar4, Jennifer F. Kherani5, Jieun Son3, Arrien A. Bertram3, Monika A. Davare6, Eric Gladstone4, Elena V. Ivanova7, Dahlia N. Henry5, Elaine M. Kelley3, Mika Lin3, Marina S.D. Milan3, Binoj C. Nair5, Elizabeth A. Olek5, Jenna E. Scanlon3, Morana Vojnic4, Kevin Ebata5, Jaclyn F. Hechtman4, Bob T. Li1,8, Lynette M. Sholl9, Barry S. Taylor10, Marc Ladanyi4, Pasi A. Janne€ 3, S. Michael Rothenberg5, Alexander Drilon1,8, and Geoffrey R. Oxnard3

ABSTRACT ◥ Purpose: The RET proto-oncogene encodes a receptor tyrosine Results: MET amplification was identified in posttreatment kinase that is activated by gene fusion in 1%–2% of non–small biopsies in 4 patients with RET fusion–positive NSCLC treated with cell lung cancers (NSCLC) and rarely in other cancer types. selpercatinib. In at least one case, MET amplification was clearly Selpercatinib is a highly selective RET kinase inhibitor that has evident prior to therapy with selpercatinib. We demonstrate that recently been approved by the FDA in lung and thyroid cancers increased MET expression in RET fusion–positive tumor cells causes with activating RET gene fusions and mutations. Molecular resistance to selpercatinib, and this can be overcome by combining mechanisms of acquired resistance to selpercatinib are poorly selpercatinib with crizotinib. Using SPPs, selpercatinib with crizo- understood. tinib were given together generating anecdotal evidence of clinical Patients and Methods: We studied patients treated on the first- activity and tolerability, with one response lasting 10 months. in-human clinical trial of selpercatinib (NCT03157129) who were Conclusions: Through the use of SPPs, we were able to offer found to have MET amplification associated with resistance to combination therapy targeting MET-amplified resistance iden- selpercatinib. We validated MET activation as a targetable mediator tified on the first-in-human study of selpercatinib. These data of resistance to RET-directed therapy, and combined selpercatinib suggest that MET dependence is a recurring and potentially with the MET/ALK/ROS1 inhibitor crizotinib in a series of single targetable mechanism of resistance to selective RET inhibition patient protocols (SPP). in advanced NSCLC.

Introduction patient selection. While several combination therapies are approved (e.g., MEK inhibition with BRAF inhibition in BRAF-mutant mela- Combination targeted therapy represents a compelling strategy for þ noma; CDK4/6 inhibition with endocrine therapy in ER breast overcoming drug resistance in metastatic cancer. However, the clinical cancer; refs. 1–4), no drug combination has yet met the standard of development of combination approaches has been challenging due to regulatory approval for effective treatment of resistance to targeted toxicity from combining two agents and the need for appropriate kinase inhibitors (TKI) in genotype-selected patients. Because resistance to targeted TKIs is universal, effective strategies to overcome acquired resistance are key to prolonging clinical benefit. To that end, 1Department of Medicine, Memorial Sloan Kettering Cancer Center, New York, New York. 2Department of Medicine, Sarah Cannon Cancer Center, Nashville, combination approaches remain a compelling investigational strategy – Tennessee. 3Lowe Center for Thoracic Oncology, Dana-Farber Cancer Institute, in oncogene-dependent non small cell lung cancer (NSCLC) with Boston, Massachusetts. 4Department of Pathology, Memorial Sloan Kettering several clinical trials ongoing (5–8). Cancer Center, New York, New York. 5Loxo Oncology, Inc., a wholly owned The RET proto-oncogene encodes a receptor tyrosine kinase which 6 subsidiary of Eli Lilly and Company, Stamford, Connecticut. Oregon Health and is activated by gene fusion in 1%–2% of NSCLC and rarely in many 7 Science University, Portland, Oregon. Belfer Center for Applied Cancer Science, other tumor types. RET gene fusions are bona fide cancer drivers and Dana-Farber Cancer Institute, Boston, Massachusetts. 8Weill Cornell Medical College, New York, New York. 9Department of Pathology, Brigham and they display the key characteristics of oncogene addiction preclini- Women’s Hospital, Boston, Massachusetts. 10Department of Epidemiology and cally (9). Selpercatinib is a highly selective and potent anti-RET TKIs Biostatistics, Memorial Sloan Kettering Cancer Center, New York, New York. which has recently reported durable responses in lung and thyroid fi Note: Supplementary data for this article are available at Clinical Cancer cancers, and these responses were maintained regardless of the speci c Research Online (http://clincancerres.aacrjournals.org/). RET alteration or prior TKI use, and also in the setting of the RET V804 “gatekeeper” mutation (10). Selpercatinib was recently approved by E.Y. Rosen and M.L. Johnson contributed equally to this article. A. Drilon and G.R. Oxnard contributed equally to this article. the FDA for use in these cancers. Mechanisms of acquired resistance to treatment with selective RET inhibitors are not well understood. While Corresponding Author: Alexander Drilon, Memorial Sloan Kettering Cancer a secondary mutation in the RET kinase domain has recently been Center, 530 East 74th Street, New York, NY 10021. Phone: 646-888-4206; fi Fax: 646-888–4263; E-mail: [email protected] reported (11), activation of bypass tracts, such as MET ampli cation, also represent a recurring mechanism of resistance to driver genotypes Clin Cancer Res 2020;XX:XX–XX in NSCLC (12, 13). Here we piloted combination therapy to target doi: 10.1158/1078-0432.CCR-20-2278 MET amplification detected in 4 patients with RET-positive NSCLC 2020 American Association for Cancer Research. (of a total 79 patients with NSCLC enrolled at all three sites) with

AACRJournals.org | OF1

Rosen et al.

