Abstract 6684: Association of human leukocyte antigen (HLA) homozygosity with unfavorable clinical outcomes in patients with non-small cell lung cancer (NSCLC) treated with PD-L1 inhibitors as frontline therapy

Background: High polymorphic diversity in HLA class I (HLA-I) is essential for successful immunologic control of cancer. Lung cancer patients with maximal HLA-I heterozygosity treated with single-agent PD-L1 inhibitors as their second-line or beyond were reported with longer overall survival. However, the role of HLA heterozygosity for patients with NSCLC who received single-agent immunotherapy with or without chemotherapy as their first-line has not been reported. Methods: Patients with unresectable stage III or IV NSCLC who underwent next-generation sequencing with HLA typing were enrolled in our study (N=27). HLA genotypes were determined using DNA sequencing data. They received either pembrolizumab (N=5) or pembrolizumab plus chemotherapy (N=22) as their first-line treatment. Kaplan-Meier curves were constructed to compare survival outcomes according to HLA (homozygous vs. heterozygous) and tumor mutation burden (TMB, above vs. below median) status. A log-rank test was applied to compare Kaplan-Meier curves between groups. Hazard ratios were derived from the Cox-proportional hazards model with adjustment for baseline characteristics including age, sex, and TMB. Results: We determined HLA-I genotypes consisting of two alleles in 27 patients with NSCLC. Among these enrollees, 23 patients (85.2%) were heterozygous at every HLA-I locus and four patients (14.8%) were homozygous at more than one HLA-I locus. Response Evaluation Criteria in Solid Tumors (RECIST) evaluable patients were 19. In the HLA-heterozygous group (N=15), three patients (20%) had partial response (PR), eight patients (53.3%) had stable disease, and four patients (26.7%) had progressive disease (PD), whereas in the HLA-homozygous group (N=4), one (25%) had PR, one (25%) had SD, and two (50%) had PD. The disease control rates (CR, PR, and SD) were 73.3% in the HLA-heterozygous group and 50% in the HLA-homozygous group. Homozygosity at HLA-I genes seemed to be associated with shorter progression-free survival (PFS, HR=3.2, 95% CI=0.71-15, p=0.13) compared with heterozygosity at HLA-I genes. The HLA-heterozygous group had longer median PFS than the HLA-homozygous group (19 vs. 9.5 months, Log-rank p=0.11). When we controlled the effect of HLA homozygosity on PFS for age, sex, and TMB, the adjusted HRs were 4.4[0.82-23], 3.1[0.69-14] and 1.9[0.36-9.8] respectively. Conclusion: No data exists on the effect of HLA homozygosity on clinical outcomes of patients treated with immunotherapy with or