Antitumor Activity of Amivantamab (JNJ-61186372), an EGFR–MET Bispecific Antibody, in Diverse Models of EGFR Exon 20 Insertion–Driven NSCLC

Published OnlineFirst May 15, 2020; DOI: 10.1158/2159-8290.CD-20-0116

Research Article

Jiyeon Yun1, Soo-Hwan Lee2, Seok-Young Kim1, Seo-Yoon Jeong1, Jae-Hwan Kim1, Kyoung-Ho Pyo1, Chae-Won Park1, Seong Gu Heo1, Mi Ran Yun2, Sangbin Lim1, Sun Min Lim3, Min Hee Hong3, Hye Ryun Kim3, Meena Thayu4, Joshua C. Curtin4, Roland E. Knoblauch4, Matthew V. Lorenzi4, Amy Roshak4, and Byoung Chul Cho3

aBstRact EGFR exon 20 insertion driver mutations (Exon20ins) in non–small cell lung cancer (NSCLC) are insensitive to EGFR tyrosine kinase inhibitors (TKI). Amivantamab (JNJ-61186372), a bispecifi c antibody targeting EGFR–MET, has shown preclinical activity in TKI-sensitive EGFR -mutated NSCLC models and in an ongoing fi rst-in-human study in patients with advanced NSCLC. However, the activity of amivantamab in Exon20ins-driven tumors has not yet been described. Ba/F3 cells and patient-derived cells/organoids/xenograft models harboring diverse Exon20ins were used to characterize the antitumor mechanism of amivantamab. Amivantamab inhibited proliferation by effectively downmodulating EGFR–MET levels and inducing immune-directed antitumor activity with increased IFNγ secretion in various models. Importantly, in vivo effi cacy of amivantamab was superior to cetuximab or poziotinib, an experimental Exon20ins-targeted TKI. Amivantamab produced robust tumor responses in two Exon20ins patients, highlighting the important translational nature of this preclinical work. These fi ndings provide mechanistic insight into the activity of amivantamab and support its continued clinical development in Exon20ins patients, an area of high unmet medical need.

SIGNIFICANCE: Currently, there are no approved targeted therapies for EGFR Exon20ins–driven NSCLC. Preclinical data shown here, together with promising clinical activity in an ongoing phase I study, strongly support further clinical investigation of amivantamab in EGFR Exon20ins–driven NSCLC.

intRoDuction mutations encompass nucleotides that translate into amino acids at position 762–823, and include a C-helix (762–766) Molecular segmentation of advanced non–small cell lung followed by a loop (position 767–775; ref. 7 ). The insertion cancer (NSCLC) based on oncogenic driver mutations has mutations of one to seven amino acids in exon 20 form a improved the overall survival and quality of life for patients wedge at the end of the C-helix in EGFR that promotes active with actionable driver mutations, and solidifi ed solid-tumor kinase conformation. EGFR exon 20 insertion driver muta- targeted therapy. Mutations in the EGFR ( 1, 2 ) gene constitu- tions (Exon20ins), a distinct and highly heterogeneous sub- tively activate downstream growth and survival signaling path- set of NSCLCs, represent 4% to 12% of all EGFR mutations ways leading to dependency on the EGFR pathway for tumor ( 7, 8 ). These Exon20ins mutations are generally insensitive to growth. Nearly 20% of Caucasians and up to 50% of Asians approved EGFR TKIs and are associated with poor prognosis, with lung adenocarcinomas harbor mutations in EGFR ( 3, 4 ). thus representing an area of high unmet medical needs ( 6, 9 ). EGFR activating mutations have been reported in the fi rst Recently, poziotinib and TAK-788 have been undergoing four exons (18 through 21) of its tyrosine kinase domain. clinical evaluation in patients whose tumors carry EGFR NSCLCs that harbor “classic” EGFR mutations in exons 18, Exon20ins mutations (10, 11). Despite initial promising effi - 19, and 21, for example, exon 19 deletions or L858R, are sen- cacy, the Zenith 20 trial demonstrated that poziotinib had sitive to treatment with fi rst-, second-, and third-generation a low response rate (RR; ∼14%) in patients with NSCLC EGFR tyrosine kinase inhibitors (TKI) such as erlotinib, with the EGFR Exon20ins mutation. Furthermore, both pozi- afatinib, and osimertinib (5, 6). In contrast, the EGFR exon 20 otinib and TAK-788 have high rates of EGFR wild-type (WT)– driven toxicity due to the lack of selectivity upon Exon20ins as compared with EGFR WT and other kinases, limiting their

1Severance Biomedical Science Institute, Brain Korea 21 PLUS Project for clinical utility (6 ). Medical Sciences, Yonsei University College of Medicine, Seoul, Republic Amivantamab (JNJ-61186372; Fig. 1A ) is an EGFR–MET of South Korea. 2 JE-UK Institute for Cancer Research, JEUK Co. Ltd., bispecifi c antibody with immune cell–directing activity that Gumi-City, Kyungbuk, Republic of South Korea. 3Division of Medical Oncol- targets activating and resistant EGFR mutations and MET ogy, Yonsei Cancer Center, Yonsei University College of Medicine, Seoul, mutations and amplifi cations. Ongoing fi rst-in-human Republic of South Korea. 4 Janssen Research and Development, Spring House, Pennsylvania. studies in patients with advanced, refractory EGFR-mutant Note: Supplementary data for this article are available at Cancer Discovery NSCLC have demonstrated preliminary clinical activity of Online (http://cancerdiscovery.aacrjournals.org/). amivantamab in patients with diverse EGFR mutations (12, Corresponding Author: Byoung Chul Cho, Yonsei University College of 13 ). Of note, amivantamab showed promising effi cacy (30% Medicine, 51 Yonseiro, Seodaemun-gu, Seoul 120-752, Republic of South RR) with a manageable safety profi le in patients with heavily Korea. Phone: 822-2228-8126; Fax: 822-393-3652; E-mail: cbc1971@ pretreated EGFR Exon20ins NSCLC. Although amivantamab yuhs.ac has been reported to harbor activity in preclinical tumor Cancer Discov 2020;10:1194–209 models driven by EGFR mutations sensitive to approved TKIs doi: 10.1158/2159-8290.CD-20-0116 (e.g., L858R and Exon 19 deletions; ref. 14 ), its activity has not ©2020 American Association for Cancer Research. yet been explored in the context of EGFR Exon20ins.

AUGUST 2020 CANCER DISCOVERY | 1195

RESEARCH ARTICLE Yun et al.

Here, we comprehensively evaluated the antitumor activity poziotinib in Ba/F3 cells overexpressing EGFR Exon20ins and mechanisms of action (MOA) of amivantamab in multi- (Supplementary Table S1). Consistent with a previous report ple EGFR Exon20ins models, including engineered cell lines, (18), poziotinib strongly inhibited the cell viability in the patient-derived cells (PDC), and patient-derived xenografts mutant EGFR Exon20ins cells (IC50 ranging from 0.8 to 10.9 (PDX). We also present evidence of clinical activity in two nmol/L). As reported in a previous study (20), poziotinib also case studies of patients with EGFR Exon20ins NSCLC treated potently suppressed proliferation of Ba/F3 cells harboring with amivantamab from an ongoing phase I clinical trial, WT EGFR (IC50 = 0.8 nmol/L). To present the selectivity for highlighting the important translational nature of this work. Exon20ins mutation in a more balanced manner, we compared antiproliferative potency between amivantamab and Results poziotinib in EGFR Exon20ins mutants over WT EGFR. Pozi- otinib exhibited lower EGFR Exon20ins–mutant selectivity Amivantamab Inhibits Proliferation of Ba/F3 Cells over WT EGFR compared with amivantamab, suggesting that Harboring Diverse EGFR Exon20ins Mutations poziotinib may adversely affect normal tissues, thereby pro- To demonstrate the antitumor activity of amivantamab ducing substantial toxicities such skin rash and diarrhea (21). in the context of Exon20ins, multiple Exon20ins were sta- To better understand the mechanisms involved in amivan- bly expressed in Ba/F3 cells. Five distinct Exon20ins were tamab-mediated cellular cytotoxicity, we assessed the effect introduced (Fig. 1B), all of which have been observed in of amivantamab treatment on cell-cycle progression and patients with NSCLC (V769_D770insASV, D770delinsGY, programmed cell death. In Ba/F3 cells expressing the EGFR H773_V774insH, Y764_V765insHH, and D770_N771ins- D770delinsGY and H773_V774insH Exon20ins mutations,