Preclinical RET-dependent models Translational Relevance RET fusion–positive human bronchioepithelial (HBEC-RET) Molecular mechanisms of acquired resistance to selpercatinib, a cell lines expressing a CCDC6-RET fusion were engineered to highly selective and potent RET kinase inhibitor, are poorly overexpress MET and a patient-derived organoid was also estab- understood. We identified MET amplification as a recurrent lished. These models were subsequently studied to investigate mechanism of resistance to targeted therapy in patients with the role of MET in selpercatinib resistance (see Supplementary non–small cell lung cancer (NSCLC) treated with selpercatinib. Methods). We show that MET amplification is sufficient to cause selpercatinib resistance in vitro, and that the addition of the MET/ALK/ROS1 SPPs of selpercatinib and crizotinib inhibitor crizotinib can rescue this phenotype. We then utilize a Each SPP was sponsored by LOXO and drafted in collaboration series of single-patient protocols to treat these patients with between LOXO and the site primary investigator. Each protocol combination therapy, and this combination treatment showed enrolled a single patient after review by the FDA and approval by clinical activity, with one response lasting 10 months. These data the site IRB. Dosing was individualized per patient and dose escalation suggest that MET dependence is a recurring and potentially was permitted as tolerated up to the established tolerable dose for each targetable mechanism of resistance to selective RET inhibition in drug. Patients initiated combination therapy directly after demon- advanced NSCLC. Prospective clinical trials are needed to validate strating resistance to prior targeted therapy (Table 1; Supplementary these findings and to identify effective combination therapies to Fig. S1). treat acquired resistance to selpercatinib. Results MET-dependent resistance to selpercatinib in patients with resistance to selpercatinib. This was made possible through the use of advanced NSCLC multiple single-patient protocols (SPP) which allowed for the quick Case 1 delivery of potentially effective combination therapy to patients with The first patient was a 36-year-old female former smoker with clear unmet clinical need. stage IV NSCLC metastatic to bone, heavily pretreated. Molecular testing identified a RET rearrangement by break-apart FISH (83% of cells) as well as MET copy-number gain (CNG) by FISH (6 copies). Patients and Methods She initiated treatment with alectinib, an ALK TKI with some Analysis of resistance to selpercatinib degree of anti-RET activity preclinically (14), and progressed in Patients were included in this analysis if they had received less than 2 months. Next-generation sequencing (NGS) analysis of selpercatinib (LOXO-292) for RET-positive NSCLC on the first- plasma circulating cell-free tumor DNA (cfDNA) then identified an in-human study of selpercatinib (NCT03157128) and exhibited EML4-RET fusion (AF 14%; ref. 15). She started treatment with evidence of MET amplification following drug resistance. All selpercatinib and experienced a clinical response (decreased tumor- patients provided written informed consent wherever necessary. related pain and anorexia) with radiographic tumor reduction Genomic analysis of tumor and plasma cfDNA was performed on imaging (21% decrease in tumor burden after 16 weeks, independently at each participating site. All specimens were studied Fig. 1A; Table 1). She progressed after 4.5 months with growth at each participating institution with independent review board of liver metastases and a new skin nodule on the neck. Biopsy of the (IRB) approval and were analyzed in accordance with the Decla- skin nodule revealed metastatic adenocarcinoma and molecular ration of Helsinki. analysis by NGS reidentified the EML4-RET fusion as well as

Table 1. Disease course and treatment history summary.

Case 1 Case 2 Case 3 Case 4

Clinical presentation 36 yo former smoker (18 48 yo former light smoker 69 yo never-smoker, 61 yo never-smoker, s/p py), multiple prior lines of (<1 py) s/p first-line second-line first-line pembrolizmab therapy, recent alectinib pembrolizumab Pretreatment genotype EML4-RET fusion on plasma KIF5B-RET fusion on tumor KIF5B-RET fusion on tumor KIF5B-RET fusion on tumor NGS, prior MET CNG NGS NGS, also high MET NGS, also MET CNG amplification Treatment duration on 6.5 months (40 mg BID ! 11 months (20 mg BID ! 3 months (80 mg BID) 6 months (160 mg BID ! selpercatinib 160 mg BID) 80 mg BID) 120 mg BID) monotherapy and dosing Treatment duration on 3.5 months (80 mg BID ! 10 months (160 mg BID; 4 months (80 mg BID ! 4 months (160 mg BID; selpercatinib and 160 mg BID; 250 mg QD 250 mg BID) 160 mg BID; 250 mg BID) 250 mg BID) crizotinib and dosing ! BID) Outcome of combination Mixed response Durable response Brief response Brief response therapy Adverse events Nausea Lower extremity edema, Myocardial infarction (not Severe colitis (not reflux attributed to study drugs) attributed to study drugs)

Abbreviations: BID, twice a day; py, pack years; yo, year old.

OF2 Clin Cancer Res; 2020 CLINICAL CANCER RESEARCH

MET-Dependent Resistance to Selective RET Inhibitor in NSCLC

Figure 1. A MET amplification identified in RET fusion–positive lung cancers treated with a selective RET inhibitor. A, Patient 1 had a symptomatic response to selpercatinib with radiographic tumor reduction on imaging (21% decrease in tumor burden) after 16 weeks, but eventually developed resistance to drug. B, Tumor NGS at 3 months of Resistance to Baseline time of resistance showed, in addition selpercatinib selpercatinib to the original EML4-RET fusion, high amplification of MET (56 copies). C and B D, In patient 4, NGS showed MET amplification in the posttreatment sample (T2), with lower level gain below threshold for amplification in the pretreatment biopsy (T1). E, The presence of MET overexpression (right) was confirmed with MET IHC both pretreatment (top) and at time of resistance (bottom). C

