External Validation of a Digital Pathology-based Multimodal Artificial Intelligence Architecture in the NRG/RTOG 9902 Phase 3 Trial.

BACKGROUND: Accurate risk stratification is critical to guide management decisions in localized prostate cancer (PCa). Previously, we had developed and validated a multimodal artificial intelligence (MMAI) model generated from digital histopathology and clinical features. Here, we externally validate this model on men with high-risk or locally advanced PCa treated and followed as part of a phase 3 randomized control trial. OBJECTIVE: To externally validate the MMAI model on men with high-risk or locally advanced PCa treated and followed as part of a phase 3 randomized control trial. DESIGN, SETTING, AND PARTICIPANTS: Our validation cohort included 318 localized high-risk PCa patients from NRG/RTOG 9902 with available histopathology (337 [85%] of the 397 patients enrolled into the trial had available slides, of which 19 [5.6%] failed due to poor image quality). OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: Two previously locked prognostic MMAI models were validated for their intended endpoint: distant metastasis (DM) and PCa-specific mortality (PCSM). Individual clinical factors and the number of National Comprehensive Cancer Network (NCCN) high-risk features served as comparators. Subdistribution hazard ratio (sHR) was reported per standard deviation increase of the score with corresponding 95% confidence interval (CI) using Fine-Gray or Cox proportional hazards models. RESULTS AND LIMITATIONS: The DM and PCSM MMAI algorithms were significantly and independently associated with the risk of DM (sHR [95% CI] = 2.33 [1.60-3.38], p < 0.001) and PCSM, respectively (sHR [95% CI] = 3.54 [2.38-5.28], p < 0.001) when compared against other prognostic clinical factors and NCCN high-risk features. The lower 75% of patients by DM MMAI had estimated 5- and 10-yr DM rates of 4% and 7%, and the highest quartile had average 5- and 10-yr DM rates of 19% and 32%, respectively (p < 0.001). Similar results were observed for the PCSM MMAI algorithm. CONCLUSIONS: We externally validated the prognostic ability of MMAI models previously developed among men with localized high-risk disease. MMAI prognostic models further risk stratify beyond the clinical and pathological variables for DM and PCSM in a population of men already at a high risk for disease progression. This study provides evidence for consistent validation of our deep learning MMAI models to improve prognostication and enable more informed decision-making for patient care. PATIENT SUMMARY: This paper presents a novel approach using images from pathology slides along with clinical variables to validate artificial intelligence (computer-generated) prognostic models. When implemented, clinicians can offer a more personalized and tailored prognostic discussion for men with localized prostate cancer.

Use of the Decipher genomic classifier among men with prostate cancer in the United States.

BACKGROUND: Management of localized or recurrent prostate cancer since the 1990s has been based on risk stratification using clinicopathological variables, including Gleason score, T stage (based on digital rectal exam), and prostate-specific antigen (PSA). In this study a novel prognostic test, the Decipher Prostate Genomic Classifier (GC), was used to stratify risk of prostate cancer progression in a US national database of men with prostate cancer. METHODS: Records of prostate cancer cases from participating SEER (Surveillance, Epidemiology, and End Results) program registries, diagnosed during the period from 2010 through 2018, were linked to records of testing with the GC prognostic test. Multivariable analysis was used to quantify the association between GC scores or risk groups and use of definitive local therapy after diagnosis in the GC biopsy-tested cohort and postoperative radiotherapy in the GC-tested cohort as well as adverse pathological findings after prostatectomy. RESULTS: A total of 572 545 patients were included in the analysis, of whom 8927 patients underwent GC testing. GC biopsy-tested patients were more likely to undergo active active surveillance or watchful waiting than untested patients (odds ratio [OR] =2.21, 95% confidence interval [CI] = 2.04 to 2.38, P < .001). The highest use of active surveillance or watchful waiting was for patients with a low-risk GC classification (41%) compared with those with an intermediate- (27%) or high-risk (11%) GC classification (P < .001). Among National Comprehensive Cancer Network patients with low and favorable-intermediate risk, higher GC risk class was associated with greater use of local therapy (OR = 4.79, 95% CI = 3.51 to 6.55, P < .001). Within this subset of patients who were subsequently treated with prostatectomy, high GC risk was associated with harboring adverse pathological findings (OR = 2.94, 95% CI = 1.38 to 6.27, P = .005). Use of radiation after prostatectomy was statistically significantly associated with higher GC risk groups (OR = 2.69, 95% CI = 1.89 to 3.84). CONCLUSIONS: There is a strong association between use of the biopsy GC test and likelihood of conservative management. Higher genomic classifier scores are associated with higher rates of adverse pathology at time of surgery and greater use of postoperative radiotherapy.In this study the Decipher Prostate Genomic Classifier (GC) was used to analyze a US national database of men with prostate cancer. Use of the GC was associated with conservative management (ie, active surveillance). Among men who had high-risk GC scores and then had surgery, there was a 3-fold higher chance of having worrisome findings in surgical specimens.

