Genomic Classifier Performance in Intermediate-Risk Prostate Cancer: Results From NRG Oncology/RTOG 0126 Randomized Phase 3 Trial

Predicting intermediate-risk prostate cancer using machine learning

PURPOSES

Intermediate-risk prostate cancer (IR PCa) is the most common risk group for localized prostate cancer. This study aimed to develop a machine learning (ML) model that utilizes biopsy predictors to estimate the probability of IR PCa and assess its performance compared to the traditional clinical model.

METHODS

Between January 2017 and December 2022, patients with prostate-specific antigen (PSA) values of ≤ 20 ng/mL underwent transrectal ultrasonography-guided prostate biopsies. Patient's age, PSA, digital rectal exam, prostate volume, PSA density (PSAD), and previous negative biopsy, number of positive cores, Gleason score, and biopsy outcome were recorded. Patients are categorized into no cancer, very low, low-, and intermediate-risk categories. The relationship between the model and IR PCa was investigated using binary generalized linear model (GLM) and assessed its discriminatory ability by calculating the area under the receiver operating characteristic curve (AUC).

RESULTS

Among 729 patients, PCa was detected in 234 individuals (32.1%), with 120 (16.5%) diagnosed with IR PCa. The AUC for the novel model compared to the clinical model was 0.806 (95% CI: 0.722-0.889) versus 0.669 (95% CI: 0.543-0.790), with a p-value of 0.018. In DCA, the GLM outperformed the clinical model by over 7%, potentially allowing for an additional 44.3% reduction in unnecessary biopsies. The PSAD emerged as the most significant predictor.

CONCLUSION

We developed a GLM utilizing pre-biopsy features to predict IR PCa. The model demonstrated good discrimination and clinical applicability, which could assist urologists in determining the necessity of a prostate biopsy.

The Clinical Impact of the Decipher Genomic Classifier in Prostate Cancer

The Decipher genomic classifier (GC) is a 22-gene expression test that refines risk stratification and informs treatment decisions in localized prostate cancer. Traditional clinicopathologic factors, including prostate-specific antigen levels and kinetics, Gleason score, histologic variants, and tumor stage, do not fully capture disease heterogeneity, leading to potential overtreatment or undertreatment. The Decipher GC has demonstrated clinical utility across risk groups, helping to distinguish candidates for active surveillance in low-risk prostate cancer, refine the need for androgen deprivation therapy in intermediate-risk disease, and guide treatment intensification in high-risk patients. In the post-radical prostatectomy setting, the GC aids in determining the need for early salvage radiation therapy and hormonal therapy. While retrospective studies support its prognostic value, limitations include heterogeneity in study designs and the lack of established predictive utility for treatment response. Ongoing prospective trials, such as NRG GU-009 and NRG GU-010, aim to validate further the Decipher GC's role in clinical decision-making and treatment personalization.

Tissue-based gene expression testing in localized prostate cancer

PURPOSE OF REVIEW

This review presents the latest research in tissue-based genomic testing in localized prostate cancer (PCa). Here we explore the current and most commonly used genomic assays, their clinical applications, current challenges, and the future of genomic testing.

RECENT FINDINGS

The management of localized PCa has evolved with the integration of genomic assays, offering a more personalized approach to risk stratification and treatment decision-making. Traditional clinical markers such as PSA levels and Gleason scores are often insufficient in capturing clinically significant cancer due to disease heterogeneity.

SUMMARY

Tissue-based genomic tests, such as Decipher, Oncotype DX (GPS), and Prolaris, have emerged as prognostic tools for assessing tumor aggressiveness and metastatic potential. Current evidence supports Decipher's prognostic capabilities with studies demonstrating risk stratification while further research is needed for Prolaris and GPS to solidify their role in PCa risk stratification. These assays are intended to guide therapeutic choices, reducing overtreatment in low-risk cases while identifying high-risk patients who may benefit from more aggressive or definitive intervention. Despite growing clinical adoption, challenges such as cost, disparities in access, and variability in physician utilization still remain. Further prospective studies and randomized trials are required to optimize clinical implementation and validate the long-term impact of genomic testing on PCa outcomes.

