Quantification of Tumor in the Diagnostic Biopsy and Correlation with Prostatectomy Report
Luciana Schultz, Carlos E Maluf, Barbara F Pizzani, Rogerio C Silva, Rodrigo H Falashi, Matheus V Costa, Maria Ines O Schultz. Instituto de Anatomia Patológica, Piracicaba, SP, Brazil; Hospital dos Fornecedores de Cana, Piracicaba, SP, Brazil
Background: The percentage of tumor reported in the biopsy (PBx) is usually millimetrically estimated while the percentage of tumor in the prostatectomy (PProst) is quantified with graphic precision. Two distinct methods to measure the millimeters of cancer in the biopsy were applied in the same population and compared with the corresponding prostatectomy reports with atention to percentage of tumor and margin status.
Design: Slides from the diagnostic biopsy and corresponding prostatectomy reports (ISUP recommendations) were retrieved from 39 patients diagnosed and treated by the same medical team between 2006-2010. PBx was originaly calculated by Additive Quantification (AQ) which sums the neoplastic foci. Linear Quantification (LQ) was also applied to report the linear extent of cancer in each core. The methods are equivalent when infiltration is continuous but when discontinuous LQ retrieves more millimeters. Percentage index was obtained as PProst/PBx. PBx and PProst were classified arbitrarily in Limited (<10%) and non-Limited (≥10%).
Results: The quantification methods diverged in 14/39 (36%) patients with a mean of 5.8mm more retrieved by LQ in this population, upgrading 13/14 cases as non-Limited. This subset of biopsies was more often associated with positive surgical margins in the corresponding prostatectomy (6/14; p=0.034) independent of tumor extension (PProst of 21% vs. 19%; mean). However, in the subset of biopsies where tumor infiltration was continuous (AQ=AL), positive margins were less frequent (3/24; p=NS) and associated with higher PProst (31% vs. 6%; mean). Percetage index was more accurate in non-Limited tumors (p=0.014) independent of quantification method. Both methods showed good correlation between PBx and PProst but LQ was superior to AQ (0.90 vs. 0.88, p=0.04 and kappa=0.63). By AQ 29/39 (74%) patients were classified as Limited in the biopsy in contrast to only 16/39 (41%) by LQ, matching the prostatectomy reports that also classified 16/39 (41%) patients as Limited.
Conclusions: Linear quantification of cancer improved the performance of PBx in predicting PProst relative to the traditional millimetric sum. The superior performance of this method seems to be particularly interesting in patients with limited volume of cancer. Limitation: This study is based on routine reports and does not directly access tumor biology.
Category: Genitourinary (including renal tumors)
Wednesday, March 6, 2013 9:30 AM
Poster Session V # 124, Wednesday Morning