D 2 E

1.5

0.5 CNA Fold change 0 T1 T2 Brain

Chest wall

increased MET copy number (56 copies; Fig. 1B). Given ongoing Case 3 clinical benefit, she continued on selpercatinib postprogression for a The third patient was a 69-year-old Asian male nonsmoker diag- total treatment time of 6.5 months. nosed with stage IV NSCLC, and after developing a solitary brain metastasis he underwent craniotomy. Tumor NGS (Foundation One) Case 2 identified a KIF5B-RET fusion and MET amplification (18 copies) in The second patient was a 48-year-old male former light brain. He initiated treatment with selpercatinib which resulted in smoker with stage IV PD-L1–positive NSCLC that harbored a shrinkage of pulmonary nodules and improvement in clinical symp- KIF5B-RET fusion identified on tumor NGS. After progression on toms, but he also developed a new left adrenal nodule, indicating first-line pembrolizumab, he initiated selpercatinib and achieved a disease progression. RNA sequencing was performed on the previously partial response after 3 months of therapy (best response of 49% resected brain metastasis (Illumina TST-170) which confirmed high- decrease in target lesions; Table 1). He eventually developed disease level MET mRNA expression. He then initiated crizotinib monother- progression after 11 months on selpercatinib. NGS analysis of apy given his MET gene amplification and overexpression, and he paired pretreatment and postprogression tumor samples identified again demonstrated a mixed response, with decrease in size of the left acquired MET amplification (9 copies) completely absent from the adrenal nodule but growth of a right adrenal nodule and left hilar pretreatment sample. adenopathy; he remained on crizotinib for 3 months.

AACRJournals.org Clin Cancer Res; 2020 OF3

Rosen et al.

Case 4 (250 mg everyday). After tolerating these doses for 4 weeks, the patient The fourth patient was a 61-year-old woman, never smoker, with was escalated sequentially until reaching RP2D/approved doses of stage IV lung adenocarcinoma metastatic to brain, PD-L1 positive. 160 mg twice a day/250 mg twice a day of selpercatinib and crizotinib, After progression on first-line pembrolizumab, tumor NGS identified respectively. Treatment was tolerated with only mild nausea. Real-time a KIF5B-RET fusion and selpercatinib was initiated. The patient had a pharmacokinetic analysis indicated selpercatinib exposure remained best response of stable disease (1% increase in the sum of tumor consistent with the patient’s exposure during selpercatinib monother- diameters at 6 weeks) but subsequently progressed after 6 months and apy, while crizotinib exposure remained consistent with published discontinued treatment. NGS analysis of posttreatment tumor showed exposures when used as monotherapy (Supplementary Fig. S2A). She MET amplification (fold change 2.1), while the pretreatment biopsy experienced clinical improvement in bone pain after 1 month of showed low-level MET CNG without frank amplification (Fig. 1C combination therapy; however, scans after 2.5 months revealed a and D). By FISH, the posttreatment biopsy showed MET amplification mixed response with improvement in liver metastases but progressive (66% of cells) while the pretreatment biopsy noted MET polysomy (3– pulmonary disease. Because of ongoing improvement in bone pain, 6 copies) in 78% of cells. The presence of MET overexpression was she continued on study therapy for a total of 3.5 months before dying of subsequently confirmed by IHC at both time points (Fig. 1E). her cancer. Case 2 (patient with partial response to selpercatinib lasting MET overexpression causes acquired resistance to selpercatinib 11 months and acquired MET amplification) initiated treatment with preclinically and may be overcome by combining selpercatinib the combination of selpercatinib and crizotinib, and pharmacokinetic with the MET inhibitor crizotinib analyses revealed the expected levels of each drug when used as To determine the potential effect of MET overexpression on sen- monotherapy (Supplementary Fig. S2B). The patient experienced a sitivity of RET fusion–positive tumor cells to selpercatinib, we over- clinical and radiographic tumor response to combination treatment, expressed MET in HBEC-RET. HBEC-RET cells were designed to with resolution of shortness of breath and maximal tumor diameter express a CCDC6-RET fusion cDNA and are sensitive to RET inhi- reduction of 38%. He responded for 10 months before discontinuing bitors (Fig. 2A and B). HBEC-RETþMET cells were far less sensitive treatment for progression in the lungs and increase in ascites (Fig. 3A). ¼ m to selpercatinib (Fig. 2C and D;IC50 10.92 mol/L) compared with He tolerated treatment well, with adverse events (AE) of lower ¼ m fl the isogenic control cells (IC50 0.09 mol/L), with a more than a 100- extremity edema, possibly related to crizotinib, and re ux. NGS of RET fold shift in the IC50 values for growth inhibition in the presence of a resistance biopsy showed persistence of the fusion but loss of the MET overexpression, suggesting that overexpression of MET drives MET gene amplification (Fig. 3B). Notably, the only other alteration resistance to selective RET inhibition. detected was the ATM splice variant [c.8988–1G>C (splice)]. In addition, we derived an organoid culture from tumor cells Case 3 (patient with mixed response on selpercatinib followed by isolated from pleural fluid of the patient described in Case 2. Analysis mixed response on crizotinib, with pretreatment MET amplification) of the organoid by MET FISH confirmed high-level MET amplifica- continued treatment with crizotinib while restarting treatment with tion, and IHC confirmed high-level MET protein expression, consis- selpercatinib at 80 mg twice a day which was subsequently dose tent with the post-selpercatinib NGS analyses (Fig. 2E). In vitro escalated. He experienced a partial response by RECIST 1.1 (maximum treatment with selpercatinib or crizotinib monotherapy was ineffec- tumor reduction 42% below baseline) after 1.5 months of combi- tive, but combined treatment with selpercatinib and crizotinib showed nation therapy; he died unexpectedly of an unrelated cardiac event a cytotoxic effect (Fig. 2F and G). As expected, only combined after 4 months. Combination treatment was otherwise well tolerated treatment resulted in decreased phospho-RET and phospho-MET without AEs. levels concomitantly (Fig. 2H). Selpercatinib alone blocked RET Case 4 (patient with best response of stable disease on selpercatinib, activity whereas the activity of AKT and ERK were retained possibly with pretreatment MET CNG and posttreatment MET amplification) demonstrating a mechanism of resistance in this patient. Interestingly, initiated treatment at the full doses of selpercatinib at 160 mg twice a crizotinib alone inhibited MET and AKT signaling but not pERK day and crizotinib at 250 mg twice a day. A brisk partial response (Fig. 2H). Finally, the combination treatment successfully led to (40%) was achieved at 4 weeks (Fig. 3C) with disease regression in a inactivation of both ERK and AKT, suggesting a potential mechanism left lung mass and a left chest wall mass. Although the patient tolerated for the utility of this drug combination in this RET fusion/MET combination treatment well without drug-related AEs, she developed amplification patient. These data demonstrate that MET amplification colitis (determined by the investigator to be unrelated to treatment) causes resistance to selpercatinib in patients with RET fusion–positive that ultimately required treatment interruption and surgery, and she NSCLC, which can be overcome preclinically by combined treatment elected to transition to hospice care. with selpercatinib and crizotinib.