SVD; refs. 15, 16). In Ba/F3 cells treated with amivantamab an accumulation of cells in G1 phase was observed in ami- ranging from 0.05 to 1 mg/mL, a significant and dose- vantamab-treated cells compared with vehicle-treated cells dependent decrease in Ba/F3 cell viability (P < 0.0001) was (Fig. 1F). As EGFR TKIs have been reported to drive apoptosis observed in all fiveEGFR Exon20ins mutations (Fig. 1C). In in NSCLC cells harboring sensitizing EGFR mutations (22, contrast, treatment with the first- and third-generation irre- 23), we investigated whether treatment with amivantamab versible EGFR TKIs gefitinib and osimertinib, respectively, resulted in engagement of the apoptotic machinery. Amivan- showed limited antiproliferative activity compared with ami- tamab treatment resulted in the induction of proapoptotic vantamab (Fig. 1C), confirming the well-known resistance proteins, including BIM and cleaved caspase-3 (Fig. 1G), sug- of Exon20ins to EGFR TKIs. No effect on cell viability was gesting that amivantamab, in addition to inhibiting down- observed when IgG1 control antibodies were used in the same stream EGFR signaling cascade, also induced apoptosis in a Ba/F3 cell lines (Supplementary Fig. S1A). In tumor mod- BIM- and caspase-dependent manner. els driven by TKI-sensitive EGFR mutations such as L858R or Exon 19 deletions, amivantamab has several proposed Amivantamab Displays Antitumor Activity MOAs including blocking ligand binding, receptor down- in PDCs and Organoids modulation, downstream signaling inhibition, and trigger- To extend our findings from Ba/F3 cells engineered to ing immune-directed antitumor activity (17). To determine express the exogenous EGFR Exon20ins mutations, we evalu- whether these MOAs are also observed in the context of ated the activity of amivantamab in several PDCs harboring Exon20ins and contribute to the observed antiproliferative the Exon20ins. The antitumor activity of amivantamab and activity, immunoblot analysis was performed (Fig. 1C) in associated mechanistic endpoints were evaluated in PDCs Ba/F3 cells overexpressing the EGFR D770delinsGY and generated from patients harboring P772ins_H773insPNP H773_V774insH Exon20ins mutations. The total EGFR lev- (DFCI-127), H773_V774insNPH (DFCI-58), and S768_ els were reduced following treatment with amivantamab, D770dup (YU-1163) Exon20ins mutations (Supplementary compared with those of untreated cells (Fig. 1D; Supplemen- Fig. S2A–S2C; Supplementary Table S2). In both DFCI- tary Fig. S1B) or cells treated with the IgG1 control antibody 127 and DFCI-58 cells, amivantamab treatment resulted (Supplementary Fig. S1C). Consistent with the reduction in decreased expression of total EGFR and MET levels, as in EGFR expression levels, the EGFR downstream signaling well as inhibition of pEGFR, pMET, pAKT, pERK, and pS6 pathways phospho-EGFR (pEGFR), phospho-AKT (pAKT), (Fig. 2A), consistent with the results observed in Ba/F3 cell phosho-ERK (pERK), and phospho-S6 (pS6) were also signifi- lines harboring EGFR Exon20ins mutations. Analysis of cell cantly reduced following amivantamab treatment (Fig. 1D), viability and colony formation revealed that amivantamab suggesting that amivantamab targeted EGFR and inhibited dose-dependently inhibited the cell growth and proliferation EGFR-related downstream signaling cascades. Similar results of PDCs, compared with IgG1 controls (Fig. 2B and C). In were observed in Ba/F3 cells expressing the V769insASV, contrast to the significant reduction in EGFR, MET, pEGFR, Y764 insHH, and D770_N771insSVD Exon20ins mutations pMET, pAKT, and pS6 in DFCI-127 and DFCI-58 cells, (Supplementary Fig. S1B). Although 100 nmol/L of gefitinib YU-1163 treated with amivantamab unexpectedly revealed and osimertinib reduced pEGFR in Ba/F3 cells overexpress- an induction of pERK (Fig. 2A). Consistent with this result, ing D770delinsGY and H773_V774insH, downstream EGFR the growth of YU-1163 was not inhibited after amivantamab signaling pathway components were not significantly inhib- treatment for 72 hours or following long-term treatment ited, which correlated with the lack of TKI effects on cell (Fig. 2B and C). From the whole-exome sequencing data of viability (Fig. 1E; Supplementary Fig. S1D). In recent studies, YU-1163, we observed a co-occurring mutation in the TP53 poziotinib has shown antitumor activity in EGFR Exon20ins gene (R280T; 96% of mutant allele frequency; Supplementary NSCLC (18, 19). We further assessed the cell viability test for Fig. S2C and S2D). According to recent studies, mutations

1196 | CANCER DISCOVERY AUGUST 2020 AACRJournals.org

Amivantamab in NSCLC with EGFR Exon20ins RESEARCH ARTICLE

A B Extracellular Transmembrane domain domain Tyrosine kinase domain

MET EGFR Amivantamab I/II III/IV Ex18 Ex19 Ex20 Ex21 Ex22 Ex23

Stable chinge + P772insPNP PDC Controlled S768dup Fab-arm exchange PDO Matched CH3 domains S768dup P773insNPH A767dup D770insG PDX K409R F405L K409R F405L EGFR Ex20 E A Y V M A S V D N P H V C (762–823) 762 763 764 765 766 767 768 769 770 771 772 773 774 775

Y764insHH D770insSVD H773insH D770delinsGY Ba/F3 V769insASV

C V769insASV D770delinsGY H773ins HY764insHH D770insSVD 130 130 130 130 130 120 120 120 120 120 110 110 110 110 110 100 ** 100 100 100 100 90 90 90 90 * 90 * * ** * * ** * * * 80 * 80 * * 80 80 80 70 70 * 70 * * 70 70 60 60 * 60 * 60 60 50 50 50 50 50 40 40 40 40 40 Cell viability (%) 30 30 30 30 30 20 20 20 20 20 10 10 10 10 10 0 0 0 0 0 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 10 10 10 10 10 10 10 10 10 10 0.1 0.5 0.1 0.5 0.1 0.5 0.1 0.5 0.1 0.5 100 100 100 100 100 100 100 100 100 100 0.05 0.05 0.05 0.05 0.05 Amivantamab Gefitinib Osimertinib Amivantamab Gefitinib Osimertinib Amivantamab Gefitinib Osimertinib Amivantamab Gefitinib Osimertinib Amivantamab Gefitinib Osimertinib (mg/mL) (nmol/L) (nmol/L) (mg/mL) (nmol/L) (nmol/L) (mg/mL) (nmol/L) (nmol/L) (mg/mL) (nmol/L) (nmol/L) (mg/mL) (nmol/L) (nmol/L)

D D770 H773_V774 E D770delinsGY H773_V774insH delinsGY insH Amivantamab Osimertinib (nmol/L) Gefitinib (nmol/L) Osimertinib (nmol/L) Gefitinib (nmol/L) 0 0.1 100.1 1 (mg/mL) pEGFR 010 100 10 100 0 10 100 10 100 pEGFR pEGFR EGFR EGFR EGFR pAKT pAKT pAKT AKT AKT AKT pERK pERK pERK ERK ERK ERK pS6 pS6 pS6

S6 S6 S6

GAPDH GAPDH GAPDH

FGD770 H773_V774 D770 H773_V774 delinsGY insH Amivantamab delinsGY insH 0 0.1 100.1 1 (mg/mL) p21 60 * * 80 Vehicle 50 * Amivantamab 0.1 mg/mL p27 60 * 40 Amivantamab 1 mg/mL p53

30 40 BIMEL

20 BIML 20 10 BIM Cell population (%) Cell population (%) s 0 0 Cleaved caspase-3 Sub G G S G – Sub G G S G – 1 1 2 1 1 2 GAPDH phase phase phase M phase phase phase phase M phase

Figure 1. Amivantamab shows antitumor activity and suppresses EGFR and MET signaling pathways in Ba/F3 cells with EGFR Exon20ins mutations. A, Schematic of the structure of amivantamab, an EGFR and MET bispecific antibody. B, Schematic of EGFR Exon20 insertions in stable Ba/F3 cells, PDC, PDO, and PDX models. C, The viability of Ba/F3 cells was determined via CellTiter-Glo. Amivantamab, gefitinib, or osimertinib were treated for 72 hours. Data are presented as averages ± SD of triplicate independent experiments. *, P < 0.0001; **, P < 0.001; Student t test. D, Ba/F3 cells overexpressing the indicated EGFR Exon20ins mutations were treated with amivantamab for 72 hours at the indicated concentrations. E, Ba/F3 cells overexpressing the indicated EGFR Exon20ins mutations were treated with osimertinib or gefitinib for 6 hours at the indicated concentrations. Immu- noblot analysis was performed for EGFR, MET, AKT, ERK, and S6 expression after amivantamab treatment. Amivantamab inhibited the cell cycle and induced synergistic apoptosis in Ba/F3 cells overexpressing the EGFR Exon20ins mutations. F, Amivantamab induced G1-phase arrest after amivantamab treatment for 72 hours in Ba/F3 cells overexpressing the indicated types of EGFR Exon20ins mutations. Cell cycles were analyzed using propidium iodide staining and FACS analysis. Data are presented as averages ± SD of triplicate independent experiments. *, P < 0.0001; Student t test. G, BIM- and caspase- dependent apoptosis were induced during amivantamab treatment. The expression of BIM and cleaved caspase-3 were detected by Western blotting.

AUGUST 2020 CANCER DISCOVERY | 1197

RESEARCH ARTICLE Yun et al.

A DFCI-127 DFCI-58 YU-1163 (P772_H773insPNP) (H773_V774insNPH) (S768_D770dup) Amivantamab 0 0.1 100.1 100.1 1 (mg/mL) pEGFR pEGFR pEGFR

EGFR EGFR EGFR

pMET pMET pMET

MET MET MET

pAKT pAKT pAKT

AKT AKT AKT

pERK pERK pERK

ERK ERK ERK

pS6 pS6 pS6

S6 S6 S6

GAPDH GAPDH GAPDH

B 120 140 120 100 120 100 100 80 * * * 80 ** 80 60 ** * * 60 60 40 lgG1 (mg/mL) 40 lgG1 (mg/mL) 40 lgG1 (mg/mL) 20 Amivantamab (mg/mL) 20 Amivantamab (mg/mL) 20 Amivantamab (mg/mL) Cell viability (%) Cell viability (%) Cell viability (%) 0 0 0 0 0.05 0.1 0.5 100.05 0.1 0.5 100.05 0.1 0.5 1

C Amivantamab DFCI-127 DFCI-58 YU-1163 0 0.1 11120 0 0.1 120 0 0.1 1 120 (mg/mL) 100 * 100 100 80 * 80 * 80 D (%) 60 D (%) 60 D (%) 60

O. 40 O. 40 * O. 40 20 20 20 0 0 0 100.1 100.1 100.1

DE YUO-036 YUO-029 (A767_V769dup) (S768_D770dup) 120 120

100 0 100 NS 0

80 * 80 * 60 0.1 60 0.1 * * 40 40 Cell viability (%) Cell viability (%) 20 lgG1 control (mg/mL) 1 20 lgG1 control (mg/mL) 1 Amivantamab (mg/mL) Amivantamab (mg/mL) 0 0 0 0.05 0.1 0.5 1 Amivantamab 0 0.05 0.1 0.5 1 Amivantamab (mg/mL) (mg/mL)

Figure 2. Amivantamab has antitumoral activity and suppresses EGFR and MET signaling pathways in PDCs and PDOs harboring EGFR Exon20ins mutations. A, PDCs with the indicated EGFR Exon20ins mutations were treated with amivantamab for 72 hours at the indicated concentrations. Immu- noblot analysis was performed for EGFR, MET, AKT, ERK, and S6 expression after amivantamab treatment. B, The viability of PDCs was determined via CellTiter-Glo. Amivantamab was treated for 72 hours. Data are presented as averages ± SD of triplicate independent experiments. *, P < 0.0001; **, P < 0.001; Student t test. C, Effects of amivantamab on the colony formation and cell proliferation of PDCs. Representative images and quantitative analysis of the colony formation assay. Data are presented as averages ± SD of triplicate independent experiments. *, P < 0.0001; Student t test. O.D., optical density. D and E, Dose–response curves of (D) YUO-036 (A767_V769dup) and (E) YUO-029 (S768_D770dup) PDOs treated with IgG1 control or amivantamab. Cell viability was measured using CellTiter-Glo 3D cell viability reagent 72 hours after drug treatment. Representative images of PDOs treated with amivantamab for 72 hours at the indicated concentrations. Data are presented as averages ± SD of triplicate independent experiments. *, P < 0.0001; Student t test; NS, not significant.

