Most clinical PET scans involve radiotracers labeled with 18F. A number of radiotracers labeled with nuclides other than 18F have been introduced, including 68Ga-labeled PSMA, 68Ga- and 64Cu-DOTATATE, and 89Zr-labeled antibodies or minibodies. These "non-standard" radionuclides have complex decay properties that are challenging for quantitative PET imaging (e.g., low signal due to low positron fraction and/or long half-life that limits injected activity, cascade gammas). The high sensitivity of a long axial field-of-view (AFOV) system can mitigate the low signal, but the wide axial angle coupled with the additional gammas can lead to more scatter and random events that pose a challenge for conventional corrections. In this work we imaged uniform cylindrical phantoms filled with non-standard radionuclides (52Mn, 55Co, 64Cu, 68Ga, 86Y, 89Zr) on the PennPET Explorer (112-cm AFOV) to assess quantitative accuracy and image uniformity in comparison to 18F. Lowering the upper energy threshold (Emax) was studied to reduce contributions from higher energy gammas. Data were reconstructed using the time-of-flight list-mode ordered subsets expectation maximization algorithm with standard corrections for scatter and random coincidences. Uptake recovery was 5-10% of the true activity for radionuclides without prompt gammas, whereas inaccurate activity estimates and significant image non-uniformities were observed for 52Mn, 55Co and 86Y. Lowering Emax and adding a DC offset to the scatter estimate to compensate for the cascade gammas led to improved accuracy and uniformity, but further refinements of the corrections are needed. This work will pave the way for quantitative human studies with these non-standard radionuclides.
