Genomic biomarkers in prostate cancer

Can we predict pathology without surgery? Weighing the added value of multiparametric MRI and whole prostate radiomics in integrative machine learning models

OBJECTIVE

To test the ability of high-performance machine learning (ML) models employing clinical, radiological, and radiomic variables to improve non-invasive prediction of the pathological status of prostate cancer (PCa) in a large, single-institution cohort.

METHODS

Patients who underwent multiparametric MRI and prostatectomy in our institution in 2015-2018 were considered; a total of 949 patients were included. Gradient-boosted decision tree models were separately trained using clinical features alone and in combination with radiological reporting and/or prostate radiomic features to predict pathological T, pathological N, ISUP score, and their change from preclinical assessment. Model behavior was analyzed in terms of performance, feature importance, Shapley additive explanation (SHAP) values, and mean absolute error (MAE). The best model was compared against a naïve model mimicking clinical workflow.

RESULTS

The model including all variables was the best performing (AUC values ranging from 0.73 to 0.96 for the six endpoints). Radiomic features brought a small yet measurable boost in performance, with the SHAP values indicating that their contribution can be critical to successful prediction of endpoints for individual patients. MAEs were lower for low-risk patients, suggesting that the models find them easier to classify. The best model outperformed (p ≤ 0.0001) clinical baseline, resulting in significantly fewer false negative predictions and overall was less prone to under-staging.

CONCLUSIONS

Our results highlight the potential benefit of integrative ML models for pathological status prediction in PCa. Additional studies regarding clinical integration of such models can provide valuable information for personalizing therapy offering a tool to improve non-invasive prediction of pathological status.

CLINICAL RELEVANCE STATEMENT

The best machine learning model was less prone to under-staging of the disease. The improved accuracy of our pathological prediction models could constitute an asset to the clinical workflow by providing clinicians with accurate pathological predictions prior to treatment.

KEY POINTS

• Currently, the most common strategies for pre-surgical stratification of prostate cancer (PCa) patients have shown to have suboptimal performances. • The addition of radiological features to the clinical features gave a considerable boost in model performance. Our best model outperforms the naïve model, avoiding under-staging and resulting in a critical advantage in the clinic. •Machine learning models incorporating clinical, radiological, and radiomics features significantly improved accuracy of pathological prediction in prostate cancer, possibly constituting an asset to the clinical workflow.

Prostate cancer biomarkers: from early diagnosis to precision treatment

Prostate cancer (PCa) is the second most prevalent cancer in men. In 2020, approximately 1,414,259 new cases were reported that accounted for 3,75,324 deaths (Sung et al. in CA 71:209-249, 2021). PCa is often asymptomatic at early stages; hence, routine screening and monitoring based on reliable biomarkers is crucial for early detection and assessment of cancer progression. Early diagnosis of disease is key step in reducing PCa-induced mortality. Biomarkers such as PSA have played vital role in reducing recent PCa deaths. Recent research has identified many other biomarkers and also refined PSA-based tests for non-invasive diagnosis of PCa in patients. Despite progress in screening methods, an important issue that influences treatment is heterogeneity of the cancer in different individuals, necessitating personalized treatment. Currently, focus is to identify biomarkers that can accurately diagnose PCa at early stage, indicate the stage of the disease, metastatic nature and chances of survival based on individual patient profile (Fig. 1). Fig. 1 Graphical abstract.

Artificial intelligence applications in prostate cancer

Artificial intelligence (AI) applications have enabled remarkable advancements in healthcare delivery. These AI tools are often aimed to improve accuracy and efficiency of histopathology assessment and diagnostic imaging interpretation, risk stratification (i.e., prognostication), and prediction of therapeutic benefit for personalized treatment recommendations. To date, multiple AI algorithms have been explored for prostate cancer to address automation of clinical workflow, integration of data from multiple domains in the decision-making process, and the generation of diagnostic, prognostic, and predictive biomarkers. While many studies remain within the pre-clinical space or lack validation, the last few years have witnessed the emergence of robust AI-based biomarkers validated on thousands of patients, and the prospective deployment of clinically-integrated workflows for automated radiation therapy design. To advance the field forward, multi-institutional and multi-disciplinary collaborations are needed in order to prospectively implement interoperable and accountable AI technology routinely in clinic.

Artificial Intelligence and Pathomics: Prostate Cancer

Artificial intelligence (AI) has the potential to transform pathologic diagnosis and cancer patient management as a predictive and prognostic biomarker. AI-based systems can be used to examine digitally scanned histopathology slides and differentiate benign from malignant cells and low from high grade. Deep learning models can analyze patient data from individual or multimodal combinations and identify patterns to be used to predict the response to different therapeutic options, the risk of recurrence or progression, and the prognosis of the newly diagnosed patient. AI-based models will improve treatment planning for patients with prostate cancer and improve the efficiency and cost-effectiveness of the pathology laboratory.

Advancements in Biomarkers of Prostate Cancer: A Review

Prostate cancer (PCa) is one of the most prevalent and deadly cancers among men, particularly affecting men of African descent and contributing significantly to cancer-related morbidity and mortality worldwide. The disease varies widely, from slow-developing forms to highly aggressive or potentially fatal variants. Accurate risk stratification is crucial for making therapeutic decisions and designing adequate clinical trials. This review assesses a broad spectrum of diagnostic and prognostic biomarkers, many of which are incorporated into clinical guidelines, including the Prostate Health Index (PHI), 4Kscore, STHLM3, PCA3, SelectMDx, ExoDx Prostate Intelliscore (EPI), and MiPS. It also highlights emerging biomarkers with preclinical support, such as urinary non-coding RNAs and DNA methylation patterns. Additionally, the review explores the role of tumor-associated microbiota in PCa, offering new insights into its potential contributions to disease understanding. By examining the latest advancements in PCa biomarkers, this review enhances understanding their roles in disease management.

