Metastatic castration-resistant prostate cancer has a high rate of mortality with a limited number of effective treatments after hormone therapy. Radiopharmaceutical therapy with 177LuLu-prostate-specific membrane antigen-617 (LuPSMA) is one treatment option; however, response varies and is partly predicted by PSMA expression and metabolic activity, assessed on 68GaPSMA-11 or 18FDCFPyL and 18FFDG PET, respectively. Automated methods to measure these on PET imaging have previously yielded modest accuracy. Refining computational workflows and standardizing approaches may improve patient selection and prognostication for LuPSMA therapy. Methods: PET/CT and quantitative SPECT/CT images from an institutional cohort of patients staged for LuPSMA therapy were annotated for total disease burden. In total, 676 68GaPSMA-11 or 18FDCFPyL PET, 390 18FFDG PET, and 477 LuPSMA SPECT images were used for development of automated workflow and tested on 56 cases with externally referred PET/CT staging. A segmentation framework, the Global Threshold Regional Consensus Network, was developed based on nnU-Net, with processing refinements to improve boundary definition and overall label accuracy. Results: Using the model to contour disease extent, the mean volumetric Dice similarity coefficient for 68GaPSMA-11 or 18FDCFPyL PET was 0.94, for 18FFDG PET was 0.84, and for LuPSMA SPECT was 0.97. On external test cases, Dice accuracy was 0.95 and 0.84 on PSMA and FDG PET, respectively. The refined models yielded consistent improvements compared with nnU-Net, with an increase of 3%-5% in Dice accuracy and 10%-17% in surface agreement. Quantitative biomarkers were compared with a human-defined ground truth using the Pearson coefficient, with scores for 68GaPSMA-11 or 18FDCFPyL, 18FFDG, and LuPSMA, respectively, of 0.98, 0.94, and 0.99 for disease volume; 0.98, 0.88, and 0.99 for SUVmean; 0.96, 0.91, and 0.99 for SUVmax; and 0.97, 0.96, and 0.99 for volume intensity product. Conclusion: Delineation of disease extent and tracer avidity can be performed with a high degree of accuracy using automated deep learning methods. By incorporating threshold-based postprocessing, the tools can closely match the output of manual workflows. Pretrained models and scripts to adapt to institutional data are provided for open use.
Jackson et al. (2025) studied this question.