Construction of an immune-related gene prognostic model for obese endometrial cancer patients based on bioinformatics analysis.

BACKGROUND

The tumor microenvironment (TME) affected the prognosis of tumors. However, its effect on the outcomes of obese endometrial cancer (EC) patients had not been reported.

METHODS

This research performed a retrospective analysis of the transcriptome profiles and medical data of 503 EC patients. Immune scores were assessed by estimation algorithms. Cox and LASSO regression analyses were utilized to pinpoint key genes linked to prognosis, and the RPS was created to forecast the outcomes of obese EC patients. The relationship among genetic mutations and RPS was examined using CNV and somatic mutation information. ssGSEA and GSVA were employed to detect immune infiltration and immune pathway enrichment associated with key genes. The TIDE algorithm and GDSC database were utilized to forecast patients' responses of patients to immunotherapy and chemotherapy, respectively. Finally, we employed the 'rms' R software package to construct the nomogram.

RESULTS

The prognosis of obese EC patients was associated with immune scores. Three key genes (EYA4, MBOAT2 and SCGB2A1) were identified. The risk prognosis score (RPS) for obese EC patients was established by risk stratification and prognostic prediction using prognostic genes. The higher the RPS, the worse the prognosis, and the more malignant the genomic alterations. The high RPS group had a significantly reduced proportion of most immune cells in comparison to the low RPS group. The high RPS group was linked to G2M, MYC and E2F related pathways such as cell proliferation, cell cycle and cell death. Cisplatin, tamoxifen and topotecan had a greater effect on the low RPS group. Notably, the nomogram had a good predictive ability.

CONCLUSION

Our study designed a reliable RPS for obese EC patients to forecast their prognosis, immune aggressiveness, and responses to immunotherapy and drug treatments.