Response to ERBB3-Directed Targeted Therapy in NRG1-Rearranged Cancers

Published OnlineFirst April 2, 2018; DOI: 10.1158/2159-8290.CD-17-1004

Research Brief

Alexander Drilon1,2, Romel Somwar1, Biju P. Mangatt3, Henrik Edgren4, Patrice Desmeules1, Anja Ruusulehto4, Roger S. Smith1, Lukas Delasos1, Morana Vojnic1, Andrew J. Plodkowski1, Joshua Sabari1, Kenneth Ng1, Joseph Montecalvo1, Jason Chang1, Huichun Tai1, William W. Lockwood5, Victor Martinez5, Gregory J. Riely1,2, Charles M. Rudin1,2, Mark G. Kris1,2, Maria E. Arcila1, Christopher Matheny3, Ryma Benayed1, Natasha Rekhtman1, Marc Ladanyi1, and Gopinath Ganji3

abstract NRG1 rearrangements are oncogenic drivers that are enriched in invasive mucinous adenocarcinomas (IMA) of the lung. The oncoprotein binds ERBB3–ERBB2 heterodimers and activates downstream signaling, supporting a therapeutic paradigm of ERBB3/ ERBB2 inhibition. As proof of concept, a durable response was achieved with anti-ERBB3 mAb therapy (GSK2849330) in an exceptional responder with an NRG1-rearranged IMA on a phase I trial (NCT01966445). In contrast, response was not achieved with anti-ERBB2 therapy (afatinib) in four patients with NRG1-rearranged IMA (including the index patient post-GSK2849330). Although in vitro data supported the use of either ERBB3 or ERBB2 inhibition, these clinical results were consistent with more profound antitumor activity and downstream signaling inhibition with anti-ERBB3 versus anti-ERBB2 therapy in an NRG1-rearranged patient-derived xenograft model. Analysis of 8,984 and 17,485 tumors in The Cancer Genome Atlas and MSK-IMPACT datasets, respectively, identifiedNRG1 rearrangements with novel fusion partners in multiple histologies, including breast, head and neck, renal, lung, ovarian, pancreatic, prostate, and uterine cancers.

SIGNIFICANCE: This series highlights the utility of ERBB3 inhibition as a novel treatment paradigm for NRG1-rearranged cancers. In addition, it provides preliminary evidence that ERBB3 inhibition may be more optimal than ERBB2 inhibition. The identification ofNRG1 rearrangements across various solid tumors supports a basket trial approach to drug development. Cancer Discov; 8(6); 1–10. ©2018 AACR.

See related commentary by Wilson and Politi, p. 676.

INTRODUCTION kinase inhibitors (TKI) are currently approved or being inves- Oncogenic gene fusions or rearrangements are identi- tigated for the treatment of these patients. Similarly, dramatic fied across a wide range of solid tumors (1). Many of these and durable responses have been reported in NTRK1/2/3– events, such as fusions involving ALK, ROS1, RET, NTRK1/2/3, rearranged solid tumors regardless of histology (2), and in FGFR1/2/3, and PDGFB, are clinically actionable. In lung ade- select cancers with FGFR1/2/3 or RET rearrangements (1). nocarcinomas, targeted therapy for patients with ALK- or Fusions involving the neuregulin-1 gene (NRG1) were iden- ROS1-rearranged tumors is highly effective, and several tyrosine tified by Fernandez-Cuesta and colleagues in non–small cell

1Memorial Sloan Kettering Cancer Center, New York, New York. 2Weill Corresponding Author: Alexander Drilon, Memorial Sloan Kettering Cornell Medical Center, New York, New York. 3GlaxoSmithKline, Collegeville, Cancer Center, 885 2nd Avenue, New York, NY 10017. Phone: 646-888- Pennsylvania. 4MediSapiens, Helsinki, Finland. 5University of British 4206; Fax: 646-888-4263; E-mail: [email protected] Columbia, Vancouver, British Columbia, Canada. doi: 10.1158/2159-8290.CD-17-1004 Note: Supplementary data for this article are available at Cancer Discovery ©2018 American Association for Cancer Research. Online (http://cancerdiscovery.aacrjournals.org/).

