ESMO 2018 CONGRESS October 2018 Munich, Germany. Developed in association with the European Thoracic Oncology Platform

Transcription

1 Developed in association with the European Thoracic Oncology Platform ESMO 218 CONGRESS October 218 Munich, Germany Supported by Eli Lilly and Company. Eli Lilly and Company has not influenced the content of this publication

2 Letter from Prof Rolf Stahel Dear Colleagues It is my pleasure to present this ETOP slide set which has been designed to highlight and summarise key findings in thoracic cancers from the major congresses in 218. This slide set specifically focuses on the ESMO 218 Congress and is available in 4 languages English, French, Chinese and Japanese. The area of clinical research in oncology is a challenging and ever changing environment. Within this environment, we all value access to scientific data and research which helps to educate and inspire further advancements in our roles as scientists, clinicians and educators. I hope you find this review of the latest developments in thoracic cancers of benefit to you in your practice. If you would like to share your thoughts with us we would welcome your comments. Please send any correspondence to etop@etop.eu-org. I would like to thank our ETOP members Solange Peters and Martin Reck for their roles as Editors for prioritising abstracts and reviewing slide content. The slide set you see before you would not be possible without their commitment and hard work. Finally, we are also very grateful to Lilly Oncology for their financial, administrative and logistical support in the realisation of this activity. Yours sincerely, Rolf Stahel President, ETOP Foundation Council

3 ETOP Medical Oncology Slide Deck Editors 218 Focus: advanced NSCLC (not radically treatable stage III & stage IV) Dr Solange Peters Multidisciplinary Oncology Center, Lausanne Cancer Center, Lausanne, Switzerland Focus: other malignancies, SCLC, mesothelioma, rare tumours Dr Martin Reck Department of Thoracic Oncology, Hospital Grosshansdorf, Grosshansdorf, Germany

4 Contents Screening and biomarkers Early stage and locally advanced NSCLC Stages I, II and III Advanced NSCLC Not radically treatable stage III and stage IV First line Later lines Other malignancies SCLC, mesothelioma and thymic epithelial tumours

5 Screening and biomarkers

6 65PD: High tumor mutational burden (TMB) and PD-L1 have similar predictive utility in 2L+ NSCLC patients (pts) treated with anti-pd-l1 and anti-ctla-4 Higgs BW, et al Study objective To compare the utility of high TMB and PD-L1 to predict outcomes in patients with non-squamous NSCLC who were treated with durvalumab and tremelimumab Methods Immunotherapy-naïve patients (n=213) with non-squamous EGFR and ALK wildtype NSCLC who had progressed after one prior platinum-based therapy were enrolled to durvalumab 2 mg/kg q4w for up to 12 months with tremelimumab 1 mg/kg q4w for 4 cycles Tumour PD-L1 expression classified as 25% or <25% using the Ventana PD-L1 (SP263) assay DNA sequencing was performed on 16 of 2 (53%) available tumour biopsies and TMB high defined as the top tertile of values (11 mut/mb) Higgs BW, et al. Ann Oncol 218;29(suppl 5):Abstr 65PD

7 PD-L1 TC, % 65PD: High tumor mutational burden (TMB) and PD-L1 have similar predictive utility in 2L+ NSCLC patients (pts) treated with anti-pd-l1 and anti-ctla-4 Higgs BW, et al Key results There was no appreciable correlation between TMB and PD-L1 Low linear correlation between PD-L1 levels and TMB PD-L1 vs. TMB status overlap TMB, mutations/mb rho= PD-L1 25% n=16 16% n=18 17% PD-L1 <25%/ TMB low n=52 5% TMB high n=17 17% Higgs BW, et al. Ann Oncol 218;29(suppl 5):Abstr 65PD

8 65PD: High tumor mutational burden (TMB) and PD-L1 have similar predictive utility in 2L+ NSCLC patients (pts) treated with anti-pd-l1 and anti-ctla-4 Higgs BW, et al Key results PFS Conclusions TMB 11 mut/mb (n=37) TMB <11 mut/mb (n=69) PD-L1 25% (n=57) PD-L1 <25% (n=136) Total* (N=213) Median, months (95%CI) 7.1 (1.7, 9.1) 1.7 (1.6, 3.6) 7.1 (3.5, 9.2) 3.3 (1.7, 3.6) 3.5 (1.8, 4.) 12-month PFS, % (95%CI) 26.8 (13.8, 41.6) 12.6 (5.6, 22.6) 34.2 (21.9, 46.8) 1. (4.8, 17.5) 17.6 (12.4, 23.6) OS Median, months (95%CI) NE (7.9, NE) 8.9 (4.8, NE) 15.5 (15.4, NE) 9.5 (6.7, NE) 15.4 (1., NE) 12-month OS, % (95%CI) 61.2 (42.4, 75.5) 43. (3.4, 55.) 71.6 (57.3, 81.9) 47.3 (38.2, 55.9) 53.8 (46.4, 6.6) *2 patients had unknown PD-L1 expression; NE, not estimable In this study of patients with non-squamous NSCLC TMB and PD-L1 expression were not correlated Patients who were TMB high ( 11 mut/mb) had improved ORR, PFS and OS Higgs BW, et al. Ann Oncol 218;29(suppl 5):Abstr 65PD

9 LM15 LM18 LM21 LM14 LM1 LM8 LM13 LM16 LM11 LM19 BM12 BM14 BM18 BM15 Neoantigen counts 1136PD: Discrepancy of tumor neoantigen burden between primary lesions and matched metastases in lung cancer Jiang T, et al Study objective To compare neoantigen landscapes in primary tumours and metastases in patients with lung cancer Key result Comparison of neoantigens in primary tumours and metastases by baseline characteristics Gender Smoking Histology Primary Female Yes SCLC Metastasis Male No LUAD LUSC 4 Age, years Gender Smoking Histology Conclusion Neoantigen landscapes were similar in primary tumours and metastases, however, only 2% of neoantigens were shared Jiang T, et al. Ann Oncol 218;29(suppl 5):Abstr 1136PD

10 Early stage and locally advanced NSCLC Stages I, II and III

11 LBA48_PR: CTONG 113: Erlotinib versus gemcitabine plus cisplatin as neoadjuvant treatment for stage IIIA N2 EGFR-mutation non-small-cell lung cancer (EMERGING): a randomised study Zhong W, et al Study objective To investigate the efficacy of erlotinib vs. gemcitabine + cisplatin as neoadjuvant/ adjuvant treatment in patients with locally advanced EGFR mutant NSCLC Key patient inclusion criteria Treatment naïve IIIA-N2 NSCLC EGFR activating mutation ECOG PS 1 (n=72) R 1:1 Erlotinib 15 mg/day for 42 days (n=37) Stratification Lymph node status Histology Smoking Sex Gemcitabine + cisplatin* q3w for 2 cycles (n=35) Non-PD S U R G E R Y Erlotinib 15 mg/day for 12 months Gemcitabine + cisplatin* q3w for 2 cycles Primary endpoint ORR *Gemcitabine 125 mg/m 2 D1, 8; cisplatin 75 mg/m 2 D1 Secondary endpoints Downstaging rates of pathological lymph nodes, pcr, PFS, OS, safety Zhong W, et al. Ann Oncol 218;29(suppl 5):Abstr LBA48_PR