Presurgical 68Ga-PSMA-11 Positron Emission Tomography for Biochemical Recurrence Risk Assessment: A Follow-up Analysis of a Multicenter Prospective Phase 3 Imaging Trial

BACKGROUND

In the initial staging of patients with high-risk prostate cancer (PCa), prostate-specific membrane antigen positron emission tomography (PSMA-PET) has been established as a front-line imaging modality. The increasing number of PSMA-PET scans performed in the primary staging setting might be associated with decreases in biochemical recurrence (BCR)-free survival (BCR-FS).

OBJECTIVE

To assess the added prognostic value of presurgical PSMA-PET for BCR-FS compared with the presurgical Cancer of the Prostate Risk Assessment (CAPRA) and postsurgical CAPRA-Surgery (CAPRA-S) scores in patients with intermediate- to high-risk PCa treated with radical prostatectomy (RP) and pelvic lymph node dissection.

DESIGN, SETTING, AND PARTICIPANTS

This is a follow-up study of the surgical cohort evaluated in the multicenter prospective phase 3 imaging trial (n = 277; NCT03368547, NCT02611882, and NCT02919111).

OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS

Each 68Ga-PSMA-11-PET scan was read by three blinded independent readers. PSMA-PET prostate uptake (low vs high), PSMA-PET extraprostatic disease (N1/M1), and CAPRA and CAPRA-S scores were used to assess the risk of BCR. Patients were followed after RP by local investigators using electronic medical records. BCR was defined by a prostate-specific antigen (PSA) level increasing to ≥0.2 ng/ml after RP or initiation of PCa-specific secondary treatment (>6 mo after surgery). Univariate and multivariable Cox models, and c-statistic index were performed to assess the prognostic value of PSMA-PET and for a comparison with the CAPRA and CAPRA-S scores.

RESULTS AND LIMITATIONS

From December 2015 to December 2019, 277 patients underwent surgery after PSMA-PET. Clinical follow-up was obtained in 240/277 (87%) patients. The median follow-up after surgery was 32.4 (interquartile range 23.3-42.9) mo. Of 240 BCR events, 91 (38%) were observed. PSMA-PET N1/M1 was found in 41/240 (17%) patients. PSMA-PET prostate uptake, PSMA-PET N1/M1, and CAPRA and CAPRA-S scores were significant univariate predictors of BCR. The addition of PSMA-PET N1/M1 status to the presurgical CAPRA score improved the risk assessment for BCR significantly in comparison with the presurgical CAPRA score alone (c-statistic 0.70 [0.64-0.75] vs 0.63 [0.57-0.69]; p < 0.001). The C-index of the postsurgical model utilizing the postsurgical CAPRA-S score alone was not significantly different from the presurgical model combining the presurgical CAPRA score and PSMA-PET N1/M1 status (p = 0.19).

CONCLUSIONS

Presurgical PSMA-PET was a strong prognostic biomarker improving BCR-FS risk assessment. Its implementation in the presurgical risk assessment with the CAPRA score improved the performance and reduced the difference with the reference standard (postsurgical CAPRA-S score).

PATIENT SUMMARY

The use prostate-specific membrane antigen positron emission tomography improved the assessment of biochemical recurrence risk in patients with intermediate- and high-risk prostate cancer who were treated with radical prostatectomy and pelvic lymph node dissection.

Canine-Inspired Chemometric Analysis of Volatile Organic Compounds in Urine Headspace to Distinguish Prostate Cancer in Mice and Men

Canines can identify prostate cancer with high accuracy by smelling volatile organic compounds (VOCs) in urine. Previous studies have identified VOC biomarkers for prostate cancer utilizing solid phase microextraction (SPME) gas chromatography-mass spectrometry (GC-MS) but have not assessed the ability of VOCs to distinguish aggressive cancers. Additionally, previous investigations have utilized murine models to identify biomarkers but have not determined if the results are translatable to humans. To address these challenges, urine was collected from mice with prostate cancer and men undergoing prostate cancer biopsy and VOCs were analyzed by SPME GC-MS. Prior to analysis, SPME fibers/arrows were compared, and the fibers had enhanced sensitivity toward VOCs with a low molecular weight. The analysis of mouse urine demonstrated that VOCs could distinguish tumor-bearing mice with 100% accuracy. Linear discriminant analysis of six VOCs in human urine distinguished prostate cancer with sensitivity = 75% and specificity = 69%. Another panel of seven VOCs could classify aggressive cancer with sensitivity = 78% and specificity = 85%. These results show that VOCs have moderate accuracy in detecting prostate cancer and a superior ability to stratify aggressive tumors. Furthermore, the overlap in the structure of VOCs identified in humans and mice shows the merit of murine models for identifying biomarker candidates.

Modern Active Surveillance in Prostate Cancer: A Narrative Review

The use of PSA screening has led to downstaging and downgrading of prostate cancer at diagnosis, increasing detection of indolent disease. Active surveillance aims to reduce over-treatment by delaying or avoiding radical treatment and its associated morbidity. However, there is not a consensus on the selection criteria and monitoring schedules that should be used. This article aims to summarize the evidence supporting the safety of active surveillance, the current selection criteria recommended and in use, the incidence of active surveillance, barriers existing to its uptake and future developments in patient selection.

Prostate cancer therapy personalization via multi-modal deep learning on randomized phase III clinical trials

Prostate cancer is the most frequent cancer in men and a leading cause of cancer death. Determining a patient's optimal therapy is a challenge, where oncologists must select a therapy with the highest likelihood of success and the lowest likelihood of toxicity. International standards for prognostication rely on non-specific and semi-quantitative tools, commonly leading to over- and under-treatment. Tissue-based molecular biomarkers have attempted to address this, but most have limited validation in prospective randomized trials and expensive processing costs, posing substantial barriers to widespread adoption. There remains a significant need for accurate and scalable tools to support therapy personalization. Here we demonstrate prostate cancer therapy personalization by predicting long-term, clinically relevant outcomes using a multimodal deep learning architecture and train models using clinical data and digital histopathology from prostate biopsies. We train and validate models using five phase III randomized trials conducted across hundreds of clinical centers. Histopathological data was available for 5654 of 7764 randomized patients (71%) with a median follow-up of 11.4 years. Compared to the most common risk-stratification tool-risk groups developed by the National Cancer Center Network (NCCN)-our models have superior discriminatory performance across all endpoints, ranging from 9.2% to 14.6% relative improvement in a held-out validation set. This artificial intelligence-based tool improves prognostication over standard tools and allows oncologists to computationally predict the likeliest outcomes of specific patients to determine optimal treatment. Outfitted with digital scanners and internet access, any clinic could offer such capabilities, enabling global access to therapy personalization.

