Quantitative 177Lu-SPECT Dosimetry

Dosimetry was performed using quantitative ¹⁷⁷Lu-SPECT scans of the abdomen 24, 48, and 72 h postinjection. The images were acquired over 15 min on a dual-headed Symbia T2 SPECT/CT system (Siemens Healthineers, Erlangen, Germany) as described previously [12 (link),15 (link)]. For attenuation correction, a low-dose CT scan (AC-CT) was acquired together with the first SPECT scan 24 h postinjection. The AC-CT was coregistered on the SPECT scans acquired 48 and 72 h after therapy (rigid body co-registration, PMOD Version 3.609, PMOD Technologies, Zurich, Switzerland). Quantitative SPECT reconstruction employed an in-house maximum a posteriori (MAP) algorithm with 20 iterations, 16 subsets, and a penalty factor of 0.001, as described previously [8 (link),10 (link)]. Scatter correction and resolution decompensation was applied in addition to the mentioned attenuation correction. For each patient, three to five lesions with the highest visual uptake in abdominal SPECT were evaluated. Tumor lesions were segmented semi-manually by placing volumes of interest (VOIs) with an iso-contour of 40% as proposed by Collarino et al. and confirmed by in-house phantom studies [18 (link)]. Lesion dosimetry was based on a mono-exponential fit model and mass-scaled sphere S-values (Figure 2).