Microscopic assessment of fresh prostate tumour specimens yields significantly increased rates of correctly annotated samples for downstream analysis

Aims: To assess if performing frozen sections of tissue biopsies from fresh radical prostatectomy specimens, prior to tissue banking, could improve the identification of the banked samples compared to standard fresh tumour banking procedures. Methods: Tissue biopsies banked from 332 fresh prostatectomy specimens were assessed for accuracy of diagnosis, comparing two separate methods of tumour identification: one in which tumour was identified in the gross specimen by visual inspection (n = 155) and one in which rapid frozen sectioning was applied (n = 177). The associations with correct tumour annotation and clinicopathological variables, including age, pre-operative prostate specific antigen (PSA) levels, pathological Gleason score, pathological T stage, tumour volume and surgical margins, were examined using univariable and multivariable binary logistic regression models. Results: For the gross visual inspection cohort the rate of correctly identifying and banking specimens containing prostate cancer was 69%. For the cohort assessed with rapid frozen sections, 94% of banked specimens actually had cancer. On multivariable analysis, we found that only frozen sectioning and tumour volume variables were independent predictors of correctly banked tumour specimens whilst all other routinely reported pathological variables had no influence on the success rates of fresh prostate tumour banking. Conclusion: The success rate for correctly banking fresh prostate tumour specimens is directly related to the tumour volume. Frozen section scrutiny of prostate samples is recommended to prevent misclassification of the banked material.