Combination treatment with selpercatinib and crizotinib Discussion overcomes MET-dependent resistance in patients We describe MET amplification as a targetable mechanism of We were motivated by the high selectivity and clean safety profile of resistance to selpercatinib in RET-rearranged NSCLC. As greater selpercatinib and by the known feasibility of adding crizotinib to other number of patients develop resistance to selpercatinib, it will be targeted TKIs in other biomarker-selected subsets of patients with important to systematically quantify the prevalence of MET amplifi- NSCLC. Therefore, we treated the above 4 patients with the combi- cation and other potentially targetable resistance mechanisms, such as nation of selpercatinib and crizotinib, each using an FDA-allowed, the secondary RET mutation that was recently described (11). We do IRB-approved SPP. note that there are 79 patients with NSCLC enrolled at our three Case 1 (patient with minor response to selpercatinib, MET CNG centers, which does offer the reader a rough estimate of the rarity of this before treatment and high MET amplification after treatment) started type of resistance. While the level of MET gene amplification clearly treatment at one-half the recommended phase II (RP2D) of selperca- increased during selpercatinib monotherapy, in 3 of 4 cases, some tinib (80 mg twice a day) and one-half the approved dose of crizotinib degree of MET gain was already present prior to exposure to

OF4 Clin Cancer Res; 2020 CLINICAL CANCER RESEARCH

MET-Dependent Resistance to Selective RET Inhibitor in NSCLC

Figure 2. A HBEC-RET B MET amplification drives resistance to sel- 10 bp percatinib and responds to MET inhibition in LC-2ad EV MET 400 8 RET – RET fusion positive models. A and B, 300 CCDC/RET fusion confirmed by RT-PCR using primers 200 6 targeting CCDC6 (exon 1, forward) and RET 400 (exon 12, reverse), and MET expression was 300 4 200 GAPDH fi con rmed by qPCR. C, Cells were treated expression 2 with the indicated concentrations of selper-

catinib for 96 hours, and then the relative Relative MET mRNA 0 number of cells was determined using pro- EV MET liferation dye. D, Viability data were ana- C 125 lyzed, and estimated IC50 values with the D

95% conﬁdence interval are shown. HBEC, IC50 for growth Cell line human bronchiolar epithelial cells. EV, emp- 100 inhibition (µmol/L) ty vector. E, Patient-derived organoid from KIF5B-RET fusion–positive NSCLC (Case 2) 75 HBEC-RET-EV 0.09 (0.04–0.17) shows MET gain by both IHC and FISH. F and G, Cell viability of dissociated cells from 50 HBEC- 10.92 (7.7–15.8) cultured organoids treated with either sel- RET+MET percatinib (0.3 mmol/L) or crizotinib (1 25 HBEC-RET+EV

mmol/L) has little effect, but the combina- Relative cell growth (%) tion is cytotoxic. H, Selpercatinib alone HBEC-RET+MET F P < 0.05 0 blocked RET activity whereas pAKT and 150 0.001 0.01 0.1 1 10 pERK were retained, while combination µ P < 0.001 treatment successfully led to inactivation of Selpercatinib ( mol/L) both AKT and ERK. 100 E 50

0 Relative luminescence (%) Criz Veh

25µm Selper 25µm Combo

H&E MET IHC MET FISH H p-MET G p-RET

p-AKT

p-ERK1/2

t-MET Vehicle Selpercatinib t-RET

t-AKT

t-ERK1/2

HSP90 Crizotinib Combo 50µm Veh Criz Selper Combo

selpercatinib. This is reminiscent of EGFR-mutant NSCLC, in which showed an ORR of 44% to osimertinib (EFGR-TKI) and savolitinib rare clones with high-level MET ampliﬁcation may be detected at (MET-TKI; ref. 8). baseline, prior to EGFR inhibitor therapy (16, 17). It is notable that a While the median progression-free survival has been reported recent early-phase EGFR-mutant/MET-ampliﬁed NSCLC trial at 18 months for selpercatinib in RET-positive NSCLC (18), our

AACRJournals.org Clin Cancer Res; 2020 OF5

Rosen et al.

A Baseline B KIF5B-RET, MET wild type Figure 3. Response to selective dual RET inhibition and RET inhibition. A, In patient 2, combination treatment yielded a clinical and radiological response, until he eventually developed disease progression. B, NGS showing acquired ampliﬁcation of MET at time of resistance to selpercatinib, then at time of Selpercatinib resistance the loss of the MET ampli- KIF5B-RET, MET amplified ﬁcation but with continued presence 4 Months Resistance of RET fusion. C, Patient 4 pretreatment (top) and on-treatment (bottom) imaging showing a partial response at 4 weeks to selpercatinib and crizotinib.

