Digital pathology (DP) is transitioning from an adjunct technology to an enterprise diagnostic platform in the United States. Despite accelerating clinical adoption, many laboratories face persistent barriers, including high capital and operating costs, workflow disruption, interoperability challenges, and a complex regulatory and reimbursement environment. This narrative review proposes a practical lifecycle framework for implementing and sustaining DP programs, with an emphasis on defining and operationalizing institutional artificial intelligence (AI) readiness for safe and sustainable adoption.

We performed a targeted narrative review informed by searches of PubMed/MEDLINE and Google Scholar for English-language publications from January 1, 2014 through December 31, 2025. Core search concepts included DP, whole slide imaging, image management/viewing systems, laboratory information system integration, validation, reimbursement, U.S. Food and Drug Administration clearance, Clinical Laboratory Improvement Amendments oversight, College of American Pathologists accreditation, interoperability standards, cybersecurity, and AI. We supplemented database searches with reference screening and review of primary guidance and public databases from regulatory and professional organizations in the United States. We prioritized peer-reviewed literature and used web-based regulatory sources when they represented the authoritative primary reference. We also incorporated our professional experience and knowledge in DP and AI.

Key implementation domains span foundational infrastructure (scanners, storage/networking, and integrated image management platforms), workflow redesign across pre-analytic, analytic, and post-analytic phases, validation and quality management, regulatory compliance and accreditation, cost capture, interoperability strategy, cybersecurity and access control, education and change management, and long-term governance. We also describe an institution-level AI readiness model that can be assessed across data quality, integration, validation, monitoring, governance, and workforce capabilities to support safe clinical AI deployment.

Successful DP implementation requires a lifecycle approach that couples technical build-out with workflow redesign and institutional governance. Early planning for compliance, interoperability, reimbursement strategy, and AI readiness can reduce implementation risk and position laboratories for sustained clinical and computational innovation.

Metabolic dysfunction–associated steatotic liver disease (MASLD), defined by the American Association for the Study of Liver Diseases as hepatic steatosis with at least one cardiometabolic risk factor, affects approximately 30–40% of adults worldwide. This condition may progress to fibrosis, cirrhosis, and hepatocellular carcinoma. The rising prevalence, alongside obesity and type 2 diabetes, underscores the need for early risk stratification and integrated therapeutic strategies. Circadian homeostasis, orchestrated by the suprachiasmatic nucleus and core clock gene feedback loops, synchronizes hepatic metabolic pathways with environmental light–dark cycles. The objective of this review is to evaluate the role of circadian disruption and metabolic dysfunction in the development of hepatic steatosis, as well as to assess current and potential treatment modalities for both disorders. Circadian disruption through shift work, artificial light at night, sleep restriction, and chrono-nutritional misalignment destabilizes hepatic clocks, promoting insulin resistance, dyslipidemia, inflammation, and steatosis. Experimental models demonstrate that clock gene dysfunction alone can induce steatohepatitis, while progressive MASLD further impairs central circadian regulation, establishing a self-reinforcing chrono-metabolic cycle. Pharmacologic therapies, including glucagon-like peptide-1 receptor agonists and thyroid hormone receptor-β agonists, improve histologic endpoints and fibrosis regression, although heterogeneity among clinical trials precludes direct comparison. Recent evidence characterizing MASLD as a predominantly nocturnal metabolic disorder further highlights persistent nighttime insulin dysregulation despite weight loss, emphasizing the potential role of circadian-targeted interventions such as melatonin. In conclusion, the peripheral circadian clock is intricately linked with MASLD pathogenesis, and metabolic dysfunction, in turn, disrupts circadian pathways. Several pharmacologic therapies offer potential for the treatment of MASLD and circadian dysfunction.

Colorectal polyp detection from endoscopic images is critical for the early diagnosis of colorectal cancer. However, traditional deep learning methods often suffer from limited generalization when deployed across datasets containing different polyp morphologies. This work aimed to investigate whether vision-language foundation models can facilitate zero-shot generalization across multiple polyp datasets without target-domain fine-tuning.