1198 | CANCER DISCOVERY AUGUST 2020 AACRJournals.org

Amivantamab in NSCLC with EGFR Exon20ins RESEARCH ARTICLE in TP53 commonly occurred with EGFR mutations in NSCLC. that amivantamab reduced EGFR and MET on PDCs com- Particularly, TP53 mutations in exon 8 in patients with pared with IgG1 control. Immunofluorescence (IF) staining NSCLC with EGFR mutations show lower responsiveness was used to visualize the internalization of EGFR and MET to EGFR TKIs and worse prognosis than patients with WT following amivantamab treatment. Treatment with 0.1 mg/ TP53 (24, 25). Indeed, accumulated studies have revealed mL amivantamab for 72 hours led to the redistribution that the R280T mutation in TP53 plays crucial roles in the of EGFR and MET receptors into internal compartments, proliferation and survival of cancer cells, and knockdown whereas IgG-treated cells showed no change in the staining of the mutant TP53 causes G2-phase arrest and apoptosis pattern for EGFR or MET (Fig. 3D; Supplementary Fig. S3). in bladder cancer cells (26, 27). As shown in Supplementary Internalization and subsequent downregulation of EGFR Fig. S2E, depletion of mutant TP53 by three different TP53- and MET receptors by lysosomes could account for the directed siRNAs significantly inhibited cell proliferation with decreased EGFR and MET protein levels observed in the a reduction in activated ERK in YU-1163 pretreated with 1 immunoblot, FACS, and IF assays following amivantamab mg/mL amivantamab. Given that mutant TP53 is associated treatment. To determine whether lysosomal degradation was with EGFR-TKI resistance (28) and the depleted mutant TP53 involved in downregulating EGFR protein levels, Ba/F3 cells restored the sensitivity of amivantamab by downregulation of overexpressing D770delinsGY and H773_V774insH were pERK, induction of pERK following amivantamab treatment treated with amivantamab in the absence and presence of in YU-1163 cells might be a key regulator of cell survival, the autophagy inhibitor bafilomycin. Bafilomycin treatment potentially through the cross-talk between mutant TP53 and inhibited the degradation of EGFR (Fig. 3E), suggesting that ERK signaling cascade (29, 30). In addition, we generated two downmodulation of the total EGFR protein level following patient-derived organoid (PDO) models from plural effu- amivantamab treatment may involve lysosomal degradation sion of patients who had A767_V769dup (YUO-036) and of internalized cell-surface receptors. Taken together, these S768_D770dup (YUO-029) to recapitulate the phenotypic results suggest that treatment with amivantamab induces and molecular landscape of the original NSCLC with EGFR receptor internalization and may contribute to the observed Exon20ins (Supplementary Fig. S2F and S2G; Supplementary antiproliferative effects of amivantamab by inhibiting EGFR- Table S2). YUO-029 was derived from the same patient from and MET-mediated signaling. whom YU-1163 PDC (S768_D770dup) was derived. As shown in Fig. 2D, YUO-036 was sensitive to amivantamab in a dose- Amivantamab Inhibits EGFR Exon20ins Mutation– dependent manner, whereas YUO-029 derived from the same Driven Growth of Ba/F3 and PDC Models In Vivo patient with YU-1163 showed no significant decrease in cell To determine whether amivantamab is active against EGFR viability following amivantamab treatment compared with Exon20ins–derived tumors in vivo, xenograft models were gen- IgG1 control (Fig. 2E). Taken together, these results indicate erated using Ba/F3 cells overexpressing EGFR D770delinsGY that amivantamab has potent antitumor activity in NSCLC and H773_V774insH Exon20ins mutations and PDCs (DFCI- patient-derived cancer cells with EGFR Exon20ins mutations 127 and YU-1163) harboring P772insPNP and S768_D770dup by downmodulation of EGFR and MET signaling pathways. EGFR Exon20ins mutations, respectively. Mice were treated with amivantamab, IgG1 control, or vehicle at 30 mg/kg EGFR and MET Are Internalized in twice per week intraperitoneally. Amivantamab-treated mice Response to Amivantamab showed reduced tumor volumes compared with vehicle- or Treatment with amivantamab results in downmodulation IgG1 control–treated mice in the Ba/F3 cells bearing NOD.Cg- of EGFR and MET, as observed in Ba/F3 cells (Fig. 1) and Prkdcscid II2rgtm1Sug/Jic (NOG) mice models (Fig. 4A and B; PDCs (Fig. 2). According to many studies, anti-EGFR mAb Supplementary Fig. S4A and S4B). Inhibition of tumor growth induces internalization of EGFR, leading to downregula- occurred early and was sustained 15 days following treatment. tion of its expression on the cell surface (31, 32). To investi- As shown in Ba/F3 cells and PDCs in vitro (Figs. 1D and 2A), gate whether amivantamab directly binds to EGFR on cells protein expression of EGFR, MET, pEGFR, and pMET were with EGFR Exon20ins mutation, Ba/F3 cells overexpress- significantly reduced following amivantamab treatment in the ing D770delinsGY or H773_V774insH were incubated with Ba/F3-bearing NOG mice models (Fig. 4C; Supplementary 0.1 mg/mL IgG1 control and 0.1 mg/mL amivantamab. FACS Fig. S4C). Similarly, in the PDC xenograft models, amivantamab- was used to measure the level of plasma membrane–bound treated mice showed a reduction in tumor volume compared EGFR. EGFR expression on the plasma membrane began with vehicle-treated mice (Fig. 4D–G), as well as a reduction to dwindle by almost 2-fold 30 minutes after amivantamab in EGFR, MET, pEGFR, and pMET protein levels (Fig. 4H treatment. The percentage changes in median fluorescence and I). Intriguingly, although amivantamab could not inhibit intensity (MFI) of EGFR relative to IgG1 control–treated the proliferation of YU-1163 PDC in vitro (Fig. 2), a dramatic cells at 30 minutes were 56% and 68% in D770delinsGY and tumor regression was observed in YU-1163–bearing BALB/c H773_V774insH, respectively, and subsequently remained at nude mice after amivantamab treatment (Fig. 4F), suggesting 40% EGFR expression relative to IgG1 control–treated cells that additional factors might contribute to the in vivo antitu- 72 hours after amivantamab treatment (Fig. 3A). To explore mor effect of amivantamab. As mentioned above, poziotinib is the internalization of MET as well as EGFR on PDCs har- a targeted agent that has shown preliminary clinical activity in boring EGFR Exon20ins, DFCI-127 and DFCI-58 PDCs were EGFR Exon20ins disease (18, 19). We compared the antitumor treated with 0.1 mg/mL amivantamab and the plasma mem- activity and safety of poziotinib with those of amivantamab brane–bound MET and EGFR were measured 72 hours after in YU-1163 (S768_D770dup)–bearing BALB/c nude mice amivantamab treatment (Fig. 3B and C). The results showed and Ba/F3 cells overexpressing D770_N771insSVD-bearing

AUGUST 2020 CANCER DISCOVERY | 1199

RESEARCH ARTICLE Yun et al.

A B C D770 H773_V774 DFCI-127 DFCI-58 delinsGY insH (P772_H773insPNP) (H773_V774insNPH)

300 300 1.02 93.5 6.69 35.5

4 4 200 200 10 10 count 30 min 100 100 103 103 0 0 0103 104 105 0103 104 105 IgG1 102 102

300 300 (0.1 mg/mL) FITC-MET FITC-MET 0 0 200 200 1.89 3.63 27.4 30.4 6 h count 3 4 5 3 4 5 100 100 0 10 10 10 0 10 10 10 PE-EGFR PE-EGFR 0 0 3 4 5 010 10 10 0103 104 105 81.97.32 12.7 19.3 300 300 104 104

200 200 24 h count 103 103 100 100

0 0 3 4 5 2 2 010 10 10 0103 104 105 10 10

400 (0.1 mg/mL) Amivantamab

300 FITC-MET 0 FITC-MET 0 300 9.19 1.56 47.1 20.9

200 3 4 5 3 4 5 200 010 10 10 010 10 10 48 h count

100 100 PE-EGFR PE-EGFR

0 0 D 3 4 5 010 10 10 0103 104 105

300 EGFR MET DAPI Merge 300

200 200 72 h count 100 100 IgG1 count 0 0 (0.1 mg/mL) 3 4 5 010 10 10 0103 104 105 EGFR EGFR DFCI-127 IgG1 control (0.1 mg/mL)

Amivantamab (0.1 mg/mL) (0.1 mg/mL) Amivantamab

% MFI ratio D770delinsGY H773_V774InsH E 30 min 55.9 68.0 D770delinsGY H773_V774insH 6 h 58.9 49.1 −++ −++ Amivantamab (1 mg/mL) 24 h 47.5 49.3 −− 100 −− 100 Bafilomycin (nmol/L) 48 h 45.7 46.9 72 h 44.4 42.9 EGFR

GAPDH

Figure 3. Amivantamab strongly promotes internalization of EGFR and MET in Ba/F3 and PDC cells expressing EGFR Exon20ins mutations. The expression of EGFR and MET on the cell surface were determined by FACS analysis. A, EGFR expression on the plasma membrane was detected in Ba/F3 cells overexpressing D770delinsGY and H773_V774insH at the indicated time. After 0.1 mg/mL IgG1 control (cont) or 0.1 mg/mL amivantamab treatment for 72 hours, PE-EGFR and FITC-MET expression on the plasma membrane was detected in (B) DFCI-127 (P772_H773insPNP) and (C) DFCI-58 (H773_V774insNPH) cells. D, Amivantamab induced redistribution of EGFR and MET in DFCI-127 PDCs. IF staining for EGFR (green) and MET (red) in a panel of DFCI-127 treated with 0.1 mg/mL IgG1 control or 0.1 mg/mL amivantamab for 72 hours. E, Pretreatment with the autophagy inhibitor bafilomycin (100 nmol/L) for 30 minutes rescued the decreased EGFR expression in 1 mg/mL amivantamab-treated Ba/F3 cell lines overexpressing D770delinsGY or H773_V774insH.

