Association of single nucleotide polymorphisms with efficacy in nivolumab-treated NSCLC patients

Abstract

Background: Proper patient selection for PD-1 checkpoint inhibitors is crucial given its limited efficacy in the majority of patients. We previously showed that a single nucleotide polymorphism (SNP) in the auto-immunity (AI) related PTPN11 gene (rs2301756) is associated with increased toxicity on nivolumab (Bins et al, ESMO immuno-oncology symp, 2016). The objective of the current analysis was to assess whether SNPs in PTPN11 and other genes associated with AI, which are putatively considered important in PD-1 influenced T-cell responses, are correlated with treatment efficacy of nivolumab in NSCLC patients. Methods: The association between 5 SNPs and efficacy was evaluated in 161 (chemotherapy pretreated) advanced NSCLC patients. Efficacy measures included early progressive disease (PD; ≥90 days after start), tumor response according to RECISTv1.1, PFS and OS. The SNPs were located on 4 genes, being PDCD1, PTPN11, ZAP70 and IFNG, respectively, encoding for the proteins PD-1, SHP-2 and ZAP70 and IFN-γ. The best model for every SNP, being either a dominant, recessive, multiplicative or additive model, was selected. SNPs were analyzed in multivariable logistic regression models if they showed a p-value <0.1 in univariable analysis, and were corrected for age and gender. PFS and OS analysis was done using the Kaplan-Meier method. Results: Overall, 54 patients showed PR or CR according to RECIST. Patients with at least one variant allele in PTPN11 had a higher response rate compared to wild-type (50% vs 30%; OR 2.4; 95% CI 1.0 - 5.5; p=0.042). Other SNPs were not associated with response rate, and none of these SNPs were associated with early PD. Until the day of analysis, at which 128 patients had PD and 107 patients were dead, none of the SNPs was associated with PFS or OS. Conclusions: Our results show that the SNP in PTPN11 predisposes for a higher response rate to nivolumab therapy. Together with our earlier finding that this SNP is associated with grade ≥ 3 toxicity, this may indicate that SHP-2 activity influences treatment outcome of anti-PD-1 therapy in terms of both toxicity and anti-tumor effects. If confirmed, SNPs in PTPN11 should be considered to be included as a biomarker in routine analysis of NSCLC patients to be treated with nivolumab.