12 LBA48_PR: CTONG 113: Erlotinib versus gemcitabine plus cisplatin as neoadjuvant treatment for stage IIIA N2 EGFR-mutation non-small-cell lung cancer (EMERGING): a randomised study Zhong W, et al Key results ORR in the ITT population was 54.1% (95%CI 37.2, 7.9) with erlotinib vs. 34.3% (95%CI 17.7, 5.8) with gemcitabine + cisplatin (OR 2.26 [95%CI.87, 5.84], p=.92) Erlotinib (n=37) Gemcitabine + cisplatin (n=35) p-value Surgery, n (%) 31 (83.8) 24 (68.6).129 Complete resection, n (%) R R1 R2 Lymph node downstage, n (%) N2 pn N2 pn1 N2 pn2 27 (73.) 27 (73.) 1 (2.7) 3 (8.1) 4 (1.8) 3 (8.1) 1 (2.7) 27 (73.) 22 (62.9) 22 (62.9) 1 (2.9) 1 (2.9) 1 (2.9) 1 (2.9) () 23 (65.7) Zhong W, et al. Ann Oncol 218;29(suppl 5):Abstr LBA48_PR

13 Progression-free survival, % LBA48_PR: CTONG 113: Erlotinib versus gemcitabine plus cisplatin as neoadjuvant treatment for stage IIIA N2 EGFR-mutation non-small-cell lung cancer (EMERGING): a randomised study Zhong W, et al Key results % PFS (ITT population) Group Events mpfs, months (95%CI) No.at risk Erlotinib Gemcitabine + cisplatin Conclusion 49.% 36.4% % Time, months Erlotinib in the neoadjuvant setting for EGFRm NSCLC improved ORR, MPR, R resection and lymph node downstaging and the safety profile was in line with previous reports Zhong W, et al. Ann Oncol 218;29(suppl 5):Abstr LBA48_PR 42 Erlotinib (n=37) (19.3, 23.6) Gemcitabine + cisplatin (n=35) (9.1, 14.7) HR.42 (95%CI.23,.76), p=.3

14 LBA49: Neoadjuvant nivolumab (N) or nivolumab plus ipilimumab (NI) for resectable non-small cell lung cancer (NSCLC) Cascone T, et al Study objective To investigate the efficacy and safety of nivolumab alone or in combination with ipilimumab in patients with untreated resectable NSCLC Key patient inclusion criteria Nivolumab 3 mg/kg D 1, 15, 29 (n=18) Surgery Adjuvant therapy Stage I IIIA NSCLC Eligible for surgery R 1:1 Stratification Stage (n=44) Ipilimumab 1 mg/kg D1 + nivolumab 3 mg/kg D1, 15, 29 (n=18) Surgery Adjuvant therapy Primary endpoint Major pathological response (MPR) Secondary endpoints Safety, perioperative mortality and morbidity, ORR, RFS, OS, complete resection rate, pcr Cascone T, et al. Ann Oncol 218;29(suppl 5):Abstr LBA49

15 LBA49: Neoadjuvant nivolumab (N) or nivolumab plus ipilimumab (NI) for resectable non-small cell lung cancer (NSCLC) Cascone T, et al Key results Total Nivolumab Nivolumab + ipilimumab Evaluable (resected) n=26 n=14 n=12 MPR + pcr, n (%) 8 (31) 4 (28) 4 (33) % viable tumour cells (pcr), n (%) 5 (19) 2 (14) 3 (25) 1 1% viable tumour cells, n (%) 3 (11) 2 (14) 1 (8) Overall (resected + unresectable) n=31 n=16 n=15 MPR + pcr, n (%) 8 (26) 4 (25) 4 (27) % viable tumour cells (pcr), n (%) 5 (16) 2 (13) 3 (2) 1 1% viable tumour cells, n (%) 3 (1) 2 (13) 1 (7) Cascone T, et al. Ann Oncol 218;29(suppl 5):Abstr LBA49

16 Overall % change in tumour size from baseline LBA49: Neoadjuvant nivolumab (N) or nivolumab plus ipilimumab (NI) for resectable non-small cell lung cancer (NSCLC) Cascone T, et al Key results (cont.) Evaluable Response, n (%) CR PR SD PD ORR, n/n (%) Total (n=32) 1 (3) 6 (19) 19 (59) 6 (19) 7/32 (22) Nivolumab (n=16) () 5 (31) 8 (5) 3 (19) 5/16 (31) Nivolumab + ipilimumab (n=16) 1 (6) 1 (6) 11 (69) 3 (19) 2/16 (12) * Nivolumab (evaluable) (n=16) Radiographic responses 2 12 * Nivolumab + ipilimumab (evaluable) (n=16) * *Considered SD in target lesion but overall PD due to new radiographic lesions * 1 Conclusions In resected patients with NSCLC, neoadjuvant nivolumab or nivolumab + ipilimumab induced an MPR rate of 31% Greater TIL proliferation and activation was demonstrated with neoadjuvant nivolumab and nivolumab + ipilimumab compared with untreated tumours Cascone T, et al. Ann Oncol 218;29(suppl 5):Abstr LBA49

17 1363O: Efficacy and safety evaluation based on time from completion of radiotherapy to randomization with durvalumab or placebo in pts from PACIFIC Faivre-Finn C, et al Study objective To investigate the efficacy and safety outcomes of durvalumab in the PACIFIC trial based on PD-L1 expression and the components of chemoradiation received prior to durvalumab or placebo Key patient inclusion criteria Stage III NSCLC No progression after 2 cycles of platinum-crt Pre-cCRT tumour tissue for PD-L1 testing if available WHO PS 1 (n=713) R 1:1 Durvalumab 1 mg/kg q2w (n=476) Stratification Age Sex Smoking history Placebo q2w (n=237) Primary endpoints PFS by BICR, OS Secondary endpoints ORR, DoR and TTDM by BICR, PFS2 by investigator, safety, PROs Faivre-Finn C, et al. Ann Oncol 218;29(suppl 5):Abstr 1363O

18 Probability of PFS Probability of PFS 1363O: Efficacy and safety evaluation based on time from completion of radiotherapy to randomization with durvalumab or placebo in pts from PACIFIC Faivre-Finn C, et al Key results 1. PFS (BICR) by PD-L1 TC 1% PFS (BICR) by PD-L1 TC <1% Events/ patients (%) Median PFS, months (95%CI) Durvalumab 84/212 (39.6) 17.8 (16.9, NR) Placebo 59/91 (64.8) 5.6 (3.6, 11.) HR.46 (95%CI.33,.64) 1. Events/ patients (%) Median PFS, months (95%CI) Durvalumab 49/9 (54.4) 1.7 (7.3, NR) Placebo 4/58 (69.) 5.6 (3.7, 1.6) HR.73 (95%CI.48, 1.11) No. at risk Durvalumab Placebo Time from randomisation, months No. at risk Durvalumab Placebo Time from randomisation, months Median OS in the PD-L1 TC 1% was NR (95%CI NR, NR) with durvalumab and 29.1 months (95%CI 17.7, NR) for placebo; HR.53 (95%CI.36,.77) Median OS in the PD-L1 TC <1% was NR (95%CI 2.8, NR) with durvalumab and NR (95%CI 27.3, NR) for placebo; HR 1.36 (95%CI.79, 2.34) Faivre-Finn C, et al. Ann Oncol 218;29(suppl 5):Abstr 1363O

