Recent Therapeutic Advances in Squamous Cell Carcinoma of the Lung

Transcription

1 Recent Therapeutic Advances in Squamous Cell Carcinoma of the Lung 2017 Conversations in Oncology in Shanghai, China Prof. Shun Lu Shanghai Lung Cancer Center Shanghai Chest Hospital Affiliated to Medical College of Shanghai Jiaotong University BI Symposium

2 Squamous Cell Carcinoma (SqCC) of the Lung Squamous histology represents approximately 20 40% of NSCLC 1,2 SqCC of the lung remains a disease with high unmet medical need SqCC of the lung is associated with poor prognosis 3 Median OS after diagnosis of advanced disease is around 4 months for untreated patients 3 The 5-year survival is ~1.6% 3 Better therapeutic options needed for patients Targetable oncogenic alterations are few and have not yet translated to a therapeutic paradigm OS = overall survival; NSCLC = non-small cell lung cancer. 1. Ho C et al. Curr Oncol. 2015;22:e16 4-e170; 2. Bryant A and Cerfolio RJ. Chest. 2007;132:185-92; 3. Cetin K et al. Clin Epidemiol. 2011;3:

3 Overview of Recent Key Phase III Second-line Treatment Studies in Patients With SqCC of the Lung Trial Treatment Median PFS (mo) HR for PFS Median OS (mo) HR for OS ORR (%) BR Erlotinib vs placebo (n=731) Squamous (n=222) 2.2 vs 1.8 NR for squamous 0.61 NR for squamous 6.7 vs vs % vs <1% NR for squamous REVEL 4 Ramucirumab + doce vs doce (n=1253) Squamous (n=328) 4.5 vs vs * 0.76* 10.5 vs vs * vs 13.6* 26.8 vs 10.5* CheckMate Nivolumab vs doce All squamous (n=272) 3.5 vs * 9.2 vs * 20.0 vs 9.0* KEYNOTE Pembrolizumab vs doce PD-L1 PS 50% (n=442) Squamous (n=222) 2 mg: 5.0 vs mg: 5.2 vs 4.1 NR for squamous 2 mg: 0.59* 10 mg: 0.59* vs vs 8.2 NR for squamous 0.54* 0.50* vs 8.0* 29.0 vs 8.0* NR for squamous LUX-Lung 8 7 Afatinib vs erlotinib (n=795) All squamous 2.6 vs * 7.9 vs * 6.0 vs 2.8* OAK 8 Atezolizumab vs doce (n=850) Squamous (n=222) 2.8 vs 4.0 NR for squamous 0.95 NR for squamous 13.8 vs vs * 0.73* 14 vs 13 NR for squamous All agents listed are FDA and EMEA approved for the treatment of SqCC of the lung. *P<0.05. doce = docetaxel; EMEA = European Medicines Agency; FDA = US Food and Drug Administration; HR = hazard ratio; mo = months; NR = not reported; ORR = objective response rate; OS = overall survival; PD-L1 = programmed death-ligand 1; PFS = progression-free survival; PS = proportion score; SqCC = squamous cell carcinoma. 1. Shepherd FA et al. N Engl J Med. 2005;352: ; 2. Clark GM. Mol Oncol. 2008;1: ; 3. Gridelli C et al. The Oncologist. 2007;12(7): ; 4. Garon E et al. Lancet. 2014;384:665-73; 5. Brahmer J et al. N Engl J Med. 2015;373:123-35; 6. Herbst R et al. Lancet. 2016;387: ; 7. Soria JC et al. Lancet Oncol. 2015;16: ; 8. Rittmeyer A et al. Lancet. 2017;389:

4 Chemotherapy EGFR-TKI/Ab I-O Anti-VEGF 4

5 CA031: Weekly nab-paclitaxel + Carboplatin vs Paclitaxel + Carboplatin ITT Squamous Subgroup ORR (%) nab-pc (N=521) sb-pc (N=531) OR 95% CI P ITT SQCC 41 (N=229) 24 (N=221) <0.001 Non-SQCC 26 (N=292) 25 (N=310) Socinski MA et al. J Clin Oncol. 2012;30:

