Checkpoint inhibitors have shown efficacy in pts with recurrent and/or metastatic head and neck cancer (HNC). Atezolizumab (atezo; anti–PD-L1) inhibits binding of PD-L1 to PD-1 and B7.1, thereby restoring tumor-specific T-cell immunity. Here, we examine single-agent safety and clinical activity of atezo in pts with advanced HNC.
Pts with HNC received atezo IV q3w (15 or 20 mg/kg or 1200 mg) in a Ph Ia study (NCT01375842). Pts were initially enrolled non-selectively (n = 10); once PD-L1 was identified as a potential biomarker, pts were selected by PD-L1 status (> 5% PD-L1 expression on IC [IC2/3]; centrally evaluated by VENTANA SP142 IHC assay, n = 22). HPV status was assessed by PCR. Treatment was originally for 16 cycles or up to 1 y; pts were subsequently treated until loss of clinical benefit. Primary endpoint was safety; tumor responses were evaluated by RECIST v1.1.
As of 31 Dec 2016, with follow-up of ≥ 14 mo, 32 pts were safety and efficacy evaluable. 84% were male; 66% had ECOG PS of 1. Median age was 62 y (range, 32-78 y), pts were heavily pre-treated (53% had ≥ 2 prior lines of therapy) and 66% were current/previous tobacco users. Most common primary tumor sites were oropharynx (50%), oral cavity (22%) and nasopharynx (19%). Median treatment duration was 3.4 mo; 21/32 pts (66%) had a treatment-related AE. 3 pts (9%) had Gr 3 treatment-related AEs (tumor lysis syndrome, hyponatremia, pruritus, colitis). 1 pt (3%) had Gr 4 treatment-related cardiac tamponade. No Gr 5 treatment-related AEs were seen. In all pts (IC0/1, n = 7; IC2/3, n = 25), confirmed ORR was 22%, mPFS was 2.6 mo (range, 0.5-48.4 mo) and mOS was 6.0 mo (range, 0.5-51.6+ mo). Clinical activity by PD-L1 subgroups is shown in the Table. Preliminary analyses indicate that there was no association between HPV status and clinical outcomes.
In advanced HNC, atezo was well tolerated. Encouraging response and long-term survival were seen independently of PD-L1 IHC or HPV status and warrant further investigation.Table:
1044O Clinical Activity per RECIST v1.1 in PD-L1 Subgroups
|(n = 7)||(n = 25)|
|ORR, n (%)||1 (14%)||6 (24%)|
|PR||1 (14%)||6 (24%)|
|DCR, n (%)||3 (43%)||7 (28%)|
|mDOR, mo (range)b||7.4||26.2 (2.8-45.8)|
|mPFS, mo (range)||5.7 (0.5-13.5)||2.6 (0.5-48.4)|
|mOS, mo (range)||9.0 (0.5-26.5)||5.6 (1.1-51.6+)|
|1-year OS rate||43%||34%|
|2-year OS rate||29%||18%|
|3-year OS rate||NE||18%|
indicates a censored value.a
Data by PD-L1 expression on TC will be presented.b
n = 1 for IC0/1, no estimate for mDOR; n = 6 for IC2/3.
IC0 = PD-L1 expression on
Clinical trial identification
Legal entity responsible for the study
F. Hoffmann-La Roche Ltd.
F. Hoffmann-La Roche Ltd.
F.S. Braiteh: Speaking and consulting fees received from Genentech. A.S. Balmanoukian: Speaker\'s bureau for Merck, Bristol-Myers Squib, Genentech, and AstraZeneca. F.S. Hodi: Consultant Merck, Bristol‐Myers Squibb, Genentech, EMD Serono, Amgen; Research support to institution from Bristol‐Myers Squibb. B. Liu, L. Molinero, X. Shen: Genentech employee and stock. C. O\'Hear: Genentech employee and Roche stock. All other authors have declared no conflicts of interest.