19 1363O: Efficacy and safety evaluation based on time from completion of radiotherapy to randomization with durvalumab or placebo in pts from PACIFIC Faivre-Finn C, et al Key results (cont.) Induction chemotherapy Platinum Taxane Etoposide Vinorelbine Dose of radiotherapy Yes No Cisplatin Carboplatin Yes No Yes No Yes No <6 Gy 6 66 Gy >66 Gy PFS (BICR) Events / patients (%) HR (95%CI) Durvalumab Placebo /123 (48.) 155/353 (43.9) 115/266 (43.2) 91/199 (45.7) 97/27 (46.9) 117/269 (43.5) 49/117 (41.9) 165/359 (46.) 58/124 (46.8) 156/352 (44.3) 16/38 (42.1) 187/47 (45.9) 1/3 (33.3) 49/68 (72.1) 18/169 (63.9) 87/129 (67.4) 65/12 (63.7) 72/112 (64.3) 85/125 (68.) 34/52 (65.4) 123/185 (66.5) 42/59 (71.2) 115/178 (64.6) 11/15 (73.3) 13/22 (64.4) 15/19 (78.9) Conclusions Durvalumab better Placebo better Durvalumab led to improvement in PFS and OS in the PD-L1 1% subgroup and PFS in the <1% PD-L1 subgroup Durvalumab improved PFS and OS regardless of chemoradiation received prior to therapy Faivre-Finn C, et al. Ann Oncol 218;29(suppl 5):Abstr 1363O

20 Advanced NSCLC Not radically treatable stage III and stage IV First line

21 LBA1: Primary results of ALESIA: A randomised, phase III, open-label study of alectinib vs crizotinib in Asian patients with treatment-naïve ALK+ advanced NSCLC Zhou C, et al Study objective To investigate safety and efficacy of 1L alectinib vs. crizotinib in Asian patients with ALK+ NSCLC in the ALESIA study Key patient inclusion criteria Asian patients with stage IIIB/IV ALK+ NSCLC Treatment naïve ECOG PS 2 (n=187) R 2:1 Alectinib 6 mg bid (n=125) Stratification ECOG PS (/1 vs. 2) Baseline brain metastases Crizotinib 25 mg bid (n=62) Primary endpoint PFS (investigator-assessed) Secondary endpoints PFS by IRC, time to CNS progression, ORR and DoR (investigator-assessed), OS, CNS ORR, safety, QoL, PK Zhou C, et al. Ann Oncol 218;29(suppl 5):Abstr LBA1

22 Progression-free survival, % Cumulative incidence*, % LBA1: Primary results of ALESIA: A randomised, phase III, open-label study of alectinib vs crizotinib in Asian patients with treatment-naïve ALK+ advanced NSCLC Zhou C, et al Key result PFS (investigator-assessed) Time to CNS progression (IRC-assessed) 1 Alectinib Crizotinib 6 Alectinib Crizotinib NE % (95%CI 23.5, 47.8) months Time, months 7.3% (95%CI 3.6, 12.8) Time, months Alectinib (n=125) Crizotinib (n=62) Patients with event, n (%) 26 (2.8) 37 (59.7) Median PFS, months (95%CI) NE (2.3, NE) 11.1 (9.1, 13.) HR (95%CI); p-value (log-rank test).22 (.13,.38); <.1 Cause-specific HR p-value (log-rank test).14 (95%CI.6,.3) <.1 *Cumulative incidence of CNS progression without prior non-cns progression or death Zhou C, et al. Ann Oncol 218;29(suppl 5):Abstr LBA1

23 DoR (unconfirmed),% LBA1: Primary results of ALESIA: A randomised, phase III, open-label study of alectinib vs crizotinib in Asian patients with treatment-naïve ALK+ advanced NSCLC Zhou C, et al Key result (cont.) Alectinib Crizotinib months Time, months Alectinib (n=114) NE Crizotinib (n=48) Median DoR, months (95%CI) NE (18.4, NE) 9.3 (7.4, NE) HR (95%CI), p-value (log-rank test) DoR (investigator-assessed).22 (.12,.4); p<.1 Response, n (%) Alectinib (n=125) Crizotinib (n=62) ORR 114 (91.2) 48 (77.4) CR 5 (4.) 3 (4.8) PR 19 (87.2) 45 (72.6) SD 7 (5.6) 8 (12.9) PD 2 (1.6) 4 (6.5) Missing or unevaluable 2 (1.6) 2 (3.2) Conclusions Alectinib led to a significant improvement in PFS and showed benefits in regards to ORR, DoR and time to CNS progression compared with crizotinib The safety profile of alectinib in Asian patients was consistent with the previous findings Zhou C, et al. Ann Oncol 218;29(suppl 5):Abstr LBA1

24 LBA5: Mechanisms of acquired resistance to first-line osimertinib: preliminary data from the phase III FLAURA study Ramalingam SS, et al Study objective To investigate the mechanisms of acquired resistance to osimertinib and standard of care (gefitinib + erlotinib) in patients who progressed or discontinued treatment in the FLAURA study Methods Patients in the FLAURA study had EGFR+ (ex19del or L858R) locally advanced or metastatic NSCLC and were treated with osimertinib or gefitinib + erlotinib until progression Paired plasma samples were taken from patients at baseline and at progression/discontinuation for ctdna genomic profiling NGS was conducted using the Guardant36 assay or GuardantOMNI assay Analysis set of valid paired NGS data were available for 272 patients Osimertinib: 113/279 (41%) Gefitinib + erlotinib: 159/277 (57%) Ramalingam SS, et al. Ann Oncol 218;29(suppl 5):Abstr LBA5

25 LBA5: Mechanisms of acquired resistance to first-line osimertinib: preliminary data from the phase III FLAURA study Ramalingam SS, et al Key results The most common acquired resistance mechanisms in the osimertinib arm (n=91) were: MET amplification (15%) and EGFR C797S mutation (7%) There was no evidence of acquired T79M The most common acquired resistance mechanisms in the gefitinib + erlotinib arm (n=129) were: T79M (47%), MET amplification (4%) and HER2 amplification (2%) Conclusions There was no evidence of acquired resistance through T79M in the osimertinib-treated patients, while MET amplification and EGFR C797S mutations were the most common New mechanisms that may lead to aggressive disease biology were not identified Ramalingam SS, et al. Ann Oncol 218;29(suppl 5):Abstr LBA5

26 LBA51: Analysis of resistance mechanisms to osimertinib in patients with EGFR T79M advanced NSCLC from the AURA3 study Papadimitrakopoulou VA, et al Study objective To investigate the mechanisms of acquired resistance to osimertinib in patients who progressed or discontinued treatment in the AURA3 study Methods Patients in the AURA3 study had T79M+ locally advanced or metastatic NSCLC and were treated with osimertinib or platinum-pemetrexed until progression Paired plasma samples were taken from patients at baseline and at progression/discontinuation for ctdna genomic profiling NGS was undertaken using the Guardant36 assay Analysis set of valid paired NGS data were available for 113 patients Osimertinib: 83/279 (3%) Platinum-pemetrexed: 3/14 (21%) Papadimitrakopoulou VA, et al. Ann Oncol 218;29(suppl 5):Abstr LBA51

