Multimodal Biomarkers That Predict the Presence of Gleason Pattern 4: Potential Impact for Active Surveillance

D. m. Berman, A. y. Lee, R. Lesurf, P. g. Patel, W. Ebrahimizadeh, J. Bayani, L. a. Lee, N. Boufaied, S. Selvarajah, T. Jamaspishvili, K.-P. Guérard, D. Dion, A. Kawashima, G. m. Clarke, N. How, C. l. Jackson, E. Scarlata, K. Siddiqui, J. b. a. Okello, A. g. AprikianM. Moussa, A. Finelli, J. Chin, F. Brimo, G. Bauman, A. Loblaw, V. Venkateswaran, R. Buttyan, S. Chevalier, A. Thomson, P. c. Park, D. r. Siemens, J. Lapointe, P. c. Boutros, J. m. s. Bartlett

Research output: Contribution to journalArticlepeer-review

Abstract / Description of output

Abstract
Purpose:

Latent grade group ≥2 prostate cancer can impact the performance of active surveillance protocols. To date, molecular biomarkers for active surveillance have relied solely on RNA or protein. We trained and independently validated multimodal (mRNA abundance, DNA methylation, and/or DNA copy number) biomarkers that more accurately separate grade group 1 from grade group ≥2 cancers.
Materials and Methods:

Low- and intermediate-risk prostate cancer patients were assigned to training (n=333) and validation (n=202) cohorts. We profiled the abundance of 342 mRNAs, 100 DNA copy number alteration loci, and 14 hypermethylation sites at 2 locations per tumor. Using the training cohort with cross-validation, we evaluated methods for training classifiers of pathological grade group ≥2 in centrally reviewed radical prostatectomies. We trained 2 distinct classifiers, PRONTO-e and PRONTO-m, and validated them in an independent radical prostatectomy cohort.
Results:

PRONTO-e comprises 353 mRNA and copy number alteration features. PRONTO-m includes 94 clinical, mRNAs, copy number alterations, and methylation features at 14 and 12 loci, respectively. In independent validation, PRONTO-e and PRONTO-m predicted grade group ≥2 with respective true-positive rates of 0.81 and 0.76, and false-positive rates of 0.43 and 0.26. Both classifiers were resistant to sampling error and identified more upgrading cases than a well-validated presurgical risk calculator, CAPRA (Cancer of the Prostate Risk Assessment; P < .001).
Conclusions:

Two grade group classifiers with superior accuracy were developed by incorporating RNA and DNA features and validated in an independent cohort. Upon further validation in biopsy samples, classifiers with these performance characteristics could refine selection of men for active surveillance, extending their treatment-free survival and intervals between surveillance.

Active surveillance (AS) is recommended for men with low- and favorable intermediate–risk prostate cancer.1 Compared to AS for low-risk men, AS for intermediate-risk men would likely benefit from more intensive surveillance to stave off disease progression. Despite increased use of advanced imaging tools, risk calculators, and molecular biomarkers, a third or more of men initially classified as low risk actually have intermediate or higher risk, heralded by subsequent detection of occult Gleason pattern 4.2,3 Strategies to identify such men have limited accuracy. They include attention to traditional risk factors such as age, tumor size and extent, and PSA level, measured by tests such as digital rectal examination, multiparametric (mp) MRI, and biopsy and blood analyses. Despite its increasing use in prostate cancer risk assessment, expert prostate mpMRI is a limited resource with low (circa 59%) sensitivity for intermediate-risk cases.4 A biomarker that more accurately distinguishes between grade group (GG) 1 and GG ≥2 could be helpful in deintensifying AS for men with truly low-risk cancers.

Several commercially available and guideline-approved tests use gene (mRNA or protein) expression levels in prostate cancer biopsies to detect adverse pathology (AP; ie, GG ≥3 or nonorgan-confined disease) in the subsequent prostatectomy. However, no existing molecular test has been adopted in current guidelines as standard of care to distinguish between GG1 and GG ≥2 cancers.1,5,6 Despite indications that such tests could be useful,6,7 uptake has been limited, perhaps because of low accuracy, which in turn may derive from limitations in the number and types of molecular features included in each test. Since cardinal molecular features of early prostate carcinogenesis include not only altered gene expression but also DNA methylation events and copy number alterations (CNAs),8-10 we hypothesized that tests combining these features could provide superior performance in separating low-grade (GG1) cancers from their higher-grade (GG ≥2) counterparts.

The personalized risk stratification for patients with early prostate cancer (PRONTO) program is a pan-Canadian effort that aims to develop a GG classifier to stratify risk in prostate cancer and achieve technical and clinical validation in statistically powered cohorts. Here, we report the development of 2 candidate classifiers comprising different types of molecular features. These classifiers, developed and independently validated, achieve superior performance by integrating tumor mRNA abundance, DNA copy number, and/or DNA methylation profiles. We demonstrate that these classifiers could add value above and beyond routinely captured clinical data and are remarkably resistant to sampling error. We discuss how adoption of classifiers with these attributes has the potential to improve current AS approaches without increasing patient morbidity. By identifying men at increased risk of occult GG ≥2 cancer, surveillance biopsies could be taken earlier to confirm the presence and extent of Gleason pattern 4 cancer. By confirming GG1 cancers, such biomarkers could identify men for whom it would be safe to forgo MRI or increase the intervals between surveillance biopsies, reducing burdens on health care systems and patients.
Original languageEnglish
Pages (from-to)257-271
JournalJournal of Urology
Volume210
Issue number2
DOIs
Publication statusPublished - 1 May 2023

Fingerprint

Dive into the research topics of 'Multimodal Biomarkers That Predict the Presence of Gleason Pattern 4: Potential Impact for Active Surveillance'. Together they form a unique fingerprint.

Cite this