The predictive values of POTEE mutations for response to immune checkpoint inhibitors in lung adenocarcinoma.

Abstract

e21110 Background: Precise selection of patients who could benefit from immune checkpoint inhibitors (ICIs)is an important challenge for immunotherapy in lung cancer. POTEE is a member of one primate-specific gene family which have been identified as cancer-related antigens and potential target for immunotherapy of cancer. Here, we investigated the correlation between POTEE mutation and the clinical outcomes of ICIs treatment in lung adenocarcinoma (LUAD). Methods: We merged three independent cohorts from publicly accessible data, the Miao(n = 47), Rizvi (n = 29) and Hellmann (n = 59), to explore the associations between POTEE mutations ( POTEE-mut) and ICIs efficacy. Then we explored the relationship between POTEE mutation and tumor mutational burden (TMB) and PD-L1 expression in the merged cohort. Further, gene set enrichment analyses (GSEA) was used to explore the potential molecular mechanism in The Cancer Genome Atlas (TCGA)-LUAD cohort. Results: In merged cohort, the mutation frequency of POTEE was 7.41% (10 in 135). POTEE-mut group had a significantly higher objective response rate (ORR) (100% vs 27.2%; Fisher’s exact test; P < 0.001) compared to POTEE wildtype group. Kaplan-Meier analysis revealed that POTEE-mut was significantly associated with longer progression-free survival (PFS) (not reached vs 6.1 months, P = 0.001). And at multivariable analysis adjusted by cohort, age, smoking status, line of immunotherapy, PD-L1 expression and TMB, POTEE-mut group showed significant improved PFS with a HR of 0.09(95% CI 0.01-0.71, P = 0.022). Then we explored the association of POTEE alterations with TMB and PD-L1 expression. In merged cohort significantly higher TMB were found in LUAD patient with POTEE mutation (Mann-Whitney U test; p < 0.001). However, there was no association between POTEE mutation and PD-L1 expression. We further explored the potential mechanism of POTTE mutation affecting immunotherapy response using TCGA-LUAD data. GSEA analysis revealed prominent enrichment of signatures related to DNA repair in patients with POTEE mutation (P < 0.001). In detail, four DNA damage response (DDR) pathways, homologous recombination repair (HRR) (NES = 2.108, P < 0.001), translesion DNA synthesis (NES = 1.587, P = 0.01), checkpoint factors (NES = 1.865, P < 0.001) and base excision repair (NES = 1.737, P = 0.002), were all significantly enriched in POTEE-mut tumor. Conclusions: Our data indicated that POTEE mutations were associated with the better response for immunotherapy of LUAD, implying that it can be used as a biomarker for predicting the efficacy of LUAD immunotherapy. POTEE mutation was associated with higher TMB and changed the activity of DNA repair related pathways in LUAD. These may be the underlying mechanisms for POTEE mutations to affect the prognosis of immunotherapy.