27 LBA51: Analysis of resistance mechanisms to osimertinib in patients with EGFR T79M advanced NSCLC from the AURA3 study Papadimitrakopoulou VA, et al Key results Loss of T79M was seen in 49% of patients Acquired EGFR mutations were observed in 21% of patients; the most common was C797S (14%) Other mutations included: MET amplification (19%); cell cycle gene alterations (12%); HER2 amplification (5%); PIK3CA amplification/mutation (5%); oncogenic fusion (4%), BRAF V6E (3%) All cases of C797X mutations were in the cis position when co-occurring with T79M Conclusion Acquired EGFR mutations with osimertinib (represents 21% of total patients) C797X: 15% (1 C797S, 1 C797G) L792H/F + C797S: 1% L792H: 1% G796S: 1% L718Q: 1% Ex2ins: 1% EGFR mutations and MET amplifications were the most common acquired resistance mechanisms to osimertinib among a diverse mixture, consistent with previous reports Papadimitrakopoulou VA, et al. Ann Oncol 218;29(suppl 5):Abstr LBA51

28 LBA52: Results of the GEOMETRY mono-1 phase II study for evaluation of the MET inhibitor capmatinib (INC28) in patients (pts) with METΔex14 mutated advanced non-small cell lung cancer (NSCLC) Wolf J, et al Study objective To investigate the efficacy and safety of capmatinib, a selective MET inhibitor, in patients with METΔex14 mutated advanced NSCLC Key patient inclusion criteria Stage IIIB/IV NSCLC METΔex14 determined centrally EGFR wt and ALK negative 1 measureable lesion ECOG PS 1 Cohort 4: Previously treated (1 2 prior lines) for advanced stage Capmatinib 4 mg bid (n=69) Cohort 5b: Treatment-naïve for advanced stage Capmatinib 4 mg bid (n=28) Primary endpoint ORR by BIRC Secondary endpoints DoR by BIRC Wolf J, et al. Ann Oncol 218;29(suppl 5):Abstr LBA52

29 Best % change from baseline Best % change from baseline LBA52: Results of the GEOMETRY mono-1 phase II study for evaluation of the MET inhibitor capmatinib (INC28) in patients (pts) with METΔex14 mutated advanced non-small cell lung cancer (NSCLC) Wolf J, et al Key results ORR (BIRC) in Cohort 4 (pre-treated) was 39.1% (95%CI 27.6, 51.6) ORR (BIRC) in Cohort 5b (treatment naïve) was 72.% (95%CI 5.6, 87.9) 25 Cohort 4 (pre-treated) * n=6 /69 * * * Cohort 5b (treatment naïve) * * * * * * * * * 25 * 5 75 * * * * * * n=24 /25 PR SD PD NE * * * * 1 *Patients still on treatment; number of patients with measurable disease at baseline and 1 post-baseline assessment Wolf J, et al. Ann Oncol 218;29(suppl 5):Abstr LBA52

30 LBA52: Results of the GEOMETRY mono-1 phase II study for evaluation of the MET inhibitor capmatinib (INC28) in patients (pts) with METΔex14 mutated advanced non-small cell lung cancer (NSCLC) Wolf J, et al Key results AE, n (%) Conclusions Cohort 4 (pre-treated) (n=69) Cohort 5b (treatment naive) (n=28) All patients (N=32) All grades Grade 3/4 All grades Grade 3/4 All grades Grade 3/4 Any 6 (87.) 33 (47.8) 27 (96.4) 14 (5.) 253 (83.8) 1 (33.1) Peripheral oedema 31 (44.9) 1 (14.5) 18 (64.3) 1 (3.6) 122 (4.4) 19 (6.3) Nausea 24 (34.8) 11 (39.3) 99 (32.8) 5 (1.7) Vomiting 13 (18.8) 4 (14.3) 58 (19.2) 6 (2.) Blood creatinine increased 15 (21.7) 7 (25.) 58 (19.2) Fatigue 9 (13.) 4 (5.8) 2 (7.1) 1 (3.6) 4 (13.2) 1 (3.3) Decreased appetite 1 (14.5) 1 (1.4) 5 (17.9) 4 (13.2) 3 (1.) Diarrhoea 8 (11.6) 3 (1.7) 35 (11.6) In patients with METΔex14 advanced NSCLC, capmatinib shows promising results The differential benefit of 1L over later-line treatment highlights the need for prompt targeted treatment in this patient group Wolf J, et al. Ann Oncol 218;29(suppl 5):Abstr LBA52

31 LBA53: IMpower13: Progression-free survival (PFS) and safety analysis from a randomised phase 3 study of carboplatin + nab-paclitaxel (CnP) with or without atezolizumab (atezo) as first-line (1L) therapy in advanced non-squamous NSCLC Cappuzzo F, et al Study objective To investigate the efficacy and safety of atezolizumab combined with carboplatin + nab-paclitaxel vs. carboplatin + nab-paclitaxel in chemotherapy-naïve patients with stage IV non-squamous NSCLC Key patient inclusion criteria Stage IV nonsquamous NSCLC Chemotherapy-naïve Pre-treated EGFR mutated or ALK translocation (TKI) (n=723; ITT-WT* n=679) R 2:1 Induction treatment (4 or 6 cycles) Atezolizumab 12 mg q3w + carboplatin AUC 6 q3w + nab-paclitaxel 1 mg/m 2 q3w Stratification Sex Baseline liver metastases PD-L1 tumour expression Carboplatin AUC 6 q3w + nab-paclitaxel 1 mg/m 2 q3w Maintenance treatment Atezolizumab Best supportive care or pemetrexed q3w Loss of clinical benefit/ toxicity PD/ toxicity Co-primary endpoints Investigator-assessed PFS and OS (ITT-WT* population) *ITT-WT population, randomised patients excluding those with EGFR or ALK mutations Secondary endpoints OS and PFS (ITT population and by PD-L1 expression), ORR, safety Cappuzzo F, et al. Ann Oncol 218;29(suppl 5):Abstr LBA53

32 PFS, % LBA53: IMpower13: Progression-free survival (PFS) and safety analysis from a randomised phase 3 study of carboplatin + nab-paclitaxel (CnP) with or without atezolizumab (atezo) as first-line (1L) therapy in advanced non-squamous NSCLC Cappuzzo F, et al OS, % Key results Investigator-assessed PFS (ITT-WT) OS (ITT-WT) PFS rate, % 6-month 12-month.8 OS rate, % 1-year 2-year Atezolizumab + CnP Atezolizumab + CnP CnP CnP Median follow-up: ~19 months HR.64 (95%CI.54,.77) p<.1.4 HR.79 (95%CI.64,.98) p= Median: 5.5 mo (95%CI 4.4, 5.9) Median: 7. mo (95%CI 6.2, 7.3). Median: 13.9 mo (95%CI 12., 18.7) Median: 18.6 mo (95%CI 16., 21.2) No. at risk Atezo + CnP Chemo Months after randomisation No. at risk Atezo + CnP Chemo Months after randomisation Cappuzzo F, et al. Ann Oncol 218;29(suppl 5):Abstr LBA53

33 LBA53: IMpower13: Progression-free survival (PFS) and safety analysis from a randomised phase 3 study of carboplatin + nab-paclitaxel (CnP) with or without atezolizumab (atezo) as first-line (1L) therapy in advanced non-squamous NSCLC Cappuzzo F, et al PFS, % PFS, % PFS, % Key results (cont.) 1 8 PD-L1-high TC3 or IC3 PFS by baseline PD-L1 status (ITT-WT) 1 8 PD-L1-low TC1/2 or IC1/2 1 8 PD-L1-negative TC and IC No. at risk Atezo+CnP Chemo Time, months Time, months Time, months Median PFS, mo (95%CI) Conclusions Atezo + CnP (n=88) 6.4 (5.49, 9.76) HR (95%CI).51 (.34,.77) CnP (n=42) 4.6 (3.22, 7) Median PFS, mo (95%CI) Atezo + CnP (n=128) 8.3 (7.16, 1.35) HR (95%CI).61 (.43,.85) In the ITT-WT population atezolizumab + chemotherapy showed benefit in PFS and OS, which was maintained across all PD-L1 subgroups No new safety signals were observed with the combination of atezolizumab + chemotherapy Cappuzzo F, et al. Ann Oncol 218;29(suppl 5):Abstr LBA53 CnP (n=65) 6. (5.29, 6.93) Median PFS, mo (95%CI) Atezo + CnP (n=235) 6.2 (5.52, 7.16) HR (95%CI).72 (.56,.91) CnP (n=121) 4.7 (4.11, 5.72)

