1. ¹Service de Pathologie, Hôpital Foch, Suresnes, France
2. ²Service de Pathologie, Institut Mutualiste Montsouris, Paris, France
3. ³Service de Pathologie, Hôpital Tenon, Paris, France
4. ⁴Service de Pathologie, Hôpital Cochin, Paris, France
5. ⁵Service de Pathologie, Hôpital Necker, Paris, France
6. ⁶Service de Pathologie, Hôpital Lariboisière, Paris, France
7. ⁷Service d'Urologie, Hôpital Foch, Suresnes, France

Correspondence: Dr V Molinié, MD, Département de Pathologie, Hôpital Foch, 40 Rue Worth, 92151 Suresnes Cedex, France. E-mail: v.molinie@hopital-foch.org

Received 23 December 2003; Revised 18 February 2004; Accepted 19 February 2004; Published online 18 June 2004.

Abstract

Prostatic needle biopsy is the preferred method for diagnosing early prostate cancer, providing specific information. In cases of histological cancer mimics, a diagnosis of atypical small acinar proliferation suspected of but not diagnosed as malignancy can be made. In such cases, and in small focus carcinomas, pathologists use 34E12, cytokeratin (CK) 5/6 or p63 immunostaining to label basal cells, and -methylacyl-CoA racemase (AMACR/p504s) immunostaining as a positive prostate cancer marker on two distinct slides. However, in cases of small foci, ambiguous lesions might disappear. The purpose of our study was to improve the sensitivity of a cocktail of two antibodies (p63/p504s) with a sample incubation on 260 prostatic specimens, in order to help make a decision in conjunction with standard histology and CK 5/6 immunostaining. We tested 101 small focus prostatic cancers, 104 atypical small acinar proliferation, 19 high-grade prostatic intraepithelial neoplasia, two atypical adenomatous hyperplasia and 34 benign mimics of cancer. After p63/p504s immunostaining, the final diagnoses retained were as follows: 154 prostatic cancers, 14 atypical small acinar proliferation, 30 high-grade prostatic intraepithelial neoplasia, three atypical adenomatous hyperplasia and 62 benign mimics of cancer. To differentiate malignant from benign lesions, we used the criteria of greater sensitivity to p504s/p63 (95%) than to CK 5/6 (57%) or p63 (86%), and higher specificity for p504s/p63 (95%) than for CK 5/6 (88%) or p63 (81%). With the p504s/p63 cocktail, 89% of the ambiguous lesions were classified vs 53% for CK 5/6. Combined use of the two antibodies, one (p504s) as a positive marker and the other (p63) as a negative marker, with a simple immunostaining procedure, may improve diagnostic performance, sensitivity and specificity, leading to a reduction in the risk of false negatives; this technique in cases of atypical small acinar proliferation should reduce the percentage of residual ambiguous lesions and the need for additional biopsies.