KRAS and EGFR Amplifications Mediate Resistance to Rociletinib and Osimertinib in Acquired Afatinib-Resistant NSCLC Harboring Exon 19 Deletion/T790M in EGFR

Published OnlineFirst October 15, 2018; DOI: 10.1158/1535-7163.MCT-18-0591

Cancer Biology and Translational Studies Molecular Cancer Therapeutics KRAS and EGFR Ampliﬁcations Mediate Resistance to Rociletinib and Osimertinib in Acquired Afatinib-Resistant NSCLC Harboring Exon 19 Deletion/T790M in EGFR Kaori Nakatani1, Toshimitsu Yamaoka2, Motoi Ohba2, Ken-Ichi Fujita2, Satoru Arata2,3, Sojiro Kusumoto4, Iori Taki-Takemoto1, Daisuke Kamei1, Shinichi Iwai1, Junji Tsurutani2, and Tohru Ohmori2,4

Abstract

The critical T790M mutation in EGFR, which mediates nib-resistant cell lines, indicating a linear and reversible resistance to first- and second-generation EGFR tyrosine kinase increase with increased osimertinib concentrations in OsiR1 inhibitors (TKI; gefitinib, erlotinib, and afatinib), has facili- and OsiR2 cells. OsiR1 cells maintained osimertinib resistance tated the development of third-generation mutation-selective with KRAS amplification after osimertinib withdrawal for 2 EGFR TKIs (rociletinib and osimertinib). We previously months. OsiR2 cells exhibited KRAS attenuation, and osimer- reported heterogeneous afatinib-resistant mechanisms, tinib sensitivity was entirely recovered. Phospho-EGFR including emergence of T790M-EGFR, and responses to (Y1068) and growth factor receptor–bound protein 2 third-generation EGFR TKIs. Here, we used afatinib-resistant (GRB2)/son of sevenless homolog 1 (SOS1) complex was lung adenocarcinoma cells [AfaR (formerly AFR3) cells], car- found to mediate osimertinib resistance in OsiR1 cells with rying exon 19 deletion/T790M in EGFR. To identify the novel sustained KRAS activation. After 2 months of osimertinib resistance mechanisms in post-afatinib treatment, RocR1/ withdrawal, this complex was dissociated, and the EGFR RocR2 and OsiR1/OsiR2 cells were established using increas- signal, but not the GRB2/SOS1 signal, was activated. Con- ing concentrations of rociletinib and osimertinib, respectively. comitant inhibition of MAPK kinase and EGFR could over- Attenuation of exon 19 deletion and T790M was confirmed in come osimertinib resistance. Thus, we identified a heteroge- both rociletinib-resistant cells; in addition, EGFR and KRAS neous acquired resistance mechanism for third-generation amplification was observed in RocR1 and RocR2, respectively. EGFR TKIs, providing insights into the development of novel Significant KRAS amplification was observed in the osimerti- treatment strategies.

Introduction present in approximately 32% of Asians and 7% of patients of other ethnicities with NSCLC (4). Somatic mutations in the EGFR tyrosine kinase inhibitors (TKI) of the first (gefitinib tyrosine kinase domain of EGFR, including in-frame deletions and erlotinib) and second (afatinib) generations are adminis- in exon 19 (exon 19 del) and the L858R point mutation in exon tered in patients with non–smallcelllungcancer(NSCLC)with 21, are common mutations accounting for 80% to 90% of activating mutations in EGFR as a first-line therapy. Several EGFR mutations in NSCLC (4). Unfortunately, most patients clinical trials have shown that the response rates to these administered EGFR TKIs develop acquired resistance after 1 to 2 therapies are 70% to 80%, and the progression-free survival years, accompanied by secondary mutations in EGFR,the timeis9to14months(1–3). Mutations in the EGFR gene are emergence of bypass signaling with MET and human EGFR 2 (HER2), and transformation to small-cell lung cancer (SCLC; – 1Department of Healthcare and Regulatory Sciences, Showa University School of refs. 2, 5 7). The most common resistance mutation is substi- Pharmacy, Tokyo, Japan. 2Advanced Cancer Translational Research Institute tution of methionine for threonine at position 790 (T790M), (formerly, Institute of Molecular Oncology), Tokyo, Japan. 3Center for Biotech- which is detected in approximately 50% to 60% of patients nology, Showa University, Tokyo, Japan. 4Division of Allergology and Respira- with NSCLC who exhibit acquired resistance to first- and tory Medicine, Department of Medicine, Showa University School of Medicine, second-generation EGFR TKIs (8, 9). To overcome acquired Tokyo, Japan. resistance due to the emergence of T790M mutations in EGFR, Note: Supplementary data for this article are available at Molecular Cancer third-generation or T790M-selective EGFR TKIs, including osi- Therapeutics Online (http://mct.aacrjournals.org/). mertinib (10) and rociletinib (11), have been developed; these Corresponding Author: Toshimitsu Yamaoka, Advanced Cancer Translational drugs inhibit activating mutations and T790M in EGFR in Research Institute, Showa University, 1-5-8 Hatanodai, Shinagawa-ku, Tokyo preclinical and clinical settings (12, 13). 142-8555, Japan. Phone: 813-3784-8146; Fax: 813-3784-2299; E-mail: Third-generation EGFR TKIs, such as rociletinib and osimer- [email protected] tinib, are oral drugs that covalently and irreversible bind to doi: 10.1158/1535-7163.MCT-18-0591 Cys797 in the ATP-binding site of EGFR. Osimertinib has 2018 American Association for Cancer Research. received approval for the treatment of patients with metastatic,

112 Mol Cancer Ther; 18(1) January 2019

Acquired Third-Generation EGFR-TKI Resistance Mechanisms

EGFR T790M mutation-positive NSCLC who have progressed vided by Boehringer-Ingelheim, and other inhibitors were during or after EGFR-TKI therapy. Although these third-gener- obtained from Selleck Chemicals. ation EGFR TKIs were active in patients with NSCLC bearing the EGFR T790M mutation in clinical trials (10, 14), development Establishment of afatinib-resistant PC-9 (AfaR) cells harboring of acquired resistance has still limited the efﬁcacy of this the EGFR T790M mutation with acquired resistance to treatment. Improving our understanding of resistance mechan- rociletinib and osimertinib isms to third-generation EGFR TKIs should increase survival To obtain cell lines with acquired resistance, PC-9 cells were ratesinpatientswithNSCLCharboringEGFR mutations and exposed to increasing concentrations of afatinib in the growth facilitate the development of novel therapeutics to overcome medium. Starting with a concentration that was one tenth of resistance. Recently, the acquisition of two secondary EGFR the half-maximal inhibitory concentration, the concentration mutations, C797S and L789I, has been reported in patients wasprogressivelyincreasedover10to12monthsto1mmol/L with NSCLC with recurrence during osimertinib treatment (15, afatinib. The resulting PC-9 afatinib-resistant cell line was 16). Although resistance mechanisms for third-generation designated AfaR cells (formerly AFR3 cells) and was maintained EGFR TKIs are not fully understood, bypass pathway activation continuously in culture medium containing 1 mmol/L afatinib. by MET or HER2 (17, 18), MAPK pathway activation by KRAS EGFR exon 19 del and T790M in exon 20 were evaluated in or MEK mutations (19) has been reported in clinical and AfaR cells, and the proliferation of these cells was effectively preclinical studies. suppressed by rociletinib and osimertinib treatment. Next, to We previously reported three distinct resistance mechanisms to obtain acquired rociletinib- or osimertinib-resistant cell lines afatinib in PC-9 lung adenocarcinoma cells, including cells bear- from AfaR cells, AfaR cells were exposed to increasing concen- ing EGFR T790M (designated AFR3 cells; refs. 20). AFR3 cells are trations of rociletinib or osimertinib in the growth medium. sensitive to third-generation EGFR TKIs, including osimertinib Starting with a dose that was approximately 1/10th of the IC50, and rociletinib. Therefore, in this study, we aimed to elucidate the the dosage was progressively increased over 10 to 12 months to mechanisms of resistance to the third-generation EGFR TKIs 1 mmol/L rociletinib or osimertinib. Two rociletinib-resistant rociletinib and osimertinib using PC-9 afatinib-resistant lung and two osimertinib-resistant AfaR cells were obtained and adenocarcinoma cells [designated AfaR (formerly named AFR3)] designated RocR1 or RocR2 for rociletinib resistance and OsiR1 by stepwise dose escalation. Our ﬁndings may lead to the devel- or OsiR2 for osimertinib resistance; these cells were maintained opment of new therapeutic options for patients with EGFR- continuously in a culture medium containing 1 mmol/L roci- mutant lung cancer. letinib or osimertinib, respectively (Fig. 1A).

