1258 - PSMA-PET Avidity and Prostate Cancer-Specific Mortality in Men with Gleason =8 Disease Presenting with Very Low PSA Levels

11:15am - 11:20am ET

Room 107

Presenter(s)

Zachary McSween, MD, MS - Icahn School of Medicine at Mount Sinai, New York, NY

Z. A. McSween^1,2, S. I. Goldberg³, P. Naik³, V. Nambi², I. Pompa⁴, G. E. Cerrato², J. A. Efstathiou³, T. S. C. Ng⁵, and S. C. Kamran²; ¹St. George's University School of Medicine, New York, NY, ²Department of Radiation Oncology, Massachusetts General Hospital, Boston, MA, ³Massachusetts General Hospital, Boston, MA, ⁴Department of Radiation Oncology, Massachusetts General Hospital, Harvard Medical School, Boston, MA, ⁵Department of Radiology, Massachusetts General Hospital, Harvard Medical School, Boston, MA

Purpose/Objective(s): Increased prostate cancer-specific mortality (PCSM) has been observed in men presenting with low prostate-specific antigen (PSA) levels and Gleason score (GS) =8 disease in prior decades using older diagnostic and treatment methods. The objective of this study was to determine whether this association persists in the contemporary era, and to assess radiotracer uptake (if any) on novel molecular scans, given the low presenting PSA levels.

Materials/Methods: The Surveillance, Epidemiology, and End Results program identified 469,045 men diagnosed with localized prostate cancer (PC) between 2011-2022. A multivariate Fine-Gray competing risks regression analysis assessed PCSM as a function of PSA and GS (8-10 vs =7). To further characterize the biology underlying this association, men with PSA = 4 ng/mL and pretreatment PSMA-PET scans were evaluated at a single institution to assess prostate radiotracer uptake. Prostate SUVmean/max values were compared between GS 8-10 vs =7 using t-test.

Results: Median follow-up was 50 months. Among 112,543 men with GS 8-10 PC, using PSA 4.1-10 ng/mL as referent, adjusted hazard ratios for PCSM were: 2.64 for PSA =2.5 ng/mL (95% CI 2.35-2.97, p<0.001), 1.62 for PSA 2.6-4 ng/mL (95% CI 1.44-1.83, p<0.001), 1.61 for PSA 10.1-20 ng/mL (95%CI 1.52-1.71, p<0.001), 2.55 for PSA 20.1-40 ng/mL (95% CI 2.40-2.71, p<0.001), and 4.98 for PSA >40 ng/mL (95% CI 4.73-5.25, p<0.001). This U-shaped distribution was not observed among 356,502 men with GS =7 PC, where adjusted HRs were 1.31 for PSA =2.5 ng/mL (95% CI 1.10-1.56, p=NS) and 0.75 for PSA of 2.6-4 ng/mL (95% CI 0.63-0.90, p=NS). A significant interaction between PSA and GS (p-interaction<0.001) indicated that PSA =2.5 ng/mL predicted poorer PCSM only in men with GS 8-10 PC.

Separately, a single-institution analysis identified 45 PC patients with PSA =4mg/mL and pretreatment PSMA-PET scans. Twenty-eight (62%) had GS =7 PC while the remaining 17 had GS 8-10 PC. The prostate SUVmean and SUVmax among those with GS 8-10 were significantly higher than those with GS =7 (SUVmean GS 8-10: 9.1, range 1.4-36.6 vs GS =7: 4.1, range 0.3-10.9, p=0.017; SUVmax GS 8-10: 19.3, range 1.8-81.4 vs GS =7: 7.4, range 1.4-23.9, p=0.02).

Conclusion: In the modern diagnostic and treatment era, the previously observed U-shaped distribution between PCSM and PSA level among men with GS 8-10 disease remains valid. Men with PSA =4 ng/mL have a higher risk of PCSM than those with PSA levels of 10.1-20 ng/mL, an association not observed among men with GS =7 disease, suggesting dedifferentiation-driven tumor aggressiveness. Despite low PSA levels at presentation, these tumors appear to still exhibit SUV uptake on PSMA-PET, with significantly increased uptake noted in men with GS 8-10 PC. Novel molecular scans remain informative in patients presenting with low PSA levels.