(UroToday.com) The 2024 American Urological Association (AUA) Annual Meeting held in San Antonio, TX between May 3rd and 6th, 2024 was host to a prostate cancer moderated poster session. Dr. David Han presented the results of a study evaluating high Decipher® scores for defining the subgroup most at risk of metastatic progression among patients with lower-grade tumors classified as NCCN high-risk based on elevated prostate-specific antigen (PSA) level alone.
Patients with a serum PSA level >20 ng/ml at the time of prostate cancer diagnosis meet the criteria for NCCN high-risk disease, independent of Grade Group or clinical stage, and are recommended for radical therapy. While high-risk prostate cancer patients generally have worse prognoses, there is evidence to suggest that this risk group is a heterogeneous population with variable cancer-specific outcomes.
The AUA recommends selective use of genomic biomarkers for risk stratification – only when they are likely to alter clinical decision-making. Genomic classifiers such as Decipher® have not been compared to traditional PSA cutoffs (e.g., >20 ng/ml for high-risk classification) for the prediction of metastasis among otherwise low- and favorable intermediate-risk patients, based on other clinicopathologic variables. Understanding the prognostic value of genomic tests such as Decipher® in these patients may allow for improved risk stratification, potentially lowering their risk status, and sparing them unnecessary overtreatment.
The objective of this study was to assess whether Decipher® score better predicts a higher risk of metastatic progression, compared to serum PSA. This was a retrospective, multicenter, individual-level analysis assessing Decipher® score post-prostatectomy. Men with pre- and post-prostatectomy Grade Group 1 or 2 disease with any range of pre-operative PSA levels (NCCN very low- to high-risk) were included. The primary outcome was distant metastasis. The association between each of Decipher® scores and PSA level with the rate of distant metastasis was evaluated using competing risks analysis with univariable and multivariable Fine and Gray regression modeling. Model-based estimates of 5- and 10-year risks of prostate cancer distant metastases were calculated.
The baseline patient characteristics are summarized below (n=453). 9% of patients had a serum PSA >20 ng/ml. Overall, 82%, 13%, and 5% had low, intermediate, and high Decipher® scores, respectively.
On multivariable modeling, a higher Decipher® score was strongly associated with the rate of metastasis, irrespective of whether the Decipher score was operationalized as a continuous (sHR: 1.41, p=0.004) or categorical variable (sHR for high versus low: 6.82, p=0.001). Conversely, a higher PSA level (either continuous as log[PSA] or dichotomized) was not associated with an increased metastatic rate.
Model-based estimates of 5- and 10-year metastasis rates by Decipher® score and PSA are shown below. If the PSA was >20 and the Decipher® score was <0.6, the mean 5- and 10-year metastasis rates were 1.8% and 5.6%, respectively. 
Dr. Han and colleagues concluded that the Decipher® score appears to better predict distant metastases, compared to PSA as the only high-risk feature among patients with Grade Group 1 or 2 disease after prostatectomy. Decipher® score may be a useful adjunctive tool when advising these patients on subsequent prognosis and management, as not all patients with PSA >20 are at high risk for metastases.
Presented by: David Han, MD, Resident Physician, Columbia University Irving Medical Center, New York, NY
Written by: Rashid Sayyid, MD, MSc - Society of Urologic Oncology (SUO) Clinical Fellow at The University of Toronto, @rksayyid on Twitter during the 2024 American Urological Association (AUA) Annual Meeting, San Antonio, TX, May 3rd - 6th, 2024