Critical evaluation of artificial intelligence as a digital twin of pathologists for prostate cancer pathology.

Prostate cancer pathology plays a crucial role in clinical management but is time-consuming. Artificial intelligence (AI) shows promise in detecting prostate cancer and grading patterns. We tested an AI-based digital twin of a pathologist, vPatho, on 2603 histological images of prostate tissue stained with hematoxylin and eosin. We analyzed various factors influencing tumor grade discordance between the vPatho system and six human pathologists. Our results demonstrated that vPatho achieved comparable performance in prostate cancer detection and tumor volume estimation, as reported in the literature. The concordance levels between vPatho and human pathologists were examined. Notably, moderate to substantial agreement was observed in identifying complementary histological features such as ductal, cribriform, nerve, blood vessel, and lymphocyte infiltration. However, concordance in tumor grading decreased when applied to prostatectomy specimens (κ = 0.44) compared to biopsy cores (κ = 0.70). Adjusting the decision threshold for the secondary Gleason pattern from 5 to 10% improved the concordance level between pathologists and vPatho for tumor grading on prostatectomy specimens (κ from 0.44 to 0.64). Potential causes of grade discordance included the vertical extent of tumors toward the prostate boundary and the proportions of slides with prostate cancer. Gleason pattern 4 was particularly associated with this population. Notably, the grade according to vPatho was not specific to any of the six pathologists involved in routine clinical grading. In conclusion, our study highlights the potential utility of AI in developing a digital twin for a pathologist. This approach can help uncover limitations in AI adoption and the practical application of the current grading system for prostate cancer pathology.

Loss of PDE4D7 expression promotes androgen independence, neuroendocrine differentiation and alterations in DNA repair: implications for therapeutic strategies.

BACKGROUND: Androgen signalling remains the seminal therapeutic approach for the management of advanced prostate cancer. However, most tumours eventually shift towards an aggressive phenotype, characterised by androgen independence and treatment resistance. The cyclic adenosine monophosphate (cAMP) pathway plays a crucial role in regulating various cellular processes, with the phosphodiesterase PDE4D7 being a vital modulator of cAMP signalling in prostate cancer cells. METHODS: Using shRNA-mediated PDE4D7 knockdown in LNCaP cells and downstream analysis via RNA sequencing and phenotypic assays, we replicate clinical observations that diminished PDE4D7 expression promotes an aggressive prostate cancer phenotype. RESULTS: Our study provides evidence that loss of PDE4D7 expression represents a pivotal switch driving the transition from an androgen-sensitive state to hormone unresponsiveness and neuroendocrine differentiation. In addition, we demonstrate that PDE4D7 loss affects DNA repair pathways, conferring resistance to poly ADP ribose polymerase (PARP) inhibitors. CONCLUSION: Reinstating PDE4D7 expression sensitises prostate cancer cells to anti-androgens, DNA damage response inhibitors, and cytotoxic therapies. These findings provide significant insight into the regulatory role of PDE4D7 in the development of lethal prostate cancer and the potential of its modulation as a novel therapeutic strategy.

Efficient Augmented Intelligence Framework for Bladder Lesion Detection.

PURPOSE: Development of intelligence systems for bladder lesion detection is cost intensive. An efficient strategy to develop such intelligence solutions is needed. MATERIALS AND METHODS: We used four deep learning models (ConvNeXt, PlexusNet, MobileNet, and SwinTransformer) covering a variety of model complexity and efficacy. We trained these models on a previously published educational cystoscopy atlas (n = 312 images) to estimate the ratio between normal and cancer scores and externally validated on cystoscopy videos from 68 cases, with region of interest (ROI) pathologically confirmed to be benign and cancerous bladder lesions (ie, ROI). The performance measurement included specificity and sensitivity at frame level, frame sequence (block) level, and ROI level for each case. RESULTS: Specificity was comparable between four models at frame (range, 30.0%-44.8%) and block levels (56%-67%). Although sensitivity at the frame level (range, 81.4%-88.1%) differed between the models, sensitivity at the block level (100%) and ROI level (100%) was comparable between these models. MobileNet and PlexusNet were computationally more efficient for real-time ROI detection than ConvNeXt and SwinTransformer. CONCLUSION: Educational cystoscopy atlas and efficient models facilitate the development of real-time intelligence system for bladder lesion detection.

Active Surveillance for Men Younger than 60 Years or with Intermediate-risk Localized Prostate Cancer. Descriptive Analyses of Clinical Practice in the Movember GAP3 Initiative.

