The following is a summary of “Role of tumor density in predicting significant cancer on targeted biopsy of the prostate,” published in the July 2023 issue of Urologic Oncology by Erlich et al.
Multiparametric magnetic resonance imaging (mpMRI) is central to diagnosing prostate cancer; however, not all imaged lesions represent clinically significant tumors. This study aimed to determine the relationship between relative tumor volume on mpMRI and clinically significant prostate cancer on biopsy. Between 2017 and 2021, the researchers examined the medical records of 340 patients who underwent combined transperineal targeted and systematic prostate biopsies. The tumor volume was estimated based on the mpMRI lesion diameter. The relative tumor volume (tumor density) was calculated by dividing the tumor’s importance by the prostate’s magnitude.
The result of the study was clinically significant malignancy detected by biopsy. The relationship between tumor density and the outcome was evaluated using logistic regression. ROC curves were used to determine the tumor density threshold. The estimated median volumes of prostate and peripheral zone tumors were 55cm3 and 0.61cm3, respectively. The median PSA density was 0.13, while the peripheral tumor density was 0.01. 231 patients (68%) had cancer, and 130 (38%) had clinically significant cancer.
Age, PSA, previous biopsy, maximal PI-RADS score, prostate volume, and peripheral zone tumor density were significant predictors of outcome based on multivariable logistic regression. Using a threshold of 0.006, the sensitivity, specificity, positive predictive value, and negative predictive value of peripheral zone tumor density were 0.90, 0.51, 0.57, and 0.88. PI-RADS 4 and 5 mpMRI lesions are associated with clinically significant prostate cancer when peripheral zone tumor density is high. Further research is required to validate their findings and evaluate the role of tumor density in averting unnecessary biopsies.
Source: sciencedirect.com/science/article/abs/pii/S1078143923000960