Protein signatures to distinguish aggressive from indolent prostate cancer

BACKGROUND

Distinguishing men with aggressive from indolent prostate cancer is critical to decisions in the management of clinically localized prostate cancer. Molecular signatures of aggressive disease could help men overcome this major clinical challenge by reducing unnecessary treatment and allowing more appropriate treatment of aggressive disease.

METHODS

We performed a mass spectrometry-based proteomic analysis of normal and malignant prostate tissues from 22 men who underwent surgery for prostate cancer. Prostate cancer samples included Grade Groups (3-5), with 8 patients experiencing recurrence and 14 without evidence of recurrence with a mean of 6.8 years of follow-up. To better understand the biological pathways underlying prostate cancer aggressiveness, we performed a systems biology analysis and gene enrichment analysis. Proteins that distinguished recurrent from nonrecurrent cancer were chosen for validation by immunohistochemical analysis on tissue microarrays containing samples from a larger cohort of patients with recurrent and nonrecurrent prostate cancer.

RESULTS

In all, 24,037 unique peptides (false discovery rate < 1%) corresponding to 3,313 distinct proteins were identified with absolute abundance ranges spanning seven orders of magnitude. Of these proteins, 115 showed significantly (p < 0.01) different levels in tissues from recurrent versus nonrecurrent cancers. Analysis of all differentially expressed proteins in recurrent and nonrecurrent cases identified several protein networks, most prominently one in which approximately 24% of the proteins in the network were regulated by the YY1 transcription factor (adjusted p < 0.001). Strong immunohistochemical staining levels of three differentially expressed proteins, POSTN, CALR, and CTSD, on a tissue microarray validated their association with shorter patient survival.

CONCLUSIONS

The protein signatures identified could improve understanding of the molecular drivers of aggressive prostate cancer and be used as candidate prognostic biomarkers.

Exome sequencing identifies RASSF1 and KLK3 germline variants in an Iranian multiple-case breast cancer family

Breast cancer is the most frequent malignancy among women in both developed and developing countries. Although several genes have been identified to harbor germline variants contributing to breast cancer risk, much of the heritability for breast cancer is yet undefined. In the present study, we have performed exome sequencing to detect susceptibility genes in an Iranian family with five first-degree family members affected with breast cancer. We identified novel candidate variants with predicted pathogenicity in RASSF1, KLK3 and FAM81B. The RASSF1 and KLK3 variants, but not the FAM81B variant, partially co-segregated with disease in the investigated pedigree and were not found in additional screenings outside the specific family. RASSF1 p.S135F is a missense substitution abolishing the ATM phosphorylation site, and KLK3 variant p.M1? is a deletion at the initiation codon that is predicted to abolish translation to the functional kallikrein protease, PSA. Our study suggests germline variation in RASSF1 and KLK3 as candidate contributors to familial breast cancer predisposition and illustrates the difficulties to determine the causal genetic risk factor among novel variants restricted to a single family.

The Role and Significance of Bioumoral Markers in Prostate Cancer

The prostate is one of the most clinically accessible internal organs of the genitourinary tract in men. For decades, the only method of screening for prostate cancer (PCa) has been digital rectal examination of 1990s significantly increased the incidence and prevalence of PCa and consequently the morbidity and mortality associated with this disease. In addition, the different types of oncology treatment methods have been linked to specific complications and side effects, which would affect the patient's quality of life. In the first two decades of the 21st century, over-detection and over-treatment of PCa patients has generated enormous costs for health systems, especially in Europe and the United States. The Prostate Specific Antigen (PSA) is still the most common and accessible screening blood test for PCa, but with low sensibility and specificity at lower values (<10 ng/mL). Therefore, in order to avoid unnecessary biopsies, several screening tests (blood, urine, or genetic) have been developed. This review analyzes the most used bioumoral markers for PCa screening and also those that could predict the evolution of metastases of patients diagnosed with PCa.

The use of molecular biomarker tests: an interview study with healthcare providers about a molecular biomarker test for prostate cancer

Aim: Investigate why healthcare providers are not always willing to use molecular biomarker tests, even though they promise to personalize disease diagnosis and treatment. Materials & methods: We interviewed 20 Dutch urological healthcare providers to ascertain why they used or did not use SelectMDx, a biomarker test for prostate cancer. Results: Whether and how it was used differed from the developers' expectations, because users and nonusers disagreed about its perceived advantages; the scientific and clinical evidence; the advantages of MRI; and the value of PCA3 testing. Financial issues and the absence of SelectMDx in professional guidelines and hospital care pathways also hampered its use. Conclusion: Eliciting users' and nonusers' views is important to better understand how biomarker tests can be embedded in clinical practice.

Radiomic and Genomic Machine Learning Method Performance for Prostate Cancer Diagnosis: Systematic Literature Review

BACKGROUND

Machine learning algorithms have been drawing attention at the joining of pathology and radiology in prostate cancer research. However, due to their algorithmic learning complexity and the variability of their architecture, there is an ongoing need to analyze their performance.

OBJECTIVE

This study assesses the source of heterogeneity and the performance of machine learning applied to radiomic, genomic, and clinical biomarkers for the diagnosis of prostate cancer. One research focus of this study was on clearly identifying problems and issues related to the implementation of machine learning in clinical studies.

