Deep-Learning Generation of Synthetic Intermediate Projections Improves 177Lu SPECT Images Reconstructed with Sparsely Acquired Projections.

The aims of this study were to decrease the 177Lu-SPECT acquisition time by reducing the number of projections and to circumvent image degradation by adding deep-learning-generated synthesized projections. Methods: We constructed a deep convolutional U-net-shaped neural network for generation of synthetic intermediate projections (CUSIPs). The number of SPECT investigations was 352 for training, 37 for validation, and 15 for testing. The input was every fourth projection of 120 acquired SPECT projections, that is, 30 projections. The output was 30 synthetic intermediate projections (SIPs) per CUSIP. SPECT images were reconstructed with 120 or 30 projections, or with 120 projections when 90 SIPs were generated from 30 projections (30-120SIPs), using 3 CUSIPs. The reconstructions were performed with 2 ordered-subset expectation maximization (OSEM) algorithms: attenuation-corrected (AC) OSEM, and attenuation, scatter, and collimator response-corrected (ASCC) OSEM. The quality of the SIPs and SPECT images was quantitatively evaluated with root-mean-square error, peak signal-to-noise ratio (PSNR), and structural similarity (SSIM) index metrics. From a Jaszczak SPECT phantom, the recovery and signal-to-noise ratio (SNR) were determined. In addition, an experienced observer qualitatively assessed the SPECT image quality of the test set. Kidney activity concentrations, as determined from the different SPECT images, were compared. Results: The generated SIPs had a mean SSIM value of 0.926 (SD, 0.061). For AC-OSEM, the reconstruction with 30-120SIPs had higher SSIM (0.993 vs. 0.989, P < 0.001) and PSNR (49.5 vs. 47.2, P < 0.001) values than the reconstruction with 30 projections. ASCC-OSEM had higher SSIM and PSNR values than AC-OSEM (P < 0.001). There was a minor loss in recovery for 30-120SIPs, but SNR was clearly improved compared with 30 projections. The observer assessed 27 of 30 images reconstructed with 30 projections as having unacceptable noise levels, whereas the corresponding values were 2 of 60 for 30-120SIPs and 120 projections. Image quality did not differ significantly between 30-120SIPs and 120 projections. The kidney activity concentration was similar between the different projection sets, excepting a minor reduction of 2.5% for ASCC-OSEM 30-120SIPs. Conclusion: Adopting SIPs for sparsely acquired projections considerably recovers image quality and could allow a reduced SPECT acquisition time in clinical dosimetry protocols.

Optimizing the Schedule of PARP Inhibitors in Combination with 177Lu-DOTATATE: A Dosimetry Rationale.

177Lu-DOTATATE for neuroendocrine tumours is considered a low-toxicity treatment and may therefore be combined with other pharmaceuticals to potentiate its efficacy. One approach is to add a poly-[ADP-ribose]-polymerase (PARP) inhibitor to decrease the ability of tumour cells to repair 177Lu-induced DNA damage. To decrease the risk of side effects, the sequencing should be optimized according to the tumour-to-normal tissue enhanced dose ratio (TNED). The aim of this study was to investigate how to enhance 177Lu-DOTATATE by optimal timing of the addition of a PARP inhibitor. Biokinetic modelling was performed based on the absorbed dose to the bone marrow, kidneys and tumour; determined from SPECT/CT and planar images from 17 patients treated with 177Lu-DOTATATE. To investigate the theoretical enhanced biological effect of a PARP inhibitor during 177Lu-DOTATATE treatment, the concept of relative biological effectiveness (RBE) was used, and PARP inhibitor administration was simulated over different time intervals. The absorbed dose rate for the tumour tissue demonstrated an initial increase phase until 12 h after infusion followed by a slow decrease. In contrast, the bone marrow showed a rapid initial dose rate decrease. Twenty-eight days after infusion of 177Lu-DOTATATE, the full absorbed dose to the bone marrow and kidney was reached. Using an RBE value of 2 for both the tumour and normal tissues, the TNED was increased compared to 177Lu-DOTATATE alone. According to the modelling, the PARP inhibitor should be introduced approximately 24 h after the start of 177Lu-DOTATATE treatment and be continued for up to four weeks to optimize the TNED. Based on these results, a phase I trial assessing the combination of olaparib and 177Lu-DOTATATE in somatostatin receptor-positive tumours was launched in 2020 (NCT04375267).

Evaluation of SSTR2 Expression in SI-NETs and Relation to Overall Survival after PRRT.

(1) Purpose: Small intestinal neuroendocrine tumors (SI-NETs) often present with distant metastases at diagnosis. Peptide receptor radionuclide therapy (PRRT) with radiolabeled somatostatin analogues is a systemic treatment that increases overall survival (OS) in SI-NET patients with stage IV disease. However, the treatment response after PRRT, which targets somatostatin receptor 2 (SSTR2), is variable and predictive factors have not been established. This exploratory study aims to evaluate if SSTR2 expression in SI-NETs could be used to predict OS after PRRT treatment. (2) Methods: Using a previously constructed Tissue Micro Array (TMA) we identified tissue samples from 42 patients that had received PRRT treatment during 2006-2017 at Sahlgrenska University hospital. Immunohistochemical expression of SSTR2, Ki-67 and neuroendocrine markers synaptophysin and Chromogranin A (CgA) were assessed. A retrospective estimation of 177Lu-DOTATATE uptake in 33 patients was performed. Data regarding OS and non-surgical treatment after PRRT were collected. Another subgroup of 34 patients with paired samples from 3 tumor sites (primary tumor, lymph node and liver metastases) was identified in the TMA. The SSTR2 expression was assessed in corresponding tissue samples (n = 102). (3) Results: The patients were grouped into Low SSTR2 or High SSTR2 groups based upon on levels of SSTR2 expression. There was no significant difference in 177Lu-DOTATATE uptake between the groups. The patients in the Low SSTR2 group had significantly longer OS after PRRT than the patients in the High SSTR2 group (p = 0.049). PRRT treated patients with low SSTR2 expression received less additional treatment compared with patients with high SSTR2 expression. SSTR2 expression did not vary between tumor sites but correlated within patients. (4) Conclusion: The results from the present study suggest that retrospective evaluation of SSTR2 expression in resected tumors cannot be used to predict OS after PRRT.