OF1 | CANCER DISCOVERY June 2018 www.aacrjournals.org

ERBB3-Directed Targeted Therapy in NRG1-Rearranged Cancers RESEARCH BRIEF lung cancers (NSCLC) with an initially approximated fre- tory volumetric analysis revealed a 90% reduction in tumor quency of 1.7% in lung adenocarcinomas (3). Transcriptome volume at the nadir. An on-treatment tumor biopsy was sequencing revealed fusion of CD74 to NRG1 exons encoding performed on day 14 of therapy, but little tumor tissue was the NRG1 III-β3 isoform. Other fusions, such as SLC3A2– present, likely due to the robust response observed; pharma- NRG1 and SDC4–NRG1, have also been identified (4, 5).NRG1 codynamic changes thus could not be assessed. The patient fusions are particularly enriched in invasive mucinous adeno- tolerated the drug with no major issues. The response was carcinomas (IMA) of the lung, where they are found in 27% durable and lasted 1 year and 7 months, exceeding the dura- to 31% of cases. These are often mutually exclusive with KRAS tion, in aggregate, of the prior systemic therapies he received. mutations, a driver known to be enriched in IMAs (4, 6–8). Although GSK2849330 bears antibody-dependent cytotox- Activating NRG1 rearrangements are drivers of cancer icity (ADCC) and complement-dependent cytotoxicity (CDC) growth. NRG1 encodes several isoforms that contain an EGF- enhancements (13), no other responses were noted in the same like domain that serves as the ligand of the receptor ERBB3 trial (NCT01966445, n = 29), which enrolled several patients (9). Chimeric transmembrane proteins encoded by NRG1 with similar and higher ERBB3 expression. None of these fusions are predicted to maintain this extracellular domain, other tumors were known to harbor an NRG1 fusion. These thus activating ERBB3 in a para/juxtacrine or autocrine data strongly suggest that ADCC, CDC, or ERBB3 expression fashion (10). NRG1 binding to ERBB3 results in heterodi- alone do not predict responses, and that our index NRG1- merization of the latter with ERBB2, activation of down- rearranged patient was an exceptional responder to therapy. stream signaling including the ERK, PI3K–AKT, and NF-κB In contrast to these outcomes, treatment with afatinib did pathways, and increased tumor cell proliferation and growth not achieve a response in 4 patients with NRG1-rearranged (9, 11). Therefore, rational targeting of ERBB3 or ERBB2 in lung cancers. After progression of disease, the same patient NRG1-rearranged tumors could be efficacious (3). with a CD74–NRG1-rearranged lung cancer described above In this article, we provide proof of principle that targeting was immediately transitioned to afatinib (40 mg daily). Repeat ERBB3 in NRG1-rearranged cancers represents a novel thera- imaging after 8 weeks of therapy revealed progressive disease, peutic paradigm. In contrast, we demonstrate that targeting with an increase in the right basilar mass and paratracheal ERBB2 is not as effective in patients in this series despite pre- lymph node, and new satellite nodules. In addition to this clinical data supporting its use and previously published clin- case, three other targeted therapy–naïve patients with NRG1- ical reports. Finally, we present large-scale genomic profiling rearranged lung cancers received afatinib at 40 mg daily, none data supporting a basket trial drug-development approach of whom responded to therapy. The clinicopathologic and for NRG1 rearrangements that extends beyond NSCLCs to molecular details of these three additional patients are sum- other cancers. marized in Supplementary Table S1. The first patient with a CD74–NRG1 fusion had a best response of stable disease at RESULTS 5 weeks (7% disease shrinkage), but frank disease progression at 13 weeks. The second patient with an SDC4–NRG1 fusion Outcomes with Targeted Therapy in Patients with had frank disease progression at 6 weeks, similar to the third NRG1-Rearranged Lung Cancers patient with a CD74–NRG1 fusion who had primary disease An 86-year-old man presented with recurrent, unresectable, progression at 8 weeks; both patients died of disease progres- advanced invasive mucinous adenocarcinoma 9 months after sion shortly after discontinuing afatinib. an initial resection and radiation for stage IIA (pT2bN0M0) disease. He was treated with carboplatin and pemetrexed, Targeted Inhibition of ERBB3/ERBB2 Results in followed by paclitaxel and bevacizumab (with a best objec- Decreased Growth of Cell Lines with Aberrant tive response of stable disease with both regimens), and NRG1 Expression finally nivolumab (with primary progressive disease). Broad, Published preclinical data on ERBB3/ERBB2 inhibition hybrid capture–based next-generation sequencing (NGS) in NRG1-altered cell lines is limited (4), and the activity using MSK-IMPACT (12) identified an in-frameCD74–NRG1 of ERBB2 inhibition with select clinically available TKIs translocation (Fig. 1A), resulting in the fusion of exons 1 to has largely been examined in cells with ectopic expression 6 of CD74 with exons 6 to 13 of NRG1; the only other altera- of CD74–NRG1 cDNA (3). We thus explored the effect of tion identified was anARID2 c.592A>T nonsense mutation. ERBB3/ERBB2 inhibition by genetic and pharmacologic The patient was enrolled in an NSCLC expansion cohort of a approaches in cell lines with endogenous NRG1 alterations. phase I trial (NCT01966445) of GSK2849330, an anti-ERBB3 We first examined the breast cancer cell line MDA-MB-175- mAb. GSK2849330 is a humanized mAb that binds with high VII, previously published to harbor a DOC4–NRG1 fusion (10, affinity to the ERBB3 domain III, where it blocks the binding 14), and confirmed the finding via RT-PCR (Fig. 2A). Orthog- of NRG1 and inhibits receptor heterodimerization (13). onal sequencing of the PCR amplicon revealed that exon 12 GSK2849330 was administered at the recommended phase of DOC4 is fused in frame with exon 2 of NRG1 [confirmed II dose of 30 mg/kg weekly during induction, followed by by targeted RNA sequencing (RNA-seq); Supplementary Fig. maintenance therapy of 30 mg/kg every 2 weeks. A confirmed S1A]. To identify a lung cancer cell line with aberrant expres- partial response to therapy was achieved, with substantial sion of NRG1, 67 lung cancer cell lines were screened using shrinkage of a right lower lobe mass (Fig. 1B) accompanied microarray expression data. The HCC-95 cell line had the by resolution of the patient’s dyspnea. Although the maximal highest level of NRG1 mRNA expression (3-fold higher than decrease in measurable tumor burden by RECIST v1.1 was the next highest cell line, Supplementary Fig. S1B). Although 32%, this underestimated total tumor shrinkage; an explora- an NRG1 fusion was not detected in this cell line using the

June 2018 CANCER DISCOVERY | OF2

RESEARCH BRIEF Drilon et al.

A Invasive mucinous adenocarcinoma

Tumor

Resection for stage IIA Normal (pT2bN0M0) disease followed by radiation

Recurrent metastatic disease CD74–NRG1 fusion identified

Anti-ERBB3 mAb: GSK2849330

Carboplatin Paclitaxel 19 Months confirmed pemetrexed Paclitaxel bevacizumab Nivolumab partial response (1.4 mo SD) (5.5 mo SD) (9 months SD) (1.3 mo PD)

B Baseline Week 6

Figure 1. Clinical response to anti-ERBB3 mAb therapy in a patient with an advanced NRG1-rearranged non–small cell lung cancer. A, A never-smoker with advanced invasive mucinous adenocarcinoma was treated with several lines of systemic therapy (chemotherapy and immune therapy) with no response to any of these treatments. His tumor was subsequently found to harbor an in-frame CD74–NRG1 fusion that includes the EGF-like extracellular domain that serves a binding site for ERBB3. Targeted therapy with GSK2849330, an anti-ERBB3 monoclonal therapeutic antibody, was initiated on a phase I clinical trial. B, A durable (19 months) and confirmed partial response (RECIST version 1.1) was achieved that exceeded the duration of disease control achieved on all prior systemic therapy regimens in aggregate. Substantial disease shrinkage of a right lower lobe mass was noted early (by week 6 as shown; circled) in the course of therapy. SD, stable disease; PD, progressive disease.

