Crowdsourcing Disease Biomarker Discovery Research: The IP4IC Study

Area of Science:

• Biomarkers
• Urology
• Machine Learning

Background:

• Biomarker discovery is challenged by sample accessibility and disease heterogeneity.
• Interstitial cystitis/bladder pain syndrome (IC/BPS) diagnosis can be difficult, especially identifying Hunner lesions.

Purpose of the Study:

• To develop a novel, cost-efficient crowdsourcing method for biomarker discovery.
• To create a Bladder Permeability Defect Risk Score (BP-RS) using noninvasive urinary cytokines to identify IC/BPS patients with Hunner lesions.

Main Methods:

• A national crowdsourcing study collected urine samples from IC/BPS patients and controls.
• Demographic and symptom data were gathered via online questionnaires.
• A machine learning algorithm analyzed 3 urinary cytokines to develop the BP-RS.

Main Results:

• The IP4IC study collected 448 urine samples from 153 IC/BPS patients (54 with Hunner lesions) and 295 controls.
• The developed BP-RS predicted IC/BPS with Hunner lesions (indicating bladder permeability defect) with 89% validity.
• Crowdsourcing enabled collection of a large, diverse sample set from 46 states.

Conclusions:

• BP-RS is the first validated urine biomarker assay for IC/BPS, specifically identifying Hunner lesions.
• Crowdsourcing offers a viable method for large-scale biomarker assay development.
• This approach facilitates the study of disease heterogeneity using accessible samples.