Corpora amylacea in benign prostatic acini are associated with concurrent, predominantly low-grade cancer

• Main Authors: Watchareepohn Palangmonthip, Ruizhe Wu, Sergey Tarima, Samuel A. Bobholz, Peter S. LaViolette, Alexander J. Gallan, Kenneth A. Iczkowski
• Format: Journal
• Published: 2020
• Subjects: Medicine
• Online Access: https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85083052266&origin=inward; http://cmuir.cmu.ac.th/jspui/handle/6653943832/70847
• Institution: Chiang Mai University

© 2020 Wiley Periodicals, Inc. Background: Corpora amylacea (CAM), in benign prostatic acini, contain acute-phase proteins. Do CAM coincide with carcinoma?. Methods: Within 270 biopsies, 83 prostatectomies, and 33 transurethral resections (TURs), CAM absence was designated CAM 0; corpora in less than 5% of benign acini: CAM 1; in 5% to 25%: CAM 2; in more than 25%: CAM 3. CAM were compared against carcinoma presence, clinicopathologic findings, and grade groups (GG) 1 to 2 vs 3 to 5. The frequency of CAM according to anatomic zone was counted. A pilot study was conducted using paired initial benign and repeat biopsies (33 benign, 24 carcinoma). Results: A total of 68.9% of biopsies, 96.4% of prostatectomies, and 66.7% of TURs disclosed CAM. CAM ≥1 was common at an older age (P =.019). In biopsies, 204 cases (75%) had carcinoma; and CAM of 2 to 3 (compared to 0-1) were recorded in 25.0% of carcinomas but only 7.4% of benign biopsies (P =.005; odds ratio [OR] = 5.1). CAM correlated with high percent Gleason pattern 3, low GG (P =.035), and chronic inflammation (CI). CI correlated inversely with carcinoma (P =.003). CAM disclosed no association with race, body mass index, serum prostate specific antigen (PSA), acute inflammation (in biopsies), atrophy, or carcinoma volume. With CAM 1, the odds of GG 3 to 5 carcinoma, by comparison to CAM 0, decreased more than 2× (OR = 0.48; P =.032), with CAM 2, more than 3× (OR = 0.33; P =.005), and with CAM 3, almost 3× (OR = 0.39, P =.086). For men aged less than 65, carcinoma predictive model was: Score = (2 × age) + (5 × PSA) − (20 × degree of CAM); using our data, area under the ROC curve was 78.17%. When the transition zone was involved by cancer, it showed more CAM than in cases where it was uninvolved (P =.012); otherwise zonal distributions were similar. In the pilot study, CAM ≥1 predicted carcinoma on repeat biopsy (P <.05; OR = 8), as did CAM 2 to 3 (P <.0001; OR = 30). CI was not significant, and CAM retained significance after adjusting for CI. Conclusion: CAM correlate with carcinoma. Whether abundant CAM in benign biopsies adds value amidst high clinical suspicion, warrants further study.