NOG mice (Supplementary Fig. S4D and S4E). Using the observed in poziotinib-treated mice compared with amivan- previously reported dosing regimen of 5 mg/kg poziotinib, tamab-treated mice (Supplementary Fig. S4H). The favorable once a day (18), sudden death occurred within 6 days of toxicity profiles with amivantamab were consistent with those treatment. Skin toxicity analyses with poziotinib and amivan- shown in an ongoing phase I study (13). tamab revealed that poziotinib-treated mice showed severe skin toxicities on the face, abdomen, and back at dose of 5 and Amivantamab Induces Antibody-Dependent 10 mg/kg, whereas 30 mg/kg amivantamab showed only mini- Cell-Mediated Cytotoxicity in Exon20ins Models mal keratosis on the face (Supplementary Fig. S4F and S4G). The process of antibody-dependent cell-mediated cytotox- In addition to skin toxicity, a dramatic loss of body weight was icity (ADCC) is known to be initiated when both the target

1200 | CANCER DISCOVERY AUGUST 2020 AACRJournals.org

Amivantamab in NSCLC with EGFR Exon20ins RESEARCH ARTICLE

ABC D770delinsGY D770delinsGY D770delinsGY 800 Amivantamab 800 700 Vehicle (30 mg/kg)

) 700 3 600 #1 #2 #2#1 600 500 Vehicle pEGFR 500 lgG1 30 mg/kg 400 lume (mm Amivantamab 30 mg/kg * 400 EGFR 300 300 pMET mor vo 200 200 Tu 100 100 MET % change in tumor volume 0 0 GAPDH 135 7911 13 15 Vehicle Amivantamab 30 mg/kg Day after treatment

H773_V774insH H773_V774insH H773_V774insH 1,000 1,000 Amivantamab 900 900 Vehicle (30 mg/kg)

) 800

3 800 #1 #2 #2#1 700 700 Vehicle 600 600 pEGFR lgG1 30 mg/kg 500 500

lume (mm * Amivantamab 30 mg/kg 400 EGFR 400 300 300 200 pMET mor vo

Tu 200 100 % change in tumor volume 100 0 MET Vehicle Amivantamab 0 30 mg/kg GAPDH 135 7911 13 15 Day after treatment

Figure 4. Amivantamab reduces tumor burden in Ba/F3 cells and PDCs with EGFR Exon20ins xenograft models. Antitumor effects of amivantamab in (A–C) Ba/F3 cells overexpressing D770delinsGY- or H773_V774insH-bearing NOG mice and (D–I) DFCI-127- or YU-1163–bearing NOG or BALB/c nude mice, respectively. (continued on next page) cell antigen and an activated Fcγ receptor are engaged, respec- monocytes and activated NK cells during ADCC, encourag- tively, by the Fab and Fc portions of an antibody. The effector ing antigen presentation and adaptive immune responses cells, mainly natural killer (NK) cells, trigger degranulation (34, 35). To explore the correlation between amivantamab- and subsequent cytokine production, resulting in the elimi- dependent ADCC and secreted IFNγ levels, we measured the nation of the target cells (33). To determine whether ADCC level of IFNγ in a medium cocultured with PDCs and PBMC plays a role in amivantamab-mediated antitumor activity, after amivantamab treatment. Consistent with the degree of ADCC assays were performed using PDCs (DFCI-127 and the ADCC effect, IFNγ levels were significantly increased with YU-1163) expressing EGFR Exon20ins mutations cocultured amivantamab treatment compared with cetuximab treatment with peripheral blood mononuclear cells (PBMC) as effector (Fig. 5E). Treatment with a FcR blocker reduced IFNγ secre- cells [effector:target (E:T) = 50:1]. Treatment with amivan- tion, indicating that IFNγ secretion was dependent on the tamab resulted in cytotoxicity in both PDCs in a dose-depend- interaction between the Fc domain of amivantamab and ent manner and to a greater extent than cetuximab, a mAb the FcR on immune cells (Fig. 5F). Induced inflammatory targeting EGFR (Fig. 5A–C). By extension, cetuximab treat- cytokines including IFNγ secreted from NK cells activated by ment led to a less-pronounced reduction in tumor volume in amivantamab bound to EGFR and MET on EGFR Exon20ins- YU-1163–bearing BALB/c nude mice models relative to that driven tumors may lead to the recruitment and activation observed with amivantamab (Supplementary Fig. S5A). Ami- of adjacent immune cells to tumor cells in vivo. To explore vantamab-mediated cellular cytotoxicity shown in Fig. 5A this, we analyzed the infiltration of macrophages and NK was significantly impaired by incubation with an Fc receptor cells into the tumor in a PDX model (YHIM-1029), which (FcR) blocker in DFCI-127 and YU-1163 PDCs (Fig. 5D), was generated from a patient-derived tumor harboring the suggesting that the amivantamab-mediated ADCC effect D770_N771insG Exon20ins mutation (Supplementary Table requires the interaction with FcRs on PBMCs. Similarly, S2), and YU-1163–bearing BALB/c nude mice models treated the antitumor effect of amivantamab was abrogated in vivo with amivantamab at 10 and 30 mg/kg dose, respectively. when amivantamab was cotreated with anti-mouse CD16/ mF4/80 and mNKp46, markers of macrophages and NK cells CD32 antibodies to block FcRγIII/FcRγII on monocytes/mac- in BALB/c nude mice, respectively, were elevated in tumors rophages and NK cells in YU-1163–bearing BALB/c nude mice following treatment with amivantamab, suggesting that the (Supplementary Fig. S5A). It is known that inflammatory mechanistic components of ADCC observed in vitro may cytokines such as IFNγ and TNFα are secreted from infected translate to recruitment of key effector cells in tumors in vivo

AUGUST 2020 CANCER DISCOVERY | 1201

RESEARCH ARTICLE Yun et al.

DEDFCI-127 H DFCI-127 (P772_H773insPNP) DFCI-127 (P772_H773insPNP) 900 1,200 1,100 Amivantamab 800 Vehicle (30 mg/kg) Vehicle 1,000 )

3 700 900 Amivantamab 30 mg/kg #1 #2 #2#1 800 600 700 pEGFR 500 * 600

lume (mm 500 EGFR 400 400 300 300 pMET

mor vo 200 200 Tu MET

% change in tumor volume 100 100 0 Vehicle Amivantamab GAPDH 0 30 mg/kg 1 2345678910 11 Day after treatment FGI YU-1163 YU-1163 YU-1163 (S768_D770dup) 500 (S768_D770dup) 1,100 Amivantamab 1,000 400 Vehicle (30 mg/kg)

) 900 3 #1 #2 #2#1 800 300 700 pEGFR 600 Vehicle 200 EGFR lume (mm 500 Amivantamab 30 mg/kg * 400 pMET 100

mor vo 300

Tu 200 MET

100 % change in tumor volume 0 Vehicle GAPDH 0 123456789101112131415 −100 Day after treatment Amivantamab 30 mg/kg

Figure 4. (Continued) Mice were treated with vehicle, IgG1 control, or amivantamab twice per week with intraperitoneal injections dosing 30 mg/kg. Data represent the mean ± SEM (n = 5/group). *, P < 0.0001 versus vehicle or IgG1 control. B, E, and G, A waterfall plot representation of the response of each tumor taken on the last day of treatment in the xenograft mice. C, H, and I, Tumor lysates of vehicle- or amivantamab-treated Ba/F3 cells or PDC xenograft mice were harvested and subjected to immunoblotting for pEGFR (Y1068), EGFR, pMET, and MET.

(Fig. 5G; Supplementary Fig. S5B). In addition, these results effects observed on tumor growth. Histopathologic examina- suggest that amivantamab has greater ADCC and antitumor tion of tumor sections obtained following amivantamab or activity than cetuximab in the context of EGFR Exon20ins vehicle treatment using hematoxylin and eosin (H&E) staining, and that ADCC is an important mechanism in mediating the and IHC staining for EGFR, MET, and Ki-67, and terminal cytotoxic effects of amivantamab. deoxynucleotidyl transferase–mediated dUTP nick end labe- ling (TUNEL) staining, further confirmed receptor inhibition Amivantamab Demonstrates Antitumor Activity and engagement of apoptotic machinery in EGFR Exon20ins– in a PDX Model Harboring the D770_N771insG driven tumors in vivo (Fig. 6E). To verify whether the antitumor Exon20ins Mutation effect of amivantamab was affected by innate immunity in the Treatment with amivantamab in the YHIM-1029 PDX in vivo models, we blocked the mouse CD16/CD32 via admin- model with D770_N771insG (Fig. 6A) resulted in a robust istration of anti-CD16/CD32 antibodies. The antitumor effect decrease in tumor volume, indicating that the antitumor of amivantamab shown in Fig. 6B was abrogated when the activity observed in Ba/F3 and PDC models was preserved in amivantamab-treated PDX-bearing BALB/c nude mice were a PDX model (Fig. 6B). In contrast, treatment with cetuximab cotreated with anti-CD16/CD32 antibodies, indicating that (10 mg/kg) or poziotinib (1 mg/kg) only modestly reduced the antitumor effects of amivantamab were partially mediated tumor volume. The dose of poziotinib was reduced to 1 by immune cells in this condition (Supplementary Fig. S6). mg/kg for this experiment due to the toxicity of poziotinib described above (Supplementary Fig. S4D–S4H). Pharma- Antitumor Activity of Amivantamab in Patients codynamic analysis showed that amivantamab treatment with EGFR Exon20ins Disease resulted in EGFR and MET downmodulation, inhibition of In an ongoing first-in-human study of amivantamab in the downstream signaling pathways pAKT, pERK, and pS6, patients with advanced NSCLC (NCT02609776), promis- and increased markers of apoptosis (Fig. 6C). In contrast, ing clinical activity has been observed in patients with EGFR tumors from mice treated with cetuximab or poziotinib main- Exon20ins disease (13). A 58-year-old patient harboring the tained the EGFR downstream signaling components pERK EGFR H773delinsNPY Exon20ins mutation achieved a partial and pS6 (Fig. 6D), which was consistent with the modest response with a 65% tumor reduction (Fig. 7A), and a 48-year-old