Cell proliferation assay Materials and Methods Cell proliferation was measured using 3-(4,5-dimethylthiazol- Cell lines, antibodies, and reagents 2-yl)-2,5- diphenyltetrazolium bromide assays (Promega), as PC-9 human NSCLC cells were established from a previously described previously (21). Briefly, cells (5 102/well) were untreated patient and donated by K. Hayata (Tokyo Medical seeded in 96-well plates and incubated overnight, and assays College, Tokyo, Japan) during the 1980s. PC-9 cells were were performed on days 0, 1, 2, 3, 5, and 6. To inhibit cell cultured in RPMI1640 medium supplemented with 10% FBS, proliferation, 5 103 cells per well were seeded in 96-well plates penicillin (100 U/mL), and streptomycin (100 mg/mL) in a 5% and incubated overnight, followed by continuous exposure to the CO2 incubator at 37 C. Cells were passaged for less than 4 indicated concentrations of inhibitor for 72 hours. The optical months prior to renewal from frozen stocks. The cell lines used density at 570 nm (OD570) was then measured with a Powerscan in this study were authenticated by short tandem repeat anal- HT microplate reader (BioTek) and expressed as a percentage of ysis at the Japanese Collection of Research Bioresources Cell the value obtained from the control cells. We prepared 6 to 12 Bank in 2016 and tested for Mycoplasma using a MycoAlert replicates, and the experiments were repeated at least three times. Mycoplasma Detection Kit in 2017 (Lonza). The following Data were graphically displayed using GraphPad Prism version antibodies were purchased from Cell Signaling Technology: 7.0 software (GraphPad, Inc.). anti-total EGFR antibodies (#4267), anti–phospho-EGFR (Y1068) antibodies (#3777), anti-EGFR (E746-A750del) anti- Western blot analysis bodies (#2085), anti-total HER2 antibodies (#4290), anti-total Treated cells were washed twice with ice-cold PBS and lysed ERBB3 antibodies (#4754), anti-total ERBB4 antibodies with modified RIPA buffer consisting of 50 mmol/L Tris (pH 7.4), (#4795), anti-total insulin-like growth factor 1 receptor 150 mmol/L NaCl, 1 mmol/L ethylenediaminetetraacetic acid, (IGF1R) antibodies (#3018), anti–phospho-IGF1R antibodies 1% Nonidet P-40, 0.25% sodium deoxycholate, 0.1% SDS, and (#3024), anti-total MET antibodies (#8198), anti-total AKT 1.0% protease- and phosphatase-inhibitor cocktails (Sigma- antibodies (#9272), anti–phospho-AKT (S473) antibodies Aldrich). Cell suspensions were centrifuged for 5 minutes at (#9271), anti-total ERK1/2 antibodies (#9102), anti–phos- 1,200 rpm at 4C, and protein concentrations were determined pho-ERK1/2 antibodies (#4370), anti–phospho-MAPK kinase using bicinchoninic acid assays (Sigma-Aldrich). Equal amounts (MEK1/2) antibodies (#2338), anti-cleaved PARP antibodies of protein were mixed and boiled in Laemmli buffer, and samples (#5625), and anti–b-actin antibodies (#4970). Antibodies tar- were separated by SDS-PAGE electrophoresis on 8% to 12% gels, geting son of sevenless homolog 1 (SOS1; sc-17793), growth followed by transfer to polyvinylidene difluoride membranes. factor receptor–bound protein 2 (GRB2; sc-8034), KRAS (sc- Membranes were probed using the appropriate primary and 30), HRAS (sc-29), and NRAS (sc-31) were purchased from secondary antibodies (diluted according to the manufacturers' Santa Cruz Biotechnology. Cetuximab was purchased from instructions, 1:1,000–2,000), were treated with enhanced chemi- Showa University Hospital (Tokyo, Japan). Afatinib was pro- luminescence solution, and were then exposed to film. b-Actin

www.aacrjournals.org Mol Cancer Ther; 18(1) January 2019 113

Nakatani et al.

and at least one additional protein were used as loading controls. developed using anti-KRAS antibodies (F234; Santa Cruz All experiments were performed in triplicate. Biotechnology).

RAS pull-down assay RNA interference Pull-down assays were performed using a glutathione S- Nontargeting (N/T) short interfering (si) RNA (controls) and transferase fusion protein corresponding to the human RAS- SMARTpool siRNAs targeting EGFR (M003114), KRAS binding domain of RAF-1, which speciﬁcally binds to the GTP- (M005069), SOS1 (M005194), and GRB2 (M019220) were pur- bound form of RAS (EMD Millipore). Western blots were chased from Dharmacon. Cells were seeded in 6-well plates in

A PC-9

Afatinib: 1 µmol/L

AfaR

Osimertinib: 1 µmol/L Rociletinib: 1 µmol/L

OsiR1 OsiR2 RocR1 RocR2

B AfaR 140 Afatinib 120 Rociletinib 100 Osimertinib Figure 1. 80 Establishment of third-generation 60 – Survival (%) 40 EGFR TKI resistant cell lines from 20 afatinib-resistant PC-9 (AfaR) cells. 0 0.001 0.01 0.1 1 10 mmol/L A, Schematic model describing the process of generating rociletinib- and osimertinib-resistant cell lines from RocR1 RocR2 140 afatinib-resistant PC-9 (AfaR) cells. Afatinib 140 Afatinib 120 Rociletinib 120 Rociletinib B, Cells were seeded into 96-well ) 100 Osimertinib 100 Osimertinib plates at 5 103 cells/50 mL growth 80 80 medium/well, preincubated 60 60 Survival (%) Survival (% 40 40 overnight, followed by treatment with 20 20 afatinib, rociletinib, and osimertinib at 0 0 the indicated concentrations for 0.001 0.01 0.1 1 10 mmol/L 0.001 0.01 0.1 1 10 mmol/L 72 hours. MTT assays were then

performed, and OD570 values were 140 OsiR1 140 OsiR2 obtained. Data represent means Afatinib Afatinib 120 120 Rociletinib Rociletinib SEMs of data obtained from 6 to 12

) 100 Osimertinib 100 Osimertinib replicate wells. C, Western blot 80 80 analysis of basal levels of various 60 60 proteins in PC-9, AfaR, RocR1, RocR2, Survival (%) Survival (% 40 40 20 20 OsiR1, and OsiR2 cells. b-Actin was 0 0 included as a loading control. D, Cells 0.001 0.01 0.1 1 10 mmol/L 0.001 0.01 0.1 1 10 mmol/L were seeded into 96-well plates at 5 C 102 cells/100 mL/well. Following MTT PC-9 AfaR RocR1 RocR2 PC-9 AfaR OsiR1 OsiR2 assays, OD570 values were obtained on EGFR days 0, 1, 2, 3, 4, 5, and 6. D Cell proliferation AfaR EGFR (ex19del) OsiR1 2.0 HER2 OsiR2

ERBB3 1.5 * RocR1 * RocR2 ERBB4 1.0 OD570 MET 0.5 IGF1R 0.0 PTEN 01234567 Days AKT

ERK1/2

Actin

114 Mol Cancer Ther; 18(1) January 2019 Molecular Cancer Therapeutics

Acquired Third-Generation EGFR-TKI Resistance Mechanisms

RPMI1640 medium supplemented with 10% FBS but no anti- Results biotics. The following day, cells were transfected with 100 pmol/ Characteristics of AfaR clones with acquired resistance to third- well siRNA using Lipofectamine 2000 reagent (Life Technolo- generation EGFR TKIs gies), according to the manufacturer's instructions, and were We established two rociletinib- and osimertinib-resistant analyzed at 72 hours after transfection. AfaR cell lines over 10 to 12 months (Fig. 1A). Afatinib-resistant PC-9 cells, called AfaR cells, harbored 15-bp deletions in exon Real-time RT-PCR 19 (exon 19 del) and the T790M mutation in EGFR,andthe Total RNA was isolated from cells using an RNeasy Mini Kit third-generation EGFR TKIs rociletinib and osimertinib effec- (Qiagen) according to the manufacturer's instructions. cDNA was tively suppressed proliferation in AfaR cells (Fig. 1B). Notably, then synthesized from each isolated RNA sample using random 6- RocR1 and RocR2 cells were partially sensitive to afatinib mers and an RT-PCR Kit (TaKaRa Bio). In addition, genomic DNA compared with AfaR cells, whereas OsiR1 and OsiR2 cells (gDNA) was extracted from cells using a QIAamp DNA Mini Kit exhibited resistance to afatinib and rociletinib (Fig. 1B). The (Qiagen). cDNA and genomic DNA samples were amplified and EGFR E746-A750 del in exon 19 was decreased in RocR1 and analyzed using Power SYBR Green PCR master mix (Applied RocR2 cells compared with that in AfaR cells, although similar Biosystems) and a fluorescence-based RT-PCR detection system expression levels were observed in OsiR1 and OsiR2 cells (GeneAmp 5700; Applied Biosystems). Specific primer sets are (Fig. 1C) and the other receptors HER2, ERBB3, ERBB4, MET, shown in our previous report (20, 22). and IGF1R were not increased in these RocR and OsiR cells. With regard to cell proliferation, RocR1 and RocR2 cells showed EGFR and KRAS sequence analysis significantly slower proliferation than parental AfaR cells. How- Exons 19–21 of the EGFR gene and exons 2–4oftheKRAS ever, the proliferation rates of OsiR1 and OsiR2 cells were gene were amplified from genomic DNA via PCR. Products similar to that of AfaR cells (Fig. 1D). Thus, these findings were purified and sequenced by FASMAC. The primers used for suggested that the exon 19 del activating mutation in EGFR was PCR and sequencing analysis are shown in in our previous greatly decreased in RocR1 and RocR2 cells but not in OsiR1 report (20, 22). and OsiR2 cells (Table 1).

KRAS EGFR EGFR allele quantification by digital PCR analysis RocR1 and RocR2 cells exhibited and gene fi EGFR Droplet digital PCR was performed using an LBx Probe for the ampli cation with loss of mutations (exon 19 del EGFR exon-19 del or T790M mutation, and samples were quan- and T790M) EGFR tified using a QX200 droplet reader (Bio-Rad). The distributions Direct sequencing of exons 19 and 20 revealed partial loss of wild-type (WT), exon-19 del, and/or T790M alleles were of E746-A750del and T790M alleles (Supplementary Fig. S1A). EGFR – determined by Riken Genesis. Allelic quantitative distribution analyses using mutation specific droplet digital PCR probes detected higher proportions of mutant EGFR alleles for exon 19 del and T790M than WT alleles in Xenograft mouse studies parental AfaR cells. Conversely, RocR1 and RocR2 cells exhibited Suspensions of 5 106 cells were injected subcutaneously significantly higher proportions of WT EGFR alleles compared into the flanks of 6- to 8-week-old female SCID mice. The care with the parental AfaR cells (Table 1). Because the third-genera- and treatment of experimental animals were performed in tion EGFR TKIs rociletinib and osimertinib were EGFR-mutation accordance with institutional guidelines. Mice were random- selective inhibitors, RocR1 and RocR2 cells, with increased expres- ized (n ¼ 5) once the mean tumor volume reached approxi- sion of WT EGFR, could prevent inhibition of EGFR phosphor- mately 230 to 300 mm3. Drugs were administered once daily by ylation. As shown in Fig. 2A, rociletinib completely suppressed oral gavage. Osimertinib (5 mg/kg/day) was suspended in 0.5% EGFR phosphorylation and downstream signaling through AKT 2-hydroxyethyl cellulose/H O. Tumors were measured twice 2 and ERK1/2 at 0.1 mmol/L. These changes led to induction of weekly using calipers, and tumor volumes were calculated apoptosis, as observed by increases in cleavable PARP levels in using the following formula: length2 width 0.5. According AfaR cells. However, in RocR1 and RocR2 cells, rociletinib did not to institutional guidelines, mice were sacrificed when their inhibit EGFR phosphorylation or AKT and ERK1/2 activation; tumor volume reached 2,000 mm3. All animal experiments therefore, induction of apoptosis was not observed. AfaR, RocR1, were approved by the Institutional Animal Care and Use and RocR2 cells were exposed to increasing concentrations of Committee of Showa University (Permit Number: 57001; afatinib, and afatinib did not inhibit EGFR phosphorylation or Tokyo, Japan).

