A randomized phase II study of pembrolizumab with or without radiation in patients with recurrent or metastatic adenoid cystic carcinoma.
6082 Background: Adenoid cystic carcinoma (ACC) is a salivary gland malignancy characterized by a high rate of distant recurrence. Systemic therapy has generally failed to produce durable benefit. Radiation (RT) is used for localized disease and as directed treatment for metastases. Here, we report the safety and efficacy of pembrolizumab (pembro) administered with or without hypofractionated RT in a phase II randomized study. Methods: Eligible patients (pts) had recurrent or metastatic ACC with evidence of progressive disease (PD) within the last 12 mos and >=1 measurable non-CNS lesion, along with 1-5 additional lesions deemed appropriate for RT to 30 Gy in 5 fractions. Pts were randomized to pembro alone (200 mg IV q3 weeks) or in combination with RT given within 7 days of cycle 1, day 1. The primary endpoint was objective response rate (ORR) outside the RT field by RECIST 1.1. Using a parallel two-stage design, if >=1 response out of 10 was observed in either arm, 10 more pts would be enrolled to that arm. If >=3 responded, the null hypothesis (ORR=5%) would be rejected in favor of a 25% ORR. Predefined secondary endpoints included progression free survival (PFS) and toxicity. Analyses of tumor growth rate (TGR) excluding RT lesions and immune biomarkers were exploratory. Results: Ten pts per arm were randomized into the trial’s first stage with median age 65 (45-79). No objective responses were seen. Stable disease (SD) was observed in 13 pts; 6 had PD as best response, 1 was unevaluable. Median PFS was 7 mos 95% CI (3 - 13 mos), with 9 pts without progression at 6 mos. 3 pts remain on study treatment (range 8-11 mos). In pts with SD, TGR decreased by >25% in 7 of 12 pts and by >75% in 4 pts. There was no difference in likelihood of SD or PFS between arms. Treatment related AEs (TRAEs) occurred in 18 pts but there were no G3-5 TRAEs. Among 8 biopsies analyzed, PD-L1+ tumor/immune cells ranged from 12-52%. Conclusions: Pembro alone or with hypofractionated RT was well tolerated. We observed no objective responses, but 65% of pts with PD prior to study entry achieved SD, the majority with decreased TGR, and 15% had prolonged SD. Additional strategies are needed to further delay progression and effect response. Clinical trial information: NCT03087019.