OF3 | CANCER DISCOVERY June 2018 www.aacrjournals.org

ERBB3-Directed Targeted Therapy in NRG1-Rearranged Cancers RESEARCH BRIEF

A175-MB-175-VII DOC4 (exon 12) NRG1 (exon 2)

MD MCF-7 HCC-95 HCC-827 control Negative AAGTTGG TTCCTTTTCCCCAACCTTGGCCTTCCTGGCCTTCCC AATG AA 600 500 400 300 DOC4–NRG1 200 100

GAPDH 160 170 180 190 200

25 BDNRG1

MDA175 VII HCC-95 MCF-7 NCI-H292 150 − +−++−− −−+−+−++−− −−+ IgG −−++−−−−++−−++−−−−++GSK2849330 HCC95 −−−+ ++−−−−+ −−++−−− ++ HRGB1

HER3 pY1289

HER3 C AKT pS473 500 60 * * 400 40 AKT 300 200 20 β-Actin

100 units) (relative (relative units) (relative 0 0 NRG1 mRNA expression NRG1 mRNA expression

MCF7 HCC95 HCC827

MDA-MB-75VII F MDA-MB-175-VII E 100 MDA-MB-175-VII MCF-7 75 4 50 Untreated IgG GSK2849330 50 Untreated IgG GSK2849330 * ) ) 40 * 6

3 6 25 * 30 * 2 0 20 Relative cell number (%) Relative 1 LU ( × 10 RLU ( × 10 -sh1 -sh1 -sh1 10 NT-sh 0 EGFREGFR-sh2 0 ERBB2ERBB23ERBB-sh3 ERBB3-sh2 0510 15 0510 15 Time (days) Time (days) MDA-MB-175-VII * HCC-95 NCI-H292 20 50 50 15 * Untreated IgG GSK2849330 Untreated IgG GSK2849330 40 * ) 40 )R * 6 6 10 30 activity 30 5 20 20 Relative capase-3/7 Relative 0 RLU ( × 10 10 RLU ( × 10 10 -sh1 -sh2 -sh1 NT-sh -sh3 0 0 0510 15 0510 15 EGFREGFR ERBB22ERBBERBB3-sh1 ERBB3-sh2 Time (days) Time (days)

Figure 2. Cell lines with aberrant expression of NRG1 are exquisitely sensitive to downregulation of ERBB3 signaling. A, Expression of the DOC4–NRG1 fusion was confirmed by RT-PCR (left) and by DNA sequencing of the PCR amplicon (right) in MDA-MB-175-VII cells. B, Copy-number breakpoints affect- ing NRG1 in HCC-95, a human lung cancer cell line, are depicted using normal human DNA as reference. The outer circle depicts human autosomes (1–22), chromosomes X and Y, and the mitochondrial genome (M). Chromosome 8, zooming into the amplifiedNRG1 locus (inset), is shown. Genome-wide distri- bution of log2 ratios is represented in concentric circles as gains (red) and losses (blue), each data point representing a probe from the array according to the locations in the human genome (NCBI 36). C, Quantitative RT-PCR was used to determine the levels of NRG1 mRNA in MDA-MB-175-VII and HCC-95 using MCF-7 and HCC827 as control cell lines for comparison. D, Cells were treated with GSK2849330 for 1 hour and then stimulated with heregulin (HRGB1) for 30 minutes before preparation of whole-cell extracts for western blotting with the antibodies described. A representative blot is shown. E, Cells were left untreated, or treated with 10 μg/mL GSK2849330 or nonspecific IgG and then measured for growth by CTG assays at the indicated time points. The y-axis represents the mean ± SD of relative luciferase units (RLU) measured in triplicate. F, Cells were infected with lentivirus containing nontargeting (NT) or shRNAs targeting the genes shown and then examined for relative number of cells remaining after 7 days of infection (top) or relative caspase 3/7 enzymatic activity (bottom). Results represent the mean ± SD of two experiments in which each condition was assayed in duplicate. *, Significantly different from NT-shRNA,P < 0.05, two tailed t test.

June 2018 CANCER DISCOVERY | OF4

RESEARCH BRIEF Drilon et al.

