SUV variability in EARL-accredited conventional and digital PET

Area of Science:

• Nuclear Medicine
• Medical Imaging
• Oncology

Background:

• Standardization of SUV measurements is critical for multi-center PET studies.
• Evaluating SUV variability between conventional and digital PET scanners is essential for reliable cancer imaging.
• This study compares SUV reproducibility using EARL-accredited scanners.

Purpose of the Study:

• To assess SUV variability in whole-body FDG-PET scans between conventional (cPET) and digital (dPET) scanners.
• To compare SUVmean, SUVpeak, and SUVmax differences using EARL-compatible reconstructions.
• To evaluate the impact of scanner technology on SUV measurements in cancer patients.

Main Methods:

• Prospective study comparing 50 patients' FDG-PET scans from cPET and dPET scanners.
• Utilized EARL-accredited protocols and reconstructions for both scanner types.
• Measured SUVmean, SUVpeak, SUVmax, and lesion diameter in up to 5 lesions per patient.
• Calculated relative SUV differences (ΔSUV) and repeatability coefficients.

Main Results:

• Included 128 lesions (average size 19 ± 14 mm).
• Average ΔSUV ranged from 6-8%, with higher values on dPET (p < 0.001).
• ΔSUVmax was significantly higher than ΔSUVmean and ΔSUVpeak (8% vs. 6-7%, p < 0.05).
• Repeatability coefficients were 27% for ΔSUVmean/peak and 33% for ΔSUVmax (p < 0.001).

Conclusions:

• Limited SUV variability (up to 8%) was observed between conventional and digital PET scanners with EARL accreditation.
• Few lesions exhibited SUV differences exceeding 30%, supporting scanner comparability.
• EARL standardization effectively ensures consistent SUV measurements across different PET scanner technologies.