34 LBA54: IMpower132: efficacy of atezolizumab (atezo) + carboplatin (carbo)/cisplatin (cis) + pemetrexed (pem) as 1L treatment in key subgroups with stage IV non-squamous non-small cell lung cancer (NSCLC) Barlesi F, et al Study objective To investigate the PFS benefit of atezolizumab combined with carboplatin or cisplatin + pemetrexed in patients from IMpower132 in an exploratory analysis stratified by race, age, smoking history or liver metastasis at baseline Key patient inclusion criteria Stage IV nonsquamous NSCLC Chemotherapy-naïve Without EGFR or ALK genetic alteration Exploratory endpoints R 1:1 Clinical and biomarker subgroup analysis Induction therapy (4 or 6 cycles) Atezolizumab + carboplatin or cisplatin + pemetrexed* Stratification Sex Smoking status ECOG PS Chemotherapy regimen (n=578) Carboplatin or cisplatin + pemetrexed* Maintenance therapy Atezolizumab + pemetrexed* Pemetrexed* PD/ loss of clinical benefit PD/ loss of clinical benefit *Atezolizumab 12 mg IV q3w; carboplatin AUC 6 mg/ml/min IV q3w; cisplatin 75 mg/m 2 IV q3w; pemetrexed 5 mg/m 2 IV q3w Barlesi F, et al. Ann Oncol 218;29(suppl 5):Abstr LBA54

35 PFS, % PFS, % PFS, % PFS, % LBA54: IMpower132: efficacy of atezolizumab (atezo) + carboplatin (carbo)/cisplatin (cis) + pemetrexed (pem) as 1L treatment in key subgroups with stage IV non-squamous non-small cell lung cancer (NSCLC) Barlesi F, et al Key results 1. Median PFS (95%CI), mos HR (95%CI) Non-Asian patients Atezolizumab + PP PP 6.9 (5.9, 8.1) 5. (4.2, 5.7).65 (.53,.81) Median PFS (95%CI), mos HR (95%CI) 1. Asian patients Atezolizumab + PP PP 1.2 (8.3, 15.3) 5.3 (4.3, 6.7).42 (.28,.63).8.6 Atezolizumab + PP (n=221) PP (n=221).8.6 Atezolizumab + PP (n=71) PP (n=65) Time, months Median PFS (95%CI), mos HR (95%CI) Former/current smokers Atezolizumab + PP PP 7.5 (6.3, 8.4) 5.1 (4.3, 5.6).61 (.5,.74) Time, months Median PFS (95%CI), mos HR (95%CI) Never smokers Atezolizumab + PP PP 8.6 (6.5, 15.4) 5.5 (4., 8.3).49 (.28,.87).8.6 Atezolizumab + PP (n=225) PP (n=226).8.6 Atezolizumab + PP (n=37) PP (n=3) Time, months Time, months Conclusion Atezolizumab added to carboplatin/cisplatin + pemetrexed improved PFS across several subgroups including patients from Asia, never smokers, those who were older and those without liver metastases at baseline, but currently has not shown improvement in OS Barlesi F, et al. Ann Oncol 218;29(suppl 5):Abstr LBA54

36 LBA55: Primary efficacy results from B-F1RST, a prospective phase II trial evaluating blood-based tumour mutational burden (btmb) as a predictive biomarker for atezolizumab (atezo) in 1L nonsmall cell lung cancer (NSCLC) Kim ES, et al Study objective To investigate the use of btmb as a predictive biomarker for atezolizumab monotherapy in 1L NSCLC Key patient inclusion criteria Stage IIIB/IVA locally advanced or metastatic NSCLC Immunotherapy naïve PD-L1 unselected ECOG PS 1 (n=152) Atezolizumab 12 mg q3w (biomarker evaluable population n=119) PD/toxicity/ loss of benefit Co-primary endpoints Investigator-assessed ORR and PFS (using pre-specified btmb cut-off of 16) Secondary endpoints Safety, investigator-assessed DoR and OS Kim ES, et al. Ann Oncol 218;29(suppl 5):Abstr LBA55

37 Overall response rate, % LBA55: Primary efficacy results from B-F1RST, a prospective phase II trial evaluating blood-based tumour mutational burden (btmb) as a predictive biomarker for atezolizumab (atezo) in 1L nonsmall cell lung cancer (NSCLC) Kim ES, et al Key results ORR per RECIST v1.1 btmb subgroups 1 cut-off 16 cut-off 2 cut-off p<.1 4 PR CR p= % 35 3 p= % % 1.1% 16.3% 5.7% 4.4% 5% ITT (N=152) BEP (n=119) High (n=49) Low (n=7) High (n=28) Low (n=91) High (n=19) Low (n=1) Kim ES, et al. Ann Oncol 218;29(suppl 5):Abstr LBA55

38 LBA55: Primary efficacy results from B-F1RST, a prospective phase II trial evaluating blood-based tumour mutational burden (btmb) as a predictive biomarker for atezolizumab (atezo) in 1L nonsmall cell lung cancer (NSCLC) Kim ES, et al Progression-free survival, % Key results (cont.) PFS in btmb high ( 16) vs. low (<16) subgroups 1 8 High, 16 (n=28) Low, <16 (n=91) Median PFS, months (9%CI) btmb high (n=28) btmb low (n=91) 4.6 (1.6, 11.) 3.7 (2.6, 4.3) 6 6-month PFS 41.6% vs. 32.8% HR (9%CI).66 (.42, 1.2) p-value month PFS 37.4% vs. 9.7% No. at risk High ( 16) Low (<16) Conclusion Time, months In patients with NSCLC treated with atezolizumab monotherapy a numerical improvement in outcomes was seen in those with a btmb cut-off of Kim ES, et al. Ann Oncol 218;29(suppl 5):Abstr LBA55

39 LBA58: Intracranial efficacy of brigatinib (BRG) vs crizotinib (CRZ) in the phase 3 ALTA-1L trial Popat S, et al Study objective To investigate the intracranial efficacy of brigatinib vs. crizotinib in ALK inhibitornaïve patients with advanced ALK+ NSCLC in the ATLA-1L study Key patient inclusion criteria Stage IIIb/IV NSCLC ALK+ (based on local ALK testing) No prior ALK inhibitor 1 prior systemic therapy R 1:1 Brigatinib 18 mg/day (9 mg/day for 7 day lead-in) (n=137) Stratification Brain metastases at baseline Prior chemotherapy PD/toxicity/ discontinuation Brain metastases were allowed (n=275) Crizotinib 25 mg bid (n=138) PD*/toxicity/ discontinuation Primary endpoint PFS (BIRC-assessed) Secondary endpoints ORR, intracranial ORR, intracranial PFS, OS, safety *Crossover to brigatinib permitted at PD Popat S, et al. Ann Oncol 218;29(suppl 5):Abstr LBA58