1202 | CANCER DISCOVERY AUGUST 2020 AACRJournals.org

Amivantamab in NSCLC with EGFR Exon20ins RESEARCH ARTICLE

A D DFCI-127 YU-1163 40 70 * 60 35 60 * 50 30 50 * 25 40 IgG1 * 40 20 IgG1 control Amivantamab 30 Amivantamab 30 15 Cetuximab Amivantamab + 20 20 FcR blocker Cytotoxicity (%) Cytotoxicity 10 (%) Cytotoxicity Cytotoxicity (%) Cytotoxicity 5 10 10 0 0 0 0 0.01 0.1 110 100 0 0.01 0.1 110100 DFCI-127 YU-1163 Concentration (µg/mL) Concentration (µg/mL)

B DFCI-127 YU-1163 E PBMC −+ −+(E:T ratio = 5:1) DFCI-127 YU-1163 IgG1 180 1,400 (10 µg/mL) * 160 1,200 * 140 1,000 120 100 800 Amivantamab 80 600 (10 µg/mL) 60 400 IFN γ (pg/mL) 40 200 20 0 0 Cetuximab

(10 µg/mL) IgG1 IgG1

Untreated Cetuximab Untreated Cetuximab Amivantamab Amivantamab

F C DFCI-127 YU-1163 DFCI-127 YU-1163 350 * 1,200 140 * 120 ** ** 300 1,000 120 100 * * 250 − Fc blocker PBMC 100 − 800 + Fc blocker 80 200 80 PBMC + 600 (%) 60 150 D. 60 400 IFN γ (pg/mL) O. 40 100 40 50 200 20 20 0 0 0 0 IgG1 IgG1

IgG1 IgG1 Untreated Untreated Cextuximab Cextuximab Amivantamab Amivantamab Cetuximab Cetuximab Amivantamab Amivantamab

G mF4/80 mNKp46 IgG1 Amivantamab IgG1 Amivantamab (10 mg/kg) (10 mg/kg) (10 mg/kg) (10 mg/kg) 1 1 YHIM-1029 # 2# # 2# (D770_N771insG) 20 × 20 ×

Figure 5. Amivantamab has superior ADCC activity compared with cetuximab. A, Amivantamab-mediated ADCC activity against NSCLC PDCs expressing EGFR Exon20ins mutations using PBMC (E:T ratio = 50: 1). ADCC assays were performed using DFCI-127 and YU-1163 PDCs as targets in the presence of IgG1 control, amivantamab, or cetuximab at various concentrations. PBMCs were cocultured for 4 hours with PDCs. *, P < 0.0001. Data are presented as averages ± SD of triplicate independent experiments. B, Amivantamab-mediated cytotoxicity against DFCI-127 and YU-1163 PDCs. PDCs were treated with IgG1, amivantamab (10 μg/mL), or cetuximab (10 μg/mL) for 24 hours in the presence or absence of PBMC (E:T ratio = 5:1). C, Quantitative analysis of the cells shown in the representative images (n = 3). *, P < 0.0001; **, P < 0.001. D, Pretreatment with FcR blocker with PBMC (E:T ratio = 50:1) reduced the amivantamab (10 μg/mL)-mediated ADCC effects. *, P < 0.0001. Data are presented as averages ± SD of triplicate independent experiments. E, IFNγ (pg/mL) levels in the cell culture media were detected by ELISA. The PDCs were cocultured with PBMCs in the presence of IgG1, amivantamab, or cetuximab at 1 μg/mL for 4 hours, and the culture medium was used for detection of IFNγ, *, P < 0.0001 versus cetuximab at the same concentration. F, PBMCs pretreated with FcR blocker reduced the IFNγ level in the culture medium in the presence of amivantamab (10 μg/mL), *, P < 0.0001; **, P < 0.001. Data are presented as averages ± SD of triplicate independent experiments. G, IHC staining for mF4/80 (macrophages) and mNKp46 (NK cells) of tumor sections in YHIM-1029 PDX-bearing BALB/c nude mice following 10 mg/kg IgG1 or 10 mg/kg amivantamab treatment.

AUGUST 2020 CANCER DISCOVERY | 1203

RESEARCH ARTICLE Yun et al.

A YHIM-1029 (D770_N771insG) C YHIM-1029 (D770_N771insG) D770_N771insG Amivantamab Vehicle (10 mg/kg)

#1 #2 #1 #2 pEGFR

EGFR GACAACGACGGGAAC pMET B MET YHIM-1029 pAKT 1,500 AKT 1,400 Vehicle 1,300 Amivantamab 10 mg/kg pERK 1,200 Cetuximab 10 mg/kg ERK

) 1,100 Poziotinib 1 mg/kg 3 1,000 pS6 900 S6 800 700 p21 600 * p27 500

Tumor volume (mm volume Tumor 400 p53 300 * Cleaved caspase-3 200 * 100 Cleaved PARP 0 BIM 1 357911 13 15 17 19 21 23 25 27 29 31 Day after treatment GAPDH

E YHIM-1029 Amivantamab Vehicle (10 mg/kg) D YHIM-1029 (D770_N771insG)

Vehicle Cetuximab (10 mg/kg) Poziotinib (1 mg/kg)

#1 #2 #1 #2 #1 #2 pEGFR

EGFR

pMET

MET

pAKT

AKT

pERK

ERK

pS6

BIM

GAPDH TUNEL Ki-67 H&E pMET MET pEGFR EGFR

Figure 6. Amivantamab reduces tumors in a PDX model with D770_N771insG EGFR mutation. A, Sanger sequencing data depicting the D770_N771 insG mutations of the EGFR gene in a PDX model. B, Patient-derived tumors implanted in BALB/c nude mice were treated with vehicle, amivantamab (10 mg/kg), cetuximab (10 mg/kg), twice per week, intraperitoneal injections, or poziotinib (1 mg/kg), once a day. Data represent the mean ± SEM (n = 7/ group). *, P < 0.0001. Western blot analysis for the downstream signaling pathways of EGFR, MET, and apoptosis markers in tumors obtained from YHIM- 1029 PDX models treated with (C) 10 mg/kg amivantamab, (D) 10 mg/kg cetuximab, or 1 mg/kg poziotinib. E, Histopathologic examination of tumor sections obtained from the PDX models following 10 mg/kg amivantamab or vehicle treatment. H&E staining and IHC staining for EGFR, pEGFR, MET, pMET, Ki-67, and TUNEL.

1204 | CANCER DISCOVERY AUGUST 2020 AACRJournals.org

Amivantamab in NSCLC with EGFR Exon20ins RESEARCH ARTICLE

A Figure 7. Amivantamab reduces tumors in patients with NSCLC with EGFR CASE #1 Exon20ins mutations. Radiologic response Pre Amivantamab Post following amivantamab 1,050 mg treatment in (A) a 58-year-old patient with the EGFR H773delinsNPY mutation and (B) a 48-year-old patient with the EGFR S768_D770dup mutation. C, A proposed model of diverse antitumor mechanisms of amivantamab in NSCLC with EGFR Exon20ins.

EGFR H773delinsNPY ORR PR (−65% tumor reduction)

B CASE #2 Pre Amivantamab Post

EGFR S768_D770dup ORR PR (−38.9% tumor reduction)

C NK cells

Amivantamab

2 Direct inhibition of tumor growth mediated by ADCC and ADCP Macrophage EGFR Exon20ins MET

Apoptosis P P P P P

P P P P EGFR P P P P Exon20ins 1 Blockade of EGFR and MET downstream signaling pathway by internalization of EGFR and MET

O H F N N N N N O HN Cl N N O NH N O N NSCLC with EGFR Exon20ins O Osimertinib Gefitinib patient with the EGFR S768_D770dup Exon20ins mutation mutations. In several Ba/F3 and PDC models expressing achieved a partial response with a 38.9% tumor reduction diverse EGFR Exon20ins mutations, amivantamab treatment (Fig. 7B). These patients were progression-free for 92 and 32 resulted in EGFR and MET internalization, inhibition of weeks, respectively, on amivantamab, with manageable toxicities. downstream signaling cascades, engagement of apoptotic machinery, and subsequent inhibition of tumor cell proliferation. Importantly, these diverse action mechanisms of Discussion amivantamab were preserved in vivo as evidenced by phar- In our study, we characterized the antitumor activity of macodynamic analyses of tumors from cell line xenografts amivantamab, a novel EGFR–MET bispecific antibody, in and PDX models treated with amivantamab (Fig. 7C). Fur- multiple preclinical models harboring EGFR Exon20ins thermore, to the best of our knowledge, we first presented