Predicting dose response to prostate cancer radiotherapy: validation of a radiation signature in the randomized phase III NRG/RTOG 0126 and SAKK 09/10 trials

BACKGROUND

The SAKK 09/10 trial randomized biochemically recurrent prostate cancer patients to salvage radiation 64 Gy versus 70 Gy, and the NRG/RTOG 0126 randomized intermediate-risk prostate cancer patients to definitive radiation 70.2 Gy versus 79.2 Gy. We investigated a previously developed Post-Operative Radiation Therapy Outcomes Score (PORTOS) to identify preferential benefit from radiation dose escalation (DE).

MATERIALS AND METHODS

PORTOS was evaluated in patients enrolled in SAKK 09/10 and NRG/RTOG 0126 with available tissue that passed quality control (n = 226, 215). PORTOS was evaluated in the published post-operative groups in SAKK 09/10 and in tertiles in NRG/RTOG 0126 as cut-offs had not been established for biopsy samples and definitive radiation patients. Clinical and molecular correlates in a real-world dataset of 42 407 prostatectomy and 31 107 biopsy samples were also analyzed.

RESULTS

In SAKK 09/10, the biomarker-treatment interaction was statistically significant between PORTOS (lower versus higher) and treatment arm for clinical progression-free survival. Only patients in the higher PORTOS group benefited from DE. In NRG/RTOG 0126, in patients with a lower tertile PORTOS, there was no difference in Phoenix biochemical failure (BF). However, for patients in the average and higher tertile PORTOS range, there was a significant benefit for DE for Phoenix BF. An interaction test indicated a significant difference in benefit for DE between higher and lower PORTOS groups. PORTOS was not strongly associated with clinicopathological variables in either trial or the large real-world dataset. In the latter, PORTOS was modestly associated with hypoxia signatures and strongly associated with immune signatures and subtypes.

CONCLUSION

In the SAKK 09/10 and RTOG 0126 randomized controlled trials, we demonstrated that PORTOS can potentially identify a subset of patients who benefit from DE, a subgroup that cannot be identified using clinicopathological or prognostic variables. These results suggest that PORTOS could be used clinically as a predictor of radiation response.

Biomarkers in prostate cancer: current status and future directions in radiotherapy-statement from the Prostate Cancer Working Group of the German Society of Radiation Oncology (DEGRO)

PURPOSE

Prostate cancer (PCa) is the most frequently diagnosed malignancy among men in Germany. Advances in diagnostics and treatment have transformed PCa into a chronic disease. Given the heterogeneity of PCa, there is a need for additional stratification tools. This review focuses on updating the evidence for genomic classifiers (GC; Decipher [Veracyte Inc. San Diego, CA, USA], Prolaris [Myriad Genetics, Inc., Salt Lake City, UT], and Oncotype DX [Exact Sciences, Madison, WI, USA] tests) and artificial intelligence (AI)-based digital histopathology biomarkers (ArteraAI Prostate Test) in the context of radiotherapy (RT) for PCa.

METHODS

The members of the Prostate Cancer Working Group of the German Society of Radiation Oncology (DEGRO) conducted an updated literature search on GCs and histopathological biomarkers in PCa, covering original articles published between January 2022 and February 2024 in the PubMed database.

RESULTS

In addition to previous reviews, 11 relevant studies were identified, of which nine studies analyzed biomarkers within prospective phase II or III trials. Eight trials focused on genomic biomarkers, of which three addressed GCs in primary localized PCa, three in recurrent PCa in the setting of salvage RT, and two in metastatic castration-sensitive PCa. In localized PCa, GCs could be validated in a retrospective analysis of randomized controlled trials. Additionally, three studies reported on AI-based histopathology biomarkers.