6 CA 031: Better Safety Profile of nab-paclitaxel AE nab-pc (%) N=514 sb-pc (%) N=524 G3 G4 G3 G4 P-value Hematological Neutropenia <0.001* Thrombocytopenia <0.001 Anemia <1 <0.001 Febrile Neutropenia <1 <1 1 <1 N/S Non-Hematological Fatigue 4 <1 6 <1 N/S Sensory Neuropathy <1 <0.001* Anorexia 2 0 <1 0 N/S Nausea <1 0 <1 0 N/S Myalgia < * Arthralgia * N/S = non-significant difference; *P<0.05: nab-p better; P<0.05; Paclitaxel better. Socinski MA et al. J Clin Oncol. 2012;30:

7 CA 031: nab-paclitaxel Pros and Cons 7

8 WJOG5208L: Nedaplatin vs Cisplatin Shukuya T et al ASCO Abstract

9 WJOG5208L: Primary Endpoints OS N + D C + D Shukuya T et al ASCO Abstract

10 Nedaplatin Plus Docetaxel Versus Cisplatin Plus Docetaxel as First-Line Chemotherapy for Advanced Squamous Cell Carcinoma of the Lung Multicenters, Open-label, Randomized, Phase III, Superiority Trial (JUST Study) Patient Disposition Primary endpoint: PFS Secondary endpoint: ORR, TTP, AE Assessed for eligibility Randomized (N=286) Stratification factors: Gender Stage ECOG PS 1 withdrew before treatment - 1 protocol violation 11 withdrew before first evaluation - 4 no visiting as plan - 1 symptom deterioration - 1 protocol violation - 5 consent withdrawal ND group (N=142) 141 cases received treatment (ITT analysis) CD group (N=144) 139 cases received treatment (ITT analysis) 130 cases (PPS analysis) 126 cases (PPS analysis) 5 withdrew before treatment - 5 consent withdrawal 13 withdrew before first evaluation - 4 no visiting as plan - 6 consent withdrawal - 3 PI decision Presented by Shun Lu at WCLC,

11 Nedaplatin Plus Docetaxel Versus Cisplatin Plus Docetaxel Primary Endpoint: PFS FAS (P=0.056) PPS (P<0.05) N C Mon N C Mon *Log-Rank two-sides test. Presented by Shun Lu at WCLC,

12 Nedaplatin Plus Docetaxel Versus Cisplatin Plus Docetaxel Best Response Evaluation Recist Ver. 1.1 ND (N=133) CD (N=128) CR 1 0 PR SD PD 7 24 ORR 51.1% 37.5% DCR 94.7% 81.3% P* value c 2 test. Presented by Shun Lu at WCLC,

13 Summary There was no significant difference of PFS between ND group and CD group in the FAS, however significant difference existed in the PPS More hematological toxicity and non-hematological toxicity were observed in the ND group and CD group, respectively Nedaplatin plus docetaxel could be a new treatment option for advanced or relapsed squamous cell lung cancer Presented by Shun Lu at WCLC,

14 Chemotherapy EGFR-TKI/Ab I-O Anti-VEGF 14

15 Somatic Mutation Prevalance (Number Mutations per Megabase) SCC NSCLC: Genetically Complex Malignancy High burden of somatic mutations/genomic alterations Pilocytic astrocytoma ALL Medulloblastoma AML Kidney chromophobe Thyroid CLL Neuroblastoma Glioblastoma Pancreas Breast Gliona low grade Lymphoma B-cell Myeloma Prostate Ovary Kidney papillary Kidney clear cell Liver Uterus Stomach Head and neck Cervix Colorectum Esophagus Lung small cell Bladder Lung adenocarcinoma Lung squamous Melanoma Overexpression/derangements of EGFR, 2,3 HER2, 4,5 HER4, 6 and/or dysregulation of their downstream pathways implicated in the pathogenesis of SCC NSCLC 1. Lawrence et al. Nature. 2013;499:214; 2. López-Malpartida et al. Lung Cancer. 2009;65:23; 3. Hirsch et al. J Clin Oncol. 2003;21:3796; 4. Heinmoller et al. Clin Cancer Res. 2003;9:5238; 5. Ugocsai et al. Anticancer Res. 2005;25:306; 6. Cancer Genome Atlas Research Network. Nature. 2012;489:

