International Society of Urological Pathology Consensus on Cancer Precursor Lesions. Working Group 1: The Prostate

Abstract

Working Group 1 at ISUP's Cancer Precursors meeting (September 2024) evaluated 5 putative precursors of invasive prostate cancer: high-grade prostatic intraepithelial neoplasia (HGPIN), intraductal carcinoma (IDC), atypical intraductal proliferation (AIP), atypical adenomatous hyperplasia (AAH)/adenosis, and proliferative inflammatory atrophy (PIA). Objectives were to compile recent evidence, interrogate current practices, and vote on recommendations, with 67% approval defined as consensus. Consensus was reached against the reporting of the low-grade form of PIN. HGPIN need not be reported when concomitant cancer or atypical small acinar proliferation suspicious for cancer exists adjacent to it, for biopsy or prostatectomy specimens. Finally, while the clinical significance of unifocal HGPIN in biopsies remains uncertain, there is stronger evidence for multifocal isolated HGPIN as a predictor of subsequent cancer detection. By consensus, multifocal HGPIN should continue being reported. Slight refinement was achieved regarding IDC criteria. The consensus opinion was that a dense cribriform to solid proliferation need not demonstrate marked nuclear atypia/ pleomorphism to qualify as IDC. The inverse scenario of marked atypia without dense cribriform/solid proliferation fell just short (65%) of consensus for IDC. Redesignating cribriform HGPIN as AIP achieved consensus. AIP found alone or with grade group 1 cancer warrants an explanatory comment. However, agreement was not attained to report AIP in the presence of invasive cancer, in either needle biopsy or prostatectomy. Finally, the optional reporting of PIA or AAH/adenosis in biopsies as pertinent negatives both fell short of consensus. This guidance should help pathologists standardize reporting, staying focused on the clinically actionable aspects of these lesions.

Keywords: atypical intraductal proliferation; consensus conference; intraductal carcinoma; precursor lesion; prostate; prostatic intraepithelial neoplasia.

FIGURE 1.

A, Tufted pattern of HGPIN is the most common of its 4 patterns. B, Micropapillary pattern of HGPIN features filiform structures lacking a fibrovascular core. C, Flat pattern of HGPIN is seen in acini at the top of the image. D, Cribriform pattern of HGPIN was, by consensus, redesignated as atypical intraductal proliferation (AIP). E, By triple (cocktail) immunostain, HGPIN characteristically retains at least a patchy basal cell layer (brown, high molecular weight cytokeratin and p63) while showing increased AMCR (red, also called P504S), similar to cancer cells. F, Situated adjacent to HGPIN (#, with brown MYC protein overexpression), the normal-appearing acini (*) show partial loss of red Glutathione S-transferase (GSTP1) through promoter hypermethylation. This signals susceptibility to oxidative stress-induced DNA damage, likely an important step in carcinogenesis. G, Intraductal carcinoma (IDC) with usual cribriform pattern. Small, dense basal cells are evident at the periphery. H, Marked nuclear enlargement alone is often found in IDC, but it is not sufficient as a sole criterion for IDC. I, Atypical intraductal proliferation (AIP) demonstrates features intermediate between HGPIN and IDC. This cribriform proliferation shows high cellularity but minimal atypia. J, AIP, showing less cellularity but more nuclear atypia, with nuclear clearing. K, Atypical adenomatous hyperplasia (AAH/adenosis) occurs as part of a BPH nodule (left). It lacks atypia and shows a corpus amylaceum (arrow). L, Proliferative inflammatory atrophy has acini with diminished cytoplasm and reactive features in an inflammatory background.

FIGURE 2.

Time-wise decline in the predictive value of isolated high-grade PIN (HGPIN) for all cancers on repeat biopsy.

FIGURE 3.

Multi-core HGPIN shows a distinctly higher predictive value for cancer on repeat biopsy than single-core HGPIN. GS indicates Gleason score.

FIGURE 4.

Perineural invasion by cancer may sometimes mimic HGPIN (A, B), but S-100 protein (C) and basal cell cytokeratin (D) confirm its identity.