Abstract
2811
Introduction: 177LuPSMA-617 is an effective treatment in metastatic, castration-resistant prostate cancer (mCRPC), but response rates and survival might be improved by early assessment of response. This study investigates the feasibility of serial quantitative 177Lu-SPECT/CT imaging in monitoring treatment response, and correlates SPECT derived imaging parameters with survival.
Methods: 56 men with mCRPC progressing after chemotherapy and a novel androgen signaling inhibitor (enzalutamide and/or abiraterone) were enrolled in the LuPIN trial evaluating up to 6 doses of 177LuPSMA-617 plus a radiation sensitizer (NOX66). Imaging with 68Ga-PSMA-11 was done at study entry and exit; and 177Lu-SPECT/CT (SPECT) was acquired 24hrs after each treatment. The SPECT system was calibrated using a cylindrical phantom filled with water and 296MBq at time of acquisition. SPECT data was reconstructed using SPECTRA Quant software (MIM Software Inc., Cleveland, USA). SPECT quantitative imaging analysis was done after the first and third doses of study treatment (C1 and C3). 177Lu-SPECT/CT and PSMA PET/CT scans were analysed semi-quantitatively using a standardised semi-automated workflow to delineate regions of interest with a minimum SUV cut-off of 3. Whole body quantitation derived total tumour volume, SUVmax, SUVmean, and total lesional activity. C1 177Lu-SPECT imaging analysis was repeated on the same acquisition to measure test-retest differences and calculate repeatability estimates using a linear mixed model that accounted for variance in score at the patient level.
Results: 32/56 men had analysable serial 177Lu-SPECT/CT imaging after both C1 and C3. In this subgroup, median PSA-PFS was 6 months (95%CI 5-10) and median OS 12 months (95%CI 12-24). A 50% reduction in serum PSA was observed in 20/32 (63%). Using whole-body quantitative imaging analysis, there was a strong correlation between baseline PSMA PET derived total tumour volume and C1 177Lu-SPECT total tumour volume (R = 0.87 (95% CI 0.74-0.93), p < 0.001). Repeatability of 177Lu-SPECT/CT quantitative analysis was assessed in all patients. Repeatability estimates were high for all parameters. Repeatability for SUVmax was 0.99 (95%CI 0.97-0.99), SUVmean 0.90 (95%CI 0.81-0.95) and tumour volume 0.99 (95%CI 0.98-0.99).
On serial imaging analysis of C1 and C3 177Lu-SPECT/CT, measures of PSMA intensity were reduced between C1 and C3 in most participants. SUVmax was reduced (median -28.9, (range -195 to +42)) in 29/32 (91%) of participants. SUVmean was reduced (median -2.6 (range -12 to +10)) in 27/32 (84%). A change in SUVmax or SUVmean was not associated with PSA-PFS. SPECT total tumour volume (TTV) was reduced in 68% (22/32, median -0.20L (-1.4 to -0.001) and increased in 31% (10/32, median 0.36L (0.1-1.4)). An increase in SPECT TTV was associated with shorter PSA-PFS (HR 4.1 (95% CI 1.5-11.2), p 0.006). PSA progression by C3 was also significantly associated with shorter PSA-PFS (HR 26.5 (95%CI 5.4-131).
Conclusions: We found that serial 177Lu-SPECT/CT quantitative analysis is feasible. An increase in SPECT TTV at cycle 3 was associated with shorter PSA-PFS, and might be a useful early indicator of limited response to 177LuPSMA-617 therapy.