Utilization of genetic testing in men with advanced prostate cancer.

BACKGROUND: Genetic evaluation of men with advanced prostate cancer is recognized as imperative both to guide treatment decisions and to trigger cascade genetic testing of family members. Here we investigate utilization patterns of genetic testing among a contemporary cohort of men with advanced prostate cancer at our institution. METHODS: We queried the Northwestern Electronic Data Warehouse from January 2021 to present for all men diagnosed with National Comprehensive Cancer Network high-risk/very high-risk, regional, or metastatic prostate cancer. Patients were excluded from analyses if treated at an outside institution and/or presented for a second opinion evaluation. Statistics were performed using t-test, Chi-squared test, and univariable and multivariable logistic regression with significance defined as p < 0.05. RESULTS: Atotal of 320 men (52.5%) had local/regional disease and 290 (47.5%) had metastatic disease, 53 (18.3%) of whom had castrate resistant prostate cancer. Rates of germline genetic testing rate were low in patients with localized disease (9.4%) and metastatic disease (34.1%). Only 19 (35.8%) men diagnosed with metastatic castrate resistant prostate cancer underwent germline genetic evaluation. Germline testing was most frequently discussed or ordered by medical oncologists (52%) followed by urologists (20%). Men who underwent germline testing were younger (p < 0.001), more likely to have Medicaid or private insurance (p = 0.002), and more likely to have metastatic disease (p < 0.001). There were no statistically significant differences in baseline PSA, ethnicity, race, or castration sensitivity status. Age (odds ratio [OR]: 0.94, 95% confidence interval [CI]: 0.91-0.97, p < 0.001) and metastatic disease (OR: 5.71, 95% CI: 3.63-9.22, p < 0.001) were significant independent predictors of genetic testing on multivariable logistic regression. CONCLUSIONS: Here we report that utilization of genetic testing is associated with metastatic disease and inversely associated with age. Overall, utilization rates of genetic testing remain low in all patient groups, including in the metastatic castrate resistant setting, where genetic testing can identify patients with homologous recombination repair deficiency who may benefit from use of targeted therapeutics such as PARP inhibitors. Genetic testing in men with aggressive prostate cancer is critical and barriers to routine implementation of testing require further study to develop strategies to improve utilization rates.

Performance of clinicopathologic models in men with high risk localized prostate cancer: impact of a 22-gene genomic classifier.