Selpercatinib + Crizotinib KIF5B-RET, MET wild type 3 Months Resistance

patients in contrast had an unusually short benefit from selperca- difficulty to assess in individual SPPs—while these patients did not tinib. The cause of this modest PFS benefitisunknown,butthis complain of intolerable toxicity while under treatment, 1 patient died maybeduetosomedegreeofMET amplification present at baseline of an apparently unrelated cardiac event, and a second patient in these patients or may be related to the aggressive nature of experienced severe colitis. Both of these AEs were thought to be the MET oncogene (19). In addition, these brief responses may unrelated to this drug combination, but the potential toxicities of be due to the presence of additional driver mutations, either such drug combinations will need to be studied in future prospective through heterogeneity of resistance to selpercatinib at distinct cohorts of patients with appropriate performance status and comor- metastatic sites, or by means of additional subclonal drivers not bidities. Finally, we are hopeful that the use of newer, more specific detected on NGS. MET inhibitors including capmatinib (refs. 7, 20; which is FDA To better understand the clinical effect of this combination, pro- approved) and tepotinib (21) in combination with selpercatinib may spective efforts will be needed to study combination therapy with result in increased efficacy and better tolerability of this drug selpercatinib plus MET inhibition. In addition, treatment tolerability is combination.

OF6 Clin Cancer Res; 2020 CLINICAL CANCER RESEARCH

MET-Dependent Resistance to Selective RET Inhibitor in NSCLC

In these SPPs, the expeditious delivery of a potentially effective for conduct of clinical trial on which M.L. Johnson served as PI), Hengrui combination therapy to patients with high unmet clinical need was Therapeutics (payment to institution for conduct of clinical trial on which enabled by the availability of an approved second agent, the willingness M.L. Johnson served as PI), Immunocore (payment to institution for conduct of clinical trial on which M.L. Johnson served as PI), Jounce Therapeutics of the sponsor to permit early use of combination therapy with an (payment to institution for conduct of clinical trial on which M.L. Johnson investigational therapy still being studied in a first-in-human trial, and served as PI), Kadmon Pharmaceuticals (payment to institution for conduct of the rapid review and allowance of each SPP by the FDA and by local clinical trial on which M.L. Johnson served as PI), Lycera (payment to IRBs. Our experience provides further evidence for the importance of institution for conduct of clinical trial on which M.L. Johnson served as PI), robust, multi-cancer gene panel–based molecular analysis in patients Neovia Oncology (payment to institution for conduct of clinical trial on which with resistance to targeted therapies to enable the identification of M.L. Johnson served as PI), OncoMed Pharmaceuticals (payment to institution for conduct of clinical trial on which M.L. Johnson served as PI), Regeneron potentially targetable acquired resistance mechanisms in a time frame Pharmaceuticals (payment to institution for conduct of clinical trial on that can help each patient. These cases provide evidence for the which M.L. Johnson served as PI), Seven and Eight Biopharmaceuticals feasibility of this approach, and this may enable other potentially (payment to institution for conduct of clinical trial on which M.L. Johnson effective combination therapies with a clear biologic rationale to be served as PI), Shattuck Labs (payment to institution for conduct of clinical trial on offered immediately to individual patients without alternative treat- which M.L. Johnson served as PI), Stem CentRx (payment to institution for conduct of ment options, while also providing clinical proof of concept that may clinical trial on which M.L. Johnson served as PI), Syndax Pharmaceuticals (payment to institution for conduct of clinical trial on which M.L. Johnson served as PI), Takeda be validated in subsequent, prospective clinical trials. Pharmaceuticals (payment to institution for conduct of clinical trial on which M.L. Johnson served as PI), Tarveda (payment to institution for conduct of clinical trial on Authors’ Disclosures which M.L. Johnson served as PI), University of Michigan (payment to institution for conduct of clinical trial on which M.L. Johnson served as PI), TCR2 Therapeutics E.Y. Rosen reports grants from Bayer (research funding) outside the submitted (payment to institution for conduct of clinical trial on which M.L. Johnson served work. M.L. Johnson reports grants and other from Loxo/Lilly (payment to institution as PI), Arcus Biosciences (payment to institution for conduct of clinical trial on which for conduct of clinical trial on which M.L. Johnson served as PI); other (payment to M.L. Johnson served as PI), Amgen (payment to institution for conduct of clinical trial institution for consulting services performed by M.L. Johnson) during the conduct of on which M.L. Johnson served as PI), TMUNITY Therapeutics (payment to the study; grants and other from AbbVie (payment to institution for conduct of institution for conduct of clinical trial on which M.L. Johnson served as PI); other clinical trial on which M.L. Johnson served as PI), AstraZeneca (payment to from Achilles Therapeutics, Association of Community Cancer Centers (payment to institution for conduct of clinical trial on which M.L. Johnson served as PI), institution for consulting services performed by M.L. Johnson); personal fees from Atreca (payment to institution for conduct of clinical trial on which M.L. Johnson Astellas (consulting/advisory role for an immediate family member) and Otsuka served as PI), Boehringer Ingelheim (payment to institution for conduct of clinical (consulting/advisory role for an immediate family member) outside the submitted trial on which M.L. Johnson served as PI), Calithera Biosciences (payment to work. S.E. Clifford reports other from Foundation Medicine outside the submitted institution for conduct of clinical trial on which M.L. Johnson served