Background: Active surveillance (AS) is a management option for men diagnosed with low-risk prostate cancer. Opinions differ on whether it is safe to include young men (≤60 yr) or men with intermediate-risk disease. Objective: To assess whether reasons for discontinuation, treatment choice after AS, and adverse pathology at radical prostatectomy (RP; N1, or ≥GG3, or ≥pT3) differ for men ≤60 yr or those with European Association of Urology (EAU) intermediate-risk disease from those for men >60 yr or those with EAU low-risk disease. Design setting and participants: We analyzed data from 5411 men ≤60 yr and 14 959 men >60 yr, 14 064 men with low-risk cancer, and 2441 men with intermediate-risk cancer, originating from the GAP3 database (21 169 patients/27 cohorts worldwide). Outcome measurements and statistical analysis: Cumulative incidence curves were used to estimate the rates of AS discontinuation and treatment choice. Results and limitations: The probability of discontinuation of AS due to disease progression at 5 yr was similar for men aged ≤60 yr (22%) and those >60 yr (25%), as well as those of any age with low-risk disease (24%) versus those with intermediate-risk disease (24%). Men with intermediate-risk disease are more prone to discontinue AS without evidence of progression than men with low-risk disease (at 1/5 yr: 5.9%/14.2% vs 2.0%/8.8%). Adverse pathology at RP was observed in 32% of men ≤60 yr compared with 36% of men >60 yr (p = 0.029), and in 34% with low-risk disease compared with 40% with intermediate-risk disease (p = 0.048). Conclusions: Our descriptive analysis of AS practices worldwide showed that the risk of progression during AS is similar across the age and risk groups studied. The proportion of adverse pathology was higher among men >60 yr than among men ≤60 yr. These results suggest that men ≤60 yr and those with EAU intermediate-risk disease should not be excluded from opting for AS as initial management. Patient summary: Data from 27 international centers reflecting daily clinical practice suggest that younger men or men with intermediate-risk prostate cancer do not hold greater risk for disease progression during active surveillance.

Artificial Intelligence-Based Prognostic Model for Urologic Cancers: A SEER-Based Study.

BACKGROUND: Prognostication is essential to determine the risk profile of patients with urologic cancers. METHODS: We utilized the SEER national cancer registry database with approximately 2 million patients diagnosed with urologic cancers (penile, testicular, prostate, bladder, ureter, and kidney). The cohort was randomly divided into the development set (90%) and the out-held test set (10%). Modeling algorithms and clinically relevant parameters were utilized for cancer-specific mortality prognosis. The model fitness for the survival estimation was assessed using the differences between the predicted and observed Kaplan-Meier estimates on the out-held test set. The overall concordance index (c-index) score estimated the discriminative accuracy of the survival model on the test set. A simulation study assessed the estimated minimum follow-up duration and time points with the risk stability. RESULTS: We achieved a well-calibrated prognostic model with an overall c-index score of 0.800 (95% CI: 0.795-0.805) on the representative out-held test set. The simulation study revealed that the suggestions for the follow-up duration covered the minimum duration and differed by the tumor dissemination stages and affected organs. Time points with a high likelihood for risk stability were identifiable. CONCLUSIONS: A personalized temporal survival estimation is feasible using artificial intelligence and has potential application in clinical settings, including surveillance management.

Expression Patterns and Corepressor Function of Retinoic Acid-induced 2 in Prostate Cancer.

BACKGROUND: Revealing molecular mechanisms linked to androgen receptor activity can help to improve diagnosis and treatment of prostate cancer. Retinoic acid-induced 2 (RAI2) protein is thought to act as a transcriptional coregulator involved in hormonal responses and epithelial differentiation. We evaluated the clinical relevance and biological function of the RAI2 protein in prostate cancer. METHODS: We assessed RAI2 gene expression in the Cancer Genome Atlas prostate adenocarcinoma PanCancer cohort and protein expression in primary tumors (n = 199) by immunohistochemistry. We studied RAI2 gene expression as part of a multimarker panel in an enriched circulating tumor cell population isolated from blood samples (n = 38) of patients with metastatic prostate cancer. In prostate cancer cell lines, we analyzed the consequences of androgen receptor inhibition on RAI2 protein expression and the consequences of RAI2 depletion on the expression of the androgen receptor and selected target genes. RESULTS: Abundance of the RAI2 protein in adenocarcinomas correlated with the androgen receptor; keratins 8, 18, and 19; and E-cadherin as well as with an early biochemical recurrence. In circulating tumor cells, detection of RAI2 mRNA significantly correlated with gene expression of FOLH1, KLK3, RAI2, AR, and AR-V7. In VCaP and LNCaP cell lines, sustained inhibition of hormone receptor activity induced the RAI2 protein, whereas RAI2 depletion augmented the expression of MME, STEAP4, and WIPI1. CONCLUSIONS: The RAI2 protein functions as a transcriptional coregulator of the androgen response in prostate cancer cells. Detection of RAI2 gene expression in blood samples from patients with metastatic prostate cancer indicated the presence of circulating tumor cells.