BACKGROUND: Prostate cancer exhibits biological and clinical heterogeneity even within established clinico-pathologic risk groups. The Decipher genomic classifier (GC) is a validated method to further risk-stratify disease in patients with prostate cancer, but its performance solely within National Comprehensive Cancer Network (NCCN) high-risk disease has not been undertaken to date. METHODS: A multi-institutional retrospective study of 405 men with high-risk prostate cancer who underwent primary treatment with radical prostatectomy (RP) or radiation therapy (RT) with androgen-deprivation therapy (ADT) at 11 centers from 1995 to 2005 was performed. Cox proportional hazards models were used to determine the hazard ratios (HR) for the development of metastatic disease based on clinico-pathologic variables, risk groups, and GC score. The area under the receiver operating characteristic curve (AUC) was determined for regression models without and with the GC score. RESULTS: Over a median follow-up of 82 months, 104 patients (26%) developed metastatic disease. On univariable analysis, increasing GC score was significantly associated with metastatic disease ([HR]: 1.34 per 0.1 unit increase, 95% confidence interval [CI]: 1.19-1.50, p < 0.001), while age, serum PSA, biopsy GG, and clinical T-stage were not (all p > 0.05). On multivariable analysis, GC score (HR: 1.33 per 0.1 unit increase, 95% CI: 1.19-1.48, p < 0.001) and GC high-risk (vs low-risk, HR: 2.95, 95% CI: 1.79-4.87, p < 0.001) were significantly associated with metastasis. The addition of GC score to regression models based on NCCN risk group improved model AUC from 0.46 to 0.67, and CAPRA from 0.59 to 0.71. CONCLUSIONS: Among men with high-risk prostate cancer, conventional clinico-pathologic data had poor discrimination to risk stratify development of metastatic disease. GC score was a significant and independent predictor of metastasis and may help identify men best suited for treatment intensification/de-escalation.

Development and Validation of a Novel Integrated Clinical-Genomic Risk Group Classification for Localized Prostate Cancer.

Purpose It is clinically challenging to integrate genomic-classifier results that report a numeric risk of recurrence into treatment recommendations for localized prostate cancer, which are founded in the framework of risk groups. We aimed to develop a novel clinical-genomic risk grouping system that can readily be incorporated into treatment guidelines for localized prostate cancer. Materials and Methods Two multicenter cohorts (n = 991) were used for training and validation of the clinical-genomic risk groups, and two additional cohorts (n = 5,937) were used for reclassification analyses. Competing risks analysis was used to estimate the risk of distant metastasis. Time-dependent c-indices were constructed to compare clinicopathologic risk models with the clinical-genomic risk groups. Results With a median follow-up of 8 years for patients in the training cohort, 10-year distant metastasis rates for National Comprehensive Cancer Network (NCCN) low, favorable-intermediate, unfavorable-intermediate, and high-risk were 7.3%, 9.2%, 38.0%, and 39.5%, respectively. In contrast, the three-tier clinical-genomic risk groups had 10-year distant metastasis rates of 3.5%, 29.4%, and 54.6%, for low-, intermediate-, and high-risk, respectively, which were consistent in the validation cohort (0%, 25.9%, and 55.2%, respectively). C-indices for the clinical-genomic risk grouping system (0.84; 95% CI, 0.61 to 0.93) were improved over NCCN (0.73; 95% CI, 0.60 to 0.86) and Cancer of the Prostate Risk Assessment (0.74; 95% CI, 0.65 to 0.84), and 30% of patients using NCCN low/intermediate/high would be reclassified by the new three-tier system and 67% of patients would be reclassified from NCCN six-tier (very-low- to very-high-risk) by the new six-tier system. Conclusion A commercially available genomic classifier in combination with standard clinicopathologic variables can generate a simple-to-use clinical-genomic risk grouping that more accurately identifies patients at low, intermediate, and high risk for metastasis and can be easily incorporated into current guidelines to better risk-stratify patients.

Implementation of a Surgeon-Level Comparative Quality Performance Review to Improve Positive Surgical Margin Rates during Radical Prostatectomy.

