Research Appraisals
Evidence-based critical appraisals of the latest medical research, systematically evaluated using Oxford CEBM methodology.
Showing 68 appraisals
Acta oncologica (Stockholm, Sweden)
User perceptions of machine learning models as decision support for colorectal cancer multidisciplinary team conferences (AID-SIM-2): a qualitative simulation study.
BACKGROUND: Multidisciplinary team (MDT) conferences are considered a cornerstone of decision-making in cancer diagnostics and care. However, the current literature has not demonstrated improved patient outcomes based on the decisions of the MDT conferences. AIM: We aimed to evaluate how four different decision-support modalities impacted the decision-making process and the internal discussions in the MDTs in a multicenter simulation study, with a focus on user perceptions. METHODS: Four colorectal cancer centers with MDTs participated. We performed four simulations in each center. Each simulation used a different decision-support tool: (1) Current standard, (2) Current standard plus a prediction model, (3) A structured data presentation tool, and (4) A structured data presentation tool plus the prediction model. Clinician- and model-estimated risks were compared, the treatment suggestions from each site were compared, questionnaires about user perceptions were conducted after Simulations 2, 3, and 4 using a Google Form link, and a semi-structured interview was conducted at each site after the last simulation. RESULTS: Similar distributions of risk groups between clinicians and models were found; however, distinct discrepancies in predictions arose, particularly with higher-risk patients, highlighting the need for standardization for more complex clinical cases. The primary perceived benefit of decision support was increased standardization of care, independent of the individual physicians' personal views. However, participants emphasized the necessity of clinician autonomy to overrule tool suggestions when identifying clinical nuances not captured by the model. CONCLUSIONS: The colorectal cancer MDTs expressed a positive view regarding the use of prediction models and other forms of decision-support in their workflow. While clinicians and prediction models had similar risk score distributions, they diverged in the assessment of specific individual patients.
1 Aug 2026
Read appraisal →Biomedical physics & engineering express
BMSNet: a boundary-guided multi-scale polyp segmentation network
Accurate polyp segmentation in colonoscopy images is essential for computer-aided diagnosis and early colorectal cancer screening. However, this task remains challenging due to large variations in polyp size, shape, and texture, as well as blurred boundaries and low contrast between polyps and surrounding mucosa. To address these challenges, we propose boundary-guided multi-scale polyp segmentation network (BMSNet), a boundary-guided multi-scale network for accurate polyp segmentation. First, an Scale-Aware Modulation Meets Transformer-T backbone initialized with officially released pretrained weights is adopted to extract multi-level features efficiently. Then, a boundary prediction module is introduced to generate preliminary boundary priors through bidirectional feature interaction and differentiable Canny-based auxiliary supervision. Based on the generated boundary information, we design a boundary-guided feature enhancement module to refine feature representations. This module combines frequency-domain enhancement, multi-scale feature interaction, and boundary-guided modulation. Finally, a feature fusion unit is introduced to progressively integrate multi-scale features and generate the final segmentation prediction. Extensive experiments on five benchmark datasets demonstrate the effectiveness of the proposed method. On ClinicDB, Kvasir, ColonDB, ETIS, and CVC-300, BMSNet achieves Dice scores of 0.943, 0.920, 0.828, 0.825, and 0.913, respectively. These results indicate that BMSNet achieves competitive segmentation performance with moderate computational complexity under the evaluated experimental setting.
28 July 2026
Read appraisal →Journal of robotic surgery
Transfer of technical skills from virtual reality robotic simulation to non-virtual surgical performance: a systematic review and meta-analysis
Virtual Reality Robotic Surgery Simulators (VR-RSS) are widely used, but their educational value depends on whether skills acquired in virtual reality (VR) training transfer to non-virtual performance. This review evaluated whether VR training improves transferable robotic surgical skills. We searched PubMed, Scopus, Web of Science, and Cochrane from inception to March 3, 2026, for randomized trials evaluating VR robotic simulator training with transfer to non-virtual settings. Screening was performed in Rayyan. Data were extracted systematically, risk of bias was assessed with RoB 2, and random-effects meta-analyses using Hedges' g were conducted in R. Eight trials were included in the review, and seven were included in meta-analysis. Sample sizes ranged from 12 to 35 participants. Although the pooled standardized mean difference (SMD) for global technical skill scores favored the VR training group, it did not reach statistical significance (g = 0.77, 95% CI -0.64-2.18; p = 0.205). Task completion time was also shorter in the VR training group, but this difference was not statistically significant (g = 0.47, 95% CI -0.35-1.29; p = 0.164). This review suggests a potential benefit of VR-based robotic simulation for transferring trained skills to non-VR environments. Future studies should include larger sample sizes and longitudinal assessments using validated metrics to better characterize these trends.
21 July 2026
Read appraisal →Food & function
Efficacy of different dietary fibers for chronic idiopathic constipation: a systematic review and network meta-analysis
This study aims to compare the efficacy and safety of different dietary fiber subtypes in treating patients with chronic idiopathic constipation (CIC) through network meta-analysis (NMA). PubMed, Embase, Cochrane Library, and Web of Science were searched from inception to October 23, 2025. The risk of bias was evaluated using the Cochrane RoB 2.0 tool, and treatment rankings were calculated using the surface under the cumulative ranking curve (SUCRA). 17 randomized controlled trials (RCTs) involving 1423 adults with CIC were included. NMA suggested that viscous soluble fibers were associated with greater efficacy compared with placebo in relieving defecation difficulty (high certainty, class 1), and were also associated with improvements in stool consistency and stool frequency relative to placebo (moderate certainty, class 1). According to SUCRA rankings, viscous soluble fibers tended to have a higher probability of favorable performance compared with other dietary fiber subtypes in improving straining during defecation (moderate certainty), stool frequency (low certainty), and stool consistency (low certainty), and was associated with a relatively favorable tolerability profile (moderate certainty). Insoluble dietary fibers showed the highest ranking probability for improvement in overall constipation symptom scores; however, this finding was supported by very low certainty evidence. Overall, the findings suggest that viscous soluble fibers have a higher probability of improving stool consistency, stool frequency, and straining during defecation. In contrast, insoluble fibers showed a relatively higher ranking for improvement in overall constipation symptom scores. These results support a symptom-oriented and individualized approach to dietary fiber selection in the management of CIC.