16 Multi-gene Alterations: More Common in Sq NSCLC Than ADC 1. Chen Z et al. Nat Rev Cancer. 2014;14: ; 2. Pikor LA et al. Lung Cancer. 2013;82: ; 3. Shi Y et al. J Thorac Oncol. 2014;9: ; 4. Pao W et al. Lancet Oncol. 2011;12:

17 Different Gene Alternation Fernandez-Cuesta L, McKay JD. Curr Opin Oncol. 2016;28(1):

18 TAILOR: Erlotinib vs Docetaxel in Second-Line Treatment for Advanced NSCLC Median overall survival: Erlotinib: 5.4 mo (95% CI: ) Docetaxel: 8.2 mo (95% Cl: ) HR=0.73 (95% CI ); P=0.05 Garassino MC et al. Lancet Oncol. 2013;14:

19 TAILOR: Erlotinib vs Docetaxel in Second-Line Treatment for Advanced NSCLC OS PFS Garassino MC et al. Lancet Oncol. 2013;14:

20 Survival Probability Docetaxel in Squamous vs Nonsquamous Histology IPD Pooled Analysis (TAILOR, DELTA, PROSE) 1.0 Overall Survival Nonsquamous Squamous Median survival: nonsquamous 10.9 mo (95% CI: ) Median survival: squamous 6.3 mo (95% CI: ) Torri et al. ASCO Abstract Time (months) Nonsquam Squamous

21 BR 21: Erlotinib vs Placebo in Second-Line Treatment for Advanced NSCLC Overall Survival Overall Survival in Patients With Squamous Histology Erlotinib Placebo Median OS (months) Erlotinib 150 mg QD (n=144) Placebo (n=78) HR 0.67 (95% CI: ) Time (months) CI = confidence interval; HR = hazard ratio; OS = overall survival; QD = once daily. 1. Shepherd FA et al. N Engl J Med. 2005;353: ; 2. Clark GM. Mol Oncol. 2008;1:

22 ErbB Receptor Family is a Valid Therapeutic Target for SqCC of the Lung Dysregulation of the ERBB pathway is frequently observed in SqCC of the lung EGFR overexpression and gene amplification aberrations of other ErbB receptors, and dysregulation of the downstream pathways, have all been implicated in the pathobiology of SqCC 1,2 These findings likely account for the benefits these patients derive from erlotinib and other EGFR-directed therapies in different treatment settings, despite the low frequency of EGFR-activating mutations 17 ErbB Receptor Frequency (%) EGFR overexpression EGFR amplification 2, EGFRvIII mutation 6 5 EGFR kinase domain mutation 7 <5% ERBB2 mutation/amplification 2 5 ERBB3 mutation 8 1 ERBB3 overexpression 9 10 ERBB Frequency of known genetic drivers in SqCC 17 EGFRvIII PI3KCA EGFR DDR2 FGFR1 Amp Unknown ~5% Amp = amplification; EGFR = epidermal growth factor receptor; FGFR = fibroblast growth factor receptor; SqCC = squamous cell carcinoma. 1. Hirsch FR et al. J Clin Oncol. 2003;21: ; 2. Lopez-Malpartida AV et al. Lung Cancer. 2009;65:25-33; 3. Lee HJ et al. Lung Cancer. 2010;68:375-82; 4. Gately K et al. Clin Lung Cancer. 2014;15:58-66; 5. Dacic S et al. Am J Clin Pathol. 2006;125:860-5; 6. Ji H et al. Proc Natl Acad Sci U S A. 2006;103: ; 7. Dearden S et al. Ann Oncol. 2013;24:2371-6; 8. Jaiswal BS et al. Cancer Cell. 2013;23:603-17; 9. Gorgoulis V et al. Pathol Res Pract. 1995;191:973-81; 10. Kan Z et al. Nature. 2010;466:869-73; 11. Shepherd FA et al. N Engl J Med. 2005;352:123-32; 12. Clark GM et al. Clin Lung Cancer. 2006;7:389-94; 13. Leon et al. ESMO Abstract 1277 (poster); 14. Pirker R et al. Lancet. 2009;373: ; 15. Pirker R et al. Lancet Oncol. 2012;13:33-42; 16. Thatcher N et al. ASCO Abstract 8008; 17. Li T et al. J Clin Oncol. 2013;31:

