Improving Immunotherapy for Melanoma

Transcription

1 Improving Immunotherapy for Melanoma David McDermott, MD Beth Israel Deaconess Medical Center Dana Farber/ Harvard Cancer Center Harvard Medical School Immunotherapy Improvement Model All patients All tumors Immune responsive patients Immune responsive tumors Improve Patient Selection Enhance Tumor Responsiveness Page 1

2 Identifying Responders to Immunotherapy in Advanced Melanoma Exploratory analysis 12 of 33 genes which correlated with response were immune/inflammatory» Wang et al. Cancer Res 2002 Gene signature enriched with immune-inflammatory genes predicts benefit to melanoma vaccines» Gajewski et al. isbtc 2008; ASCO 2009; AACR 2011 CCR5 Delta32 polymorphism associated with adverse outcome to immunotherapy» Ugurel et al. Cancer Immunol Immunother 2008 Melanocyte-Derived Gene Signature Class 1 defined by MITF and melanocyte antigen expression eg. MITF, ML-IAP, GP100, Tyrosinase, MelanA Class 1 Class 2 Class 2 is defined by Immune genes eg. Annexin A1, IL6R, Oncostatin M, MCSF, GMCSF Unpublished Data; Jean-Philippe Brunet, Todd Golub, Tom Flotte, Stefan Wagner, DFHCC Melanoma Program Page 2

3 HD IL-2 Selection: Efficacy Data Class 2 Better PFS p = Better RR p = sided FET OS similar p = 0.19 Response (%) DASL Class 1: Antigenic (n=21) DASL Class 2 Immune (n=7) Complete 2 (10%) 2 (29%) Partial 6 (28%) 4 (57%) Total 8 (38%) 6 (86%) Durable (>18 mo) 3 + (14%) 3 + (43%) Survival (mo) Median OS Median PFS Ryan Sullivan et al Progression-Free Survival by DASL Class Class 1: med 2.5 Class mo 1 Class 2: med 19.4 Class mo2 0.0 Probability Months p = Ryan Sullivan et al Page 3

4 Relationship of MAP Kinase Pathway mutations and Response to HD IL-2 Outcome NRAS BRAF&WT p-value N=15 N=88 CR/PR 7 (47%) 27 (19%) 0.04 OS (years) PFS (days) A significantly larger proportion of patients with NRAS mutant tumors achieved CR/PR compared to those with BRAF/WT tumors. Improvements seen in PFS and OS. More data in needed. Joseph, Sullivan et al- JIT 2011 Accepted Impact of VEGF Levels on HD IL-2 Efficacy Sabatino M et al JCO 2009 Page 4

5 HD IL-2 Tissue-based Predictive Biomarkers: Conclusions A novel immune-based gene expression profile appears to predict for a better PFS and possibly response to HD IL-2 Association of clinical benefit with immune response signature suggests possible mechanism for HD IL-2 anti-tumor effect Confirmatory Trial underway in CWG BRAF/NRAS mutational status may predict for favorable response to HD IL-2 Opportunity for combination studies (IL-2 + vemurafenib Elevated LDH/VEGF may predict for lack of response to HD IL-2 (new treatment approach necessary?) Joseph, Sullivan et al Immunotherapy Improvement Model All patients All tumors Immune responsive patients Immune responsive tumors Improve Patient Selection Enhance Tumor Responsiveness Page 5

6 Focus Immune Response with Vaccination Page 6

7 IL-2 +/- peptide vaccine Is this result p proof of concept p that the immune response can be focused and will it be relevant with novel immunotherapies? Immune Checkpoint Inhibitors + vaccine CTLA 4 Antibody PD-1 Antibody Page 7

8 Immunotherapy Improvement Model All patients All tumors Immune responsive patients Immune responsive tumors Improve Patient Selection Enhance Tumor Responsiveness Ipilimumab : Pivotal trials Ipilimumab (3 mg/kg) vs gp100 peptide vaccine vs combination in HLA A2+ patients Overall survival advantage reported (Hodi et al NEJM, 2010) No advantage for addition of vaccine DTIC +/- ipilimumab (10 mg/kg) Overall survival advantage reported (Robert et al NEJM 2011) Ipilimumab in patients with CNS metastases-cwg Significant activity observed (Margolin et al ASCO 2010) Adjuvant studies (Stage III patients) Ipilimumab vs observation (Europe-underway) Ipilimumab vs IFN (US-Cooperative Group, underway) Page 8