21 July 2026
Read appraisal →Journal of robotic surgery
Technology-enhanced training in basic robotic surgical skills: a systematic review
Robotic surgery has seen substantial growth, however training access remains limited by resources and time-intensive credentialling pathways. Technology enhanced learning may address these barriers whilst improving training efficiency and assessment objectivity. We aimed to systematically evaluate technology-enhanced approaches to basic robotic surgical skills training. This systematic review followed PRISMA guidelines with prospective registration (OSF: V93MZ). MEDLINE (via OVID), Embase, and the Cochrane Library (CENTRAL and CDSR) were searched on 23 May 2025 without date restrictions. Technologies included virtual reality, augmented reality, sensor-based systems, video review, and performance feedback. Risk of bias was assessed using the RoB-2 and ROBINS-I tools, and certainty of evidence using GRADE. Narrative synthesis was utilised given methodological heterogeneity, with reported outcomes mapped to the Kirkpatrick model of training evaluation. Seventeen studies involving 484 participants were included (10 randomised trials, 7 prospective cohort studies; 16 single-centre). Technologies comprised virtual reality (7), sensor-based systems (5), performance feedback (3), video review (1), and augmented reality (1); no eligible artificial intelligence studies were identified. Virtual reality, utilised within bespoke robotic simulators, demonstrated accelerated learning curves, although some systems required more repetitions than traditional robotic simulator-based training. The single augmented reality study improved instrument velocity. Sensor-based force measurement devices improved force application and ergonomics. Video review improved accuracy but temporarily increased completion time. Multimodal feedback systems outperformed conventional feedback. Evidence was concentrated at Kirkpatrick level 2 (learning); one study reported only level 1 (reaction) data, with no studies reporting level 3 or 4 outcomes. Risk of bias was low in randomised studies and low-moderate in non-randomised studies. Technology-enhanced basic robotic skills training has evidence supporting improved performance outcomes, accelerated learning curves, and enhanced objective assessment capabilities. Gaps persist for basic skills training in augmented reality, artificial intelligence, and docking and setup. Standardised competency assessment frameworks are required to evaluate these rapidly progressing tools.
21 July 2026
Read appraisal →Science advances
Multistage microrobots with pH-responsive release of platelet membrane-coated nanoparticles
Targeted drug delivery in the gastrointestinal tract remains challenging because therapeutics must overcome multiple hierarchical barriers before reaching diseased tissue. Here, we present a multistage delivery platform that integrates magnetic microrobots, a pH-responsive protective coating, and platelet membrane-coated nanoparticles (PNPs) in one platform. A fillable design enables the formation of an internal magnetic layer for microrobot actuation, while the pH-responsive coating protects the cargo during transit and selectively degrades upon pH change, releasing cancer cell-targeting PNPs. In an in vitro colon cancer model that reproduces key gastrointestinal features, including flow, pH variation, and villi-like structures, this strategy increased nanoparticle retention and enhanced cancer cell cytotoxicity compared to nanoparticles administered alone. Ex vivo studies in porcine stomach and intestine further demonstrated robust locomotion on compliant and folded tissue surfaces. These results establish an environment-responsive hierarchical delivery strategy for more precise oral delivery in complex gastrointestinal settings.
20 July 2026
Read appraisal →Pediatric surgery international
ARM in ARM? Investigating the co-occurrence of anorectal malformations and labioscrotal anomalies.
Anorectal malformations (ARMs) are known to be associated with other defects, however there is no evidence addressing the co-occurrence of ARMs with anomalies of structures derived from the labioscrotal folds. We performed a systematic review (Prospero: CRD420251013945) aiming at investigating correlations and patterns between ARMs and labioscrotal anomalies (LSAs), when presenting simultaneously. Our search identified 1642 articles. After abstract screening, 344 full texts were assessed and 144 met the inclusion criteria. In total, these reported on 319 individual cases with both ARM and LSA. Individuals with complex ARMs had proportionally the most LSAs associated with their ARM - 58.4% of these were associated with only a single LSA. For 'high' and 'low' ARMs, 66.0% and 77.0% of cases were associated with a single LSA respectively. As data reporting prevalence of each type of ARM lacks information about LSAs, the potential to predict the presence of LSAs according to type of ARM is limited. Nevertheless, this review demonstrates a frequent association of ARMs with LSAs, that has not been previously described. This evidence highlights the importance of early screening and timely management of LSAs in patients with any type of ARM, with close collaboration between paediatric surgeons, urologists and other clinicians.
20 July 2026
Read appraisal →Pediatric surgery international
Single-stage versus staged reconstruction for high and intermediate anorectal malformations in neonates: a systematic review and meta-analysis
The surgical approach to high and intermediate anorectal malformations remains controversial. This systematic review compared continence and infection outcomes between the two approaches. MEDLINE, Embase, and Cochrane CENTRAL were searched per a registered protocol (PROSPERO CRD420261291319). Comparative studies of single-stage versus staged repair in neonates with high or intermediate anorectal malformations were included. Risk of bias was assessed using ROBINS-I. Random-effects meta-analysis with prediction intervals was performed. Four comparative studies (746 patients: 399 single-stage, 347 staged) were included. Pooled analysis showed no significant difference in incontinence outcomes (RR 0.93, 95% CI: 0.52-1.66, p = 0.80; 95% prediction interval: 0.28-3.08) with substantial heterogeneity (I²=83%). Infection data could not be pooled due to methodological concerns in outcome reporting. All included studies demonstrated moderate to serious risk of bias. GRADE certainty was LOW for incontinence and VERY LOW for infection. Current evidence does not demonstrate superiority of either approach for bowel continence (LOW certainty). Infection evidence was insufficient for synthesis (VERY LOW). Wide prediction intervals indicate substantial uncertainty; available data cannot confirm or refute inferior outcomes after single-stage repair. Prospective studies with standardized outcomes are needed. Protocol registration: PROSPERO CRD420261291319.
