A multivariable prediction model combining 18F-PSMA PET/CT and mpMRI for clinically significant prostate cancer: development and validation - Report - MDSpire
Advertisement
A multivariable prediction model combining 18F-PSMA PET/CT and mpMRI for clinically significant prostate cancer: development and validation
Clinical Report: Development and Validation of a Multivariable Model Integrating 18F-PSMA PET/CT and mpMRI
Overview
This study developed and validated a multivariable model that integrates clinical parameters, mpMRI, and 18F-PSMA PET/CT to predict clinically significant prostate cancer (csPCa). The model demonstrated high accuracy and clinical utility, potentially optimizing biopsy decisions.
Background
Prostate cancer is the most commonly diagnosed malignancy in men, necessitating effective diagnostic strategies to distinguish clinically significant disease from indolent forms. Traditional PSA screening has limitations, leading to overdiagnosis and overtreatment. The integration of imaging modalities like mpMRI and PSMA PET/CT may enhance risk stratification and improve clinical decision-making.
Data Highlights
Parameter
Training AUC
Internal Test AUC
Temporal Validation AUC
Model
0.916
0.914 (95% CI: 0.882–0.941)
0.837 (95% CI: 0.778–0.891)
Key Findings
The final model included PRIMARY score, PI-RADS score, and PSAD as predictors.
At the Youden Index cutoff (≥84%), the model achieved a sensitivity of 79.3% and specificity of 76.6%.
At the recommended screening cutoff (≥46%), sensitivity increased to 96.0%.
The model showed good calibration with a Brier score of 0.096.
It provided superior clinical utility compared to individual imaging parameters.
Clinical Implications
The multivariable model can assist clinicians in making informed decisions regarding biopsy in patients with suspected prostate cancer. By accurately predicting csPCa, it may reduce unnecessary interventions for low-risk disease and improve patient outcomes.
Conclusion
The integration of 18F-PSMA PET/CT and mpMRI into a multivariable model offers a promising approach for risk stratification in prostate cancer, potentially enhancing clinical decision-making and patient management.