AUGUST 2020 CANCER DISCOVERY | 1205

RESEARCH ARTICLE Yun et al. evidence of clinical activity of amivantamab in two case stud- IFNγ secreted from amivantamab mediated active immune ies of patients with EGFR Exon20ins NSCLC from an ongo- cells, including NK cells and macrophages, may restimulate ing phase I trial, highlighting the important translational and recruit surrounding immune cells to the tumor, although nature of our preclinical work. additional studies are required to investigate this hypothesis. Currently, there are no targeted therapies approved for In addition, immunocytokines secreted from NK cells lead EGFR Exon20ins–positive advanced NSCLC. Owing to its to upregulation of ICAM1 on target cells, rendering them small size and flexibility, poziotinib, an oral pan-HER inhibi- more susceptible to target cell cytolysis (41). Indeed, phar- tor, has demonstrated greater activity than approved EGFR macodynamic analyses of tumors from mice treated with TKIs in vitro and in PDX models of EGFR Exon20ins–mutant amivantamab revealed increased tumor levels of NK cells and NSCLC (18). In a single-center phase II trial, poziotinib macrophages. In contrast to many other therapeutic antibod- showed a 43% confirmed RR in heavily pretreated advanced ies used in the clinical setting, amivantamab was designed and EGFR Exon20ins–mutant NSCLC (21). However, in a subse- engineered with a low fucose backbone, which enhances its quent pivotal phase II trial (NCT03318939), poziotinib yielded binding to FcγRIIIa (17), which is present on NK cells, mono- only 14.8% RR in a similar patient population (https://www. cytes, and macrophages. The human FcγRIIIa, critical for precisiononcologynews.com/drug-discovery-development/ ADCC, binds antibodies with low-level core fucosylation more spectrums-poziotinib-failed-meet-primary-phase-ii-trial- tightly and consequently mediates more potent and effective endpoint#.XjJWkWgzaUk). TAK-788 has also shown preclinical ADCC killing of cancer cells (42). Thus, the enhanced binding activity against activating EGFR and HER2 mutations includ- of amivantamab to FcγRIIIa may lead to increased induction ing EGFR Exon20ins. Although preliminary, in a phase I/II of Fc effector functions in comparison with other (normal study (NCT02716116), TAK-788 produced clinical activity fucose) hIgG1 antibodies such as cetuximab. in a small subset of EGFR Exon20ins–mutant NSCLC (11). In this study, we used two different xenograft models, Importantly, treatment with poziotinib or TAK-788 was asso- NOG mice and BALB/c nude mice for in vivo study. As ciated with a high incidence of EGFR WT–driven toxicity such demonstrated in many studies, NOG mice have impaired as diarrhea and rashes, further limiting their clinical utility. innate immunity and extremely low NK-cell activity, whereas Therefore, there is a substantial clinical need to identify new BALB/c nude mice have intact innate immunity and active therapies for patients with EGFR Exon20ins. NK cells (43, 44). For this reason, more potent ADCC activ- In our study, amivantamab was clearly superior to pozi- ity was expected in BALB/c nude mice than NOG mice. otinib or cetuximab in terms of efficacy and tolerability in Therefore, a minimal amivantamab-mediated ADCC activity xenografts. Multiple facets of the MOA of amivantamab may resulted in the modest efficacy with amivantamab in Ba/F3- contribute to the superior antitumor activity of amivantamab and DFCI-127–bearing NOG mice (Fig. 4A and D; Supple- in the context of Exon20ins. In addition to the different mentary Fig. S4A). On the other hand, the significant tumor MOA described in the various Exon20ins models described regression was observed with amivantamab in YU-1163–bear- above, the ability of amivantamab to simultaneously bind ing and YHIM-1029–bearing BALB/c nude mice (Figs. 4F and two distinct epitopes of EGFR and MET may result in the 6B), which resulted from multiple MOA of amivantamab to concurrent interference with highly interconnected signaling block the EGFR and MET downstream signaling pathway pathways. To this end, amivantamab has been shown to have and elicit ADCC. higher efficacy in decreasing tumor growth in the H1975- It has been observed that EGFR and tumor suppressor HGF model compared with the combination of anti-EGFR TP53 genes are commonly mutated in patients with NSCLC and anti-MET monovalent antibodies (36). This finding can with independent prognostic implications. Furthermore, in be partly explained by an “avidity effect” whereby tumor patients with concomitant mutations in EGFR and TP53, cells expressing both targets, that is, EGFR and MET, bind there have been reports of decreased responsiveness to EGFR to both arms of amivantamab with higher affinity than do TKIs (45). A similar effect was observed in our study in the cells that express only one target or engage a single Fab arm. YU-1163 PDC and YUO-029 PDO following treatment with A previous study has described correlations between binding amivantamab. On the other hand, amivantamab exhibited a affinity, receptor density, and receptor phosphorylation with potent in vivo activity in YU-1163–bearing BALB/c nude mice, amivantamab (37). Overall, bispecific antibodies show greater suggesting that ADCC activity of amivantamab, shown in antitumor efficacy compared with a combination of monospe- Fig. 5A–G, was believed to be involved in the in vivo antitumor cific mAbs via potential synergistic effects, and they increase activity. These results suggest that the combination of effector selectivity by simultaneous targeting both receptors, favoring cell–dependent and -independent MOAs elicited by amivan- overexpressing cells as a consequence of avidity effects (38). tamab (Fig. 7C) may result in antitumor activity in tumors Amivantamab, produced by an engineered cell line defec- harboring a coalescence of intractable mutations, for exam- tive for protein fucosylation, has a low-level core fucosylation. ple in EGFR Exon20ins disease, and concomitant deleterious The human FcγRIIIa, critical for ADCC, binds antibodies with mutations such as the TP53 mutation present in our preclini- low core fucosylation with higher affinity and consequently cal model and reported in the broader patient population. mediates more potent and effective NK cell–mediated killing In our study, amivantamab demonstrated less skin tox- of cancer cells (12). In this study, amivantamab demonstrated icity than poziotinib. Amivantamab-treated BALB/c nude more robust ADCC in EGFR Exon20ins mutation models mice appeared phenotypically normal with only minimal than cetuximab, an EGFR-directed antibody that has not signs of keratosis on the face, indicating that amivantamab shown robust utility in NSCLC (39, 40). This ADCC activity was well tolerated in this preclinical model. In contrast, was correlated with secreted IFNγ levels. It is possible that the treatment with poziotinib resulted in severe keratosis,

1206 | CANCER DISCOVERY AUGUST 2020 AACRJournals.org

Amivantamab in NSCLC with EGFR Exon20ins RESEARCH ARTICLE significant weight loss, and even sudden death (Supplemen- 10 μmol/L Y-27632 (Enzo). R-spondin conditioned medium was tary Fig. S4D–S4H). produced from HA-R-Spondin1-Fc 293T Cells (Amsbio). For passag- In conclusion, our data demonstrated that amivantamab ing, organoids were collected, mechanically sheared with a 25-gauge functions through multiple MOA to elicit antitumor activity needle, and washed with cold PBS before the organoid pellets were in multiple preclinical models of EGFR Exon20ins disease. resuspended in the Matrigel and seeded into 24-well plates at ratios of 1:2 to 1:4. The culture medium was replenished at least twice a Consequently, amivantamab warrants further clinical inves- week. Cell viability tests were performed as described previously (47). tigation, as evidenced by clinical results from 2 patients with Briefly, organoids were trypsinized into single cells and cultured for EGFR Exon20ins NSCLC who have been treated with ami- 5 to 10 days. Then, the organoids were collected, resuspended in vantamab in the clinical setting. This represents important the medium containing 5% Matrigel, and plated in a 96-well plate progress toward the identification of an effective therapeutic (Corning) at a concentration of 2,000 organoids/μL. The medium option for patients with NSCLC with EGFR Exon20ins, an with the IgG1 control or amivantamab at diverse concentrations was area of high unmet medical need. added and incubated for 72 hours. Cell viability was measured using CellTiter-Glo 3D Culture Reagent (Promega) on a microplate lumi- nometer according to the manufacturer’s instructions. Methods PDX Models Ba/F3 Cell Lines and Drug Compounds PDXs were created using 6- to 8-week-old female SCID (NOG) and All mutant Ba/F3 cell lines were purchased from the German Collec- nude (nu/nu) mice obtained from Orient Bio. All methods complied tion of Microorganisms and Cell Cultures and were obtained from the with the guidelines of our Institutional Animal Research Committee Dana-Farber Cancer Institute, Harvard University (Boston, MA). All (Yonsei University College of Medicine, Seoul, Republic of South cells were maintained in RPMI1640 medium supplemented with 10% Korea) and were approved by the Association for Assessment and FBS and puromycin in a humidified incubator with 5% CO . Amivan- 2 Accreditation of Laboratory Animal Care (AAALAC). After removal tamab and IgG1 controls were provided by Janssen. Gefitinib, osimer- of the necrotic and supporting tissues from core biopsy specimens, tinib, cetuximab, and poziotinib were purchased from SelleckChem. small specimens of the tumor tissue (3 mm × 3 mm × 3 mm) from each patient were implanted subcutaneously in 1 to 2 mice. After the Antibodies tumor reached 1.5 cm in diameter, it was excised, dissected into small Primary antibodies specific for pEGFR (2234), EGFR (4267), pMET specimens (3 mm × 3 mm × 3 mm), and reimplanted into nude mice. (3077), MET (8198), pERK (4370), ERK (9107), pAKT (9271), AKT (9272), pS6 (4858), S6 (2217), p27 (2252), cleaved PARP (5625S), In Vivo Xenograft Studies cleaved caspase-3 (9661), and BIM (2933) were purchased from Cell Female athymic BALB-c/nu mice were obtained from Orient Bio Signaling Technology; p21(sc-817) and p53 (sc-126) were purchased at 5 to 6 weeks of age. All mice were handled in accordance with the from Santa Cruz Biotechnology, Inc.; and GAPDH (PAB13195) was Animal Research Committee’s Guidelines at Yonsei University Col- purchased from Abnova. For the IHC assay, mF4/80 (#70076) and lege of Medicine, and all facilities were approved by AAALAC. Ba/F3 mNKp46 (AF2225) were purchased from Cell Signaling Technology cells and PDCs (1 × 107 cells) were injected subcutaneously into the and R&D Systems, respectively. NOG and BALB-c/nu mice, respectively, and growth was measured twice weekly; after establishment of palpable lesions, mice were PDCs assigned to testing. Once the tumor volume reached approximately YU-1163 (S768_D770dup) cell lines were derived from malignant 150 to 200 mm3, mice were randomly allocated into groups of 5 effusions from patients with NSCLC and cultured on collagen-coated animals to receive either vehicle, IgG1 control, or amivantamab. The plates in ACL-4 medium supplemented with 5% FBS. The cells main- tumor size was measured every 2 days using calipers. The average tained the driver oncogenes that were observed in the patients. Cells tumor volume in each group was expressed in mm3 and calculated were enriched in an epithelial cell adhesion molecule (EPCAM)– according to the equation for a prolate spheroid: tumor volume = positive cell population with a purity of over 95% before they were sub- 0.523 × (large diameter) × (small diameter)2. jected to further assays. DFCI-58 (H773_V774insNPH) and DFCI-127 (P772_H773insPNP) cell lines were obtained from the Dana-Farber Antiproliferation Assay Cancer Institute and were cultured in ACL-4 medium and RPMI Ba/F3 cells or PDCs expressing EGFR Exon20ins mutations were medium with 10% FBS, respectively. All patient samples were collected seeded onto 96-well plates in 100 μL. After treatment with IgG1 after written informed consent from the patients was obtained. The control, amivantamab, gefitinib, or osimertinib for 72 hours, cell study protocols were approved by the respective institutional review viability was measured by quantifying the total amount of ATP using boards. the CellTiter-Glo 2.0 Assay Kit (Promega) according to the manufacturer’s instructions. PDO Culture PDOs (YUO-029 and YUO-036) were established as described pre- Colony Formation Assay viously (46). Briefly, malignant effusions from 2 patients with NSCLC Cells were seeded onto 6-well culture plates and incubated for were collected and centrifuged, and the cell pellets were mixed with 12 days at 37°C with amivantamab (0, 0.1, or 1 mg/mL). Cells were growth factor–reduced Matrigel (Corning) and seeded into 48-well washed with PBS, fixed, and stained with 4% paraformaldehyde in plates. Solidified gels were overlaid with advanced DMEM/F12 (Inv- 5% crystal violet for 10 minutes. Colonies were eluted with 1% SDS, itrogen) containing 1× Glutamax (Invitrogen), 10 mmol/L HEPES and the optical density value was determined using ELISA at 470 nm. (Invitrogen), 1× Antibiotic-Antimycotic (Invitrogen), 1× B-27 (Inv- itrogen), 20% R-spondin conditioned medium, 5 mmol/L Nicotina- mide (Sigma), 1.25 mmol/L N-acetylcysteine (Sigma), 500 nmol/L ADCC Assays SB-202190 (Sigma), 500 nmol/L A83-01 (Tocris), 100 ng/mL Mouse The ADCC assay was conducted using the Lactase Dehydrogenase Noggin (PeproTech), 100 ng/mL human FGF10 (PeproTech), 25 ng/ (LDH) Cytotoxicity Detection Kit (Roche) in accordance with the mL human FGF7 (PeproTech), 50 μg/mL Primocin (InvivoGen), and manufacturer’s instructions. Human PBMCs obtained from healthy

