Solange Peters, MD-PhD Cancer Center, Lausanne Switzerland EGFR AND NSCLC

Solange Peters, MD-PhD Cancer Center, Lausanne Switzerland EGFR AND NSCLC

Interactive Clinical Case

History Maria José 1963 Since end 2010, progressive dyspnea Attends Emergency Room 02.01.2011 after 3 days with cough and fever >39 PMH: None DH: None Current smoker: 35 pack years

White left lung Chest X-Ray Additional Evaluations Bronchoscopy: no endobronchial lesion but extrinsic compression of LLL main bronchus -> Lavage and brushing negative Blood tests and hemogram normal except crp 200mg/l -> ad antibiotherapy co-amoxicillin

First Intervention Pleural effusion drainage by thoracoscopy (>2l, citrine exudative fluid) Multiple pleural biopsies 20.1.2011 Dg: Lung adenocarcinoma Poorly differentiated TTF1 +, CK7+, CK 5/6-, p63 -

p63 TTF-1 Lung Adenocarcinoma

CT scan (ct3 cn0 pm1)

PET/CT scan ct3 cn0 pm1

Brain MRI Normal Very well Baseline Symptoms Fully active, 2 small kids Mild persisting dry cough Performance status zero

Stage IV Lung Adenocarcinoma Heavy smoker 1)Cisplatin-pemetrexed 2)Ciplatin-pemetrexed-bevacizmab 3)Other platinum doublet 4)Other doublet with bev 5)EGFR mutation testing 6)ALK FISH

Stage IV Lung Adenocarcinoma Heavy smoker EGFR Exon 19 deletion (E746-A750)

Chemo or EGFR-TKI? Overall Response Rate (%) Cis/Pem Cis/Gem Erlotinib Platinum Doublet Scagliotti et al, JCO 2008 Rosell et al, Lancet Oncol. 2012

Clinical Course Grade 0 Rash PS 0 No toxicity Ongoing active smoking (!)

Restaging CT scan Baseline Post 4wks

Restaging CT scan Post 6 months Post 1yr

Restaging CT scan Post 12 months months Post 20 months

Audience Question: How should treatment be continued? Continue EGFR TKI therapy with close follow up Chemoradiotherapy Chemotherapy Continue EGFR TKI + Thoracic radiosurgery Thoracic radiotherapy Surgery

Cancer genome and driver genes 3142 genes mutated in Cancer (Sanger/COSMIC database) 10% (286) of these are driver genes (10% oncogenes, 90% tumour supressor genes affected) Oncogenes About 33 genes described within 12 core pathway By definition, same codon mutated in >2 different tumors Tumor Suppressor Genes 15% of mutations leads to truncation Cannot be inhibited by current approaches

2012 update on adenocarcinoma driver mutations KRAS Unknown EGFR NRAS MEK1 MET PTEN HER2 BRAF PIK3CA

2012 update on adenocarcinoma driver gene alterations EML4-ALK ALK REA RET REA ROS REA PTEN LOSS PIK3CA AMP Unknown HER2 AMP MET AMP

2012 update on squamous carcinoma driver mutations/ gene alterations PTEN PIK3CA KRAS AKT1 DDR2 Unknown LKB1 BRAF EGFR VIII HER2 MET AMP AMP CCND1 AMP EML4ALK PIK3CA AMP PTEN loss FGFR1 AMP SOX2 AMP FGFR2 MULTIPLE HER2 MET Squamous is even more complex grouping by pathways? Prez-Moreno, CCR 2012; Paik, ASCO 2012, TCGA, Nature 2012

Molecular standards in NSCLC Minimal testing Peters et al, Ann Oncol 2012

Epidermal growth factor receptor (EGFR) signaling pathway (1980) EGFR, HER2 Cell membrane SOS EGFR-TK PI3K Akt ATP P ATP GRB2 STAT MAPK RAS RAF MEK Gene transcription cell-cycle progression Nucleus Proliferation Survival signals Invasion Angiogenesis Metastasis

EGFR discovery - 11 patients in 2004 Lynch, NEJM 2004

Inhibition of EGFR signaling pathway with activating mutation (2004) EGFR, HER2 ATP EGFR Inhibitors (gefitinib, erlotinib) ATP ATP EGFR-TK pathways PI3K Akt Cell membrane SOS EGFR-TK ATP ATP P ATP GRB2 STAT MAPK RAS RAF MEK Gene transcription cell-cycle progression Nucleus Proliferation Activating mutation ( 15% of patients) Survival signals Invasion Angiogenesis Metastasis