CONCLUSION

Genomic classifiers and AI-based digital histopathology models might have superior prognostic and predictive value compared to established clinical and pathological parameters in localized, recurrent, and metastatic PCa. Despite promising results, prospective validation of these biomarkers in randomized trials remains limited. This review underscores the need for further prospective trials to confirm the usefulness of these biomarkers in PCa.

Integrating Radiosensitivity Index and Radiation Resistance Related Index Improves Prostate Cancer Outcome Prediction

Purpose

This study aimed to establish a nomogram combining 31-gene signature (31-GS), radiosensitivity index (RSI), and radiation-resistance-related gene index (RRRI) to predict recurrence in prostate cancer (PCa) patients.

Methods and Materials

Transcriptome data of PCa were obtained from gene expression omnibus and the cancer genome atlas to validate the predictive potential of 3 sets of published biomarkers, namely, 31-GS, RSI, and RRRI. To adjust these markers for the characteristics of PCa, we analyzed 4 PCa-associated radiosensitivity predictive indices based on 31-GS, RSI, and RRRI by the Cox analysis and least absolute shrinkage and selection operator regression analysis. Time-dependent receiver operating characteristic curves, decision curve analyses, integrated discrimination improvement, net reclassification improvement and decision tree model construction were used to compare the radiosensitivity predictive ability of these 4 gene signatures. Key modules and associated functions were identified through a weighted gene co-expression network analysis and gene function enrichment analysis. A nomogram was built to improve the recurrence-prediction capability.

Results

We validated and compared the predictive potential of 2 published predictive indices. Based on the 31-GS, RSI, and RRRI, we analyzed 4 PCa-associated radiosensitivity predictive indices: 14Genes, RSI, RRRI, and 20Genes. Among them, 14Genes showed the most promising predictive performance and discriminative capacity. Genes in the key module defined by the 14Genes model were significantly enriched in radiation therapy-related cell death pathways. The area under receiver operating characteristic curve and decision tree variable importance for 14Genes was the highest in the cancer genome atlas and Gene Expression Omnibus Series (GSE) cohorts.

Conclusions

This study successfully established a radiosensitivity-related nomogram, which had excellent performance in predicting recurrence in patients with PCa. For patients who received radiation therapy, the 20Genes and RRRI models can be used to predict recurrence-free survival, whereas 20Genes is more radiation therapy-specific but needs further external validation.

Genomic classifiers and prognosis of localized prostate cancer: a systematic review

BACKGROUND

Refinement of the risk classification for localized prostate cancer is warranted to aid in clinical decision making. A systematic analysis was undertaken to evaluate the prognostic ability of three genomic classifiers, Decipher, GPS, and Prolaris, for biochemical recurrence, development of metastases and prostate cancer-specific mortality in patients with localized prostate cancer.

METHODS

Data sources: MEDLINE, Embase, and Web of Science were queried for reports published from January 2010 to April 2022.

STUDY SELECTION

prospective or retrospective studies reporting prognosis for patients with localized prostate cancer.

DATA EXTRACTION

relevant data were extracted into a customized database by one researcher with a second overreading. Risk of bias was assessed using a validated tool for prognostic studies, Quality in Prognosis Studies (QUIPS). Disagreements were resolved by consensus or by input from a third reviewer. We assessed the certainty of evidence by GRADE incorporating adaptation for prognostic studies.

RESULTS

Data synthesis: a total of 39 studies (37 retrospective) involving over 10,000 patients were identified. Twenty-two assessed Decipher, 5 GPS, and 14 Prolaris. Thirty-four studies included patients who underwent prostatectomy. Based on very low to low certainty of evidence, each of the three genomic classifiers modestly improved upon the prognostic ability for biochemical recurrence, development of metastases, and prostate cancer-specific mortality compared to standard clinical risk-classification schemes.

LIMITATIONS

downgrading of confidence in the evidence stemmed largely from bias due to the retrospective nature of the studies, heterogeneity in treatment received, and era in which patients were treated (i.e., prior to the 2000s).

