Validation of Distinct Bladder Pain Phenotypes Utilizing the MAPP Research Network Cohort

Area of Science:

• Urology
• Pain Medicine
• Computational Biology

Background:

• Interstitial cystitis/bladder pain syndrome (IC/BPS) is a complex condition with likely multiple underlying causes.
• Previous attempts to classify IC/BPS patients into distinct phenotypes have been limited.
• Validating these phenotypes in a larger, multi-center cohort is crucial for understanding disease heterogeneity.

Purpose of the Study:

• To validate three previously proposed clinical phenotypes of IC/BPS patients.
• To utilize unsupervised machine learning (ML) analysis in a large, multi-center cohort.
• To compare ML-derived phenotypes with those from a single-center cohort.

Main Methods:

• Employed k-means unsupervised clustering on data from 130 premenopausal IC/BPS participants in the Multidisciplinary Approach to the Study of Chronic Pelvic Pain (MAPP) network.
• Utilized the female Genitourinary Pain Index and O'Leary-Sant Indices for data collection.
• Compared patient-reported symptoms between ML-derived clusters and previously defined single-center phenotypes.

Main Results:

• Unsupervised ML successfully categorized IC/BPS participants into three distinct phenotypes: myofascial pain, non-urologic pelvic pain, and bladder-specific pain.
• Phenotypes were characterized by specific pain and urinary symptom patterns, with statistically significant defining features.
• The ML model achieved this classification using only 11 features, a substantial reduction from previous methods, while maintaining accuracy.

Conclusions:

• The reproducible identification of IC/BPS phenotypes using independent ML analysis of a multicenter database supports their validity.
• These findings suggest that the identified phenotypes represent true pathophysiologic differences among IC/BPS patients.
• Distinguishing bladder-specific pain from myofascial and genital pain is critical for targeted therapeutic approaches.