Citations

Abstract Large language models (LLMs) and vision‐language models represent a fundamentally different category of artificial intelligence (AI) compared to prior image analysis approaches in digital pathology, which have largely been based on convolutional neural network architectures. This review from the American Society of Cytopathology Clinical Practice Committee examines the current evidence for LLM and vision‐language model applications in cytopathology, including structured reporting, diagnostic assistance, quality control, education, and workflow integration. The distinction between applications with preliminary evidence and those that remain hypothetical is described. A detailed assessment of the challenges that must be addressed before clinical deployment, including hallucination risk, limited explainability, bias, data privacy, validation gaps, and infrastructure barriers is discussed. A review of the regulatory landscape in the United States and European Union as it applies to AI‐enabled software as a medical device is provided. Recommendations addressing cytopathology‐specific benchmarks, multi‐institutional validation, transparent governance, and incremental deployment beginning with low‐risk applications are suggested. In the current environment, LLMs have the potential to augment cytopathology practice, but responsible adoption requires rigorous validation and sustained collaboration among cytopathologists, AI researchers, and regulatory bodies.

ABSTRACT Many real‐world patients with advanced biliary tract cancer (BTC) are excluded by TOPAZ‐1 criteria; this study assessed cisplatin and gemcitabine plus durvalumab (CGD) efficacy and safety in these patients. Data from 1358 patients with advanced BTC treated with first‐line CGD across 55 international centers were retrospectively analyzed. Patients were classified as TOPAZ‐1‐in (meeting all inclusion criteria) or TOPAZ‐1‐out (meeting ≥ 1 exclusion criterion). Primary endpoints were overall survival (OS) and progression‐free survival (PFS); secondary endpoints included objective response rate (ORR), disease control rate (DCR), and safety. Furthermore, OS and PFS of the TOPAZ‐1‐out cohort were compared with reconstructed survival data from the phase III trial. 912 (67.1%) patients were classified as TOPAZ‐1‐in , and 446 (32.9%) as TOPAZ‐1‐out . After a median follow‐up of 14.5 months (95% CI: 13.7–36.5), OS was 16.1 months in TOPAZ‐1‐in and 12.5 months in TOPAZ‐1‐out (HR 0.69, 95% CI: 14.6–16.5, p = 0.0001); PFS was 8.2 versus 6.5 months (HR 0.73, 95% CI: 7.1–8.1, p < 0.0001). Median OS in the TOPAZ‐1‐out cohort (12.5 months) was nearly identical to the phase III trial experimental arm (12.9 months; HR 1.15, p = 0.13); for PFS the HR was 1.12 (95% CI 0.93–1.42; p = 0.09). Among TOPAZ‐1‐out patients, active infection, elevated bilirubin, and ECOG PS > 1 were linked to poorer OS, while no detrimental impact emerged for ALT/AST abnormalities, corticosteroid use, renal or hematologic parameters, or prior surgery within 6 months, suggesting treatment effectiveness was broadly maintained across clinical subgroups. Furthermore, TOPAZ‐1‐out patients experienced no significant increase in adverse events compared with the TOPAZ‐1‐in group. Real‐world data suggest CGD may be effective in TOPAZ‐1‐out patients, with no evident increase in toxicity, supporting the potential expanded use of the regimen in clinical practice.

