Assessment of the Effectiveness of the Metabolomic Approach in Screening for Endometrial Cancer

The study is structured into several phases, described below:

• Recruitment and data collection of patients selected for model training at IRCCS Pascale The study will be conducted on patients attending the IRCCS G. Pascale according to normal clinical practice. The patient's participation in the study will begin with the collection of their data and biological samples and will end with the follow-up visit.
• Recruitment and data collection of patients selected for the validation of the trained models at Diagnostica Medica.

This phase of the study will be conducted on patients who will attend the laboratories of the company "Dott.ri Armando and Pierpaolo Cavallo - Diagnostica Medica" according to standard diagnostic and healthcare practices. Specifically, patients who undergo the required checks within the normal diagnostic and healthcare pathway through laboratory medicine, and who meet the enrollment criteria, will be invited to participate in the study.

• Collection of biological samples from enrolled patients In the event of the patient's actual enrollment in the study, blood samples necessary for the study will be collected.
• Analysis of Biological Samples The analysis will consist of both a bioanalytical and a bioinformatic phase. The bioanalytical phase will involve the extraction of the metabolome from DBS (Dried Blood Spots) using a mixture of solvents optimized to extract both polar and apolar metabolites, as well as those with acidic, basic, and neutral properties. This will be followed by a purification phase of the extracted metabolites using liquid-phase partitioning and a post-lyophilization derivatization phase to enhance chromatographic separation. The definition of the metabolomic profile will be carried out using mass spectrometry techniques with a single quadrupole coupled with gas chromatography (GCMS) and high-resolution mass spectrometry techniques (UPLC-Q-Orbitrap).
• Measurement Methods for Results The results of the experiment will be evaluated in terms of sensitivity (defined as the ability to correctly identify affected subjects), specificity (defined as the ability to correctly identify subjects not affected by endometrial carcinoma), positive and negative predictive value (defined respectively as the probability of a subject being correctly classified as affected or not affected), and overall accuracy (defined as the total number of diagnostic errors made by the test). Preliminary evaluations of the diagnostic effectiveness of the metabolomic signature, measured by cross-validation estimates of classification models, showed sensitivity ranging from 93-100%, specificity ranging from 92-98%, and overall accuracy between 95-100%.

The study is expected to last a total of 2 years. All collected data will be integrated and analyzed using mathematical and statistical tools, both descriptive and multivariate.