23 Afatinib is the First Irreversible ErbB Family Blocker Afatinib covalently binds and irreversibly blocks EGFR, HER2, and ErbB4 Targeting the whole ErbB Family enhances the effect on important signalling pathways Li et al. Oncogene. 2008;27:4702; Solca et al. J Pharmacol Exp Ther. 2012;343:

24 LUX-Lung 8: Study Design Advanced SqCC NSCLC (Stage IIIB/IV) PD after 4 cycles of a first-line platinum doublet ECOG PS 0 or 1 No prior anti-egfr therapy No active brain metastases Randomisation 1:1 (N=795) Afatinib (n=398) 40 mg qd Erlotinib (n=397) 150 mg qd Treatment until disease progression or unacceptable AEs Stratification: East Asian vs non-east Asian Tumour tissue collected for correlative science Radiographic tumour assessment at baseline; Weeks 8, 12, 16; every 8 weeks thereafter Primary endpoint: PFS; key secondary endpoint: OS AE = adverse event; EGFR = epidermal growth factor receptor; ECOG PS = Eastern Cooperative Oncology Group performance status; NSCLC = non-small cell lung cancer; OS = overall survival; PD = disease progression; PFS = progression-free survival; qd = once daily; SqCC = squamous cell carcinoma. 1. Soria JC et al. Lancet Oncol. 2015;16:

25 Probability of PFS (%) Probability of OS (%) LUX-Lung 8: Significant Improvement in PFS and OS With Afatinib Compared With Erlotinib Updated PFS analysis by Independent Review (n=795) No. at risk Time (months) Afatinib Erlotinib Afatinib 40 mg QD (n=398) Afatinib Erlotinib Erlotinib 150 mg QD (n=397) Primary analysis of OS (key secondary endpoint) (n=795) % 22.0% 28.2% % No. at risk Time (months) Afatinib Erlotinib Afatinib 40 mg QD (n=398) Afatinib Erlotinib Erlotinib 150 mg QD (n=397) Patients progressed or died, n (%) 299 (75.1) 306 (77.1) Median PFS (months) Patients died, n (%) 307 (77.1) 325 (81.9) Median OS (months) CI Erlotinib = confidence interval; 397 HR = hazard 99 ratio; OS = overall 34 survival; PFS 17 = progression-free 10 survival; 2 QD = once daily Soria JC et al. Lancet Oncol. 2015;16: HR 0.81; 95% CI: ; P= HR 0.81; 95% CI: ; P=

26 LUX-Lung 8: Consistent OS Benefit With Afatinib Over Erlotinib Across All Subgroups Factors No. of Patients HR (95% CI) Overall ( ) Age <65 years ( ) 65 years ( ) Gender Male ( ) Female ( ) Race Non-East Asian ( ) East Asian ( ) ECOG at baseline ( ) ( ) Smoking history Never smoker ( ) Light ex-smoker ( ) Current and other ex-smoker ( ) Histology Squamous ( ) Mixed ( ) Best response to first-line chemotherapy CR/PR ( ) SD ( ) Unknown ( ) Soria JC et al. Lancet Oncol. 2015;16: /16 1/ Favours Afatinib Favours Erlotinib 26

27 Maximum Decrease From Baseline SLD (%) Percent Maximum Decrease From Baseline SLD (%) Greater and More Durable Tumour Control With Afatinib Compared With Erlotinib Afatinib P= Afatinib Erlotinib % increase (n=62) 0 <20% increase (n=90) >0 <30% decrease (n=81) 30% decrease (n=22) Patient Index Sorted by Maximum Decrease (%) DCR Soria et al. Lancet Oncol. 2015;16: P= ORR Duration of response was 7.29 months for afatinib and 3.71 months for erlotinib Erlotinib % increase (n=74) 80 0 <20% increase (n=101) 60 >0 <30% decrease (n=77) 40 30% decrease (n=13) Patient Index Sorted by Maximum Decrease (%) 27