We introduced a zero-shot colorectal polyp detection framework based on Contrastive Language-Image Pretraining (CLIP) to improve cross-dataset detection performance. Key innovations include: (1) a background patch contrastive loss using pseudo-normal tissue patches to teach the model to distinguish normal mucosa from polyps; (2) attribute-enhanced text prompts that incorporate domain-specific descriptors of polyp appearance, improving the model’s semantic generalization to novel polyp morphologies; and (3) an enhanced CLIP visual adapter with per-layer adaptive feature fusion and generalized mean pooling to capture multi-scale features for better polyp localization. During training, we use one annotated colorectal polyp dataset (e.g., CVC-ColonDB) to learn patch-level image-text correspondence. The model is then evaluated in a zero-shot manner on different polyp datasets (CVC-ClinicDB, Kvasir-SEG, and CVC-300), where we measure both pixel-level and image-level anomaly detection performance.

The framework demonstrated robust zero-shot generalization on unseen test cohorts. Without any dataset-specific fine-tuning, the model achieved a mean pixel-level AUROC of 0.94 and a mean average precision of 0.81 across the 12 leave-one-dataset-out zero-shot transfer settings. In the CVC-ColonDB-source benchmark, the model achieved a mean Dice coefficient of 0.84 across CVC-ClinicDB, Kvasir-SEG, and CVC-300. This high level of performance was consistent across datasets with distinct visual characteristics, underscoring the ability of the model to detect diverse polyp morphologies that it had not been explicitly trained to recognize.

Our findings demonstrate that an anomaly-aware vision-language model significantly improves cross-dataset polyp detection generalization without requiring normal images for training. This multimodal strategy may facilitate the robust deployment of artificial intelligence-based colorectal screening systems by enabling reliable detection of diverse polyp morphologies across different clinical settings. Extension to non-polyp colorectal pathologies (e.g., ulcerative colitis and colorectal tumors) remains an important direction for future work, pending the availability of pixel-level annotated datasets for these lesion categories.

Acute leukemias with chimeric fusion genes involving FET (FUS, EWSR1, and TAF15) family proteins and ETS (E26 transformation-specific)-like transcription factors often present with unique clinical and pathological characteristics. This mini-review aims to summarize the clinical and pathological features of acute leukemia cases harboring rearrangements involving the fused in sarcoma (FUS) or Ewing sarcoma breakpoint region 1 (EWSR1) genes.

An extensive literature review was performed on reported acute leukemia cases with fusions involving FUS or EWSR1. The details of the reported cases, as well as summarized information, are presented.

Rare cases of acute leukemia have been found to harbor either FUS or EWSR1 gene rearrangements with ETS or non-ETS proteins as partners and demonstrate heterogeneous clinical and pathological features. Acute leukemias carrying FUS gene rearrangements present with diverse immunophenotypes and are predominantly, but not exclusively, acute myeloid leukemia (AML), with ERG as the most frequent fusion partner. In contrast, acute leukemias with EWSR1 gene rearrangements more commonly present as B-cell acute lymphoblastic leukemia (ALL) and mixed phenotypic acute leukemia (MPAL), with ZNF384 as the predominant partner. At present, FUS::ERG-positive AML is the only specific entity with a FET::ETS fusion that is formally recognized in the World Health Organization 5th edition hematolymphoid tumor classification (WHO-HEM5) and the International Consensus Classification (ICC) systems. Cytogenetic karyotyping and fluorescence in situ hybridization remain crucial tools for detecting chromosomal translocations in over half of acute leukemias harboring FUS or EWSR1 gene rearrangements. However, a subset of patients may exhibit a normal karyotype and require advanced molecular diagnostic methods. EWSR1-rearranged leukemias can be difficult to distinguish from Ewing sarcoma and therefore require particular attention.