PURPOSE: Success in the era of value-based payment will depend on the capacity of health systems to improve quality while controlling costs. Comparative quality performance review can be used to drive improvements in surgical outcomes and thereby reduce costs. We sought to determine the efficacy of a comparative quality performance review to improve a surgeon-level measure of surgical oncologic quality, that is the positive surgical margin rate at the time of radical prostatectomy. MATERIALS AND METHODS: Eight surgeons who performed consecutive radical prostatectomies at a single high volume institution between January 1, 2015 and December 31, 2015 were included in analysis. Individual surgeons were provided with confidential report cards every 6 months detailing their case mix, case volume and pT2 radical prostatectomy positive surgical margin rate relative to 1) their own self-matched data, 2) the de-identified data of their colleagues and 3) institutional aggregate data during the study period. Positive surgical margin rates were compared before and after intervention. Hierarchal logistic regression analysis was used to examine the association of study period on the odds of positive surgical margins, adjusted for prostate specific antigen level and National Comprehensive Cancer Network® risk group. RESULTS: Overall, 1,822 (1,392 before and 430 after intervention) radical prostatectomies were performed that met study inclusion criteria. The aggregate departmental unadjusted positive surgical margin rates were 10.6% and 7.4% in the pre-intervention and post-intervention groups, respectively. After adjusting for higher risk cancer in the post-intervention group, there was a significant protective association of post-intervention status on positive margins (OR 0.64, 95% CI 0.43-0.97, p = 0.03). All 5 surgeons with positive surgical margin rates higher than the aggregate department rate in the pre-intervention period showed improvement after intervention. CONCLUSIONS: Comparative quality performance review can be implemented at the surgeon level and can promote improvement in an objective measure of surgical oncology quality.

Utility of Risk Models in Decision Making After Radical Prostatectomy: Lessons from a Natural History Cohort of Intermediate- and High-Risk Men.

BACKGROUND: Current guidelines suggest adjuvant radiation therapy for men with adverse pathologic features (APFs) at radical prostatectomy (RP). We examine at-risk men treated only with RP until the time of metastasis. OBJECTIVE: To evaluate whether clinicopathologic risk models can help guide postoperative therapeutic decision making. DESIGN, SETTING, AND PARTICIPANTS: Men with National Comprehensive Cancer Network intermediate- or high-risk localized prostate cancer undergoing RP in the prostate-specific antigen (PSA) era were identified (n=3089). Only men with initial undetectable PSA after surgery and who received no therapy prior to metastasis were included. APFs were defined as pT3 disease or positive surgical margins. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: Area under the receiver operating characteristic curve (AUC) for time to event data was used to measure the discrimination performance of the risk factors. Cumulative incidence curves were constructed using Fine and Gray competing risks analysis to estimate the risk of biochemical recurrence (BCR) or metastasis, taking censoring and death due to other causes into consideration. RESULTS AND LIMITATIONS: Overall, 43% of the cohort (n=1327) had APFs at RP. Median follow-up for censored patients was 5 yr. Cumulative incidence of metastasis was 6% at 10 yr after RP for all patients. Cumulative incidence of metastasis among men with APFs was 7.5% at 10 yr after RP. Among men with BCR, the incidence of metastasis was 38% 5 yr after BCR. At 10 yr after RP, time-dependent AUC for predicting metastasis by Cancer of the Prostate Risk Assessment Postsurgical or Eggener risk models was 0.81 (95% confidence interval [CI], 0.72-0.97) and 0.78 (95% CI, 0.67-0.97) in the APF population, respectively. At 5 yr after BCR, these values were lower (0.58 [95% CI, 0.50-0.66] and 0.70 [95% CI, 0.63-0.76]) among those who developed BCR. Use of risk model cut points could substantially reduce overtreatment while minimally increasing undertreatment (ie, use of an Eggener cut point of 2.5% for treatment of men with APFs would spare 46% from treatment while only allowing for metastatic events in 1% at 10 yr after RP). CONCLUSIONS: Use of risk models reduces overtreatment and should be a routine part of patient counseling when considering adjuvant therapy. Risk model performance is significantly reduced among men with BCR. PATIENT SUMMARY: Use of current risk models can help guide decision making regarding therapy after surgery and reduce overtreatment.

Reclassification rates are higher among African American men than Caucasians on active surveillance.