The prognostic potential of alkaline phosphatase and lactic acid dehydrogenase in bmCRPC patients without significant PSA response under enzalutamide.

BACKGROUND: In patients with bone metastatic castration-resistant prostate cancer (bmCRPC) on systemic treatment, it is difficult to differentiate between continuous rise of prostate specific antigen (PSA) representing progression, and PSA-surge, which is followed by clinical response or stable disease. The purpose of this study was to evaluate the prognostic value of dynamic changes of alkaline phosphatase (ALP) and lactic acid dehydrogenase (LDH) levels as a predictor of clinical efficacy or therapeutic resistance of patients who do not show a sufficient initial PSA decline of ≥50% from baseline during early therapy with Enzalutamide. METHODS: Forty-eight men with bmCRPC on Enzalutamide 07/2010-09/2019 with initially rising PSA were analyzed. We monitored PSA, LDH and ALP at week 0, 2, 4, and every 4 weeks thereafter and analyzed the correlation between ALP rising at 12 weeks with or without LDH-normalization and the association with survival. For this we used Kaplan Meier analysis and uni- and multivariate cox-regression models. RESULTS: In Kaplan-Meier analysis, ALP rising at 12 weeks with or without LDH-normalization was associated with significantly worse median progression-free survival (PFS) of 3 months vs. 5 months (Log rank P = 0.02) and 3 months vs. 5 months (P = 0.01), respectively and overall survival (OS) with 8 months vs. 15 months (P = 0.02) and 8 months vs. 17 months (P < 0.01). In univariate analysis of PFS, ALP rising at 12 weeks alone, ALP rising at 12 weeks without LDH-normalization and application of Enzalutamide after chemotherapy showed a statistically significant association towards shorter PFS (hazard ratio (HR): 0.51, P = 0.04; HR: 0.48, P = 0.03; HR: 0.48, P = 0.03). Worse OS was significantly associated with ALP rising at 12 weeks alone, ALP rising at 12 weeks without LDH-normalization, and application of Enzalutamide after chemotherapy (HR: 0.47, P = 0.02; HR: 0.36, P < 0.01; HR: 0.31, P < 0.01). In multivariate analysis only the application of Enzalutamide after chemotherapy remained an independent prognostic factor for worse OS (HR: 0.36, P = 0.01). CONCLUSIONS: Dynamic changes of ALP (non-rise) and LDH (normalization) under therapy with Enzalutamide may be associated with clinical benefit, better PFS, and OS in patients with bmCRPC who do not show a PSA decline.

A Multivariable Approach Using Magnetic Resonance Imaging to Avoid a Protocol-based Prostate Biopsy in Men on Active Surveillance for Prostate Cancer-Data from the International Multicenter Prospective PRIAS Study.

BACKGROUND: There is ongoing discussion whether a multivariable approach including magnetic resonance imaging (MRI) can safely prevent unnecessary protocol-advised repeat biopsy during active surveillance (AS). OBJECTIVE: To determine predictors for grade group (GG) reclassification in patients undergoing an MRI-informed prostate biopsy (MRI-Bx) during AS and to evaluate whether a confirmatory biopsy can be omitted in patients diagnosed with upfront MRI. DESIGN, SETTING, AND PARTICIPANTS: The Prostate cancer Research International: Active Surveillance (PRIAS) study is a multicenter prospective study of patients on AS (www.prias-project.org). We selected all patients undergoing MRI-Bx (targeted ± systematic biopsy) during AS. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: A time-dependent Cox regression analysis was used to determine the predictors of GG progression/reclassification in patients undergoing MRI-Bx. A sensitivity analysis and a multivariable logistic regression analysis were also performed. RESULTS AND LIMITATIONS: A total of 1185 patients underwent 1488 MRI-Bx sessions. The time-dependent Cox regression analysis showed that age (per 10 yr, hazard ratio [HR] 0.84 [95% confidence interval {CI} 0.71-0.99]), MRI outcome (Prostate Imaging Reporting and Data System [PIRADS] 3 vs negative HR 2.46 [95% CI 1.56-3.88], PIRADS 4 vs negative HR 3.39 [95% CI 2.28-5.05], and PIRADS 5 vs negative HR 4.95 [95% CI 3.25-7.56]), prostate-specific antigen (PSA) density (per 0.1 ng/ml cm3, HR 1.20 [95% CI 1.12-1.30]), and percentage positive cores on the last systematic biopsy (per 10%, HR 1.16 [95% CI 1.10-1.23]) were significant predictors of GG reclassification. Of the patients with negative MRI and a PSA density of <0.15 ng/ml cm3 (n = 315), 3% were reclassified to GG ≥2 and 0.6% to GG ≥3. At the confirmatory biopsy, reclassification to GG ≥2 and ≥3 was observed in 23% and 7% of the patients diagnosed without upfront MRI and in 19% and 6% of the patients diagnosed with upfront MRI, respectively. The multivariable analysis showed no significant difference in upgrading at the confirmatory biopsy between patients diagnosed with or without upfront MRI. CONCLUSIONS: Age, MRI outcome, PSA density, and percentage positive cores are significant predictors of reclassification at an MRI-informed biopsy. Patients with negative MRI and a PSA density of <0.15 ng/ml cm3 can safely omit a protocol-based prostate biopsy, whereas in other patients, a multivariable approach is advised. Being diagnosed with upfront MRI appears not to significantly affect reclassification risk; hence, a confirmatory MRI-Bx cannot totally be omitted yet. PATIENT SUMMARY: A protocol-based prostate biopsy while on active surveillance can be omitted in patients with negative magnetic resonance imaging (MRI) and prostate-specific antigen density <0.15 ng/ml cm3. A confirmatory biopsy cannot simply be omitted in all patients diagnosed with upfront MRI.

