Targeted Assessment of G0S2 Methylation Identifies a Rapidly Recurrent, Routinely Fatal Molecular Subtype of Adrenocortical Carcinoma

Abstract

Purpose: Adrenocortical carcinoma (ACC) is a rare, aggressive malignancy with few therapies; however, patients with locoregional disease have variable outcomes. The Cancer Genome Atlas project on ACC (ACC-TCGA) identified that cancers of patients with homogeneously rapidly recurrent or fatal disease bear a unique CpG island hypermethylation phenotype, "CIMP-high." We sought to identify a biomarker that faithfully captures this subgroup. Experimental Design: We analyzed ACC-TCGA data to characterize differentially regulated biological processes, and identify a biomarker that is methylated and silenced exclusively in CIMP-high ACC. In an independent cohort of 114 adrenocortical tumors (80 treatment-naive primary ACC, 22 adrenocortical adenomas, and 12 non-naive/ nonprimary ACC), we evaluated biomarker methylation by a restriction digest/qPCR-based approach, validated by targeted bisulfite sequencing. We evaluated expression of this biomarker and additional prognostic markers by qPCR. Results: We show that CIMP-high ACC is characterized by upregulation of cell cycle and DNA damage response programs, and identify that hypermethylation and silencing of G0S2 distinguishes this subgroup. We confirmed G0S2 hypermethylation and silencing is exclusive to 40% of ACC, and independently predicts shorter disease-free and overall survival (median 14 and 17 months, respectively). Finally, G0S2 methylation combined with validated molecular markers (BUB1B-PINK1) stratifies ACC into three groups, with uniformly favorable, intermediate, and uniformly dismal outcomes. Conclusions: G0S2 hypermethylation is a hallmark of rapidly recurrent or fatal ACC, amenable to targeted assessment using routine molecular diagnostics. Assessing G0S2 methylation is straightforward, feasible for clinical decision-making, and will enable the direction of efficacious adjuvant therapies for patients with aggressive ACC.

CCNB2 and AURKA overexpression may cause atypical mitosis in Japanese cortisol-producing adrenocortical carcinoma with TP53 somatic variant

“…The Cancer Genome Atlas (TCGA) project for ACC (ACC-TCGA) is the largest opensource database for comprehensive genomic and molecular analysis of ACC [14], and many studies use the ACC-TCGA dataset for genomic and molecular secondary analyses [19,20]. Those large-scale studies include whole genome or exome sequencing data using next-generation sequencing (NGS) technology.…”

Genetic Programs Driving Oncogenic Transformation: Lessons from In Vitro Models

“…In this manner, G0S2 acts as a master regulator of the tissue-specific balance of triglyceride storage vs. mobilization [32]. Correlations with cancer are unclear with anti-proliferative effects described by some reports [33][34][35]. Finally, SRPX (sushi repeat containing protein X-linked), known also as ETX1 or DRS, was initially isolated as deleted in patients with X-linked retinitis pigmentosa [36], as well as downregulated by v-src [37].…”

“…This behavior could depend on G0S2. In fact in ACC, which is a rare, aggressive malignancy, G0S2 hypermethylation is a hallmark of rapidly recurrent or fatal disease, amenable to targeted assessment using routine molecular diagnostics [35]. Very low levels of G0S2 mRNA expression characterize tumors with G0S2 hypermethylation.…”

Investigating the Role of Mineralocorticoid Receptor Signaling in Cancer Biology in the Genomic Era

“…H295R cells have also been shown to express components of aldosterone signaling pathway including ENaC subunits, NEDD4L, SGK1, MR, and HSD11B2 [66,67]. Although several factors, such as age, resection margin and proliferation scores, uterine steroid profiles, CpG island hypermethylation status, as well as levels of selected biomarkers, have been tested for their contribution in ACC prognosis, the importance of MR expression status is yet to be evaluated in ACC patients [65,68,69]. Future studies should investigate the role of activation/inactivation of MR-aldosterone signaling in diagnosis/prognosis of ACC for which significant amounts of genomic and transcriptomic data have recently become available (please see the next section for details).…”