PDCD1 Polymorphisms May Predict Response to Anti-PD-1 Blockade in Patients With Metastatic Melanoma

Abstract

A significant number of patients (pts) with metastatic melanoma do not respond to anti-programmed cell death 1 (PD1) therapies. Identifying predictive biomarkers therefore remains an urgent need. We retrospectively analyzed plasma DNA of pts with advanced melanoma treated with PD-1 antibodies, nivolumab or pembrolizumab, for five PD-1 genotype single nucleotide polymorphisms (SNPs): PD1.1 (rs36084323, G>A), PD1.3 (rs11568821, G>A), PD1.5 (rs2227981, C>T) PD1.6 (rs10204225, G>A) and PD1.9 (rs2227982, C>T). Clinico-pathological and treatment parameters were collected, and presence of SNPs correlated with response, progression free survival (PFS) and overall survival (OS). 115 patients were identified with a median follow up of 18.7 months (range 0.26 - 52.0 months). All were Caucasian; 27% BRAF V600 mutation positive. At PD-1 antibody commencement, 36% were treatment-naïve and 52% had prior ipilimumab. The overall response rate was 43%, 19% achieving a complete response. Overall median PFS was 11.0 months (95% CI 5.4 - 17.3) and median OS was 31.1 months (95% CI 23.2 - NA). Patients with the G/G genotype had more complete responses than with A/G genotype (16.5% vs. 2.6% respectively) and the G allele of PD1.3 rs11568821 was significantly associated with a longer median PFS than the AG allele, 14.1 vs. 7.0 months compared to the A allele (p=0.04; 95% CI 0.14 - 0.94). No significant association between the remaining SNPs and responses, PFS or OS were observed. Despite limitations in sample size, this is the first study to demonstrate an association of a germline PD-1 polymorphism and PFS in response to anti-PD-1 therapy in pts with metastatic melanoma. Extrinsic factors like host germline polymorphisms should be considered with tumor intrinsic factors as predictive biomarkers for immune checkpoint regulators.

Keywords: PD1; immunotherapy; metastatic melanoma; polymorphism; predictive biomarker.

Figure 1

Expected (E)-to-observed (O) ratio of responses according to PD-1 SNP. Responses for all 4 SNPs (A–D) are shown as complete response (CR), partial response (PR), stable disease (SD) or progressive disease (PD) by dividing number of expected responses by number of observed responses within each genotype. The solid line at y=1 represents a theoretical E:O ratio of 1:1. NA signifies data points where one genotype had 0 events.

Figure 2

Kaplan-Meier curves of progression-free survival stratified by the individual SNPs. (A) the median PFS for PD1.3 rs11568821 GG allele was 14.1 months versus 7.0 months for the AG allele (HR 0.836 (95%CI (0.50-1.39) p=0.49); (B) the median PFS for PD1.5 rs2227981 CT allele 16 months (HR 0.79 (95%CI (0.48-1.27) p=0.329), 2.1 months for the TT allele (HR 1.78 (95%CI (0.70-4.57) p=0.228) versus 8.1 months for the CC allele; (C) the median PFS for PD1.6 rs10204525 21.3 months for the AG allele versus 9.2 months for the GG allele (HR 0.75 (95%CI (0.40-1.40) p=0.364); (D) the median PFS for PD1.9 rs2227982 CC allele was 11.0 months versus 5.4 months for the CT allele (HR 0.73 (95%CI (0.27-2.01) p=0.543).