Establishment, immunological analysis, and drug prediction of a prognostic signature of ovarian cancer related to histone acetylation

Abstract

In recent years, epigenetic modifications have been increasingly regarded as an important hallmark of cancer. Histone acetylation, as an important part of epigenetic modification, plays a key role in the progress, treatment, and prognosis of many cancers. In this study, based on the TCGA database, we performed LASSO regression and the Cox algorithm to establish a prognostic signature of ovarian cancer associated with histone acetylation modulator genes and verified it externally in the GEO database. Subsequently, we performed an immunological bioinformatics analysis of the model from multiple perspectives using the CIBERSORT algorithm, ESTIMATE algorithm, and TIDE algorithm to verify the accuracy of the model. Based on the prognostic model, we divided ovarian cancer patients into high-risk and low-risk groups, and assessed survival and the efficacy of accepting immunosuppressive therapy. In addition, based on the analysis of characteristics of the model, we also screened targeted drugs for high-risk patients and predicted potential drugs that inhibit platinum resistance through the connectivity map method. We ultimately constructed a histone acetylation modulator-related signature containing 10 histone acetylation modulators, among which HDAC1, HDAC10, and KAT7 can act as independent prognostic factors for ovarian cancer and are related to poor prognosis. In the analysis of the tumor microenvironment, the proportion of the B-infiltrating cells and the macrophages was significantly different between the high- and low-risk groups. Also, the samples with high-risk scores had higher tumor purity and lower immune scores. In terms of treatment, patients in the high-risk group who received immunotherapy had a higher likelihood of immune escape or rejection and were less likely to respond to platinum/paclitaxel therapy. Finally, we screened 20 potential drugs that could target the model for reference.