Outcomes of advanced melanoma patients who discontinued pembrolizumab (pembro) after complete response (CR) in the French early access program (EAP)

Abstract

Background: Information on outcomes after cessation of anti-PD-1 mAb for CR are scarce outside KN-001 & 006 trials. We investigated these outcomes in the French pembro EAP, where pts were less selected than in registration trials. Methods: A multicenter ambispective cohort of advanced melanoma pts initiating pembro between May 2014-Sept 2015 (CCTIRS, #15.640) was searched for pts who achieved CR. Overall and progression-free survivals with Log rank tests were used to compare curves. Results: 687 pts (159 with brain metastases) were included from 41 centers, which covered 75% of French pembro EAP. 79 pts (13%) achieved CR, among whom 68 (10%) stopped pembro for CR. Compared with the remaining 619 pts surveyed, these 68 pts had lower baseline serum LDH levels (54.5% vs 73.8%, respectively, were normal, p = 0.016), lower ECOG PS (ECOG 0 for 298 (52.6%) and 54 (80.6%), respectively, p = 0.001), and fewer metastatic sites (p = 0.003). Of note, 10% of these 68 pts had >3 metastatic sites, 10% had brain metastases, 53% had 2010 AJCC stage M1c melanoma, and 67% had received previous treatment for advanced melanoma (mainly ipilimumab). Frequency of BRAFV600, NRAS, and c-kit mutation status was not different between pts who did or did not achieve CR. Median duration on pembro in these 68 pts was 21.2 m (range: 2.5-48.4, SD 10.0). Pembro was used alone to achieve CR in 79% of pts, while 4% received concurrent radiotherapy alone, 10% concurrent surgery alone, and 6% both treatments. At the cut-off date (13/09/2018), after a median follow-up since pembro initiation of 36.3 m and of 15.3 m since pembro cessation (2-31, SD 8.3), only 3% of pts experienced melanoma recurrence. Median recurrence-free survival was not achieved. These 2 pts received radiotherapy, with one partial response achieved. 5 pts died, with 2 deaths unrelated to melanoma (1 lymphoma, 1 chylothorax), 1 related to melanoma progression, and 2 of unknown causes. Conclusions: In this large real-life cohort, 10% of pts reached CR and discontinued pembro. All but 3% were recurrence-free after a median 15.3 m of treatment, thus validating the discontinuation of anti-PD1 in complete responders. Clinical trial identification: CCTIRS, #15.640. Editorial acknowledgement: RIC-Mel network and their team. Legal entity responsible for the study: MSD France. Funding: MSD France. Disclosure: P. Saiag: Honoraria (self), Advisory / Consultancy: MSD. L. Mortier: Honoraria (self), Advisory / Consultancy: MSD. C. Dutriaux: Honoraria (self), Advisory / Consultancy: MSD. L. Benmahammed: Full / Part-time employment: MSD. O. Morsli: Full / Part-time employment: MSD. C. Train: Full / Part-time employment: ClinSearch. A. Spampinato: Full / Part-time employment: MSD. M.T. Leccia: Honoraria (self), Advisory / Consultancy: MSD. N. Meyer: Advisory / Consultancy: MSD. J.J. Grob: Honoraria (self), Advisory / Consultancy: MSD.