1. Evolution of systemic therapy for non-small cell lung cancer:From empiric to molecular-driven approach Rolf Stahel University Hospital Zürich Switzerland Ermatingen, April 6, 2010
2. Past developments in advanced NSCLC Cisplatin-based combination prolongs survival Edna Rapp, JCO 1988 Reduced nausea and vomiting with ondansetron Ann Intern Med 1990 Meta-analysis using data on individual patients from 52 randomized trials definitively confirms survival advantage Non-small cell lung cancer collaborative group, BMJ 1995 Similar results with platin combined with any third generation agent: a plateau has been reachedSchiller, NEJM 2001 2nd line chemotherapy with docetaxel, pemetrexedShepherd, JCO 2000; Hanna, JCO 2004
3. Case 1: Presentation 58-y/o woman, married, no children, works as executive secretary suffers from back pain since December 2008. After 2 months unsuccessful treatment by a chiropractioner, her family doctor orders an MRI. Here, a tumor leading to destruction of the first lumber vertebra was identified and the patient referred to oncology. The medical history and physical examination is otherwise unrevealing, patients stopped smoking 30 years ago
5. Case 1: Diagnosis and therapy Howtoproceed? 18.2.2009 Surgicaldecompressionand dorsal stabilisation Histology: Adenocarcinoma, TTF-1 positive, ER negative Diagnosis: Adenocarcinomaofthelungwith mediastinal andcervicallymphnodemetastasisandbonemetastasis Howtotreat?
6. Case 1: Diagnosis and therapy 23.2.2009 Initiation of chemotherapy with cisplatin, pemetrexed and bevacizumab, well tolerated 16.3. Hospitalization for second cycle. Updated results of pathology: EGFR genotype (exons 18 bis 21): Deletion in exon 19 (p.746E_750Adel) EGFR-FISH: positive (high-grade polysomy)EGFR-IHC: protein expression score 3 (DAKO Score 0-3) Continuation of cisplatin, pemetrexed, bevacizumab for total of 4 cycles
7. Evolution of systemic therapy for non-small cell lung cancer: Advanced NSCLC:Histological classification is necessary for decision making Advanced NSCLC: Reconsideration of extended therapy Molecular classification: Present necessities and future directions Consequences for second line therapy Consequences for adjuvant therapy
8. 1. Histological classification is necessary for today‘s decision making A diagnosis of “non-small cell lung cancer” is no longer acceptable as sufficient basis for treatment decisions: Cisplatin superior to carboplatin in adenocarcinoma Ardizzoni, JNCI 2007 Benefit of bevacizumab added to first line chemotherapy in non-squamous cell carcinoma Sandler, JCO 2006; Reck JCO 2009 Differential effect of pemetrexed in non-squamous vs squamous cell carcinoma Scagliotti, JCO 2008 Histology will help guide decision about which molecular analysis is performed
11. Bevacizumab added to chemotherapy:Design of 2 phase III trials E45991 CP x 6 (n=444) Previously untreated, stage IIIB, IV or recurrent non-squamous NSCLC (n=878) PD Bevacizumab (15mg/kg) every 3 weeks + CP x 6 (n=434) Bev PD Primary endpoint: OS Bevacizumab (15mg/kg) every 3 weeks + CG x 6 (n=351) Bev 2 RANDOMISE PD AVAiL2 1 Placebo (15mg/kg) +CG x 6 Previously untreated, stage IIIB, IV or recurrent non-squamous NSCLC (n=1,043) Placebo PD (n=347) 1 Placebo (7.5mg/kg) +CG x 6 Primary endpoint: PFS Bevacizumab (7.5mg/kg) every 3 weeks + CG x 6 (n=345) Bev 2 PD CP = carboplatin/paclitaxel CG = cisplatin/gemcitabine Sandler, 2006; Reck, JCO 2009
12. Bevacizumab added to chemotherapy:Significant increase of response Sandler, 2006; Reck, JCO 2009