Going through simple comparisions Comparison of overall survival between patients who began first-line systemic therapy before or after gefitinib approval (A) All patients (B) (B) Patients with EGFR mutations. Takano, JCO 2011

Mutations in the EGFR gene Confer sensitivity/resistance to EGFR TKIs EGFR transcript Unclear effect on sensitivity to EGFR TKIs P694X 18 Extracellular domain Exons 1 16 EGFR Tyrosinekinase domain Exon 17 Exons 18 24 Exons 25 28 19 D761Y 18 D770_N771 insnpg 20 T790M L858R Regulatory domain E709X G719A/S Deletions Transmembrane domain V700D L861X 21 L730F P733L G735S V738F V742A T751I S752Y D761N E746K A763V N765A S768I T783A L792P L798F G810S L833V N826S H835L L838V H850N T847I V851X I853T G863DA864T A859T E866K Riely, Clin Cancer Res 2006

Studies of EGFR TKIs versus chemotherapy as first-line therapy in EGFR Act Mut+ NSCLC Study EGFR TKI n Median PFS in TKI arm (months) OPTIMAL Erlotinib 154 13.1 <0.0001 0.16 First Signal Gefitinib 42 8.4 0.084 0.61 IPASS Gefitinib 261 9.5 <0.0001 0.48 WJTOG 3405 Gefitinib 177 9.2 <0.001 0.48 NEJSG 002 Gefitinib 200 10.8 <0.001 0.36 EURTAC Erlotinib 174 9.4 <0.0001 0.42 LUX-3 Afatinib 308 13.6 <0.0001 0.47 P value HR

EURTAC: erlotinib first-line in EGFR mutation+ NSCLC Phase III study initiated by the Spanish Lung Cancer Group (GECP) Recruitment in Spain, Italy and France Erlotinib 150 mg/day until PD Chemonaïve advanced NSCLC EGFR mutation-positive (exon 19 or L858R) ECOG PS 0 2 R N=174 Primary endpoint: PFS Secondary endpoints: ORR, 1-year survival, OS, safety, QoL, localisation of PD ORR = objective response rate; QoL = quality of life Platinum-based doublet chemotherapy

EURTAC: PFS in ITT population 1.0 Erlotinib (n=86) Chemotherapy (n=87) PFS probability 0.8 HR=0.37 (0.25 0.54) 0.6 Log-rank p<0.0001 0.4 0.2 5.2 0 0 9.7 3 6 9 12 63 49 54 20 32 8 21 5 Patients at risk Erlotinib Chemo 86 87 15 18 Time (months) 17 4 9 3 21 24 27 30 33 7 1 4 0 2 0 2 0 0 0

Sensitivity to EGFR-TKI according to different EGFR mutations N EGFR RR (%) PFS (months) OS (months) 278 Classical exon 19-21 74.1 8.5 19.6 272 Wild-type 16.5 2.0 10.4 11 Exon 20 insertion 0 1.4 4.8 15 G719 53.3 8.1 16.4 15 L861 60.0 6.0 15.2 15 Uncommon mutations 20.0 1.6 11.1 Wu, Clin Cancer Res 2011

The risk of a wrong selection: the TORCH study R a n d o m Standard First-line Chemotherapy* PD Second-line Erlotinib** First-line Erlotinib** PD Second-line Chemotherapy* 1:1 Experimental Strata: histology smoking status gender country (Italy, Canada) age ethnicity *Chemotherapy: Cisplatin, 80 mg/m2, day 1 Gemcitabine, 1200 mg/m2, day 1 and 8 every 3 weeks, for 6 cycles **Erlotinib: 150 mg/day p.o. until progression PRIMARY END-POINT: NON INFERIORITY FOR OS

Overall survival Patients Events Median OS Months (95% CI) Standard 380 189 12.0 (10.3 14.8) Experimental 380 226 8.5 (7.2 10.5) Hazard Ratio 1.36 (95%CI 1.12 1.65) Log-rank test p = 0.002 Patients at risk Standard Experimental 380 380 226 197 108 88 34 34 11 16 1 4 2