METHODS

Following the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) protocol, 816 titles were identified from the PubMed, Scopus, and OvidSP databases. Studies that used machine learning to detect prostate cancer and provided performance measures were included in our analysis. The quality of the eligible studies was assessed using the QUADAS-2 (quality assessment of diagnostic accuracy studies-version 2) tool. The hierarchical multivariate model was applied to the pooled data in a meta-analysis. To investigate the heterogeneity among studies, I2 statistics were performed along with visual evaluation of coupled forest plots. Due to the internal heterogeneity among machine learning algorithms, subgroup analysis was carried out to investigate the diagnostic capability of machine learning systems in clinical practice.

RESULTS

In the final analysis, 37 studies were included, of which 29 entered the meta-analysis pooling. The analysis of machine learning methods to detect prostate cancer reveals the limited usage of the methods and the lack of standards that hinder the implementation of machine learning in clinical applications.

CONCLUSIONS

The performance of machine learning for diagnosis of prostate cancer was considered satisfactory for several studies investigating the multiparametric magnetic resonance imaging and urine biomarkers; however, given the limitations indicated in our study, further studies are warranted to extend the potential use of machine learning to clinical settings. Recommendations on the use of machine learning techniques were also provided to help researchers to design robust studies to facilitate evidence generation from the use of radiomic and genomic biomarkers.

Correlative analysis between two commercially available post-prostatectomy genomic tests

BACKGROUND

Multiple genomic tests are available following radical prostatectomy (RP), however, there is a lack of head-to-head evidence for these tests. We sought to compare the performance of two genomic tests in predicting post-RP oncological outcomes.

METHODS

A cohort of 16 post-RP patients with adverse pathological features who had obtained both Decipher (D) and Prolaris (P) testing. The Pearson correlation was used to compare scores from D and cell cycle progression (CCP) from P. Then, we derived a microarray CCP (mCCP) from D and correlated with P-CCP. The associations of D and mCCP with biochemical recurrence (BCR) and metastasis (M) was evaluated in multivariable survival analysis (MVA) in a large cohort of RP patients treated at Johns Hopkins University (1992-2010). In addition, we characterized the expression of the 31 P-CCP genes and mCCP scores in a cohort of 17,967 RP samples from Decipher platform.

RESULTS

There was significant correlation between the D score and P-CCP (r = 0.67, p = 0.004), and between the 10-year probability of BCR reported by P and 5-year probability of M reported by D (r = 0.69, p = 0.003). In this cohort, mCCP derived from the D platform was highly correlated to the reported P-CCP scores from the P platform (r = 0.88, p = 6.7e-6). In a comparative retrospective RP cohort, both mCCP and D were significantly associated with M outcome (p < 0.01 for both). On MVA, D was a predictor of M (HR 1.3, 95% CI [1.12-1.52], p = 0.0005), while mCCP was not a predictor of M (p = 0.62). In the D platform cohort, the 31 P-CCP genes were correlated to each other, and TOP2A was the most correlated to mCCP (r = 0.7).

CONCLUSIONS

We found that P and D scores post-RP were correlated and help in identifying patients who at high risk of BCR in this cohort. In a larger cohort with longer follow-up, D was predictor of M, whereas mCCP was not.

A review of current clinical biomarkers for prostate cancer: towards personalised and targeted therapy

Background:

Prostate cancer is the most commonly diagnosed cancer in men and it is responsible for about 10% of all cancer mortality in Canadian men. The current ‘gold standard’ for the diagnosis of prostate cancer is a prostate biopsy and the decision on when to biopsy a patient with non-suspicious Digital Rectal Examination (DRE) result and total prostate specific antigen (tPSA) of 4–10 ng/ml can be challenging. In order to shift the treatment paradigm of prostate cancer toward more personalised and targeted therapy, there is the need for a clear system that makes its detection binary so as to decrease the rate of inaccurate detections. Therefore in recent years, there have been several investigations into the development of various biomarkers with high sensitivity and specificity for screening, early detection and personalised patient-specific targeted medicine from diagnosis to treatment of the disease.

Materials and methods:

This paper reports on nine currently available clinical biomarkers used in screening for early detection and diagnosis, to reduce the number of unnecessary biopsies, in risk assessment of aggressive disease and in monitoring treatment response of prostate cancer.

Conclusion:

Current clinical prostate cancer biomarkers have the potential for a personalised risk assessment of aggressive disease and the risk of developing distant metastatic disease and have been proven to be useful tools to guide clinicians in personalised patient-specific targeted treatment and in the shared decision making between patients and their physicians regarding prostate biopsy and treatment. Using biomarkers to select patients with a significant probability of aggressive prostate cancer would potentially avoid premature death from the disease, while at the same time would safely preclude patients who do not require unnecessary invasive intervention.

The preventative effect of bone marrow-derived mesenchymal stem cell exosomes on urethral stricture in rats

Background

Urethral stricture (US) is a major challenge in urology and there is an urgent need for effective therapies for its treatment. Exosomes derived from bone marrow mesenchymal stem cells (BMSCs-Exos) have been shown to be effective in preventing scar and fibrosis formation after tissue injury. However, the potential utility of BMSCs-Exos in the prevention of US remains unknown. We hypothesized that local administration of BMSCs-Exos may influence urethral healing and scar formation in a rat model of US.

Methods

A previously established model of rat US was used in this study. Sprague Dawley rats were randomly assigned into sham, US, and US + BMSCs-Exos groups. Micro-ultrasound assessment, histopathology, immunohistochemistry, and gene expression analysis were performed at four weeks post-surgery.

Results

US rats exhibited thick urethral walls with a narrowed lumen, when compared with sham rats. However, these changes were suppressed in the US + BMSCs-Exos group. The preventative effects of BMSCs-Exos on US formation were also apparent histologically. US + BMSCs-Exos rats demonstrated decreased expression of several fibrosis-related genes in urethral tissues, including Col I, fibronectin, and elastin, when compared with US rats. BMSCs-Exos treatment also led to an increase in the expression of angiogenesis-related genes in these tissues, including VEGF, eNOS, and bFGF.