MSK Solid Fusion Assay, NRG1 was subsequently found treated with vehicle, afatinib (15 mg/kg daily), GSK2849330 to be amplified on the basis of array comparative genomic (25 mg/kg, biweekly), or control IgG (25 mg/kg, biweekly). hybridization analysis (Fig. 2B). Both MDA-MB-175-VII and Afatinib treatment caused a significant reduction in tumor HCC-95 significantly overexpressedNRG1 relative to tissue growth compared with vehicle (Fig. 3B; Supplementary Fig. type-matched control cell lines (Fig. 2C). In addition, using S5A); however, no tumor regression was observed. a phospho–receptor tyrosine kinase (RTK) array that allows In contrast, GSK2849330 elicited a more profound antitu- for profiling the activation state of 49 RTKs (15), we observed mor response as evidenced by durable and complete tumor that the MDA-MB-175-VII cell line had high levels of phospho- regression (Fig. 3B; Supplementary Fig. S5A) that persisted even rylation of EGFR, ERBB2, and ERBB3, as expected, but little after treatment was discontinued after 35 days of dosing (Sup- activation of any of the remaining 46 RTKs (Supplementary plementary Fig. S5B). Pharmacodynamic analysis of samples Fig. S1C). collected 4 hours after a single dose treatment of GSK2849330 To further investigate these cell lines with aberrant NRG1 demonstrated near-complete inhibition of phospho-ERBB3 expression, we examined the effect of the anti-ERBB3 mAb and phospho-AKT relative to IgG control (Fig. 3C). Afatinib- GSK2849330 on proliferation and activation of the ERBB3 treated samples, however, showed relatively lower inhibition of pathway. MDA-MB-175-VII and HCC-95 cells showed higher phospho-ERBB3 and little to no change in phospho-AKT. basal levels of phospho-ERBB3 compared with control cell Taken together, these in vivo results were analogous to the lines (MCF-7 and H292) with low NRG1 expression (Fig. 2D). differential clinical responses observed with ERBB3- versus Addition of exogenous heregulin did not further stimulate ERBB2-directed targeted therapy as described earlier, where a ERBB3, as evidenced by little effect on phospho-ERBB3 or deep and durable objective response to therapy was noted with phospho-AKT levels in the MDA-MB-175-VII or HCC-95 cell GSK2849330, as opposed to afatinib therapy, where no responses lines, unlike the effects observed in control cell lines. MDA- were observed in patients with NRG1-rearranged tumors. MB-175-VII and HCC-95 cells treated with GSK2849330 showed suppression of ERBB2, ERBB3, and AKT phospho- NRG1 Fusions Are Identified Across rylation compared with treatment with a nonspecific IgG a Variety of Cancers control. Moreover, relative to the control lines, both cell lines Recognizing the clinical activity of targeted therapy in exhibited significant growth inhibition upon GSK2849330 NRG1-rearranged lung cancers, we proceeded to determine treatment over the IgG control (Fig. 2E). We observed simi- the prevalence of NRG1 fusions in advanced solid tumors of lar results repeating the same experiments with a variety both lung and non-lung origin. Using MSK-IMPACT alone, of ERBB2/EGFR inhibitors, such as afatinib, erlotinib, and among 17,485 patients with a variety of advanced solid neratinib (Supplementary Figs. S2 and S3), complementing tumors, NRG1 rearrangements were detected in NSCLCs prior experience with trastuzumab, pertuzumab, or T-DM1 [3/2,079 patients (0.14%) with lung adenocarcinomas], pan- treatment in vitro (16). These data collectively show that creatic adenocarcinoma [1/791 patients (0.13%) with pancre- these models, despite their molecular differences in terms of atic adenocarcinoma], and ER+/HER2− breast cancer [1/2,703 NRG1 alterations, are similar in their constitutive activation, patients (0.04%) with breast carcinoma]. dependence on the pathway, and sensitivity to inhibitors. Knowing that NRG1 fusions are more common in IMAs To dissect the roles of ERBB family members on in vitro of the lung, and that the NRG1 fusions of 2 patients who growth and survival, we tested shRNAs targeting EGFR, ERBB2, received targeted therapy in this series (Supplementary Table or ERBB3 in the NRG1 fusion–positive cell line MDA-MB-175- S1) were detected by targeted RNA-seq (MSK Solid Fusion VII (Supplementary Fig. S4). ERBB2 or ERBB3 knockdown Assay, Archer FusionPlex) and not NGS of DNA (MSK- reduced cell growth significantly (Fig. 2F, top) and induced IMPACT), we performed additional targeted RNA-seq of >8-fold increase in caspase-3/7 activity (Fig. 2F, bottom). IMAs of the lung that did not harbor KRAS mutations using However, these cells were less sensitive to EGFR inhibition, the MSK Solid Fusion Assay. In patients with KRAS wild- phenocopying the reduced cell viability observed with ner- type IMAs, NRG1 fusions were detected in 4 of 36 patients atinib and afatinib relative to erlotinib (Supplementary Fig. (11%). Targeted RNA-seq also detected an NRG1 fusion in a S3). Collectively, these results support that NRG1 alteration pancreatic adenocarcinoma not previously detected by MSK- drives basal activation and dependence on the ERBB2/ERBB3 IMPACT alone. signaling pathway and confers sensitivity to ERBB2/ERBB3 Collectively, MSK-IMPACT and the MSK Solid Fusion inhibitors such as GSK2849330 and afatinib in vitro, reflecting Assay detected 10 NRG1 fusions in total: 7 in lung adeno- the clinical experience, here and elsewhere (5, 17, 18). carcinomas, 2 in pancreatic adenocarcinomas, and 1 in a breast carcinoma. All fusions included exon 6 that encodes Differential Growth Inhibition Is Observed with the EGF-like domain of NRG1 that likely binds to ERBB3 ERBB3 versus ERBB2 Targeting in OV-10-0050, to trigger downstream signaling. CD74 was the most com- a PDX Model Harboring an NRG1 Fusion mon upstream partner (Fig. 4A). Additional fusions included Profiling a variety of patient-derived xenograft (PDX) mod- the known SDC4–NRG1 and the novel ROCK1–NRG1 and els by RNA-seq identified OV-10-0050, an ovarian model with FOXA1–NRG1 fusions (Fig. 4B). outlier expression of NRG1 mRNA (Fig. 3A). Further analysis We then performed an analysis of RNA-seq data for 8,984 of this sample revealed a novel CLU–NRG1 fusion resulting tumors from The Cancer Genome Atlas (TCGA). NRG1 rear- from the intragenic fusion of exon 2 of CLU with exon 6 of rangements with a wide variety of novel 5′ and 3′ partners NRG1, retaining the EGF-like extracellular domain. Mice were found across several tumor histologies. These included bearing OV-10-0050 tumors implanted subcutaneously were breast, lung, and pancreatic cancers, matching the results

OF5 | CANCER DISCOVERY June 2018 www.aacrjournals.org

ERBB3-Directed Targeted Therapy in NRG1-Rearranged Cancers RESEARCH BRIEF

A Sample ID # of 5′ → 3′ 5′ breakpoint3′ breakpoint5′ gene 3′ gene 5′ gene 3′ gene Reads FPKM FPKM CNV CNV OV-10- 861 CLU-exon2 chr8: chr8: 110.87 14.97 2.29 2.25 0050 →NRG1-exon6 27467992 32,585,467

Outlier NRG1 Ig1*:Ig-like C2-type 1 mRNA expression Signal peptide CLU Clusterin NM_001831.3 33 OV-10-0050 CLU→NRG1 EGF-like 506 168 20 NRG1 Ig1* EGF-like 640 NM_013964.3 18 16 Fusion junction sequence 14 12 10 8 6 4 2 0 LI-3-0509 U-03-0008 U-03-0022 U-03-0159 U-03-0257 U-03-0217 U-03-1199 U-03-0011 U-03-1082 U-03-0126 U-03-0004 U-03-0164 U-03-0875 U-03-0140 U-03-0103 U-03-0840 U-03-0255 U-03-0357 U-03-0010 U-03-1036 U-03-1091 GI-20-0008 ST-02-0395 LU-01-0547 LU-01-0236 LU-01-0021 LU-01-0057 LU-01-0011 LU-01-0275 LU-01-0529 LU-01-0462 LU-01-0413 LU-01-0006 LU-01-0251 LU-01-0720 LU-01-0683 LU-01-0252 LU-01-0695 LU-01-0322 LU-01-0055 LU-01-0556 ES-06-0222 BR-05-0044 BR-05-0014 HN-13-0008 CH-17-0044 CH-17-0050 CO-04-0331 CO-04-0343 ME-21-0009 CO-04-0327

VehicleAfatinib B C 1,500 pAKT

1,000 AKT

pHER2 500 HER2 0 pHER3 –50 HER3 Change in tumor volume (%) Change in tumor volume

–100 Actin Afatinib Vehicle (15 mg/kg QD)

1,500 IgG control GSK2849330

pAKT 1,000 AKT 500 pHER2

0 HER2

pHER3 –50

Change in tumor volume (%) Change in tumor volume HER3 –100 Actin Vehicle IgG GSK2849330 (25 mg/kg BIW) (25 mg/kg BIW)

