3350 - Per Lesion Metabolic Change on Serial PSMA PET Imaging for Metastatic Prostate Cancer Patients
Presenter(s)
N. Syed1, J. J. Shi1, W. Issa2, C. Shen3, A. Garant3, N. B. Desai3, R. Hannan3, T. Zhang2, J. Wang3, and D. X. Yang3; 1School of Medicine, University of Texas Southwestern Medical Center, Dallas, TX, 2Department of Hematology and Oncology, University of Texas Southwestern Medical Center, Dallas, TX, 3Department of Radiation Oncology, University of Texas Southwestern Medical Center, Dallas, TX
Purpose/Objective(s): In recent years, there has been a rapid increase in utilization of PSMA PET imaging for initial and subsequent re-staging of metastatic prostate cancer. While preliminary analysis is underway evaluating serial PSMA PET response criteria in Prostate Cancer Clinical Trials Working Group 4 (PCWG4), there is currently a lack of consensus guidelines and patterns of per lesion response are not yet well characterized.
Materials/Methods: We conducted a single-institution retrospective analysis of patients with metastatic prostate cancer who underwent serial PSMA PET imaging while receiving systemic therapy. SUVmax values of all metastatic lesions were recorded and each lesion's % SUVmax change between images was calculated. Lesions were classified as complete response (CR) if no longer identified on radiology report, as partial response (PR) if at least 30% reduction in SUVmax, and as non-response (NR) otherwise (stable or progressive lesions). PSA levels closest to each PSMA PET scan were recorded.
Results: A total of 94 lesions for 19 patients were evaluated across 38 paired Ga68 PSMA PET/CT scans. Median time between PSMA PET scans were 9.2 months [IQR 5.8 to 13.3]. 13 patients received androgen deprivation therapy alone, with the other patients receiving additional systemic therapy. There were 29 lymph node, 46 bone, 11 visceral, and 8 prostate/locally recurrent lesions. The median initial PSA was 4.32 and the median post-therapy PSA was 0.73. The median initial SUVmax for CR, PR, and NR lesions was 4.65 [IQR 3.05 to 8.30], 13.05 [IQR 6.40 to 27.70], and 5.03 [IQR 3.0 to 16.55], respectively. The median re-staging SUVmax for PR and NR lesions after systemic therapy were 3.61 [IQR 2.31 to 6.00] and 11.51 [IQR 3.80 to 20.66]. The median % SUVmax change for PR and NR lesions were -64.56% [IQR -74.51 to -54.41] and 10.94% [IQR -10.99 to 110.17]. For 8 patients with >90% PSA decline between serial PSMA PET scans, the distribution of CR, PR, and NR lesions were 27.08%, 35.00%, and 37.92%, respectively. For the patients without a deep PSA response, there were 10.17% CR, 42.90% PR, and 46.93% NR lesions (p <0.01).
Conclusion: Although there was variation between individual lesions, overall PSA response appears to be correlated with per lesion metabolic changes on serial PSMA PET imaging as demonstrated by a significant increase in the proportion of CR lesions. These findings provide a benchmark for evaluating metabolic imaging response in metastatic prostate cancer and warrant further study in larger, prospective cohorts.