9 Pivotal 2 nd Line Phase III Trial Study Design Pre-treated Metastatic Melanoma (N=676) R A N D O M I Z E Ipilimumab + gp100 Ipilimumab + placebo gp100 + placebo (N=403) (N=137) (N=136) Hodi et al NEJM, 2010 Kaplan-Meier Analysis of Survival Ipi + gp100 Ipi + pbo gp100 + pbo Survival Rate Ipi + gp100 N=403 Ipi + pbo N=137 1 year 44% 46% 25% 2 year 22% 24% 14% gp100 + pbo N=136 Page 9

10 Commentary - Anti CTLA-4 Enables immune responses and anti-tumor responses in some individuals Activity powerful enough to work in CNS and overcome concurrent immunosuppression No additional clinical benefit with peptide vax Activity seen in patients with prior IL-2; not restricted by BRAF status Ipilimumab represents an option for the majority of patients with advanced melanoma; timing of therapy, severe autoimmune toxicities require attention Ipilimumab: Current/future studies Ipi i +/- GM-CSF (E1608) Accrual completed in 6 months Ipi + PLX (DFHCC, MSKCC, UCLA) Ipi + HD IL-2 (CWG proposed) Ipi + Bev (DFHCC) Hodi et al ASCO 2011 Page 10

11 Ipilimumab + Interleukin-2 Three Serial Phase II Trials at NCI Protocol 2 = Concomitant Therapy 36 Patients Response Rate = 21% CR Rate = 17% All 6 ongoing > 5 years Potential ti mechanism? Confirmatory Trial planned in CWG Ipilimumab + IL-2 = Toxicity? Maker et al, Annals of Surg Onc 2005 Page 11

12 Rational Combinations with CTLA-4 Blockade: Targeting VEGF Immune mediated vasculopathy Immune modulatory effects of VEGF Decreased DC maturation T cell trafficking Hodi et al ASCO 2010 Ipilimumab Plus Bevacizumab: Phase I Clinical Trial Cohort 1: 10 mg/kg ipilimumab plus 7.5 mg/kg bevacizumab Cohort 2: 10 mg/kg ipilimumab plus 15 mg/kg bevacizumab Induction: Every 3 weeks x 4 cycles Maintenance: Bevacizumab continued every 3 weeks, ipilimumab continued every 3 months Page 12

13 Ipilimumab Plus Bevacizumab: Inflammatory Adverse Events 22 Patients Giant cell (temporal) arteritis; palpable purpura 5 cases hypophysitis 3 cases thyroiditis grade 4 hepatitis Alopecia areata 2 cases bilateral uveitis/retinitis grade 2 colitis Ipilimumab Plus Bevacizumab: Clinical Efficacy (22 patients) BORR: 6 partial responses (27%) 1 complete response (5%) 7 durable stable disease (32%) 8 progressive disease (36%) Median follow up = 14 months 6-month PFS = 59% 1-year OS = 72% Page 13

14 CTLA-4 plus VEGF Blockade: Summary Effects on both tumor immunity and tumor vasculature were observed VEGF-A blockade may have additive effects to immune therapy platform of CTLA-4 blockade Inflammatory adverse events Striking immune activation Possible benefits in efficacy Manageable toxicities Randomized phase II warranted Hodi et al ASCO 2011 Page 14

15 Immunotherapy Improvement Model All patients All tumors Immune responsive patients Immune responsive tumors Improve Patient Selection Block Tumor Defenses Approaches to Cancer Immunotherapy: Covering the Barbed Wire PD1-PD L1 = Barbed Wire Page 15