FISH Table 1. The proportions of EGFR exon 19 del and T790M alleles in PC-9, AfaR, KRAS and chromosome 12 FISH analyses were performed by RocR1, RocR2, OsiR1, and OsiR2 cells, as measured by digital PCR analysis Chromosome Science Lab Inc. using the KRAS/Chr12cen probe. ddPCR analysis Mutation rates (%): mut/(mut þ wt) Ex 19 del T790M Statistical analysis PC-9 86.2 (5062/5869) 0 (0/4992) Data are presented as means SEMs and were analyzed using AfaR 94.9 (9079/9563) 72.6 (6658/9166) GraphPad Prism version 7.0 software (GraphPad, Inc.). Statistical RocR1 58.8 (8628/14668) 11.5 (1013/8842) signiﬁcance was evaluated by two-tailed Student t tests and, unless RocR2 52.1 (25366/48725) 1.98 (504/25392) otherwise noted, differences with P values of less than 0.05 were OsiR1 92.7 (10058/10846) 75.9 (7163/9437) considered statistically signiﬁcant. OsiR2 91.5 (10014/10949) 79.9 (7849/9821)

www.aacrjournals.org Mol Cancer Ther; 18(1) January 2019 115

Nakatani et al.

A B AfaR RocR1 RocR2 AfaR RocR1 RocR2

Afatinib: µmol/L 0 0.01 0.1 1 0 0.01 0.1 1 0 0.01 0.1 1 Rociletinib: µmol/L 0 0.01 0.1 1 0 0.01 0.1 1 0 0.01 0.1 1 P-EGFR P-EGFR

EGFR EGFR

P-AKT P-AKT

AKT AKT

P-ERK1/2 P-ERK1/2

ERK1/2 ERK1/2

Cle PARP Cle-PARP

Actin Actin

F RocR1 C D siRNA N/T KRAS PC-9 AfaR RocR1 RocR2 Afatinib: 1 µmol/L --++ PC-9 AfaR RocR1 KRAS Pull-down: Raf-RBD P-EGFR IB: KRAS NRAS EGFR Input IB: KRAS HRAS P-AKT

AKT Actin E RocR1 1.5 * P-ERK1/2

1.0 ERK1/2

OD570 Cle-PARP 0.5

KRAS 0.0 µ Afatinib: 1 mol/L - + - + Actin siRNA N/T KRAS

H AfaR RocR1 RocR2

G siRNA EGFR gDNA N/T EGFR N/T EGFR N/T EGFR I * EGFR 10 1.5 P-EGFR * *

P-AKT 1.0 * 5 AKT OD570 0.5

Copy number fold change P-ERK1/2 0 ERK1/2 PC-9 AfaR 0.0 RocR1 RocR2 siRNA N/T EGFR N/T EGFR N/T EGFR Cle-PARP AfaR RocR1 RocR2

Actin

116 Mol Cancer Ther; 18(1) January 2019 Molecular Cancer Therapeutics

Acquired Third-Generation EGFR-TKI Resistance Mechanisms

AKT and ERK1/2 activation in AfaR cells. In RocR1 and RocR2 ylation and AKT activation in a concentration-dependent man- cells, afatinib effectively inhibited EGFR phosphorylation, and ner, whereas ERK1/2 was still activated. Therefore, addition of the downstream signaling was inhibited especially in RocR2 cells. selumetinib, a MEK inhibitor, to block ERK1/2 activation Moreover, in RocR2 cells, induction of apoptosis was observed, enhanced the induction of apoptosis (Supplementary Fig. indicating that afatinib sensitivity was recovered (Fig. 2B). Afati- S2C). Taken together, these results showed that loss of the nib inhibited activating mutations and WT EGFR to a greater rociletinib target (mutated EGFR [exon 19 del/T790M]) and extent than EGFR T790M, and third-generation EGFR TKIs selec- KRAS amplification caused rociletinib resistance in RocR1 cells; tively inhibited activating EGFR mutations and T790M, but combination of a MEK inhibitor and afatinib could overcome weakly inhibited WT EGFR (12, 13). Afatinib was administered this resistance. to RocR1 cells at increasing concentrations, leading to concen- EGFR amplification and overexpression were observed in tration-dependent inhibition of EGFR and attenuation of down- RocR2 cells (Figs. 1C and 2G). When EGFR was knocked down stream AKT and ERK1/2 phosphorylation. However, this suppres- by siRNA in AfaR cells, downstream AKT and ERK1/2 activation sion of AKT and ERK1/2 phosphorylation was modest (Fig. 2B) as was inhibited; furthermore, the induction of apoptosis was compared with the levels observed following rociletinib admin- observed (Fig. 2H), and cell proliferation was inhibited (Fig. istration to parental AfaR cells (Fig. 2A). Furthermore, elevated 2I). Moreover, RocR2 cells required EGFR expression and phos- KRAS gene copy numbers and KRAS protein overexpression were phorylation for survival, similar to AfaR cells. Therefore, afatinib observed in RocR1 cells, but not in PC-9, AfaR, and RocR2 cells administration inhibited cell proliferation, and addition of cetux- (Supplementary Fig. S2A). In addition, each of the cell lines imab enhanced suppression of cell proliferation, although cetux- exhibited similar levels of NRAS and HRAS protein expression imab alone did not inhibit cell proliferation (Supplementary Fig. (Fig. 2C). Pull-down assays detected elevated KRAS activity in S2D). In RocR2 cells, afatinib inhibited the phosphorylation of RocR1 cells (Fig. 2D) but not constitutive mutational activation of EGFR in a concentration-dependent manner and downstream of KRAS, which is indicative of WT KRAS exons 2–4 (Supplementary AKT and ERK1/2, and addition of cetuximab enhanced the Fig. S1B; ref. 23). inhibition of EGFR/AKT/ERK1/2 and stimulated apoptosis (Sup- As reported previously, the relationship between amplification plementary Fig. S2E). Other studies have shown that ligands of and mutation of KRAS may be mutually exclusive (24). Although EGFR increased, leading to rociletinib resistance; however, in RocR1 cells did not show reduced cell proliferation when KRAS RocR2 cells, EGFR ligands were not increased compared with levels were downregulated by siRNA, additional treatment with that in AfaR cells (Supplementary Fig. S2F; ref. 27). afatinib in KRAS-knockdown cells significantly reduced cell proliferation (Fig. 2E). The attenuation of KRAS by siRNA reduced KRAS amplification and overexpression were observed in OsiR1 ERK1/2 activation, but AKT activation was modestly decreased. and OsiR2 cells Although afatinib inhibited EGFR phosphorylation and down- In OsiR1 and OsiR2 cells, the distribution of EGFR mutations stream of AKT and ERK1/2 activation, when KRAS was down- (exon 19 del and T790M) was similar to that in AfaR cells regulated by siRNA in RocR1 cells, AKT and ERK1/2 activation was (Table 1). AfaR cells showed reduced phosphorylation of EGFR markedly inhibited, and apoptosis was obviously induced (Fig. and activation of AKT and ERK1/2, and apoptosis occurred in 2F). This suggested that EGFR signaling may still be activated for response to osimertinib exposure in a concentration-dependent survival in RocR1 cells. manner (Fig. 3A). In OsiR1 and OsiR2 cells, EGFR phosphoryla- MAPK signaling is a well-known KRAS effector pathway, and tion was sustained following osimertinib exposure; downstream targeting this pathway using MEK inhibitors has been shown to AKT and ERK1/2 activation was also sustained (Fig. 3A). More- have limited activity in patients with KRAS mutations (25). In over, in a xenograft model, osimertinib reduced the growth of RocR1 cells, selumetinib, a MEK1/2 inhibitor (26), did not xenograft tumors derived from AfaR cells, but not the growth of suppress cell proliferation, and afatinib alone suppressed cell tumors derived from OsiR1 and OsiR2 cells (Fig. 3B). In addition, proliferation only modestly. In combination with selumetinib, KRAS amplification and overexpression were observed in OsiR1 afatinib enhanced inhibition of cell proliferation (Supplemen- and OsiR2 cells (Fig. 3C and D). KRAS amplification in OsiR1 and tary Fig. S2B). Moreover, afatinib suppressed EGFR phosphor- OsiR2 cells was significantly increased by 300- and 30-fold

