IDDF2021-ABS-0065 The tumor immunological characterizations of colon adenocarcinoma reveal clinically relevant population-specific differences

Abstract

Background Immunomodulation has become a topical area of interest in many tumors, being colon adenocarcinoma (COAD), not an exception. Considering the molecular heterogeneity of COADs and their non-immunogenic character, immunotherapy only showed a viable role in a certain subset of COADs. This study aimed to determine immune subtypes (ISs) of COADs for the selection of suitable patients from an extremely heterogeneous population. Methods Gene expression profiles and corresponding clinical information were collected from TCGA and GEO databases, respectively. Consensus clustering analysis was performed to identify the ISs. Immunogenomics methods were integrated to characterize the immune environment of each IS subtype. Linear discriminant analysis was conducted to establish an immune subtyping characteristic index (ISCI) for IS classification of COADs. Co-expression network analysis was used for detected hub genes. DAVID was used for functional annotation. Results Based on 17 prognostic immune characteristics, COADs were stratified into three ISs characterized by differential molecular, cellular and clinical features. Patients with the IS1 tumor had immune ‘hot’ and immunosuppressive phenotype, IS3 tumor had immune ‘hot’ phenotype, whereas those with the IS2 tumor had immune ‘cold’ phenotype. Patients with the IS1 tumor had the worst disease-free survival compared with the other IS types, while those with the IS3 tumor had the best prognosis. Moreover, the proportion of highest adenomatous polyposis coli (APC) mutations in IS2 subtype is significantly higher than that in the IS1 and IS3; the proportion of TP53 mutations in IS1 subtypes were significantly higher than IS2 and IS3; while the proportion of KRAS mutations in IS1 subtypes were significantly lower than IS2 and IS3. IS3 subtype is predicted more sensitive to Cisplatin than other ISs, while IS1 is predicted more sensitive to 5-FU. Furthermore, the ISCI developed based on immune subtypes of COADs revealed immune infiltration degree in individual patients and can be utilized to determine the IS of patients with COAD. Based on 17 prognostic immune characteristics, COADs were stratified into three ISs. Patients with the IS1 tumor had the worst disease-free survival compared with the other IS types, while those with the IS3 tumor had the best prognosis (IDDF2021-ABS-0065 Figure 1, IDDF2021-ABS-0065 Figure 2, IDDF2021-ABS-0065 Figure 3, IDDF2021-ABS-0065 Figure 4). The proportion of highest adenomatous polyposis coli (APC) mutations in IS2 subtype is significantly higher than that in the IS1 and IS3; the proportion of TP53 mutations in IS1 subtypes were significantly higher than IS2 and IS3; while the proportion of KRAS mutations in IS1 subtypes were significantly lower than IS2 and IS3 (IDDF2021-ABS-0065 Figure 5, IDDF2021-ABS-0065 Figure 6). Moreover, distinct molecular, cellular and clinical features were observed among different IS tumors: patients with the IS1 tumor had immune ‘hot’ and immunosuppressive phenotype, IS3 tumor had immune ‘hot’ phenotype, whereas those with the IS2 tumor had immune ‘cold’ phenotype (IDDF2021-ABS-0065 Figure 7, IDDF2021-ABS-0065 Figure 8, IDDF2021-ABS-0065 Figure 9, IDDF2021-ABS-0065 Figure 10). IS3 subtype is predicted more sensitive to Cisplatin than other ISs, while IS1 is predicted more sensitive to 5-FU (IDDF2021-ABS-0065 Figure 11). Furthermore, the ISCI developed based on immune subtypes of COADs revealed immune infiltration degree in individual patient and can be utilized to determine the IS of patients with COAD (IDDF2021-ABS-0065 Figure 12, IDDF2021-ABS-0065 Figure 13). Specifically, high ISCI level is correlated with high CTLA4, PDCD1 and CD274 (PD-L1) expression (IDDF2021-ABS-0065 Figure 14), and the ISCI level of the partial response (PR)/complete response (CR) group is significantly lower than of stable (SD) and progressive disease (PD) group (IDDF2021-ABS-0065 Figure 15). Conclusions The immune subtyping and ISCI system are indicative for the prediction of tumor prognosis of COADs. Identification of immune subtypes may facilitate the optimal selection of COAD patients responsive to adequate therapeutic strategies.