Hiroshi Yoshida

Aim: High‐grade serous ovarian cancer is a deadly cancer with a poor prognosis. Serum microRNAs have been used to detect and diagnose various cancers. We aimed to investigate the prognostic impact of serum microRNA profiles pre‐, peri‐, and post‐primary treatment on the progression‐free survival of patients with advanced high‐grade serous ovarian cancer receiving neoadjuvant chemotherapy. Methods: Patients were prospectively recruited, and pre‐, peri‐, and post‐treatment blood samples were collected. MicroRNA microarray results, together with the clinical and surgical data, were analyzed using Cox regression with leave‐one‐out least absolute shrinkage and selection operator for progression‐free survival. Results: Overall, 29 patients were included. A combined analysis of pre‐treatment serum microRNAs and clinical factors showed a higher c‐index (0.86) for distinguishing high‐ and low‐risk groups for poor progression‐free survival outcomes compared to analyzing each factor alone (0.70 and 0.64, respectively). The high‐risk group exhibited significantly worse progression‐free survival (log‐rank, p < 0.01). The serum microRNA profile at 3 months post‐treatment showed outstanding discrimination (c‐index of 1.00). Changes in the microRNA profile between pre‐ and peri‐treatment had no prognostic impact; however, the difference between pre‐treatment and 3 months post‐treatment exhibited excellent discrimination (c‐index of 1.00). Conclusions: In patients with advanced high‐grade serous ovarian cancer who received neoadjuvant chemotherapy, a combination of the pre‐treatment serum microRNA profiles and clinical factors, or the 3‐month post‐treatment serum microRNA profile could distinguish between those in high‐ and low‐risk groups for poor progression‐free survival outcomes.

Supplementary Table S9. Gene set enrichment analysis statistics for representative pathways.

Supplementary Table S2. Significance thresholds used for GSEA and ssGSEA analyses.

Supplementary Figure S7. Gene expression profiles across different RNA clusters in The Cancer Genome Atlas Program dataset.

Supplementary Table S1. List of samples analyzed in this study.

Supplementary Table S5. Detailed evaluation of discordance among TP53 mutation status, p53 immunohistochemistry, and CN-based molecular classification in non-POLE/non-MSI cases.

Supplementary Figure S10. Prognostic impact of transcriptomic clusters in TP53-mutated, high-grade endometrial carcinoma.

Exome sequencing revealed genomic landscape across high-grade endometrial cancers. A, Bar plots showing the number of mutations and signature proportions. The top shows the number of mutations per megabase. The middle shows the relative contribution of different mutational signatures. The bottom two show the length of genomic regions with CN loss or gain and the relative contribution of different CN signatures. B, Summary of genetic and clinical features for each sample. Each column corresponds to a sample. Each row corresponds to a different genetic or clinical feature or gene mutation status. C, Bar plots showing the frequency of specific gene mutations by subtype. The left shows the frequency of mutations in various genes by histologic classification. The right shows mutation frequency by molecular subtype. APOBEC, apolipoprotein B mRNA-editing enzyme, catalytic polypeptide; clear-S, clear cell carcinoma mixed with serous carcinoma; CS, carcinosarcoma; EM, endometrioid; G1/G2, grades 1–2 endometrioid; MSS, microsatellite stable; VUS, variant of unknown significance.