13. Bevacizumab added to chemotherapy: Significant Increase of PFS PFS in E45991 PFS in AVAiL2 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 Bev 7.5mg/kg + CG Bev 15mg/kg + CP CP Bev 15mg/kg + CG Placebo + CG Probability of PFS 0 6 12 18 24 30 0 6 12 18 24 30 Duration of PFS (months) Duration of PFS (months) Median PFSBev 15mg/kg Median PFSBev 7.5mg/kg Median PFSBev 15mg/kg 6.2 months HR=0.66 p<0.001 6.7 months HR=0.75 p=0.003 6.5 months HR=0.82 p=0.03 Sandler, 2006; Reck, JCO 2009
14. OS benefit in adenocarcinoma histology: Pre-planned subgroup analysis (E4599) Bevacizumab-based therapy (n=602) extends OS to 14.2 months 31% reduction in the risk of death (HR=0.69) 1.0 0.75 0.50 0.25 0 Avastin + CP (n=300) CP (n=302) Probability of OS 10.3 14.2 0 6 12 18 24 30 36 42 48 Duration of OS (months) Pre-planned subgroup analysis (E4599) Sandler, JTO 2008
15. Case 1: Response assessment after 4 cycles of therapy February 09 June 09
16. Case 1: How to proceed Systemic therapy: Discontinue therapy? Maintain pemetrexed? Maintain bevacizumab? Change to erlotinib? Other: Radiotherapy to lumber spine? Bisphosponates or denosumab? Maintenance bevacizumab monthly
17. 2. Advanced NSCLC: Reconsideration of extended therapy Significant prolongation of PFS by extended chemotherapy Soon, JCO 2009 Significant improvement OAS with pemetrexed maintenance after standard cisplatin-based combination chemotherapy (without pemetrexed) in patients with non-squamous cell lung cancerCiuleanu, Lancet 2009 Significant prolongation of OAS with erlotinib maintenance after standard chemotherapy, independent of histology Cappuzzo, Lancet Oncology 2010 In booth studies benefit in particular for patients with stable disease
18. Increased time to PD Maintenance approach Maintenance therapy PD PD Diagnosis CR/PR/SD Maintenance in advanced NSCLC: treating before disease progression First-line treatmentPlatinum doublet chemotherapy (4–6 cycles) Traditional approach 2nd/3rd line treatment Break from treatment Diagnosis PD PD CR/PR/SD
20. Duration of therapy in advanced NSCLC: Progression-free survival Gem Doc Soon, JCO 2009
21. Pemetrexed 15.5 mos Pemetrexed 9.9 mos Placebo 10.3 mos Placebo 10.8 mos Maintenance pemetrexed vs placebo:Overall survival by histology Non-squamous (n=481) Squamous (n=182) HR=0.70 (95% CI: 0.56-0.88) P =0.002 HR=1.07 (95% CI: 0.49–0.73) P =0.678 Survival Probability Time (months) Time (months) Belani, ASCO 2009; Ciuleanu, Lancet 2009
22. SATUTRN: OAS with erlotinib maintanance in EGFR WT tumors Erlotinib 150mg/day PD Chemonaïve advanced NSCLC n=1,949 4 cycles of 1st-line platinum-based doublet Non-PD n=889 1:1 Placebo PD Mandatory tumour sampling Maintenance treatment with erlotinib (SATURN) Cappuzzo,. Lancet Oncol 2010
23. SATURN: OS in EGFR WT group with SD on first-line chemotherapy 1.0 0.8 0.6 0.4 0.2 0 Erlotinib (n=114) Placebo (n=103) HR=0.65 (0.48–0.87) Log-rank p=0.0041 OS probability 8.7 12.4 0 3 6 9 12 15 18 21 24 27 30 33 36 Time (months) Measured from time of randomisation into the maintenance phase Coudert, ECLC 2010
24. 0.4 0.6 0.8 1.0 1.4 1.2 Both pivotal maintenance studies suggestgreater benefit in patients with SD vs PR/CR JMEN OS Pemetrexed SATURN OS Erlotinib ITT population (n=663) CR/PR (n=322) SD (n=337) ITT population(n=889) CR/PR (n=394) SD (n=487) 0.4 0.8 0.6 1.0 1.4 1.2 HR HR Favours erlotinib Favours placebo Favours placebo Favours pemetrexed
26. 3. Molecular classification: Present necessities and future directions Adenocarcinoma of the lung is not a uniform disease and needs to be classified by additional molecular analysis Present needs include EGFR mutation status and determination of EML4-ALK fusion gene Knowledge about resistance mechanisms to available agents and the opportunity of agents against new molecular targets mandate change in the trial design Potential driver mutations are also being identified in squamous cell lung cancer
28. First line EFGR TKI or chemotherapy for non-squamous cell lung cancer harboring activating EGFR mutation