Abstract Objectives To systematically evaluate the certainty of evidence for treatment strategies across age groups and mood phases in bipolar disorder, and develop an open access web platform to facilitate shared decision making. Design Living umbrella review, evaluation, analysis, and communication hub (U-REACH) project. Data sources PubMed, PsycInfo, and Cochrane library databases, from inception to 19 November 2024. Eligibility criteria for selecting studies Systematic reviews with network or pairwise meta-analyses of randomised controlled trials of pharmacological, nutraceutical, psychosocial, brain stimulation, or circadian rhythm based treatments, administered as monotherapy (without concurrent interventions), augmentation treatment (interventions added to an ongoing treatment regimen), or combination treatment (simultaneous initiation of two different interventions), examining any age group, bipolar disorder phase (ie, acute bipolar depression, mania or mixed episodes, or maintenance), treatment, control, or outcome. Results 77 studies met the inclusion criteria (21 network meta-analyses and 56 pairwise meta-analyses), including 116 unique pharmacological (n=74), brain stimulation (n=18), nutraceutical (n=13), psychosocial (n=8), and circadian rhythm based (n=3) treatments as monotherapy, augmentation, or combination therapy, along with five control interventions. These studies covered 133 unique outcomes (45 efficacy outcomes and 88 safety outcomes) resulting in 2510 meta-analyses with Grading of Recommendations, Assessment, Development, and Evaluations (GRADE) ratings of the certainty of the evidence as high (n=236), moderate (n=827), low (n=986), and very low (n=461). A communication hub, the Evidence Based Interventions for Bipolar Disorder (EBI-BD) platform, was developed and the full results are freely available ( https://ebibd-database.org ), including the preference based tool (12 interventions and 17 safety outcomes). Interventions effective across outcomes varied by phases. For bipolar depression, effective interventions in adults were cariprazine, divalproex or valproate, fluoxetine, ketamine (augmentation), lamotrigine, lumateperone, lurasidone, olanzapine, olanzapine with fluoxetine, and quetiapine, whereas effective interventions in children and adolescents were lurasidone and olanzapine with fluoxetine. For mania episodes, effective interventions in adults were aripiprazole, asenapine, carbamazepine, cariprazine, divalproex or valproate, haloperidol, lithium, olanzapine, paliperidone, quetiapine (also augmentation), risperidone (also as augmentation), tamoxifen, and ziprasidone, whereas effective interventions in children and adolescents were aripiprazole, asenapine, olanzapine, quetiapine, and risperidone. For maintenance, interventions effective across outcomes in adults were aripiprazole (also the long acting injectable formulation), asenapine, divalproex or valproate, lithium, olanzapine, group psychoeducation (augmentation), quetiapine, and risperidone long acting injectable formulation. Interventions effective across phases were aripiprazole (also as augmentation and as a long acting injectable formulation), asenapine, cariprazine, cognitive behavioural therapy (augmentation), divalproex or valproate, lamotrigine, lithium, olanzapine (also as augmentation), paliperidone, quetiapine (also as augmentation), and risperidone (also as augmentation and the long acting injectable formulation) (adults). Treatment effects by neuroscience based nomenclature classes are also reported. Conclusions The EBI-BD tool can help clinicians make evidence based, personalised treatment decisions for bipolar disorder. This resource can inform clinical guidelines and provides a foundation for continuously improving bipolar disorder care as new evidence emerges. Trial registration Open Science Framework https://osf.io/pjmvn/ Readers’ note This is a living systematic review and may be updated in the next two years if additional evidence emerges.

Abstract Although the human genome encodes a vast repertoire of noncoding RNAs that regulate gene expression, the noncoding genome remains underexplored due to technical challenges. Specifically, during transcriptomic sequencing data alignment, the overlap between noncoding and coding loci can create ambiguous read alignments that are subsequently discarded from downstream analysis. For this reason, most of the noncoding genome is excluded from standard genomic annotations used for sequencing alignment. To address this challenge and enable concurrent profiling of the coding and noncoding transcriptome, we systematically integrated standard coding (GENCODE) and noncoding (LncBook) genome annotations, preserving coding gene annotations and removing overlapping noncoding regions. The resulting integrated genome annotation expanded the number of annotated noncoding genes from 40,785 to 138,296 while preserving all coding genes and reducing ambiguous read assignment. To evaluate the utility of our integrated genome annotation for uncovering novel, biologically relevant noncoding RNAs (ncRNAs), we realigned CD138-positive bulk RNA-seq (N = 942) and CD138-negative single-cell RNA-seq (N = 478) data from the MMRF CoMMpass study, generating a comprehensive coding-noncoding atlas of the myeloma bone marrow microenvironment with noncoding genes representing 51% of highly variable genes and displaying significant cell type specificity. Tumor expression profiling based on this integrated profiling identified 15 clusters, including two enriched for amp(1q21) or t(4;14) and associated with shorter progression-free survival (PFS). Differential expression and systematic filtering yielded 19 candidate high-risk ncRNAs, including previously uncharacterized ENSG00000310209 , which was associated with poor PFS (HR = 1.141, P = 0.0025), increased IRF4 activity, Wnt pathway activation, CCL5 signaling, and the accumulation of anergic-like CD8+ T cells. These findings establish integrated coding-noncoding analysis as a strategic approach for discovering functional ncRNAs from transcriptomic sequencing data.