28 LUX-Lung 8: Similar Adverse Event Profile for Afatinib and Erlotinib Consistent With EGFR Inhibition Afatinib AE profile was consistent with its mechanistic profile and was manageable Pattern of AEs was consistent in both arms with similar rates of severe, serious, and fatal AEs Events Afatinib (n=392) (%) Erlotinib (n=395) (%) Any AE Drug-related AEs AEs leading to dose reduction AEs leading to discontinuations CTCAE grade 3 or higher Serious AEs Drug-related fatal AEs 1.5 a 1.3 b a Interstitial lung disease (n=2), pneumonia, respiratory failure, acute renal failure, and general physical health deterioration (1 patient each). b Interstitial lung disease, pneumonitis, pneumonia, intestinal obstruction, and peritonitis (1 patient each). Soria et al. Lancet Oncol. 2015;16:

29 Disease-Related Symptom Relief and Quality of Life Significantly more patients had improved overall health-related quality-of-life (36% vs 28%; P=0.041) and improvement in cough (43% vs 35%; P=0.029) than with erlotinib Afatinib significantly delayed time to deterioration of dyspnoea compared with erlotinib (median 2.6 months [95% CI ] vs 1.9 months [ ]; HR 0.79 [95% CI ]; P=0.0078) Coughing (Q1 from QLQ-LC13) Dyspnoea Dyspnea (Q3 Q5 from QLQ-LC13) Symptom Improvement Afatinib 35.2 Erlotinib 43.4 P= P=0.06 Coughing (Q1 from QLQ-LC13) Dyspnoea (Q3 Q5 from QLQ-LC13) Time to Deterioration No. of Patients HR (95% Cl) ( ) ( ) Pain (Q9, Q19 from QLQ-C30) P=0.78 Pain (Q9, Q19 from QLQ-C30) ( ) GHS/QoL (Q29 Q30 from QLQ-C30) P=0.04 GHS/QoL (Q29 Q30 from QLQ-C30) ( ) GHS = global health status. Soria et al. Lancet Oncol. 2015;16: Patients With Improvement in Symptoms (%) 1/4 1/ Favours afatinib Favours erlotinib 29

30 Estimated OS probability Post-hoc Analysis of LUX-Lung 8 Long-term Responders (LTRs) OS: Primary Analysis (ITT population) Afatinib (n=398) Erlotinib (n=397) 12 and 18-month OS rates indicated that some patients derived prolonged benefit with afatinib Post-hoc analysis identified 21 patients (LTRs) who received 12 months of afatinib treatment % 36.4% 14.4% Time (months) 22.0% Median treatment duration was 17.6 months (range: months) Possible molecular/clinical biomarkers indicative of long-term response to afatinib were evaluated Baseline characteristics Efficacy/safety of afatinib Molecular genomic analysis VeriStat classification* *Serum protein test used to assign a Good or Poor classification, with prognostic and predictive utility for EGFR-targeted agents in NSCLC2. 1. Yang J et al. ELCC Poster #102P. 30.

31 Treatment Response* and OS in LTRs Median OS was 21.1 months (range: months) Median PFS (independent central review) was 16.6 months (range: months) *Stable disease unless noted otherwise (patient 2 was classified as non-evaluable); Patients were ordered and numbered by treatment duration, with patient 1 being on treatment longest; First observed response at time of tumour measurement; Last observed response at time of tumour measurement; Treatment ongoing until death; Received 1 line of chemotherapy after afatinib; CR = complete response; PR = partial response. Yang J et al. ELCC Poster #102P. 31

32 Genomic Aberrations in LTRs ErbB family mutations were more frequent in LTRs than in the overall afatinib-treated population LTRs (n=10*) All afatinib-treated patients (n=132*) ErbB2, 20.0% ErbB3, 0% ErbB4, 10.0% EGFR, 6.8% ErbB2, 6.8% ErbB3, 4.6% ErbB4, 2.3% EGFR, 20.0% ErbB WT, 50.0% ErbB WT, 81.1% *Next-generation sequencing was undertaken in 10/21 LTRs and 132/398 afatinib-treated patients overall; WT = wild-type. Yang J et al. ELCC Poster #102P. 32

