Abstract
241590
Introduction: In the phase 3, randomized VISION study, outcome endpoints all favored 177LuPSMA-617 (LuPSMA) + standard of care (SOC) compared to SOC alone, leading to LuPSMA approval for metastatic castration-resistant prostate cancer (mCRPC). Beyond clinical trial-guided eligibility criteria, it remains challenging to identify those most likely to benefit or experience adverse events (AEs) from newly approved drugs. In VISION, ~50% of men in the LuPSMA arm did not have a PSA-50 response, and ~47% had hematologic AEs. Further, patients (pts) with a "superscan" bone scan were excluded, but the amount of bone disease on PSMA-PET/CT was not considered. This study aimed to evaluate if the extent and heterogeneity of bone disease on PSMA-PET/CT were associated with hematologic AEs and PSA-50 response.
Methods: We queried an IRB-approved prospectively maintained registry to evaluate all pts with mCRPC who received SOC LuPSMA at our institution between 6/22 and 7/23. All met VISION criteria of 1 PSMA-avid lesion (uptake > liver) to be eligible for SOC LuPSMA. Pre-therapy PSMA-PET/CT scans were analyzed for disease burden using the aPROMISE platform (EXINI Diagnostics). The following data were extracted: total tumor SUVmean, total bone volume, total volume of PSMA-avid disease in the bone (voxel SUV > 3), percentage of bone disease with SUV > 10, and percentage of bone disease with SUV 3-10. Clinical data was extracted from the registry and medical record, including dose delays, dose reductions or discontinuations due to hematologic AEs, the reason for stopping LuPSMA, and PSA levels during treatment. PSA-50 response was defined as ≥50% decline in PSA level at any time during LuPSMA therapy. Wilcoxon signed-rank tests were used to compare differences in variables about the extent of bone disease in patients grouped by incidence of hematologic AEs or reasons for stopping LuPSMA and differences in total tumor burden between PSA-50 responders and non-responders. Associations between discretized bone disease heterogeneity and incidence of hematologic AEs or PSA-50 response were assessed by Fisher exact tests.
Results: 97 men (mean age 72 ± 8 years) received at least 1 cycle of LuPSMA (median 4 cycles, range 1-6). 14 pts had dose delays (n=9), dose reductions (n=9) or dose discontinuations (n=6) due to hematologic AEs. Of the 95 pts with PSMA-avid bone disease, the median percentage of total bone volume with PSMA-avid disease was 1.9% (range <0.1%-89%); median percentage of voxels with bone disease > SUV 10 was 17.7% (range 0-65.4%) and with bone disease SUV 3-10 was 53% (range 0-95.6%). Pts with a higher percentage of PSMA-avid bone disease in the total bone volume (=voxel SUV > 3) were more likely to require dose delays or reductions (p<0.001) or dose discontinuation (p=0.006) due to hematologic AEs compared to those with a lower percentage. No significant associations were found between heterogeneity of PSMA uptake in bone disease (percent of voxels with SUV > 10 vs. 3-10) and hematologic AEs. The reason for stopping treatment was completion of all planned 6 cycles (n=36), progressive disease (n=42), toxicity (n=10), and other (n=9). Compared to pts who completed 6 cycles of LuPSMA, a higher volume of bone disease was observed for those who stopped LuPSMA for toxicity (p=0.0025) or progressive disease (p=0.0031). The overall PSA-50 rate was 53% (n=51). Of the 96 pts evaluable for PSA-50, those with a much higher proportion of bone disease with SUV 3-10 over SUV>10 had a significantly lower PSA-50 response rate (30.3% vs. 64.5%, p=0.002). PSA-50 responders were more likely to have a higher total tumor SUVmean (median 8.7, range 2.8-22.6) vs. PSA-50 non-responders (median 6.5, range 2.8-15.6, p=0.01).
Conclusions: This data validates total tumor SUVmean as a predictor of PSA-50 response. Both total volume and heterogeneity of PSMA-avid bone disease appear associated with the likelihood of PSA-50 response. Not unexpectedly, a larger percentage of total volume of bone involved with PSMA-avid disease was associated with hematologic AEs and the reason for stopping LuPSMA. These findings may be helpful to guide clinical decision-making around dosing of LuPSMA therapy in routine practice and to select pts for clinical trials.