Abstract Artificial intelligence is spreading through public health agencies and healthcare systems faster than workforces are trained to govern it. We describe RHAMI (Responsible Health AI Readiness and Maturity Index), the first human-centered automated composite index for health organizations to score AI readiness and maturity as one measurable system of science, economics and ethics, and the physician certification-to-Master’s pathway built to teach what it measures—both designed jointly by frontline public health professionals, clinicians, ethicists, engineers, leaders, and policy-makers in low- and middle-income countries and high-income countries.

Summary Background Body composition is recognized as a major determinant of health outcomes, but its multidimensional nature makes clinical adoption challenging. We sought to develop and validate a body composition index (BCI) for all-cause mortality risk assessment, integrating variables of six body composition tissues. Methods We analyzed 28509 consecutive patients undergoing myocardial perfusion imaging with routine low-dose chest CT attenuation correction (CTAC) scans acquired during myocardial perfusion imaging (MPI) at 12 centers across four countries. An artificial intelligence-based BCI was developed in a cohort of 15037 patients’ CTACs by integrating the CT-derived metrics of bone, skeletal muscle, and four adipose tissue compartments, coronary artery calcium score, and basic demographic variables (age, sex, BMI). The performance of BCI for mortality prediction was validated in an internal cohort of 6444 patients and an external cohort of 7028 patients by prognosis, calibration, net benefit, and explainability. Model-based simulation of tissue metrics modification was performed to evaluate estimated mortality risk reduction. Findings During a median of 3.5 (IQR [1.9, 5.1]) years, 4697 (16%) patients died. In the external testing cohort, the BCI demonstrated excellent discrimination for mortality (area under receiver operating characteristic curve 0.78 (95% CI [0.76, 0.79]) and Harrell’s concordance index 0.75 [0.73, 0.76]), calibration, and net benefit overall and across pre-specified subgroups stratified by patient characteristics and imaging protocols. Visceral adipose tissue attenuation was the most influential body composition measure, followed by skeletal muscle volume. Simulated improvement in body composition was associated with significant mortality risk reduction. Interpretation An index combining six body composition measures obtained opportunistically from routine chest CT provides robust mortality risk stratification. By converting complex body composition information into a single interpretable score, the BCI can facilitate clinical implementation of opportunistic CT biomarkers and guide individualized preventive strategies. Funding National Institute of Health Research in context Evidence before this study We searched PubMed and Google Scholar on July 7, 2026 for English language studies using the terms (“mortality prediction” OR “death prediction”) AND (“integrated body composition analysis” OR “body composition metrics integration”) AND (“predictive model” OR “model development”), and identified 41 publications describing predictive models incorporating body composition metrics. Recent advances in artificial intelligence (AI) have enabled increasingly granular quantification of body composition from CT, with multiple tissues and tissue-specific metrics showing independent prognostic value for all-cause mortality (ACM) prediction. However, reported associations of body composition metrics with mortality risk were highly inconsistent or even contradictory across studies. This heterogeneity likely reflected the differences in study design or the inherent complexity of body composition analysis, such as the interaction among variables derived from multiple tissues. Thus, it is difficult to interpret body composition results reliably in clinical practice. We hypothesized that integrating multiple body composition metrics could achieve comprehensive and reliable prediction of mortality across patient populations. Added value of this study In this large international, multicenter longitudinal cohort study of 28509 consecutive patients with suspected or known coronary artery disease (CAD) undergoing myocardial perfusion imaging across 12 sites in four countries, we developed and externally validated an artificial intelligence-based body composition index (BCI) for all-cause mortality prediction. The BCI integrated chest CT-derived metrics of six body composition tissue compartments – bone, skeletal muscle (SM), subcutaneous adipose tissue (SAT), intramuscular adipose tissue (IMAT), visceral adipose tissue (VAT), and epicardial adipose tissue (EAT), together with demographics (age, sex, BMI) and CT-derived coronary artery calcium (CAC) score, all derived automatically from routine low-dose attenuation correction CT scans without additional imaging or laboratory tests. In an external validation cohort of 7028 patients, the BCI achieved robust discrimination for mortality prediction, substantially outperforming a baseline model using demographics and calcium score alone, and exceeded the discriminative performance of prior body composition models. Performance was consistent across subgroups defined by sex, age, body mass index, cardiometabolic risk factors, imaging modality, and acquisition protocol, supporting generalizability across clinical settings. The BCI demonstrated excellent calibration and net clinical benefit. Explainability analysis identified visceral adipose tissue attenuation and skeletal muscle volume as the dominant contributors to mortality risk, offering tissue-specific therapeutic targets that complement generic weight-based metrics. Model-based simulations suggested that improvements in these specific compartments could reduce estimated mortality risk more effectively than equivalent changes in BMI alone, pointing toward precision therapeutic targets beyond weight reduction. Our study improved over previous literatures on body composition-based predictive models in one or more of the following ways: (1) we considered six types of body composition tissues while existing studies considered only three or less tissues, thus significantly higher granularity and improved ability to capture refined tissue interactions; (2) our model was rigorously validated in external sites, showing cross-site generalizability while most previous work lacked external validation; (3) our model achieved significant prediction performance improvement over previous body composition predictive models, offering better clinical outcomes prediction and facilitating the suitability for widespread clinical adoption; (4) our study was conducted with a large multi-center cohort across multiple countries, achieving statistical power and demonstrating reliability towards patients characteristics and imaging acquisition protocols while previous studies involved small single-center cohorts; (5) our model required only CT scans and basic demographics which were widely available in clinical routine while the existing models required dedicated laboratory tests or specialized imaging modalities, enhancing the applicability in routine clinical practice. Implications of all the available evidence BCI offers an explainable and clinically deployable tool for improved prediction of mortality risk from existing chest CT scans. Its use could inform updates to existing risk models, guide precision preventive strategies, and support development of therapeutics targeting specific body composition phenotypes.

