Integrated multi-omics and experimental approaches identify fascin actin-bundling protein 1 as an unfavorable prognostic biomarker in adrenocortical carcinoma

Abstract

Adrenocortical carcinoma (ACC) is a rare epithelial tumor originating from adrenal cortical cells, notable for its high degree of malignancy and poor prognosis. Owing to heterogeneity, patient outcomes vary significantly. Current biomarkers for ACC risk stratification have notable limitations. However, with the advancement of multi-omics sequencing technology, we can utilize multi-omics data to explore the heterogeneity of ACC, thereby identifying novel biomarkers. In this study, we establish multicenter transcriptomics and ATAC-seq data from the TCGA and GEO databases to perform weighted gene coexpression network analysis (WGCNA) clustering and conduct comprehensive analyses of various ACC samples. These findings are integrated with univariate Cox regression, receiver operating characteristic (ROC) curve analysis, and survival analysis to identify potential biomarkers. We establish FSCN1 as an independent risk factor associated with poor ACC prognosis. ATAC-seq data demonstrate higher chromatin accessibility of FSCN1 in ACC patients with progressive disease. Immunohistochemical analysis confirms the expression of FSCN1 at the protein level, while functional cell assays reveal its role in promoting tumor invasion and proliferation. Functional enrichment analyses highlight the biological characteristics of FSCN1, and estimation of TME-infiltrating cells suggests that FSCN1 expression contributes to poor prognosis by inhibiting CD8+ T-cell infiltration within the ACC microenvironment. Finally, multi-omics analyses elucidate the role of FSCN1 at the mutation level. Taken together, our findings highlight FSCN1 as a promising novel biomarker and potential therapeutic target, underscoring its value in guiding the strategic management of ACC.

1. Introduction

Adrenocortical carcinoma (ACC) is an uncommon yet markedly malignant epithelial tumor that arises from the cells of the adrenal cortex. It accounts for a minor 0.2% of total cancer-related mortality, yet its malignancy and aggressive invasion ability are noteworthy. It is frequently discovered at advanced stages, often with distant metastasis, which leads to a discouraging late-stage survival rate of less than 30% [1,2]. Surgical intervention currently offers the only potential curative approach for ACC. However, this option is feasible only for stage I, II and some stage III patients who are candidates for complete resection. Despite surgical intervention, more than 50% of patients experience either recurrence or metastasis postoperatively. Unfortunately, nonsurgical treatment modalities are limited in their scope and often yield severe side effects. For example, mitotane, a pharmaceutical therapy, has limited efficacy [2]. Conventional therapeutic strategies, such as chemotherapy and radiotherapy, still lack robust evidence supporting their effectiveness in primary ACC [2,3].

Presently, biomarkers for ACC demonstrate inadequate specificity and sensitivity. A minority of these biomarkers are supported by robust evidence, leading to limited clinical utility. Consequently, a deeper understanding of the mechanisms underlying ACC progression, alongside the identification of novel prognostic indicators and therapeutic targets, is crucial for improving patient outcomes.

In recent years, the application of high-throughput sequencing technologies has resulted in exponential growth in biological data. In 2000, the National Center for Biotechnology Information (NCBI) in the United States launched the Gene Expression Omnibus (GEO) database, a comprehensive repository of high-throughput gene expression data provided by international research organizations [4]. Additionally, in 2006, the United States initiated the TCGA project, which offered an extensive range of publicly accessible multi-omics data from cancer patients.