Diagnostic utility of immunohistochemical staining for p63, a sensitive marker of prostatic basal cells.

Diagnostically reliable identification of prostatic basal cells has depended on staining for high molecular weight cytokeratin. The diagnosis of malignancy is often based on the absence of basal cells. False-negative staining is occasionally observed. Thus, a second method of identifying basal cells might prove useful. Selective expression of p63, a homologue of p53, has been demonstrated in prostatic basal cells. We investigated the diagnostic utility of p63 staining in 70 consecutive specimens for which the differential diagnosis included prostatic adenocarcinoma: 68 needle biopsies and 2 transurethral resection blocks. High molecular weight cytokeratin staining was the gold standard when material was available. A total of 61 specimens were diagnosed as carcinoma, 4 as atrophy, 2 as high-grade prostatic intraepithelial neoplasia, 2 as unclassified collections of benign glands, and 1 as carcinoma versus high-grade prostatic intraepithelial neoplasia. Sections mounted on charged slides were used for p63 staining for 14 specimens. Sections previously hematoxylin and eosin stained on uncharged slides were used for 56 specimens. In every case in which there was successful p63 staining (55 specimens), basal cells in benign lesions were properly marked and other cell types were not stained. Uninformative staining in the remaining 15 specimens was due to failure of tissue adherence in 14 specimens in which sections were on uncharged slides and, in 1 specimen, to poor positive internal control staining of benign glands. Thus, p63 staining was informative in 55 of 56 specimens (98%) for which there was material for examination. No case with satisfactory p63 and high molecular weight staining showed disagreement between the two stains. An additional group of 21 transurethral resection specimens was stained (p63 and high molecular weight cytokeratin). There was less false-negative staining for p63 compared with the case of high molecular weight cytokeratin. No false-positive staining was seen. We conclude that p63 staining is at least as sensitive and specific for the identification of basal cells in diagnostic prostate specimens as is high molecular weight cytokeratin staining.