CONCLUSIONS

Genomic classifiers provide a small but consistent improvement upon the prognostic ability of clinical classification schemes, which may be helpful when treatment decisions are uncertain. However, evidence from current management-era data and of the predictive ability of these tests is needed.

Short-term Darolutamide (ODM-201) Concomitant to Radiation Therapy for Patients with Unfavorable Intermediate-risk Prostate Cancer: The Darius (AFU-GETUG P15) Phase 2 Trial Protocol

BACKGROUND

Combination of androgen deprivation therapy (ADT) with external beam radiation therapy (EBRT) is a standard of care for patients with intermediate-risk prostate cancer (PCa). However, 6 months of ADT generates multiple side effects impacting quality of life (QoL). Darolutamide (an androgen receptor targeting agent [ARTA]) is associated with low blood-brain barrier penetrance and less drug-drug interaction.

OBJECTIVE

To assess the efficacy of a combination of 6 months of darolutamide with EBRT to treat patients with unfavorable intermediate-risk PCa.

DESIGN, SETTING, AND PARTICIPANTS

The DARIUS trial is a multicenter randomized non comparative phase 2 trial, randomizing the 6-months darolutamide + EBRT arm versus 6-months ADT + EBRT in patients with unfavorable intermediate-risk PCa.

OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS

The primary endpoint is a biological response defined as prostate-specific antigen ≤0.1 ng/ml at month six of darolutamide or ADT. The key secondary endpoints are biochemical recurrence-free survival, disease-free survival, safety, and QoL. Ancillary studies using radiomics and genomic classifier are planned. Sixty-two patients will be included.

RESULTS AND LIMITATIONS

In this population of patients requiring ADT combined with EBRT, the use of an ARTA alone, such as darolutamide, may demonstrate antitumoral efficacy while minimizing toxicity and maintaining QoL. Limitations are mainly inherent to the open-label design of this study.

CONCLUSIONS

Six months of darolutamide + EBRT compared with 6 months of ADT + EBRT may be efficient in terms of a biological response, avoiding toxicity and altered QoL attributable to ADT in patients with unfavorable intermediate-risk PCa.

PATIENT SUMMARY

The ongoing DARIUS clinical trial assesses short-term (6 months) darolutamide treatment in association with external beam radiation therapy in men with localized prostate cancer. The trial investigates whether single-agent darolutamide can improve the biological response while maintaining a favorable tolerability profile.

Integration of Genomic Tests in Prostate Cancer Care: Implications for Clinical Practice and Patient Outcomes

Prostate cancer (PCa) is a common malignancy in men and is among the leading causes of cancer-related death worldwide. Genomic tests assess disease aggressiveness and guide treatment, particularly in low- and intermediate-risk PCa. We reviewed the literature on the use of four genomic tests (Prolaris®, Promark®, Oncotype DX®, and Decipher®) in assessing the prognosis of PCa and their use in treatment decision-making. Most of the studies showed that Prolaris® has a strong correlation with biochemical recurrence, metastasis risk, PCa-specific mortality (PCSM), and pathological features. Similarly, three studies on Promark® indicated a connection between results and pathological features in the subsequent prostatectomy, time to metastasis, and biochemical recurrence. Fourteen studies on Oncotype DX® showed a clear correlation between high scores, death, and PCSM. One study found that routine biopsy pathology reports, combined with serum PSA levels, provide a risk assessment comparable to Oncotype DX® testing. Results from 22 studies on Decipher® were controversial. The test was associated with conservative management, suggesting that patients with a high GC score are more likely to need radiation after surgery. Comparative studies indicated that Oncotype DX® is preferable for assessing PCSM, Decipher® for predicting metastasis, and Prolaris® for predicting recurrence. With the incidence rate of PCa dramatically increasing, genomic tests appear to be useful adjunctive precision medicine tools with significant potential in improving prognostic discrimination, facilitating better risk stratification, and guiding personalized treatment, especially in the intermediate-risk patient group. Large-scale, prospective, multi-sectional studies are required to validate the utility of these tests prior to their integration into clinical practice.