Conclusions

Our findings therefore demonstrate that the local administration of BMSCs-Exos prevents urethral stricture formation by regulating fibrosis and angiogenesis. These findings provide a basis for an innovative strategy involving the clinical application of exosomes to counteract US formation.

A MYC and RAS co-activation signature in localized prostate cancer drives bone metastasis and castration resistance

Understanding the intricacies of lethal prostate cancer poses specific challenges due to difficulties in accurate modeling of metastasis in vivo. Here we show that NPK EYFP mice (for Nkx3.1 CreERT2/+ ; Pten flox/flox ; Kras LSL-G12D/+ ; R26R-CAG-LSL-EYFP/+) develop prostate cancer with a high penetrance of metastasis to bone, thereby enabling detection and tracking of bone metastasis in vivo and ex vivo. Transcriptomic and whole-exome analyses of bone metastasis from these mice revealed distinct molecular profiles conserved between human and mouse and specific patterns of subclonal branching from the primary tumor. Integrating bulk and single-cell transcriptomic data from mouse and human datasets with functional studies in vivo unravels a unique MYC/RAS co-activation signature associated with prostate cancer metastasis. Finally, we identify a gene signature with prognostic value for time to metastasis and predictive of treatment response in human patients undergoing androgen receptor therapy across clinical cohorts, thus uncovering conserved mechanisms of metastasis with potential translational significance.

A 38-gene model comprised of key TET2-associated genes shows additive utility to high-risk prostate cancer cases in the prognostication of biochemical recurrence

BACKGROUND

Early treatment of patients at risk for developing aggressive prostate cancer is able to delay metastasis and reduce mortality; as such, up-front identification of these patients is critical. Several risk classification systems, including CAPRA-S, are currently used for disease prognostication. However, high-risk patients identified by these systems can still exhibit wide-ranging disease outcomes, leading to overtreatment of some patients in this group.

METHODS

The master methylation regulator TET2 is downregulated in prostate cancer, where its loss is linked to aggressive disease and poor outcome. Using a random forest strategy, we developed a model based on the expression of 38 genes associated with TET2 utilizing 100 radical prostatectomy samples (training cohort) with a 49% biochemical recurrence rate. This 38-gene model was comprised of both upregulated and downregulated TET2-associated genes with a binary outcome, and was further assessed in an independent validation (n = 423) dataset for association with biochemical recurrence.

RESULTS

38-gene model status was able to correctly identify patients exhibiting recurrence with 81.4% sensitivity in the validation cohort, and added significant prognostic utility to the high-risk CAPRA-S classification group. Patients considered high-risk by CAPRA-S with negative 38-gene model status exhibited no statistically significant difference in time to recurrence from low-risk CAPRA-S patients, indicating that the expression of TET2-associated genes is able to separate truly high-risk cases from those which have a more benign disease course.

CONCLUSIONS

The 38-gene model may hold potential in determining which patients would truly benefit from aggressive treatment course, demonstrating a novel role for genes linked to TET2 in the prognostication of PCa and indicating the importance of TET2 dysregulation among high-risk patient groups.

Prediction of prostate cancer Gleason score upgrading from biopsy to radical prostatectomy using pre-biopsy multiparametric MRI PIRADS scoring system

An increase or ‘upgrade’ in Gleason Score (GS) in prostate cancer following Transrectal Ultrasound (TRUS) guided biopsies remains a significant challenge to overcome. to evaluate whether MRI has the potential to narrow the discrepancy of histopathological grades between biopsy and radical prostatectomy, three hundred and thirty men treated consecutively by laparoscopic radical prostatectomy (LRP) between July 2014 and January 2019 with localized prostate cancer were included in this study. Independent radiologists and pathologists assessed the MRI and histopathology of the biopsies and prostatectomy specimens respectively. A multivariate model was constructed using logistic regression analysis to assess the ability of MRI to predict upgrading in biopsy GS in a nomogram. A decision-analysis curve was constructed assessing impact of nomogram using different thresholds for probabilities of upgrading. PIRADS scores were obtained from MRI scans in all the included cases. In a multivariate analysis, the PIRADS v2.0 score significantly improved prediction ability of MRI scans for upgrading of biopsy GS (p = 0.001, 95% CI [0.06–0.034]), which improved the C-index of predictive nomogram significantly (0.90 vs. 0.64, p < 0.05). PIRADS v2.0 score was an independent predictor of postoperative GS upgrading and this should be taken into consideration while offering treatment options to men with localized prostate cancer.

Concordance and Performance of 4Kscore and SelectMDx for Informing Decision to Perform Prostate Biopsy and Detection of Prostate Cancer

OBJECTIVES

To compare both the concordance between the 4Kscore and SelectMDx for informing decision to perform prostate biopsy (PB) and the performance of these tests for detecting clinically significant prostate cancer (csPCa). Several biomarkers were developed to inform decisions whether to perform a PB based on the probability of detecting csPCa. There is a paucity of studies directly comparing them METHODS: Between 11/2018 and 4/2019, all new referrals with the diagnosis of elevated PSA were advised to undergo 4Kscore and SelectMDx in order to guide the selection of candidates for PB. Men were advised to undergo PB if the reported biomarker risk for detecting csPCA was ≥7.5%, or if they presented a Prostate Image Reporting and Data System ≥1 MRI. Cohen's Kappa was used to assess the concordance between the binary 4Kscore and SelectMDx results using externally validated cutoffs of 7.5% and 12%. Receiver operating characteristics curve and area under the curve (AUC) assessed the performance of each biomarker for distinguishing csPCa.

RESULTS

Of 128 consecutive patients referred, 114 (89.1%) underwent 4Kscore and SelectMDx. The kappa coefficient between the biomarkers using the 7.5% cutoff was 0.184 (poor concordance) and 0.22 using the 12% cutoff. The two biomarkers yielded discordant guidance whether to proceed with PB in 46% and 38% of cases, respectively. csPCa was found in 22 of the 50 patients who underwent PB (44%). The AUC for 4Kscore and SelectMDx was 0.830 (95% confidence interval [CI]: 0.710 - 0.949) and 0.672 (95%CI: 0.517 - 0.828; P = .036) for csPCa, respectively.