33 VeriStrat Analysis of LUX-Lung 8 to Predict Treatment Benefit Between Afatinib vs Erlotinib VeriStrat is a predictive and prognostic serum-based protein test for patients with advanced NSCLC who test negative for EGFR mutations or whose EGFR mutation status is unknown Commercially available test (Biodesix) designed to identify which patients are likely to benefit from an oral EGFR tyrosine kinase inhibitor The test classifies patients as either VeriStrat Good (VS-G) or VeriStrat Poor (VS-P) In addition to being prognostic, VeriStrat is predictive of differential treatment benefit when selecting between single-agent treatment options Main Objectives of LUX-Lung 8 Analysis: 1. To assess the utility of VeriStrat as a predictive test of differential clinical benefit between afatinib and erlotinib 2. To investigate the ability of VeriStrat to stratify patients treated with afatinib as a single agent 33

34 Afatinib Significantly Improved OS and PFS Over Erlotinib in VS-G Patients Gadgeel S et al. Lung Cancer. 2017;109(2017):

35 In the Entire VeriStrat-classified Population, VSV-G Patients Had Better OS and PFS Gadgeel S et al. Lung Cancer. 2017;109(2017):

36 Anti-EGFR Mab: Necitumumab Fully human anti-egfr monoclonal antibody OS 11.5 VS 9.9 MO HR = 0.84; P=0.01 SQUIRE 1 Cisplatin/gemcitabine ± necitumumab in squamous NSCLC Primary endpoint: survival INSPIRE 2 Cisplatin/pemetrexed ± necitumumab in non-squamous NSCLC Primary endpoint: survival Enrolment (n=634) was prematurely closed due to thromboembolic events Thatcher N et al. Lancet Oncol. 2015;16(7):763-74; 2. Paz-Ares L et al. Lancet Oncol. 2015;16(3):

37 SQUIRE: EGFR H Score 200 OS Paz-Ares L et al. Lancet Oncol. 2015;16:

38 Chemotherapy EGFR-TKI/Ab I-O Anti-VEGF 38

39 CheckMate-017 (Nivolumab vs Docetaxel): OS Felip E et al. Ann Oncol. 2017;28(Supplement 5):abstract 1301PD; Horn L et al. J Clin Oncol. 2017;doi: /JCO [Epub ahead of print]. 39

40 CheckMate-017: OS in PD-L1 Subgroups Felip E et al. Ann Oncol. 2017;28(Supplement 5):abstract 1301PD. 40

41 CheckMate-017: PFS a Investigator-assessed. NC = not calculable. Felip E et al. Ann Oncol. 2017;28(Supplement 5):abstract 1301PD. 41

42 KEYNOTE-010: Pembrolizumab vs Docetaxel OS in Patients With PD-L1 TPS 50% 1 Treatment arm Median (95% CI), months HR (95% CI) P Pembrolizumab 2 mg/kg 14.9 (10.4 NR) 0.54 ( ) Pembrolizumab 10 mg/kg 17.3 (11.8 NR) 0.50 ( ) < Docetaxel 8.2 ( ) Post hoc multivariate analyses showed that nonsquamous histology was associated with better OS among patients treated with pembrolizumab (nonsquamous vs squamous histology HR 0.55 [ ; P<0.0001]) 2 CI = confidence interval; HR = hazard ratio; OS = overall survival; PD-L1 = programmed death-ligand 1; TPS = tumour proportion score. 1. Herbst RS et al. Lancet. 2016;387: ; 2. Herbst RS et al. J Clin Oncol. 2017;35(suppl; abstr 9090). 42

43 KEYNOTE 010: OS in Key Subgroups (PD-L1 TPS 1 %*) *Data for pembrolizumab were pooled. CI = confidence interval; ECOG = Eastern Cooperative Oncology Group; EGFR = epidermal growth factor receptor; OS = overall survival; PD-L1 = programmed death-ligand 1; TPS = tumour proportion score. Herbst RS et al. Lancet. 2016;387:

