Search

De novo metastatic castration-sensitive prostate cancer (mCSPC) is small subgroup of prostate cancer associated with poor prognosis. The aim of our study was to assess the expression of programmed death-ligand 1 (PD-L1) in tumor cells of de novo mCSPC patients. Patients referred to our institution from January 2007 to October 2017 with de novo mCSPC and available diagnostic tissue were included. We tested the PD-L1 pharmaDx qualitative immunohistochemical assay, a monoclonal rabbit anti-PD-L1, clone 28-8 intended for use in the detection of PD-L1 protein in formalin-fixed paraffin-embedded. Kaplan-Meier method was used to estimate survivals according to the expression of PD-L1. The study population included 32 de novo mCSPC patients, analyzed for PD-L1 expression using 2 cut-off values (1% and 5%) to define the positivity. Total of 46.9% of cases had tumor PD-L1 expression ≥1%, and 31.3% had expression ≥5%. No differences were found between the PD-L1 expression ≥1% and the involvement of liver, lung, and number of bone metastases, and the disease volume based on CHAARTED classification. PD-L1 tumors had higher incidence of Gleason score ≥8 compared with PD-L1 tumors (P=0.037), while the incidence of lymph node metastases was higher in PD-L1 tumors (P=0.044). No difference in the median overall survival (mOS) was observed between the PD-L1 population and the PD-L1 patients (43.8 vs. 29.6 mo; P=0.88). The tumor PD-L1 expression cannot be considered a prognostic factor for de novo mCSPC, even if its prognostic and predictive significance have to be thoroughly investigated to better define the selected group of prostate cancer patients that might benefit from checkpoint blockade immunotherapy.

Purpose: Our goal was to evaluate retrospectively 30 cases of serous cystadenoma (SCA) to determine its main imaging features as well as to discuss the differential diagnosis problems versus the other cystic lesions of the pancreas. Method: Thirty SCAs were analyzed; they were all benign lesions, proven at surgery. Twenty-three tumors were evaluated with US, 26 with CT, and 5 with MRI. Results: Three different morphostructural patterns were identified: microlacunar (n = 19), mixed (n = 6), and macrolacunar (n = 5). The diagnosis of SCA, possible in either the microlacunar or the mixed patterns, was achieved in 74% of cases with US (17/23) and in 61.5% with CT (16/26). Among the 19 patients evaluated with both modalities, the joint information allowed a correct diagnosis in 16 cases (84%). The five macrolacunar tumors were undistinguishable from other cystic masses of the pancreas. Conclusion: The diagnosis of SCA can be considered certain in the microlacunar, likely in the mixed, and not possible in the macrolacunar type.