CONCLUSION

The discordance observed between the 4Kscore and SelectMDx is disconcerting. The 4Kscore when combined with magnetic resonance imaging was superior to the SelectMDx for detecting csPCa. Prospective comparative studies must be performed to optimize implementation of biomarkers for selecting candidates for PB.

Decipher identifies men with otherwise clinically favorable-intermediate risk disease who may not be good candidates for active surveillance

BACKGROUND

We aimed to validate Decipher to predict adverse pathology (AP) at radical prostatectomy (RP) in men with National Comprehensive Cancer Network (NCCN) favorable-intermediate risk (F-IR) prostate cancer (PCa), and to better select F-IR candidates for active surveillance (AS).

METHODS

In all, 647 patients diagnosed with NCCN very low/low risk (VL/LR) or F-IR prostate cancer were identified from a multi-institutional PCa biopsy database; all underwent RP with complete postoperative clinicopathological information and Decipher genomic risk scores. The performance of all risk assessment tools was evaluated using logistic regression model for the endpoint of AP, defined as grade group 3-5, pT3b or higher, or lymph node invasion.

RESULTS

The median age was 61 years (interquartile range 56-66) for 220 patients with NCCN F-IR disease, 53% classified as low-risk by Cancer of the Prostate Risk Assessment (CAPRA 0-2) and 47% as intermediate-risk (CAPRA 3-5). Decipher classified 79%, 13% and 8% of men as low-, intermediate- and high-risk with 13%, 10%, and 41% rate of AP, respectively. Decipher was an independent predictor of AP with an odds ratio of 1.34 per 0.1 unit increased (p value = 0.002) and remained significant when adjusting by CAPRA. Notably, F-IR with Decipher low or intermediate score did not associate with significantly higher odds of AP compared to VL/LR.

CONCLUSIONS

NCCN risk groups, including F-IR, are highly heterogeneous and should be replaced with multivariable risk-stratification. In particular, incorporating Decipher may be useful for safely expanding the use of AS in this patient population.

Beyond the biomarker role: prostate-specific antigen (PSA) in the prostate cancer microenvironment

The prostate-specific antigen (PSA) blood test is the accepted biomarker of tumor recurrence. PSA levels in serum correlate with disease progression, though its diagnostic accuracy is questionable. As a result, significant progress has been made in developing modified PSA tests such as PSA velocity, PSA density, 4Kscore, PSA glycoprofiling, Prostate Health Index, and the STHLM3 test. PSA, a serine protease, is secreted from the epithelial cells of the prostate. PSA has been suggested as a molecular target for prostate cancer therapy due to the fact that it is not only active in prostate tissue but also has a pivotal role on prostate cancer signaling pathways including proliferation, invasion, metastasis, angiogenesis, apoptosis, immune response, and tumor microenvironment regulation. Here, we summarize the current standing of PSA in prostate cancer progression as well as its utility in prostate cancer therapeutic approaches with an emphasis on the role of PSA in the tumor microenvironment.

RASSF10 is frequently epigenetically inactivated in kidney cancer and its knockout promotes neoplasia in cancer prone mice

Kidney cancer incidences are rising globally, thereby fueling the demand for targeted therapies and precision medicine. In our previous work, we have identified and characterized the Ras-Association Domain Family encoding ten members that are often aberrantly expressed in human cancers. In this study, we created and analyzed the Rassf10 knockout mice. Here we show that Rassf10 haploinsufficiency promotes neoplasia formation in two established mouse cancer models (Rassf1A^-/- and p53^-/-). Haploinsufficient Rassf10 knockout mice were significantly prone to various diseases including lymphoma (Rassf1A^-/- background) and thymoma (p53^-/- background). Especially Rassf10^-/- and p53-deficient mice exhibited threefold increased rates of kidney cysts compared with p53^-/- controls. Moreover, we observed that in human kidney cancer, RASSF10 is frequently epigenetically inactivated by its CpG island promoter hypermethylation. Primary tumors of renal clear cell and papillary cell carcinoma confirmed that RASSF10 methylation is associated with decreased expression in comparison to normal kidney tissue. In independent data sets, we could validate that RASSF10 inactivation clinically correlated with decreased survival and with progressed disease state of kidney cancer patients and polycystic kidney size. Functionally, we revealed that the loss of Rassf10 was significantly associated with upregulation of KRAS signaling and MYC expression. In summary, we could show that Rassf10 functions as a haploinsufficient tumor suppressor. In combination with other markers, RASSF10 silencing can serve as diagnostic and prognostic cancer biomarker in kidney diseases.

Expression of stromal elements of prostatic adenocarcinoma in different gleason scores

Purpose:

To quantify and compare the expression of stromal elements in prostate adenocarcinoma of different Gleason scores with non-tumor area (control).

Methods:

We obtained 132 specimens from samples of prostate peripheral and transition zone. We analyzed the following elements of the extracellular matrix: collagen fibers, elastic system, smooth muscle fibers and blood vessels. The tumor area and non-tumor area (control) of the TMA (tissue microarray) were photographed and analyzed using the ImageJ software.

Results:

The comparison between the tumor area and the non-tumor area showed significant differences between stromal prostate elements. There was an increase of collagen fibers in the tumor area, mainly in Gleason 7. Elastic system fibers showed similar result, also from the Gleason 7. Blood vessels showed a significant increase occurred in all analyzed groups. The muscle fibers exhibited a different behavior, with a decrease in relation to the tumor area.

Conclusions:

There is a significant difference between the extracellular matrix in prostate cancer compared to the non-tumor area (control) especially in Gleason 7. Important modifications of the prostatic stromal elements strongly correlate with different Gleason scores and can contribute to predict the pathological staging of prostate cancer.

