Abstract
383
Objectives: Accurate dosimetry is essential to 177Lu-DOTATATE peptide receptor radionuclide therapy (PRRT) of inoperable neuroendocrine tumors (NET). The usual dosimetry method is to fit a mono-exponential function to the organ’s time-activity curve (TAC) obtained on SPECT/CT, assuming that organ/lesion-specific TAC would reach peak activity within 1-2 hours post injection of 177Lu-DOTATATE and then decrease exponentially. However, our kinetic data showed that individual TACs were often multi-phasic, therefore very different from a simple mono-exponential clearance, thus leading to inaccurate or erroneous dose calculation. We aimed to develop a more reliable and clinically practical dosimetry method based on 3-time-point SPECT/CT.
Methods: During year 2015-17, 23 consecutive patients (male: 12, female: 11; mean age: 56±14 y) with inoperable somatostatin receptor-expressing NET were treated by 177Lu-DOTATATE co-infused with amino acid solution for renal protection. 177Lu-DOTATATE was infused for 30 min followed by a 15-min rinse of the tubing with physiological saline. Post-treatment SPECT/CT (window, 113 and 208±15% KeV, 30 projections, 180°, 20 s/projection, 128×128) was performed at 4, 24 and 48 h. Regions of interest (ROI) of 3-cm spheres were placed in normal kidney, liver, bone marrow and the NET ROI was drawn automatically (MIMContouring 6.4) on 4h SPECT/CT. All ROIs were copied to 24 and 48h SPECT/CT to generate a 3-time-point TAC for each ROI. The TAC data were calibrated with the known 177Lu source saline (4.96±1.26 MBq in 100 ml) measured on the same day with the same camera settings as in the patient study. The resulting TAC was divided into 2 phases: 4-24 h, 24-48 h and fitted to 2 mono-exponential functions respectively. The cumulated activity (A) was then calculated as: (∫024h 1st exponential function) + (∫24h∞ 2nd exponential function). The absorbed dose (D) was measured based on MIRD scheme: D=A×S value.
Results: The mean injected activity (IA) was 6.50±1.25 GBq (range: 3.70-7.99 GBq). The proposed bi-phasic exponential model was applicable to all patients. The liver D was 0.94±0.84 Gy, right kidney 1.91±1.24 Gy, left kidney 1.83±1.23 Gy, bone marrow 0.14±0.08 Gy. The metastatic NET sites were found in liver (15/23=65%), bone (13/23=57%), lymph node (11/23=48%), soft tissue (5/23=22%) and lung (2/23=9%). The absorbed D to the metastatic NET in liver was 7.91±3.17 Gy, bone 1.89±0.31 Gy, lymph node 2.96±2.14 Gy, soft tissue 1.46±0.69 Gy, lung 0.93±0.11 Gy. The mono-exponential model was either unable to fit or not applicable to calculate the absorbed D in 4/23 (17%) patients, and there was >20% absorbed D difference between the 2 models in 4/23 (17%) liver, 5/23 (22%) right kidney, 4/23 (17%) left kidney, and 9/23 (39%) axial marrow. Conclusions: The proposed bi-phasic dosimetric approach using 3-time-point SPECT/CT is more accurate in absorbed dose calculation for 177Lu-DOTATATE PRRT compared with the mono-phasic exponential model. It can be easily applied in clinical practice to provide better dosimetry monitoring especially in multi-cycle therapeutic regimen.