Target Volume Optimization for Localized Prostate Cancer

PURPOSE

To provide a comprehensive review of the means by which to optimize target volume definition for the purposes of treatment planning for patients with intact prostate cancer with a specific emphasis on focal boost volume definition.

METHODS

Here we conduct a narrative review of the available literature summarizing the current state of knowledge on optimizing target volume definition for the treatment of localized prostate cancer.

RESULTS

Historically, the treatment of prostate cancer included a uniform prescription dose administered to the entire prostate with or without coverage of all or part of the seminal vesicles. The development of prostate magnetic resonance imaging (MRI) and positron emission tomography (PET) using prostate-specific radiotracers has ushered in an era in which radiation oncologists are able to localize and focally dose-escalate high-risk volumes in the prostate gland. Recent phase 3 data has demonstrated that incorporating focal dose escalation to high-risk subvolumes of the prostate improves biochemical control without significantly increasing toxicity. Still, several fundamental questions remain regarding the optimal target volume definition and prescription strategy to implement this technique. Given the remaining uncertainty, a knowledge of the pathological correlates of radiographic findings and the anatomic patterns of tumor spread may help inform clinical judgement for the definition of clinical target volumes.

CONCLUSION

Advanced imaging has the ability to improve outcomes for patients with prostate cancer in multiple ways, including by enabling focal dose escalation to high-risk subvolumes. However, many questions remain regarding the optimal target volume definition and prescription strategy to implement this practice, and key knowledge gaps remain. A detailed understanding of the pathological correlates of radiographic findings and the patterns of local tumor spread may help inform clinical judgement for target volume definition given the current state of uncertainty.

Hormonal Agents in Localized and Advanced Prostate Cancer: Current Use and Future Perspectives

Prostate cancer (PC) is generally a hormone-dependent tumor. Androgen deprivation therapy ( has been the standard of care in metastatic disease for more than 80 years. Subsequent studies have highlighted the efficacy of ADT even in earlier disease settings such as in localized disease or in the case of biochemical recurrence (BCR). Improved knowledge of PC biology and ADT resistance mechanisms have led to the development of novel generation androgen receptor pathway inhibitors (ARPI). Initially used only in patients who became resistant to ADT, ARPI have subsequently shown to be effective when used in patients with metastatic hormone-naive disease and in recent years their effectiveness has also been evaluated in localized disease and in case of BCR. The objective of this review is to describe the current role of agents interfering with the androgen receptor in different stages of PC and to point out future perspectives.

Association Between the Decipher Genomic Classifier and Prostate Cancer Outcome in the Real-world Setting

BACKGROUND AND OBJECTIVE

Although the prognostic significance of the Decipher prostate cancer genomic classifier (GC) has been established largely from analyses of archival tissue, less is known about the associations between the results of Decipher testing and oncologic outcomes among patients receiving contemporaneous testing and treatment in the real-world practice setting. Our objective was to assess the associations between the Decipher GC and risks of metastasis and biochemical recurrence (BCR) following prostate biopsy and radical prostatectomy (RP) among patients tested and treated in the real-world setting.

METHODS

A retrospective cohort study was conducted using a novel longitudinal linkage of transcriptomic data from the Decipher GC and real-world clinical data (RWD) aggregated from insurance claims, pharmacy records, and electronic health record data across payors and sites of care. Kaplan-Meier and Cox proportional hazards regressions were used to examine the associations between the GC and study outcomes, adjusting for clinical and pathologic factors.