Figure 2. RocR1 and RocR2 exhibited KRAS and EGFR gene ampliﬁcation, respectively, with loss of exon 19 deletion and T790M in EGFR following acquisition of rociletinib resistance. A and B, AfaR, RocR1, and RocR2 cells were exposed to different concentrations of rociletinib (A) or afatinib (B) for 24 hours, and cell lysates were subjected to Western blot analysis. b-Actin was included as a loading control. Immunoblot analysis of PARP cleavage was used to screen for the induction of apoptosis. Blots are representative of three independent experiments. C, The expression levels of KRAS, NRAS, and HRAS were determined by Western blot analysis. b-Actin was included as a loading control. D, Western blot analysis of KRAS expression in PC-9, AfaR, and RocR1 cells and correlations with the results of RAF-RBD pull-down assays. E, RocR1 cells were transfected with nontargeting (N/T) siRNA or siRNA directed against KRAS, and transfected cells were then reseeded in the presence or absence of 1 mmol/L afatinib. After a 72-hour incubation, MTT assays were performed, and OD570 values were obtained. Data represent the means SEMs of the data obtained from 6 replicate wells; , P < 0.01. F, RocR1 cells were transfected with N/T siRNA or siRNA directed against KRAS for 48 hours and then treated with 1 mmol/L afatinib for 24 hours. KRAS knockdown was determined by Western blot analysis. Lysates were subjected to immunoblot analysis using the indicated primary antibodies. b-Actin was included as a loading control. Immunoblot analysis of PARP cleavage was used to screen for the induction of apoptosis. G, Increases in EGFR copy numbers were detected using quantitative PCR analysis of genomic DNA extracted from PC-9, AfaR, RocR1, and RocR2 cells; , P < 0.01. H, AfaR, RocR1, and RocR2 cells were transfected with N/T siRNA or siRNA directed against EGFR, and EGFR knockdown was determined by Western blot analysis. Lysates were then subjected to immunoblot analysis using the indicated primary antibodies; b-actin was included as a loading control. Immunoblot analysis for PARP cleavage was used to screen for the induction of apoptosis. Blots are representative of three independent experiments. I, AfaR, RocR1, and RocR2 cells were transfected with N/T siRNA or siRNA directed against EGFR, incubated for 72 hours, and subjected to MTT assays. Data (OD570 values) represent the means SEMs of the data obtained from 6 replicate wells; , P < 0.01.

www.aacrjournals.org Mol Cancer Ther; 18(1) January 2019 117

Nakatani et al.

B 2,000 A AfaR OsiR1 OsiR2 AfaR: Control )

3 AfaR: Osimertinib Osimertinib: µmol/L 0 0.01 0.1 1 0 0.01 0.1 1 0 0.01 0.1 1 1,500

P-EGFR 1,000 * EGFR 500 Tumor volume (mm P-AKT 0 0102030 AKT Days after administration

P-ERK1/2 OsiR1: Control

) 2,000 OsiR1: Osimertinib 3 ERK1/2 OsiR2: Control 1,500 OsiR2: Osimertinib Cle-PARP 1,000

Actin 500 Tumor volume (mm

0 0102030 Days after administration C D PC-9 AfaR OsiR1 OsiR2 KRAS gDNA * KRAS * E 300 PC-9 AfaR OsiR1 OsiR2 NRAS Pull-down: Raf-RBD 200 IB: KRAS

HRAS Input 100 IB: KRAS Actin Copy number fold change 0

PC-9 AfaR OsiR1 OsiR2

F OsiR1 G OsiR2 OSI906: 1 µmol/L OSI906: 1 µmol/L

Selumetinib: µmol/L 0 0.01 0.1 1 0 0.01 0.1 1 Selumetinib: µmol/L 0 0.01 0.1 1 0 0.01 0.1 1

P-EGFR P-EGFR

EGFR EGFR

P-IGF1R P-IGF1R

IGF1R IGF1R

P-AKT P-AKT

AKT AKT

P-MEK1/2 P-MEK1/2

P-ERK1/2 P-ERK1/2

ERK1/2 ERK1/2

Cle-PARP Cle-PARP

Actin Actin

Figure 3. Acquired increases in KRAS expression in OsiR1 and OsiR2 cells, which were sensitive to cotreatment with a MEK inhibitor and an IGF1R inhibitor. A, AfaR, OsiR1, and OsiR2 cells were exposed to osimertinib at different concentrations for 24 hours, and cell lysates were subjected to Western blot analysis using specific antibodies. b-Actin was included as a loading control. Immunoblot analysis of PARP cleavage was used to screen for the induction of apoptosis. Blots are representative of three independent experiments. B, Female SCID mice were injected with AfaR, OsiR1, or OsiR2 cells. Once tumors reached a volume of approximately 230 to 300 mm3, 5 mice per group were randomized into groups and treated with or without 5 mg/kg osimertinib daily via orogastric gavage for 25 days. Tumors were measured twice weekly with calipers. Data in the figure represent mean tumor volumes SEMs; , P < 0.01. C, The expression levels of KRAS, NRAS, and HRAS were determined by Western blot analysis. b-Actin was included as a loading control. D, KRAS gene copy numbers were detected via quantitative PCR analysis of genomic DNA extracted from PC-9, AfaR, OsiR1, and OsiR2 cells; , P < 0.01. E, Western blot analysis of KRAS expression in PC-9, AfaR, OsiR1, and OsiR2 cells and correlations with the results of RAF-RBD pull-down assays. OsiR1 (F) and OsiR2 (G) cells were exposed to selumetinib at different concentrations in the presence or absence of 1 mmol/L OSI906 for 24 hours, and cell lysates were subjected to Western blot analysis using specific antibodies. b-Actin was included as a loading control. Immunoblot analysis of PARP cleavage was used to screen for the induction of apoptosis. Blots are representative of three independent experiments.

118 Mol Cancer Ther; 18(1) January 2019 Molecular Cancer Therapeutics

Acquired Third-Generation EGFR-TKI Resistance Mechanisms

OsiR1: Osimertinib (µmol/L) in medium OsiR2: Osimertinib (µmol/L) in medium A PC-9 AfaR 0.08 0.1 0.2 0.4 0.60.8 1 C PC-9 AfaR0.060.08 0.1 0.2 0.4 0.80.61 KRAS KRAS

P-EGFR P-EGFR

EGFR EGFR

P-AKT P-AKT

AKT AKT

P-ERK1/2 P-ERK1/2

ERK1/2 ERK1/2 Actin Actin B D KRAS gDNA KRAS gDNA

400 * * * 150 * * 300 * * * 100 * 200 * * 50 * * 100 * * Copy number fold change 0 Copy number fold change 0 1 1 0.1 0.2 0.4 0.6 0.8 0.1 0.2 0.4 0.6 0.8 0.06 0.08 PC-9 AfaR 0.08 PC-9 AfaR Osimertinib: µmol/L Osimertinib: µmol/L OsiR1 OsiR2 E OsiR1 G OsiR2 PC-9 AfaR OsiR1 F2W F1M F2M PC-9 AfaR OsiR2 F2W F1M F2M

KRAS KRAS

P-EGFR P-EGFR

EGFR EGFR

P-AKT P-AKT

AKT AKT

P-ERK1/2 P-ERK1/2

ERK1/2 ERK1/2

Actin Actin

F H KRAS gDNA 400 KRAS gDNA 30 e * * * 300 * * 20 200 * * 10 100 Copy number fold change Copy number fold chang 0 0

PC-9 AfaR F2W F1M F2M AfaR F2W F1M F2M OsiR1 PC-9 OsiR2 OsiR1 OsiR2

Figure 4. Acquired increases in KRAS and changes in KRAS levels following removal of osimertinib in OsiR1 and OsiR2 cells. A and C, AfaR cells were cultured in the presence of osimertinib with increasing concentrations from 0.01 to 1 mmol/L for 10 to 12 months to establish OsiR1 (A) and OsiR2 (C) cells. OsiR1 and OsiR2 cell lysates were generated and subjected to Western blot analysis using the indicated primary antibodies. b-Actin was included as a loading control. B and D, Increases in KRAS gene copy numbers were detected via quantitative PCR analysis of genomic DNA extracted from PC-9, AfaR, OsiR1 (B), and OsiR2 (D) cells; , P < 0.01, compared with AfaR cells. OsiR1 (E) and OsiR2 (G) cells were cultured in the absence of osimertinib for 2 weeks (F2W), 1 month (F1M), or 2 months (F2M), and lysates were generated and subjected to Western blot analysis using the indicated primary antibodies. b-Actin was included as a loading control. KRAS copy numbers were evaluated in PC-9, AfaR, OsiR1 (F), and OsiR2 (H) cells (F2W, F1M, and F2M) by real-time PCR analysis; , P < 0.01, compared with AfaR cells. www.aacrjournals.org Mol Cancer Ther; 18(1) January 2019 119

Nakatani et al.

OsiR1-F2M A OsiR1-F2M B Osimertinib: 1 µmol/L Osimertinib: µmol/L 0 0.01 0.1 0.3 1 Selumetinib: µmol/L 0 0.01 0.1 1 0 0.01 0.1 1

P-EGFR P-EGFR

EGFR EGFR P-AKT P-AKT AKT

P-ERK1/2 AKT

ERK1/2 P-ERK1/2

Cle-PARP ERK1/2 Actin Cle-PARP

Actin C OsiR1-F2M 120 Osimertinib 100 Selumetinib Selumetinib 80 + Osimertinib 1 µmol/L OsiR2-F2M 60 D 120 Afatinib 40 Survival (%) 100 Osimertinib 20 80 0 0.001 0.01 0.1 1 10 mmol/L 60

40 Survival (%) 20 OsiR2-F2M 0 0.001 0.01 0.1 1 10 µmol/L E Osimertinib: µmol/L Afatinib: µmol/L

0 0.001 0.01 0.1 1 0 0.001 0.01 0.1 1 P-EGFR

EGFR F 2,000 OsiR2-F2M: Control

P-AKT )