29. IPASS: Objective RR in EGFR mutation positive and negative patients Overallresponserate (%) Gefitinib Carboplatin / paclitaxel EGFR M+ odds ratio (95% CI) = 2.75(1.65, 4.60), p=0.0001 EGFR M- odds ratio (95% CI) = 0.04(0.01, 0.27), p=0.0013 71.2% 47.3% 23.5% 1.1% (n=132) (n=129) (n=91) (n=85) Odds ratio >1 implies greater chance of response on gefitinib Mok, ESMO 2008; NEJM 2009
30. Improvement of lung cancer symptoms Adapted from Mok, NEJM 2009; supplementary appendix
31. Mean change from baseline in LCS by EGFR mutation status (EFQ) 10 8 EGFR M+ Gefitinib (n=131)Carboplatin / paclitaxel (n=128) 6 4 Change from baseline 2 0 -2 -4 -6 57 0 3 6 9 12 15 18 24 30 36 42 48 54 10 8 6 EGFR M- Gefitinib (n=89)Carboplatin / paclitaxel (n=80) 4 2 0 Change from baseline -2 -4 -6 -8 -10 3 6 9 12 15 18 24 30 0 21 27 33 Week White dotted line indicates baseline; Purple dotted lines indicate a clinically meaningful change from baseline in LCS; Mean change from baseline by EGFR mutation status was calculated post hoc; Error bars are the 95% CI of the mean; N = number of evaluable patients at baseline; Data after Week 54 (EGFR M+) and Week 30 (EGFR M-) not presented as n<20;
32. IPASS: Overall survivalin EGFR mutation positive and negative patients EGFR mutation + EGFR mutation - Gefitinib (n=91)Carboplatin /paclitaxel(n=85) HR (95% CI) 1.18 (0.86, 1.63); p=0.309No. events G 82 (90%)C / P 74 (87%)Median OS G 11.2 monthsC / P 12.7 months Gefitinib (n=132)Carboplatin /paclitaxel(n=129) HR (95% CI) 1.00 (0.76, 1.33); p=0.990No. events G 104 (79%)C / P 95 (74%)Median OS G 21.6 monthsC / P 21.9 months 1.0 1.0 0.8 0.8 0.6 0.6 0.4 0.4 Probability of survival Probability of survival 0.2 0.2 0.0 0.0 52 52 0 4 8 12 16 20 44 24 28 32 36 40 48 0 4 8 12 16 20 44 24 28 32 36 40 48 Patients at risk: Gefitinib C / P Time from randomisation (months) Time from randomisation (months) 132 129 126 123 103 95 70 68 24 26 11 15 121 112 88 80 58 55 46 48 38 40 6 7 3 0 0 0 5 1 69 76 52 57 40 44 29 33 26 25 1 1 19 19 16 16 11 11 8 3 0 1 91 85 0 0 Cox analysis with covariates; a hazard ratio <1 implies a lower risk of death on gefitinib No formal adjustment for multiple testing was made, therefore statistical significance at the traditional 5% level cannot be claimed Yang, ESMO 2010
34. First line erlotinib in patients with (activating) EGFR mutations 2105 pt screened for mutations (non-squamous cell) 350 (16.6%) had mutations Current smoker 5.8% Former smoker 9.5% Non-smoker 37.7% 217 ptstreated RR 71% Median PFS 14 months Median survival 27 months Rosell, NEJM 2009
35. Case 1: Second line treatment with EGFR TKI July 10 March 10 July 10
36. Case 1: Re-treatment with EGFR TKI atreduced dose andshort-term prednison February 11 August 10
37. Presence of EGFR mutations an imperfect predictor of outcometo EGFR TKIs PFS 27 months in patients with low BRCA1 mRNA levels, 18 months in those with intermediate levels, 10 months in those with high levels: Randomized phase I/II study with gefitinib and olaparibstratified for BCCA1 mRNA (SLCG) Rosell, CCR 2011
38. Presence of EGFR mutations an imperfect predictor of outcometo EGFR TKIs EGFR T790Mmutation before therapy in 35%. Shorter PFS (8 vs 18 ms): Stratified phase II study with erl/bev (SLCG/ETOP) ?Irreversible TKIs (PF299, afatinib) Rosell, CCR 2011 Naumov, CCR 2009
39. Findings associated with resistance to EGFR TKI T790M mutation: 50% of tissue samples from patients with acquired gefitinib resistance and a proportion of patients at diagnosisKosaka, Clin Cancer Res 2006 ?Irreversible TKIs (PF299, afatinib) MET amplification of : 22% of patients with acquiredTKI resistance: Engelman, Science 2007 ? Combination of TKI and met inhibitor
40.
