Abstract 1068P
Background
Information on RT in anti-PD-1-treated melanoma pts is limited.
Methods
Search for all pts with unresectable AJCC stages III-IV melanoma initiating anti-PD-1 without ipilimumab between 1/1/15-30/8/19 in a prospective database of a referral center. Radiologists evaluated radiated and non-radiated lesions (RECIST 1.1) every 3 m. Main criteria were the complete response (CR) rate. Overall (OS) and progression-free (PFS) survivals were calculated using Kaplan-Meier. Multivariate analysis studied factors associated with CR or partial response (PR). Database lock: 30/11/20.
Results
206 pts (59% M1c/d, 50% ≥3 metastasis sites, 33% ECOG PS≥1, 33% >1 line, 32% LDH>ULN) received nivolumab (83%) or pembrolizumab (17%). Median follow-up was 19.5 m. A total of 100 pts (49%) received first RT early (<3 m of PD-1 blockade, in 39 pts with rapidly progressing lesions) or late (>3 m, in 61 pts with confirmed non-response to anti-PD-1 therapy). First RT was hypo-fractionated RT to 1-2 targets (26 Gy in 4 weekly radiations (68 pts), radiosurgery (SRS) for brain mets (25 pts), or standard RT. 39 pts received a second RT. Globally, 66 (32% [95%CI:25.7-38.4]) pts achieved CR (with anti-PD-1 discontinuation in 64 pts, with 9 pts (14%) relapsing to date). RT added 24 CR to the 42 CR observed without. Median PFS and OS were 15.6 [95%CI: 10.7-22.8] and 24.6 [16.9-41.8] m, respectively. In pts with P, PFS and OS were 16.6 [13.2-25.0] and 36.4 m[24.2-NA], respectively, in radiated pts and 2.4[2.2-3.5] and 4.9 m[2.6-7.7], respectively, in non-radiated pts (P<0.001). In radiated pts, rates for radiated lesions of CR, PR, stable (SD), progressive (PD) disease were 26%, 26%, 8%, and 37% (non-evaluable:3%), respectively, and 24%, 14%, 1%, and 61%, respectively, for non-radiated lesions. Abscopal response was observed in non-radiated lesions in 30% of pts radiated late for non-response to anti-PD-1. AJCC staging, naïve vs non-naïve, ECOG PS, LDH serum level, <3 sites with metastasis vs ≥3, oligo vs multimetastatic disease, hypo-fractionated RT vs SRS, early vs late RT were not associated with CR+PR in radiated patients. No unusual adverse event was seen.
Conclusions
High-fraction doses RT may enhance anti-PD-1 efficacy and CR rate above the 15-21% seen in registration studies.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
Has not received any funding.
Disclosure
P. Saiag: Financial Interests, Personal and Institutional, Advisory Board: MSD; Financial Interests, Personal, Advisory Board: Merck Serono; Financial Interests, Personal, Advisory Board: BMS; Financial Interests, Personal, Advisory Board: Pierre-Fabre; Financial Interests, Personal, Advisory Board: Novartis; Financial Interests, Personal, Advisory Board: Roche. A. Blom Fily: Financial Interests, Personal, Advisory Board: Merck Serono. C. Longvert: Financial Interests, Personal, Advisory Board: Novartis; Financial Interests, Personal, Advisory Board: MSD. E. Funck-Brentano: Financial Interests, Personal, Advisory Board: Novartis; Financial Interests, Personal, Advisory Board: MSD; Financial Interests, Personal, Advisory Board: BMS. All other authors have declared no conflicts of interest.