3 OsiR2-F2M: Osimertinib 1,500 AKT

1,000 P-ERK1/2 *

ERK1/2 500 Tumor volume (mm

Cle-PARP 0 0102030 Actin Days after administration

120 Mol Cancer Ther; 18(1) January 2019 Molecular Cancer Therapeutics

Acquired Third-Generation EGFR-TKI Resistance Mechanisms

compared with that in AfaR cells. In exons 2–4 of the KRAS gene, the levels of KRAS increased, although basal expression of EGFR no mutations were identified (Supplementary Fig. S1B). Notably, was maintained in both OsiR1 and OsiR2 cells (Fig. 4A and C). KRAS activation was also increased markedly (Fig. 3E). FISH These findings suggested that KRAS amplification occurred via analysis showed that KRAS amplification was observed as a adaptation to osimertinib, rather than selection. homogeneously stained region in OsiR1 and OsiR2 cells (Sup- After establishment of osimertinib-resistant cells, OsiR1 and plementary Fig. S3A). OsiR2 cells were cultured in the presence of 1 mmol/L osimertinib for at least 2 to 3 months. OsiR1 and OsiR2 cells were then The combination of a MEK inhibitor and an IGF1R cultured in osimertinib-free medium. Approximately 2 months inhibitor overcame KRAS amplification in later, the amplified KRAS expression was sustained in OsiR1 cells osimertinib-resistant cells (Fig. 4E and F); however, in OsiR2 cells, KRAS expression was Using siRNA against EGFR or KRAS, knockdown of EGFR or decreased to the level of the control parental AfaR cells (Fig. 4G KRAS reduced AKT and ERK1/2 activation, suggesting that these and H). In both OsiR1 and OsiR2 cells, EGFR phosphorylation downstream signals were induced by either EGFR or KRAS was increased markedly after osimertinib removal only for (Supplementary Fig. S3B). Interestingly, IGF1R phosphoryla- 2 weeks; otherwise, basal EGFR expression was maintained tion was increased in KRAS-knockdown OsiR1 and OsiR2 cells. (Fig. 4E and G). Other studies have shown that KRAS attenuation leads to negative feedback activation of PI3K via insulin receptor sub- Acquired resistance to osimertinib was sustained in OsiR1 strate, and cells are then sensitive to the combination of IGF1R cells but reversed in OsiR2 cells, according to KRAS KRAS and MEK inhibitors in the presence of activating muta- expression, after 2 months of osimertinib withdrawal tions (28). When treated with the MEK inhibitor selumetinib, After approximately 2 months of culture in osimertinib-free both OsiR1 and OsiR2 cells showed inhibition of ERK1/2 growth medium, OsiR1 and OsiR2 cells exhibited EGFR muta- activation in a concentration-dependent matter. Activation of tions (exon 19 del/T790M) similar to those of AfaR cells and to AKT was decreased only modestly but was completely inhibited OsiR1 and OsiR2 cells without osimertinib withdrawal (Sup- by addition of the IGF1R inhibitor OSI906, and induction of plementary Table S1). Interestingly, EGFR phosphorylation was apoptosis was enhanced (Fig. 3F and G). Cell proliferation was inhibited by osimertinib in a concentration-dependent matter inhibited by selumetinib in both OsiR1 and OsiR2 cells, and in OsiR1 cells at 2 months after osimertinib withdrawal. How- further inhibition was observed following addition of OSI906 ever, the downstream signals of AKT and ERK1/2 were inhibited (Supplementary Fig. S3C and S3D). modestly, suggesting that overexpression of KRAS prevented complete inhibition (Fig. 5A). Next, for further inhibition of Acquired KRAS amplification and changes in KRAS ERK1/2 activation, the MEK inhibitor selumetinib was admin- amplification following removal of osimertinib in OsiR1 and istered to OsiR1 cells subjected to osimertinib withdrawal for 2 OsiR2 cells months and then treated with 1 mmol/L osimertinib. The results While establishing osimertinib-resistant cells, AfaR cells were showed that downstream AKT and ERK1/2 activation was exposed to osimertinib by increasing the concentration in completely inhibited, and apoptosis induction was observed growth medium; KRAS and EGFR expression levels were then following selumetinib treatment in a concentration-dependent determined. In both OsiR1 and OsiR2 cells, KRAS expression manner (Fig. 5B). The proliferation of OsiR1 cells after 2 started to change as the concentration of osimertinib increased months of osimertinib withdrawal was suppressed by the from 0.06–0.08 to 1 mmol/L. Although the acquired increase in combination of selumetinib and osimertinib (Fig. 5C). In KRAS was maximized at 0.4 to 0.6 mmol/L osimertinib in both OsiR2 cells after 2 months of osimertinib withdrawal, OsiR1 and OsiR2 cells, amplified KRAS expression was sustained decreased KRAS expression was observed after removal of in OsiR1 cells and showed a 300-fold increase compared with osimertinib, and cell proliferation was suppressed by osimer- that in AfaR cells (Fig. 4A and B). However, in OsiR2 cells, KRAS tinib but not by afatinib (Fig. 5D). Osimertinib also inhibited expression was decreased from a 120-fold increase to only a 30- EGFR phosphorylation and the activation of AKT and ERK1/2, fold increase compared with that in AfaR cells (Fig. 4C and D). and induction of apoptosis was clearly observed; however, Conversely, phospho-EGFR expression levels were decreased as these effects were not noted when cells were exposed to afatinib

Figure 5. Acquired resistance to osimertinib was sustained in OsiR1 cells but reversed in OsiR2 cells, according to KRAS expression, after 2 months of osimertinib withdrawal. A, OsiR1-F2M cells were exposed to graded concentrations of osimertinib for 24 hours, and cell lysates were subjected to immunoblot analysis using speciﬁc antibodies. b-Actin was included as a loading control. Immunoblot analysis of PARP cleavage was used to screen for the induction of apoptosis. B, OsiR1-F2M cells were exposed to selumetinib at different concentrations for 24 hours in the presence or absence of 1 mmol/L osimertinib, and cell lysates were subjected to Western blot analysis using the indicated primary antibodies. b-Actin was included as a loading control. Immunoblot analysis of PARP cleavage was used to screen for the induction of apoptosis. Blots are representative of three independent experiments. C, OsiR1-F2M cells were seeded into 96-well plates at 5 103 cells/50 mL growth medium/ well, preincubated overnight, and treated with osimertinib or selumetinib in the presence or absence of 1 mmol/L osimertinib at the indicated concentrations for 72 hours. An MTT assay was performed, and OD570 values were measured. Data represent the means SEMs for 6 to 12 wells. D, OsiR2-F2M cells were seeded into 96-well plates at 5 103 cells/50 mL growth medium/well, preincubated overnight, and treated with osimertinib or afatinib at the indicated concentrations for 72 hours. MTT assays were performed, and OD570 values were measured. Data represent the means SEMs for 6 to 12 wells. E, OsiR2-F2M cells were exposed to different concentrations of osimertinib or afatinib for 24 hours, and cell lysates were subjected to immunoblot analysis using the indicated primary antibodies. b-Actin was included as a loading control. Immunoblot analysis of PARP cleavage was used to screen for the induction of apoptosis. F, Female SCID mice were injected with OsiR2-F2M cells. Once tumors reached a volume of approximately 230 to 300 mm3, 5 mice per group were treated with or without 5 mg/kg osimertinib daily via orogastric gavage for 25 days. Tumors were measured twice weekly with calipers. Data in the ﬁgure represent mean tumor volumes SEMs; , P < 0.01.

www.aacrjournals.org Mol Cancer Ther; 18(1) January 2019 121

Nakatani et al.

OsiR1 C B A siRNA AfaR OsiR1 OsiR1-F2M N/T GRB2

P-EGFR AfaR OsiR1 OsiR1-F2M IB: EGFR

P-EGFR IB:GRB2 EGFR IP: GRB2

EGFR P-AKT IB: GRB2 AKT Actin WCL IB:Actin

P-ERK1/2

ERK1/2

D OsiR1 OsiR1-F2M E GRB2

siRNA N/T SOS1 N/T SOS1 AfaR OsiR1 OsiR1-F2M Actin Osimertinib: 1 µmol/L −−+ −+ + P-EGFR P-EGFR EGFR

P-AKT EGFR IP: SOS1 AKT SOS1 AfaR OsiR1 OsiR1-F2M P-ERK1/2 Osimertinib: 1 µmol/L −−+ −+ + F ERK1/2 P-EGFR

SOS1 EGFR

Actin GRB2

WCL SOS1 G OsiR1 KRAS siRNA N/T GRB2 Osimertinib: 1 µmol/L − + −+− + −+ Actin Selumetinib: 1 µmol/L −+− + −+− + Pull-down: P-EGFR KRAS Raf-RBD EGFR H P-AKT * * AKT 2.0 2.0 *

1.5 1.5 0 P-ERK1/2 0 1.0 1.0 OD57 ERK1/2 OD57 0.5 0.5 Cle-PARP 0.0 0.0 Osimertinib: µmol/L 0 1 0 0.01 0.1 1 0 1 0 0.01 0.1 1 GRB2 Selumetinib: µmol/L 00 11111 00 11111 siRNA N/T GRB2 Actin OsiR1

