Histological evaluation of prostatic cancer (III): Reproducibility of assessment of tumour volume and its possible significance for prognosis

Abstract

The aim of this study was to evaluate the INTER- and INTRA-observer reproducibility of prostatic cancer volume defined as the percentage of chips with carcinoma in transurethrally resected tissue. All histological slides from 80 consecutive cases of prostatic cancer were blindly evaluated on two independent occasions by each of four pathologists. The correlation was expressed by Spearman's rank correlation coefficient (R). Both the overall INTER- and mean INTRA-observer correlation of tumour volume were high with R being respectively 0.905 and 0.918. With reference to stage A1 and A2, data were transformed to an ordinal scale with 2 categories (less than = 5%, greater than 5%). The overall INTER- and mean INTRA-observer agreement of this classification were 0.926 and 0.943 respectively--the corresponding kappa coefficients being 0.68 and 0.74. It is concluded that in transurethrally resected tissue a highly reproducible index of the resected prostatic cancer volume can be given by the percentage of chips with tumour. The prognostic significance of this parameter is discussed.