The Coronavirus disease-19 (COVID-19) pandemic dramatically impacted on oncologic patients' care. Since the introduction of vaccines and the demonstration of their benefit on frail patients, COVID-19 vaccinations were indicated also for oncologic population. However, data about the impact of anticancer-treatments and the timing between vaccinations and systemic therapy de-livery were not available. We aimed to evaluate potential factors influencing the outcome of COVID-19 vaccination in cancer patients. We prospectively collected data of patients undergoing COVID-19 vaccination with gas-tro-entero-pancreatic and neuroendocrine neoplasms, treated at our Institute, between 03/2021 and 12/2021. We enrolled 46 patients, 63.1% males; at the time of data collection 86.9% received two-doses of Pfizer-BioNTech and the remains Moderna vaccine. All patients obtained a subsequent im-mune-response. A significantly lower values of IgG for patients treated with chemotherapy versus other anti-cancer agents (p=0.004). No significant effect on immune-response was reported for both vaccinations performed ≤7 vs >7 days from the last systemic treatment (p=0.77) and lymphocytes count (p=0.11). The findings suggest that the optimal timing for COVID-19 vaccination and lymphocytes count are not the issue but rather the quality of the subset of lymphocytes before the vaccination to deter-mine the efficacy level of immune-response in this population.

Gastroenteropancreatic neuroendocrine neoplasms (GEP NENs) are rare cancers consisting of neuroendocrine carcinomas (NECs) and neuroendocrine tumors (NETs), which have been increasing in incidence in recent years. Few cell lines and pre-clinical models exist for the study of GEP NECs and NETs, limiting the ability to discover novel imaging and treatment modalities. To address this gap, we isolated tumor cells from cryopreserved patient GEP NECs and NETs and injected them into the flanks of immunocompromised mice to establish patient-derived xenograft (PDX) models. Two of 6 mice developed tumors (NEC913 and NEC1452). Over 90% of NEC913 and NEC1452 tumor cells stained positive for Ki67. NEC913 PDX tumors expressed neuroendocrine markers such as chromogranin A (CgA), synaptophysin (SYP), and somatostatin receptor-2 (SSTR2) whereas NEC1452 PDX tumors do not express SSTR2. Exome sequencing revealed loss of p53 and RB1 in both tumors. To demonstrate an application of these novel NEC PDX models for SSTR2-targeted peptide imaging, the NEC913 and NEC1452 cells were bilaterally injected into mice. Near infrared-labelled octreotide was administered and fluorescent signal was specifically observed for the NEC913 SSTR2 positive tumors. These 2 GEP NEC PDX models serve as valuable resource for GEP NEN therapy testing.

Deep learning (DL) algorithms used for DOTATATE PET lesion detection typically require large, well-annotated training datasets. These are difficult to obtain due to low incidence of gastroenteopancreatic neuroendocrine tumors (GEP-NETs), and the high cost of manual annotation. Furthermore, networks trained and tested with data acquired from site specific PET/CT instrumentation, acquisition and processing protocols have reduced performance when tested with offsite data. This lack of generalizability requires even larger, more diverse training datasets. The objective of this study is to investigate the feasibility of improving DL algorithm performance by better matching the background noise in training datasets to higher noise, out-of-domain testing datasets. 68Ga-DOTATATE PET/CT datasets were obtained from two scanners: Scanner1, a state-of-the-art digital PET/CT (GE DMI PET/CT; n=83 subjects), and Scanner2, an older-generation analog PET/CT (GE STE; n=123 subjects). Set1, the data set from Scanner1 was reconstructed with standard clinical parameters (5 minutes; Q.Clear) and list-mode reconstructions (VPFXS 2, 3, 4, and 5-minutes). Set2, data from Scanner2, representing out-of-domain clinical scans, used standard iterative reconstruction (5 minutes; OSEM). Reconstructed datasets were divided into training, validation, and testing datasets in a 60%, 20%, 20% proportion, and a deep neural network was trained with each: Network1 for Scanner1 and Network2 for Scanner2. DL performance (Network1) was tested with out-of-domain test data (Set2). To evaluate the effect of training sample size, we tested DL model performance using a fraction (25%, 50%, and 75%) of Set1 for training. Scanner1, list-mode 2-minute reconstructed data demonstrated the most similar noise level compared that of Set2, resulting in the best performance (F1=0.713). This was not significantly different compared to the highest performance, upper-bound limit using in-domain training for Network2 (F1=0.755; p-value=0.103). Regarding sample size, the F1 score significantly increased from 25% training data (F1=0.478) to 100% training data (F1=0.713; p<0.001). List-mode data from modern PET scanners can be reconstructed to better match the noise properties of older scanners. Using existing data and their associated annotations dramatically reduces the cost and effort in generating these datasets, and significantly improves the performance of existing DL algorithms. List-mode reconstructions can provide an efficient, low-cost method to improve DL algorithm generalizability.