44 KEYNOTE-024: Pembrolizumab Versus Chemotherapy in 1L Advanced NSCLC 1 2 Key eligibility criteria Untreated Stage IV NSCLC PD-L1 TPS 50% Pembrolizumab 200 mg IV Q3W (2 years) ECOG PS 0 1 No activating EGFR mutation or ALK translocation No untreated brain metastases No active autoimmune disease requiring systemic therapy R (1:1) N=305 Platinum-doublet chemotherapy (4 6 cycles) PD * Crossover to Pembrolizumab 200 mg Q3W for 2 years Key endpoints Primary: PFS (RECIST v1.1 per blinded, independent central review) Secondary: OS, ORR and safety Exploratory: DOR *To be eligible for crossover, PD had to be confirmed by blinded, independent central radiology review and all safety criteria had to be met. ALK = anaplastic lymphoma kinase; DOR = duration of response; ECOG PS = Eastern Cooperative Group performance status; EGFR = epidermal growth factor receptor; NSCLC = non-small cell lung cancer; PD = disease progression; ORR = objective response rate; OS = overall survival; PFS = progression free survival; R = randomization; PD-L1 = programmed death ligand 1; Q3W = every three weeks; TPS = tumor proportion score. 1. Reck M et al. Ann Oncol. 2016;27(suppl 6):abstract LBA08; 2. Reck M et al. N Engl J Med. 2016;375:

45 KEYNOTE-024: Pembrolizumab Versus Platinumdoublet Chemotherapy 1 3 PFS (primary endpoint) OS (secondary endpoint) Update from ASCO 2017, abstract 9000 Chemo = chemotherapy; CI = confidence interval; HR = hazard ratio; mo = months; OS = overall survival; PFS = progression free survival; Pembro = pembrolizumab. 1. Reck M et al. N Engl J Med. 2016;375: ; 2. NICE Single Technology Appraisal committee papers [ID990]; 3. Brahmer J et al. J Clin Oncol. 2017;35(suppl; abstr 9000). 45

46 Overall Survival (%) OAK Trial: Overall Survival, ITT (n=850) Minimum follow up = 19 months HR, 0.73 a (95% CI, 0.62, 0.87) P= Atezolizumab Docetaxel Median 9.6 mo (95% CI, 8.6, 11.2) Median 13.8 mo (95% CI, 11.8, 15.7) Months a Stratified HR. Rittmeyer A et al. Lancet. 2017;389(1066):

47 OAK: OS by PD-L1 Expression TC0 and IC0 45% TC3 or IC3 16% TC1/2/3 or IC1/2/3 54% TC2/3 or IC2/3 31% Subgroup Atezolizumab n=425 Median OS, mo Docetaxel n=425 TC3 or IC TC2/3 or IC2/ TC1/2/3 or IC1/2/3 a TC0 and IC ITT a a Stratified HR for ITT and TC1/2/3 or IC1/2/3. Rest unstratified. Barlesi et al. Atezolizumab Phase III OAK Study Hazard Ratio a In favour of atezolizumab In favour of docetaxel 47

48 Overall Survival (%) Overall Survival (%) OAK: OS by Histology Non-squamous 1 Squamous 1 Atezolizumab Docetaxel HR, 0.73 a (95% CI, 0.60, 0.89) P= Atezolizumab Docetaxel HR, 0.73 a (95% CI, 0.54, 0.98) P= Minimum follow up = 19 months Minimum follow up = 19 months Median 11.2 mo (95% CI, 9.3, 12.6) Median 15.6 mo (95% CI, 13.3, 17.6) Median 7.7 mo (95% CI, 6.3, 8.9) Median 8.9 mo (95% CI, 7.4, 128) Months Months In an analysis of long-term survivors (n=119) with a minimum follow-up of 26 months, 15.1% of patients had squamous histology, compared to 84.9% of patients with non-squamous histology 2 In non-long-term responders (n=279), 30.1% of patients had squamous histology, compared to 69.9% of patients with non-squamous histology 2 1. Rittmeyer A et al. Lancet. 2017;389(1066): ; 2. Satouchi M et al. WCLC. 2017;abstract OA

49 OAK: OS in Non-Squamous and Squamous NSCLC by BOR Subgroups de Marinis F et al. Annals of Oncology. 2017;28(Supplement 5) [poster 1310P]. 49

