1106P NivoLag-when: International real-world study of combination immunotherapy sequences in melanoma
Background
Three immune checkpoint inhibitor (ICI) regimens are standard of care in metastatic melanoma (MM): anti-PD1 combined with anti-CTLA4 (ipi/nivo), with anti-LAG3 (rela/nivo) or as monotherapy. Data describing the efficacy of sequential regimens are limited.
Methods
This multicenter retrospective and prospective study assessed patients (pts) who received rela/nivo followed by ipi/nivo (A); ipi/nivo followed by rela/nivo (B); or anti-PD1 followed by rela/nivo (C). Primary endpoint was objective response rate (ORR) to second treatment (Tx). Secondary endpoints were progression-free survival (PFS), overall survival (OS) and safety.
Results
With a median follow-up of 62.5 months (mo), 130 pts (A=26; B=43; C=61) from Europe, USA and Australia were included. Favorable baseline characteristics were observed in C, with lower prevalence of stage M1c-d (26.3%) compared with A (88.4%) and B (67.5%) and fewer pts with high LDH (14.8%) compared with A (46.2%) and B (32.6%). In C, 12 pts had received anti-PD1 as adjuvant Tx, of which 8 were primary resistant (PRes), as defined by the SITC (Kluger JITC 2020). B had more pretreated pts with ≥2 lines in 44.2%, vs 7.7% in A and 8.2% in C. ORR were 15.4%, 19% and 26.8% in A, B and C respectively. ORR were numerically lower in PRes to prior Tx than for secondary resistance (SRes) in A and B. In C, ORR was 33.4% (95% CI 9.9-65.1) for pts who received prior adjuvant anti-PD1 vs 23% (95% CI 11.8-38.6) in the metastatic setting. Disease control rates (DCR) were comparable in A and B. Six-mo DCR were 11.5% in A, 23.3% in B and 29.5% in C. No unexpected toxicity was observed and recurrent adverse events occurred in 7.7% in A, 11.6% in B and 13.1% in C. Discontinuation due to toxicity was 31% in A, 4.6% in B and 4.9% in C.
Conclusions
Despite the limitation of a retrospective study, our results suggest that pts with MM respond more frequently to sequential ICI combination when they have previously responded to ICI (SRres), whereas PRes does not impair ORR in pts treated with adjuvant anti-PD1.