40 PFS, % of patients PFS, % of patients LBA58: Intracranial efficacy of brigatinib (BRG) vs crizotinib (CRZ) in the phase 3 ALTA-1L trial Popat S, et al Key results Whole body BIRC-assessed PFS by brain metastases at baseline 1 8 With brain metastases HR.2 (95%CI.9,.46) p<.1 by log-rank test 1 8 Without brain metastases HR.72 (95%CI.44, 1.18) p=.2 by log-rank test Brigatinib (n=4) Crizotinib (n=41) Time, months 2 Brigatinib (n=97) Crizotinib (n=97) Time, months Treatment Pts with events, n (%) Median PFS, months (95%CI) 1-year PFS rate, % (95%CI) Brigatinib (n=4) 8 (2) NR 75 (56, 87) Crizotinib (n=41) 24 (59) 5.6 (3.8, 11.1) 25 (8, 46) Treatment Pts with events, n (%) Median PFS, months (95%CI) 1-year PFS rate, % (95%CI) Brigatinib (n=97) 28 (29) NR 63 (5, 74) Crizotinib (n=41) 39 (4) 11.1 (9.2, NR) 49 (36, 61) Conclusion In patients with advanced ALK+ NSCLC, brigatinib demonstrated superior intracranial efficacy vs. crizotinib, with significantly higher intracranial response and improved intracranial PFS in those with brain metastases Popat S, et al. Ann Oncol 218;29(suppl 5):Abstr LBA58

41 1377O: Phase 2 study of tepotinib + gefitinib (TEP+GEF) in MET-positive (MET+)/epidermal growth factor receptor (EGFR)-mutant (MT) non-small lung cancer (NSCLC) Wu Y, et al Study objective To investigate the efficacy and safety of tepotinib + gefitinib vs. chemotherapy in Asian patients with advanced MET+/EGFR+T79M- NSCLC Key patient inclusion criteria Locally-advanced/metastatic IV NSCLC EGFR+, T79M, MET+ Asian Resistance to prior EGFR TKI No prior HGF/MET pathwaydirected therapy (n=55) Primary endpoint PFS (investigator-assessed) R* Tepotinib 5 mg/day + gefitinib 25 mg/day (n=31) Stratification Secondary endpoints ORR, safety PD/ toxicity MET+ type (IHC2+, IHC3+ or MET amplification) Prior EGFR-TKI therapy Chemotherapy: Pemetrexed 5 mg/m 2 q3w+ cisplatin 75 mg/m 2 or carboplatin AUC 5 or 6 q3w (n=24) PD/ toxicity *Initially 1:1 and changed to 2:1 at protocol amendment (3 Sept 216) Wu Y, et al. Ann Oncol 218;29(suppl 5):Abstr 1377O

42 KM estimate for PFS 1377O: Phase 2 study of tepotinib + gefitinib (TEP+GEF) in MET-positive (MET+)/epidermal growth factor receptor (EGFR)-mutant (MT) non-small lung cancer (NSCLC) Wu Y, et al Key results Investigator-assessed PFS (ITT population) Tepotinib + gefitinib (n=31) Chemotherapy (n=24) Events Median PFS, months (9%CI) 4.9 (3.9, (4.2, 6.8) Stratified HR (9%CI).71 (.36, 1.39) Tepotinib + gefitinib Chemotherapy No. at risk Tep + gef Chemo Time, months In the MET IHC3+ subgroup (n=34) median PFS was 8.3 months with tepotinib + gefitinib and 4.4 months with chemotherapy (HR.35 [95%CI.17,.74]) In the MET amplification subgroup (n=19) median PFS was 21.2 months with tepotinib + gefitinib and 4.2 months with chemotherapy (HR.17 [95%CI.5,.57]) Wu Y, et al. Ann Oncol 218;29(suppl 5):Abstr 1377O

43 1377O: Phase 2 study of tepotinib + gefitinib (TEP+GEF) in MET-positive (MET+)/epidermal growth factor receptor (EGFR)-mutant (MT) non-small lung cancer (NSCLC) Wu Y, et al Key results Responders* n/n Overall (n=55) 14/31 MET IHC3+ (n=34) 13/19 MET amplification (n=19) 8/12 Tepotinib + gefitinib ORR, % (9%CI) 45.2 (29.7, 61.3) 68.4 (47., 85.3) 66.7 (39.1, 87.7) Responders* n/n 8/24 5/15 3/7 Chemotherapy ORR, % (9%CI) 33.3 (17.8, 52.1) 33.3 (14.2, 57.7) 42.9 (12.9, 77.5) Odds ratio (9%CI) 1.99 (.56, 6.87) 4.33 (1.3, 18.33) 2.67 (.37, 19.56) Tepotinib + gefitinib was generally well tolerated, with TEAEs leading to discontinuation in 9.7% of patients compared with 4.3% in the chemotherapy group Conclusion In patients with advanced NSCLC and MET amplifications, improvements in PFS were observed with tepotinib + gefitinib, indicating that MET may be considered as a biomarker for treatment with tepotinib *No confirmation required Wu Y, et al. Ann Oncol 218;29(suppl 5):Abstr 1377O

44 1379PD: Impact of the EML4-ALK variant on the efficacy of alectinib (ALC) in untreated ALK+ advanced NSCLC (ansclc) in the global phase III ALEX study Dziadziuszko R, et al Study objective To assess efficacy of alectinib vs. crizotinib in the ALEX trial in patients with ALK+ advanced NSCLC stratified by EML4-ALK fusion variants Key patient inclusion criteria ALK+ NSCLC No prior treatment for NSCLC ECOG PS 2 (n=33) R 1:1 Alectinib 6 mg bid Crizotinib 25 mg bid PD/death/ withdrawal PD/death/ withdrawal Methods FOUNDATIONACT and FOUNDATIONONE NGS platforms were used to determine the ALK fusion variants in plasma and tissue BEP subgroups Dziadziuszko R, et al. Ann Oncol 218;29(suppl 5):Abstr 1379PD

45 Survival probability 1379PD: Impact of the EML4-ALK variant on the efficacy of alectinib (ALC) in untreated ALK+ advanced NSCLC (ansclc) in the global phase III ALEX study Dziadziuszko R, et al Key results PFS (investigator-assessed) Plasma BEP subgroup Tissue BEP subgroup 1. Crizotinib Median PFS, months EML4-ALK V1: 7.4 EML4-ALK V2: 8.8 EML4-ALK V3a/b: Alectinib Median PFS, months EML4-ALK V1: NE EML4-ALK V2: 14.9 EML4-ALK V3a/b: NE 1. Crizotinib Median PFS, months EML4-ALK V1: 12.9 EML4-ALK V2: 8.8 EML4-ALK V3a/b: Alectinib Median PFS, months EML4-ALK V1: NE EML4-ALK V2: 11.5 EML4-ALK V3a/b: NE Time, months Time, months Time, months Time, months ALK V1 ALK V2 ALK V3a/b Conclusion In patients with ALK+ NSCLC alectinib demonstrated greater efficacy than crizotinib, regardless of EML4-ALK variant Dziadziuszko R, et al. Ann Oncol 218;29(suppl 5):Abstr 1379PD