Figure 3. In vivo growth inhibition elicited by afatinib and GSK2849330 in a CLU–NRG1-rearranged PDX mouse model. A, A panel of PDX models was profiled by RNA-seq. An ovarian cancer PDX model, OV-10-0050, showed a high level ofNRG1 mRNA expression, resulting from a novel CLU–NRG1 intragenic rearrangement (inset) where exon 2 of CLU (chromosome 8) was fused with exon 6 or NRG1 (chromosome 8) with retention of the extracellular EGF-like domain in the fusion product, as shown. B, Female BALB/c nude mice were subcutaneously implanted with CLU–NRG1-rearranged tumors and treated with afatinib at 15 mg/kg daily (QD) for 28 days, resulting in a significant inhibition of tumor growth compared with the vehicle arm n( = 6, top). The same PDX mouse model was treated with the anti-ERBB3 mAb GSK2849330 at 25 mg/kg biweekly (BIW) for 35 days, resulting in a dramatic response of complete tumor regression compared with vehicle- or IgG-treated controls (n = 10, bottom). Changes from baseline tumor volume (TV), as measured by [TV (final) − TV (initial)]× 100/TV (initial), are shown in the waterfall plots. C, Pharmacodynamic changes in tumor tissues as measured 4 hours after a single dose of treatment with afatinib, vehicle, GSK2849330, or IgG control (n = 3) are shown.

June 2018 CANCER DISCOVERY | OF6

RESEARCH BRIEF Drilon et al.

AC CD74 NRG1 Partner NRG1 987 65 432112 345678910111213 Exon 5 (ENST00000361243) Exon 2 (ENST00000356819) t(8;15) AKAP13–NRG1 NM_001025159 NM_013956 [BRCA] (9 exons) 5q32 (13 exons) 8q12 Exon 3 (ENST00000367816) Exon 2 (ENST00000356819) t(1;8) ATPB1–NRG1 ][PAAD Exon 3 (ENST00000215742) Exon 2 (ENST00000356819) 1234 566789 10 111213 t(8;22) THAP7–NRG1* [LUSC] Exon 6 (ENST00000260356) Exon 6 (ENST00000356819) t(8;15) THBS1–NRG1 [HNSC] B Exon 9 (ENST00000394836) Exon 6 (ENST00000356819) Partner NRG1 t(8;11) RAB3IL1–NRG1 [OV] Exon 6 Exon 6 Exon 4 (ENST00000372733) Exon 6 (ENST00000356819) t(8;20) SDC4–NRG1 [LUAD] Exon 7 Exon 6 Exon 1 (ENST00000342988) Exon 6 (ENST00000356819) t(8;18) SMAD4–NRG1* [LUSC] Exon 8 Exon 6 t(5;8) CD74–NRG1 Exon 3 (ENST00000401827) Exon 6 (ENST00000356819) t(8;8) PDE7A–NRG1 Exon 7 Exon 2 [HNSC] Exon 2 (ENST00000325083) Exon 6 (ENST00000356819) t(8;8) PCM1–NRG1 Exon 6 Exon 3 [KIRC]

Exon 2 Exon 4 t(8;20) SDC4–NRG1 Protein domains EGF Exon 2 Exon 2 NRG1 Partner t(8;18) ROCK1–NRG1 Protein domains Exon 3 (ENST00000356819) Exon 2 (ENST00000518111) Exon 2 Exon 2 t(8;8) NRG1–STMN2 TM t(8;14) FOXA1–NRG1 [PRAD] EGF Exon 1 (ENST00000520407) Exon 2 (ENST00000341744) t(8;20) NRG1–PMEPA1 [UCS]

D NRG1 expression

2.44e–04

1.53e–05

9.54e–07 RSEM (scaled estimate)

5.96e–08 OV UCS KIRC PAAD LUAD LUSC BRCA PRAD HNSC Tumor type

Figure 4. NRG1 rearrangements are found in multiple solid tumors. A, A schematic of CD74–NRG1, the most commonly identified genomic rearrangement identified in lung cancers in this series, is shown. TheCD74 gene (chromosome 5q) is disrupted and inverted downstream of exon 5 and subsequently ligated to a position upstream of exon 6 of the NRG1 gene. B, Structural features of fusions, involving the indicated 5′ and 3′ chromosomal partners, identified by next-generation DNA sequencing (MSK-IMPACT;n = 17,485 tumors profiled) or targeted RNA-seq (MSK Solid Fusion Assay) are shown. The EGF-like domain is maintained in all fusions identified. C, Fusion junction exons, the associated chromosomal partners, and corresponding transcripts (Ensembl database version 75) are shown for all fusions detected by analyzing the TCGA dataset (n = 8,984 tumors analyzed), with the EGF-like domain indicated, wherever applicable. NRG1 fusion partners are not drawn to scale. The asterisk indicates that two distinct fusions were found in the same tumor sample. D, NRG1 RNA-seq expression is plotted across fusion-positive TCGA cancer types (x-axis) with fusion-positive samples indicated in red. Note that only one NRG1 expression value is plotted for the lung squamous cell carcinoma (LUSC) sample with two distinct fusions. BRCA, breast-invasive carcinoma; HNSC, head and neck squamous cell carcinoma; KIRC, kidney renal clear cell carcinoma; LUAD, lung adenocarcinoma; OV, ovarian serous cystadenocarcinoma; PAAD, pancreatic adenocarcinoma; PRAD, prostate adenocarcinoma; UCS, uterine carcinosarcoma.