OBJECTIVE: To evaluate the risk of reclassification on serial biopsy for Caucasian and African American (AA) men with very low-risk (VLR) prostate cancer enrolled in a large prospective active surveillance (AS) registry. METHODS: The Johns Hopkins AS registry is a prospective observational study that has enrolled 982 men since 1994. Including only men who met all National Comprehensive Cancer Network VLR criteria (clinical stage ≤T1, Gleason score ≤6, prostate-specific antigen [PSA] level <10 ng/mL, PSA density <0.15 ng/mL/cm(3), positive cores <3, percent cancer per core ≤50), we analyzed a cohort of 654 men (615 Caucasians and 39 AAs). The association of race with reclassification on serial biopsy was assessed with competing-risks regressions. RESULTS: AA men on AS were more likely than Caucasians to experience upgrading on serial biopsy (36% vs 16%; adjusted P <.001). Adjusting for PSA level, prostate size, volume of cancer on biopsy, treatment year, and body mass index, AA race was an independent predictor of biopsy reclassification (subdistribution hazard ratio, 1.8; P = .003). Examining specific modes of reclassification, AA race was independently associated with reclassification by grade (subdistribution hazard ratio, 3.0; P = .002) but not by volume. CONCLUSION: AA men with VLR prostate cancer followed on AS are at significantly higher risk of grade reclassification compared with Caucasians. Therefore, if the goal of AS is to selectively monitor men with low-grade disease, AA men may require alternate selection criteria.

Racial disparities in oncologic outcomes after radical prostatectomy: long-term follow-up.

OBJECTIVE: To report race-based outcomes after radical prostatectomy (RP) in a cohort stratified by National Comprehensive Cancer Network (NCCN) risk category with updated follow-up. MATERIALS AND METHODS: Studies describing racial disparities in outcomes after RP are conflicting. We studied 15,993 white and 1634 African American (AA) pretreatment-naïve men who underwent RP at our institution (1992-2013) with complete preoperative and pathologic data. Pathologic outcomes were compared between races using appropriate statistical tests; biochemical recurrence (BCR) for men with complete follow-up was compared using multivariate models that controlled separately for preoperative and postoperative covariates. RESULTS: Very low- and low-risk AA men were more likely to have positive surgical margins (P <.01), adverse pathologic features (P <.01), and be upgraded at RP (P <.01). With a median follow-up of 4.0 years after RP, AA race was an independent predictor of BCR among NCCN low-risk (HR, 2.16; P <.001) and intermediate-risk (hazard ratio [HR], 1.34; P = .024) classes and pathologic Gleason score ≤ 6 (HR, 2.42; P <.001) and Gleason score 7 (HR, 1.71; P <.001). BCR-free survival for very low-risk AA men was similar to low-risk white men (P = .890); BCR-free survival for low-risk AA men was similar to intermediate-risk white men (P = .060). CONCLUSION: When stratified by NCCN risk, AA men with very low-, low-, or intermediate-risk prostate cancer who undergo RP are more likely to have adverse pathologic findings and BCR compared with white men. AA men with "low risk" prostate cancer, especially those considering active surveillance, should be counseled that their recurrence risks can resemble those of whites in higher risk categories.

Pathological examination of radical prostatectomy specimens in men with very low risk disease at biopsy reveals distinct zonal distribution of cancer in black American men.

PURPOSE: Of men with very low risk prostate cancer at biopsy recent evidence shows that black American men are at greater risk for adverse oncologic outcomes after radical prostatectomy. We studied radical prostatectomy specimens from black and white men at very low risk to determine whether there are systematic pathological differences. MATERIALS AND METHODS: Radical prostatectomy specimens were evaluated in men with National Comprehensive Cancer Network® (NCCN) very low risk prostate cancer. At diagnosis all men underwent extended biopsy sampling (10 or more cores) and were treated in the modern Gleason grade era. We analyzed tumor volume, grade and location in 87 black and 89 white men. For each specimen the dominant nodule was defined as the largest tumor with the highest grade. RESULTS: Compared to white men, black men were more likely to have significant prostate cancer (61% vs 29%), Gleason 7 or greater (37% vs 11%, each p <0.001) and a volume of greater than 0.5 cm(3) (45% vs 21%, p = 0.001). Dominant nodules in black men were larger (median 0.28 vs 0.13 cm(3), p = 0.002) and more often anterior (51% vs 29%, p = 0.003). In men who underwent pathological upgrading the dominant nodule was also more frequently anterior in black than in white men (59% vs 0%, p = 0.001). CONCLUSIONS: Black men with very low risk prostate cancer at diagnosis have a significantly higher prevalence of anterior cancer foci that are of higher grade and larger volume. Enhanced imaging or anterior zone sampling may detect these significant anterior tumors, improving the outcome in black men considering active surveillance.