46 138PD: Efficacy of lorlatinib in patients (pts) with ROS1-positive advanced non-small cell lung cancer (NSCLC) and ROS1 kinase domain mutations Solomon BJ, et al Study objective To investigate molecular profiling of patients with ROS1+ advanced NSCLC in the B74611 study of lorlatinib Methods Patients with locally advanced or metastatic NSCLC with ROS1 rearrangements were enrolled in the phase 1/2 study Patients were treatment-naïve in the advanced setting of PD after 1 prior ROS1 inhibitor (phase 1) or any number of prior therapies (phase 2) Lorlatinib was orally administered in continuous 21-day cycles with escalating doses (1 mg/day to 1 mg bid) in phase 1 and 1 mg/day in phase 2 Molecular profiling of tumour tissue and blood was performed All patients had tissue samples collected before enrolment and tissue collection was encouraged on PD; tumour tissue was analysed with a ROS1 kinase domain mutationfocused NGS panel Blood samples were collected at screening, at the beginning of cycle 3, and at the endof-treatment visit for circulating-free DNA (cfdna) analysis using an NGS panel or digital PCR Solomon BJ, et al. Ann Oncol 218;29(suppl 5):Abstr 138PD

47 Best percentage change from baseline 138PD: Efficacy of lorlatinib in patients (pts) with ROS1-positive advanced non-small cell lung cancer (NSCLC) and ROS1 kinase domain mutations Solomon BJ, et al Key results ORR by presence of ROS1 mutations ROS1 TKI-naïve (n=17) No mutation (n=17) Best overall response, n (%) 1 mutation (n=) Prior crizotinib (n=36)* No mutation (n=27) 1 mutation (n=8) Prior noncrizotinib TKI or 2 TKIs (n=5) No mutation (n=4) 1 mutation (n=1) CR 2 (11.8) 1 (3.7) Percentage change in tumour size from baseline in patients with 1 ROS1 kinase domain mutation Best overall response Partial response Stable disease Progressive disease PR 9 (52.9) 8 (29.6) 2 ( 25.) SD 5 (29.4) 8 (29.6) 5 (62.5) 3 (75.) 1 (1) PD 1 (5.9) 3 (11.1) 1 (12.5) 1 (25.) Indeterminate 7 (25.9) Responders, n (%) 11 (64.7) 9 (33.3) 2 (25.) *One patient had a non-analysable or uninformative sample ROS1 kinase domain mutation in: cfdna Tumour tissue Both cfdna and tumour tissue Conclusions ROS1 kinase domain mutations occurred in 1 14% of samples In patients with ROS1 kinase domain resistance mutations, lorlatinib exhibited some anti-tumour activity Solomon BJ, et al. Ann Oncol 218;29(suppl 5):Abstr 138PD

48 1381PD Gefitinib with or without pemetrexed in nonsquamous non-small cell lung cancer with EGFR mutation: Final overall survival results from a randomized phase II study Chih-Hsin Yang J, et al Study objective To investigate the efficacy and safety of pemetrexed + gefitinib vs. gefitinib in non-squamous NSCLC with EGFR mutations Key patient inclusion criteria Stage IV or recurrent non-squamous NSCLC Activating EGFR mutations East Asian 18 years ( 2 years in Japan and Taiwan) No prior systemic therapy for stage IV or recurrent NSCLC ECOG PS 1 (n=191) R 2:1 Pemetrexed 5 mg/m 2 q3w + gefitinib 25 mg/day (n=126) Gefitinib 25 mg/day (n=65) Primary endpoint PFS Secondary endpoints OS, time to progressive disease, tumour response rates, DoR, QoL, biomarker analysis, safety Chih-Hsin Yang J et al. Ann Oncol 218;29(suppl 5):abstr 1381PD

49 Probability Probability 1381PD Gefitinib with or without pemetrexed in nonsquamous non-small cell lung cancer with EGFR mutation: Final overall survival results from a randomized phase II study Chih-Hsin Yang J, et al Key results OS Median OS, months (95%CI) Pemetrexed + gefitinib 43.4 months (33.4, 5.8) Gefitinib 36.8 months (26.7, 42.6) Adjusted HR.77 (95%CI.5, 1.2) 1-sided p= Updated PFS Median PFS, months (95%CI) Pemetrexed + gefitinib 16.2 months (12.6, 18.7) Gefitinib 11.1 months (9.7, 13.8) Adjusted HR.67 (95%CI.5,.9) 1-sided p= No.at risk P+G G Time, months No.at risk P+G G Time, months (28.6%) patients had TEAEs in the pemetrexed + gefitinib group and 7 (1.8%) in gefitinib group Conclusion In patients with EGFR+ advanced NSCLC pemetrexed + gefitinib significantly prolonged PFS, with a numerically longer, but not statistically significant, improvement in OS Chih-Hsin Yang J et al. Ann Oncol 218;29(suppl 5):abstr 1381PD

50 1382PD: Phase III study of gefitinib (G) versus gefitinib+carboplatin+pemetrexed (GCP) as 1st-line treatment for patients (pts) with advanced non-small cell lung cancer (NSCLC) with EGFR mutations (NEJ9) Seike M, et al Study objective To investigate the efficacy and safety of gefitinib + carboplatin + pemetrexed vs. gefitinib in patients with non-squamous NSCLC Induction phase Maintenance phase Key patient inclusion criteria Treatment naïve patients with non-squamous IIIB/IV NSCLC EGFR+ R 1:1 Gefitinib 25 mg/day + carboplatin AUC6 + pemetrexed 5 mg/m 2 Stratification Sex EGFR mutation Smoking history Gefitinib + pemetrexed q3w PS 1 (n=345) Gefitinib 25 mg/day Platinumbased regimen Primary endpoints Secondary endpoints PFS, PFS2, OS ORR, safety, QoL Seike M, et al. Ann Oncol 218;29(suppl 5):Abstr 1382PD

51 Progression-free survival, % Overall survival, % 1382PD: Phase III study of gefitinib (G) versus gefitinib+carboplatin+pemetrexed (GCP) as 1st-line treatment for patients (pts) with advanced non-small cell lung cancer (NSCLC) with EGFR mutations (NEJ9) Seike M, et al Results PFS Updated OS No. at risk Gefitinib Gef + carb + pem Time, months No. at risk Gefitinib Gef + carb + pem Time, months Gefitinib Gefitinib + carb + pem Median PFS, months (95%CI) 11.2 (9., 13.4) 2.9 (18., 24.) HR (95%CI); p-value.49 (.388,.62); <.1 Gefitinib Gefitinib + carb + pem mos, months (95% CI) 38.8 (31.1, 47.3) 5.9 (41.8, 62.5) HR (95%CI); p-value.72 (.58,.95);.2 Conclusions PFS and OS were significantly improved in patients treated with combination therapy of gefitinib + carboplatin + pemetrexed There was no difference in PFS2 Seike M, et al. Ann Oncol 218;29(suppl 5):Abstr 1382PD

52 1385PD: A randomised phase III trial evaluating the addition of denosumab to standard first-line treatment in advanced NSCLC the ETOP and EORTC SPLENDOUR trial Peters S, et al Study objective To investigate the efficacy of denosumab as an add-on to standard 1L doublet chemotherapy + BSC in patients with NSCLC in the SPLENDOUR study Denosumab 12 mg every 3 4 weeks + chemotherapy (4 6 cycles) (n=255) Key patient inclusion criteria Stage IV NSCLC (n=514) R 1:1 Stratification Bone metastases PS Histology Region Chemotherapy (4 6 cycles) + BSC* (n=259) Primary endpoint OS Secondary endpoints PFS, OS by bone metastases, safety *Zoledronic acid in case of skeletal involvement Peters S, et al. Ann Oncol 218;29(suppl 5):Abstr 1385PD

