Endometrial cancer at risk of relapse

María J. Macias

Institute for Research in Biomedicine (IRB Barcelona), Spain

Every year 400,000 new cases of endometrial cancer are diagnosed worldwide. However, with current molecular staging models, almost two-thirds of patients do not receive a clear prognosis. Moreover, after treatment, a percentage of women suffer relapses and develop metastases.

Over the past three years, the team leading this project has conducted an innovative study combining transcriptomics, clinical data and machine learning algorithms to find biomarkers to identify those women most likely to suffer a relapse and develop the most aggressive form of the disease. Based on the study of molecular and clinical data from 200 patients at the Hospital de la Santa Creu i Sant Pau in Barcelona, characteristics have been defined that enable very accurate prediction of the cases in which the disease will relapse and allow for the diagnosis of tumours that current standard protocols are unable to classify.

In this project, the aim is to validate those results by expanding the number of samples used to train the first artificial intelligence algorithm, also incorporating patient data from other hospitals. Once the model has been validated, its clinical application will be explored as a tool to design new protocols by better defining the groups at risk of relapse, adjusting adjuvant treatments and avoiding overtreatment. In addition, this model would allow optimising the monitoring and treatment of high-risk patients, as well as identifying new therapeutic targets.

Validation of a new methodology and biomarkers for predicting relapse in endometrial cancer patients

Stage 1