Response to immune checkpoint inhibitors (ICIs) is governed in part by the spatial organization of the tumour microenvironment, yet most deep learning models applied to digital pathology treat tissue regions independently and therefore do not represent the neighbourhood interactions that shape immune activation. This study develops and retrospectively evaluates a Neighborhood-aware Masked Graph Spiking Neural Network (NAMG-SNN) that classifies ICI responders and non-responders from pre-treatment haematoxylin and eosin whole-slide images. Whole-slide images are stain-normalized, tiled and encoded into patch-level embeddings that combine statistical, textural, morphological and deep semantic descriptors. A spatial graph is then constructed over the retained patches and pruned by an adaptive similarity mask that suppresses weak or artefactual neighbourhood edges. The masked graph is propagated through a leaky integrate-and-fire spiking network whose accumulated firing rates form a spatial transcriptomic representation, which is finally classified into responder and non-responder categories. The model was evaluated on an integrated cohort assembled from The Cancer Genome Atlas (TCGA), the Panoptic segmentation of nUclei and tissue in Melanoma pathology Challenge (PUMA challenge) and the Phoenix virtual spatial transcriptomics resource, using patient-level splitting. On the integrated test data the framework reached 97.31% accuracy, 96.94% precision, 97.82% recall, 97.38% F1-score and 98.87% area under the receiver operating characteristic curve (AUC), with a Matthews correlation coefficient of 0.947 and Cohen’s kappa of 0.946, exceeding convolutional, transformer and graph baselines trained under identical conditions. Transcriptomic reconstruction reached a Pearson correlation of 0.962, mean absolute error of 0.051, root mean square error of 0.081 and structural similarity index (SSIM) of 0.966. Ablation confirmed that neighbourhood-aware graph construction, adaptive masking and spike-based learning each contributed to performance. These results are retrospective and computational.
The advent of immune checkpoint inhibitors (ICIs) has revolutionized cancer treatment, offering an effective therapeutic option for advanced cancers. However, their widespread use has led to the occurrence of immune-related adverse events (irAEs), including pulmonary toxicity. ICI-induced pulmonary toxicity represents one of the most relevant irAEs because of its heterogeneous clinical and radiological presentation and its potential life-threatening course. Computed tomography (CT) is the imaging modality of choice for evaluating lung irAEs and the specific patterns associated with ICI-induced pneumonitis. A literature search of the PubMed and Scopus databases was conducted to identify studies published over the past decade investigating pulmonary adverse effects associated with ICI therapy. Relevant studies were reviewed with particular attention to the clinical and radiological features of ICI-induced pulmonary toxicity, including the main CT patterns of pneumonitis and the emerging applications of radiomics and machine learning. This narrative review aims to provide a comprehensive overview of the current evidence on the radiological manifestations of lung irAEs, with particular emphasis on CT findings and radiomics and its potential role in the prediction and diagnosis of lung irAEs. Ultimately, this research highlights the importance of a multidisciplinary approach in the diagnosis and management of ICI-induced pulmonary toxicity, with radiologists playing a critical role in identifying and monitoring these potentially life-threatening events.
Pancreatic cancer is one of the most common malignant neoplasms, the incidence has increased in the population among patients who have recovered from various SARS-CoV-2 strains with subsequent post-COVID syndrome symptoms development. Given the heterogeneity of cancer development mechanisms within the population, the individual complex treatment implementation is essential for the correction of impaired immunometabolic processes. The aim is to assess pathological changes in humoral and cellular markers of innate and adaptive immunity in groups of patients with non-metastatic and metastatic pancreatic cancer. 64 patients (38 to 79 years) with pancreatic cancer were examined: group 1—non-metastatic (n = 44); group 2—metastatic (n = 20). The following research methods were used: ELISA (enzyme-linked immunosorbent assay), flow cytometry, immunoturbidimetry, spectrophotometry, light microscopy, impedance spectrometry. In first group patients had an increase in the C3 component complement, IL-1β, TNF-α, TLR4, TLR9 and DAMPs (Damage-Associated Molecular Patterns) cytotoxic oligonucleotide fraction. In second group patients had an increase in cortisol, C3 component complement, TLR9, oligopeptide and oligonucleotide DAMPs fractions and a maximum increase in the T regulatory cells CD3+CD4+CD25+СD127-. In both groups was increase in NK cells CD3-CD56+CD16+ and a wide spectrum of antinuclear antibodies to altered nuclear structures that acquired antigenic properties. In both groups, we identified cohorts of patients who with significant changes in the studied parameters set, had a decrease in serum electrical conductivity (on average by 45%). Individual changes in immunometabolic markers in patients of different ages with non-metastatic and metastatic pancreatic cancer have diagnostic and prognostic value.
