ساخت اکانت کاربری رایگان

مقالات پذیرفته شده کنگره

  • صفحه نخست
  • مقالات پذیرفته شده اولین کنگره بین المللی ژنتیک سرطان
بازنشر

  • Cross-Cohort Transcriptomic Meta-Analysis Identifies GAB1 as a Reproducible Signal of Platinum Resistance in High-Grade Serous Ovarian Cancer

  • Mohammadreza Mahdipour,1,* Rahil Nasari Fard,2
    1. Department of Biotechnology, College of Science, University of Tehran, Tehran, Iran
    2. University of Tehran Dormitory Complex, Opposite 15th Street, North Kargar Street, Tehran, Iran


  • Introduction: High-grade serous ovarian cancer (HGSOC) is commonly treated with platinum-based chemotherapy, and most patients initially respond to treatment. However, the development of platinum resistance remains a major clinical challenge and is closely associated with recurrent disease. Gene-expression studies have reported numerous genes associated with platinum response, but many of these findings have not been reproduced across independent patient cohorts. We asked whether integrating several HGSOC datasets at the level of gene-specific effect sizes could identify transcriptomic changes that are more consistent across studies and experimental platforms.
  • Methods: We searched the Gene Expression Omnibus for studies of human HGSOC tumors with bulk transcriptomic data and sufficient clinical information to classify samples as platinum-resistant or platinum-sensitive. Four independent cohorts met the inclusion criteria: GSE51373, GSE131978, GSE114206, and GSE273637. Together, they included 91 tumor samples, comprising 47 platinum-resistant and 44 platinum-sensitive cases. The studies were generated using different technologies, including Affymetrix and Agilent microarrays and bulk RNA sequencing. To avoid introducing cross-platform batch effects, expression matrices were not combined directly. Instead, each cohort was processed separately, probes or transcripts were mapped to gene symbols, and expression values were summarized at the gene level. For GSE131978, samples generated on two Affymetrix platforms were treated as platform-specific strata within the same study. For each gene, the difference between resistant and sensitive tumors was estimated within each cohort using Hedges’ g. These cohort-level effect sizes were then combined using a random-effects meta-analysis. Between-study heterogeneity was assessed using I² and tau², and P values were adjusted across genes using the Benjamini-Hochberg false discovery rate (FDR). To examine whether the main findings depended on a particular dataset, the meta-analysis was repeated after removing each cohort in turn. Hallmark pathway enrichment was explored as a secondary analysis.
  • Results: After gene-level harmonization, 15,501 genes could be evaluated across all four cohorts. GAB1 was the only gene that met the prespecified genome-wide FDR threshold. GAB1 expression was higher in platinum-resistant tumors in each of the four cohorts. The cohort-specific Hedges’ g values were 1.44 for GSE51373, 0.72 for GSE131978, 1.52 for GSE114206, and 1.04 for GSE273637. The pooled random-effects estimate was 1.10 (95% CI, 0.64–1.56; P=2.8×10⁻⁶; FDR=0.043), with no observed between-study heterogeneity (I²=0%). The direction of the GAB1 association was also stable in the leave-one-cohort-out analyses. When each dataset was removed in turn, the pooled effect remained positive, ranging from 0.96 to 1.31. However, none of these reduced analyses retained genome-wide significance after FDR correction, indicating that the statistical evidence was strongest when all four cohorts were considered together. Exploratory Hallmark analysis suggested that genes expressed at higher levels in resistant tumors were enriched for epithelial-mesenchymal transition, angiogenesis, and hypoxia-related programs. In contrast, E2F-target and G2M-checkpoint programs were relatively reduced in the resistant group.
  • Conclusion: Across four independent patient cohorts, GAB1 showed the clearest and most consistent gene-expression difference between platinum-resistant and platinum-sensitive HGSOC tumors. The association was observed in both microarray and RNA-seq datasets and remained positive when individual cohorts were removed from the analysis. However, its genome-wide FDR significance was not preserved in the leave-one-cohort-out analyses. These findings identify GAB1 as a candidate associated with the platinum-resistant phenotype rather than an established predictive biomarker. Additional clinical cohorts and functional studies will be needed to determine whether GAB1 contributes directly to platinum resistance and whether it has potential clinical relevance.
  • Keywords: High-grade serous ovarian cancer (HGSOC); Platinum resistance; Transcriptomics; Gene expression; Met

به خانواده بزرگ کنسر ژنتیکس و ژنومیکس سرطان بپیوندید!

ایجاد حساب کاربری رایگان