17 July 2026
Read appraisal →Nutricion hospitalaria
Immunonutrition in surgical patients with colorectal cancer: a systematic review
Introduction: surgery in patients with colorectal cancer often leads to complications, affecting recovery, prolonging hospital stay, and impacting quality of life. Immunonutrition plays an important role by modulating the immune system and may contribute to the reduction of these complications. Objective: to assess the effect of immunonutrient supplementation on complications in surgical patients with colorectal cancer. Materials and methods: a literature search was conducted in PubMed, Web of Science, Embase, and Cochrane, following the PRISMA model. A total of 20 clinical trials and two observational studies published between 2000 and 2025 were selected, analyzing postoperative complications, physical or nutritional status, immune response, and length of hospital stay. Results: a reduction in postoperative complications was observed with combined immunonutrient supplementation (arginine, omega-3 fatty acids, nucleotides, L-alanine, and L-glutamine), particularly in wound infections. No benefits were found with isolated supplementation. Conclusion: combined immunonutrient administration may improve postoperative complications, especially wound infections, although results regarding hospital stay are inconsistent. Isolated supplementation with omega-3 fatty acids shows no benefit. Introducción: la cirugía en pacientes con cáncer colorrectal conlleva complicaciones, alterando la recuperación, alargando la estancia hospitalaria y afectando a la calidad de vida. La inmunonutrición ejerce una acción importante al modular el sistema inmunológico y podría mejorar estas complicaciones. Objetivo: evaluar el efecto de la suplementación con inmunonutrientes en las complicaciones en pacientes quirúrgicos por cáncer colorrectal. Material y métodos: se realizó una búsqueda bibliográfica en PubMed, Web of Science, Embase y Cochrane, siguiendo el modelo PRISMA. Se seleccionaron 20 ensayos clínicos y dos observaciones, publicados entre 2000 y 2025, que analizaron complicaciones posoperatorias, el estado físico o nutricional, la respuesta inmune y la duración de estancia hospitalaria. Resultados: se halló una reducción de las complicaciones posoperatorias con una suplementación combinada de inmunonutrientes como arginina, ácidos grasos omega-3, nucleótidos, L-alanina y L-glutamina, especialmente en infecciones de herida, sin beneficios en la administración aislada. Conclusión: la administración combinada podría mejorar las complicaciones posquirúrgicas, especialmente de la infección de la herida, con resultados contradictorios para estancia hospitalaria. La suplementación aislada de ácidos grasos n-3 no aporta beneficios.
15 July 2026
Read appraisal →Hernia : the journal of hernias and abdominal wall surgery
Clinical outcomes in abdominal wall autologous flap reconstruction: a systematic review
PURPOSE: Complex abdominal wall defects present major reconstructive challenges when restoration of fascial continuity must be combined with durable soft-tissuecoverage, visceral protection, and, in selected cases, functional repair. Although autologous flap reconstruction is well established for selected indications, its overall role across defect types has yet to be established. METHODS: A systematic review was performed in accordance with PRISMA guidelines (PROSPERO: CRD42023491916). PubMed/MEDLINE, Embase, Scopus, Web of Science, Cochrane Library, and Google Scholar were searched from inception to April 2026 for studies reporting adult autologous flap-based abdominal wall reconstruction. Given substantial clinical and methodological heterogeneity, data were analysed descriptively. RESULTS: 172 studies comprising 4,421 patients were included. Pedicled flaps predominated (n=4,095, 92.6%), with free flaps used in 146 patients (3.3%) and mixed approaches in 180 (4.1%). Hernia-related pathology was the commonest indication (n=3751, 84.8%), followed by malignancy (n=470, 10.6%). Frequently used flap categories included anterolateral thigh, tensor fascia lata, rectus-based, and latissimus dorsi-based flaps. Flap selection was primarily defect-driven: fascia-bearing regional and turnover techniques were used mainly for structurally focused hernia repair, whereas composite flaps were used more often for oncologic, fistulating, traumatic, contaminated, or full-thickness defects. Common short-term complications include seroma, surgical site infection, and wound dehiscence. Flap failure and reported long term hernia recurrence were low. Free flaps were used more often for larger defects weighted mean 401 cm² vs 250 cm² for pedicled flaps), although size ranges overlapped. CONCLUSIONS: Autologous flap reconstruction is a durable and versatile option for selected complex abdominal wall defects. Reconstructive strategy should be guided primarily by defect composition.
10 July 2026
Read appraisal →Journal of gastrointestinal cancer
Radiomics Models as Tools for Predicting Genetic Mutations in Colorectal Cancer: A Systematic Review and Meta-Analysis
BACKGROUND: The evaluation of genetic mutations is crucial for personalized therapy in colorectal cancer (CRC), but the invasive tissue biopsy is subject to sampling bias and other complications. Radiomics has emerged as a non-invasive tool to predict these mutations from standard medical images. In this systematic review and meta-analysis, we aimed to evaluate the diagnostic accuracy and methodological quality of radiomics models for predicting key genetic mutations in CRC. METHODS: A comprehensive search of PubMed, Scopus, Web of Science, and Embase was conducted in accordance with PRISMA guidelines. Studies evaluating radiomics models for predicting genetic mutations in CRC patients using pre-operative CT, MRI, or PET/CT were included. A meta-analysis of diagnostic accuracy was performed to calculate the pooled sensitivity, specificity. Methodological quality was assessed using the Radiomics Quality Score (RQS) and QUADAS-2 tools. RESULTS: Sixteen studies were included in the quantitative analysis. The pooled sensitivity and specificity were 0.75 (95% CI, 0.67-0.81) and 0.78 (95% CI, 0.70-0.85), respectively, with an overall AUC of 0.79. Subgroup analyses revealed that radio-clinical models integrating both clinical and radiomics features achieved superior sensitivity compared to models with only radiological input. However, the overall methodological quality of the included studies was low, with a mean RQS of 45%. CONCLUSION: Conventional radiomics models demonstrate promising results for the non-invasive prediction of genetic mutations in CRC, with sensitivity enhanced by the integration of clinical data. Despite this potential, significant methodological shortcomings and heterogeneity across studies highlight the need for standardized protocols and large-scale, prospective validation before these models can be translated into routine clinical practice.