Abstract Cellular senescence has been implicated in the pathophysiology of radiotherapy-associated bone loss. Based on our previous work, clearance of senescent cells using genetic and pharmacological tools alleviates the anomalies associated with radiation-associated bone deterioration. The pro-inflammatory senescence associated secretome referred to as senescence associated secretory phenotype (SASP), is a hallmark of cellular senescence. The modulation of SASP by senomorphic drugs, potentially can suppress the pro-inflammatory secretome of senescent cells, irrespective of the underlying senescence mechanism. In this study we tested a senomorphic drug, ruxolitinib, a Janus kinase inhibitor (JAKi), during acute and chronic radiotherapy related effects on the bone. Our clinical data indicate an early increase in several pro-inflammatory SASP proteins following radiotherapy of spinal metastasis in prostate cancer patients. Longitudinal assessment of SASP-related genes confirmed this acute elevation in several SASP markers in systemic circulation following irradiation of male mouse femurs. In studies done in male mice, following three preclinical radiotherapy regimens of 30Gy (5 x 6Gy), 60Gy (5 x 12Gy) and a single dose of 24Gy, suppression of SASP by JAKi was efficacious in suppressing radiation-induced bone damage. In comparison and as shown before, the senolytic combination of D+Q was also able to alleviate radiation-associated bone loss. Early and intermittent suppression of SASP using JAK inhibitors in male mice alleviated chronic bone deterioration, diminished telomere dysfunction, lowered senescence and SASP marker expression and reduced bone-marrow adiposity. Overall, our study shows that early targeting of SASP proteins could be a potential therapeutic to prevent radiotherapy-related chronic bone loss and risk of fractures.

