Multiparametric model for prostate cancer precision diagnostics

Projektin yksityiskohdat


For the purpose of clinical assessment, PC could be divided into three categories: 1) clinically meaningless (no need to treat), 2) curable and 3) aggressive and/or metastatic at diagnosis, which are currently incurable. The challenge is, however, in predicting the disease course for an individual patient. Still, as only biopsy-based guidelines for clinicians to stratify patients into active surveillance (AS) or radical treatment exist, no guidelines yet rely on predictive markers. Contemporary multiparametric MRI (mpMRI) combines morphological and functional imaging and is able to detect significant PC in biopsynaïve males and in men with prior negative biopsies. Currently, no proper validation of MpMRI with histological findings exists. The big data wave arriving from novel omics-based technologies will eventually revolutionize and individualize cancer care. Clinicians treating cancer patients need new tools for translating this data into knowledge and actionable treatments. We and the collaborators have developed new models for outcome prediction for men with advanced disease and this work is going-on now with broader spectrum real-life patients-derived data. Firstly, this research will address the diagnostic accuracy of mpMRI based on the PI-RADS system with definitive pathology for whole-mount sections of RP specimens as the golden standard. This will establish a possibility to verify the accuracy, specificity and sensitivity of the mpMRI and to compare the PI-RADS scoring system in comparison to the Gleason score. Secondly, the abovementioned research is adapted for implementation in clinical biopsy practice in which mpMRI-TRUS fusion biopsies and targeted biopsies will be correlated with the mpMRI-findings. Thirdly, the research plan includes tissue marker validation part to determine predictive tissue markers for aggressive behaviour in PC. In the near future, in collaborative project, we will incorporate laboratory test values into the prediction model together with imaging and biomarker derived data.
Todellinen alku/loppupvm01/01/2016 → …