Assessment of MYC/PTEN Status by Gene-Protein Assay in Grade Group 2 Prostate Biopsies.

This study leveraged a gene-protein assay to assess MYC and PTEN status at prostate cancer biopsy and examined the association with adverse outcomes after surgery. MYC gain and PTEN loss were simultaneously assessed by chromogenic in situ hybridization and immunohistochemistry, respectively, using 277 Grade Group 2 needle biopsies that were followed by prostatectomy. The maximal size of cribriform Gleason pattern 4 carcinoma (CRIB), the presence of intraductal carcinoma (IDC), and percentage of Gleason pattern 4 carcinoma at biopsy were also annotated. MYC gain or PTEN loss was present in 19% and 18% of biopsies, respectively, whereas both alterations were present in 9% of biopsies. Tumors with one or both alterations were significantly more likely to have non-organ-confined disease (NOCD) at radical prostatectomy. In logistic regression models, including clinical stage, tumor volume on biopsy, and presence of CRIB/IDC, cases with MYC gain and PTEN loss remained at higher risk for NOCD (odds ratio, 6.23; 95% CI, 1.74-24.55; P = 0.005). The area under the curve for a baseline model using CAPRA variables (age, prostate-specific antigen, percentage of core involvement, clinical stage) was increased from 0.68 to 0.69 with inclusion of CRIB/IDC status and to 0.75 with MYC/PTEN status. Dual MYC/PTEN status can be assessed in a single slide and is independently associated with increased risk of NOCD for Grade Group 2 biopsies.

High throughput assessment of biomarkers in tissue microarrays using artificial intelligence: PTEN loss as a proof-of-principle in multi-center prostate cancer cohorts.

Phosphatase and tensin homolog (PTEN) loss is associated with adverse outcomes in prostate cancer and has clinical potential as a prognostic biomarker. The objective of this work was to develop an artificial intelligence (AI) system for automated detection and localization of PTEN loss on immunohistochemically (IHC) stained sections. PTEN loss was assessed using IHC in two prostate tissue microarrays (TMA) (internal cohort, n = 272 and external cohort, n = 129 patients). TMA cores were visually scored for PTEN loss by pathologists and, if present, spatially annotated. Cores from each patient within the internal TMA cohort were split into 90% cross-validation (N = 2048) and 10% hold-out testing (N = 224) sets. ResNet-101 architecture was used to train core-based classification using a multi-resolution ensemble approach (×5, ×10, and ×20). For spatial annotations, single resolution pixel-based classification was trained from patches extracted at ×20 resolution, interpolated to ×40 resolution, and applied in a sliding-window fashion. A final AI-based prediction model was created from combining multi-resolution and pixel-based models. Performance was evaluated in 428 cores of external cohort. From both cohorts, a total of 2700 cores were studied, with a frequency of PTEN loss of 14.5% in internal (180/1239) and external 13.5% (43/319) cancer cores. The final AI-based prediction of PTEN status demonstrated 98.1% accuracy (95.0% sensitivity, 98.4% specificity; median dice score = 0.811) in internal cohort cross-validation set and 99.1% accuracy (100% sensitivity, 99.0% specificity; median dice score = 0.804) in internal cohort test set. Overall core-based classification in the external cohort was significantly improved in the external cohort (area under the curve = 0.964, 90.6% sensitivity, 95.7% specificity) when further trained (fine-tuned) using 15% of cohort data (19/124 patients). These results demonstrate a robust and fully automated method for detection and localization of PTEN loss in prostate cancer tissue samples. AI-based algorithms have potential to streamline sample assessment in research and clinical laboratories.

Risk Stratification of Prostate Cancer Through Quantitative Assessment of PTEN Loss (qPTEN).

BACKGROUND: Phosphatase and tensin homolog (PTEN) loss has long been associated with adverse findings in early prostate cancer. Studies to date have yet to employ quantitative methods (qPTEN) for measuring of prognostically relevant amounts of PTEN loss in postsurgical settings and demonstrate its clinical application. METHODS: PTEN protein levels were measured by immunohistochemistry in radical prostatectomy samples from training (n = 410) and validation (n = 272) cohorts. PTEN loss was quantified per cancer cell and per tissue microarray core. Thresholds for identifying clinically relevant PTEN loss were determined using log-rank statistics in the training cohort. Univariate (Kaplan-Meier) and multivariate (Cox proportional hazards) analyses on various subpopulations were performed to assess biochemical recurrence-free survival (BRFS) and were independently validated. All statistical tests were two-sided. RESULTS: PTEN loss in more than 65% cancer cells was most clinically relevant and had statistically significant association with reduced BRFS in training (hazard ratio [HR] = 2.48, 95% confidence interval [CI] = 1.59 to 3.87; P < .001) and validation cohorts (HR = 4.22, 95% CI = 2.01 to 8.83; P < .001). The qPTEN scoring method identified patients who recurred within 5.4 years after surgery (P < .001). In men with favorable risk of biochemical recurrence (Cancer of the Prostate Risk Assessment - Postsurgical scores <5 and no adverse pathological features), qPTEN identified a subset of patients with shorter BRFS (HR = 5.52, 95% CI = 2.36 to 12.90; P < .001) who may be considered for intensified monitoring and/or adjuvant therapy. CONCLUSIONS: Compared with previous qualitative approaches, qPTEN improves risk stratification of postradical prostatectomy patients and may be considered as a complementary tool to guide disease management after surgery.