10 July 2026
Read appraisal →Journal of gastrointestinal cancer
Tumour-Associated Neutrophils in Colorectal Cancer: a Systematic Review, Meta-analysis, and Comprehensive Literature Review
PURPOSE: Neutrophils are at the forefront of innate immune response. The prognostic impact of tumour-associated neutrophils (TANs) remains unclear. This study aimed to evaluate the prognostic role of TANs in colorectal cancer (CRC). METHODS: A systematic review and meta-analysis were conducted using PubMed, Embase, and Scopus to identify studies correlating TANs with time-to-event survival analysis in patients with stages I-IV CRC from the date of inception until April 2024. Articles were included if they involved the identification of TANs through hematoxylin-eosin (HE) or immunohistochemistry (IHC) and reported survival analysis. Data was collected from the tumour core (TC) and invasive margin (IM). The hazard ratios (HRs) were extracted from the study results. Heterogeneity was evaluated using Cochran's Q and I². Primary and secondary endpoints were overall survival (OS) and disease-free survival (DFS). RESULTS: Of 4,292 citations found, 18 studies fulfilled eligibility criteria, encompassing 7,406 patients. IHC was used to identify TANs in 12 trials, while six performed HE. Five studies included patients with disease stages I-III, and 13 had stages I-IV. High TANs at the IM were associated with improved OS (HR = 0.64, 95% Confidence Interval [CI] 0.50-0.81, I2 = 48%) and DFS (HR = 0.47, 95% CI 0.25-0.88, I2 = 32%). At the TC, high TANs showed a trend towards better OS (HR = 0.82, 95% CI 0.61-1.10, I2 = 83%) and DFS (HR = 0.48, 95% CI 0.21-1.07, I2 = 93%), although not statistically significant and with high heterogeneity. CONCLUSION: In this systematic review and meta-analysis, high TANs at the invasive margin (IM), the tumour-host interface, was associated with improved prognosis in CRC. In contrast, no statistically significant association was observed for high TANs at the tumour core (TC).
9 July 2026
Read appraisal →Nature communications
A deep learning framework for efficient pathology image analysis
Artificial intelligence has transformed digital pathology by enabling biomarker prediction from high-resolution whole-slide images. However, current methods are computationally inefficient, processing thousands of redundant tiles per slide and requiring complex aggregation models. We introduce EAGLE (Efficient Approach for Guided Local Examination), a deep learning framework that emulates pathologists by selectively analyzing informative regions. EAGLE combines task-agnostic tile selection with detailed feature extraction and is benchmarked against leading slide- and tile-level foundation models across 43 tasks from nine cancer types spanning morphology, biomarker prediction, treatment response and prognosis. EAGLE outperforms patch aggregation methods by up to 23% and achieves the highest overall classification performance. It processes one slide in 2.27 s, reducing computational time by more than 99% compared with existing models. This efficiency supports rapid and auditable workflows by enabling review of the exact tiles used for each prediction and reducing dependence on high-performance computing. By reliably identifying informative regions and minimizing artifacts, EAGLE provides robust and auditable outputs, supported by systematic negative controls and attention concentration analyses. Its unified embedding enables rapid slide search, integration into multi-omics pipelines and emerging clinical foundation models.
2 July 2026
Read appraisal →International journal of medical informatics
Artificial intelligence in gastroenterology clinical practice: Scoping review of large language model applications.
BACKGROUND: Advances in artificial intelligence have brought renewed attention to tools that can work with the large amount of written information generated in clinical practice. Among these, large language models (LLMs) stand out for their ability to interpret and generate medical text in a flexible, context-aware way. Gastroenterology, like many specialties, produces a wide range of narrative and semi-structured data, and this has encouraged researchers to explore how LLMs might help clinicians manage everyday tasks. Recent studies have examined their potential contributions to patient education, communication between care teams and patients, decision support, and routine documentation, reflecting a growing interest in how these systems might fit into real-world clinical workflows. OBJECTIVE: This scoping review aimed to map current applications of LLMs in gastroenterology clinical practice, including subspecialty focus, study designs, model types, and reported outcomes. METHOD: Following PRISMA guidelines, a systematic search was conducted in PubMed, Scopus, and Web of Science for studies published between January 2022 and August 2025. Eligible studies included original research assessing LLM applications in gastroenterology clinical practice. Data were extracted on subspecialty, application domain, LLM type, data source, and outcomes. We employed thematic analysis to address our primary research question. RESULTS: 73 out of 2895 studies identified in the initial search met the inclusion criteria. Six subspecialties and six application domains emerged from our review. Hepatology (20/73 studies, 27.3%) and endoscopy (17/73 studies, 23.2%) were the most represented subspecialties. The most frequently investigated application domains were patient education and communication (38 studies) and decision support and clinical guidance (24 studies). Most studies were simulation-based or literature-based cases, although an increasing number have used real-world clinical data, particularly in recent years. The majority evaluated general-purpose models such as GPT-3.5 and GPT-4, with some incorporating retrieval augmentation or fine-tuning. Reported outcomes varied by application domain and included measures of accuracy, concordance, completeness, relevance, safety, reliability, usability, user satisfaction, efficiency, time savings, and educational value. Commonly described limitations included variable reliability, incomplete responses, and challenges in generalizing from simulated to clinical settings. CONCLUSIONS: Research on LLMs in gastroenterology has expanded across multiple subspecialties and application domains. Current evidence is primarily based on simulation studies, with limited but growing evaluation using real-world clinical data. Further work is needed to assess performance in prospective and applied clinical contexts.
