Correia, Cristina

Accurate estimation of Alzheimer’s disease (AD) severity from structural magnetic resonance imaging (MRI) remains difficult, as disease-associated anatomical alterations are often subtle and publicly available datasets are typically too small to support robust deep learning model training. This feasibility study sought to determine how much Alzheimer’s disease spectrum-related information could be extracted from a small structural MRI cohort using a deliberately lightweight two-dimensional convolutional neural network (2D CNN), and whether transfer learning improves model performance. This study was intended as a methodological proof of concept rather than the development of a clinically deployable diagnostic tool. Structural scans and Clinical Dementia Rating (CDR) labels from the OASIS-1 dataset were filtered to 214 subjects: 124 cognitively normal (CN), 65 with mild cognitive impairment (MCI; CDR = 0.5), and 25 with AD-level impairment (CDR ≥ 1). A compact 2D CNN trained from scratch and a transfer learning model (frozen ImageNet MobileNetV2 features) were evaluated on four binary tasks (CN vs. AD, MCI vs. AD, CN vs. MCI, and CN vs. any impairment) under identical pre-processing and subject-level repeated 5-fold cross-validation (10 repeats), with the decision threshold tuned only on an inner split. Discrimination was summarized by ROC-AUC with 95% confidence intervals (CIs), permutation tests against chance, and per-task sensitivity and specificity. The from-scratch CNN recovered only a broad normal-versus-impaired signal (CN vs. any impairment AUC 0.59) and was at chance on adjacent-stage tasks (MCI vs. AD 0.41; CN vs. MCI 0.51). Transfer learning improved every task: CN vs. AD AUC 0.745 (95% CI 0.730–0.763), CN vs. any impairment 0.642, CN vs. MCI 0.601, and MCI vs. AD 0.599. On an independent OASIS-2 cohort, the transfer learning CN vs. AD model retained AUC 0.748. In this small-data regime, transfer learning recovers substantially more Alzheimer-spectrum signals than a from-scratch CNN, but performance remains modest because it is bounded by CDR-based, non-biomarker-confirmed labels, suggesting the model separates CDR-defined cognitive-status groups rather than detecting AD pathology.

TRAIL is a TNF family ligand that trimerizes TRAIL-R1 (DR4) or TRAIL-R2 (DR5) to induce apoptosis, necroptosis, and/or NF-κB activation in receptor-bearing cells. We previously identified TRAILshort as a splice variant of TRAIL that lacks cysteine 230, cannot trimerize, and acts as a dominant-negative ligand that blocks TRAIL-mediated apoptosis. TRAILshort is expressed on cell surfaces and within extracellular vesicles, enabling it to confer TRAIL resistance to both producing and bystander cells. In this study, we showed that elevated TRAILshort levels were associated with chronic viral infections, cancer, and autoimmune diseases, suggesting a link to impaired immune regulation. Using unbiased phosphoproteomics and mechanistic studies, we demonstrated that TRAILshort binding to DR5 recruited and activated the phosphatase Src homology region 2 domain–containing phosphatase 1 (SHP-1), leading to zeta-chain-associated protein kinase 70 (ZAP-70) dephosphorylation, disruption of ZAP-70–CD3ζ interactions, and impaired T cell receptor signaling, thereby reducing T cell activation, proliferation, and cytokine production in response to antigen or CD3/CD28 ligation. Genetic or pharmacologic SHP-1 inhibition reverses these effects. In humanized mouse models, TRAILshort promoted the persistence of transformed mouse embryonic fibroblasts (MEFs) and L428 and antagonized CD19-directed CAR T cell activity, revealing TRAILshort as an immunomodulator of T cell function with therapeutic implications, including blocking TRAILshort to restore T cell immunity or delivering TRAILshort to enforce tolerance. TRAILshort, a variant of TRAIL, suppresses T cell function and immune responses in cancer and infectious diseases, highlighting its potential as a therapeutic target.