as PI), Lilly work. R. Somwar reports grants and nonfinancial support from Loxo Oncology (Loxo (payment to institution for conduct of clinical trial on which M.L. Johnson served Oncology provided research funds as well as selpercatinib to conduct in vitro as PI), EMD Serono (payment to institution for conduct of clinical trial on which experiments) during the conduct of the study; grants from Helsinn Health Care M.L. Johnson served as PI), Genentech/Roche (payment to institution for conduct (research grant), Elevation Oncology (research grant), and Merus (research) outside of clinical trial on which M.L. Johnson served as PI), GlaxoSmithKline (payment the submitted work. J.F. Kherani reports other from Loxo Oncology (employee of to institution for conduct of clinical trial on which M.L. Johnson served as PI), Loxo Oncology at Lilly) during the conduct of the study and other from Loxo Gritstone Oncology (payment to institution for conduct of clinical trial on which Oncology (employee of Loxo Oncology at Lilly) outside the submitted work. B.C. Nair M.L. Johnson served as PI), Guardant Health (payment to institution for conduct reports other from Loxo Oncology, Inc., a subsidiary of Eli Lilly and Company (full- of clinical trial on which M.L. Johnson served as PI), Incyte (payment to time employment) during the conduct of the study. E.A. Olek reports other from Eli institution for conduct of clinical trial on which M.L. Johnson served as PI), Lilly and Company (employee and stock grant) during the conduct of the study. K. Janssen (payment to institution for conduct of clinical trial on which M.L. Ebata reports other from Loxo Oncology, Inc., a subsidiary of Eli Lilly and Company Johnson served as PI), Loxo Oncology (payment to institution for conduct (employment, stock) during the conduct of the study and other from Loxo Oncology, of clinical trial on which M.L. Johnson served as PI), Merck (payment to Inc., a subsidiary of Eli Lilly and Company (employment, stock) outside the submitted institution for conduct of clinical trial on which M.L. Johnson served as PI), work. J.F. Hechtman reports personal fees from Illumina and grants from Eli Lilly Mirati Therapeutics (payment to institution for conduct of clinical trial on which outside the submitted work. B.T. Li reports grants and personal fees from Roche M.L. Johnson served as PI), Novartis (payment to institution for conduct of Genentech (consulting/advisory board and clinical trial funding), Guardant Health clinical trial on which M.L. Johnson served as PI), Pfizer (payment to institution (consulting/advisory board and clinical research funding), Hengrui Therapeutics for conduct of clinical trial on which M.L. Johnson served as PI), Sanofi (payment (consulting/advisory board and clinical trial funding); personal fees from Thermo to institution for conduct of clinical trial on which M.L. Johnson served as PI), Fisher Scientific (consulting/advisory board), Mersana Therapeutics (consulting/ WindMIL (payment to institution for conduct of clinical trial on which M.L. advisory board); grants from Lilly [clinical trial funding and advisory board Johnson served as PI), Ribon Therapeutics (payment to institution for conduct (uncompensated)], AstraZeneca [clinical trial funding and advisory board of clinical trial on which M.L. Johnson served as PI); other (payment to institution (uncompensated)], Daiichi Sankyo (clinical trial funding), Amgen [clinical trial for consulting services performed by M.L. Johnson); grants from Acerta (payment funding and advisory board (uncompensated)], Illumina (clinical trial funding), to institution for conduct of clinical trial on which M.L. Johnson served as PI), GRAIL (clinical trial funding), BioMedValley Discoveries (clinical trial funding); Adaptimmune (payment to institution for conduct of clinical trial on which grants and nonfinancial support from MORE Health (clinical research funding, M.L. Johnson served as PI), Apexigen (payment to institution for conduct of academic travel support); nonfinancial support from Resolution Bioscience clinical trial on which M.L. Johnson served as PI), Array BioPharma (payment (academic travel support), Jiangsu Hengrui Medicine (academic travel support); to institution for conduct of clinical trial on which M.L. Johnson served as PI), and other from Boehringer Ingelheim [advisory board (uncompensated)] outside the BeiGene (payment to institution for conduct of clinical trial on which M.L. Johnson submitted work. In addition, B.T. Li has a patent for US62/685,057 issued (methods served as PI), BerGenBio (payment to institution for conduct of clinical trial for detecting HER2 dimerization in patients with HER2-mutant lung cancers and on which M.L. Johnson served as PI), Checkpoint Therapeutics (payment to predicting responsiveness to ado-trastuzumab emtansine) and a patent for US62/ institution for conduct of clinical trial on which M.L. Johnson served as PI), 514,661 issued (predicting Cancer Treatment Outcome with T-DM1). L.M. Sholl Corvus Pharmaceuticals (payment to institution for conduct of clinical trial on reports personal fees from EMD Serono and grants from Roche/Genentech (to her which M.L. Johnson served as PI), CytomX (payment to institution for conduct of institution) outside the submitted work. B.S. Taylor reports grants from Genentech, clinical trial on which M.L. Johnson served as PI), Daiichi Sankyo (payment Inc and personal fees from Boehringer Ingelheim and Loxo Oncology at Lilly outside to institution for conduct of clinical trial on which M.L. Johnson served as PI), the submitted work. M. Ladanyi reports grants from LOXO Oncology and personal Dynavax (payment to institution for conduct of clinical trial on which M.L. Johnson fees from Lilly Oncology outside the submitted work. P.A. J€anne reports personal fees served as PI), Genmab (payment to institution for conduct of clinical trial on from LOXO Oncology (consulting fees for drug development) and grants and which M.L. Johnson served as PI), Genocea Biosciences (payment to institution personal fees from Eli Lilly and Company (consulting fees for drug development

AACRJournals.org Clin Cancer Res; 2020 OF7

Rosen et al.

