[904] Variation among Urogenital Pathologists in Diagnostic Opinion of Single Small Atypical Foci in Prostate Biopsies

TH Van der Kwast, A Evans, G Lockwood, D Tkachuk, K Watson, DG Bostwick, JI Epstein, P Humphrey, R Montironi, J Srigley, Members of the Pathology Committee of the ERSPC. University Health Network, Toronto, Canada; Bostwick Laboratories, Glen Allen; Johns Hopkins Hospital, Baltimore; Washington University School of Medicine, St. Louis; Polytechnic University of the Marche Region, Ancona, Italy; Credit Valley Hospital, Mississauga, Canada

Background: We investigated the level of diagnostic agreement (kappa) for small foci of atypical glands on prostate biopsy among experts specialized in urogenital pathology and all-round pathologists with extensive experience in reading prostate biopsies.
Design: We retrieved 20 prostate biopsies with small atypical foci, originally reported by urogenital pathologists of University Health Network, Toronto as benign (n=4), suspicious for adenocarcinoma (n=8) and adenocarcinoma (n=8). HE stained slides and if available- matched immunostained slides (n=10) were digitalized. Five experts in urogenital pathology (UGE) and eight members of the pathology committee of the European Randomized study of Screening for Prostate Cancer ERSPC were asked to examine the virtual slides, available through internet. Their diagnoses were transformed into three categories, benign, suspicious and adenocarcinoma. Multirater kappa statistics were applied to determine agreement and significant differences between experts and ERSPC members.
Results: Five UGE and 7 of 8 invited ERSPC pathologists participated in the study. The kappa of the UGE (0.39 (95%CI: 0.29-0.49)) was higher than that of the ERSPC pathologists (0.21 (95%CI: 0.14 0.27). A 100% agreement was reached by UGE for 7 biopsies (4 adenocarcinomas, 2 benign, 1 suspicious) and by ERSPC pathologists for 1 biopsy (benign). UGE and ERSPC pathologists rendered, in 5 and 9 biopsies, respectively a diagnosis ranging from benign to carcinoma. The UGE diagnosed cancer (49%) more often (p< 0.001) than the ERSPC pathologists (32%), mainly at the expense of the number of benign diagnoses. If the agreement was calculated for originally diagnosed cancers and benign diagnoses the kappa's of GUE increased to 0.42 and 0.55, respectively and those of ERSPC pathologists to 0.26 and 0.27 respectively.
Conclusions: Although UGE demonstrate a better agreement in their diagnostic opinion of small atypical foci in prostate biopsies than their non-specialized ERSPC counterparts, their level of agreement remains poor. UGE rendered more often a diagnosis of carcinoma than the ERSPC pathologists.
Category: Genitourinary (including renal tumors)

Wednesday, March 11, 2009 9:30 AM

Poster Session V # 115, Wednesday Morning