PROPOSe: A Real-life Prospective Study of Proclarix, a Novel Blood-based Test to Support Challenging Biopsy Decision-making in Prostate Cancer.

BACKGROUND: Prostate-specific antigen (PSA)-based detection of prostate cancer (PCa) often leads to negative biopsy results or detection of clinically insignificant PCa, more frequently in the PSA range of 2-10 ng/ml, in men with increased prostate volume and normal digital rectal examination (DRE). OBJECTIVE: This study evaluated the accuracy of Proclarix, a novel blood-based diagnostic test, to help in biopsy decision-making in this challenging patient population. DESIGN, SETTING, AND PARTICIPANTS: Ten clinical sites prospectively enrolled 457 men presenting for prostate biopsy with PSA between 2 and 10 ng/ml, normal DRE, and prostate volume ≥35 cm3. Transrectal ultrasound-guided and multiparametric magnetic resonance imaging (mpMRI)-guided biopsy techniques were allowed. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: Serum samples were tested blindly at the end of the study. Diagnostic performance of Proclarix risk score was established in correlation to systematic biopsy outcome and its performance compared with %free PSA (%fPSA) and the European Randomised Study of Screening for Prostate Cancer (ERSPC) risk calculator (RC) as well as Proclarix density compared with PSA density in men undergoing mpMRI. RESULTS AND LIMITATIONS: The sensitivity of Proclarix risk score for clinically significant PCa (csPCa) defined as grade group (GG) ≥2 was 91% (n = 362), with higher specificity than both %fPSA (22% vs 14%; difference = 8% [95% confidence interval {CI}, 2.6-14%], p = 0.005) and RC (22% vs 15%; difference = 7% [95% CI, 0.7-12%], p = 0.028). In the subset of men undergoing mpMRI-fusion biopsy (n = 121), the specificity of Proclarix risk score was significantly higher than PSA density (26% vs 8%; difference = 18% [95% CI, 7-28%], p < 0.001), and at equal sensitivity of 97%, Proclarix density had an even higher specificity of 33% [95% CI, 23-43%]. CONCLUSIONS: In a routine use setting, Proclarix accurately discriminated csPCa from no or insignificant PCa in the most challenging patients. Proclarix represents a valuable rule-out test in the diagnostic algorithm for PCa, alone or in combination with mpMRI. PATIENT SUMMARY: Proclarix is a novel blood-based test with the potential to accurately rule out clinically significant prostate cancer, and therefore to reduce the number of unneeded biopsies.

The CAPRA&PDE4D5/7/9 Prognostic Model Is Significantly Associated with Adverse Post-Surgical Pathology Outcomes.

