Abstract 1498
Background
The class I human leucocyte antigen (HLA) molecules play a critical role in tumor recognition by T cells and the loss of expression seems to be an escape mechanism of antitumoral immunity. Novel immune-targeting cancer vaccines are currently developped in HLA-A2 positive patients for modulating the T cells immune response. We hypothesized that HLA-A2 status could influence the prognosis and response to immunotherapy.
Methods
Advanced NSCLC patients treated with nivolumab, pembrolizumab or atezolizumab were prospectively included from Nov. 2013 to Apr. 2016 in our institute. HLA-A2 status was analysed by flow cytometry; PDL1 was analysed by immunohistochemistry. Clinical and biological data were collected at baseline and after cycle 1. Statistical analysis was performed with SPSS v.20.
Results
Out of 125 patients treated, HLA-A2 status was available for 30 patients. 50% were male, median age was 61 years (29-77); 86% were smokers and 83% had performance status 0-1. 18 (60%) were adenocarcinoma, 7 (23.3%) squamous and 5 (16.7%) others histologies. 2 NSCLC were EGFRmut, 2 ALK positive, 7 KRASmut, 19 wildtype. PDL1 expression was positive in 8 patients (26.7%), negative in 2 (6.7%) and unknown in 20 (66.7%). The median of previous lines of treatment was 1 (1-8). The patient and tumor characteristics were well balanced according to HLA-A2. The median progression free survival to immunotherapy (iPFS) was 1.47 months [confidence interval (CI) 95% 1.18-1.7]. HLA-A2 positive status was correlated with longer iPFS [2.04 vs. 1.3 months, log-rank; p = 0.020]. The median overall survival (OS) was 38.05 months [CI 95% 8.94-67.16]. The median OS was 55.9 months [CI 95% 9.4-102.15] vs. 22.5 months [CI 95% 0- 54.8] in HLA-A2 positive vs. negative patients (p = 0.340). In univariate analysis, there was no correlation between outcome and patient or tumor characteristics.
Conclusions
Our preliminary results suggest that HLA-A2 status could influence the outcome in NSCLC patients treated with immune checkpoint inhibitors. An updated analysis on 59 patients will be presented.
Clinical trial identification
Legal entity responsible for the study
Dr Benjamin Besse
Funding
Gustave Roussy
Disclosure
All authors have declared no conflicts of interest.