A phase II trial of reirradiation combined with pembrolizumab in patients with locoregional inoperable recurrence or second primary squamous cell carcinoma of the head and neck (HNSCC).

Abstract

6049 Background: Better outcomes are needed for patients (pts) with locoregional inoperable recurrence or second primary HNSCC after prior radiation therapy. Methods: We conducted a phase II single arm trial of hyperfractionated reirradiation (ReRT, 1.2 Gy BID for total 60Gy to gross disease + margin without elective nodal radiation) plus pembrolizumab (P) (200mg q3 weeks starting day 1 of ReRT). P was continued until confirmed CR or if no CR, until disease progression or up to 24 months. All pts underwent a 3 month post ReRT PET/CT and subsequently CT neck/chest after every 3rd cycle of P. Pts included had: Locoregional recurrence or second primary HNSCC (excluding skin or salivary), were unresectable or not willing to undergo resection, one RT course completed at least 6 months prior, with >50% tumor volume treated at doses >45Gy, ECOG PS 0-1, a target lesion by RECIST, no distant metastasis, and no prior anti-PD-1/PD-L1. Primary endpoint was PFS, and the hypothesis was that the median PFS would increase from the historical control of 8 months (mo) to 12 mo, α=0.05, power 78%, n=48. Secondary endpoints: ORR by RECIST, OS, toxicity, QOL (EORTC). The Kaplan-Meier method was used to estimate PFS and OS. Results: In the 48 evaluable pts, median age was 65.5 (range 44-79), 62.5% male. Primary site: hypopharynx (8.3%), larynx (6.3%), nasopharynx (10.4%), oral cavity (43.7%), oropharynx (31.3%, one HPV+ pt). Median doses of P was 8, median RT dose was 60 Gy and 20.8% received proton RT, remainder IMRT. Median follow up was 49.1 months. The median PFS was 8.3 mo (95%CI 5.5-10.3), and median OS was 13.8 mo (95% CI 9.5-21.8). 1 and 3-year PFS were 28.9% and 22.3% respectively, and for OS 54.2% and 25.7%, respectively. 42 pts were evaluable for response and the ORR was 54.8% (33.3% CR, 21.4%PR), 21.4% SD, 23.8% PD. In an exploratory analysis, the median PFS was 13.8 mo (95% CI 9.1-NR) and the median OS was 57.8 mo (19.6-NR) in patients that had a response (CR/PR, n=23), with a 1 and 3-year PFS of 51.8% and 42.4%, respectively, and for OS 78.3% and 51.7%, respectively. All patients had at least one treatment related AE (TRAE), defined as possibly, probable, or definitely related to treatment. 39% of pts had a G3 TRAE and the most common ones were: aspiration, dehydration, anemia, dysphagia, dyspnea. 4 patients had a G5 TRAE (found down deceased, epistaxis, aspiration pneumonia, iRAE pneumonitis). Conclusions: Our prospective trial uniquely evaluated ReRT plus pembrolizumab in patients with locoregional recurrence/second primary HNSCC that did not undergo salvage surgery. The primary endpoint of PFS was not improved compared to historical control. PD-L1 biomarker analysis is ongoing and will be reported subsequently. Clinical trial information: NCT02289209 .