Purpose/Objective(s) Peer review is reflective of the entire patient treatment process with clinical information, therapeutic parameters and potential treatment variations. Our hypothesis is that using artificial intelligence (AI), we can enhance the efficacy of the peer review process by screening cases with a potential for treatment interruption. Materials/Methods From 3,899 radiotherapy patients (7,168 plans) treated from 2014-2021 in our department, 36 features of clinical and therapeutic parameters were used as input for two AI models: multivariable least absolute shrinkage and selection operator (LASSO) logistic regression model and pattern recognition feed forward neural network (NN). LASSO is a shrinkage regularization method that assigns coefficients to significantly useful features. NN passes input features through multiple hidden nodes to assign weights and biases for feature selection. Each method results in continuous probability of treatment interruption. Performance metrics of accuracy, sensitivity and specificity were calculated for evaluations. Results Overall, 8.1% of all cases had treatment interruptions, most commonly in head and neck (18.8%) compared to other sites (19% vs 7-9%, p<0.01). For LASSO model, testing set sensitivity, specificity and accuracy ranged from 65-89%, 31-53% and 35-54%, respectively, with higher sensitivity than specificity. Spine/Extremity and Brain sites had the highest accuracy (54%). Higher sensitivity than specificity indicated that the model was more able to predict true positives. For NN model, testing set sensitivity, specificity and accuracy ranged from 23-62%, 66-90% and 64-86%, respectively. Higher specificity than sensitivity was observed. The Brain site had highest accuracy (86%). Accuracy of NN was higher than LASSO for all treatment sites (68% vs. 50%), which is particularly true for the brain (86% vs. 54%) and the pelvis/prostate (77% vs. 35%). This result indicates that the linear hyperplane of LASSO was insufficient to classify the underlying complex clinical dataset accurately. Including more input features is expected to improve the accuracy for both models. Conclusion Our results provide proof-of-concept that AI can be used as a screening tool to aid the peer review process. It can help to predict treatment interruptions, which include replanning or treatment cessation. Early identification of patients at risk of radiotherapy interruptions using this method may potentially translate into higher treatment completion rates. The study is being continued at our institution to further explore the capabilities of AI in the peer review process.

A total of 847 inbred Lewis rats of mixed sex were used in this pancreaticoduodenal (Pd) donor aging study. Pd grafts were taken from 9- to 12-month-old donors and transplanted into 3-month-old recipients (thus, the first generation Pd graft, or 1 Pd). After 9 to 12 months, the same Pd grafts were again harvested and transplanted into 3-month-old rats (thus the 2 Pd generation). This cycle was repeated to obtain the 3, 4, and 5 Pd series. Sequential transplantation was able to extend the Pd grafts' mean survival time to 32 months for fourteen 4 Pd grafts, and to 39.2 months for four 5 Pd grafts (the longest lived graft survived for 42 months). The pancreas and duodenal sections of the grafts remained normal throughout the entire study. However, the aortic sections of the grafts (which were harvested to include the superior mesenteric and celiac arteries) all exhibited moderate to massive atherosclerotic changes by the 5 Pd mean survival age of 39.2 months. Such histological changes commenced even before 21 months of Pd graft age in some animals, gradually progressing to dilation of the aorta (and subsequent narrowing of aortic tributaries), as well as formation of an eggshell-like inner membrane shielding the aortic intima, by 42 months. Such atherosclerotic changes precluded transplantations beyond the 5 Pd series.

Murine chronic graft-vs-host disease (CGVHD) to minor histocompatibility antigens (B10.D2 → BALB/c) is characterized by inflammatory destruction of intrahepatic bile ducts, scleroderma-like skin lesions, and lymphoid involution. Spleen cells isolated from this model proliferate poorly when stimulated with mitogens. Previous reports indicate defective lymphocyte proliferation in this model is the result of active suppression induced by the graft-vs-host reaction in the spleen and is mediated by Thy 1.2−, sIg−, plastic nonadherent, splenic natural suppressor (NS) cells. To determine whether the intense CGVHD in the liver is associated with induction of suppression, we compared the suppressor activity of hepatic and splenic mononuclear inflammatory cells isolated concurrently during murine CGVHD. Both hepatic and splenic MC suppressed the proliferation of mitogen-stimulated normal spleen cells in a non-MHC, non-MIs restricted manner. T cells contributed to the suppressor activity of both populations. However, the suppressor activity of hepatic MC was mediated largely by a macrophage-enriched population of MC while that of splenic MC was mediated largely by NS cells.