122 Mol Cancer Ther; 18(1) January 2019 Molecular Cancer Therapeutics

Acquired Third-Generation EGFR-TKI Resistance Mechanisms

(Fig. 5E). Thus, the distribution of EGFR T790M mutations was to that in OsiR1 cells without osimertinib withdrawal (Fig. still dominant in OsiR2 cells subjected to osimertinib with- 6F). Moreover, activated EGFR signals induced Shc, Gab1, and drawal for 2 months. Furthermore, xenograft tumors of these Gab2, except GRB2/SOS1 in OsiR1 cells, after osimertinib cells were sensitive to osimertinib administration (Fig. 5F), and withdrawal for 2 months (Supplementary Fig. S4B). Using OsiR2 cells were resensitized to osimertinib after 2 months of siRNA against GRB2, we found that downregulation of GRB2 osimertinib withdrawal. Taken together, these findings sup- decreased EGFR phosphorylation and that osimertinib inhib- ported that EGFR phosphorylation was inhibited by osimerti- ited EGFR phosphorylation in this context. Moreover, the nibinbothOsiR1andOsiR2cellsafter2monthsofosimerti- combination of osimertinib and selumetinib enhanced the nib withdrawal. induction of apoptosis in OsiR1 cells (Fig. 6G). The proliferation of OsiR1 cells was significantly inhibited by the com- Association of the phospho-EGFR and GRB2/SOS complex bination of osimertinib and selumetinib in GRB2-knockdown promoted KRAS activity and resistance to osimertinib OsiR1 cells (Fig. 6H). in OsiR1 cells In OsiR1 cells, the expression of EGFR was similar to that in AfaR cells and OsiR1 cells after osimertinib withdrawal for Discussion 2 months; however, phosphorylation of EGFR was strongly In this study, we demonstrated that the third-generation EGFR suppressed(Fig.6A).Wenextattemptedtodeterminethe TKIs rociletinib and osimertinib showed novel resistance mechan- associations between adapter proteins and EGFR. EGFR and isms in acquired afatinib-resistant AfaR cells carrying exon 19 del GRB2 were associated strongly in OsiR1 cells, but not in AfaR and T790M in EGFR. Gene amplification of KRAS and EGFR was and OsiR1 cell after osimertinib withdrawal for 2 months (Fig. observed in rociletinib-resistant RocR1 and RocR2 cells, with loss 6B). Then, using siRNA against GRB2 to dissociate the EGFR/ of exon 19 del/T790M in EGFR. Moreover, osimertinib resistance GRB2 complex, EGFR phosphorylation was attenuated in was induced by KRAS amplification in OsiR1 and OsiR2 cells OsiR1 cells, and the activation of downstream AKT and carrying exon 19 del/T790M in EGFR, similar to parental AfaR ERK1/2 signaling was also inhibited (Fig. 6C), suggesting that cells. Notably, in OsiR1 cells, withdrawal of osimertinib treatment the EGFR/GRB2 association may have a critical role in osimer- for 2 months resulted in an increase in EGFR phosphorylation; tinib resistance. osimertinib partially inhibited EGFR phosphorylation and, sub- The GRB2/SOS1 complex has been shown to activate RAS as sequently, KRAS activity. This result could be explained by the a guanine nucleotide exchange factor (GEF; ref. 29). To dissociation of the phosphorylated EGFR and GRB2/SOS1 com- determine the role of SOS1 in OsiR1 cells, we used siRNA plex in OsiR1 cells after osimertinib withdrawal for 2 months and against SOS1 (Fig. 6D). SOS1 downregulation decreased the partial activation of EGFR–KRAS signals. Therefore, the com- ERK1/2 phosphorylation, but not AKT phosphorylation. bination of osimertinib plus a MEK inhibitor was able to over- Moreover, the GRB2/SOS1 complex bound to phospho-EGFR come the resistance in OsiR1 cells after osimertinib withdrawal for (Y1068), even following osimertinib exposure, and osimerti- 2 months. Furthermore, in OsiR2 cells, KRAS expression was nib induced the dissociated of the phospho-EGFR and GRB2/ decreased to the level in parental AfaR cells; hence, sensitivity to SOS1 complex in AfaR cells; no association was observed in osimertinib was recovered completely. OsiR1 cells after osimertinib withdrawal for 2 months (Fig. The acquired resistance mechanisms to third-generation EGFR 6E; Supplementary Fig. S4A). Thus, these findings suggested TKIs have been reported in both clinical and preclinical settings. that the association between the phospho-EGFR (Y1068) and Such mechanisms include C797S (15), G796D (30), and L798I GRB2/SOS1 complex led to sustained EGFR phosphorylation, (16) mutations in EGFR, suggesting that these mutations may even following osimertinib treatment, and maintained KRAS decrease the binding activity to EGFR. Bypass signaling pathways activation. KRAS activity was partially inhibited by osimertinib function through activation of HER2, MET, and FGFR or upre- treatment in OsiR1 cells after osimertinib withdrawal for 2 gulation of ligands of FGFR or EGFR (27, 31–33). Downstream months, although KRAS expression in these cells was similar activation by the mutation/amplification of NRAS, KRAS,orYes1;

Figure 6. Association of phospho-EGFR (Y1068) with the GRB2/SOS1 complex sustained KRAS activity and led to osimertinib resistance. A, Western blot analysis of basal levels of phospho-EGFR (Y1068)/EGFR expression in AfaR, OsiR1, and OsiR1-F2M cells. b-Actin was included as a loading control. B, Association between EGFR and GRB2. AfaR, OsiR1, and OsiR1-F2M cells were lysed and subjected to immunoprecipitation with anti-GRB2 antibodies followed by immunoblotting with anti-EGFR and anti-GRB2 antibodies. Equal amounts of proteins from whole-cell lysates (WCL) were subjected to immunoblotting with anti-GRB2 or anti–b-actin antibodies. C, OsiR1 cells were transfected with nontargeting (N/T) siRNA or siRNA directed against GRB2, and GRB2 knockdown was determined by Western blot analysis. Lysates were then subjected to immunoblot analysis using the indicated primary antibodies; b-actin was included as a loading control. D, OsiR1 and OsiR1-F2M cells were transfected with N/T siRNA or siRNA directed against SOS1, and SOS1 knockdown was determined by Western blot analysis. Lysates were then subjected to immunoblot analysis using the indicated primary antibodies; b-actin was included as a loading control. E, Association between phospho-EGFR (Y1068) and the GRB2/SOS1 complex. AfaR, OsiR1, and OsiR1-F2M cells were exposed to 1 mmol/L osimertinib for 24 hours, and cell lyses were subjected to immunoprecipitation with anti-SOS1 antibodies followed by immunoblotting with anti–phospho-EGFR (Y1068), anti-EGFR, and anti-SOS1 antibodies. F, AfaR, OsiR1, and OsiR1-F2M cells were exposed to 1 mmol/L osimertinib for 24 hours, and cell lysates were subjected to immunoblot analysis using the indicated primary antibodies; b-actin was included as a loading control. Active KRAS in AfaR, OsiR1, and OsiR1-F2M cells was determined by RAF-RBD pull-down assays. G, OsiR1 cells were transfected with N/T siRNA or siRNA directed against GRB2 for 48 hours and then treated with 1 mmol/L osimertinib and/or selumetinib for 24 hours. GRB2 knockdown was determined by Western blot analysis. Lysates were subjected to immunoblot analysis using the indicated antibodies, and PARP cleavage was used to screen for the induction of apoptosis. b-Actin was included as a loading control. H, OsiR1 cells were transfected with N/T siRNA or siRNA directed against GRB2, and the transfected cells were reseeded in the presence or absence of osimertinib and/or selumetinib at the indicated concentrations. After incubation for 72 hours, MTT assays were performed, and OD570 values were obtained. Data represent means SEMs for 6 to 12 wells.

www.aacrjournals.org Mol Cancer Ther; 18(1) January 2019 123

Nakatani et al.

histologic transformation to SCLC; and phenotypic epithelial– removal in OsiR1 and OsiR2 cells. In OsiR1 cells, expression of mesenchymal transition have been shown to act as resistance the KRAS gene was increased in parallel with osimertinib con- mechanisms to other generation EGFR TKIs (19, 34–36). Previ- centrations up to 0.6 mmol/L and was then sustained at a high ously, loss of activating mutations in EGFR has been reported to level. In contrast, in OsiR2 cells, amplification of KRAS increased, act as a resistance mechanism to first- and second-generation with a peak at 0.4 mmol/L, and then decreased at 1 mmol/L EGFR TKIs (20, 37). Moreover, wild-type EGFR amplification osimertinib. Interestingly, at osimertinib concentrations of 0.2 has been observed in rociletinib resistance in PC-9 cells carrying to 0.6 mmol/L, activation of AKT and ERK1/2 caused stronger exon 19 del in EGFR and clinical tissue samples post-rociletinib inhibition in OsiR1 and OsiR2 cells, suggesting that strong therapy (27, 31, 38). To the best of our knowledge, this is a replication stress may have occurred. At the intermediate stage, first report describing partial loss of both exon 19 del and T790M KRAS amplification and overexpression were sustained, even with increased expression of WT EGFR. Furthermore, loss of after 2 months of osimertinib withdrawal in OsiR1 cells. both exon 19 del and T790M in EGFR is not sufficient for roci- However, in OsiR2 cells, reduced KRAS amplification and letinib resistance. Amplification of KRAS or EGFR was observed in expression were observed. Taken together, these findings clearly RocR1 and RocR2 cells, respectively; hence, inhibition of the KRAS indicated that the observed increases in KRAS gene copies were effector protein MEK, combined with afatinib, was required to not due to selection of preexisting populations of cells, but overcome RocR1 cell proliferation. For RocR2 cells, further strong rather due to an adaptive process in OsiR1 and OsiR2 cells inhibition of EGFR activation was necessary, and addition of because KRAS amplification did not show a linear increase. afatinib to cetuximab was required to suppress cell proliferation. However, the regulatory mechanism of KRAS alterations We have recently shown that WT KRAS amplification and remains unclear, and requires further investigation. hyperactivation leads to acquired resistance to afatinib in PC-9 Notably, we found that association of phospho-EGFR (Y1068) cells (i.e., AFR1), which exhibit reversible KRAS amplification with the GRB2/SOS1 complex sustained KRAS activity and osi- following withdrawal of afatinib for 2 months, decreasing to the mertinib resistance in OsiR1 cells. RAS proteins are upregulated by level of the parental PC-9 cells. Subsequently, afatinib sensitivity GEFs, such as SOS1, which enhance the exchange of GDP to GTP was recovered (20). Notably, in OsiR2 cells, KRAS amplification to promote the activation state, and downregulated by GTPase- and hyperactivation was reduced after 2 months of osimertinib activating proteins (GAP), such as p120GAP and neurofibromin 1 withdrawal, and KRAS expression then decreased to the level of (NF1; ref. 46). The reduced NF1 expression was reported to confer that in the parental AfaR cells. Subsequently, osimertinib sensi- resistance to EGFR inhibition in patients with NSCLC harboring tivity was recovered. Similar results were reported by Cepero and EGFR-activating mutations, although the osimertinib-resistant colleagues, demonstrating that KRAS and MET gene amplification cell lines did not show reduced NF1 expression (Supplementary was involved in the mechanism of MET inhibitor resistance in Fig. S4C; ref. 47). The association of phosphorylated EGFR with gastric cancer cells. This KRAS and MET amplification was revers- the GRB2/SOS1 complex leads to RAS activation. GRB2 is an ible after drug withdrawal for 1 month, resulting in loss of adaptor protein and is essential for signal transduction. GRB2 resistance (39). They postulated that amplified KRAS and MET binds to phospho-tyrosine residues in receptors via its SH2 genes may be induced by "replication stress," which interferes domain and binds to proteins containing proline-rich motifs, with DNA replication and disrupts chromatin organization in this such as SOS proteins via its SH3 domain (29). In OsiR1 cells, region, predisposing it to breakage (40). Furthermore, withdrawal GRB2 downregulation by siRNA decreases the activation of AKT of the inhibitor resulted in excessive signal transduction, which and ERK1/2, resulting in suppression of KRAS activation through may induce cellular stress and lead to loss of extrachromosomal disruption of the GRB2/SOS1 complex in OsiR1 cells. Interest- DNA (41). However, in osimertinib-resistant cells (OsiR1 and ingly, GRB2 knockdown leads to decreased phosphorylation of OsiR2), FISH analysis showed that the acquired increased copies EGFR. Furthermore, osimertinib inhibits EGFR phosphorylation of KRAS were not located on chromosome 12, where the KRAS when GRB2 is knocked down by siRNA, suggesting that GRB2 gene is located, or on extrachromosomal DNA, but existed on a may sustain EGFR phosphorylation. The role of GRB2 in EGFR homogenously stained region containing an amplified chromo- dynamics, including EGFR internalization, is unclear, and further somal segment, which stimulated cancer progression (42). Thus, studies are needed to elucidate these mechanisms. Moreover, the it is likely that cells gaining KRAS amplification may have a association between phosphorylated EGFR and GRB2/SOS1 was selective advantage under selective pressure with osimertinib. not evaluated in OsiR1 cells after osimertinib withdrawal for 2 Interestingly, in postprogression plasma samples after osimerti- months; however, adaptor/scaffold molecules, such as Shc, Gab1, nib therapy, amplification of KRAS was identified in circulating and Gab2, were activated and then inhibited by osimertinib, tumor DNA, suggesting this is one of the putative mechanisms of indicating that phosphorylated EGFR signals were induced down- resistance to osimertinib (43). Compound EGFR mutations, stream to avoid GRB2/SOS1. defined as multiple mutations in the EGFR tyrosine kinase Osimertinib treatment after first-line afatinib treatment exhib- domain, are frequently detected using advanced sequencing tech- ited a remarkable median time to disease progression of 31.5 nologies. It has been reported that compound EGFR mutations months longer than that of other subsequent therapies using with EGFR exon 19 deletion are rarer than those with L858R chemotherapeutics and first-generation EGFR-TKIs in retrospec- mutations (44, 45). Therefore, gene amplification or bypass tive analysis of Lux-lung 3, 6, and 7 (48). However, these are signals are more likely to act as resistance mechanisms than premature and retrospective results, the administration of osi- secondary or tertiary mutations involving EGFR exon 19 dele- mertinib in afatinib-relapsed NSCLC patients carrying activating tions. However, this is still a subject of investigation. mutations/T790M in EGFR would be promising. Furthermore, There are different patterns of altered KRAS copy numbers combination therapy with EGFR-TKI and MEK-inhibitor has been observed at various stages of resistance, that is, increased osimer- developed for lung cancers carrying activating EGFR mutations tinib concentrations and intermediate stages of osimertinib (NCT02143466 and NCT02025114). Consequently, the

