MP53-07: Histological Variants of Prostate Cancer are Under-reported in Prostate Biopsies Assessed Outside Tertiary Referral Centers. A Plea for Biopsy Review for Optimal Patient Management

Simone Scuderi*, Francesco Pellegrino, Roberta Lucianò, Nazario Tenace, Milan, Italy, Francesca Paparesta, Laura Pepi, Ancona, Italy, Leonardo Quarta, Lorenzo Toneatto, Elio Mazzone, Francesco Barletta, De Angelis Mario, Daniele Robesti, Donato Cannoletta, Antony Pellegrino, Vito Cucchiara, Riccardo Leni, Giorgio Gandaglia, Armando Stabile, Milan, Italy, Alessia Cimadamore, Rodolfo Montironi, Ancona, Italy, Francesco Montorsi, Alberto Briganti, Milan, Italy

Introduction: Recent studies reported the prognostic impact of histological variant (HV) at prostate biopsy in prostate cancer (PCa) patients. However, the assessment of HV would require additional effort and expertise from pathologists. No study has yet evaluated how pathologist expertise can affect the HV detection at prostate biopsy. This is key to offer patients tailored treatments and prognostic counselling. We aimed at comparing the HV detection at prostate biopsy evaluated at tertiary referral centers vs. peripheral hospitals.

Methods: We prospectively recruited 681 patients who received a prostate biopsy outside referral tertiary centers between 2015 and 2021. All histological specimens were reviewed by two dedicated uro-pathologists at two high-volume academic tertiary referral centers. We compared the detection of prostate cancer HVs (namely, the presence of cribriform, glomeruloid, poorly formed/fused, foamy, mucinous, ductal, intraductal and neuroendocrine variants) at the initial assessment done at non-academic peripheral hospital versus centralized pathological review. To characterize patients with HV, we compared their clinical characteristics to those without HVs using Wilcoxon rank sum test and Pearson's Chi-squared test. Multivariable regression logistic analysis (MVA) tested whether pathologist experience was associated with the detection of HV after adjusting for confounders.

Results: The HV detection by uro-patologist review was markedly higher compared to the initial report (139 [20%] vs 17 [2.4%]). Moreover, pathologists at peripheral hospitals identified HVs in 14 cases which were not confirmed by central pathologic review. The most frequent HV were intraductal and cribriform after central review (62 and 40 cases, respectively), whereas cribriform and mucinous were the most frequent in the initial report (6 and 4 cases, respectively). Patients harboring HV had significantly higher median PSA (8 vs 6 ng/m p<0.001), and prevalence of high Gleason grade group (GG=3 in 62% vs 29%, p<0.001) and positive DRE (74% vs 37%, p<0.001). Overall, 67 of the 718 complete pathological reports reported an HV. At MVA, uro-patologist review resulted associated to the outcome (odds ratio [OR]: 14.5, p<0.001).

Conclusions: Our study demonstrates that HVs of PCa are underdiagnosed by pathologists working outside referral centers. Histological review by a dedicated uro-pathologist is needed to assess the presence of HV which might affect patient clinical management.

Source of Funding: none