As more cases and additional data become available, it may be justified to expand this category of acute leukemias to include other specific acute leukemia entities with fusions involving FET::ETS, such as FUS::FLI1 and FUS::FEV, in addition to FUS::ERG-positive AML. However, additional data are required to support such subclassification. In contrast, AML cases with EWSR1 rearrangements are exceedingly rare and display considerable variability. Cases of B-ALL or B/myeloid MPAL with the EWSR1::ZNF384 fusion may be more appropriately classified together with other ZNF384-rearranged leukemia subtypes. Advanced molecular diagnostic methods, especially RNA-based next-generation sequencing, are suggested to improve the accurate diagnosis of acute leukemias with FUS or EWSR1 fusions. Additional pathologic workup, particularly immunohistochemical staining with hematopoietic markers, is highly recommended to differentiate EWSR1-rearranged leukemia from Ewing sarcoma.

Primary biliary cholangitis (PBC) significantly impairs health-related quality of life (HRQL), yet the impact of disease stage and fatigue on HRQL and psychological status remains insufficiently quantified. This study aimed to investigate differences in HRQL across disease stages and the impact of fatigue in patients with PBC.

This cross-sectional study recruited 219 patients with PBC from two Chinese tertiary hospitals (2011–2024). After excluding one preclinical case, 218 patients were analyzed. Quality of life was assessed using the validated Chinese versions of the SF-36 and Chronic Liver Disease Questionnaire (CLDQ); psychological status was assessed using the Self-Rating Anxiety Scale and Self-Rating Depression Scale. Between-group differences were quantified by mean differences (MDs) and odds ratios (ORs) with 95% confidence intervals (CIs). Baseline characteristics were balanced across stages (all P > 0.05).

Of the 218 patients (90.4% female; mean age, 57.2 ± 10.3 years), 41 were in the clinical stage, 75 in the fibrosis stage, and 102 in the cirrhosis stage. SF-36 scores were lowest in the cirrhosis stage (e.g., Physical Functioning MD, 17.26; 95% CI, 6.93–27.59 vs. clinical stage), with similar declines in CLDQ domains. Anxiety was highest in the clinical stage (58.5%; OR vs. cirrhosis, 4.13; 95% CI, 1.92–8.92), whereas depression was highest in the cirrhosis stage (55.9%; OR vs. clinical stage, 4.50; 95% CI, 1.95–10.38). Fatigue prevalence was 66.1% and increased with disease stage. Patients with fatigue had lower SF-36 scores in Physical Functioning, Bodily Pain, Vitality, Mental Health, and Physical Component Summary (e.g., Physical Component Summary MD, 38.22; 95% CI, 10.41–66.02).

HRQL declines progressively with PBC stage. Fatigue is strongly associated with impaired HRQL and is closely interrelated with anxiety and depression. Stage-specific psychological patterns suggest the need for tailored supportive interventions.

Hepatitis B virus (HBV) infection and hepatitis C virus (HCV) infection are among the leading causes of chronic liver diseases worldwide. Through the same transmission routes, HBV/HCV coinfection is widespread and aggravates liver damage. In this study, we aimed to assess the safety and efficacy of sofosbuvir/velpatasvir (SOF/VEL) and the pre-treatment of tenofovir alafenamide fumarate (TAF) on HBV reactivation in HBV/HCV coinfected patients.

A multicenter, prospective, single-arm, open-label 12-week trial, followed by a 12/48-week observational clinical trial, was conducted. Ninety-six adults with chronic HBV/HCV coinfection were enrolled from May 2021 to December 2024 in thirteen centers in China. Seventy-seven non-cirrhotic patients were included in Group 1 and nineteen compensated cirrhotic patients in Group 2. All subjects were enrolled to receive SOF/VEL once daily for 12 weeks. Non-cirrhotic subjects received TAF once daily for 28 weeks, and compensated cirrhotic subjects received TAF once daily for 64 weeks simultaneously. Statistical significance was set at P < 0.05.