and sponsored research) during the conduct of the study; grants and personal fees fees and other from Foundation Medicine (employment) and Roche (equity) outside from AstraZeneca (consulting fees for drug development and sponsored research), the submitted work. No disclosures were reported by the other authors. Boehringer Ingelheim (consulting fees for drug development and sponsored research), Daiichi Sankyo (consulting fees for drug development and sponsored Authors’ Contributions research), Takeda Oncology (consulting fees for drug development and sponsored research), PUMA (consulting fees for drug development and sponsored research); and E.Y. Rosen: Conceptualization, formal analysis, investigation, writing-original personal fees from Pfizer (consulting fees for drug development), Roche/Genentech draft, writing-review and editing. M.L. Johnson: Conceptualization, formal analysis, (consulting fees for drug development), Chugai (consulting fees for drug supervision, writing-original draft. S.E. Clifford: Formal analysis, validation, development), Ignyta (consulting fees for drug development), SFJ Pharmaceuticals investigation. R. Somwar: Investigation, visualization. J.F. Kherani: Resources, (consulting fees for drug development), Araxes Pharmaceuticals (consulting fees for formal analysis, investigation, writing-original draft. J. Son: Formal analysis, drug development), Voronoi (consulting fees for drug development), Biocartis validation, investigation, visualization, writing-original draft. A.A. Bertram: (consulting fees for drug development), Novartis (consulting fees for drug Investigation. M.A. Davare: Resources. E. Gladstone: Investigation. E.V. Ivanova: development), Sanofi Oncology (consulting fees for drug development), Mirati Investigation. D.N. Henry: Investigation. E.M. Kelley: Investigation. M. Lin: Therapeutics (consulting fees for drug development), Transcenta (consulting fees Investigation. M.S.D. Milan: Investigation. B.C. Nair: Investigation. E.A. Olek: for drug development), Silicon Therapeutics (consulting fees for drug development), Investigation. J.E. Scanlon: Investigation. M. Vojnic: Formal analysis, Revolution Medicines (sponsored research), and Astellas Pharmaceuticals (sponsored investigation. K. Ebata: Resources, writing-original draft, project administration. research) outside the submitted work. In addition, P.A. J€anne has a patent for EGFR J.F. Hechtman: Resources, investigation. B.T. Li: Resources, investigation. Mutations issued and licensed to Lab Corp (receives postmarketing royalties from L.M. Sholl: Formal analysis, validation. B.S. Taylor: Formal analysis, supervision. € DFCI owned patent licensed to Lab Corp). S.M. Rothenberg reports other from M. Ladanyi: Formal analysis, supervision, investigation. P.A. Janne: Formal analysis, Loxo Oncology (employee, stock options) during the conduct of the study and supervision, validation, investigation, writing-original draft. S.M. Rothenberg: other from Pfizer (employee, stock options) outside the submitted work. A. Drilon Conceptualization, supervision, writing-original draft, project administration. reports other from Loxo/Lilly (honoraria/advisory), Ignyta/Genentech/Roche A. Drilon: Conceptualization, resources, formal analysis, supervision, (honoraria/advisory), and Blueprint Medicines (honoraria/advisory) during investigation, writing-original draft, writing-review and editing. G.R. Oxnard: the conduct of the study; other from Bayer (honoraria/advisory), Takeda/Ariad/ Conceptualization, formal analysis, supervision, investigation, writing-original Millenium (honoraria/advisory), TP Therapeutics (honoraria/advisory), AstraZeneca draft, writing-review and editing. (honoraria/advisory), Pfizer (honoraria/advisory), Helsinn (honoraria/advisory), Beigene (honoraria/advisory), BergenBio (honoraria/advisory), Hengrui Acknowledgments Therapeutics (honoraria/advisory), Exelixis (honoraria/advisory), Tyra Biosciences This work was supported by the NCI grants R01CA222823 (to P.A. J€anne) and (honoraria/advisory), Verastem (honoraria/advisory), MORE Health (honoraria/ R35CA220497 (to P.A. J€anne), NIH 2T32 CA009512–29A1 (to E.Y. Rosen), NIH advisory), AbbVie (honoraria/advisory), 14ner/Elevation Oncology (honoraria/ Cancer Center Grant P30CA008748 to Memorial Sloan Kettering Cancer Center, and advisory), Remedica Ltd. (honoraria/advisory), ArcherDX (honoraria/advisory), The Expect Miracles Foundation (to E.V. Ivanova) and Robert and Renee Belfer Monopteros (honoraria/advisory), Novartis (honoraria/advisory), EMD Serono Foundation (to E.V. Ivanova). (honoraria/advisory), Melendi (honoraria/advisory), Exelixis (associated research to institution), GlaxoSmithKlein (associated research to institution), Teva (associated research to institution), Taiho (associated research to institution), and The costs of publication of this article were defrayed in part by the payment of page PharmaMar (associated research to institution) outside the submitted work; royalties charges. This article must therefore be hereby marked advertisement in accordance from Wolters Kluwer; other from Merck (food/beverage), Puma (food/beverage), with 18 U.S.C. Section 1734 solely to indicate this fact. Merus (food/beverage), and Boehringer Ingelheim; and CME honoraria from Medscape, OncLive, PeerVoice, Physicians Education Resources, Targeted Oncology, Research to Practice, Axis, Peerview Institute, Paradigm Medical Received June 12, 2020; revised September 3, 2020; accepted October 16, 2020; Communications, WebMD, and MJH Life Sciences. G.R. Oxnard reports personal published first October 20, 2020.