Objectives: To investigate the association of the prognostic risk score CAPRA&PDE4D5/7/9 as measured on pre-surgical diagnostic needle biopsy tissue with pathological outcomes after radical prostatectomies in a clinically low−intermediate-risk patient cohort. Patients and Methods: RNA was extracted from biopsy punches of diagnostic needle biopsies. The patient cohort comprises n = 151 patients; of those n = 84 had low−intermediate clinical risk based on the CAPRA score and DRE clinical stage 2, or pathological pT stage > pT3a, or tumor penetrated prostate capsular status, or pN1 disease); (ii) any ISUP pathological Gleason >2; (iii) any ISUP pathological Gleason >1. In the n = 84 patients with low to intermediate clinical risk profiles, the clinical-genomics CAPRA&PDE4D5/7/9_BCR risk score was significantly lower in patients with favorable vs. unfavorable outcomes. In univariable logistic regression modeling the genomics PDE4D5/7/9_BCR as well as the clinical-genomics CAPRA&PDE4D5/7/9_BCR combination model were significantly associated with all three post-surgical pathology outcomes (p = 0.02, p = 0.0004, p = 0.04; and p = 0.01, p = 0.0002, p = 0.01, respectively). The clinically used PRIAS criteria for the selection of low-risk candidate patients for active surveillance (AS) were not significantly associated with any of the three tested post-operative pathology outcomes (p = 0.3, p = 0.1, p = 0.1, respectively). In multivariable analysis adjusted for the CAPRA score, the genomics PDE4D5/7/9_BCR risk score remained significant for the outcomes of adverse pathology (p = 0.04) and ISUP pathological Gleason >2 (p = 0.004). The negative predictive value of the CAPRA&PDE4D5/7/9_BCR risk score using the low-risk cut-off (0.1) for the three pathological endpoints was 82.0%, 100%, and 59.1%, respectively for a selected low-risk cohort of n = 22 patients (26.2% of the entire cohort) compared to 72.1%, 94.4%, and 55.6% for n = 18 low-risk patients (21.4% of the total cohort) selected based on the PRIAS inclusion criteria. Conclusion: In this study, we have shown that the previously reported clinical-genomics prostate cancer risk model CAPRA&PDE4D5/7/9_BCR which was developed to predict biological outcomes after surgery of primary prostate cancer is also significantly associated with post-surgical pathology outcomes. The risk score predicts adverse pathology independent of the clinical risk metrics. Compared to clinically used active surveillance inclusion criteria, the clinical-genomics CAPRA&PDE4D5/7/9_BCR risk model selects 22% (n = 8) more low-risk patients with higher negative predictive value to experience unfavorable post-operative pathology outcomes.

Decision aid and cost compensation influence uptake of PSA-based early detection without affecting decisional conflict: a cluster randomised trial.

International guidelines recommend to inform men about the benefits and harms of prostate specific antigen (PSA) based early detection of prostate cancer. This study investigates the influence of a transactional decision aid (DA) or cost compensation (CC) for a PSA test on the decisional behaviour of men. Prospective, cluster-randomised trial to compare two interventions in a 2 × 2 factorial design: DA versus counselling as usual, and CC versus noCC for PSA-testing. 90 cluster-randomised physicians in the administrative district of Muenster, Germany recruited 962 participants aged 55-69 yrs. in 2018. Primary endpoint: the influence of the DA and CC on the decisional conflict. Secondary endpoints: factors which altered the involvement of the men regarding their decision to take a PSA-test. The primary endpoint was analysed by a multivariate regression model. The choice to take the PSA test was increased by CC and reduced by the DA, the latter also reduced PSA uptake in men who were offered CC. The DA led to an increase of the median knowledge about early detection, changed willingness to perform a PSA test without increasing the level of shared decision, giving participants a stronger feeling of having made the decision by themselves. The DA did not alter the decisional conflict, as it was very low in all study groups. DA reduced and CC increased the PSA uptake. The DA seemed to have a greater impact on the participants than CC, as it led to fewer PSA tests even if CC was granted.Trial registration: German Clinical Trial Register (Deutsches Register Klinischer Studien DRKS00007687). Registered: 06/05/2015. https://www.drks.de/drks_web/navigate.do?navigationId=trial.HTML&TRIAL_ID=DRKS00007687 .

Low Tumor-to-Stroma Ratio Reflects Protective Role of Stroma against Prostate Cancer Progression.