At the end of SOF/VEL treatment, the overall sustained virologic response was 97.9%, of which 100% was achieved in Group 2. HCV RNA, HBV DNA, and HBV RNA levels were substantially decreased in all patients. Alanine aminotransferase (ALT) (61.5 vs. 21.9, P < 0.001) and aspartate aminotransferase (AST) (50.8 vs. 25.7, P < 0.001) levels decreased, and albumin (ALB) (42.4 vs. 45.1, P < 0.001) level increased compared to pre-treatment in Group 1 at 12 weeks post-treatment. ALT (64.1 vs. 25.2, P < 0.001), AST (65.7 vs. 29.7, P < 0.001), alkaline phosphatase (ALP) (111.6 vs. 88.2, P < 0.05), and alpha-fetoprotein (AFP) (17.9 vs. 4.7, P < 0.05) levels decreased, and ALB (41.3 vs. 42.5, P = 0.051) and platelet count (PLT) (114.0 vs. 127.2, P = 0.052) levels showed a trend toward increase compared to pre-treatment in Group 2 at 48 weeks post-treatment. Liver stiffness measurement (LSM) (22.6 vs. 12.7, P < 0.01), aspartate aminotransferase to platelet ratio index (APRI) (1.6 vs. 0.6, P < 0.001), and fibrosis-4 index (FIB-4) (4.7 vs. 2.6, P < 0.05) significantly decreased after treatment in Group 2. Two patients in Group 1 with genotype 3 showed HBV reactivation and HCV relapse, respectively. No drug-related adverse events were observed in the study.

SOF/VEL effectively achieves a sustained virologic response and improves liver function, with an acceptable safety profile in chronic HBV/HCV coinfected patients, including those with compensated cirrhosis, who achieved modest improvement in non-invasive fibrosis indices. Pre-administration of TAF may mitigates the risk of HBV reactivation in this population.

Pancreatic cancer remains a highly lethal malignancy owing to the difficulty of early detection. In 2021, the Chinese Consensus on Early Screening and Surveillance for Pancreatic Cancer in High-risk Individuals was first established. However, the evidence landscape has evolved rapidly, necessitating an updated, evidence-based framework tailored to the Chinese healthcare context. This revised consensus aims to standardize the early screening and surveillance process for high-risk populations in China. A multidisciplinary expert panel comprising 53 specialists from 17 provincial-level regions systematically reviewed the literature using the GRADE (Grading of Recommendations Assessment, Development and Evaluation) methodology. A modified Delphi process was employed, with consensus predefined as ≥75% agreement. The panel formulated 26 evidence-based recommendations covering screening objectives, the definition of high-risk populations (hereditary susceptibility, new-onset diabetes, chronic pancreatitis, and pancreatic cystic neoplasms), age at screening initiation, surveillance intervals, imaging modalities (magnetic resonance imaging/magnetic resonance cholangiopancreatography, endoscopic ultrasound, computed tomography), surgical indications, and lifestyle modifications. Of these recommendations, 14 are strong and 12 are weak, supported by evidence levels ranging from A to D. Implementation of this consensus in clinical practice will help improve the early diagnosis of stage I pancreatic cancer and high-grade precursor lesions, thereby advancing standardized multidisciplinary care and ultimately improving patient outcomes in China.

Liver fibrosis is a pivotal and reversible stage in the progression of chronic liver disease. However, the mechanisms underlying fibrosis reversal remain unclear, and effective diagnostic biomarkers are lacking in clinical practice. In this study, we aimed to elucidate the role and molecular mechanisms of glutathione S-transferase Mu 3 (GSTM3) in liver fibrosis reversal, and preliminarily determine whether GSTM3 can serve as a novel biomarker for liver fibrosis reversal.

Carbon tetrachloride-induced mouse models of liver fibrosis and spontaneous reversal were established. Proteomic analysis was used to identify proteins shared between liver tissue and serum that were continuously downregulated during fibrosis reversal. The expression of GSTM3 was evaluated in the livers of mice undergoing fibrosis reversal and in clinical samples from patients with liver fibrosis. Hepatic stellate cells (HSCs) were transfected with Gstm3-silencing RNA or an overexpression plasmid to assess the effects on fibrosis markers. RNA sequencing analyses were performed, and the underlying molecular mechanisms were investigated.

