Background: Although most studies believe that systematic biopsy (SB) and targeted biopsy (TB) should be performed simultaneously in patients with suspected prostate cancer, we believe that patients with the Prostate Imaging-Reporting and Data System (PI-RADS) score of 4/5 may be able to perform TB only.
Methods: We retrospectively analyzed the pathological results of patients undergoing transperineal prostate biopsy with PI-RADS 4 and 5 in our center. We use the data from 2019 to 2020 as the training set to establish the prediction model and the data from 2021 as the verification set to test the effectiveness. Through stepwise logistics regression analysis, we integrate statistically significant clinical factors and establish a model to further predict whether the target area is tumor.
Results: The results showed that age (O), total number of lesions (T), histological region (R), PI-RADS score (S), and PSA density (P) were significantly correlated with the results of TB, and the formula was: p = 1/[1 + e^(- 11.387 + 0.058 × O + (- 0.736 × T) + 0.587 × R + 1.574 × S + 7.338 × P)]. The area under the curve (AUC) of the receiver operating characteristic (ROC) curve of the prediction model was 0.840 (95% CI 0.802-0.877), with the optimal threshold of 0.762. And the corresponding specificity and sensitivity were 0.765 and 0.752. In the validation set, the AUC of the prediction model was 0.816 (95% CI 0.759-0.874), which means that it has good prediction efficiency.
Conclusion: The P.R.O.S.T score can effectively screen PI-RADS 4/5 lesions, which may help physicians shunt patients who need prostate biopsy to reduce unnecessary systematic biopsies.
Keywords: Biopsy; MRI; PI-RADS; Prostate cancer; Scoring system.
© 2022. The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.