This study evaluated whether simulated low-dose PET protocols can accurately quantify pancreatic beta-cell mass (BCM), with the specific aim of informing potential imaging in individuals under 18 years old with type 1 diabetes (T1D). Retrospective, full-count dynamic [11C]-(+)-PHNO PET/CT data from adult participants were used to create reduced-count reconstructions and to compare quantitative and semi-quantitative outcome measures against the full-count standard.
The analyses used previously published full-count PET data from eleven adult participants: six healthy controls and five individuals with T1D (combined 6 males, 5 females; ages 22–53). Dynamic [11C]-(+)-PHNO PET/CT scans were originally acquired in list-mode and reconstructed from the full-count datasets for this retrospective evaluation.
To simulate low-dose protocols, the full-count list-mode data were down-sampled to produce reduced-count datasets at 50%, 25%, and 10% of the original counts. Each reduced-count dataset was re-reconstructed using the same reconstruction approach as the full-count data. Pancreas and spleen regions were delineated to extract time-activity curves (TACs) for subsequent kinetic and semi-quantitative analyses.
Pancreas and spleen TACs from each count level were assessed. Volume of distribution (VT, mL/cm3) estimates were derived using a reversible 1-tissue compartment model (1TC) with a tmax of 30 minutes for all count levels. VT estimates from the pancreas and spleen (1TC; tmax = 30 min) were used to compute non-displaceable binding potential (BPND) for pancreas relative to spleen.
Across reduced-count reconstructions, TACs were reported to remain similar to full-count TACs despite visually increased image noise at lower counts. Quantitatively, 1TC VT estimates for reduced-count levels and across evaluated tmax values showed very high concordance with full-count estimates (all R2 ≥ 0.99), indicating robust kinetic parameter recovery from down-sampled data.
The study compared compartmental-model–derived BPND (pancreas vs spleen; 1TC, tmax = 30 min) with a simplified semi-quantitative approach: standardized uptake value ratio minus one (SUVR-1) calculated from 20–30 minutes using spleen as reference. Pancreas SUVR-1 (20–30 min) correlated highly with pancreas BPND across all count levels (all R2 ≥ 0.80). This strong correlation supports the use of simplified reference-region metrics when employing reduced-count or low-dose protocols.
Radiation dosimetry was simulated using adult, adolescent, and pediatric phantoms to estimate mean effective dose equivalents for the reduced injected dose scenarios. For the simulated protocol with 1/10th of the original injected activity, mean effective dose equivalents for all age groups remained below both yearly occupational and research scan radiation dose limits, according to the authors' reported assessments.
Qualitatively, reduced-count images showed progressively increasing noise at the 50%, 25%, and 10% levels compared to full-count images. Despite this visual degradation, extracted TACs and kinetic parameter estimates (VT and derived BPND) remained consistent with full-count results, indicating that increased image noise did not materially affect the quantitative outcomes evaluated in this dataset.
Using adult full-count [11C]-(+)-PHNO PET data down-sampled to simulate reduced-count acquisitions, the investigators found that low-count reconstructions and simplified reference-region analyses provide quantitative results comparable to the full-count standard. Correlations between SUVR-1 and model-derived BPND remained strong across count levels, and dosimetry modeling suggested that a 1/10th injected dose protocol would keep mean effective doses within yearly occupational and research limits for adult, adolescent, and pediatric phantoms. The authors interpret these findings as evidence that accurate BCM quantification using simulated low-dose PET protocols may be feasible for individuals with T1D under 18 years old.
Ethics approvals were obtained from the Yale University Human Investigation Committee and the Yale-New Haven Hospital Radiation Safety Committee, and all participants provided written informed consent. The authors declare no competing interests. The datasets used in this retrospective analysis are available from the corresponding author on reasonable request.
Note: The source article reports results from down-sampled reconstructions of adult scans; direct empirical data from pediatric subjects were not reported in the source and thus pediatric clinical performance was not directly measured in this study.