41. Progression after one or two lines of chemotherapy (incl. one platinum- based regimen) and ≥12 weeks of treatment with erlotinib or gefitinibMiller, ESMO 2010
42. MET and ALK in lung cancer Pao & Girard, Lancet Oncol, 2011
45. 44| Clinical trials and translational research in lung cancer: The advances over the last years Advanced squamous cell carcinoma: Molecular determinants ETOP | EMCTO | Lugano, February 24, 2011
46. 45| Clinical trials and translational research in lung cancer: The advances over the last years All histologies: EGFR antibody in EGFR high expressing tumors ETOP | EMCTO | Lugano, February 24, 2011
47. 46| Molecular based clinical trials in lung cancer: Issues Molecular pathology both at diagnosis and at relapse for definition or stratification of study population mandatory Centralized analysis or standardization of methodologies between sites Availability of integrated services at sites Rarity of molecular subgroups Large networks of sites necessary to detect eligible patients Optimal number or sites for a given trial Emphasis on early decision in molecularly-driven trials New models of collaboration with diagnostic and pharmaceutical companies ETOP | Milano, February 2, 2011
49. 48| Specific aims of ETOP according to bylaws To serve as a meeting platform for European study groups and institutions dealing with thoracic malignancies To foster intergroup studies among, but not exclusively, European study groups and institutions To sponsor and/or perform own studies To foster scientific exchange on laboratory and clinical issues among interested parties and beyond To provide knowledge to partners in the field ETOP | Ermatingen, April 6,, 2011
66. 51| Stepwise evolution Step 1: Retrospective analysis of 1500 completely resected NSCLC from a limited number of 7-10 sites: Mutation testing, immunohistochemistry, selected FISH on formalin-fixed, paraffin-embedded tumor tissue Step 2: Expansion to biopsies from advanced disease and a phased approach increasing to the number of participating sites with the aim to have participation from at least one site from all countries represented in ETOP Further steps and issues under considerations:Enlargement of biobank, exon sequencing (selected frozen tissue), circulating biomarkers, technology platforms, resource utilization and health economics research ETOP | LUNGSCAPE | Milano, February 2, 2011
67.
68. If EGFR TKI in first line, second line platin-based combination (first line chemotherapy)
74. Lung Adjuvant Cisplatin Evaluation Chemotherapy vs control No. Deaths Hazard ratio No. Deaths Hazard ratio Category HR [95% CI] Category HR [95% CI] (Chemotherapy / Control) (Chemotherapy / Control) / No. Entered / No. Entered Stage IA 102 / 347 1.41 [0.96;2.09] Stage IA 102 / 347 1.41 [0.96;2.09] Stage IB 509 / 1371 0.92 [0.78;1.10] Stage IB 509 / 1371 0.92 [0.78;1.10] Stage II 880 / 1616 Stage II 880 / 1616 0.83 [0.73;0.95] 0.83 [0.73;0.95] Stage III 865 / 1247 0.83 [0.73;0.95] Stage III 865 / 1247 0.83 [0.73;0.95] 0.5 1.0 1.5 2.0 2.5 0.5 1.0 1.5 2.0 2.5 | | Chemotherapy better Control better Chemotherapy better Control better Test for trend: p = 0.051 Test for trend: p = 0.051 CT may be detrimental for stage IA, but stage IA patients were generally not given the potentially best combination cisplatin+vinorelbine (13% of stage IA patients versus ~43% for other stages) Pignon, ASCO 2006, JCO 2008
75. Adjuvant gefitinib in unselected NSCLC After “curative chemoradiotherapyfor stage III: After curative resection: Kelly, JCO 2008 Goss, ASCO 2010
76. Stage IB with tumor size over 4 cm CALGB 9633: adjuvant carboplatin/paclitaxel in stage IB Adjuvant cisplatin and vinorelbinein stage I and II NSCLC: JBR.10 update Strauss, JCO 2008 Butts, JCO 2010