References 1. Long GV, Stroyakovskiy D, Gogas H, Levchenko E, de Braud F, Larkin J, et al. 10. Subbiah V, Velcheti V, Tuch BB, Ebata K, Busaidy NL, Cabanillas ME, et al. Combined BRAF and MEK inhibition versus BRAF inhibition alone in mela- Selective RET kinase inhibition for patients with RET-altered cancers. noma. N Engl J Med 2014;371:1877–88. Ann Oncol 2018;29:1869–76. 2. Flaherty KT, Infante JR, Daud A, Gonzalez R, Kefford RF, Sosman J, et al. 11. Solomon BJ, Tan L, Lin JJ, Wong SQ, Hollizeck S, Ebata K, et al. RET solvent front Combined BRAF and MEK inhibition in melanoma with BRAF V600 mutations. mutations mediate acquired resistance to selective RET inhibition in RET-driven N Engl J Med 2012;367:1694–703. malignancies. J Thorac Oncol 2020;15:541–9. 3. Sherr CJ, Beach D, Shapiro GI. Targeting CDK4 and CDK6: from discovery to 12. Turke AB, Zejnullahu K, Wu Yi-L, Song Y, Dias-Santagata D, Lifshits E, et al. therapy. Cancer Discov 2016;6:353–67. Preexistence and clonal selection of MET amplification in EGFR mutant NSCLC. 4. O’Leary B, Finn RS, Turner NC. Treating cancer with selective CDK4/6 Cancer Cell 2010;17:77–88. inhibitors. Nat Rev Clin Oncol 2016;13:417–30. 13. Dagogo-Jack I, Yoda S, Lennerz JK, Langenbucher A, Lin JJ, Rooney MM, et al. 5. Oxnard GR, Hu Y, Mileham KF, Husain H, Costa DB, Tracy P, et al. Assessment MET alterations are a recurring and actionable resistance mechanism in ALK- of resistance mechanisms and clinical implications in patients with EGFR positive lung cancer. Clin Cancer Res 2020;26:2535–45. T790M-positive lung cancer and acquired resistance to osimertinib. 14. Lin JJ, Kennedy E, Sequist LV, Brastianos PK, Goodwin KE, Stevens S, et al. JAMA Oncol 2018;4:1527–34. Clinical activity of alectinib in advanced RET-rearranged non-small cell lung 6. Yu HA, Arcila ME, Rekhtman N, Sima CS, Zakowski MF, Pao W, et al. Analysis cancer. J Thorac Oncol 2016;11:2027–32. of tumor specimens at the time of acquired resistance to EGFR-TKI therapy in 15. Supplee JG, Milan MSD, Lim LP, Potts KT, Sholl LM, Oxnard GR, et al. 155 patients with EGFR-mutant lung cancers. Clin. Cancer Res 2013;19:2240–7. Sensitivity of next-generation sequencing assays detecting oncogenic fusions 7. Wu YL, Zhang L, Kim D-W, Liu X, Lee DH, Chih-Hsin Yang J, et al. Phase Ib/II in plasma cell-free DNA. Lung Cancer 2019;134:96–9. study of capmatinib (INC280) plus gefitinib after failure of epidermal growth 16. Scagliotti G, Moro-Sibilot D, Kollmeier J, Favaretto A, Cho EK, Grosch H, et al. A factor receptor (EGFR) inhibitor therapy in patients with EGFR-mutated, MET randomized-controlled phase 2 study of the MET antibody emibetuzumab in factor-dysregulated non-small-cell lung cancer. J Clin Oncol 2018;36:3101–9. combination with erlotinib as first-line treatment for EGFR mutation–positive 8. Oxnard GR, Yang JC-H, Yu H, Kim S-W, Saka H, Horn L, et al. TATTON: a NSCLC patients. J Thorac Oncol 2020;15:80–90. multi-arm, phase Ib trial of osimertinib combined with selumetinib, savolitinib, 17. Cappuzzo F, J€annePA,SkokanM,FinocchiaroG,RossiE,LigorioC, or durvalumab in EGFR-mutant lung cancer. Ann Oncol 2020;31:507–16. et al. MET increased gene copy number and primary resistance to 9. Mulligan LM. Progress and potential impact of RET kinase targeting in cancer. gefitinib therapy in non-small-cell lung cancer patients. Ann Oncol Expert Rev Proteomics 2016;13:631–3. 2009;20:298–304.

OF8 Clin Cancer Res; 2020 CLINICAL CANCER RESEARCH

MET-Dependent Resistance to Selective RET Inhibitor in NSCLC

18. Drilon A, Oxnard G, Wirth L, Besse B, Gautschi O, Tan SWD, et al. PL02.08 20. Baltschukat S, Engstler BS, Huang A, Hao H-X, Tam A, Wang HQ, et al. registrational results of LIBRETTO-001: a phase 1/2 trial of LOXO-292 in Capmatinib (INC280) is active against models of non–small cell lung cancer and patients with RET fusion-positive lung cancers. J Thorac Oncol 2019;14:S6–S7. other cancer types with defined mechanisms of MET activation. Clin Cancer Res 19. Tong JH, Yeung SF, Chan AWH, Chung LY, Chau SL, Lung RWM, et al. MET 2019;25:3164–75. amplification and exon 14 splice site mutation define unique molecular sub- 21. Paik PK, Felip E, Veillon R, Sakai H, Cortot AB, Garassino MC, et al. Tepotinib in groups of non-small cell lung carcinoma with poor prognosis. Clin Cancer Res non–small-cell lung cancer with MET exon 14 skipping mutations. N Engl J Med 2016;22:3048–56. 2020;383:931–43.

AACRJournals.org Clin Cancer Res; 2020 OF9

Overcoming MET-Dependent Resistance to Selective RET Inhibition in Patients with RET Fusion−Positive Lung Cancer by Combining Selpercatinib with Crizotinib

Ezra Y. Rosen, Melissa L. Johnson, Sarah E. Clifford, et al.

Clin Cancer Res Published OnlineFirst October 20, 2020.

Updated version Access the most recent version of this article at: doi:10.1158/1078-0432.CCR-20-2278

Supplementary Access the most recent supplemental material at: Material http://clincancerres.aacrjournals.org/content/suppl/2020/10/20/1078-0432.CCR-20-2278.DC1

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/early/2020/11/20/1078-0432.CCR-20-2278. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.