Advancing Risk Stratification in Endometrial Cancer: Integrating Dynamic Circulating Tumor DNA and Proteomics-Based Liquid Biopsies for Enhanced Prognostic Precision

Endometrial cancer (EC) is the most common gynecologic malignancy in high-income countries. Although the 2023 International Federation of Gynecology and Obstetrics (FIGO) staging system has improved prognostic stratification by integrating molecular classification with conventional clinicopathologic features, it remains primarily based on static information obtained at the time of diagnosis and surgery. Consequently, it may not fully capture the biological heterogeneity of EC or the dynamic changes associated with minimal residual disease (MRD) and disease recurrence. Emerging evidence suggests that circulating tumor DNA (ctDNA) and plasma proteomic profiling may provide complementary, non-invasive biomarkers capable of improving risk assessment and identifying patients at increased risk of relapse.

The DaPHNE study is an observational, ambispective, multicenter study designed to investigate whether integrating liquid biopsy-derived molecular information with the FIGO 2023 staging system improves prognostic accuracy in patients with endometrial cancer undergoing curative-intent surgery. The study hypothesizes that a dynamic multimodal approach incorporating serial ctDNA and proteomic assessments before surgery (T0) and 4-6 weeks after surgery (T1) will provide more accurate prediction of disease progression than the current staging system alone.

The study comprises two phases. Phase 1 is a retrospective analysis of approximately 300 patients with histologically confirmed endometrial cancer treated with radical surgery and no residual disease, for whom plasma samples and clinical data are available through the Institutional Biobank. Phase 2 is a prospective validation study enrolling approximately 100 patients from participating institutions using the same eligibility criteria and biospecimen collection procedures.

Peripheral blood samples collected at T0 and T1 will undergo plasma isolation followed by ctDNA extraction. Genomic analyses will include targeted next-generation sequencing using the Guardant360 assay and methylation profiling with the Illumina 5-Base DNA Prep workflow. Plasma proteomic profiling will be performed on postoperative samples using the Olink Reveal platform, with paired T0 and T1 samples analyzed when available to evaluate temporal molecular changes associated with MRD and early recurrence.

Clinical, pathological, treatment, imaging, and follow-up data will be collected in a secure REDCap database. Molecular findings will be integrated with clinicopathologic variables to develop and validate multimodal prognostic models. The study will compare the prognostic performance of the standard FIGO 2023 staging system with an enhanced longitudinal staging model (L-FIGO 2023), which incorporates baseline liquid biopsy information, and will further explore a dynamic staging approach (Δ-FIGO) based on molecular changes between preoperative and postoperative time points.