124 Mol Cancer Ther; 18(1) January 2019 Molecular Cancer Therapeutics

Acquired Third-Generation EGFR-TKI Resistance Mechanisms

combination of third-generation EGFR-TKIs and MEK inhibitors Authors' Contributions may be effective for treating lung cancers with EGFR mutations Conception and design: T. Yamaoka, S. Kusumoto, T. Ohmori and RAS amplification through acquisition of EGFR-TKIs. There- Development of methodology: T. Yamaoka fore, our study indicates that the intermittent treatment with 2 Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): K. Nakatani, T. Yamaoka, S. Arata months of osimertinib withdrawal is better than continuous Analysis and interpretation of data (e.g., statistical analysis, biostatistics, treatment with EGFR-TKIs. computational analysis): K. Nakatani, T. Yamaoka, K.-I. Fujita, T. Ohmori To the best of our knowledge, this is the first report describing Writing, review, and/or revision of the manuscript: T. Yamaoka, M. Ohba, the resistance mechanisms to third-generation EGFR TKIs, that is, K.-I. Fujita, S. Kusumoto, S. Iwai rociletinib and osimertinib, after the emergence of T790M EGFR Administrative, technical, or material support (i.e., reporting or organizing following afatinib treatment. Amplification and subsequent over- data, constructing databases): K. Nakatani, T. Yamaoka, M. Ohba, S. Iwai Study supervision: T. Yamaoka, M. Ohba, I. Taki-Takemoto, D. Kamei, S. Iwai, expression of oncogenes have been observed in several types of J. Tsurutani, T. Ohmori neoplasms and are thought to play important roles in the advancement of tumor cells toward increased malignancy (49). In this study, we characterized novel mechanisms of acquired Acknowledgments resistance to T790M EGFR-selective third-generation EGFR TKIs. We thank the members of the Institute of Molecular Oncology for their Moreover, KRAS and EGFR gene expression should be evaluated thoughtful discussions and helpful advice, Mr. Hosono, and Mr. Kagami for in clinical samples following reacquisition of sensitivity to osi- technical assistance with animal care, and Editage for their assistance with English language editing. This work was supported in part by a research award mertinib in post-afatinib therapy. We propose a novel concept for from the Showa University School of Medicine Alumni Association and JSPS KRAS the adaptation of T790M-selective inhibitors, that is, or KAKENHI (grant number: 16K09590 to T. Yamaoka). EGFR amplification with or without loss of EGFR mutations (exon 19 del and T790M). In addition, we describe a preclinical ratio- The costs of publication of this article were defrayed in part by the nale for reversal of acquired KRAS activity; further clinical analyses payment of page charges. This article must therefore be hereby marked and clinical trials are needed to evaluate rechallenge with osi- advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate mertinib at 2 months after osimertinib withdrawal. this fact.

Disclosure of Potential Conflicts of Interest Received June 2, 2018; revised August 24, 2018; accepted October 9, 2018; No potential conflicts of interest were disclosed. published first October 15, 2018.

References 1. Maemondo M, Inoue A, Kobayashi K, Sugawara S, Oizumi S, Isobe H, et al. 12. Walter AO, Sjin RT, Haringsma HJ, Ohashi K, Sun J, Lee K, et al. Discovery of Gefitinib or chemotherapy for non-small-cell lung cancer with mutated a mutant-selective covalent inhibitor of EGFR that overcomes T790M- EGFR. N Engl J Med 2010;362:2380–8. mediated resistance in NSCLC. Cancer Discov 2013;3:1404–15. 2. Mok TS, Wu YL, Thongprasert S, Yang CH, Chu DT, Saijo N, et al. Gefitinib 13. Cross DA, Ashton SE, Ghiorghiu S, Eberlein C, Nebhan CA, Spitzler PJ, et al. or carboplatin-paclitaxel in pulmonary adenocarcinoma. N Engl J Med AZD9291, an irreversible EGFR TKI, overcomes T790M-mediated resis- 2009;361:947–57. tance to EGFR inhibitors in lung cancer. Cancer Discov 2014;4:1046–61. 3. Rosell R, Moran T, Queralt C, Porta R, Cardenal F, Camps C, et al. Screening 14. Mok TS, Wu YL, Ahn MJ, Garassino MC, Kim HR, Ramalingam SS, et al. for epidermal growth factor receptor mutations in lung cancer. N Engl J Osimertinib or platinum-pemetrexed in EGFR T790M-positive lung can- Med 2009;361:958–67. cer. N Engl J Med 2017;376:629–40. 4. Mitsudomi T, Yatabe Y. Mutations of the epidermal growth factor receptor 15. Thress KS, Paweletz CP, Felip E, Cho BC, Stetson D, Dougherty B, et al. gene and related genes as determinants of epidermal growth factor receptor Acquired EGFR C797S mutation mediates resistance to AZD9291 in non- tyrosine kinase inhibitors sensitivity in lung cancer. Cancer Sci 2007;98: small cell lung cancer harboring EGFR T790M. Nat Med 2015;21:560–2. 1817–24. 16. Chabon JJ, Simmons AD, Lovejoy AF, Esfahani MS, Newman AM, Har- 5. Zhou C, Wu YL, Chen G, Feng J, Liu XQ, Wang C, et al. Erlotinib versus ingsma HJ, et al. Circulating tumour DNA profiling reveals heterogeneity of chemotherapy as first-line treatment for patients with advanced EGFR EGFR inhibitor resistance mechanisms in lung cancer patients. Nat Com- mutation-positive non-small-cell lung cancer (OPTIMAL, CTONG- mun 2016;7:11815. 0802): a multicentre, open-label, randomised, phase 3 study. Lancet Oncol 17. Mizuuchi H, Suda K, Murakami I, Sakai K, Sato K, Kobayashi Y, et al. 2011;12:735–42. Oncogene swap as a novel mechanism of acquired resistance to epidermal 6. Sequist LV, Yang JC, Yamamoto N, O'Byrne K, Hirsh V, Mok T, et al. growth factor receptor-tyrosine kinase inhibitor in lung cancer. Cancer Sci Phase III study of afatinib or cisplatin plus pemetrexed in patients with 2016;107:461–8. metastatic lung adenocarcinoma with EGFR mutations. J Clin Oncol 18. Planchard D, Loriot Y, Andre F, Gobert A, Auger N, Lacroix L, et al. EGFR- 2013;31:3327–34. independent mechanisms of acquired resistance to AZD9291 in EGFR 7. Ohashi K, Maruvka YE, Michor F, Pao W. Epidermal growth factor receptor T790M-positive NSCLC patients. Ann Oncol 2015;26:2073–8. tyrosine kinase inhibitor-resistant disease. J Clin Oncol 2013;31:1070–80. 19. Eberlein CA, Stetson D, Markovets AA, Al-Kadhimi KJ, Lai Z, Fisher PR, et al. 8. Kobayashi S, Boggon TJ, Dayaram T, Janne PA, Kocher O, Meyerson M, et al. Acquired resistance to the mutant-selective EGFR inhibitor AZD9291 is EGFR mutation and resistance of non-small-cell lung cancer to gefitinib. associated with increased dependence on RAS signaling in preclinical N Engl J Med 2005;352:786–92. models. Cancer Res 2015;75:2489–500. 9. Campo M, Gerber D, Gainor JF, Heist RS, Temel JS, Shaw AT, et al. Acquired 20. Yamaoka T, Ohmori T, Ohba M, Arata S, Murata Y, Kusumoto S, et al. resistance to first-line afatinib and the challenges of prearranged progres- Distinct afatinib resistance mechanisms identified in lung adenocarcinoma sion biopsies. J Thorac Oncol 2016;11:2022–6. harboring an EGFR mutation. Mol Cancer Res 2017;15:915–28. 10. Janne PA, Yang JC, Kim DW, Planchard D, Ohe Y, Ramalingam SS, et al. 21. Ando K, Ohmori T, Inoue F, Kadofuku T, Hosaka T, Ishida H, et al. AZD9291 in EGFR inhibitor-resistant non-small-cell lung cancer. N Engl J Enhancement of sensitivity to tumor necrosis factor alpha in non-small Med 2015;372:1689–99. cell lung cancer cells with acquired resistance to gefitinib. Clin Cancer Res 11. Sequist LV, Soria JC, Goldman JW, Wakelee HA, Gadgeel SM, Varga A, et al. 2005;11:8872–9. Rociletinib in EGFR-mutated non-small-cell lung cancer. N Engl J Med 22.YamaokaT,OhmoriT,OhbaM,ArataS,KishinoY,MurataY,etal. 2015;372:1700–9. Acquired resistance mechanisms to combination Met-TKI/EGFR-TKI