16 PD-1: Role in T Cell Activation What is PD-1? Member of CD28 family involved in T cell regulation Expressed by activated T cells, memory T cells, and regulatory T cells Down regulates T cell activity upon binding to PD-L1/L2 Tumor PD-L1 expression may correlate with negative prognosis potential mechanism of tumor self defense Sznol M, et al. Presented at ASCO 2010 J Clin Oncol 28:15s, 2010 (suppl; abstr 2506) Clinical Activity: Best Overall Response Diagnosis RCC Dose (# evaluable for response) 1 mg/kg (2) 10 mg/kg (14) CR PR upr SD PD NA mg/kg (16) MEL 3 mg/kg (14) mg/kg (16) NSCLC 3 mg/kg (1) 10 mg/kg (10) Best overall response measured using RECIST (v 1.0) Stable Disease (SD) = patients not meeting criteria for CR/PR/PD NA = Not available Note: No responses (CR, upr, PR) observed in subjects with mcrpc (n=12) and CRC (n=6) Sznol M, et al. Presented at ASCO 2010 Data cutoff date: 5/3/2010 J Clin Oncol 28:15s, 2010 (suppl; abstr 2506) Page 16

17 Change in Tumor Burden Melanoma Chan nge (%) in Tumor Burden New lesions after cycle Patients Group 1mg/kg 3mg/kg 10mg/kg Sznol M, et al. Presented at ASCO 2010 J Clin Oncol 28:15s, 2010 (suppl; abstr 2506) Objective Response in Melanoma: 1 MG/KG A A B B History: 66 yr/male patient Diagnosed in 2001 Progression on HD IL-2 Presented with pulmonary and bulky liver lesions A: Baseline (11/4/2008) B: Cycle 1 assessment (1/26/2009) Pt. met PR criteria after 3 cycles (12 bi-weekly doses) Pt. currently in cycle 10, response ongoing at 12+ months Sznol M, et al. Presented at ASCO 2010 J Clin Oncol 28:15s, 2010 (suppl; abstr 2506) Page 17

18 SAFETY RESULTS: ALL PATIENTS For the entire study group, maximum tolerated dose (MTD) was not reached at doses of 1, 3, and 10 mg/kg Drug-related adverse events (AEs) appear to have an immunomodulatory mechanism of action There was no apparent relationship between drug dose and AE frequency The most common drug-related (investigator attributed*) grade 3/4 AEs were investigations a (4.0%), endocrine disorders f (2.4%), and fatigue (1.6%) 1 treatment-related death (grade 4 drug-related pneumonitis; patient died of sepsis while being treated with immunosuppressive agents) 35 PD-1 Ab: Anti-Tumor Activity in RCC ORR (CR/PR/irPR) Duration of response (mo) DCR-4mo DCR-6mo PFS (mo) 6/18 (33%) 13 (8.5 NR) 13/18 (72%) 11/18 (61%) 8.0 (4.0 14) Page 18

19 PD-1 Antibody Therapy Early trials show tumor responses in >30% of heavily pretreated patients with advanced melanoma (ASCO ) Responses also in other cancers; durable to date Adverse event profile consistent with an immunomodulatory mechanism of action PD-1Ab combination studies underway Ipilimumab Phase I Multi-peptide vax Stage III and IV J Weber, Moffitt Other PD-1/PDL-1 agents being considered Curetech PD1 Ab (DFHCC) Many other companies with agents in pipeline PD-1 Pathway Agents in Development Company Agent Structure Status Amplimmune/GSK AMP-224 Fc fusion protein to PD-L2 BMS/Ono BMS Fully human, IgG 4 Ab Curetech/Teva CT-011 Humanized monoclonal Phase I Phase II RCC, others solid tumors Phase II melanoma Genentech/Roche PDL-1 Ab Entering Phase I Merck MK-3475 Humanized, IgG 4 ab Entering Phase I Page 19

20 Immunotherapy Improvement Model All patients and tumors Immune responsiveness Elimination of Tregs CTLA4Ab PD1 Ab CD137 agonist Ab, IL-15, IL-21 Immune responsive phenotype Immune responsive tumors Tumor Responsiveness PDL1 Ab Combination therapy Earlier therapy Focus Response Vaccine? Selection Identify the patients in the overlap Achieving Dr Kirkwood s Dream through Targeted Immunotherapy "Bummer of a birthmark, Hal" Page 20