2 July 2026
Read appraisal →Journal of gastrointestinal and liver diseases : JGLD
Colon Cleaning after Inadequate Bowel Preparation: A Pooled-data Analysis
BACKGROUND AND AIMS: An adequate bowel preparation is essential for performing high-quality ccolonoscopy; however, it remains ineffective for a certain percentage of patients. We conducted a literature search to identify evidence on salvage strategies following inadequate bowel preparation. METHODS: The search of medical literature was conducted using Medline and Google Scholar, restricted to studies published in English as a full paper. From retrieved studies, the success rate was calculated according to the rescue approaches used, including different oral- or device-assisted procedures, and pooling data of similar approaches. RESULTS: By considering data of 17 studies with 2,167 patients, an adequate colon cleaning was overall achieved in 79.3% (95%CI: 77.5-80.1). By grouping data of similar approaches, the success rate was 78.1% with oral high-volume polyethylene glycol (PEG), 75.3% with low-volume PEG, 82.5% with oral sennosides, and 84.6% following colonoscope or device-assisted preparation. CONCLUSIONS: Effective bowel cleansing is not achieved in nearly 20% of patients due to inadequate bowel preparation and although multiple approaches have been investigated, none has consistently provided adequate bowel preparation for all patients.
29 June 2026
Read appraisal →Public health research & practice
Optimising SMS content for bowel cancer screening participation in Australia: cross-sectional national survey findings.
OBJECTIVES AND IMPORTANCE OF STUDY: This study aimed to identify short message service (SMS) reminder content perceived as most likely to prompt bowel cancer screening, examine differences across sociodemographic subgroups, and explore preferences for timing and frequency. METHODS: Australian residents (N = 1016) aged 50-74 years completed an online survey rating five SMS reminders presented in random order. Outcomes included perceived usefulness, likelihood of encouraging kit return, likelihood of irritation and clarity. Preferences for timing and frequency were also assessed. Bayesian multilevel modelling (cumulative probit) compared ratings across SMS types, with effect sizes expressed as standard deviation (s.d.) differences in perceived likelihood that each SMS would encourage kit return compared with a reminder-only message. Interactions with age, gender, socioeconomic status and screening history were explored. RESULTS: Compared with a 'reminder-only' message, SMS content that encouraged storing the kit near the toilet (s.d. 0.44), conveyed general practitioner endorsement (s.d. 0.32) and gave instructions (s.d. 0.22) was more likely to prompt kit return. Responses varied slightly by age and area-level socioeconomic status. Most participants preferred two to three SMS reminders (mean 2.89, s.d. 6.73). CONCLUSIONS: SMS reminders using behavioural prompts and clear, concise content may support improved screening participation. Tailored SMS messages that reflect public preferences may increase kit return rates, support national screening goals and reduce bowel cancer mortality.
27 June 2026
Read appraisal →Journal of robotic surgery
Rethinking health technology assessment in robotic surgery: an EFISDS-TROGSS position paper. Official position paper of the European Federation - International Society for Digestive Surgery (EFISDS) and The Robotic Global Surgical Society (TROGSS).
Robotic-assisted surgery (RAS) has evolved from a procedural innovation into an increasingly integrated component of contemporary digital surgical ecosystems. Nevertheless, most current Health Technology Assessment (HTA) frameworks continue to evaluate robotic systems primarily through comparator-based models focused on isolated perioperative and oncological outcomes. In this EFISDS-TROGSS position paper, we critically examine the methodological limitations of conventional HTA paradigms when applied to robotic surgical platforms, using the recent Italian AGENAS appraisal as a representative case study. While the AGENAS document represents one of the most comprehensive national evaluations of RAS performed to date, its heterogeneous recommendations across procedures highlight unresolved tensions regarding perioperative benefit, real-world implementation, learning curves, organizational impact, and long-term healthcare value. We argue that RAS should increasingly be interpreted not simply as a surgical device, but as a platform technology interacting with simulation-based training, digital infrastructure, surgical data science, artificial intelligence, telecommunication systems, and institutional organization. Rather than a surgical device alone, RAS should be interpreted and regarded as a combination of technological advances and approaches that integrate various degrees of artificial intelligence autonomy, image navigation, telesurgery, and other benefits to empower the surgical team. Conventional HTA models, originally developed for relatively discrete therapeutic interventions, may incompletely capture the multidimensional interaction between robotic technologies and modern healthcare systems. Particular attention is dedicated to real-world evidence, implementation maturity, reimbursement limitations, and the growing mismatch between current Diagnosis-Related Group (DRG) structures and technologically integrated surgical care. Finally, we propose more flexible and multidimensional assessment frameworks integrating procedural outcomes with organizational sustainability, digital interoperability, workforce implications, and longitudinal healthcare value.
20 June 2026
Read appraisal →International journal of medical informatics
Automated Flow and local LLM-Driven clinical Context Engineering: Precision colorectal cancer recurrence registry
OBJECTIVE: This study develops and validates a deployable, privacy-preserving automated tool to address the labor-intensive nature of colorectal cancer (CRC) recurrence registration. Our tool utilizes a reproducible, two-stage 'Clinical Context Engineering' workflow to mimic expert clinical reasoning and overcome the limitations of handling longitudinal clinical data ambiguity. METHODS: We retrospectively studied 3053 CRC patients (2010-2018). A local Large Language Model (LLM) (Qwen3:14B) analyzed ∼ 19,900 pathology and ∼ 43,900 imaging reports using iterative, patch-based analysis ("raw LLM"). Clinical validation rules were applied to generate "rule-based LLM" output, enhancing explainability and trustworthiness. Both automated methods and the manual Taiwan Cancer Registry (TCR) database were compared against a 20% manual reference standard (N = 602). Full prompts and validation code are provided for complete reproducibility. RESULTS: Under a strict 60-day temporal tolerance, the Rule-Based LLM achieved 90.7% accuracy, comparable to standard TCR processes (92.0%), and 77.2% sensitivity. The application of clinical validation rules significantly improved specificity from 87.7% (Raw LLM) to 93.9% (Rule-Based LLM). In time window analysis, the Rule-Based LLM identified 87.1% of recurrences within 60 days of the reference date. CONCLUSION: Our locally deployed, privacy-preserving, and explainable Clinical Context Engineering framework offers a viable, non-inferior alternative to standard TCR processes, reducing workload while maintaining data quality and fostering trust in AI-assisted cancer registry automation.
16 June 2026
Read appraisal →Med (New York, N.Y.)
