Integrated pan-cancer analysis revealed therapeutic targets in the ABC transporter protein family

Abstract

Next-generation sequencing technology enables uniform and impartial assessment of cancer diagnoses and prognosis. However, such studies are mostly type-specific, and capturing shared genomic abnormalities responsible for neoplastic transformation and progression is a challenging task. Pan-cancer analysis offers insights into the shared and unique molecular mechanisms driving cancer. We conducted an integrated gene-expression analysis using 10,629 samples from 30 distinct cancer types characterized by The Cancer Genome Atlas (TCGA). A gene co-expression network was constructed and genes overlapping between the selected modules and Differentially Expressed Genes (DEGs) were designated as genes of interest. Following a comprehensive literature review, ATP binding cassette subfamily A member 10 (ABCA10) and ATP binding cassette subfamily B member 5 (ABCB5) were selected as key candidates for downstream analysis due to the absence of systematic pan-cancer analysis of these genes. This study presents a unique contribution as the first comprehensive pan-cancer analysis of ABCA10 and ABCB5, highlighting their roles in tumor biology and clinical outcomes. We employed a variety of bioinformatics tools to explore the role of these genes across different tumors. Our research demonstrated that ABCA10 shows reduced expression, while ABCB5 displays variable expression patterns across tumors, indicating their opposing roles and flexible functions in pan-cancer. In many cancer patients, these expression patterns are correlated with worse survival outcomes. Furthermore, immunotherapy responses and immune infiltration across a variety of tumor types are associated with the expression levels of both ABCA10 and ABCB5. These results imply that ABCA10 and ABCB5 could serve as valuable predictive markers and potential therapeutic targets across various cancers.

Fig 1. Methodology and datasets used in this study.

(A) The flowchart of methodology. (B) Cancer types and the RNASeq sample count from TCGA used in presented work.

Fig 2. AAIC and DEGs visualization.

(A) Array-array intensity correlation (AAIC). (B) Volcano plot of DEGs.

Fig 3. Gene set enrichment analysis of DEGs.

(A) Displays the pathways enriched in upregulated DEGs. (B) Illustrates the pathways enriched in downregulated DEGs. Normalized enrichment score (NES), false discovery rate (FDR).

Fig 4. WGCNA and identification of genes of interest.

(A) Evaluation of network topology performed across a range of soft-threshold powers. (B) Module–trait correlation heatmap (* P<0.05; ** P<0.01; *** P<0.001). (C) Gene clustering dendrograms (D) Venn diagram of the DEGs and modules.

Fig 5. ABCA10 and ABCB5 expression levels across various cancers and their stages.

Fig 6. Comparison of low and high expression levels of ABCA10 and ABCB5 across various tumors using survival analysis.

Fig 7. Enrichment of genes correlated with ABCA10 and ABCB5.

Fig 8. Correlation of ABCA10 and ABCB5 expression with the extent of immune cell infiltration.

Fig 9. The relationship of ABCA10 and ABCB5 gene expression with drug sensitivity.