184P - Patient-specific HLA-I genotypes predict response to immune checkpoint blockade

Background

Specific shared HLA Class 1 alleles have been linked to the response to immune checkpoint blockade (ICB). In this study, we aimed to identify the HLA-A subtypes associated with maximum benefit from ICB.

Methods

We compiled a prospectively maintained clinical dataset of patients who underwent CLIA-approved germline HLA status testing as part of various advanced immune and cell therapy trials undertaken at the Christie NHS Foundation Trust. The probability of overall survival (OS) or progression-free survival (PFS) was estimated by the Kaplan-Meier method, and their associations with the HLA-A alleles were tested with Cox regression analysis.

Results

A total of 285 patients from 11 immune and cell therapy clinical trials were eligible for final analysis. The median age at diagnosis was 57 years (range: 19 – 82). The tumor categories were NSCLC (n=93), melanoma (n=60), and other cancers (n= 123). We identified 15 HLA-A subtypes, the most common alleles being HLA-A02 (42.2%), HLA-A01 (35.4%), and HLA-A03 (26.0%). A total of 145 patients received ICB either as first-line (56.5%) or as second or later-line (43.5%). Of these, 136 patients had evaluable clinical response status. 24.2% were positive for HLA-A02:01 and were eligible for screening for our HLA-specific trials. Patients with the HLA-A01 and HLA-A30 subtypes were significantly associated with a higher likelihood of clinical response to ICB (odds ratio = 2.58, 95%CI: 1.20 – 5.62, p = 0.017 and 1.80, 95% CI: 1.58-2.18, p= 0.011, respectively). In the NSCLC subgroup, patients with HLA-A01 subtypes were significantly associated with better OS (hazard ratio (HR): 0.42, 95%CI: 0.22-0.81, p= 0.010); meanwhile, HLA-A02 was associated with worse OS (HR: 1.95, 95%CI: 1.09-3.49, p= 0.024). In the other cancers subgroup, HLA-A11 and A24 were associated with better OS (HR: 0.20, 95%CI: 0.05-0.85, p= 0.030 and HR: 0.31, 95% CI: 0.10-0.89, p= 0.029, respectively).

Conclusions

HLA-A status predicted clinical outcomes of patients receiving ICB. HLA genotyping could be incorporated early into the diagnostic work-up of patients with solid cancers as a predictive and selective biomarker.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The Christie NHS Foundation Trust.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.