TY - JOUR T1 - <strong>Selective internal radiation therapy of hepatic tumors: can pre-therapeutic 99mTc-MAA SPECT replace 90Y PET for dosimetric evaluation?</strong> JF - Journal of Nuclear Medicine JO - J Nucl Med SP - 1747 LP - 1747 VL - 59 IS - supplement 1 AU - Andrea Skanjeti AU - Jérémie Tordo AU - Agnes Rode AU - Alexandre Bani Sadr AU - Francesco Giammarile Y1 - 2018/05/01 UR - http://jnm.snmjournals.org/content/59/supplement_1/1747.abstract N2 - 1747Objectives: ntroduction. Glass 90Y microsphere product has been initially approved by FDA for the Selective Internal Radiation Therapy (SIRT) of hepatocellular carcinoma (HCC). It is currently available in the global market and is used for the local treatment of several hepatic tumors. After a baseline angiography and an administration of 99mTc-MAA, a SPECT/CT (SPECT) is frequently performed to predict the post-therapeutic distribution of microspheres in the tumor, in normal liver, pulmonary shunt, and any other extrahepatic distribution. After Glass 90Y microsphere SIRT, 90Y-PET/CT (PET) can be realized due to the positron emission of 90Y. As several dosimetric software packages are now available in clinical practice, we aimed in this study to compare two dosimetric approaches based on SPECT and PET data, respectively, to determine whether 90Y PET/CT is useful. Material and Methods. Ten patients (6 males, median age 65 years old, 6 pts with HCC, 2 pts with hepatic metastases and 2 pts with Cholangiocarcinoma) underwent a total of 12 SIRT (2 patients twice) after contrast enhanced CT, baseline angiography and SPECT. Administration of 99mTc-MAA and SIRT were reproducible as injected by the same trained radiologist. Simplicit90Y™, a dosimetry software was used to corregistrate all the scans and to estimate tumoral irradiation dose as well as whole liver irradiation dose according to SPECT and PET distribution. Medians were compared using Wilcoxon paired test. Overlap in distribution volume of microspheres between PET and SPECT respectively was estimated by the Sørensen Dice Index (SDI) for three different cut-offs: 5%, 10% and 20% of the maximal uptake, respectively. Results. The aimed lobar irradiation was set at 120Gy for all cases excepted two in which it was established at 80Gy for clinical reasons. For each patient, activity to be administered was calculated as suggested by the supplier on the basis of required dose, lobar volume and pulmonary shunt. The median pulmonary shunt was 1.9%, median administered activity was 2.15GBq. No statistical differences between tumoral dose estimated according to SPECT and PET respectively were observed (p&gt;0.05), however, absolute differences ranged from 0.3 to 65.3Gy. No statistical differences between whole liver dose estimated according to SPECT and PET respectively were observed (p&gt;0.05), while absolute differences ranged from 0 to 20.9Gy, they were significantly smaller than absolute differences of the tumoral irradiation dose estimation (p=0.0161). The median overlap for distribution volume of microspheres was estimated at 0.5 for the 20% threshold, 0.67 for the 10% threshold (p=0.001) and 0.76 for the 5% threshold (p=0.005). Conclusion. The overlap of the microsphere distribution based on SPECT and PET is imperfect, in particular for high uptake distribution. Therefore the estimation of tumoral irradiation dose based on SPECT could be biased. Although no significant differences were seen between SPECT-based and PET-based tumoral doses estimations, the absolute differences were noteworthy, while for whole liver doses remain significantly smaller. Due to a small population, a confirmatory study is required, however, given the imperfect overlap of microsphere distribution on SPECT and PET, 90Y PET/CT seems to be useful for precise tumoral irradiation dose estimation. ER -