Introduction/Background Molecular biology has transformed EC understanding and risk assessment. Genomic analysis informs surgical treatment, prognosis, and adjuvant therapy. Historically, lymphadenectomy in high-risk groups lacked clear benefits for adjuvant treatment and recurrence prediction. Despite declining use, a subgroup (Non-Specific Molecular Profile) lacks well-defined molecular-based recurrence risk prediction, impacting preoperative planning, recurrence follow-up, and optimal adjuvant therapy. This study aims to develop new genomic indicators predicting EC recurrence risk, guiding staging lymphadenectomy, adjuvant treatment, and monitoring for maximum benefit.

Methodology This study combines clinical, experimental, and computational methods. It includes all EC patients at Hospital de la Santa Creu i Sant Pau (2010–2018), assessed by the Oncology-Gynaecology Committee, and treated according to current Clinical Practice Guidelines. Retrospectively collected clinical data categorized patients into risk subgroups (A to H) based on endometrial biopsy results, lymphadenectomy, and 3-year recurrence. Tumour histological blocks were post-surgery confirmed, and RNA extraction enabled microarray analysis to identify candidate genes predicting EC recurrence through mRNA transcriptomic analysis.

Results Genetic material extracted from 120 samples: 97 initial project samples and 33 additional samples at three-year follow-up. Table 1 shows patient distribution. Transcriptomic data differentiates between groups A (low risk, no recurrence) and B (low risk, recurrence), identifying low-risk patients in group B experiencing recurrence (figures 1 and 2). Comparative analysis of transcriptomic profiles between subgroups A and F (low-risk, no recurrence vs. high-risk with negative lymphadenectomy and recurrence) reveals differences (figures 3 and 4). Notable distinctions occur in high-risk patients with recurrence, regardless of positive (D) or negative (F) lymphadenectomy.

Conclusion A low-risk subgroup may face recurrence, while a high-risk subgroup might not, impacting surgical and adjuvant treatment decisions that may not align with individual risks. Our preliminary study identifies differentially expressed candidate genes in tumours from these subgroups, offering insight into the behaviour of tumours with uncertain aggressiveness.

Disclosures All authors declare no conflicts of interest.