JoVE - Journal of Visualized Experiments
JoVE Visualize
Contact Us

Novel clinical risk calculator for improving cancer predictability of mpMRI fusion biopsy in prostates

  • 0School of Medicine, Department of Urology, Texas Tech University Health Sciences Center, 3601-4 Street STOP 7260, Lubbock, TX, 79430-7260, USA.
International Urology and Nephrology +

|

April 5, 2024

|

April 5, 2024

View abstract on PubMed

Summary

This summary is machine-generated.

New predictive models improve prostate cancer detection for PI-RADS 3 and 4 lesions. This approach helps stratify patients, potentially reducing unnecessary biopsies and associated complications.

Area Of Science

  • Radiology
  • Oncology
  • Medical Imaging

Background

  • Prostate Imaging-Reporting and Data System (PI-RADS) is used with multiparametric MRI (mpMRI) to evaluate prostate lesions.
  • Improving the prediction of clinically significant prostate cancer (csPCa) for PI-RADS grades 3-5 remains an area of active research.
  • Current PI-RADS classification shows limitations in accurately predicting csPCa, particularly for intermediate scores.

Purpose Of The Study

  • To develop and validate an easily implementable method to enhance the predictability of csPCa using PI-RADS scores.
  • To improve the stratification of patients with PI-RADS 3 and 4 lesions.
  • To reduce the rate of unnecessary prostate biopsies.

Main Methods

  • A cohort of 151 patients with PI-RADS 3-5 lesions on mpMRI underwent Fusion and random US Biopsy.
  • Data were collected from January 2019 to December 2022.
  • Two predictive models for csPCa were developed using logistic regression with readily available clinical parameters.

Main Results

  • The study included patients with PI-RADS 3 (n=49), PI-RADS 4 (n=82), and PI-RADS 5 (n=20).
  • csPCa was confirmed in 12/49 (PI-RADS 3), 42/82 (PI-RADS 4), and 18/20 (PI-RADS 5) patients.
  • The developed predictive models achieved an Area Under the Curve (AUC) of 0.8133 and 0.8206, indicating good predictive performance.

Conclusions

  • PI-RADS classification demonstrates relevant predictability issues for grades 3 and 4.
  • The proposed risk calculators offer improved stratification for patients with PI-RADS 3 and 4.
  • These models can serve as valuable diagnostic tools, aiding clinical decision-making and potentially sparing patients from complications associated with unnecessary biopsies.

Keywords: