The expression and location of proliferating cell nuclear antigen (PCNA) immunostaining in epithelial, endothelial and stromal nuclei were assessed in prostatic intra-epithelial neoplasia (PIN). It was then compared with patterns in benign lesions and in invasive adenocarcinomas of the prostate. The PCNA-positive nuclei showed homogeneous or granular types of staining, or a mixture of both, and a gradation in the intensity of staining. Nuclei with granular and mixed patterns appeared lighter brown than those with a homogeneous pattern, which are darker and more often noted in PIN and invasive adenocarcinomas than in benign lesions. For epithelial PCNA-stained nuclei, the proportions in the two grades of PIN were greater than in benign prostatic hyperplasia (mean 3.16%, SE 0.31%) and prostatic atrophic ducts and acini (mean 0.56%, SE 0.09%), the values decreasing from the nuclei in the basal position towards those in the luminal layer. In grade 1, the category mean values were 9.51% (SE 1.14%) in the basal, 7.02% (SE 1.27%) in the intermediate and 6.02% (SE 0.90%) in the luminal position. In grade 2, the category mean values were 13.81% (SE 1.42%) in the basal position, 10.99% (SE 1.17%) in the intermediate and 7.91% (SE 1.43%) in the luminal position. In small and large acinar adenocarcinomas, the proportions of positive nuclei were 8.66% (SE 0.30%) and 9.06% (SE 0.30%), respectively. The category mean values in the cribriform adenocarcinomas were 14.40% (SE 0.61%) in the basal position, 11.84% (SE 1.30%) in the intermediate and 9.26% (SE 0.66%) in the luminal position. As in PIN, the proportions of immunostained nuclei in the adenocarcinoma with cribriform pattern decreased from the basal towards the luminal layer. In the solid/trabecular adenocarcinomas, the category mean value in the cell layer adjacent to the stroma was 17.60% (SE 2.92%), whereas in the other cell layers it was lower than that in the cells adjacent the stroma (mean 13.88%, SE 1.71%). For capillary endothelial and stromal cells, the percentages of PCNA-stained nuclei were much lower than those in the epithelial component. The lowest mean values were obtained in benign lesions, whereas the highest were in invasive adenocarcinomas, the percentages in PIN being intermediate.

The aim of our study was to assess the frequency and location of apoptotic bodies (ABs) in haematoxylin and eosin-stained sections of prostatic intraepithelial neoplasia (PIN) and then to compare the patterns with those in benign prostatic hyperplasia (BPH) and prostatic invasive adenocarcinoma (PAC). ABs were identified in all epithelial cell layers of the ducts, acini and tumour islands, as well as in the lumina contained in such structures. In the epithelial cell layers, ABs were found in general in the intercellular space and occasionally in the cytoplasm of epithelial cells. The frequency of ABs increased from BPH through PIN up to PAC. The proportions of ABs in PIN lesions of low grade (PINlow) and high grade (PINhigh) were greater than in BPH, the values decreasing from the nuclei in the basal position towards those in the luminal layer. In PINlow, the mean category values were 0.85% (standard error, SE, 0.311%) in the basal, 0.623% (SE 0.065%) in the intermediate and 0.474% (SE 0.138%) in the luminal position. In PINhigh, the mean category values were 1.006% (SE 0.16%) in the basal position, 0.713% (SE 0.182%) in the intermediate and 0.618% (SE 0.172%) in the luminal position. The proportions of ABs in adenocarcinoma with cribriform pattern decreased from the basal towards the luminal layer, as for PIN: 1.806% (SE 0.346%) in the basal position, 1.15% (SE 0.172%) in the intermediate and 0.886% (SE 0.137%) in the luminal position. In the solid/trabecular adenocarcinomas, the mean category value in the cell layer adjacent to the stroma was 2.154% (SE 0.203%), whereas in the other cell layers it was 2.052% (SE 0.239%). In small and large acinar adenocarcinomas, the proportions of positive nuclei were 1.022% (SE 0.1%) and 0.922% (SE 0.163%), respectively. The evaluation of the frequency and location of ABs gives accurate information on cell death in PIN in comparison with BPH and PAC.

High-grade prostatic intraepithelial neoplasia (HGPIN) has been established as a precursor to prostatic adenocarcinoma. HGPIN shares many morphological, genetic, and molecular signatures with prostate cancer. Its predictive value for the development of future adenocarcinoma during the prostate-specific antigen screening era has decreased, mostly owing to the increase in prostate biopsy cores. Nevertheless, a literature review supports that large-volume HGPIN and multiple cores of involvement at the initial biopsy should prompt a repeat biopsy of the prostate within 1 year. No treatment is recommended for HGPIN to slow its progression to cancer.

No abstract

The diagnostic value of nucleolar margination, defined as the percentage of nucleoli touching the nuclear membrane, was investigated in 359 cytological preparations of benign and malignant lesions of the thyroid, breast, prostate and central nervous system. Premalignant lesions of the uterine cervix and non-invasive papillary carcinomas of the bladder were also examined. It was observed that the percentages in benign lesions were, in general, lower than in the malignant and that the values increased progressively with increasing grade in the cervix and bladder. When the overlap index was calculated, this gave exact information on the usefulness of nucleolar margination in distinguishing benign from malignant lesions, particularly in the prostate and thyroid and, to a lesser extent, in the breast and central nervous system. As for lesions of different grades, the calculation of the index allowed the identification of two subgroups, one corresponding to low grades (mild cervical dysplasia or urothelial papillary carcinoma of grade 1), the other subgroup to high grades (severe cervical dysplasia and carcinoma in situ, or papillary carcinoma of grade 3). Moderate dysplasia cases and grade 2 papillary carcinomas do not appear as separate intermediate categories but rather show values falling into the range of either the higher or lower grades. The margination values obtained from the cytological preparations corresponded well to those in the histological slides obtained from the resected specimens. In conclusion, nucleolar margination appears to be a feature which is easy to evaluate in a reproducible way and useful in cytological diagnosis.