Identifying New Potential Biomarkers in Adrenocortical Tumors Based on mRNA Expression Data Using Machine Learning

Abstract

Adrenocortical carcinoma (ACC) is a rare disease, associated with poor survival. Several "multiple-omics" studies characterizing ACC on a molecular level identified two different clusters correlating with patient survival (C1A and C1B). We here used the publicly available transcriptome data from the TCGA-ACC dataset (n = 79), applying machine learning (ML) methods to classify the ACC based on expression pattern in an unbiased manner. UMAP (uniform manifold approximation and projection)-based clustering resulted in two distinct groups, ACC-UMAP1 and ACC-UMAP2, that largely overlap with clusters C1B and C1A, respectively. However, subsequent use of random-forest-based learning revealed a set of new possible marker genes showing significant differential expression in the described clusters (e.g., SOAT1, EIF2A1). For validation purposes, we used a secondary dataset based on a previous study from our group, consisting of 4 normal adrenal glands and 52 benign and 7 malignant tumor samples. The results largely confirmed those obtained for the TCGA-ACC cohort. In addition, the ENSAT dataset showed a correlation between benign adrenocortical tumors and the good prognosis ACC cluster ACC-UMAP1/C1B. In conclusion, the use of ML approaches re-identified and redefined known prognostic ACC subgroups. On the other hand, the subsequent use of random-forest-based learning identified new possible prognostic marker genes for ACC.

Keywords: adrenocortical carcinoma; bioinformatic clustering; biomarker prediction; in silico analysis; machine learning.

Figure 1

UMAP cluster representation of different mRNA expression patterns. Representation of the UMAP (uniform manifold approximation and projection) clustering of the ACC-TCGA dataset without outliers (A) and the overlap with C1A/C1B clustering from the original publication of Zheng et al. [15] (B). Kaplan–Meier curve of overall survival of ACC-TCGA patients assigned to the two clusters by UMAP. Shaded area: confidence interval with alpha = 0.05 (C). Representation of the UMAP clustering of the ENSAT dataset [19] (D). NAG = normal adrenal gland; EIA = endocrine inactive adenoma; MACS-CPA = mild autonomous cortisol secretion adenoma-cortisol producing adenoma; CS-CPA = Cushing syndrome-cortisol producing adenoma which together make the ACA = adrenocortical adenoma; ACC = adrenocortical carcinoma.

Figure 2

Selection of mRNA expression pattern of 10 genes from the TCGA-ACC dataset, as identified by RF analysis, that were previously shown to be involved in adrenal function. SLC2A1: solute carrier family 2 member 1 (A), SOAT1: sterol-O acyltransferase (B), EIF2S1: eukaryotic translation initiation factor 2 α (C), MYC proto-oncogene MYC (D), SMAD2: sma—and mad-related protein 2 (E), BUB3 budding uninhibited by benzimidazoles 3 homolog (F), ASB4: ankyrin repeat And SOCS box containing 4 (G), MED27: mediator complex subunit 27 (H), FSCN1: fascin actin-bundling protein 1 (I), GNAI3: guanine nucleotide binding protein (G protein), alpha inhibiting activity polypeptide 3 (J). ns, not significant. **** p < 0.0001. Y-axis units: FPKM.

Figure 3

Selection of mRNA expression pattern of 10 genes from the validation dataset, as identified by RF analysis, that were previously shown to be involved in adrenal function. SLC2A1: solute carrier family 2 member 1 (A), SOAT1: sterol-O acyltransferase (B), EIF2S1: eukaryotic translation initiation factor 2 α (C), MYC: proto-oncogene MYC (D), SMAD2: sma—and mad-related protein 2 (E), BUB3: budding uninhibited by benzimidazoles 3 homolog (F), ASB4: ankyrin repeat and SOCS box containing 4 (G), MED27: mediator complex subunit 27 (H), FSCN1: fascin actin-bundling protein 1 (I), GNAI3: guanine nucleotide binding protein (G protein), alpha inhibiting activity polypeptide 3 (J). NAG = normal adrenal gland; ACA = adrenocortical adenoma; ACC = adrenocortical carcinoma. ns, not significant. * p < 0.05, ** p < 0.01, ns, not significant. Y-axis units: FPKM.