AUGUST 2020 CANCER DISCOVERY | 1207

RESEARCH ARTICLE Yun et al. volunteers were used as the effector cells. ADCC was conducted using MOGAM Institute, Dong-A ST, Champions Oncology, Janssen, an E:T cell ratio ranging from 50:1 to 5:1 and incubated for 4 to Yuhan, Ono, and Dizal Pharma, reports receiving speakers bureau

24 hours at 37°C in 5% CO2. Amivantamab concentrations of 100 honoraria from Novartis, Bayer, MSD, AstraZeneca, MOGAM Insti- to 0.01 μg/mL were tested. The LDH activity of the cell culture tute, Dong-A ST, Champions Oncology, Janssen, Yuhan, Ono, and supernatants was measured, and the percentage cytotoxicity was Dizal Pharma, has ownership interest (including patents) in Thera calculated as described in the manufacturer’s protocol. CanVac Inc, Gencurix Inc, Bridgebio Therapeutics, and KANAPH Therapeutic Inc, and has received other remuneration from Daan IF Analysis Biotherapeutics. No potential conflicts of interest were disclosed by PDCs were seeded on 0.01% poly-L-lysine (Sigma-Aldrich) coated the other authors. coverslips. The following day, cells were treated with IgG1 control or amivantamab at 0.1 mg/mL. After 72 hours, the coverslips were Authors’ Contributions fixed in 4% formaldehyde for 15 minutes, permeabilized with 0.5% Conception and design: J. Yun, J.-H. Kim, S.M. Lim, M.V. Lorenzi, Triton X-100 for 5 minutes, and incubated with primary antibody B.C. Cho for 1 hour at room temperature. The primary antibodies used in the Development of methodology: J. Yun, J.-H. Kim, S.M. Lim, B.C. Cho study were rabbit monoclonal anti-EGFR and anti-MET (Santa Cruz Acquisition of data (provided animals, acquired and managed Biotechnology) and ab992 (Millipore) at a dilution of 1:100. The cov- patients, provided facilities, etc.): J. Yun, S.-Y. Kim, J.-H. Kim, erslips were rinsed twice with PBS, followed by incubation with the M.H. Hong, H.R. Kim, R.E. Knoblauch, B.C. Cho appropriate fluorophore-conjugated secondary antibody (Invitrogen) Analysis and interpretation of data (e.g., statistical analysis, for 1 hour at room temperature. The cells were counterstained with biostatistics, computational analysis): J. Yun, J.-H. Kim, K.-H. Pyo, 4′,6-diamidino-2-phenylindole (DAPI; 300 nmol/L; Invitrogen), and S.G. Heo, S. Lim, J.C. Curtin, R.E. Knoblauch, B.C. Cho the coverslips were mounted on slides using Faramount Aqueous Writing, review, and/or revision of the manuscript: J. Yun, S.-H. Lee, Mounting Medium (Dako). M.R. Yun, M. Thayu, J.C. Curtin, R.E. Knoblauch, M.V. Lorenzi, A. Roshak, B.C. Cho IHC Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): S.-H. Lee, S.-Y. Jeong, IHC was performed using the Automated Staining System (BOND J.-H. Kim, C.-W. Park, B.C. Cho Rx, Leica Biosystems). Briefly, 4-mm paraffin-embedded tumor Study supervision: S.M. Lim, B.C. Cho sections were deparaffinized and rehydrated. Slides then under- went heat-induced epitope retrieval with citrate buffer at 100°C for Acknowledgments 20 minutes. Antibodies were used at 1:100 dilution and hematoxylin solution was used for counterstaining. Stained slides were visualized This research was supported by the Basic Science Research Pro- with a Vectra Polaris and the Phenochart program. gram through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT & Future Planning (NRF- In Vivo Pharmacodynamic Study 2016R1A2B3016282 and 2018R1C1B6008916), Republic of South Korea. Mice bearing tumor tissues were treated with vehicle, IgG1 control, or amivantamab (10 or 30 mg/kg) twice per week intraperitoneally, or Received January 31, 2020; revised April 17, 2020; accepted May 7, cetuximab (10 mg/kg) or poziotinib (1 mg/kg) once daily. The tumor 2020; published first May 15, 2020. samples were collected 48 hours after 15 days of treatment, and EGFR and MET downstream signaling was evaluated by immunoblotting.

Patients References The clinical study (NCT02609776) was approved by the appro- . 1 Fang WF, Huang YH, Hong SD, Zhang ZH, Wang MH, Gan JD, et al. priate institutional review boards at each participating site and EGFR exon 20 insertion mutations and response to osimertinib in conducted in accordance with Good Clinical Practice guidelines and non-small-cell lung cancer. BMC Cancer 2019;19:595. the ethical principles of the Declaration of Helsinki. All patients pro- 2. Nagano T, Tachihara M, Nishimura Y. Mechanism of resistance to vided written informed consent. epidermal growth factor receptor-tyrosine kinase inhibitors and a potential treatment strategy. Cells 2018;7:212. 3. Kris MG, Johnson BE, Berry LD, Kwiatkowski DJ, Iafrate AJ, Wistuba Statistical Analysis II, et al. Using multiplexed assays of oncogenic drivers in lung cancers Data were collected from three independent experiments and to select targeted drugs. JAMA 2014;311:1998–2006. either presented descriptively or analyzed by one-way ANOVA, fol- 4. Travis WD. 2015 WHO classification of the pathology and genetics of lowed by the Dunnett test or Student t test. Dose–response curves tumors of the lung. J Thorac Oncol 2015;10:S68. were prepared using GraphPad Prism (Ver. 5, GraphPad Software 5. Gazdar AF. Activating and resistance mutations of EGFR in non- Inc.). small-cell lung cancer: role in clinical response to EGFR tyrosine kinase inhibitors. Oncogene 2009;28:S24–S31. Disclosure of Potential Conflicts of Interest 6. Vyse S, Huang PH. Targeting EGFR exon 20 insertion mutations in non-small cell lung cancer. Signal Transduct Target Ther 2019;4:5. J.C. Curtin is an associate director, oncology translational research 7. Yasuda H, Park E, Yun CH, Sng NJ, Lucena-Araujo AR, Yeo WL, et al. at Janssen Research and Development. R.E. Knoblauch is an executive Structural, biochemical, and clinical characterization of epidermal medical director at Janssen. M.V. Lorenzi is vice president, oncology growth factor receptor (EGFR) exon 20 insertion mutations in lung at Janssen, Pharmaceutical Companies of Johnson and Johnson and cancer. Sci Transl Med 2013;5:216ra177. has ownership interest (including patents) in Johnson & Johnson. 8. Riess JW, Gandara DR, Frampton GM, Madison R, Peled N, Bufill JA, B.C. Cho is a professor at Novartis, AstraZeneca, Takeda, Janssen, et al. Diverse EGFR exon 20 insertions and co-occurring molecular Medpacto, Blueprint Medicines, MSD, Boehringer-Ingelheim, Roche, alterations identified by comprehensive genomic profiling of NSCLC. BMS, Ono, Yuhan, Pfizer, Eli Lilly, and Janssen, reports receiving J Thorac Oncol 2018;13:1560–8. commercial research grants from Novartis, Bayer, MSD, AbbVie, 9. Oxnard GR, Lo PC, Nishino M, Dahlberg SE, Lindeman NI, Medpacto, GIInnovation, Eli Lilly, Blueprint Medicines, AstraZeneca, Butaney M, et al. Natural history and molecular characteristics of

