New tool may improve risk stratification of men with high-risk prostate cancer
A proxy for prostate-specific membrane antigen PET/CT upstaging was strongly predictive of meaningful, long-term cancer-specific outcomes, according to results of a multinational cohort study published in JAMA Network Open.
The findings also showed the PSMA PET/CT-specific nomogram, developed by researchers at University of California, Los Angeles and trained solely on radiographic findings on PSMA PET/CT at initial diagnosis, even outperformed existing risk-stratification tools built with input from downstream clinical outcomes, Michael Xiang, MD, PhD, assistant clinical professor in the department of radiation oncology at David Geffen School of Medicine at UCLA, told Healio.
“PSMA PET/CT is a new and exciting technology for the evaluation of patients with prostate cancer; however, due to its recency, how well it actually correlates with meaningful long-term outcomes will not be known for at least several more years as the data mature,” Xiang said. “To address this gap, our study looked at this question from a different angle — using a surrogate or proxy for PSMA PET/CT upstaging based on other known disease characteristics — and investigated the correlation of the proxy to long-term outcomes.”
The analysis included 5,275 men (median age, 66 years; median PSA level, 10.5 ng/mL)) diagnosed with high risk or very high-risk prostate cancer from April 1995 to August 2018. About three-quarters of the men (76%) had Gleason grade 8 to 10 disease, and 14% had stage T3 to T4 disease.
Most of the men (55%) underwent curative-intent radical prostatectomy, 1,669 (32%) with external beam radiotherapy and 723 (14%) with external beam radiotherapy plus brachytherapy, all with or without androgen deprivation.
Xiang and colleagues calculated the PSMA PET/CT upstage probability from the nomogram using the biopsy Gleason score, percentage positive systematic biopsy cores, clinical T category and PSA level. They used Fine-Gray and Cox regressions to analyze biochemical recurrence, distant metastasis, prostate cancer-specific mortality and OS.
Median follow-up was 5.1 years (interquartile range, 3.1-7.9). Nearly a quarter of men (n = 1,221) were followed up for at least 8 years.
Results showed PSMA upstage probability was significantly prognostic of all clinical endpoints, with 8-year concordance (C) indices of 0.63 (95% CI, 0.61-0.65) for biochemical recurrence, 0.69 (95% CI, 0.66-0.71) for distant metastases, 0.71 (95% CI, 0.67-0.75) for prostate cancer-specific mortality and 0.6 (95% CI, 0.57-0.62) for OS.
The PSMA nomogram outperformed existing risk-stratification tools except for similar performance for prostate cancer-specific mortality (eg, distant metastases) to Staging Collaboration for Cancer of the Prostate (8-year C index = 0.69; 95% CI, 0.66-0.71 vs. 0.65, 95%CI, 0.62-0.68). Eight-year C indices for PSMA were 0.57 (95% CI, 0.54-0.6) for the Memorial Sloan Kettering Cancer Center nomogram and 0.53 (95% CI, 0.51-0.56) for Cancer of the Prostate Risk Assessment groups.
“(I was a little surprised) the proxy for PSMA PET/CT actually performed better than existing risk-stratification tools,” Xiang said. “I have no doubt the actual performance of PSMA PET/CT will be even better.”
Secondary cohorts from the SEER database and National Cancer Database validated the study’s results. One of the next steps, Xiang said, is to integrate the PSMA PET/CT proxy and/or actual PET/CT results with other tools, such as genomics-based tests, to further improve risk stratification and prognostication.
“Prospective studies are needed to determine if patients predicted to be at particularly high risk or have a poor prognosis should have their therapy tailored accordingly, such as with strategies aimed at treatment intensification,” Xiang told Healio.
For more information:
Michael Xiang, MD, PhD, can be reached at Department of Radiation Oncology, University of California, Los Angeles, 200 Medical Plaza Drive, Suite #B265, Los Angeles, CA 90095; email: mxiang@mednet.ucla.edu.
Collapse
Xiang M, et al. JAMA Netw Open. 2021;doi:10.1001/jamanetworkopen.2021.38550.