[A preliminar mathematical model for patients with a previous negative prostate biopsy]

It is usual to identify patients with a negative prostate biopsy who are still at risk of prostate cancer. We try to analyse if the classical variables used in the prostate cancer screening are useful for those patients with a previous negative prostate biopsy, and if there is a possibility for making a nomogram witch would help us in the decision to repeat the biopsy.

We studied 179 patients with at least 1 initial negative biopsy. At each biopsy session we recorded: Patient age, serum prostate specific antigen (PSA), free PSA/total PSA, PSA slope, digital rectal examination, prostate volume, PSA density, cancer suspicion in previous transrectal ultrasounds findings, number of negative cores previously obtained, history of precarcinomatous lesions and time between biopsies. Through Logistic regression analysis we determined the association of each variable a positive biopsy. A nomogram was constructed using all variables and discrimination was calculated as the concordance index.

Overall 46% of patients had cancer at the repeated biopsy session. In the univariate analysis: Age, digital rectal examination, prostate volume, PSA density, cancer suspicion in ultrasounds findings, and precarcinomatous lesions were associated with repeat positive biopsy for cancer (all p <0.05). In the multivariate study, age, digital rectal examination, prostate volume and history of precarcinomatous lesions were associated with repeat positive biopsy. A nomogram was constructed that had a concordance index of 0.80.