Distinct molecular etiologies of male and female hepatocellular carcinoma

Abstract

Background: Sex-differences in cancer occurrence and mortality are evident across tumor types; men exhibit higher rates of incidence and often poorer responses to treatment. Targeted approaches to the treatment of tumors that account for these sex-differences require the characterization and understanding of the fundamental biological mechanisms that differentiate them. Hepatocellular Carcinoma (HCC) is the second leading cause of cancer death worldwide, with the incidence rapidly rising. HCC exhibits a male-bias in occurrence and mortality, but previous studies have failed to explore the sex-specific dysregulation of gene expression in HCC.

Methods: Here, we characterize the sex-shared and sex-specific regulatory changes in HCC tumors in the TCGA LIHC cohort using combined and sex-stratified differential expression and eQTL analyses.

Results: By using a sex-specific differential expression analysis of tumor and tumor-adjacent samples, we uncovered etiologically relevant genes and pathways differentiating male and female HCC. While both sexes exhibited activation of pathways related to apoptosis and cell cycle, males and females differed in the activation of several signaling pathways, with females showing PPAR pathway enrichment while males showed PI3K, PI3K/AKT, FGFR, EGFR, NGF, GF1R, Rap1, DAP12, and IL-2 signaling pathway enrichment. Using eQTL analyses, we discovered germline variants with differential effects on tumor gene expression between the sexes. 24.3% of the discovered eQTLs exhibit differential effects between the sexes, illustrating the substantial role of sex in modifying the effects of eQTLs in HCC. The genes that showed sex-specific dysregulation in tumors and those that harbored a sex-specific eQTL converge in clinically relevant pathways, suggesting that the molecular etiologies of male and female HCC are partially driven by differential genetic effects on gene expression.

Conclusions: Sex-stratified analyses detect sex-specific molecular etiologies of HCC. Overall, our results provide new insight into the role of inherited genetic regulation of transcription in modulating sex-differences in HCC etiology and provide a framework for future studies on sex-biased cancers.

Keywords: Gene expression; HCC; Hepatocellular carcinoma; Sex; Sex as a biological variable; eQTL.

Fig. 1

Patterns of gene expression and molecular etiologies of male and female HCC. a Sex-biased gene expression in HCC. A volcano plot of DEGs between male (N = 26) and female (N = 18) HCC tumor samples. X-linked genes are indicated in pink, Y-linked in green, and autosomal in black. Significant genes were selected based on an FDR-adjusted p-value threshold of 0.01 and absolute log₂(FC) threshold of 2. Multiple transcripts of the long non-coding RNA XIST are independently expressed. Genes that were not expressed in a sex-biased way in healthy liver (GTEx) or in the tumor-adjacent tissues are indicated with an asterisk. b An example of a gene exhibiting a sex-bias in HCC but not in healthy liver or tumor-adjacent tissues. DTX1 expression in log(CPM) is shown for male and female samples in each tissue. c A multi-dimensional scaling plot of the paired TCGA LIHC tumor and tumor-adjacent samples of each sex. Euclidean distances between samples were calculated based on 100 genes with the largest standard deviations between samples. Tissue type (dimension 1) and sex (dimension 2) drive the overall patterns of gene expression in HCC. d Venn-diagram of the overlap of DEGs in the sex-specific and combined analyses of matched tumor and tumor-adjacent samples. Substantially more DEGs were identified in the sex-specific analyses. e Sex-specific and sex-shared DEGs were analyzed for the overrepresentation of functional pathways. Sex-specific patterns of pathway enrichment point to differential processes driving the etiology of male and female HCC. f Examples of sex-specific and sex-shared pathways

Fig. 2

Absolute log₂-fold changes of DEGs detected from tumor vs. tumor-adjacent comparisons in the combined analysis of both sexes, male, and female analysis (a), in the combined analysis only (b), in the male analysis only (c), and in the female analysis only (d). Absolute log₂-fold changes are given for female samples, male samples, and across all samples. Global p-values for ANOVA are shown for each DEG type. Adjusted p-values based on Kruskal-Wallis tests are shown for each pairwise comparison

Fig. 3

Sex-specific genetic effects on tumor gene expression in HCC. a QQ-plot of eQTL associations in the combined analysis of both sexes (grey), male-specific analysis (blue), and female-specific analysis (red). b Genomic annotations of eQTLs in the combined analysis of both sexes, male-specific analysis, and female-specific analysis. c Overlap of eGenes detected in combined and sex-specific analyses. d An example of a male-specific eQTL. POGLUT1 expression in tumors is modulated by a germline variant in cis in male HCC, but not in female HCC nor in the combined analysis of both sexes, indicating effect modification by sex. Numbers of individuals with each genotype, adjusted significance, and effect size (β) are given for each model

Fig. 4

Absolute effect sizes of sex-shared and sex-specific eQTLs in males, females, and the whole study sample. Due to the larger sample size, sex-shared low-effect eQTLs are only detected as significant in the combined analysis (a). Sex-shared large effect eQTLs are detected in the combined analysis as well as the sex-specific analyses (b). Sex-specific eQTLs exhibit a larger effect in one sex than the other, and the effect is diluted in the combined analysis (c, d). Sex-shared large effect eQTLs can be detected in sex-specific and combined analyses. Global p-values for ANOVA are shown for each eQTL type. Adjusted p-values based on Kruskal-Wallis tests are shown for each pairwise comparison