Impact Of Inter-Cycle Changes In Dose Heterogeneity, Biokinetics And Tumour Volume On Dose-Based Personalisation Of Administered Activity For 177lu-Based Treatment

268 Aim: The application of dosimetry as a clinical tool to adapt the administered activity of multi-cycle 177Lutetium-based peptide receptor radionuclide therapy (PRRT) remains one of the most controversial issues within the nuclear medicine community. Recent reports to optimise AA per cycle based on the predicted dose to the kidneys (dose-limiting organ) have been promising, however, critics claim an avoidance of biology in this approach. Indeed, the assumption of identical intra-cycle kinetics, tumour volume and dose distribution allows for a clinically-feasible dosimetry approach but may lead to overestimation or underestimation of the optimal AA per cycle. A further step towards AA optimisation of multi-cycle PRRT would be to investigate the influence of earlier cycles on the biokinetics, dose distribution and tumour volume of subsequent cycles. This step requires 3D voxelised dosimetry, longitudinal SPECT/CT after each cycle and radiobiologic modelling to account for the radiobiological effects of the changes. Here, we describe a simulation study based on actual dosimetry data from patients who had undergone four-cycles of weight-based Lu177-DOTATATE therapy as part of the LuDo trial (SRCTN98918118)[4]. We determine whether the predictions of previously reported dose-based optimisation strategies prove robust to variable intra-cycle biokinetics, dose distribution and tumour volume. 3 children with relapsed or primary refractory neuroblastoma were assessed for suitability for 177Lu-DOTATATE therapy by imaging with 68Ga-DOTATATE PET/CT. After each cycle, four SPECT/CT images of the abdomen region were acquired at nominal/approximately times of 2, 24, 48 or 72h post-injection. In this study, we used both rigid and deformable image registration for each clinical case to register the 4 sequential SPECT/CT scans in the same frame of reference of the first CT scan. Dose rate images were calculated for each time point from the SPECT activity images using a local energy deposition (LED) approach. A voxel-by-voxel fit to an mono-exponential expression was then applied to the aligned dataset to obtain the clearance half-time for each voxel, and the absorbed dose calculated by analytical integration. Once the physical indices of spatial absorbed dose distribution were computed, the radiobiological metrics of BED and EUD were calculated at the voxel and organ level respectively[5]. In this study, there is considerable deviation between the predicted intra-cycle kidney absorbed dose per AA and those calculated based on the SPECT/CT data. The intra-cycle kidney pharmacokinetics as quantified by the effective half-life increased 30% between cycles 1 and 4 while the heterogeneity of the dose distribution changed significantly within each cycle based on the EUD radiobiological metric. Patient-specific dosimetry simulations based on data from previous multi-cycle Lu177-DOTATATE clinical study demonstrate the importance of consider changing intra-cycle dose heterogeneity, tumour volume and kidney biokinetics in personalising patient administered activity. his study shows understanding the correlation between changes in biology and absorbed dose could help improve the planning of subsequent cycles of radionuclide therapy treatments.