OF7 | CANCER DISCOVERY June 2018 www.aacrjournals.org

ERBB3-Directed Targeted Therapy in NRG1-Rearranged Cancers RESEARCH BRIEF obtained using MSK-IMPACT and the MSK Solid Fusion tumab (U3-1287, AMG-888), seribantumab (MM-121), lumre- Assay, but also additional tumor types, such as ovarian and tuzumab, AV-203, and LJM716 (20–24), but early testing did head and neck cancers. not focus on molecular enrichment for NRG1 fusions. A case The identified fusions and their corresponding frequencies could also be made for the use of ERBB2 inhibitors, such as were AKAP13–NRG1 (breast cancer; 0.01%), THBS1–NRG1 afatinib, lapatinib, neratinib, trastuzumab, T-DM1, or pertu- and PDE7A–NRG1 (head and neck cancer; 0.49%), SDC4– zumab, in patients with NRG1-rearranged cancers; however, NRG1 (lung adenocarcinoma; 0.22%), PCM1–NRG1 (renal our study and previous data (16) suggest that agents such as clear cell carcinoma; 0.19%), THAP7–NRG1 and SMAD4– afatinib, trastuzumab, pertuzumab, or T-DM1 may provide NRG1 (squamous cell lung cancer; 0.21%), RAB3IL1–NRG1 more limited benefits. Finally, combinations of one or more (ovarian cancer; 0.24%), ATP1B1–NRG1 (pancreatic cancer; of these agents may maximize antitumor activity in NRG1- 0.55%), NRG1–PMEPA1 (uterine carcinosarcoma; 1.75%), and rearranged tumors as previously explored preclinically (16). NRG1–STMN2 (prostate cancer; 0.3%; Fig. 4C). We recommend that screening for NRG1-rearranged tumors Of these rearrangements, the NRG1–PMEPA1, ATP1B1– be performed. In this article, NRG1 rearrangements were more NRG1, SDC4–NRG1, PDE7A–NRG1, and RAB3IL1–NRG1 fusions optimally detected when a combination of DNA-based tar- were clearly predicted to be in-frame. For the remaining fusions, geted hybrid capture–based NGS and targeted RNA-seq with frame retention prediction was more difficult; however, fusion anchored multiplex PCR was performed. Considering the mRNA was still clearly detected (Fig. 4D). All fusions, except diversity of NRG1 fusion partners, the latter offers the distinct NRG1–PMEPA1, NRG1–STMN2, and PCM1–NRG1, retained the advantage of detecting unknown fusion partners using a uni- EGF-like domain of NRG1 and were thus predicted to be func- directional targeted approach. Consistent with previous data tional and activating, as described above. (3, 6, 25), the NRG1 rearrangements we identified occurred predominantly in lung cancers. These cancers thus represent a target population to screen in the absence of other actionable DISCUSSION drivers, especially in driver-negative lung IMAs. In this article, we provide early clinical proof-of-principle Finally, using two large genomic datasets, we show that data supporting the use of ERBB3-directed targeted therapy in NRG1 rearrangements can be detected across a variety of patients with advanced NRG1-rearranged cancers. GSK2849330, other cancers, including breast, head and neck, renal, ovar- an anti-ERBB3 mAb, elicited a dramatic and durable response ian, pancreatic, prostate, and uterine cancers. This supports in an exceptional responder with an advanced CD74–NRG1- the utility of a basket trial approach to drug development for rearranged IMA on a phase I trial. This activity is supported this genomic subset. We previously demonstrated the clinical by durable tumor regression observed in a PDX mouse model utility of this approach with NTRK1/2/3–rearranged tumors, and antiproliferative activity noted in MDA-MB-175-VII with where age- and histology-agnostic responses were observed anti-ERBB3 mAb therapy. On a molecular level, these NRG1- with TRK-directed targeted therapy across a wide variety of rearranged models demonstrate functional similarity in their adult and pediatric solid tumors (2). Should NRG1 fusions dependence on ERBB3 by basal pathway activation and near- be similarly amenable to targeted therapy regardless of his- complete suppression of downstream signaling upon ERBB3 tology, tumor-agnostic development should be adopted to inhibition. NRG1 fusions can be added to a growing list of explore the activity of ERBB3/ERBB2 inhibition. therapeutically actionable driver alterations, including ALK, ROS1, RET, NTRK1/2/3, FGFR1/2/3, and PDGFB fusions (1, 19). In contrast, ERBB2-directed targeted therapy was not as METHODS effective in this population. None of the 4 patients with NRG1- Molecular Profiling rearranged tumors in this report responded to afatinib (two of Targeted hybrid capture NGS was performed using the MSK-IMPACT whom developed rapidly progressive disease resulting in death), (Integrated Mutational Profiling of Actionable Cancer Targets; data although significant responses to afatinib were recently reported accessible on www.cbioportal.org), designed to capture base substitu- in other patients with NRG1-rearranged tumors (5, 17, 18). tions, small indels, copy-number alterations, and select rearrangements Our cases temper the previous observations of responses with in more than 340 cancer-related genes as described previously (12, 19). afatinib and underscore the potential for enhanced antitumor Targeted RNA-seq was performed using the MSK Solid Fusion Assay, activity with ERBB3 inhibition. It is noteworthy that our index an NGS-based assay designed to detect fusions which utilizes Anchored case demonstrated a much longer response to GSK2849330 Multiplex PCR technology (Archer FusionPlex, Illumina MiSeq). Unidi- (19 months) relative to all other published cases of afatinib rectional gene-specific primers targeted specific exons in more than 35 genes known to be involved in chromosomal rearrangements. responses (≤12 months; refs. 5, 17, 18). This was consistent with the substantially reduced antitumor activity and subop- Targeted Therapy timal pathway inhibition noted in our NRG1-rearranged PDX GSK2849330 was administered on an NSCLC expansion cohort model treated with afatinib relative to GSK2849330. Although of a phase I trial (NCT01966445) conducted in accordance with the genomic context and heterogeneity may play a role, we specu- International Ethical Guidelines for Biomedical Research Involv- late that these differences could be explained by NRG1- ing Human Subjects after approval by institutional board review. mediated activation of ERBB3 limiting responses to and/or Informed written consent was obtained from subjects. Patients were offering an escape from ERBB2 inhibition (9). eligible if they had ERBB3/NRG1-expressing tumors and advanced Targeted therapy trials should evaluate dedicated cohorts of disease. GSK2849330 was administered at the recommended phase II NRG1-rearranged tumors. Several ERBB3 inhibitors, besides dose (30 mg/kg i.v. weekly for 5 doses, followed by 30 mg/kg every other GSK2849330, have been explored in the clinic, such as patri- week). Imaging was performed at baseline and every 8 weeks. Response