Tumor-to-stroma ratio (TSR) is a prognostic factor that expresses the relative amounts of tumor and intratumoral stroma. In this study, its clinical and molecular relevance was evaluated in prostate cancer (PCa). The feasibility of automated quantification was tested in digital scans of tissue microarrays containing 128 primary tumors from 72 PCa patients stained immunohistochemically for epithelial cell adhesion molecule (EpCAM), followed by validation in a cohort of 310 primary tumors from 209 PCa patients. In order to investigate the gene expression differences between tumors with low and high TSR, we applied multigene expression analysis (nCounter® PanCancer Progression Panel, NanoString) of 42 tissue samples. TSR scores were categorized into low (<1 TSR) and high (≥1 TSR). In the pilot cohort, 31 patients (43.1%) were categorized as low and 41 (56.9%) as high TSR score, whereas 48 (23.0%) patients from the validation cohort were classified as low TSR and 161 (77.0%) as high. In both cohorts, high TSR appeared to indicate the shorter time to biochemical recurrence in PCa patients (Log-rank test, p = 0.04 and p = 0.01 for the pilot and validation cohort, respectively). Additionally, in the multivariate analysis of the validation cohort, TSR predicted BR independent of other factors, i.e., pT, pN, and age (p = 0.04, HR 2.75, 95%CI 1.07-7.03). Our data revealed that tumors categorized into low and high TSR score show differential expression of various genes; the genes upregulated in tumors with low TSR score were mostly associated with extracellular matrix and cell adhesion regulation. Taken together, this study shows that high stroma content can play a protective role in PCa. Automatic EpCAM-based quantification of TSR might improve prognostication in personalized medicine for PCa.

ALP bouncing and LDH normalization in bone metastatic castration-resistant prostate cancer patients under therapy with Enzalutamide: an exploratory analysis.

BACKGROUND: In bone metastatic castration-resistant prostate cancer (bmCRPC) treated with Enzalutamide commonly used prostate-specific antigen (PSA) can be misleading since initial PSA-flares may occur. In other therapies, bouncing of alkaline phosphatase (ALP-bouncing) was shown to be a promising surrogate for survival outcome. Low lactate dehydrogenase (LDH) is usually associated with better outcome. We evaluated the prognostic ability of ALP-bouncing, LDH, PSA, and the combination of these markers after initiation of Enzalutamide. METHODS: Eighty-nine patients with bmCRPC and dynamic changes of PSA, LDH and ALP were analyzed. ALP-bouncing, an increase after therapy start followed by a decline below baseline during the first 8 weeks, LDH-normalization and PSA-decline were analyzed regarding their association with survival using Kaplan-Meier analyses and uni- and multivariate (UV and MV) Cox-regression models. RESULTS: In Kaplan-Meier analysis a PSA-decline >50%, LDH-normalization and ALP-bouncing were associated with longer median progression-free survival (PFS) with 7 [95% confidence interval (CI): 4.2-9.8] vs. 3 (2.3-3.7) months for PSA-decline (log-rank P<0.01), 6 (4.1-8) vs. 2 (1.2-2.8) for LDH-normalization (P<0.01) and 8 (0-16.3) vs. 3 (1.9-4.1) for ALP-bouncing (P=0.01). Analysis of overall survival (OS) showed similar, not for all parameters significant, results with 17 (11.7-22.3) vs. 12 (7.0-17.1) months for PSA (P=0.35), 17 (13.2-20.8) vs. 7 (5.8-8.2) for LDH-normalization (P<0.01) and 19 (7.9-30.1) vs. 12 (7.7-16.3) for ALP-bouncing (P=0.32). In UV analysis, ALP-bouncing [hazard ratio (HR): 0.5 (0.3-1.0); P=0.02], PSA-decline >50% [HR: 0.5 (0.3-0.7); P<0.01] and LDH-normalization [HR: 0.4 (0.2-0.6); P<0.01] were significantly associated with longer PFS. For OS, LDH-normalization significantly prognosticated longer survival [HR: 0.4 (0.2-0.6); P<0.01]. In MV analysis, LDH-normalization was associated with a trend towards better OS [HR: 0.5 (0.2-1.1); P=0.09]. Comparing ALP-bouncing, LDH-normalization and PSA-decline with a PSA-decline alone, Kaplan-Meier analysis showed significantly longer PFS [11 (0.2-21.8) vs. 4 (0-8.6); P=0.01] and OS [20 (17.7-22.3) vs. 8 (0.3-15.7); P=0.02] in favor of the group presenting with the beneficial dynamics of all three markers. In UV analysis, the presence of favorable changes in the three markers was significantly associated with longer PFS [HR: 0.2 (0.1-0.7); P<0.01] and OS [HR: 0.3 (0.1-0.8); P=0.02]. CONCLUSIONS: ALP-bouncing and LDH-normalization may add to identification of bmCRPC-patients with favorable prognosis under Enzalutamide.

Specific spatial distribution patterns of tumor foci are associated with a low risk of biochemical recurrence in pT2pN0R0 prostate cancer.

