Site Editor

Sandy Srinivas, MD

Advertisement
Advertisement

Prostate Cancer Risk Prediction Tools and How to Accommodate for Missing Risk Factor Data

By: Anna Fanelli
Posted: Friday, September 16, 2022

A study conducted by Matthias Neumair, PhD, of the Technical University of Munich, and colleagues examined the accommodation of heterogeneous missing data patterns for prostate cancer risk prediction by comparing six commonly used logistic regression methods from multiple heterogeneous cohorts. Some of these cohorts do not collect all risk factors, and the team therefore developed an online risk prediction tool to accommodate the missing risk factors from the end-user. The study findings were reported in BMC Medical Research Methodology.

β€œIn addition to contributing to model development techniques for systematic missing data across heterogeneous cohorts, we have provided helpful methods for the end-user of online risk tools, namely the fit of multiple models for different risk factor missing data patterns,” concluded the study authors.

In the study, 10 North American and European cohorts from the Prostate Biopsy Collaborative Group (PBCG) were used for fitting a risk prediction tool for clinically significant prostate cancer (defined as Gleason grade group β‰₯ 2 on standard transrectal ultrasound scan prostate biopsy). One large European PBCG cohort was withheld for external validation, in which calibration-in-the-large (CIL), calibration curves, and area-underneath-the-receiver-operating characteristic curve (AUC) were evaluated. Tenfold leave-one-cohort-internal validation further validated the optimal missing data approach.

Among 12,703 biopsies from 10 training cohorts, 3,597 (28%) had clinically significant prostate cancer, compared with 1,757 of 5,540 (32%) in the external validation cohort. In external validation, the available-cases method that pooled individual patient data containing all risk factors input by an end-user had the best CIL. Also, in external validation, the method underpredicted risks as percentages by 2.9% on average and obtained an AUC of 75.7%. Imputation had the worst CIL (–13.3%).

For end-users of the risk tool, two risk factors were mandatory: serum prostate-specific antigen (PSA) level and age, and 10 were optional: digital rectal exam; prostate volume; prior negative biopsy; 5-alpha-reductase inhibitor use; prior PSA screen; African ancestry; Hispanic ethnicity; and first-degree prostate-, breast-, and second-degree prostate cancer family histories.

Disclosure: One of the authors has received royalties from the 4K score. The other study authors reported no conflicts of interest.


By continuing to browse this site you permit us and our partners to place identification cookies on your browser and agree to our use of cookies to identify you for marketing. Read our Privacy Policy to learn more.