June 2018 CANCER DISCOVERY | OF8

RESEARCH BRIEF Drilon et al. was assessed by RECISTv1.1. An exploratory assessment of volumetric Authors’ Contributions tumor response was performed. Adverse events were captured using Conception and design: A. Drilon, R. Somwar, B.P. Mangatt, CTCAE v4.0. Commercial afatinib was administered at 40 mg daily. M.G. Kris, M.E. Arcila, C. Matheny, G. Ganji Cell Lines and Culture Development of methodology: A. Drilon, J. Sabari, W.W. Lockwood, M.E. Arcila MDA-MB-175-VII, NCI-292, and MCF7 were obtained from the Acquisition of data (provided animals, acquired and managed ATCC in 2015. HCC-95 cells were obtained from the Korean Cell patients, provided facilities, etc.): A. Drilon, R. Somwar, Line Bank in 2016. Cell lines were grown in a humidified incubator P. Desmeules, R.S. Smith, L. Delasos, M. Vojnic, A.J. Plodkowski, (37°C, 5% CO2), authenticated by short tandem repeat profiling, and J. Sabari, J. Montecalvo, J. Chang, W.W. Lockwood, G.J. Riely, N. Rekhtman, tested for Mycoplasma (ATCC Universal Mycoplasma Detection Kit G. Ganji upon receipt and every 6 months, last tested in September 2017). Analysis and interpretation of data (e.g., statistical analysis, bio- MCF7 was cultured in Eagle Minimum Essential Medium (Gibco) statistics, computational analysis): A. Drilon, R. Somwar, H. Edgren, supplemented with 0.01 mg/mL insulin and 10% FBS. All other cell A. Ruusulehto, R.S. Smith, M. Vojnic, A.J. Plodkowski, J. Sabari, lines were cultured in RPMI supplemented with 10% FBS. J. Montecalvo, W.W. Lockwood, V. Martinez, G.J. Riely, C.M. Rudin, M.G. Kris, M.E. Arcila, C. Matheny, R. Benayed, N. Rekhtman, G. Ganji In Vitro and In Vivo Experiments Writing, review, and/or revision of the manuscript: A. Drilon, NRG1 fusion testing (MDA-MB-175-VII, HCC-95) was performed R. Somwar, B.P. Mangatt, H. Edgren, P. Desmeules, A. Ruusulehto, by RT-PCR and targeted RNA-seq (MSK Solid Fusion Assay). RTK R.S. Smith, M. Vojnic, A.J. Plodkowski, J. Sabari, K. Ng, phosphorylation was assessed using a phospho-RTK array (profiles W.W. Lockwood, V. Martinez, G.J. Riely, C.M. Rudin, M.G. Kris, the activation state of 49 RTKs; ref. 15). To generate growth curves M.E. Arcila, C. Matheny, N. Rekhtman, M. Ladanyi, G. Ganji for cells treated with GSK2849330 or control IgG, MDA-MB-175- Administrative, technical, or material support (i.e., reporting or VII (500 cells/well), HCC-95 (200 cells/well), NCI-H292 (200 cells/ organizing data, constructing databases): A. Drilon, P. Desmeules, well), and MCF7 (300 cells/well) were plated in 96-well plates (Nunc M. Vojnic, H. Tai, M.G. Kris, G. Ganji 136102) in triplicate. Cells were treated with 10 μg/mL GSK2849330 Study supervision: A. Drilon, M.G. Kris, M. Ladanyi, G. Ganji or control IgG. Cell proliferation was measured (CellTiter-Glo Lumi- nescent Cell Viability Assay, Promega). For shRNA infection, 250,000 cells were plated (6-well plates) and infected (24 hours later) with Acknowledgments viral supernatant. Infected cells were selected with 5 μg/mL puro- We would like to thank all our patients, their families, and the site mycin. For caspase-3/7 activation, 25,000 puromycin-selected cells staff for their participation and contributions to this study. We also were plated (white clear bottom 96-well plates) and caspase-3/7 acknowledge BioWa, Inc. (U.S. subsidiary of Kyowa Hakko Kirin Co., enzymatic activity measured (APO-ONE homogenous caspase-3/7 Ltd.) for their POTELLIGENT Technology and COMPLEGENT Tech- Activity Assay Kit, Promega). 0V-10-0050 was treated (vehicle, IgG nology in developing GSK2849330, an ADCC- and CDC-enhanced control at 25 mg/kg, or GSK2849330 at 25 mg/kg biweekly) on day anti-ERBB3 mAb. Baseline characterization of the OV-10-0050 PDX 35 postimplantation at an average tumor size of approximately 163 model was carried out at WuXi AppTec Co., Ltd. (Shanghai, China). mm3. Tumor size was measured twice weekly in two dimensions and A. Drilon, G.J. Riely, C.M. Rudin, and M.G. Kris are supported by the used for calculations of both T-C and T/C values. NIH by NCI Cancer Support Grant P30 CA008748, which did not directly fund study costs. The NCT01966445 study was sponsored TCGA Sample Analysis by GlaxoSmithKline. RNA-seq data from a total of 8,984 cancer samples (dbGaP Study Received September 7, 2017; revised March 7, 2018; accepted Accession: phs000178.v9.p8) from 33 different cancer types in TCGA March 28, 2018; published first April 2, 2018. were analyzed using the MediSapiens FusionSCOUT fusion gene detection pipeline. Candidate fusion genes were filtered on the basis of: gene–gene distance, sequence homology, either partner annotated as a pseudogene, and the presence of the fusion in RNA-seq data References from ∼600 normal tissue samples. Fusion junctions were identified by . 1 Schram AM, Chang MT, Jonsson P, Drilon A. Fusions in solid tumours: constructing all possible exon–exon combinations between each pair diagnostic strategies, targeted therapy, and acquired resistance. Nat of genes, followed by alignment of unmapped reads against synthetic Rev Clin Oncol 2017;14:735–48. junctions. Reading frame status was predicted on the basis of the end 2. Drilon A, Laetsch TW, Kummar S, Dubois F, Lassen UN, Demetri and start phases of junction exons. NRG1 expression by cancer type GD, et al. Efficacy of larotrectinib in TRK fusion-positive cancers in was drawn using the RSEM scaled estimate expression values (Broad adults and children. N Engl J Med 2018;378:731–9. Institute GDAC Firehose pipeline). 3. Fernandez-Cuesta L, Plenker D, Osada H, Sun R, Menon R, Leenders F, et al. CD74-NRG1 fusions in lung adenocarcinoma. Cancer Discov Additional details regarding cell lines, molecular profiling, shRNA 2014;4:415–22. infection, caspase-3/7 activity, western blotting, PDX generation and 4. Shin DH, Lee D, Hong DW, Hong SH, Hwang JA, Lee BI, et al. treatment, fusion gene analysis for TCGA samples, and expression/ Oncogenic function and clinical implications of SLC3A2-NRG1 copy-number analysis of NSCLC lines are provided in the Supple- fusion in invasive mucinous adenocarcinoma of the lung. Oncotarget mentary Methods. 2016;7:69450–65. 5. Jones MR, Lim H, Shen Y, Pleasance E, Ch’ng C, Reisle C, et al. Suc- Disclosure of Potential Conflicts of Interest cessful targeting of the NRG1 pathway indicates novel treatment strategy for metastatic cancer. Ann Oncol 2017;28:3092–7. A. Ruusulehto has ownership interest (including patents) in 6. Nakaoku T, Tsuta K, Ichikawa H, Shiraishi K, Sakamoto H, Enari M, MediSapiens Ltd. G.J. Riely reports receiving commercial research et al. Druggable oncogene fusions in invasive mucinous lung adeno- support from Ariad/Takeda, Novartis, Pfizer, and Roche/Genen- carcinoma. Clin Cancer Res 2014;20:3087–93. tech. M.G. Kris is a consultant/advisory board member for Astra- 7. Shim HS, Kenudson M, Zheng Z, Liebers M, Cha YJ, Hoang Ho Q, Zeneca. No potential conflicts of interest were disclosed by the et al. Unique genetic and survival characteristics of invasive mucinous other authors. adenocarcinoma of the lung. J Thorac Oncol 2015;10:1156–62.

