BackgroundImmune check-point blockade (ICB) is one of the emerging therapeutic options for advanced hepatocellular carcinoma (HCC). However, low response rates amongst patients necessitates the development of robust predictive biomarkers that identify patients who likely benefit from ICB. Previously our group found that immunohistochemical scoring of CD38 in the tumour microenvironment predicts responsiveness to anti-PD-1/anti-PD-L1 immunotherapy in HCC.1 Recently BMS 4-gene inflammatory signature, comprising the 4 genes CD8, PD-L1, LAG-3 and STAT1, has been shown to be associated with better overall response to immunotherapy in various cancer types.2–4 In the present study, we examined the relationship between tissue expression of BMS 4-gene inflammatory signature and the responsiveness of HCC to immunotherapy, and whether BMS 4-gene inflammatory signature increases the predictive power of CD38.MethodsHCC tissue samples from 124 Asian patients that underwent conventional treatment and from 49 Asian patients that underwent ICB were analysed for CD8, PD-L1, LAG-3, STAT1, CD38 and CD68 tissue expression using immunohistochemistry and multiplex immunohistochemistry, followed by survival and statistical analysis.ResultsSurvival analysis of the 124 samples showed that high LAG-3 tissue expression was associated with shorter progression-free survival (PFS). On the other hand, immunohistochemical analyses on the 49 patient samples treated with ICB revealed that high LAG-3 density and high total LAG-3+CD8+ T cell proportion were associated with improved response to ICB (figure 1). However, CD8, PD-L1 and STAT1 levels did not significantly correlate with improved survival. The addition of total LAG-3+ cell proportion to total CD38+ cell proportion significantly increased the predictive value for both PFS (DeltaLRChi2=9.97; P=0.0016; table 1) and overall survival (OS) (DeltaLRChi2=8.84; P=0.0021; table 1), compared with total CD38+ cell proportion alone. Similarly findings were obtained after adding total LAG-3+CD8+ cell proportion to total CD38+ cell proportion (PFS: DeltaLRChi2=7.21; P=0.0072; OS: DeltaLRChi2=8.06; P=0.0045; table 1), compared with total CD38+ cell proportion alone. Lastly, when the total LAG-3+CD8+ cell proportion was added to total CD38+ and CD38+CD68+ cell proportion, the predictive value of the biomarker was significantly increased (PFS: DeltaLRChi2=6.10; P=0.0136; OS: DeltaLRChi2=6.18; P=0.0129; table 1). Ongoing works include further validation of the findings in various cohorts, and correlating with clinical outcome of the patients.Abstract 89 Table 1Log-likelihood of models with added predictive termsAbstract 89 Figure 1HCC patients‘ response to ICB in relation to LAG-3 density. (A) Kaplan-Meier curve showing the association between a high LAG-3 density and improved overall survival after treatment with ICB. (B) Kaplan-Meier curve showing the association between a high total LAG-3+CD8+ T cell proportion and improved overall survival after treatment with ICB. (C) Kaplan-Meier curve showing the association between a high LAG-3 density and improved progression-free survival after treatment with ICB. (D) Kaplan-Meier curve showing the association between a high total LAG-3+CD8+ T cell proportion and improved progression-free survival after treatment with ICB.ConclusionsHigh LAG-3 expression on tissue-infiltrating immune cells predicted greater response to ICB. LAG-3+ and LAG3+CD8+ cell proportion added predictive value to CD38+ cells for predicting survival outcome in immunotherapy-treated HCC. LAG-3 may be used in conjunction with CD38 to predict responsiveness to ICB in HCC.ReferencesNg HHM, Lee RY, Goh S, et al. Immunohistochemical scoring of CD38 in the tumor microenvironment predicts responsiveness to anti-PD-1/PD-L1 immunotherapy in hepatocellular carcinoma. J Immunother Cancer 2020;8.Hodi FS, Wolchok JD, Schadendorf D, et al. Abstract CT037: Genomic analyses and immunotherapy in advanced melanoma. AACR 2019.Lei M, Siemers NO, Pandya D, et al. Analyses of PD-L1 and Inflammatory Gene Expression Association with Efficacy of Nivolumab ± Ipilimumab in Gastric Cancer/Gastroesophageal Junction Cancer. Clinical Cancer Research 2021;27:3926–35.Sangro B, Melero I, Wadhawan S, et al. Association of inflammatory biomarkers with clinical outcomes in nivolumab-treated patients with advanced hepatocellular carcinoma. J Hepatol 2020.Ethics ApprovalThis study was approved by the Centralised Institutional Review Board of SingHealth (CIRB ref: 2009/907/B).ConsentWritten informed consent was obtained from the patient for publication of this abstract and any accompanying images. A copy of the written consent is available for review by the Editor of this journal.