Mut+ NSCLC: EGFR-TKI Acquired Resistance Baseline Tumor regression (RR up to 90%) Progression (median 9 months) Cave disease Flare: Hospitalization and/or death attributable to disease progression after discontinuation of gefitinib or erlotinib and before initiation of study drug; >20% patients median time 8 months

Mechanisms of TKI acquired resistance Lovly and Pao, Science Trans Med 2012

Evolutionary modeling based on growth kinetics No comparative data evaluating rechallenge strategy No prospective customized trial in TKI resistant NSCLC BUT provisional evidences: accumulating retrospective data Chmielecki J, Sci Transl Med 2011

Continuation on TKI upon progression A strategy per se Oxnard, ASCO 2012

Continuation on TKI upon progression A strategy per se Faehling, ASCO 2012

Rechallenge with TKI Author TKI PR (%) SD (%) DCR (%) mpfs mos mos mos Tomizawa 10 N=20 GEF 24 41 65 4 10 Koizumi 12 N=20 GEF 15 30 45 2 12 Heon 12 N=24 GEF 30% ERLO 70% 4 63 67 3.3 NR Tomizawa, Lung Cancer 2010; Koizumi, Clin Lung Cancer 2012, Heon ASCO 2012

EGFR resistance mediating mechanisms Combination chemotherapy + EGFR-TKI Change of EGFRTKI Chemotherapy IGF-Inhibitors Second Generation EGFR-TKIs PIK3 TKI Combination EGFR-TKI + EGFR AB HGF AB cmet TKI Met Mab Sequist et al, Sci Transl Med 2011

T790M: history of disease Found Germline (40%)! Discordant recent results! Erlotinib is still effective against tumours with de novo T790M n Median PFS (months) T790M+ 45 12 T790M 84 18 1.0 0.8 PFS probability T790M at baseline : 2.8% sequencing 25% MALDI-TOF 35% deep sequencing T790M at TKI resistance: 33.3% sequencing 83.3% MALDI-TOF 0.6 0.4 0.2 0 0 10 20 30 40 50 Time (months) Rosell, CCR 2011

Overcoming EGFR T790M? Promising irreversible pan-her agents in overcoming resistance show unexpected discouraging activity

EGFR TKI resistant patients treated with cetuximab-afatinib: update Janjigian, ESMO 2012 Confirmed partial response: 30% (29 of 96 pts), identical in T790M subgroups Median duration of response: 8 months Clinical benefit (partial response or stable disease): 75% (72 of 96 pts)

Side-road (bypass tracks) resistance Pao, Nat Rev Cancer, 2010

Side-roads EGFR pathways shown to be activated - MET amplification HGF over expression PIK3CA mutations PTEN loss FGRF overexpression AXL overexpression CRKL amplification NFkB activation BRAF mutation Yano, JTO 2011; Sos, Cancer Res, 2011; Ware, PLoS one 2010, Zhang Nat Gen 2012; Ohashi PNAS 2012; Tabara, Plos one 2012; Ng, Nat Med 2012

Wilson, Nature 2012

Other EGFR TKI resistance biological processes - Anti-apoptotic pathway (BIM deletion) Loss of EGFR mutant gene Transformation to SCLC EMT Yano, JTO 2011; Sos, Cancer Res, 2011; Ware, PLoS one 2010, Zhang Nat Gen 2012; Ohashi PNAS 2012; Tabara, Plos one 2012; Ng, Nat Med 2012

Conclusions: EGFR activating and sensitizing mutations in NSCLC The majority of lung adenocarcinoma in Caucasian patients harbor potentially druggable genetic oncogenic alteration EGFR alteration and TKI availability has opened an new era of customized treatment dramatically modifying the survival of selected patients Resistance mechanisms open the complexity of further treatment development at various levels: Tumour heterogeneity Pathways signaling redundancy and feed-back loops Clonal selection and kinetics of cancer cell populations Multiple concomitant drivers existence Need for TKI/targeted agents combination trials (phase I, vertical or horizontal blockade of pathways)

Conclusions: EGFR activating and sensitizing mutations in NSCLC (2) Other strategies targeting apoptotic pathways, HSP function, and anticancer immune response are emerging and seem promising in NSCLC, in combination or not with available approaches. A similar picture, with some more complexity, is emerging on lung squamous cell carcinoma Further progress mandates a move from unselected NSCLC trial designs to molecularly-driven clinical trials Such trials will require the building of large, flexible and international collaborative groups

Thanks for your attention