Phase 2 open-label study of brentuximab vedotin (BV) + pembrolizumab (pembro) in patients (pts) with treatment (tx)-naive metastatic head and neck squamous cell carcinoma (HNSCC).

Abstract

6015 Background: BV, an antibody-drug conjugate (ADC) targeting CD30, is hypothesized to deplete T regulatory cells (Tregs) that express CD30 and resensitize tumors to anti–PD-1 therapy. SGN35-033 (NCT04609566) is an ongoing multicohort study evaluating the efficacy and safety of BV + pembro in pts with solid tumors; we report results for cohort 6 in tx-naive HNSCC with PD-L1 combined positive score (CPS) ≥1, where pembro has historically demonstrated ORR of 19% and mPFS of 3.2 months. Methods: Cohort 6 included pts with metastatic HNSCC with PD-L1 CPS ≥1 by local testing and no prior therapy for metastatic disease or exposure to a PD-1/PD-L1 inhibitor. Pts received BV 1.8 mg/kg + pembro 200 mg every 3 wks. The primary endpoint was confirmed ORR assessed by investigator per RECIST 1.1. Secondary endpoints included DOR, PFS, and safety. Exploratory endpoints included OS and biomarker analyses. A genAI tool (01/02/25; Pfizer; GPT-4o) developed the 1st draft; authors assume content responsibility. Results: As of 10/25/24, 32 pts received ≥1 dose of BV + pembro. In all pts, the confirmed ORR was 34% (95% CI, 18.6-53.2), with median follow-up duration of 9.7 mo and median DOR not reached (95% CI, 3.9 mo-not estimable [NE]). BOR is in the Table. Responses were seen regardless of HPV status and across PD-L1 CPS ≥1 subgroups, with responses in 11 of 32 (34%) vs 9 of 25 (36%) in CPS ≥1 vs CPS ≥20, respectively. The KM estimate of DOR ≥6 mo was 89%. Median PFS was 7.2 mo (95% CI, 3.2 mo-NE); 6-mo PFS rate was 56%. Biomarker analyses of peripheral blood showed that Tregs expressed relatively higher CD30 vs other T cells. There was a trend of Treg depletion and increased T-cell proliferation and activation after BV + pembro. Observed PK of BV when combined with pembro in HNSCC was similar to that of BV monotherapy. All pts had ≥1 tx-emergent adverse event (TEAE); 24 pts (75%) had a grade ≥3 TEAE; 10 pts (31%) had tx-related grade ≥3 TEAEs. The most common tx-related grade ≥3 TEAEs were lymphocyte count decreased (13%), ALT increased, fatigue, neutropenia, and neutrophil count decreased (6% each). Tx-related serious TEAEs were reported in 2 pts (6%). No new safety signals were identified. Conclusions: BV + pembro demonstrated promising clinical efficacy with a safety profile consistent with each individual agent in pts with tx-naive metastatic HNSCC with PD-L1 CPS ≥1. Biomarker analyses support the hypothesized immunomodulatory mechanism of action of BV + pembro. These encouraging data are consistent with prior findings in PD-1–refractory NSCLC and melanoma and support continued investigation of CD30-directed ADCs + anti–PD-1 therapy in solid tumors. Clinical trial information: NCT04609566 . n=32 BOR, n (%) a Complete response 1 (3) Partial response (PR) 10 (31) Stable disease 12 (38) PD 5 (16) a 1 pt (3%) had unconfirmed PR at data cutoff; 3 pts (9%) discontinued tx with no postbaseline response assessment.