Model combining mpMRI and clinical data may reduce unnecessary systematic prostate biopsies

Author(s):

Validation testing suggested the combined model would prevent unnecessary systematic prostate biopsies in over 40% of study patients.

A novel predictive model based on cognitive MRI-targeted biopsy (MRTB) and clinical factors showed the potential to prevent some patients with suspected prostate cancer from receiving unnecessary systematic biopsies, according to a retrospective study published in Insights into Imaging.¹

“The predictive model might be useful in making a decision about which patient(s) could safely avoid unnecessary (systematic biopsy) in addition to MRTB in biopsy-naïve patients and may offer the best compromise between the risk of missing csPCa, biopsy complications, and medical burden,” the authors wrote.

For their analysis, the study investigators identified independent predictive factors for clinically significant prostate cancer (csPCa) with cognitive MRTB, and subsequently compared clinical and MRI models with a combination model of multiparametic MRI (mpMRI) and clinical factors.

(Editor’s note: This article has been adapted from its original publication on our sister site Diagnostic Imaging.)

The cohort was comprised of 201 patients in the model development cohort and 73 patients in the validation cohort, all of whom had combined cognitive MRTB and systematic biopsy (SB), according to the study.

The researchers found that prostate-specific antigen density (PSAD), index lesion (IL) on the peripheral zone, age and PI-RADS 4 and 5 scores were independent predictive factors for detecting csPCa via cognitive MRTB.

The combined model, incorporating mpMRI findings and clinical parameters, demonstrated an 18% higher AUC (88%) than the clinical model (70%) in validation testing and was slightly higher than the AUC for the MRI-only model (86%). The study authors also found that the combined model offered more net benefit (predicting csPCa while reducing additional systematic biopsy) than the clinical and MRI-only models at risk thresholds between 10 to 80 % with the optimal risk threshold being between 50% to 65%

Specifically, at a risk threshold of 60%, validation testing revealed the combined model would prevent unnecessary systematic biopsy in 32 of 73 patients (43.8%) and miss one csPCa diagnosis in 34 patients (2.9%), according to the study authors.

“The combined model, (incorporating) both clinical and mpMRI characteristics, achieved greater net benefit and would help clinical decision-making comprehensively,” wrote study co-author Bin Song, MD, who is affiliated with the Department of Radiology at the West China Hospital of Sichuan University in China, and colleagues. “With a risk threshold set at 60%, 43.8% of patients with very high risk (greater than 60%) of detecting csPCa on MRTB would be advised to avoid (systematic biopsy) both in the development and validation cohorts. Accordingly, a total of 1440 systematic cores would be saved at the expense of missing only two cases of csPCa.”

The study authors noted that previous literature reports have recommended avoiding additional systematic biopsy beyond mpMRI in patients with PI-RADS 5 due to marginal increases in diagnosing csPCa.

In validation testing for patients with PI-RADS 5 lesions, the researchers found that performing only MRTB would prevent systematic biopsies in 24.7% of patients and miss one case of csPCa out of 34 patients (2.9%).

Overall, Song and colleagues said the combination model approach, emphasizing cognitive MRTB and relevant clinical factors, offers a more balanced biopsy approach to managing benefits and risks in patients with suspected csPCa.

Reference

1. Cheng X, Chen Y, Xu J, Cai D, Liu Z, Zeng H, Yao J, Song B. Development and validation if a a predictive model based on clinical and MpMRI findings to reduce additional systematic prostate biopsy. Insights Imaging. 2024;15(1):3. doi: 10.1186/s13244-023-01544-0