EARL-based quantitative evaluation of administered activity reduction for [68Ga]Ga-DOTATOC PET/CT on a 32-cm axial FOV digital BGO system: Phantom analysis and retrospective list-mode clinical assessment

Purpose The increased sensitivity of long-axial field-of-view (FOV) digital PET/CT systems enables potential reductions in administered activity and/or acquisition time with respect to standard-field-of-view systems and established clinical protocols, while preserving quantitative performance. This study aimed to evaluate the feasibility of reducing administered activity for [68Ga]Ga-DOTATOC PET/CT on a 32-cm axial FOV digital bismuth germanium oxide (BGO) system using an EARL-based phantom methodology, complemented by a preliminary clinical assessment.Methods A NEMA IEC body phantom was acquired at progressively decreasing 68Ga activity concentrations; recovery coefficients (RCmean and RCmax) were compared with EARL 18F Standard 1 limits and published 68Ga-specific reference curves. The influence of reconstruction settings on harmonization compliance was investigated. Clinical evaluation was performed in two patients by retrospectively simulating reduced-count conditions and assessing image quality using a 5-point Likert scale evaluated by six experienced readers.Results Phantom results demonstrated compliance with EARL 18F Standard 1 recovery limits down to approximately 1.7 kBq/mL for 2 min/bed acquisitions. At 0.8 kBq/mL, corresponding to an administered activity of 1.5 MBq/kg, background variability remained acceptable despite limited RCmax deviations. Comparison with published reference curves showed good agreement for RCmean, whereas RCmax compliance was lower. Reconstruction settings substantially influenced harmonization, with no single reconstruction achieving uniform agreement. Clinical analysis showed that count reductions of approximately 12-24 % did not significantly impair perceived diagnostic image quality.Conclusions On the evaluated 32-cm axial FOV digital BGO PET/CT system, administered activity reductions toward 2.0 MBq/kg appear feasible while preserving acceptable quantitative performance and clinical image quality.