53 OS, % PFS, % 1385PD: A randomised phase III trial evaluating the addition of denosumab to standard first-line treatment in advanced NSCLC the ETOP and EORTC SPLENDOUR trial Peters S, et al Key results OS PFS 1 8 Denosumab BSC 1 8 Denosumab BSC No. at risk Denosumab BSC Years No. at risk Denosumab BSC Months Treatment Events/ patients Denosumab 177/258 BSC 178/255 Non-parametric Median, years (95%CI).68 (.62,.87).73 (.63,.92) 1 year, % (95%CI) 4.2 (33.8, 46.5) 4.2 (33.8, 46.5) HR (95%CI).96 (.78, 1.18) Cox model p-value (Score test) Treatment Events/ patient Denosumab 228/259 BSC 227/254 Non-parametric Median, years (95%CI).39 (.35,.44).39 (.34,.44) 1 year, % (95%CI) 13.1 (9.1, 17.9) 9.3 (6., 13.6) HR (95%CI).97 (.81, 1.17) Cox model p-value (Score test) Peters S, et al. Ann Oncol 218;29(suppl 5):Abstr 1385PD

54 OS, % OS, % 1385PD: A randomised phase III trial evaluating the addition of denosumab to standard first-line treatment in advanced NSCLC the ETOP and EORTC SPLENDOUR trial Peters S, et al Key results (cont.) 1 8 OS Bone metastases Denosumab BSC 1 8 OS No bone metastases Denosumab BSC No. at risk Denosumab BSC N Months Months No. at risk Denosumab BSC N Treatment Events/ patients Denosumab 98/138 BSC 98/137 Non-parametric Median, years (95%CI).62 (.46,.73).61 (.51,.82) 1 year, % (95%CI) 34.4 (26.1, 43.) 35.8 (27.5, 44.2) HR (95%CI) 1.3 (.78, 1.37) Cox model p value (Score test) Treatment Events/ patients Denosumab 79/121 BSC 8/118 Non-parametric Median (years) (95%CI).89 (.67, 1.35).92 (.71, 1.5) 1 year, % (95%CI) 46.9 (37.2, 55.9) 45.4 (35.7, 54.6) Hazard Ratio (95%CI).89 (.65, 1.22) Cox model P-Value (Score test) Conclusions Denosumab added to 1L platinum-based chemotherapy did not improve OS There were no safety concerns with denosumab Peters S, et al. Ann Oncol 218;29(suppl 5):Abstr 1385PD

55 Advanced NSCLC Not radically treatable stage III and stage IV Later lines

56 LBA63: Long-term survival in patients (pts) with advanced NSCLC in the KEYNOTE-1 study overall and in pts who completed 2 years of pembrolizumab (pembro) Herbst RS, et al Study objective Long-term follow-up of KEYNOTE-1 investigating pembrolizumab vs. docetaxel in patients with advanced NSCLC and PD-L1 TPS 1% who had progressed after platinum-containing chemotherapy Key patient inclusion criteria Advanced NSCLC PD-L1 TPS 1% Progression after 1 line of chemotherapy No active brain metastases Endpoints in TPS 5% and 1% populations OS, PFS ORR, DoR, safety R 1:1:1 Pembrolizumab 2 mg/kg q3w for 24 months Pembrolizumab 1 mg/kg q3w for 24 months Docetaxel 75 mg/m 2 q3w Stratification ECOG PS ( vs. 1) Region (East Asia vs. non-east Asia) PD-L1 status (TPS 5% vs. 1 49%) Herbst RS, et al. Ann Oncol 218;29(suppl 5):Abstr LBA63

57 Overall survival, % Overall survival, % LBA63: Long-term survival in patients (pts) with advanced NSCLC in the KEYNOTE-1 study overall and in pts who completed 2 years of pembrolizumab (pembro) Herbst RS, et al Key results PD-L1 TPS 5% population OS PD-L1 TPS 1% population N Events, n (%) Median OS, mo (95%CI) HR (95%CI); p-value N Events, n (%) Median OS, mo (95%CI) HR (95%CI); p-value Pembrolizumab (69) 16.9 (12.3, 21.4) Docetaxel (84) 8.2 (6.4, 9.8).53 (.42,.66); <.1 Pembrolizumab (79) 11.8 (1.4, 13.1) Docetaxel (86) 8.4 (7.6, 9.5).69 (.6,.8); < % 13% % 11% No. at risk Pembro Docetaxel Time, months No. at risk Pembro Docetaxel Time, months Herbst RS, et al. Ann Oncol 218;29(suppl 5):Abstr LBA63

58 LBA63: Long-term survival in patients (pts) with advanced NSCLC in the KEYNOTE-1 study overall and in pts who completed 2 years of pembrolizumab (pembro) Herbst RS, et al Key results (cont.) Overall, median treatment duration was 3.5 months (range ) in the pembrolizumab group (n=682) and 2. months (range ) in the docetaxel group (n=39) 79 patients completed 35 cycles or 2 years of pembrolizumab with a median follow-up of 43.4 months (range ) 75/79 (95%) patients had a CR or PR 48/75 (64%) patients had ongoing response, median DoR NR (range months) Median OS was NR (95%CI NR, NR) The Kaplan-Meier estimate of 36-month OS rate was 98.7% (95%CI 91.2, 99.8) 25/79 (32%) patients had PD after stopping 35 cycles or 2 years of pembrolizumab In 14 patients who received a second course of pembrolizumab after 35 cycles or 2 years of treatment and subsequent PD, 6 (43%) had PR and 5 (36%) had SD Conclusions In patients with PD-L1-expressing advanced NSCLC, pembrolizumab continued to prolong OS compared with docetaxel In patients who completed 2 years of pembrolizumab AEs were manageable and responses were durable Herbst RS, et al. Ann Oncol 218;29(suppl 5):Abstr LBA63

59 1378O: ARCTIC: durvalumab + tremelimumab and durvalumab monotherapy vs SoC in 3L advanced NSCLC treatment Kowalski DM, et al Study objective To investigate the efficacy and safety of durvalumab vs. SoC and the combination of durvalumab + tremelimumab vs. SoC in subgroups based on PD-L1 expression ( 25% or <25%) in the ARCTIC study Key patient inclusion criteria Stage IIIB/IV NSCLC 2 prior treatments Immunotherapy-naïve Stratification Planned SoC, histology Study A PD-L1 TC 25% (n=126) Study B EGFR/ALK wild-type PD-L1 TC WHO PS /1 <25% SoC* (n=469) Primary endpoint OS, PFS (investigator assessed R Durvalumab 1 mg/kg q2w for up to 12 mo (n=62) 1:1 SoC* (n=64) R 3:2:2:1 Durvalumab 2 mg/kg + tremelimumab 1 mg/kg q4w for 12 weeks then durvalumab 1 mg/kg for 34 weeks (n=174) (n=118) Durvalumab 1 mg/kg q2w for up to 12 mo (n=117) Tremelimumab 1 mg/kg q4w for 24 weeks then q12w for 24 weeks (n=6) Secondary endpoints 1-year OS and PFS rates, ORR, safety *Erlotinib, gemcitabine, or vinorelbine Kowalski DM, et al. Ann Oncol 218;29(suppl 5):Abstr 1378O