LAUSR

Purpose It is desirable to achieve accurate quantification in PET/CT imaging as routinely assessed using SUV measurement, including for Gallium-68 (68Ga) labelled tracers. A PET/CT scan of a uniformly filled 68Ga phantom on a GE D710 scanner unexpectedly resulted in a low SUV mean of 0.88, outside of the local tolerance of 1.0  ±  5%. A possible cause of underestimation of quantification is inaccuracy of the activity measurement of 68Ga in the dose calibrator associated with the PET/CT scanner. Methods A source of 68Ga (maximum activity of 599 MBq in 4ml in a P6 vial) was measured in 5 Capintec CRC-55t calibrators (including that used with the D710 PET/CT scanner) using the default 68Ga pre-set calibrator factor of 416. The source was then measured in an externally located Fidelis secondary standard calibrator. Based upon the results, manual adjustments were made to the Capintec factors to match the decay corrected Fidelis measurement. A repeat PET/CT scan of a uniformly filled 68Ga phantom was then performed. Results The cross-calibration results showed that the 5 Capintec calibrators systematically overestimated the activity measurement by 8–9% compared to the Fidelis when using the default factor of 416. The factors were adjusted to 457–463 to match the decay corrected Fidelis measurement. After adjustment of factors, a repeat uniform phantom scan resulted in a SUV mean of 0.97, within the local tolerance. Conclusions Underestimation of PET/CT quantification of 68Ga was caused by overestimation of 68Ga activity measurement using the pre-set dose calibrator factor settings on a Capintec CRC-55t. Improvement in quantification accuracy was achieved by adjusting the calibrator factor based upon an external cross-calibration. Similar results have been recently reported by a group in Australia [1] . We recommend SUV checks using a uniform phantom and regular calibrator cross-calibration exercises for all isotopes used clinically or in clinical trials for quantitative PET/CT imaging.