Introduction
Targeted radionuclide therapy with Lutetium-177 labelled PSMA (177Lu-PSMA) has emerged as a leading option for metastatic castration-resistant prostate cancer (mCRPC). Traditional dosimetry often relies on multiple time-point imaging to capture radiopharmaceutical pharmacokinetics, which can create a substantial imaging burden, workload, and costs. Single time-point (STP) dosimetry, particularly at about 48 hours post-administration, offers a practical alternative for estimating absorbed doses to kidneys and tumours. This article discusses a phantom-validated workflow and its clinical integration to enable fast SPECT acquisitions for STP dosimetry in 177Lu-PSMA–treated patients.
Phantom study: calibrating fast SPECT for quantitative dosimetry
A dual-head SPECT/CT system (GE Discovery 870 DR) with a MEGP collimator was used. A Jaszczak phantom with known 177Lu activity quantified the camera’s response, enabling a camera-specific calibration factor in line with MIRD Pamphlet No. 26. The goal was to translate raw counts into accurate activity concentrations for subsequent dose calculations.
Acquisition parameters were optimized for speed: 5 seconds per frame, 60 frames per head, totalling ~7 minutes per bed. Reconstructed data used OSEM (16 iterations, 9 subsets) with a 128×128 matrix, incorporating a Gaussian model to mitigate collimator–detector blur. Attenuation correction employed a low-dose CT scan. This setup laid the groundwork for reliable single-time-point dosimetry at 48 hours after tracer injection.
Clinical protocol: STP dosimetry at 48 hours
Eight treatment cycles across five patients were imaged at ~48 hours post-administration, with 177Lu-PSMA I&T activities ranging from 5.6 to 8.1 GBq (mean 6.8 GBq). The same imaging and reconstruction workflow from the phantom study was applied to patient scans to ensure consistency. STP voxel-based dosimetry was then performed with the Hanscheid method in Hermes Dosimetry Software to estimate cumulated activity and absorbed doses in kidneys and tumours.
The Hanscheid dual-time-point concept was adapted for a single-measurement approach, using the 48 h image to approximate the activity present at that time and to project absorbed doses via established conversion factors and Monte Carlo dose simulations.
Results: dose distribution in kidneys and tumours
Kidney doses remained within a narrow, acceptable range: mean 2.04 Gy with a standard deviation of 0.37 Gy (range 1.71–2.67 Gy). This supports renal safety given the commonly cited renal tolerance around 23 Gy. In contrast, tumour doses showed substantial inter- and intra-patient variability (n=25 tumours across eight cycles): 0.98 to 16.05 Gy, mean 5.14 Gy, with a standard deviation of 3.81 Gy. Most tumours received higher doses than kidneys, underscoring the targeted efficacy of 177Lu-PSMA therapy in heterogeneously PSMA-expressing lesions.
Within individuals, tumour doses reflected site-specific biology. For example, some skeletal and lung lesions exhibited high absorbed doses in certain cycles (up to 16.05 Gy in the lumbar spine for one patient), while others showed lower uptake. Such variability highlights the potential relevance of individualized dosimetry for tailoring treatment planning and monitoring response.
Clinical implications and discussion
STP dosimetry combined with fast SPECT acquisitions offers a practical route to quantify therapeutic dose with reduced imaging burden. The kidney doses observed align with safety expectations, while the wider range of tumour doses echoes the heterogeneity of metastatic disease and supports the use of patient-specific dosimetry to guide subsequent cycles and dose adjustments. These findings complement known correlations between higher tumour doses and PSA responses in prospective studies, suggesting that STP approaches can help identify patients who may benefit from dose optimization or alternative radiopharmaceutical strategies in scenarios of suboptimal uptake.
Limitations and future directions
The study’s small sample size limits statistical power. Fast acquisitions may marginally affect image quality and quantitative accuracy, particularly for small or low-uptake lesions. Ongoing work aims to confirm these findings in larger cohorts and across different scanners, and to refine calibration schemes and reconstruction parameters to further stabilize dose estimates from a single 48-hour SPECT snapshot. Prospective trials could also evaluate whether STP-based dosing improves clinical outcomes and resource utilization compared with traditional multi-time-point dosimetry.
Conclusion
Fast SPECT acquisitions enabling STP dosimetry for 177Lu-PSMA in mCRPC demonstrate feasibility and clinical relevance. By delivering robust absorbed dose estimates at a single, strategically chosen time point, this approach can reduce imaging and workflow burdens while preserving patient safety and treatment efficacy. The transition from phantom validation to routine clinical practice marks a meaningful step toward more efficient, dosimetry-guided nuclear medicine care.