Abstract Introduction Postoperative radiotherapy is an adjuvant therapy for brain metastases(BrMs) that improves local recurrence(LR). The radiosensitivity index(RSI) is a 10-gene index of pan-cancer molecular radiosensitivity previously validated for predicting radiotherapy response in clinical datasets. In this external validation study, we evaluate RSI as a predictor of local recurrence(LR) after radiotherapy in resected BrM. Methods Patients from University Health Network Toronto(UHN) and McGill University who had a resected BrM and subsequently underwent whole brain radiation(WBRT) or stereotactic radiosurgery(SRS). RSI was calculated using a ranked linear regression model from the weighted expression of 10 radiosensitivity genes, with higher values corresponding to radioresistance. Radiophenotype was determined using a previously validated RSI cutoff of &lt; 0.375 to identify radiosensitive tumors. LR was defined as recurrence in the resection cavity and overall survival(OS) was measured from time of BrM diagnosis. Results 106 BrM patients were included, comprising 29 breast, 37 GI, 33 lung, and 7 melanoma primary sources. Mean age at surgery was 59.2 years, 31.1%(n = 33) were male, 67.9%(n = 72) had WBRT, 23.6%(n = 25) had a subtotal resection, and 30.2%(n = 32) had postoperative chemotherapy. Overall, 34.0%(n = 36) patients had LR. Patients with breast BrMs had median RSI 0.70(IQR:0.55-0.75, 10.3% radiosensitive) with 48.3% LR(n = 14); gastrointestinal BrMs had median RSI 0.43(IQR: 0.35-0.46, 29.7% radiosensitive) and 24.3% LR(n = 9); lung BrMs had median RSI of 0.67(IQR:0.66-0.71, 9% radiosensitive) and 33.3% LR(n = 11); melanoma BrMs had median RSI of 0.72(IQR:0.53-0.92, 0% radiosensitive) and 28.6% LR(n = 2). After adjusting for demographics, primary organ, radiotherapy, resection extent, among others, RSI did not predict LR(p = 0.56) or OS(p = 0.61). Discussion RSI did not predict either LR or OS in patients with surgically resected BrMs status post radiotherapy, further establishing the need for improved predictive scores in this population.

Abstract INTRODUCTION Associations of arterial stiffness, measured through pulse wave velocity (PWV), with incident dementia may be driven by age‐related structural changes to the arterial wall or by elevated blood pressure. METHODS Dementia‐free White or Black participants in the community‐based Atherosclerosis Risk in Communities (ARIC) cohort study with measures of PWV were examined. The associations of carotid‐femoral PWV (cfPWV) and total, structural, and load‐dependent PWV with dementia were estimated using Poisson regression models and Cox proportional hazards models. Dementia cases were defined by physician‐adjudicated review, telephone interviews, hospitalization records, and death certificates. RESULTS Of 4818 participants included in the study population (mean age: 75 years; 41% male, 21% Black), 1067 (22%) developed dementia over a median follow‐up of 9.0 years. Higher cfPWV, total PWV, structural PWV, and load‐dependent PWV were associated with greater dementia risk. DISCUSSION Both structural and pressure‐related arterial stiffness mechanisms may underlie the link between PWV and dementia.

Abstract Background Lupus nephritis remains a major cause of chronic kidney disease and kidney failure in systemic lupus erythematosus. Glucocorticoids are central to treatment but are limited by systemic toxicity. We evaluated whether a previously characterized collagen IV α3-targeted liposomal nanoparticle formulation carrying low-dose prednisolone could attenuate established lupus nephritis in MRL/lpr mice. Methods Female MRL/lpr mice with disease present at treatment initiation and C57BL/6J control mice received saline or collagen IV α3-targeted prednisolone-loaded nanoparticles (Col4-α3-Pred-NPs). Renal outcomes were assessed by longitudinal proteinuria, glomerular filtration rate (GFR), survival, kidney histopathology, renal IgG and C3d deposition, dUTP/TUNEL-associated injury staining, and renal cytokine/chemokine profiling. Body weight, food and water intake, and blood glucose were monitored as measures of general condition and preliminary tolerability. Results Col4-α3-Pred-NPs improved survival in MRL/lpr mice, reduced cumulative proteinuria burden, and attenuated terminal GFR decline compared with saline-treated MRL/lpr controls. Treatment reduced glomerular and tubulointerstitial injury, lowered composite EGTI histopathology scores, decreased terminal kidney enlargement, reduced glomerular IgG deposition and renal dUTP-positive injury signals, and reduced renal signals for IL-28A/B, IL-7, PD-ECGF, IL-11, CCL6/C10, and IL-15. C3d deposition was not significantly altered. Nanoparticle treatment was not associated with sustained treatment-related increases in blood glucose or body-weight loss during the measured study period. Conclusions Collagen IV α3-targeted liposomal delivery of low-dose prednisolone attenuated established lupus nephritis in MRL/lpr mice and improved renal structural, functional, inflammatory, and survival outcomes. These findings support further evaluation of glomerulus-targeted nanotherapy as a potential strategy to improve the precision and therapeutic index of glucocorticoid treatment in lupus nephritis.