Neoadjuvant mFOLFOXIRI with or without cadonilimab versus mFOLFOX6 in locally advanced colorectal cancer: A randomized phase 2 trial (OPTICAL-2)
BACKGROUND: The efficacy of neoadjuvant chemotherapy combined with dual immune checkpoint blockade in proficient mismatch repair/microsatellite-stable (pMMR/MSS) locally advanced colorectal cancer (LACRC) remains uncertain. METHODS: In this open-label, randomized phase 2 trial (ClinicalTrials.gov: NCT05571644), eligible patients were randomly assigned (1:1:1) to receive 6 cycles of neoadjuvant mFOLFOXIRI plus cadonilimab, mFOLFOXIRI, or mFOLFOX6, followed by surgery and adjuvant chemotherapy. The primary endpoint was pathological complete response (pCR) in the intention-to-treat population. Secondary endpoints included major pathological response (MPR), safety, locoregional recurrence, disease-free survival, and overall survival. FINDINGS: Between July 2023 and August 2024, 123 patients were randomly assigned (41 per group). Neoadjuvant mFOLFOXIRI plus cadonilimab significantly improved pCR compared with mFOLFOX6 (26.8% versus 9.8%; odds ratio [OR], 3.4; 95% confidence interval [CI], 1.04 to 13.24; p = 0.046), whereas mFOLFOXIRI alone did not (14.6% versus 9.8%; OR, 1.6; 95% CI, 0.4 to 6.7; p = 0.50). MPR rates were 68.3%, 46.3%, and 43.9%, respectively, and were significantly higher with mFOLFOXIRI plus cadonilimab than with mFOLFOX6 (p = 0.026). Downstaging to ypStage 0-I occurred in 66%, 45%, and 41% of patients in the three groups, respectively, and was significantly higher in the mFOLFOXIRI plus cadonilimab group (p = 0.027). Grade ≥3 adverse events were manageable, with liver enzyme elevation and neutropenia being the most common. CONCLUSIONS: Neoadjuvant mFOLFOXIRI plus cadonilimab significantly improved pCR compared with mFOLFOX6 in LACRC, with a manageable safety profile, supporting further investigation. FUNDING: This work was funded by the National Science Fund for Distinguished Young Scholars (82425045) and the National Natural Science Foundation of China (82272800).
14 June 2026
Read appraisal →Journal of clinical oncology : official journal of the American Society of Clinical Oncology
Amivantamab Monotherapy in Chemorefractory RAS/BRAF Wild-Type Metastatic Colorectal Cancer: Results From OrigAMI-1, an Open-Label, Phase Ib/II Study
PURPOSE: Amivantamab, an EGFR-MET bispecific antibody with immune cell-directing activity, is approved in non-small cell lung cancer (NSCLC). Effective treatments are limited for chemorefractory metastatic colorectal cancer (mCRC). METHODS: OrigAMI-1 (ClinicalTrials.gov identifier: NCT05379595) is a phase Ib/II study evaluating amivantamab monotherapy in chemorefractory (2-3 prior lines) mCRC. Participants had centrally confirmed RAS/BRAF/EGFR ectodomain wild-type status, without ERBB2/HER2 amplification. Participants with left-sided mCRC without (cohort A) or with (cohort B) prior anti-EGFR antibody treatment, or right-sided mCRC (cohort C) regardless of prior anti-EGFR treatment, received intravenous amivantamab 1,050 mg (1,400 mg for ≥80 kg) once every 2 weeks. The primary end point was objective response rate (ORR) per RECIST v1.1. RESULTS: By October 31, 2024, 94 participants received amivantamab monotherapy (median follow-up, 11.9 months). The median age was 60 years, and 65% of participants were male, with a median of 2 prior lines (94%, prior bevacizumab). In left-sided cohorts, the ORR was 29% (5 of 17) in cohort A and 19% (10 of 54) in cohort B; the median duration of response (DoR) was 9.0 months and 6.1 months, and the median progression-free survival (PFS) was 5.7 months and 4.6 months, respectively. In the right-sided cohort, the ORR was 22% (10 of 23; 43% had prior anti-EGFR), the median DoR was 9.8 months, and the median PFS was 3.7 months. Most frequent treatment-related grade ≥3 adverse events (AEs) were rash (7%), dermatitis acneiform (4%), and hypoalbuminemia (4%). One participant discontinued amivantamab because of a treatment-related AE. CONCLUSION: Amivantamab monotherapy demonstrated promising, durable antitumor activity in chemorefractory mCRC, regardless of prior anti-EGFR therapy and the primary tumor location. The amivantamab safety profile in mCRC is consistent with experience in NSCLC. Amivantamab plus chemotherapy is currently being explored in two phase III studies in first-line and second-line mCRCs.
11 June 2026
Read appraisal →Techniques in coloproctology
Robotic versus laparoscopic TME for rectal cancer: meta-analysis of pathological quality indicators
BACKGROUND: Total mesorectal excision (TME) quality is a key determinant of oncological outcomes in rectal cancer. While robotic surgery offers technical advantages over laparoscopy in the confined pelvis, its superiority regarding pathological outcomes remains debated. We conducted a meta-analysis comparing robotic and laparoscopic TME focusing on quality indicators of TME and risk factors for incomplete TME. METHODS: A PROSPERO-registered systematic search of PubMed and EMBASE up to May 2025. Comparative studies reporting pathological outcomes of robotic versus laparoscopic TME were included. Primary endpoints were TME completeness, circumferential resection margin (CRM) positivity and distal resection margin (DRM) positivity. Secondary analysis included baseline characteristics (male gender, BMI, bulky tumours, distance to anal verge, neoadjuvant radiotherapy) and local recurrence rate. RESULTS: Fifty-six studies (27,648 patients; robotic 10,629, laparoscopic 17,019) were included. Robotic surgery was associated with significantly more complete TME specimens (OR 1.50, 95% CI 1.23-1.82, p < 0.001) and fewer positive DRMs (OR 0.68, 95% CI 0.48-0.97, p = 0.031). CRM positivity was comparable between groups (OR 0.93, 95% CI 0.77-1.12, p = 0.44). In random-effects analysis, there was a non-significant trend towards fewer local recurrences after robotic TME (OR 0.75, 95% CI 0.54-1.05, p = 0.09). Robotic cohorts more frequently included male patients, distal tumours and neoadjuvant chemoradiotherapy, suggesting preferential selection of technically challenging cases. CONCLUSIONS: Robotic TME is associated with higher specimen completeness and lower DRM positivity compared with laparoscopic TME, while CRM positivity and local recurrence rates appear broadly similar. These data support the use of robotics as a primary minimally invasive option for mid-low rectal cancer; however, as a result of low certainty of evidence, these findings should be interpreted cautiously. SYNOPSIS: This meta-analysis provides the most up-to-date synthesis of pathological outcomes comparing robotic and laparoscopic TME for rectal cancer, incorporating data from 56 studies including the recent REAL and COLRAR randomized trials. Unlike previous reviews, it exclusively analyses total mesorectal excision procedures, excluding partial or high anterior resections, thereby eliminating a major source of heterogeneity. The findings demonstrate that robotic TME is associated with higher specimen completeness and fewer positive distal margins, even in technically demanding mid- and low-rectal cancers, supporting the role of robotics as the preferred minimally invasive approach in mid-low, difficult tumours.