Pediatric tuberculosis remains difficult to monitor using objective biomarkers. Total serum adenosine deaminase (ADA) reflects cellular immune activation, whereas vitamin D and zinc may modify antimycobacterial immunity. This pilot randomized trial evaluated whether adjunctive vitamin D or zinc altered serum ADA during the intensive phase of pediatric tuberculosis treatment. Thirty-three treatment-naive children with tuberculosis were randomized equally to standard anti-tuberculosis treatment (ATT) alone, ATT plus zinc, or ATT plus vitamin D. Total serum ADA was measured at baseline and after two months. The prespecified completer analysis compared change scores, while the primary adjusted analysis used analysis of covariance (ANCOVA) with post-treatment ADA as the outcome and baseline ADA as a covariate. A multiple-imputation intention-to-treat sensitivity analysis included all randomized participants. Twenty-two participants completed follow-up (ATT, n = 8; zinc, n = 8; vitamin D, n = 6). Unadjusted ADA decreased in all groups and most markedly in the vitamin D arm, but the between-group difference in change was not significant (p = 0.364). Baseline-adjusted ANCOVA also showed no treatment-group difference (F2,18 = 0.79; p = 0.470). Compared with ATT alone, the adjusted effects were −4.35 U/L for vitamin D (95% CI −15.70 to 7.00) and +2.26 U/L for zinc (95% CI −8.07 to 12.59). The intention-to-treat sensitivity analysis produced the same overall interpretation. In this underpowered pilot trial, adjunctive vitamin D or zinc did not significantly reduce total serum ADA beyond standard therapy. The large crude decline in the vitamin D arm was attenuated after adjustment for baseline imbalance and regression to the mean. Larger multicentre trials should confirm micronutrient repletion, measure ADA2, and prespecify methods for missing data.
The tumor microenvironment (TME) plays a major role in cancer progression, immune evasion, and resistance to immunotherapy. Coumarins, a structurally diverse class of natural and synthetic compounds, have attracted increasing attention because of their anticancer, anti-inflammatory, antioxidant, and immunomodulatory properties, suggesting potential applications in TME-oriented cancer immunotherapy. This review aimed to summarize the immunotherapeutic potential of natural and synthetic coumarin derivatives in modulating TME and enhancing anticancer immune responses. A narrative review of the literature was conducted focusing on the effects of coumarin derivatives on immune-cell regulation, immune checkpoint signaling, angiogenesis, extracellular matrix remodeling, metabolic reprogramming, and hypoxia-related pathways within TME. Evidence from preclinical studies involving natural and synthetic coumarins was critically evaluated. Preclinical findings indicate that coumarin derivatives can modulate several components of TME by promoting M1 macrophage polarization, activating dendritic cells and natural killer cells, enhancing T-cell–mediated immunity, and suppressing immunosuppressive cell populations such as regulatory T cells and M2 macrophages. Several coumarins also demonstrated the ability to inhibit tumor glycolysis, attenuate hypoxia-associated immune resistance, normalize angiogenesis, and regulate inflammatory signaling pathways. Synthetic coumarins showed additional promise through structural optimization and targeted delivery approaches, while their potential complementary effects with immune checkpoint inhibitors remain to be established experimentally. However, available evidence remains limited, and further mechanistic and translational investigations are required. Natural and synthetic coumarins represent promising multifunctional immunomodulators capable of reshaping TME and improving anticancer immune responses. Their diverse biological activities support their future development as adjuncts or enhancers of cancer immunotherapy.
