Immune cell surface proteins from tumor samples to predict patient clinical outcome.

Abstract

e15630 Background: The incidence and mortality of Hepatocellular carcinoma (HCC) are rapidly increasing in the US. Treatments for advanced HCC are limited despite best efforts to develop therapeutics that target the deregulated pathways of HCC. Recent studies reveal a direct causal relationship between cancer and immune dysfunction, whereby tumor cells and their microenvironment can evade immune attack by exploiting various immunoregulatory mechanisms in a process termed “cancer immune editing”. The objective is to determine the relationship between immune cell surface protein transcripts from tumor samples and the clinical outcome of pts by performing exploratory analyses of liver cancer data from the cancer genome atlas (TCGA). Methods: We analyzed RNAseq data from a cohort of 75 HCC samples in the TCGA collection. We explored the association of expression of a group of specific markers for infiltrating lymphocytes (several subtypes) with overall survival (OS) and these results were confirmed by IHC analysis of the same cell subtypes in HCC samples from patients at GUH. Results: Among the 75 HCC cases analyzed, 25 pts had Hepatitis C (HCV), 25 had Hepatitis B (HBV), and 25 had HCV and HBV co-infection. Fourteen of the 75 cases were found to express the CD8B isoform. The LOG RANK survival test showed a significant association between CD8B levels and OS (ChiSq = 5.2, 1df; p = 0.0222), as did the Cox proportional hazard model for survival. The latter test showed that pts with HCC that did not express CD8B were at a higher risk of death compared with those who had CD8B-expressing HCC. Additional immune cell subtype specific transcripts are being tested. Based on analysis of differentially expressed gene differences between immune cell subtype ‘low’ vs. ‘high’ groups, pathways highly relevant to lymphocyte signaling and immune response and infiltration were identified. Thus, 5 tumor-infiltrating immune cell subtypes were found to have wide variability among patients. Using IHC, these cell subtypes were assessed in HCC samples from patients and correlated to clinical outcome. Conclusions: Immune cell specific transcripts in tumor samples may serve as a predictor of response to treatments that target immune evasion in cancer patients.