1 June 2026
Read appraisal →Journal of gastrointestinal cancer
Durable Survival Outcomes After Liver Transplantation for Colorectal Liver Metastases: A Systematic Review and Single-arm Meta-analysis of Prospective Clinical Trials
BACKGROUND: Liver transplantation (LT) has emerged as a potential treatment option for selected patients with colorectal liver metastases (CRLM), but existing meta-analyses are limited by heterogeneous populations and mixed treatment modalities. METHODS: A systematic review and single-arm meta-analysis were conducted in accordance with PRISMA guidelines. Studies reporting outcomes following LT for CRLM were included. Primary outcomes were overall survival (OS) and progression-free survival (PFS) at predefined time points. Secondary outcomes included adverse events. Pooled proportions were calculated using random-effects models with logit transformation and Hartung-Knapp adjustment. Heterogeneity was assessed using the I² statistic. RESULTS: Four prospective studies (1 RCT, 3 prospective cohorts) comprising 118 patients were included. Studies were conducted in Norway (SECA-I, n = 21; SECA-II, n = 15), the United Kingdom (n = 50), and a European multicentre RCT (TransMet, n = 32). Median age ranged from 55 to 58 years; all patients had liver-only metastases with resected primary tumors and had received pre-transplant chemotherapy. The pooled 3-year OS was 69% (95% CI, 52-82%; I² = 0.0%), and the pooled 5-year OS was 62% (95% CI, 36-83%; I² = 14.6%). The pooled 3-year PFS was 33% (95% CI, 7-76%), with substantial heterogeneity (I² = 62.8%). The pooled incidence of adverse events was 41% (95% CI, 25-60%; I² = 0.0%). Sensitivity analyses confirmed the robustness of pooled estimates. The retrospective literature also supported the finding that LT can achieve meaningful long-term survival in carefully selected patients. CONCLUSIONS: In carefully selected patients with CRLM, LT is associated with durable long-term survival and an acceptable safety profile, supporting further prospective evaluation and refinement of selection criteria.
31 May 2026
Read appraisal →BMC surgery
Improvement of mobility and motivation in patients with elective colorectal resection using tracking devices and utilizing self-awareness (IMPETUS): a randomized controlled trial in a traditional non-ERAS clinical setting
BACKGROUND: Early postoperative mobilization is a cornerstone of modern perioperative care, including in elective colorectal surgery. Here we evaluated the effectiveness of smart wearables for improving patients' mobility and motivation during the recovery period after elective colorectal resection. METHODS: This prospective randomized two-armed clinical trial enrolled 62 patients undergoing elective colorectal resection. These patients were randomized into an intervention group that received hourly active reminders to mobilize via a smartwatch, and a control group that received only passive monitoring. All patients were given a wearable device that tracked their daily step counts. The primary end-point was the total number of steps taken within five postoperative days. Secondary end-points included complication rates, length of hospital stay, and subjective motivation, assessed using a structured questionnaire. RESULTS: Compared to the control group, the intervention group achieved significantly higher step counts (p = 0.039), and reported increased motivation for mobilization (p = 0.016). The groups did not significantly differ in overall complication rates or duration of hospital stay. CONCLUSION: Smart wearable devices with hourly active reminder functionalities effectively promoted postoperative mobility and motivation. TRIAL REGISTRATION: German Clinical Trials Register (DRKS), DRKS00039731. Registered 24.03.2026. Retrospectively registered.
29 May 2026
Read appraisal →Medicine
Survival prediction of colorectal cancer using 101 machine learning methods based on immune-related genes: A machine learning study
Colorectal cancer (CRC) is characterized by high incidence and mortality rates. Early identification of high-risk CRC patients and timely intervention are crucial for reducing mortality rates. This study aims to develop a prognostic model incorporating immune-related genes using 101 machine learning algorithms. We utilized data from 630 CRC patients in The Cancer Genome Atlas Program database and the Gene Expression Omnibus database as a validation set. We identified immune-related genes from the ImmPort repository and conducted differential gene expression analysis. We developed 101 machine learning models, including random survival forest, elastic net, LASSO, Ridge regression, stepwise Cox, Coxboost, partial least squares regression for Cox, supervised principal component analysis, generalized boosted modeling, and survival support vector machines. The model with the highest mean concordance index was selected as the optimal prognostic model. The Ridge regression model demonstrated the most robust performance with a mean concordance index of 0.67. Patients with higher immune-related gene risk score (IRGRS) had significantly lower survival rates (log-rank test, P < .001). Univariate and multivariate Cox regression analyses confirmed IRGRS as an independent prognostic factor. Nomograms successfully predicted overall survival at 1, 2, and 3 years. IRGRS correlated with immune cell infiltration levels and stromal activity in the tumor microenvironment. High IRGRS scores were associated with higher tumor immune dysfunction and exclusion scores, suggesting potential differences in response to immune checkpoint inhibitors. In addition, high IRGRS patients exhibited greater sensitivity to certain chemotherapy drugs. The IRGRS score is a potent predictor of survival in CRC patients. This study offers a new perspective for CRC prognosis and supports the development of personalized treatment strategies.