Proteomic analysis revealed significantly decreased GSTM3 levels in both hepatic tissue and serum in mice undergoing fibrosis reversal, and its expression was negatively correlated with the extent of reversal. GSTM3 levels were markedly increased in the hepatic tissue and serum of patients with liver fibrosis. GSTM3 expression was upregulated in transforming growth factor-β-stimulated HSCs. GSTM3 knockdown inhibited the expression of fibrosis markers, such as collagen type I α1 and tissue inhibitor of metalloproteinase 1, whereas its overexpression promoted their expression. Mechanistic studies indicated that GSTM3 knockdown activated peroxisome proliferator-activated receptor γ (PPARγ) signaling and downregulated its downstream targets, cluster of differentiation 36 and fatty acid-binding protein 4, thereby suppressing HSC activation.

GSTM3 knockdown promotes liver fibrosis reversal via PPARγ signaling-mediated inhibition of HSC activation. Therefore, GSTM3 is a promising therapeutic target and diagnostic biomarker for liver fibrosis.

Irinotecan is a camptothecin derivative that exerts its antitumor effects after conversion into the active metabolite 7-ethyl-10-hydroxycamptothecin (SN-38), and it is widely used for the treatment of various advanced solid tumors. However, irinotecan-induced neutropenia (IIN) remains one of its most common and clinically significant hematological toxicities, increasing the risk of infection, treatment interruption, dose reduction, and poor therapeutic outcomes. This review aims to summarize the pathogenesis, risk factors, and management strategies of IIN to provide practical guidance for its clinical prevention and standardized management. Current evidence suggests that IIN is mainly associated with SN-38-mediated topoisomerase I inhibition and mitochondrial injury in hematopoietic stem cells, leading to impaired neutrophil production. The risk of IIN is influenced by irinotecan dosage, UGT1A and transporter gene polymorphisms, baseline patient characteristics, organ function, and concomitant chemotherapy. Preventive and therapeutic strategies include genotype-guided dose adjustment, careful dose optimization, oral alkalinizing agents, granulocyte colony-stimulating factor support, and modulation of intestinal microbiota. In conclusion, IIN is a multifactorial and potentially manageable adverse reaction. Integrating pharmacogenetic testing, individualized dosing, supportive care, and emerging approaches such as microbiota-based interventions may improve the safety and continuity of irinotecan-based chemotherapy.

Artificial intelligence (AI) is increasingly reshaping diagnostic pathology, with breast pathology representing one of the most advanced and clinically impactful areas of adoption. Despite rapid progress, many practicing pathologists remain unfamiliar with core AI concepts and their practical implications. This review provides a concise and accessible overview of AI in breast pathology, focusing on foundational principles, current clinical applications, and future directions.

Pertinent literature was reviewed. Personal experiences were also summarized and incorporated.

Key AI concepts, including algorithms, models, architectures, machine learning, deep learning, neural networks, and multimodal and foundational models, are introduced to establish a common framework. Important distinctions among generative, black-box, and explainable AI are highlighted, emphasizing the need for transparency and interpretability in clinical settings. The evolution of AI in breast pathology is reviewed, from early rule-based computer-assisted diagnostic systems to modern deep learning approaches leveraging large-scale whole-slide imaging datasets. Current applications span multiple domains, including detection of lymph node metastases, Nottingham grading, classification of benign and malignant lesions, and automated quantification of critical biomarkers. AI-based approaches to prognosis, risk stratification, prediction of treatment response, and analysis of the tumor microenvironment are also discussed. Finally, the review addresses challenges associated with real-world implementation, including data quality, bias, regulatory considerations, cost, infrastructure, and workflow integration.

As AI continues to evolve toward large-scale, multimodal, and explainable models, it is expected to function as an augmentative tool rather than a replacement for pathologists, supporting diagnostic accuracy, standardization, and personalized management in breast cancer care.