RESUMEN
PURPOSE: In Peptide Receptor Radionuclide Therapy (PRRT) with [177Lu]Lu-DOTATATE of gastro-entero-pancreatic neuroendocrine tumours (GEP NETs) a question remains open about the potential benefits of personalised dosimetry. This observational prospective study examines the association of individualized dosimetry with progression free survival (PFS) in G1-G2 GEP NETs patients following the standard [177Lu]Lu-DOTATATE therapeutic regimen. METHODS: The analysis was conducted on 42 patients administered 4 times, and on 165 lesions. Dosimetry was performed after the first and the forth cycle, with two SPECT/CT scans at day 1 and 7 after administration. Global mean Tumour absorbed Dose of each patient (GTD) was calculated after cycle 1 and 4 as the sum of lesion doses weighted by lesion mass, normalized by the global tumour mass. Cumulative GTD_TOT was calculated as the mean between cycle 1 (GTD_1) and 4 (GTD_4) multiplied by 4. Patients were followed-up for median 32.8 (range 18-45.5) months, through blood tests and contrast enhanced CT (ceCT). This study assessed the correlation between global tumour dose (GTD) and PFS longer or shorter than 24 months. After a ROC analysis, we stratified patients according to the best cut-off value for two additional statistical analyses. At last a multivariate analysis was carried out for PFS > / < 24 months. RESULTS: The median follow-up interval was 33 months, ranging from 18 to 45.5 months. The median PFS was 42 months. The progression free survival rate at 20 months was 90.5%. GTD_1 and GTD_TOT were statistically associated with PFS > / < 24 m (p = 0.026 and p = 0.03 respectively). The stratification of patients on GTD_1 lower or higher than the best cut-off value at 10.6 Gy provided significantly different median PFS of 21 months versus non reached, i.e. longer than 45.5 months (p = 0.004), with a hazard ratio of 8.6, (95% C.I.: [2 - 37]). Using GTD_TOT with the best cut-off at 43 Gy, the same PFS values were obtained as after cycle 1 (p = 0.035). At multivariate analysis, a decrease in GTD_1 and, with lower impact, a higher global tumour volume were significantly associated with PFS < 24 months. We calculated the Tumour Control Probability of obtaining PFS > 24 months as a function of GTD_1. DISCUSSION: Several statistical analyses seem to confirm that simple tumour dosimetry with 2 SPECT/CT scans after the first administration allows to predict PFS values after 4 × 7.4 GBq administrations of 177Lu[Lu]-DOTATATE in G1-G2 GEP NETs. This result qualitatively confirms recent findings by a Belgian and a French study. However, dosimetric thresholds are different. This probably comes from different cohort baseline characteristics, since the median PFS in our study (42 m) was longer than in the other studies (28 m and 31 m). CONCLUSION: Tumour dosimetry after the first administration of [177Lu]Lu-DOTATATE offers an important prognostic value in the clinical decision-making process, especially for the future as alternative emitters or administration schedule may become available.
RESUMEN
Deep learning (DL) strategies applied to magnetic resonance (MR) images in positron emission tomography (PET)/MR can provide synthetic attenuation correction (AC) maps, and consequently PET images, more accurate than segmentation or atlas-registration strategies. As first objective, we aim to investigate the best MR image to be used and the best point of the AC pipeline to insert the synthetic map in. Sixteen patients underwent a 18F-fluorodeoxyglucose (FDG) PET/computed tomography (CT) and a PET/MR brain study in the same day. PET/CT images were reconstructed with attenuation maps obtained: (1) from CT (reference), (2) from MR with an atlas-based and a segmentation-based method and (3) with a 2D UNet trained on MR image/attenuation map pairs. As for MR, T1-weighted and Zero Time Echo (ZTE) images were considered; as for attenuation maps, CTs and 511 keV low-resolution attenuation maps were assessed. As second objective, we assessed the ability of DL strategies to provide proper AC maps in presence of cranial anatomy alterations due to surgery. Three 11C-methionine (METH) PET/MR studies were considered. PET images were reconstructed with attenuation maps obtained: (1) from diagnostic coregistered CT (reference), (2) from MR with an atlas-based and a segmentation-based method and (3) with 2D UNets trained on the sixteen FDG anatomically normal patients. Only UNets taking ZTE images in input were considered. FDG and METH PET images were quantitatively evaluated. As for anatomically normal FDG patients, UNet AC models generally provide an uptake estimate with lower bias than atlas-based or segmentation-based methods. The intersubject average bias on images corrected with UNet AC maps is always smaller than 1.5%, except for AC maps generated on too coarse grids. The intersubject bias variability is the lowest (always lower than 2%) for UNet AC maps coming from ZTE images, larger for other methods. UNet models working on MR ZTE images and generating synthetic CT or 511 keV low-resolution attenuation maps therefore provide the best results in terms of both accuracy and variability. As for METH anatomically altered patients, DL properly reconstructs anatomical alterations. Quantitative results on PET images confirm those found on anatomically normal FDG patients.