Molecular Composition of Genomic TMPRSS2-ERG Rearrangements in Prostate Cancer

There is increasing interest in the use of cell-free circulating tumor DNA (ctDNA) as a serum marker for therapy assessment in prostate cancer patients. Prostate cancer is characterized by relatively low numbers of mutations, and, in contrast to many other common epithelial cancers, commercially available single nucleotide mutation assays for quantification of ctDNA are insufficient for therapy assessment in this disease. However, prostate cancer shares some similarity with translocation-affected mesenchymal tumors (e.g., leukemia and Ewing sarcoma), which are common in pediatric oncology, where chromosomal translocations are used as biomarkers for quantification of the tumor burden. Approximately 50% of prostate cancers carry a chromosomal translocation resulting in generation of the TMPRSS2-ERG fusion gene, which is unique to the tumor cells of each individual patient because of variability in the fusion breakpoint sites. In the present study, we examined the structural preconditions for TMPRSS2-ERG fusion sites in comparison with mesenchymal tumors in pediatric patients to determine whether the sequence composition is suitable for the establishment of tumor-specific quantification assays in prostate cancer patients. Genomic repeat elements represent potential obstacles to establishment of quantification assays, and we found similar proportions of repeat elements at fusion sites in prostate cancer to those reported for mesenchymal tumors, where genomic fusion sequences are established as biomarkers. Our data support the development of the TMPRSS2-ERG fusion gene as a noninvasive tumor marker for therapy assessment, risk stratification, and relapse detection to improve personalized therapy strategies for patients with prostate cancer.

Assessment of the Utility of Gene Positioning Biomarkers in the Stratification of Prostate Cancers

There is a pressing need for additional clinical biomarkers to predict the aggressiveness of individual cancers. Here, we examine the potential usefulness of spatial genome organization as a prognostic tool for prostate cancer. Using fluorescence in situ hybridization on formalin-fixed, paraffin embedded human prostate tissue specimens, we compared the nuclear positions of four genes between clinically relevant subgroups of prostate tissues. We find that directional repositioning of SP100 and TGFB3 gene loci stratifies prostate cancers of differing Gleason scores. A more peripheral position of SP100 and TGFB3 in the nucleus, compared to benign tissues, is associated with low Gleason score cancers, whereas more internal positioning correlates with higher Gleason scores. Conversely, LMNA is more internally positioned in many non-metastatic prostate cancers, while its position is indistinguishable from benign tissue in metastatic cancer. The false positive rates were relatively low, whereas, the false negative rates of single or combinations of genes were high, limiting the clinical utility of this assay in its current form. Nevertheless, our findings of subtype-specific gene positioning patterns in prostate cancer provides proof-of-concept for the potential usefulness of spatial gene positioning for prognostic applications, and encourage further exploration of spatial gene positioning patterns to identify novel clinically relevant molecular biomarkers, which may aid treatment decisions for cancer patients.

Gleason pattern 5 is associated with an increased risk for metastasis following androgen deprivation therapy and radiation: An analysis of RTOG 9202 and 9902

BACKGROUND/PURPOSE

Stratification of Gleason score (GS) into three categories (2-6, 7, and 8-10) may not fully utilize its prognostic discrimination, with Gleason pattern 5 (GP5) previously identified as an independent adverse factor.

MATERIALS/METHODS

Patients treated on RTOG 9202 (n = 1292) or RTOG 9902 (n = 378) were pooled and assessed for association of GS and GP5 on biochemical failure (BF), local failure (LF), distant metastasis (DM), and overall survival (OS). Fine and Gray's regression and cumulative incidence methods were used for univariate and multivariate analyses.

RESULTS

With median follow-up of 9.4 years, patients with GS 8-10 with GP5 had worse outcome than GS 4 + 4 for DM on both RTOG9202 (p = 0.038) and RTOG9902 (p < 0.001) with a trend toward worse OS (p = 0.059 and p = 0.089, respectively), but without differences in BF or LF. At 10-years DM was higher by 11% (RTOG 9202) and 18% (RTOG 9902) with GP5 compared to GS 4 + 4. On multivariate analysis restricted to long-term androgen deprivation therapy the presence of GP5 substantially increased distant metastasis (HR = 0.43, 95%CI: 0.24-0.76, p = 0.0039) with a trend toward worse OS (HR:0.74, 95% CI:0.54-1.0, p = 0.052) without association with LF (HR:0.55, 95%CI:0.28-1.09, p = 0.085) or BF (HR:1.15, 95%CI:0.84-1.59, p = 0.39). We did not observed substantial differences between Gleason 3 + 5, 5 + 3, or Gleason 9-10.

CONCLUSIONS

These results validate GP5 as an independent prognostic factor which is strongest for DM. As a result GP5 should be considered when stratifying patients with GS 8 and may be a patient population in which to evaluate newly approved systemic therapies or additional local treatments.

The pathological upgrading after radical prostatectomy in low-risk prostate cancer patients who are eligible for active surveillance: How safe is it to depend on bioptic pathology?

BACKGROUND

Active surveillance (AS) is one of the treatment alternatives in low-risk prostate cancer (PCa). The pathological upgrading after radical prostatectomy (RP) were investigated in patients who were eligible for AS in the present study.

METHODS

Between August 2006 and July 2017, 627 patients underwent RP in our institution. One hundred and thirty-six patients who were eligible for AS at the time of RP were included in this study. The previously defined AS criteria Gleason 3 + 3=6 adenocarcinoma at maximum two biopsy cores, prostate-specific antigen (PSA) < 10 ng/mL and clinical T stage ≤ 2a were used in the study. The demographics, clinical, and histopathological outcomes were retrospectively compared between two groups, which were divided in accordance with the upgrading status at final pathology as Group 1 (n = 67, upgrading) and Group 2 (n = 69, nonupgrading).