African American men with very low-risk prostate cancer exhibit adverse oncologic outcomes after radical prostatectomy: should active surveillance still be an option for them?

PURPOSE: Active surveillance (AS) is a treatment option for men with very low-risk prostate cancer (PCa); however, favorable outcomes achieved for men in AS are based on cohorts that under-represent African American (AA) men. To explore whether race-based health disparities exist among men with very low-risk PCa, we evaluated oncologic outcomes of AA men with very low-risk PCa who were candidates for AS but elected to undergo radical prostatectomy (RP). PATIENTS AND METHODS: We studied 1,801 men (256 AA, 1,473 white men, and 72 others) who met National Comprehensive Cancer Network criteria for very low-risk PCa and underwent RP. Presenting characteristics, pathologic data, and cancer recurrence were compared among the groups. Multivariable modeling was performed to assess the association of race with upgrading and adverse pathologic features. RESULTS: AA men with very low-risk PCa had more adverse pathologic features at RP and poorer oncologic outcomes. AA men were more likely to experience disease upgrading at prostatectomy (27.3% v 14.4%; P < .001), positive surgical margins (9.8% v 5.9%; P = .02), and higher Cancer of the Prostate Risk Assessment Post-Surgical scoring system (CAPRA-S) scores. On multivariable analysis, AA race was an independent predictor of adverse pathologic features (odds ratio, [OR], 3.23; P = .03) and pathologic upgrading (OR, 2.26; P = .03). CONCLUSION: AA men with very low-risk PCa who meet criteria for AS but undergo immediate surgery experience significantly higher rates of upgrading and adverse pathology than do white men and men of other races. AA men with very low-risk PCa should be counseled about increased oncologic risk when deciding among their disease management options.

Molecular effects of genistein on male urethral development.

PURPOSE: The increasing incidence of hypospadias is partly attributed to increased gestational exposure to endocrine disruptors. We investigated the effects of genistein, the primary phytoestrogen in soy, on the molecular program of male urethral development. MATERIALS AND METHODS: Female mice were fed diets supplemented with genistein (500 mg/kg diet) or control diets before breeding and throughout gestation. Urethras from embryonic day 17.5 male fetuses were harvested, and RNA was prepared, amplified, labeled and hybridized on whole genome microarrays. Data were analyzed using packages from the R/Bioconductor project. Immunohistochemical analysis and immunoblotting were used to confirm the activity of MAPK and the presence of Ntrk1 and Ntrk2 during urethral development. RESULTS: Gestational exposure to genistein altered the urethral expression of 277 genes (p <0.008). Among the most affected were hormonally regulated genes, including IGFBP-1, Kap and Rhox5. Differentially expressed genes were grouped into functional pathways of cell proliferation, adhesion, apoptosis and tube morphogenesis (p <0.0001), and were enriched for members of the MAPK (p <0.00001) and TGF-ß (p <0.01) signaling cascades. Differentially expressed genes preferentially contained ELK1, Myc/Max, FOXO, HOX and ER control elements. The MAPK pathway was active, and its upstream genistein affected tyrosine kinase receptors Ntrk1 and Ntrk2 were present in the developing male urethra. CONCLUSIONS: Gestational exposure to genistein contributes to hypospadias by altering pathways of tissue morphogenesis, cell proliferation and cell survival. In particular, genes in the MAPK and TGF-ß signaling pathways and those controlled by FOXO, HOX and ER transcription factors are disrupted.

Risk factors, prevention and early detection of prostate cancer.

Approximately one in six men in the United States will develop prostate cancer during their lifetime. Genetic and environmental variables play a role in determining prostate cancer risk. This article highlights the latest evidence regarding the risk factors for prostate cancer. The current screening strategies using prostate-specific antigen and digital rectal examination are also discussed, as well as the limitations of these protocols and potential methods for improving early detection.