Clinical Report: Revised MSKCC Nomogram for Predicting Metastatic Progression
Overview
The updated MSKCC nomogram demonstrates superior predictive accuracy for metastatic progression in prostate cancer compared to the CAPRA and D’Amico models. Implementing nomogram-based thresholds for active surveillance eligibility could significantly increase patient selection without elevating metastatic risk.
Background
Prostate cancer is the most prevalent cancer among men in Europe and a leading cause of cancer mortality. Accurate prognostic models are essential for guiding treatment decisions, particularly in distinguishing candidates for active surveillance from those needing immediate intervention. The ability to predict metastatic disease is critical, as it is a primary driver of prostate cancer mortality.
Data Highlights
Model
AUCt
Brier Score
MSKCC
0.81
0.15
CAPRA
0.77
0.11
D’Amico
0.64
0.03
Key Findings
The MSKCC nomogram outperformed CAPRA and D’Amico models in predicting metastatic prostate cancer (mPCa).
Using the 95th percentile of MSKCC's predicted probabilities increased active surveillance eligibility from 7.8% to 57.0%.
The observed 5-year risk of metastasis was low at 1.7% when using the MSKCC nomogram thresholds.
Overall mortality prediction followed the same ranking as mPCa prediction among the models.
Prognostic models need to account for competing risks to improve accuracy in clinical decision-making.
Clinical Implications
Clinicians should consider utilizing the MSKCC nomogram for better risk stratification in prostate cancer patients, particularly for those being evaluated for active surveillance. This approach may enhance patient selection and facilitate informed shared decision-making between patients and healthcare providers.
Conclusion
The revised MSKCC nomogram provides a more accurate tool for predicting metastatic progression in prostate cancer, potentially transforming active surveillance practices and improving patient outcomes.
by Nicolas Destefanis, Daniela Zugna, Valentina Fiano, Renata Zelic, Michelangelo Fiorentino, Francesca Giunchi, Piero Fariselli, Mauro Giulio Papotti, Paola Cassoni, Marco Oderda, Paolo Gontero, Luca Lianas, Mauro Del Rio, Giuseppe Carlo Iorio, Umberto Ricardi, Olof Akre, Andreas Pettersson, Lorenzo Richiardi