BACKGROUND: The previous attempts for pT2 substaging of prostate cancer (PCa) were insufficient in providing prognostic subgroups and the search for new prognostic parameters to subcategorize pT2 PCa is, therefore, needed. Therefore, the current study investigated the association between tumor distribution patterns and the biochemical recurrence (BCR)-free survival rate in pT2pN0R0 PCa. METHODS: Following radical prostatectomy, the anatomical distribution of PCa in 743 men with pT1-pT3pN0 disease was analyzed to determine 20 types of PCa distribution patterns. Then, 245 men with pT2pN0R0 PCa was considered for prognostic evaluation with a mean follow-up period of 60 months. The spatial distribution patterns of PCa were evaluated using a cMDX©-based map model of the prostate. An analysis including 552,049 comparison operations was performed to assist in the evaluation of the similarity levels of the distribution patterns. A k-mean cluster analysis was applied to determine groups with similar distribution patterns. A decision-tree analysis was performed to divide these groups according to frequency of BCR. The BCR-free survival rate was analyzed using Kaplan-Meier curves. Predictors of progression were investigated using a Cox proportional hazards model. RESULTS: BCR occurred in 8.2% of the 245 men with pT2pN0R0 PCa. The median time of recurrence was 60 months (interquartile range [IQR]: 42-77). In univariate and multivariate analyses, the prostate volume and the distribution patterns were independent predictors for BCR, whereas the sub-staging of pT2 tumors, Gleason grading, prostate-specific antigen (PSA) level, and relative tumor volume were not. In the patients with pT2pN0R0 disease, PCa distribution patterns with the apical involvement were significantly associated with the risk of BCR (P = 0.001). CONCLUSION: The spread tumor patterns with the apical involvement are associated with a high-risk of BCR in the pT2 tumor stage. The vertical tumor spread could be considered in developing improved prognostic pT2 sub-categories.

EGFR as a stable marker of prostate cancer dissemination to bones.

BACKGROUND: Prostate cancer (PCa) is among the most commonly diagnosed malignancies in men. Although 5-year survival in patients with localised disease reaches nearly 100%, metastatic disease still remains incurable. Therefore, there is a need for markers indicating metastatic dissemination. METHODS: EGFR overexpression (EGFRover) was tracked in 1039 primary tumours, circulating tumour cells from 39 d'Amico high-risk patients and metastatic samples from 21 castration-resistant PCa cases. EGFR status was compared to clinical parameters and multiple molecular factors were assessed using immunohistochemistry and gene ontology analysis. The functional aspect of EGFR was evaluated by plating PC-3 cells on soft and rigid matrices. RESULTS: EGFRover was found in 14% of primary tumours, where it was associated with shorter metastasis-free survival and was an independent indicator of worse overall survival. EGFRover correlated with a pro-migratory and pro-metastatic phenotype of tumour cells as well as rich collagen fibre content. All circulating tumour cells (detected in 13% of cases) were positive for EGFR, independent of their EMT-related phenotype. EGFRover was more prevalent in castration-resistant bone metastases (29% of patients) and supported growth of human PCa cells on rigid matrices mimicking bone stiffness. CONCLUSIONS: EGFRover is a stable, EMT-independent marker of PCa disseminating to rigid organs, preferentially bones.

Combination possibility and deep learning model as clinical decision-aided approach for prostate cancer.

This study aims to introduce as proof of concept a combination model for classification of prostate cancer using deep learning approaches. We utilized patients with prostate cancer who underwent surgical treatment representing the various conditions of disease progression. All possible combinations of significant variables from logistic regression and correlation analyses were determined from study data sets. The combination possibility and deep learning model was developed to predict these combinations that represented clinically meaningful patient's subgroups. The observed relative frequencies of different tumor stages and Gleason score Gls changes from biopsy to prostatectomy were available for each group. Deep learning models and seven machine learning approaches were compared for the classification performance of Gleason score changes and pT2 stage. Deep models achieved the highest F1 scores by pT2 tumors (0.849) and Gls change (0.574). Combination possibility and deep learning model is a useful decision-aided tool for prostate cancer and to group patients with prostate cancer into clinically meaningful groups.

Low Numbers of Vascular Vessels Correlate to Progression in Hormone-Naïve Prostate Carcinomas Undergoing Radical Prostatectomy.

Vascularization influences tumor development by supporting the nutrition and dissemination of tumor cells. On the other hand, a low number of vascular vessels (VVlow) may induce hypoxia, accounting for selection of resistant clone(s) of tumor cells. This study aimed to evaluate the prognostic significance of vascular (VV) and lymphatic vessels (LV) in prostate cancer (PCa). Tumor samples from 400 PCa patients undergoing radical prostatectomy (RP) were prepared in duplex as tissue microarrays. Numbers of VV and LV were evaluated using immunohistochemistry detecting CD34 and podoplanin, respectively, and correlated to clinical data, biochemical recurrence (BR), and proteins analyzed in tumor cells. VVlow and LV were found in 32% and 43% of patients with informative PCa samples, respectively. VVlow correlated with a shorter time to BR 3, 5, and 10 years after RP in hormone-naïve patients (p = 0.028, p = 0.027 and p = 0.056, respectively). It was also shown to be an independent prognostic factor 5 years after surgery (multivariate analysis, p = 0.046). Tumors characterized by VVlow expressed the epithelial cell adhesion molecule, EpCAM, less frequently (p = 0.016) and revealed a borderline correlation to increased levels of tumor cell invasion marker Loxl-2 (p = 0.059). No correlations were found for LV. In summary, VVlow in hormone-naïve patients undergoing RP has prognostic potential and seems to be related to an aggressive phenotype of tumor cells.

