808O - Inhibition of PD-L1 by MPDL3280A leads to clinical activity in pts with metastatic urothelial bladder cancer (UBC)
| Date | 29 September 2014 |
| Event | ESMO 2014 |
| Session | Genitourinary tumours, non prostate immunotherapy |
| Topics | Urothelial Cancers Cancer Immunology and Immunotherapy |
| Presenter | Joaquim Bellmunt |
| Citation | Annals of Oncology (2014) 25 (suppl_4): iv280-iv304. 10.1093/annonc/mdu337 |
| Authors |
J. Bellmunt1, D.P. Petrylak2, T.B. Powles3, F. Braiteh4, N. Vogelzang5, C. Cruz6, H. Burris7, J.P. Eder8, G. Fine9, M.S.L. Teng10, X. Shen11, J. Bruey12, Z. Boyd11, P. Hegde11, D. Chen9, Y. Loriot13
|
Abstract
Aim
New treatments are needed for metastatic UBC. PD-L1 expression is prevalent in UBC and may help protect cancer cells from immune-mediated destruction. MPDL3280A is a human anti-PD-L1 mAb with an engineered Fc-domain which disrupts PD-L1 binding to its receptors PD-1 and B7.1.
Methods
In a Ph I study, UBC pts received MPDL3280A 15 mg/kg IV q3w for up to 1 y. Objective response (OR; unconfirmed and confirmed) was assessed by RECIST v1.1. PD-L1 IHC was centrally evaluated on tumor biopsies. Circulating biomarkers were evaluated.
Results
As of Jan 1, 2014, 68 UBC pts were treated with MPDL3280A. 67 were evaluable for efficacy (dosed by Nov 20, 2013; with ≥ 6 wk follow-up) and included 30 PD-L1 IHC 2/3, 35 PD-L1 IHC 0/1 and 2 PD-L1 unknown pts. The 67 efficacy evaluable pts were 72% male, had a median age of 65 y (36-86) and 59% were ECOG PS 1. 75% had visceral metastases and 33% had liver metastases. 72% received ≥ 2 prior therapies, 93% received prior platinum-based chemo, 79% received prior cisplatin and 42% received MPDL3280A ≤ 3 mo from last prior chemo. For IHC 2/3 pts, ORR was 43% (95% CI 26-63; 2 CRs, 11 PRs); the median duration of response (DOR) was not reached (range 0.1+ - 30.3+ wk). For IHC 0/1 pts, ORR was 11% (95% CI 4-26; 4 PRs); the median DOR was not reached (range 0.1+ - 6.0+ wk). 16/17 responders were still responding as of clinical cutoff date and included pts with poor prognostic factors. For the 17 responders, the median time to first response was 42 d (38-85 d). The 68 safety-evaluable pts received MPDL3280A for a median duration of 65 d (1-259 d). All grade study drug-related AEs occurred in 57% of pts, most often decreased appetite, fatigue and nausea. Treatment-related G3/4 AEs were seen in 4% of pts (no G4 AEs or treatment-related deaths were observed); there was 1 event each of asthenia, thrombocytopenia and decreased blood phosphorus. Treatment resulted in transient increases in pharmacodynamics biomarkers, including Ki-67 + CD8+ T cells and plasma proteins (eg, IL-18). Updated data will be presented, including PFS.
Conclusions
MPDL3280A was well tolerated in this heavily pretreated UBC population. The majority of responses were rapid and on-going, and PD-L1 status correlated with response to MPDL3280A.
Disclosure
J. Bellmunt: Genentech consultant, stock; F. Braiteh: has served as a consultant on the speakers bureau for Amgen/Onyxx, Bayer, BMS, Caris Life Sciences, Celgene, Genomic Health, Incyte, INSYS, Novartis, Pfizer, Sanofi and as consultant to Agendia, Foundation Medicine, and Saladax; N. Vogelzang: has served as a consultant for Celgene and Roche/Genentech and has received research funding from Novartis and honoraria from Pfizer; G. Fine, M.S.L. Teng, X. Shen, J. Bruey, Z. Boyd, P. Hegde and D. Chen: is an employee of Genentech, Inc.; Y. Loriot: has served as a consultant for Astellas, Sanofi, Celgene and Pierre Fabre, and received research grants from Astellas and Sanofi. All other authors have declared no conflicts of interest.