www.aacrjournals.org Mol Cancer Ther; 18(1) January 2019 125

Nakatani et al.

exposure in Met-amplified EGFR-TKI-resistant lung adenocarcinoma of T790M determination on liquid re-biopsy. Lung Cancer 2018; harboringanactivatingEGFRmutation. Mol Cancer Ther 2016; 115:21–7. 15:3040–54. 36. Sequist LV, Waltman BA, Dias-Santagata D, Digumarthy S, Turke AB, Fidias 23. Roberts PJ, Stinchcombe TE. KRAS mutation: should we test for it, and does P, et al. Genotypic and histological evolution of lung cancers acquiring it matter? J Clin Oncol 2013;31:1112–21. resistance to EGFR inhibitors. Sci Transl Med 2011;3:75ra26. 24. Valtorta E, Misale S, Sartore-Bianchi A, Nagtegaal ID, Paraf F, Lauricella C, 37. Tabara K, Kanda R, Sonoda K, Kubo T, Murakami Y, Kawahara A, et al. et al. KRAS gene amplification in colorectal cancer and impact on response Loss of activating EGFR mutant gene contributes to acquired resistance to EGFR-targeted therapy. Int J Cancer 2013;133:1259–65. to EGFR tyrosine kinase inhibitors in lung cancer cells. PLoS One 25. Infante JR, Fecher LA, Falchook GS, Nallapareddy S, Gordon MS, Becerra C, 2012;7:e41017. et al. Safety, pharmacokinetic, pharmacodynamic, and efficacy data for the 38. Piotrowska Z, Niederst MJ, Karlovich CA, Wakelee HA, Neal JW, Mino- oral MEK inhibitor trametinib: a phase 1 dose-escalation trial. Lancet Kenudson M, et al. Heterogeneity underlies the emergence of EGFRT790 Oncol 2012;13:773–81. wild-type clones following treatment of T790M-positive cancers with a 26. Yeh TC, Marsh V, Bernat BA, Ballard J, Colwell H, Evans RJ, et al. Biological third-generation EGFR inhibitor. Cancer Discov 2015;5:713–22. characterization of ARRY-142886 (AZD6244), a potent, highly selective 39. Cepero V, Sierra JR, Corso S, Ghiso E, Casorzo L, Perera T, et al. MET and mitogen-activated protein kinase kinase 1/2 inhibitor. Clin Cancer Res KRAS gene amplification mediates acquired resistance to MET tyrosine 2007;13:1576–83. kinase inhibitors. Cancer Res 2010;70:7580–90. 27. Nukaga S, Yasuda H, Tsuchihara K, Hamamoto J, Masuzawa K, Kawada I, 40. Hellman A, Zlotorynski E, Scherer SW, Cheung J, Vincent JB, Smith DI, et al. et al. Amplification of EGFR wild-type alleles in non-small cell lung cancer A role for common fragile site induction in amplification of human cells confers acquired resistance to mutation-selective EGFR tyrosine kinase oncogenes. Cancer Cell 2002;1:89–97. inhibitors. Cancer Res 2017;77:2078–89. 41. Nathanson DA, Gini B, Mottahedeh J, Visnyei K, Koga T, Gomez G, et al. 28. Ebi H, Corcoran RB, Singh A, Chen Z, Song Y, Lifshits E, et al. Receptor Targeted therapy resistance mediated by dynamic regulation of extrachro- tyrosine kinases exert dominant control over PI3K signaling in human mosomal mutant EGFR DNA. Science 2014;343:72–6. KRAS mutant colorectal cancers. J Clin Invest 2011;121:4311–21. 42. Schwab M. Oncogene amplification in solid tumors. Semin Cancer Biol 29. Buday L, Downward J. Epidermal growth factor regulates p21ras through 1999;9:319–25. the formation of a complex of receptor, Grb2 adapter protein, and Sos 43. Ramalingam SS, Yang JC, Lee CK, Kurata T, Kim DW, John T, et al. nucleotide exchange factor. Cell 1993;73:611–20. Osimertinib as first-line treatment of EGFR mutation-positive advanced 30. Zheng D, Hu M, Bai Y, Zhu X, Lu X, Wu C, et al. EGFR G796D mutation non-small-cell lung cancer. J Clin Oncol 2018;36:841–9. mediates resistance to osimertinib. Oncotarget 2017;8:49671–9. 44. Kohsaka S, Nagano M, Ueno T, Suehara Y, Hayashi T, Shimada N, et al. A 31. Ortiz-Cuaran S, Scheffler M, Plenker D, Dahmen L, Scheel AH, Fernandez- method of high-throughput functional evaluation of EGFR gene variants of Cuesta L, et al. Heterogeneous mechanisms of primary and acquired unknown significance in cancer. Sci Transl Med 2017;9.pii: eaan6566. resistance to third-generation EGFR inhibitors. Clin Cancer Res 2016; 45. Kobayashi S, Canepa HM, Bailey AS, Nakayama S, Yamaguchi N, Goldstein 22:4837–47. MA, et al. Compound EGFR mutations and response to EGFR tyrosine 32. Martinez-Marti A, Felip E, Matito J, Mereu E, Navarro A, Cedres S, et al. Dual kinase inhibitors. J Thorac Oncol 2013;8:45–51. MET and ERBB inhibition overcomes intratumor plasticity in osimertinib- 46. Vigil D, Cherfils J, Rossman KL, Der CJ. Ras superfamily GEFs and GAPs: resistant-advanced non-small-cell lung cancer (NSCLC). Ann Oncol validated and tractable targets for cancer therapy? Nat Rev Cancer 2017;28:2451–7. 2010;10:842–57. 33. Kim TM, Song A, Kim DW, Kim S, Ahn YO, Keam B, et al. Mechanisms of 47. de Bruin EC, Cowell C, Warne PH, Jiang M, Saunders RE, Melnick MA, et al. acquired resistance to AZD9291: a mutation-selective, irreversible EGFR Reduced NF1 expression confers resistance to EGFR inhibition in lung inhibitor. J Thorac Oncol 2015;10:1736–44. cancer. Cancer Discov 2014;4:606–19. 34. Ichihara E, Westover D, Meador CB, Yan Y, Bauer JA, Lu P, et al. SFK/FAK 48. Sequist LV, Wu Y, Schuler M, Kato T, Yang JC, Tanaka H, et al. PS02.20 signaling attenuates osimertinib efficacy in both drug-sensitive and drug- subsequent therapies post-afatinib among patients with EGFR mutation- resistant models of EGFR-mutant lung cancer. Cancer Res 2017;77:2990– positive (EGFRmþ) NSCLC in LUX-Lung 3, 6 and 7. J Thorac Oncol 3000. 2017;12. 35. Minari R, Bordi P, Del Re M, Facchinetti F, Mazzoni F, Barbieri F, et al. 49. Brison O. Gene amplification and tumor progression. Biochim Biophys Primary resistance to osimertinib due to SCLC transformation: issue Acta 1993;1155:25–41.

126 Mol Cancer Ther; 18(1) January 2019 Molecular Cancer Therapeutics

KRAS and EGFR Amplifications Mediate Resistance to Rociletinib and Osimertinib in Acquired Afatinib-Resistant NSCLC Harboring Exon 19 Deletion/T790M in EGFR

Kaori Nakatani, Toshimitsu Yamaoka, Motoi Ohba, et al.

Mol Cancer Ther 2019;18:112-126. Published OnlineFirst October 15, 2018.

Updated version Access the most recent version of this article at: doi:10.1158/1535-7163.MCT-18-0591

Supplementary Access the most recent supplemental material at: Material http://mct.aacrjournals.org/content/suppl/2018/10/13/1535-7163.MCT-18-0591.DC1

Cited articles This article cites 48 articles, 18 of which you can access for free at: http://mct.aacrjournals.org/content/18/1/112.full#ref-list-1

Citing articles This article has been cited by 1 HighWire-hosted articles. Access the articles at: http://mct.aacrjournals.org/content/18/1/112.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 Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mct.aacrjournals.org/content/18/1/112. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.