The Association of the Long Prostate Cancer Expressed PDE4D Transcripts to Poor Patient Outcome Depends on the Tumour's TMPRSS2-ERG Fusion Status.

OBJECTIVES: To investigate the added value of assessing transcripts for the long cAMP phosphodiesterase-4D (PDE4D) isoforms, PDE4D5 and PDE4D9, regarding the prognostic power of the 'CAPRA & PDE4D7' combination risk model to predict longitudinal postsurgical biological outcomes in prostate cancer. PATIENTS AND METHODS: RNA was extracted from both biopsy punches of resected tumours (606 patients; RP cohort) and diagnostic needle biopsies (168 patients; DB cohort). RT-qPCR was performed in order to determine PDE4D5, PDE4D7, and PDE4D9 transcript scores in both study cohorts. By RNA sequencing, we determined the TMPRSS2-ERG fusion status of each tumour sample in the RP cohort. Kaplan-Meier survival analyses were then applied to correlate the PDE4D5, PDE4D7 and PDE4D9 scores with postsurgical patient outcomes. Logistic regression was then used to combine the clinical CAPRA score with PDE4D5, PDE4D7, and PDE4D9 scores in order to build a 'CAPRA & PDE4D5/7/9' regression model. ROC and decision curve analysis was used to estimate the net benefit of the 'CAPRA & PDE4D5/7/9' risk model. RESULTS: Kaplan-Meier survival analysis, on the RP cohort, revealed a significant association of the PDE4D7 score with postsurgical biochemical recurrence (BCR) in the presence of the TMPRSS2-ERG gene rearrangement (logrank p<0.0001), compared to the absence of this gene fusion event (logrank p=0.08). In contrast, the PDE4D5 score was only significantly associated with BCR in TMPRSS2-ERG fusion negative tumours (logrank p<0.0001 vs. logrank p=0.4 for TMPRSS2-ERG+ tumours). This was similar for the PDE4D9 score although less pronounced compared to that of the PDE4D5 score (TMPRSS2ERG- logrank p<0.0001 vs. TMPRSS2ERG+ logrank p<0.005). In order to predict BCR after primary treatment, we undertook ROC analysis of the logistic regression combination model of the CAPRA score with the PDE4D5, PDE4D7, and PDE4D9 scores. For the DB cohort, this demonstrated significant differences in the AUC between the CAPRA and the PDE4D5/7/9 regression model vs. the CAPRA and PDE4D7 risk model (AUC 0.87 vs. 0.82; p=0.049) vs. the CAPRA score alone (AUC 0.87 vs. 0.77; p=0.005). The CAPRA and PDE4D5/7/9 risk model stratified 19.2% patients of the DB cohort to either 'no risk of biochemical relapse' (NPV 100%) or the 'start of any secondary treatment (NPV 100%)', over a follow-up period of up to 15 years. Decision curve analysis presented a clear, net benefit for the use of the novel CAPRA & PDE4D5/7/9 risk model compared to the clinical CAPRA score alone or the CAPRA and PDE4D7 model across all decision thresholds. CONCLUSION: Association of the long PDE4D5, PDE4D7, and PDE4D9 transcript scores to prostate cancer patient outcome, after primary intervention, varies in opposite directions depending on the TMPRSS2-ERG genomic fusion background of the tumour. Adding transcript scores for the long PDE4D isoforms, PDE4D5 and PDE4D9, to our previously presented combination risk model of the combined 'CAPRA & PDE4D7' score, in order to generate the CAPRA and PDE4D5/7/9 score, significantly improves the prognostic power of the model in predicting postsurgical biological outcomes in prostate cancer patients.

Development and Prospective Randomized Evaluation of a Decision Aid for Prostate-specific Antigen-based Early Detection of Prostate Cancer in Men Aged Between 55 and 69Yr: The PSAInForm Trial.

For men interested in early detection of prostate cancer, the potential impact on decisional conflict of a decision aid with or without cost compensation for the prostate-specific antigen test will be investigated.