1208 | CANCER DISCOVERY AUGUST 2020 AACRJournals.org

Amivantamab in NSCLC with EGFR Exon20ins RESEARCH ARTICLE

lung cancers harboring EGFR exon 20 insertions. J Thorac Oncol 28. Mantovani F, Collavin L, Del Sal G. Mutant p53 as a guardian of the 2013;8:179–84. cancer cell. Cell Death Differ 2019;26:199–212. 10. Yang ZD, Tchekmedyian N, Chu DT, Reddy G, Bhat G, Socinski MA. 29. Alam SK, Yadav VK, Bajaj S, Datta A, Dutta SK, Bhattacharyya M, A phase 2 study of poziotinib in patients with EGFR or HER2 exon et al. DNA damage-induced ephrin-B2 reverse signaling promotes 20 mutation-positive non-small cell lung cancer. J Clin Oncol 36:15s, chemoresistance and drives EMT in colorectal carcinoma harboring 2018 (suppl; abstr TPS9106). mutant p53. Cell Death Differ 2016;23:707–22. 11. Janne PA, Neal JW, Camidge DR, Spira AI, Piotrowska Z, Horn L, et al. 30. Sauer L, Gitenay D, Vo C, Baron VT. Mutant p53 initiates a feedback Antitumor activity of TAK-788 in NSCLC with EGFR exon 20 inser- loop that involves Egr-1/EGF receptor/ERK in prostate cancer cells. tions. J Clin Oncol 2019;37:9007. Oncogene 2010;29:2628–37. 12. Cho BC, Lee JS, Han JY, Cho EK, Haura E, Lee KH, et al. JNJ-61186372 31. Sunada H, Magun BE, Mendelsohn J, Macleod CL. Monoclonal- (JNJ-372), an EGFR-cMET bispecific antibody, in advanced non-small antibody against epidermal growth-factor receptor is internalized cell lung cancer (NSCLC): an update on phase I results. Ann Oncol without stimulating receptor phosphorylation. Proc Natl Acad Sci 2018;29:viii493–547. U S A 1986;83:3825–9. 13. Haura EB, Cho BC, Lee JS, Han JY, Lee KH, Sanborn RE, et al. JNJ- 32. Okada Y, Kimura T, Nakagawa T, Okamoto K, Fukuya A, Goji T, et al. 61186372 (JNJ-372), an EGFR-cMet bispecific antibody, in EGFR- EGFR downregulation after anti-EGFR therapy predicts the antitu- driven advanced non-small cell lung cancer (NSCLC). J Clin Oncol mor effect in colorectal cancer. Mol Cancer Res 2017;15:1445–54. 37:15s, 2019 (suppl; abstr 9009). 33. Wang W, Erbe AK, Hank JA, Morris ZS, Sondel PM. NK cell-mediated 14. Moores SL, Chiu ML, Bushey BS, Chevalier K, Luistro L, Dorn K, antibody-dependent cellular cytotoxicity in cancer immunotherapy. et al. A novel bispecific antibody targeting EGFR and cMet is effective Front Immunol 2015;6:368. against EGFR inhibitor-resistant lung tumors. Cancer Res 2016;76: 34. Fauriat C, Long EO, Ljunggren HG, Bryceson YT. Regulation of 3942–53. human NK-cell cytokine and chemokine production by target cell 15. Kosaka T, Tanizaki J, Paranal RM, Endoh H, Lydon C, Capelletti M, recognition. Blood 2010;115:2167–76. et al. Response heterogeneity of EGFR and HER2 exon 20 insertions 35. Reefman E, Kay JG, Wood SM, Offenhauser C, Brown DL, Roy S, et al. to covalent EGFR and HER2 inhibitors. Cancer Res 2017;77:2712–21. Cytokine secretion is distinct from secretion of cytotoxic granules in 16. Frega S, Lorenzi M, Fassan M, Indraccolo S, Calabrese F, Favaretto A, NK cells. J Immunol 2010;184:4852–62. et al. Clinical features and treatment outcome of non-small cell lung 36. Zheng S, Moores S, Jarantow S, Pardinas J, Chiu M, Zhou H, et al. cancer (NSCLC) patients with uncommon or complex epidermal growth Cross-arm binding efficiency of an EGFR X C-Met bispecific anti- factor receptor (EGFR) mutations. Oncotarget 2017;8:32626–38. body. Clin Pharmacol Ther 2017;101:S14. 17. Grugan KD, Dorn K, Jarantow SW, Bushey BS, Pardinas JR, Laquerre 37. Jarantow SW, Bushey BS, Pardinas JR, Boakye K, Lacy ER, Sanders R, S, et al. Fc-mediated activity of EGFR x c-Met bispecific antibody JNJ- et al. Impact of cell-surface antigen expression on target engagement 61186372 enhanced killing of lung cancer cells. Mabs 2017;9:114–26. and function of an epidermal growth factor receptor x c-MET bispe- 18. Robichaux JP, Elamin YY, Tan Z, Carter BW, Zhang SX, Liu SW, cific antibody. J Biol Chem 2015;290:24689–704. et al. Mechanisms and clinical activity of an EGFR and HER2 exon 38. Sellmann C, Doerner A, Knuehl C, Rasche N, Sood V, Krah S, et al. 20-selective kinase inhibitor in non-small cell lung cancer. Nat Med Balancing selectivity and efficacy of bispecific epidermal growth fac- 2018;24:638–46. tor receptor (EGFR) x c-MET antibodies and antibody-drug conju- 19. Colwell J. Poziotinib shows promise for rare lung cancer. Cancer Discov gates. J Biol Chem 2016;291:25106–19. 2018;8:OF4. 39. Wong SF. Cetuximab: an epidermal growth factor receptor monoclo- 20. Cha MY, Lee KO, Kim M, Song JY, Lee KH, Park J, et al. Antitumor nal antibody for the treatment of colorectal cancer. Clin Ther 2005; activity of HM781-36B, a highly effective pan-HER inhibitor in 27:684–94. erlotinib-resistant NSCLC and other EGFR-dependent cancer mod- 40. Ramalingam S, Forster J, Naret C, Evans T, Sulecki M, Lu HL, et al. els. Int J Cancer 2012;130:2445–54. Dual inhibition of the epidermal growth factor receptor with cetuxi- 21. Heymach J, Negrao M, Robichaux J, Carter B, Patel A, Altan M, et al. A mab, an IgG1 monoclonal antibody, and gefitinib, a tyrosine kinase phase II trial of poziotinib in EGFR and HER2 exon 20 mutant non- inhibitor, in patients with refractory non-small cell lung cancer small cell lung cancer (NSCLC). J Thorac Oncol 2018;13:S323–S4. (NSCLC): a phase I study. J Thorac Oncol 2008;3:258–64. 22. Costa DB, Halmos B, Kumar A, Schumer ST, Huberman MS, 41. Wang RP, Jaw JJ, Stutzman NC, Zou ZC, Sun PD. Natural killer Boggon TJ, et al. BIM mediates EGFR tyrosine kinase inhibitor- cell-produced IFN-γ and TNF-α induce target cell cytolysis through induced apoptosis in lung cancers with oncogenic EGFR mutations. up-regulation of ICAM-1. J Leukocyte Biol 2012;91:299–309. PLoS Med 2007;4:1669–80. 42. Satoh M, Iida S, Shitara K. Non-fucosylated therapeutic antibod- 23. Shi PY, Oh YT, Deng L, Zhang GJ, Qian GQ, Zhang S, et al. Overcom- ies as next-generation therapeutic antibodies. Expert Opin Biol Th ing acquired resistance to AZD9291, a third-generation EGFR inhibi- 2006;6:1161–73. tor, through modulation of MEK/ERK-dependent Bim and Mcl-1 43. Puchalapalli M, Zeng XK, Mu L, Anderson A, Glickman LH, Zhang M, degradation. Clin Cancer Res 2017;23:6567–79. et al. NSG mice provide a better spontaneous model of breast cancer 24. Canale M, Petracci E, Delmonte A, Chiadini E, Dazzi C, Papi M, et al. metastasis than athymic (Nude) mice. PLoS One 2016;11:e0163521. Impact of TP53 mutations on outcome in EGFR-mutated patients 44. Okada S, Vaeteewoottacharn K, Kariya R. Application of highly treated with first-line tyrosine kinase inhibitors. Clin Cancer Res immunocompromised mice for the establishment of patient-derived 2017;23:2195–202. xenograft (PDX) models. Cells 2019;8:889. 25. Conde E, Angulo B, Tang MY, Morente M, Torres-Lanzas J, Lopez- 45. Aggarwal C, Davis CW, Mick R, Thompson JC, Ahmed S, Jeffries Encuentra A, et al. Molecular context of the EGFR mutations: evi- S, et al. Influence of TP53 mutation on survival in patients with dence for the activation of mTOR/S6K signaling. Clin Cancer Res advanced EGFR-mutant non-small-cell lung cancer. JCO Precis 2006;12:710–7. Oncol 2018;2018:10.1200/PO.18.00107. 26. Lin CL, Liang YJ, Zhu HL, Zhang JJ, Zhong XY. R280T mutation 46. Sachs N, Papaspyropoulos A, Zomer-van Ommen DD, Heo I, Bot- of p53 gene promotes proliferation of human glioma cells through tinger L, Klay D, et al. Long-term expanding human airway organoids GSK-3 beta/PTEN pathway. Neurosci Lett 2012;529:60–5. for disease modeling. EMBO J 2019;38:e100300. 27. Zhu HB, Yang K, Xie YQ, Lin YW, Mao QQ, Xie LP. Silencing of 47. Roerink SF, Sasaki N, Lee-Six H, Young MD, Alexandrov LB, Behjati mutant p53 by siRNA induces cell cycle arrest and apoptosis in S, et al. Intra-tumour diversification in colorectal cancer at the single- human bladder cancer cells. World J Surg Oncol 2013;11:22. cell level. Nature 2018;556:457–62.

AUGUST 2020 CANCER DISCOVERY | 1209

Antitumor Activity of Amivantamab (JNJ-61186372), an EGFR −MET Bispecific Antibody, in Diverse Models of EGFR Exon 20 Insertion−Driven NSCLC

Jiyeon Yun, Soo-Hwan Lee, Seok-Young Kim, et al.

Cancer Discov 2020;10:1194-1209. Published OnlineFirst May 15, 2020.

Updated version Access the most recent version of this article at: doi:10.1158/2159-8290.CD-20-0116

Supplementary Access the most recent supplemental material at: Material http://cancerdiscovery.aacrjournals.org/content/suppl/2020/05/15/2159-8290.CD-20-0116.DC 1

Cited articles This article cites 46 articles, 14 of which you can access for free at: http://cancerdiscovery.aacrjournals.org/content/10/8/1194.full#ref-list-1

Citing articles This article has been cited by 5 HighWire-hosted articles. Access the articles at: http://cancerdiscovery.aacrjournals.org/content/10/8/1194.full#related-urls

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/10/8/1194. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.