Enhancing Prostate Cancer Diagnosis: AI-Driven Virtual Biopsy for Optimal MRI-Targeted Biopsy Approach and Gleason Grading Strategy.

An optimal approach to MRI fusion targeted prostate biopsy (PBx) remains unclear (number of cores, inter-core distance, Gleason grading (GG) principle). The aim of this study was to develop a precise pixel-wise segmentation diagnostic AI algorithm for tumor detection and GG as well as an algorithm for virtual prostate biopsy that are used together to systematically investigate and find an optimal approach to targeted PBx. Pixel-wise AI algorithms for tumor detection and GG were developed using a high-quality, manually annotated dataset (slides n=442) after fast-track annotation transfer into segmentation style. To this end, a virtual biopsy algorithm was developed that can perform random biopsies from tumor regions in whole-mount whole-slide images with pre-defined parameters. A cohort of 115 radical prostatectomy (RP) patient cases with clinically significant, MRI-visible tumors (n=121) was used for systematic studies of the optimal biopsy approach. Three expert genitourinary (GU) pathologists participated in the validation. The tumor detection algorithm (aware version sensitivity/specificity 0.99/0.90, balanced version 0.97/0.97) and GG algorithm (quadratic kappa range vs pathologists 0.77-0.78) perform on par with expert GU pathologists. In total, 65,340 virtual biopsies were performed to study different biopsy approaches with the following results: 1) four biopsy cores is the optimal number for a targeted PBx, 2) cumulative GG strategy is superior to using maximal Gleason score for single cores, 3) controlling for minimal inter-core distance does not improve the predictive accuracy for the RP Gleason score, 4) Using tertiary Gleason pattern principle (for AI tool) in cumulative GG strategy might allow better predictions of final RP Gleason score. The AI algorithm (based on cumulative GG strategy) predicted the RP Gleason score of the tumor better than 2 of the 3 expert GU pathologists. In this study, using an original approach of virtual prostate biopsy on the real cohort of patient cases, we find the optimal approach to the biopsy procedure and the subsequent Gleason grading of a targeted PBx. We publicly release two large datasets with associated expert pathologists' GG and our virtual biopsy algorithm.

Prognostic impact of concomitant loss of PBRM1 and BAP1 protein expression in early stages of clear cell renal cell carcinoma.

PURPOSE: To evaluate the prognostic impact of immunohistochemical expression of BAP1 and PBRM1 in patients with early stage (pT1-pT2N0M0) clear cell renal cell carcinoma (ccRCC). PATIENTS AND METHODS: A total of 441 consecutive patients treated surgically for stages I and II (TNM-AJCC 2010) ccRCC between 1990 and 2016 were selected. All cases were reviewed for uniform reclassification and the most representative tumor areas were selected for the construction of a tissue microarray. Sixty-two patients had frozen tumoral tissue available in the tumor bank of our institution for quantitative real-time reverse transcriptase polymerase chain reaction analysis. RESULTS: Of the 441-immunostained ccRCC specimens, 91 (20.6%) and 107 (24.3%) showed negative-expression of PBRM1 and BAP1, respectively. Fifty-eight (13.2%) showed negative-expression of both markers (PBRM1-/BAP-). There was an association between both markers expression pattern and classical parameters, such as pT stage (P<0.001), tumor size (P<0.001), and tumor grade (P<0.001). Both independent PBRM1 and BAP1 negative-expression were associated with lower rates of disease-specific survival and recurrence-free survival. When patients were grouped into presence of positive expression of one or both markers vs. PBRM1-/BAP1- patients, disease-specific survival and rates were 95.3% vs. 77.6%, respectively (P<0.001). PBRM1-/BAP1-group presented a higher risk of cancer specific death (hazard ratio = 2.722, P = 0.007) and disease recurrence (hazard ratio = 2.467, P = 0.004) in multivariate analysis. CONCLUSION: Patients with early stage tumors that present concomitant loss of both PBRM1 and BAP1 demonstrated worse survival rates and represent a relevant risk group for tumor recurrence and death.

GATA3 expression in small cell carcinoma of bladder and prostate and its potential role in determining primary tumor origin.

GATA3 is a sensitive marker for urothelial carcinoma. We here evaluate, for the first time, GATA3 expression in small cell carcinoma of bladder and prostate and assess its utility in the differential diagnosis with small cell carcinoma of lung primary. Archival tissues from 60 small cell carcinomas (12 bladder, 15 lung, and 33 prostate primary cases) were used to build 2 tissue microarrays. We also assessed whole slide sections from 10 additional primary small cell carcinomas of bladder. GATA3 nuclear expression was evaluated using standard immunohistochemistry. Intensity (weak, moderate, and strong) and extent of expression were assessed in each tissue microarray spot. Extent positivity was categorized as focal (1%-25%), multifocal (>25%), and diffuse (>75%). Nuclear GATA3 expression was encountered in 7 bladder (7/22, 32%) and 2 lung (2/15, 13%) small cell carcinomas. All 33 primary prostate small cell carcinomas were negative. Among bladder tumors, strong and diffuse (>75%) GATA3 labeling was seen in 3 cases (3/22, 14%); focal positivity was observed in the 4 remaining cases (4/22, 18%). Both positive lung cases had only focal positivity. Our study is the first to reveal GATA3 expression in the small subset of lung small cell carcinoma that should be taken into consideration in assigning site of origin in advanced small cell carcinoma cases. Our novel finding of GATA3 positivity in one-third of bladder small cell carcinoma is of potential value in differentiating small cell carcinomas of prostate origin from those of bladder origin.