50 Chemotherapy EGFR-TKI/Ab I-O Anti-VEGF 50

51 Overall Survival REVEL: Ramucirumab Plus Docetaxel vs Docetaxel Overall Survival in the Squamous Histology Subgroup Ramucirumab + docetaxel Placebo + docetaxel Ramucirumab + Docetaxel 10 mg/kg + 75 mg/m 2 q3wk (n=157) Placebo + docetaxel 75 mg/m 2 q3wk (n=171) Median OS (months) HR 0.883; 95% CI: P= Patients at risk Ramucirumab + docetaxel Placebo + docetaxel Time (months) CI = confidence interval; HR = hazard ratio; OS = overall survival. Garon EB et al. Lancet. 2014;384:

52 REVEL: SQCC Subgroup PFS OS * *Difference in OS was non-significant The most common grade 3 or worse neutropenia, febrile neutropenia, fatigue, leucopenia and hypertension in the ramucirumab group. The numbers of deaths from adverse events and grade 3 or worse pulmonary haemorrhage did not differ between groups Garon EB et al. The Lancet. 2014;384(9944):

53 Second-line Therapy NCCN Guidelines Version Non-Small Cell Lung Cancer. NCCN NSCLC 2017V5 53

54 S1400 (MASTER LUNG-1) 2nd Line on SqNSCLC Biomarker Profiling (NGS/CLIA) Biomarker Nonmatch Multiple Phase II-III Arms With Rolling Opening and Closure CT PD-L1i MEDI4736 PiK3CA mut CCND1 amp or CDKN2 loss + RB WT FGFR amp, mut, fusion MET expr (IHC score) PIK3i CT CDK 4/6i CT FGFRi + CT CT HGFi + E E Primary Endpoint OS Primary Endpoint OS Primary Endpoint OS Primary Endpoint OS GDC0032 Palbociclib AZD4547 AMG102+E Project Chair; V. Papadimitrakopoulou; Steering Committee Chair: R. Herbst; SWOG Lung Chair: D. Gandara. 54

55 Current Treatment Recommendations for Metastatic SqCC of the Lung Stage IV SqCC Never or former light smoker (<15 pack/year) BSC (II, B) Molecular test (ALK/EGFR) Molecular test negative Molecular test positive Targeted therapy PS 3 4 PS 0 2 I) Age II) PS <70 years and PS cycles: Cisplatin gemcitabine (I, A) Cisplatin docetaxel (I, A) Cisplatin vinorelbine (I, A) Carboplatin paclitaxel (I, A) Carboplatin nab-paclitaxel (I, B) Cisplatin gemcitabine necitumumab (if EGFR expression by IHC) (I, B; MCBS 1) <70 years and PS 2 or >70 years and PS cycles: Carboplatin-based doublets (II, B) Single-agent chemotherapy (gemcitabine, vinorelbine or docetaxel) (I,A) Disease progression BSC PS 3 4 Nivolumab (I, A; MCBS 5) Pembrolizumab if PD-L1>1% (I, A; MCBS 3 if PD-L1 >1%; MSBC 5 if PD-L1 >50% Docetaxel (I, B) Ramucirumab docetaxel (I, B; MCBS 2) Erlotinib (II, C) Afatinib (I, C; MCBS 1) *ESMO guidelines do not recommend maintenance therapy in the treatment of squamous cell carcinoma NSCLC. 1 BSC = Best Standard of Care; EGFR = epidermal growth factor receptor; MCBS = Magnitude of Clinical Benefit Scale; NSCLC = non-small cell lung cancer; PD-L = programmed death-ligand; PS = Performance Status; SqCC = squamous cell carcinoma. 1. Novello S et al. Ann Oncol. 2016;27(suppl 5):v1 v27. 55

56 Summary Platinum doublet chemotherapy remains the gold standard of care for 1st line treatment of advanced lung SqCC Afatinib should be considered the treatment of choice in the 2nd line setting in certain populations (eg, patients who are not eligible for I-O therapy) as a treatment option in SqCC patients who have failed chemotherapy and I-O therapy Anti-angiogenic agents are generally contraindicated in SqCC due to safety concerns, with the exception of Ramucirumab Despite of the complex genomic profile of SqCC, some potentially functionable molecular targeted agents are under investigation I-O therapy has emerged as a promising novel treatment option for advanced SqCC I-O = immune checkpoint inhibitor; SqCC = squamous cell carcinoma. 56

57 For more information about other BI events and collaborations, please visit