OF9 | CANCER DISCOVERY June 2018 www.aacrjournals.org

ERBB3-Directed Targeted Therapy in NRG1-Rearranged Cancers RESEARCH BRIEF

8. Trombetta D, Graziano P, Scarpa A, Sparaneo A, Rossi G, Rossi A, in antitumor response to combination therapy. Clin Cancer Res et al. Frequent NRG1 fusions in Caucasian pulmonary mucinous 2014;20:456–68. adenocarcinoma predicted by Phospho-ErbB3 expression. Oncotarget 17. Cheema PK, Doherty M, Tsao MS. A case of invasive mucinous pul- 2018;9:9661–71. monary adenocarcinoma with a CD74-NRG1 fusion protein targeted 9. Fernandez-Cuesta L, Thomas RK. Molecular pathways: targeting with afatinib. J Thorac Oncol 2017;12:e200–e2. NRG1 fusions in lung cancer. Clin Cancer Res 2015;21:1989–94. 18. Gay ND, Wang Y, Beadling C, Warrick A, Neff T, Corless CL, et al. 10. Schaefer G, Fitzpatrick VD, Sliwkowski MX. Gamma-heregulin: a Durable response to afatinib in lung adenocarcinoma harboring novel heregulin isoform that is an autocrine growth factor for the NRG1 gene fusions. J Thorac Oncol 2017;12:e107–e10. human breast cancer cell line, MDA-MB-175. Oncogene 1997;15: 19. Zehir A, Benayed R, Shah RH, Syed A, Middha S, Kim HR, et al. 1385–94. Mutational landscape of metastatic cancer revealed from prospective 11. Murayama T, Nakaoku T, Enari M, Nishimura T, Tominaga K, clinical sequencing of 10,000 patients. Nat Med 2017;23:703–13. Nakata A, et al. Oncogenic fusion gene CD74-NRG1 confers cancer 20. Le Clorennec C, Bazin H, Dubreuil O, Larbouret C, Ogier C, Lazrek stem cell-like properties in lung cancer through a IGF2 autocrine/ Y, et al. Neuregulin 1 allosterically enhances the antitumor effects paracrine circuit. Cancer Res 2016;76:974–83. of the noncompeting anti-HER3 antibody 9F7-F11 by increasing its 12. Cheng DT, Mitchell TN, Zehir A, Shah RH, Benayed R, Syed A, et al. binding to HER3. Mol Cancer Ther 2017;16:1312–23. Memorial sloan kettering-integrated mutation profiling of action- 21. Huang J, Wang S, Lyu H, Cai B, Yang X, Wang J, et al. The anti-erbB3 able cancer targets (MSK-IMPACT): a hybridization capture-based antibody MM-121/SAR256212 in combination with trastuzumab next-generation sequencing clinical assay for solid tumor molecular exerts potent antitumor activity against trastuzumab-resistant breast oncology. J Mol Diagn 2015;17:251–64. cancer cells. Mol Cancer 2013;12:134. 13. Clarke N, Hopson C, Hahn A, Sully K, Germaschewski F, Yates J, 22. Meulendijks D, Jacob W, Martinez-Garcia M, Taus A, Lolkema MP, et al. Preclinical pharmacologic characterization of GSK2849330, a Voest EE, et al. First-in-human phase I study of lumretuzumab, monoclonal AccretaMab® antibody with optimized ADCC and CDC a glycoengineered humanized anti-HER3 monoclonal antibody, in activity directed against HER3. Eur J Cancer 2014;50:98–9. patients with metastatic or advanced HER3-positive solid tumors. 14. Adelaide J, Huang HE, Murati A, Alsop AE, Orsetti B, Mozziconacci Clin Cancer Res 2016;22:877–85. MJ, et al. A recurrent chromosome translocation breakpoint in breast 23. Meetze K, Vincent S, Tyler S, Mazsa EK, Delpero AR, Bottega S, and pancreatic cancer cell lines targets the neuregulin/NRG1 gene. et al. Neuregulin 1 expression is a predictive biomarker for response Genes Chromosomes Cancer 2003;37:333–45. to AV-203, an ERBB3 inhibitory antibody, in human tumor models. 15. Roudabush FL, Pierce KL, Maudsley S, Khan KD, Luttrell LM. Trans- Clin Cancer Res 2015;21:1106–14. activation of the EGF receptor mediates IGF-1-stimulated shc phos- 24. Garner AP, Bialucha CU, Sprague ER, Garrett JT, Sheng Q, Li S, et al. An phorylation and ERK1/2 activation in COS-7 cells. J Biol Chem antibody that locks HER3 in the inactive conformation inhibits tumor 2000;275:22583–9. growth driven by HER2 or neuregulin. Cancer Res 2013;73:6024–35. 16. Phillips GD, Fields CT, Li G, Dowbenko D, Schaefer G, Miller K, 25. Duruisseaux M, McLeer-Florin A, Antoine M, Alavizadeh S, Poulot V, et al. Dual targeting of HER2-positive cancer with trastuzumab Lacave R, et al. NRG1 fusion in a French cohort of invasive mucinous emtansine and pertuzumab: critical role for neuregulin blockade lung adenocarcinoma. Cancer Med 2016;5:3579–85.

June 2018 CANCER DISCOVERY | OF10

Response to ERBB3-Directed Targeted Therapy in NRG1 -Rearranged Cancers

Alexander Drilon, Romel Somwar, Biju P. Mangatt, et al.

Cancer Discov Published OnlineFirst April 2, 2018.

Updated version Access the most recent version of this article at: doi:10.1158/2159-8290.CD-17-1004

Supplementary Access the most recent supplemental material at: Material http://cancerdiscovery.aacrjournals.org/content/suppl/2018/03/31/2159-8290.CD-17-1004.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://cancerdiscovery.aacrjournals.org/content/early/2018/05/11/2159-8290.CD-17-1004. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.