RESULTS

Gleason upgrading (GU) was found in 67 (49.3%) patients, and 17 patients (12.5%) were upstaged to pT3a. The upgrading to Gleason 3 + 4 was reported in 38.7% of patients, however, 7.4%, and 3.7% of the patients were upgraded to Gleason 4 + 3, and Gleason 4 + 4, respectively. The 10.3% of the patients had extraprostatic involvement, and the rate (19.4% vs 1.4%, P = .002) was significantly higher in Group 1. PSA density (P = .001), tumor size (P < .001), tumor percentage (P < .001), apical involvement (P = .013), and perineural invasion (P < .001) in RP specimen were higher in Group 1. Multivariate analysis showed that perineural invasion (OR = 4.26; 95%CI: 1.76-10.33; P = .001) and pathologic T stage (OR = 5.45; 95%CI: 1.08-27.4; P = .04) were independently associated with GU.

CONCLUSIONS

Since 12.5% of the patients upstaged to pT3a disease, and there is a possible risk of Gleason 4 pattern, upgrading of the tumor should carefully be kept in mind before offering AS to low-risk patients with PCa.

Assessment of men's risk thresholds to proceed with prostate biopsy for the early detection of prostate cancer

PURPOSE

To delineate the range of "risk thresholds" for prostate biopsy to determine how improved prostate cancer (CaP) risk prediction tools may impact shared decision-making (SDM).

METHODS

We conducted a cross-sectional survey study involving men 45-75 years old attending a multispecialty urology clinic. Data included demographics, personal and family prostate cancer history, and prostate biopsy history. Respondents were presented with a summary of the details, risks, and benefits of prostate biopsy, then asked to indicate the specific risk threshold (% chance) of high-grade CaP at which they would proceed with prostate biopsy.

RESULTS

Of a total of 103 respondents, 18 men (17%) had a personal history of CaP, and 31 (30%) had undergone prostate biopsy. The median risk threshold to proceed with prostate biopsy was 25% (interquartile range 10-50%). Risk thresholds did not vary by race, education, or employment. Personal history of CaP or prostate biopsy was significantly associated with lower mean risk thresholds (19% vs. 32% [P = 0.02] and 23% vs. 33% [P = 0.04], respectively). In the lowest versus highest risk threshold quartiles, there were significantly higher rates of CaP (36% vs. 1%, P = 0.01) and prior prostate biopsy (46% vs. 17%, P < 0.01).

CONCLUSIONS

Men have a wide range of risk thresholds for high-grade CaP to proceed with prostate biopsy. Men with a prior history of CaP or biopsy reported lower risk thresholds, which may reflect their greater concern for this disease. The extent to which refined risk prediction tools will improve SDM warrants further study.

SPINK1 expression is enriched in African American prostate cancer but is not associated with altered immune infiltration or oncologic outcomes post-prostatectomy

BACKGROUND

The SPINK1 molecular subtype is more common in African-American (AA) men with prostatic adenocarcinoma (PCa) than European Americans (EA). Studies have suggested that SPINK1 expression is associated with more aggressive disease. However, the size, follow-up, and racial diversity of prior patient cohorts have limited our understanding of SPINK1 expression in AA men. The objective was to determine the associations between SPINK1 subtype, race, and oncologic outcomes after radical prostatectomy (RP).

METHODS

A total of 186 AA and 206 EA men who underwent RP were matched according to pathologic grade. We examined SPINK1 status by immunohistochemistry on tissue microarrays using a genetically validated assay. Cox proportional hazard analyses assessed the association of SPINK1 status with oncologic outcomes in race-specific multivariate models. A second objective was to determine the correlation between CD3/CD8 T cell densities with SPINK1 status and race, using immunostaining and automated image analysis.

RESULTS

SPINK1-positive subtype was present in 25% (45/186) of AA and 15% (30/206) of EA men (p = 0.013). There were no differences in pathologic grade, pathologic stage, biochemical recurrence (BCR)-free survival, or metastasis-free survival between SPINK1-positive and SPINK1-negative tumors in the overall cohort or by race. In multivariate analyses, SPINK1 expression was not associated with BCR (AA: HR 0.99, 95% CI 0.56-1.75, p = 0.976; EA: HR 0.88, 95% CI 0.43-1.77, p = 0.720) or metastasis (AA: HR 0.79, 95% CI 0.25-2.49, p = 0.691; EA: HR 1.55, 95% CI 0.58-4.11, p = 0.381) in either AA or EA men. There were no significant differences in surrounding CD3/CD8 lymphocyte densities between SPINK1-positive and SPINK1-negative tumors in either race.

CONCLUSIONS

SPINK1-positive subtype is more prevalent in AA than EA men with PCa. Contrary to previous studies, we found that SPINK1 protein expression was not associated with worse pathologic or oncologic outcomes after RP in either AA men or EA men.

Discovery of Metabolic Biomarkers Predicting Radiation Therapy Late Effects in Prostate Cancer Patients

Patients presenting with prostate cancers undergo clinical staging evaluations to determine the extent of disease to guide therapeutic recommendations. Management options may include watchful waiting, surgery, or radiation therapy. Thus, initial risk stratification of prostate cancer patients is important for achieving optimal therapeutic results or cancer cure and preservation of quality of life. Predictive biomarkers for risks of complications or late effects of treatment are needed to inform clinical decisions for treatment selection. Here, we analyzed pre-treatment plasma metabolites in a cohort of prostate cancer patients (N = 99) treated with Stereotactic Body Radiation Therapy (SBRT) at Medstar-Georgetown University Hospital in a longitudinal, quality-of-life study to determine if individuals experiencing radiation toxicities can be identified by a molecular profile in plasma prior to treatment. We used a multiple reaction mass spectrometry-based molecular phenotyping of clinically annotated plasma samples in a retrospective outcome analysis to identify candidate biomarker panels correlating with adverse clinical outcomes following radiation therapy. We describe the discovery of candidate biomarkers, based on small molecule metabolite panels, showing high correlations (AUCs ≥ 95%) with radiation toxicities, suitable for validation studies in an expanded cohort of patients.