KEY FINDINGS AND LIMITATIONS

Metastasis from prostate cancer and BCR after radical prostatectomy, Decipher GC continuous score, and risk categories were evaluated. We identified 58 935 participants who underwent Decipher testing, including 33 379 on a biopsy specimen and 25 556 on an RP specimen. The median age was 67 yr (interquartile range [IQR] 62-72) at biopsy testing and 65 yr (IQR 59-69) at RP. The median GC score was 0.43 (IQR 0.27-0.66) among biopsy-tested patients and 0.54 (0.32-0.79) among RP-tested patients. The GC was independently associated with the risk of metastasis among biopsy-tested (hazard ratio [HR] per 0.1 unit increase in GC 1.21 [95% confidence interval {CI} 1.16-1.27], p < 0.001) and RP-tested (HR 1.20 [95% CI 1.17-1.24], p < 0.001) patients after adjusting for baseline clinical and pathologic risk factors. In addition, the GC was associated with the risk of BCR among RP-tested patients (HR 1.12 [95% CI 1.10-1.14], p < 0.001) in models adjusted for age and Cancer of the Prostate Risk Assessment postsurgical score.

CONCLUSIONS AND CLINICAL IMPLICATIONS

This real-world study of a novel transcriptomic linkage conducted at a national scale supports the external prognostic validity of the Decipher GC among patients managed in contemporary practice.

PATIENT SUMMARY

This study looked at the use of the Decipher genomic classifier, a test used to help understand the aggressiveness of a patient's prostate cancer. Looking at the results of 58 935 participants who underwent testing, we found that the Decipher test helped estimate the risk of cancer recurrence and metastasis.

Prognostic utility of biopsy-based PTEN and ERG status on biochemical progression and overall survival after SBRT for localized prostate cancer

Introduction/background

Phosphatase and tensin homolog (PTEN) genomic deletions and transmembrane protease, serine 2/v-ets avian erthyroblastosis virus E26 oncogene homolog (ERG) rearrangements are two of the most common genetic abnormalities associated with prostate cancer. Prior studies have demonstrated these alterations portend worse clinical outcomes. Our objective is to evaluate the impact of biopsy-determined PTEN losses and TMPRSS2-ERG fusion on biochemical progression-free survival (bPFS) and overall survival (OS) in patients who receive SBRT for localized prostate cancer.

Methods/materials

Patients received SBRT for localized prostate cancer on a prospective quality-of-life (QoL) and cancer outcomes study. For each patient, the single biopsy core with the highest grade/volume of cancer was evaluated for PTEN and ERG abnormalities. Differences in baseline patient and disease characteristics between groups were analyzed using ANOVA for age and χ2 for categorical groupings. bPFS and OS were calculated using the Kaplan Meier (KM) method with Log-Rank test comparison between groups. Predictors of bPFS and OS were identified using the Cox proportional hazards method. For all analyses, p <0.05 was considered statistically significant.

Results

Ninety-nine consecutive patients were included in the analysis with a median follow-up of 72 months. A statistically significant improvement in bPFS (p = 0.018) was observed for wild type ERG patients with an estimated 5-year bPFS of 94.1% vs. 72.4%. Regarding PTEN mutational status, significant improvements in were observed in both bPFS (p = 0.006) and OS (p < 0.001), with estimated 5-year bPFS rates of 91.0% vs. 67.9% and 5-year OS rates of 96.4% vs. 79.4%. When including both ERG and PTEN mutational status in the analysis, there were statistically significant differences in both bPFS (p = 0.011) and OS (p < 0.001). The estimated 5-year bPFS rates were 100%, 76.6%, 72.9%, and 63.8% for patients with ERG+/PTEN+, ERG-/PTEN+, ERG+/PTEN-, and ERG-/PTEN- phenotypes respectively. The estimated 5-year OS rates were 93.9%, 100%, 80.0%, and 78.7% for patients with ERG+/PTEN+, ERG-/PTEN+, ERG+/PTEN-, and ERG-/PTEN- phenotypes respectively.

Conclusion

ERG rearrangements and PTEN deletions detected on biopsy samples are associated with poorer oncologic outcomes in prostate cancer patients treated with SBRT and merit further study in a dedicated prospective trial.