Abstract 7612: Genetic predictors of anti-PD(L)-1 therapy response in hepatocellular carcinoma

Abstract Introduction: Immunotherapy targeting PD-1/PD-L1 shows promise in treating hepatocellular carcinoma (HCC), but standard predictors like PD-L1 expression are unreliable for gauging treatment response. RNA sequencing has identified a BMS 4-gene inflammatory signature that could more accurately forecast clinical outcomes. However, genomic alterations and tumor mutational burden (TMB) haven't consistently predicted anti-PD (L) -1 therapy response in HCC. Therefore, the aim of this research is to identify potential genetic alternation that could serve as predictive marker for anti-PD (L) -1 therapy responses in advanced HCC. Methods: Formalin-fixed paraffin-embedded (FFPE) tumor tissues from 38 HCC patients were analyzed using targeted next-generation sequencing with the ACTOncoTM panel before anti-PD (L) -1 therapy. This analysis identified somatic variants across 440 genes and calculated TMB. In silico tools like NetMHC and IEDB predicted neo-peptide-HLA class I binding affinity. Additionally, peripheral blood mononuclear cells provided germline data for paired TMB calculations, with tumor responses assessed by RECIST criteria. Results: Within a cohort of 167 patients with HCC undergoing ICI therapy, 35 (20. 9%) responded to treatment, while 132 (79. 1%) did not. Available formalin-fixed paraffin-embedded (FFPE) samples from this population included those from 14 responders and 30 non-responders. After conducting quality control measures, which assessed tumor purity, DNA quantity and integrity, as well as sequencing quality, 13 samples from responders and 25 from non-responders qualified for further analysis. Gene alteration analysis identified mutations in TP53 (47%), MUC16 (34%), CTNNB1 (32%), USH2A (16%), ARID1A (13%), NOTCH3 (13%), and ADAMTS16 (11%) of the cases. A TP53 mutation was linked to poor overall survival, whereas mutations in CTNNB1 and the RTK/RAS/RAF pathway did not correlate with clinical outcomes. The median TMB, predicted at 3. 2 ± 4. 3 mutations per megabase (muts/Mb), was nominally higher in patients who achieved disease control compared to non-responders (4. 6 vs. 3. 7 muts/Mb, p=0. 53). Utilizing paired TMB analysis, those with disease control exhibited a higher TMB trend than non-responders (6. 0 vs. 2. 2 muts/Mb, p=0. 06). Notably, the 13 cases with MUC16 mutations were associated with a significantly higher predicted TMB (10. 0 vs. 3. 3 muts/Mb, p=0. 0015). Conclusions: This study suggests that certain genetic alterations, especially TP53 mutations, may have predictive value for the response to anti-PD (L) -1 therapy in HCC. Moreover, an elevated TMB, particularly when associated with MUC16 mutations, seems to be correlated with improved disease control. Citation Format: San-Chi Chen, Nai-Jung Chiang, MingHuang Chen, Muh-Hwa Yang. Genetic predictors of anti-PD (L) -1 therapy response in hepatocellular carcinoma abstract. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts) ; 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84 (6Suppl): Abstract nr 7612.