24 May 2026
Read appraisal →BMC cancer
Efficacy and risk analysis of monotherapy and dual immunotherapy in dMMR/MSI-H metastatic colorectal cancer: a meta-analysis based on randomized controlled trials
BACKGROUND: Immune checkpoint inhibitors (ICIs) have transformed the treatment landscape for metastatic colorectal cancer (mCRC) exhibiting DNA mismatch repair deficiency (dMMR) or microsatellite instability-high (MSI-H). However, the comparative efficacy and safety of monotherapy versus dual immunotherapy remain inadequately quantified. This meta-analysis aims to evaluate the benefits and risks of these strategies by quantitatively synthesizing data from randomized controlled trials (RCTs) to inform clinical decision-making. DESIGN: We systematically searched major databases (PubMed, Web of Science, Cochrane, Embase) from inception to September 2025 for relevant RCTs. Outcomes included objective response rate (ORR), progression-free survival (PFS), overall survival (OS), duration of response (DOR), and adverse events. Statistical analyses were performed using RevMan 5.3 and Stata 18.0. RESULTS: Six RCTs (1,460 patients) were included. Pooled analysis showed that PD-1/PD-L1 monotherapy significantly improved ORR compared with chemotherapy (OR = 1.52, 95% CI: 1.02-2.27, P = 0.04), based on two RCTs (n = 429), and demonstrated a significantly lower incidence of grade ≥ 3 adverse events (OR = 0.14, 95% CI: 0.08-0.23, P < 0.00001; 2 studies). Dual immunotherapy with nivolumab plus ipilimumab significantly improved ORR versus monotherapy (OR = 1.77, 95% CI: 1.25-2.49, P = 0.001), based on one RCT (n = 582), with a significantly increased toxicity risk (OR = 1.73, 95% CI: 1.22-2.44, P = 0.002). For PFS, KEYNOTE-177 (HR = 0.59, 95% CI: 0.45-0.79) and CheckMate 8HW (HR = 0.62, 95% CI: 0.48-0.81) showed significant benefits, while SAMCO-PRODIGE 54 and CheckMate 9 × 8 did not demonstrate statistically significant PFS improvement. OS hazard ratios across four studies ranged from 0.74 to 1.03, with all 95% confidence intervals crossing 1.0, indicating no statistically significant survival benefit in any individual study. DOR was consistently longer with immunotherapy, with median not reached for pembrolizumab versus 10.6 months for chemotherapy in KEYNOTE-177. CONCLUSIONS: Effect estimates for PFS, OS, and DOR generally favored immunotherapy; however, confidence intervals were wide and statistical significance was not consistently observed across trials. PD-1/PD-L1 monotherapy demonstrated a favorable safety profile and a statistically significant improvement in ORR compared with chemotherapy (OR = 1.52, 95% CI: 1.02-2.27, P = 0.04). Dual immunotherapy significantly improved ORR compared with monotherapy (OR = 1.77, 95% CI: 1.25-2.49, P = 0.001), but this benefit was accompanied by significantly increased toxicity (OR = 1.73, 95% CI: 1.22-2.44, P = 0.002) and did not consistently translate into statistically significant PFS or OS benefits across trials. These findings highlight the need for personalized risk-benefit assessment. Future research should focus on biomarker-driven patient selection and strategies to overcome resistance.
23 May 2026
Read appraisal →Journal of medical Internet research
Benchmarking Large Language Models and Prompt Engineering Strategies in Microsatellite Instability Cancers: Evaluation Study
BACKGROUND: The reliability of general-purpose large language models (LLMs) for complex clinical tasks in specialized domains such as microsatellite instability (MSI) cancers remains critically uncharacterized. The absence of a domain-specific benchmark to evaluate and guide the optimization of their capabilities across diverse clinical tasks poses unevaluated risks to patient safety. OBJECTIVE: This study aimed to develop and validate Microsatellite Instability Cancer Benchmark (MSIC-Bench), a novel, two-tiered benchmark for MSI cancer, covering both consensus and frontier knowledge. Using this framework, we aimed to systematically assess LLM performance across various prompting strategies, identify task-specific weaknesses, and reveal effective pathways for performance improvement. METHODS: We developed MSIC-Bench, a 511-question benchmark derived from clinical guidelines and a curated knowledge base. Three state-of-the-art LLMs (GPT-4o, [OpenAI], Gemini 2.5 Pro [Google], and Claude Opus 4 [Anthropic]) were evaluated using 4 prompting strategies, including vanilla, chain-of-thought, reflection of thoughts, and retrieval-augmented generation (RAG), under both multiple-choice and open-ended modalities. Performance was assessed on accuracy, safety (honesty), error composition, and token usage. RESULTS: LLMs demonstrated a significant "scaffolding effect," with accuracy dropping substantially in open-ended scenarios. For non-RAG strategies, the primary failure mode was an internal knowledge deficit. The integration of RAG proved to be the most effective intervention. A domain-aligned RAG strategy not only significantly improved accuracy in complex decision-making tasks but also fundamentally shifted the system's primary bottleneck from knowledge deficits to retrieval failures. In terms of safety, RAG induced a favorable shift from high-risk fabrications to safer refusals, though this introduced a safety-utility trade-off in the form of false refusals. Notably, our hybrid-RAG configuration, which combined both knowledge sources, demonstrated the most robust and generalizable performance across all tasks. CONCLUSIONS: Current LLMs lack the specialized knowledge required for reliable application in MSI oncology. A well-designed RAG architecture is the pivotal intervention to address this gap. However, its success is not automatic; it transforms the nature of system failure, making retrieval precision and knowledge base quality the new critical determinants of performance and safety. Our findings establish a clear directive for developing trustworthy clinical artificial intelligence: focus must shift toward optimizing the retrieval component and curating high-quality, comprehensive knowledge sources. MSIC-Bench provides a robust framework to guide these future efforts.
22 May 2026
Read appraisal →Journal of medical Internet research
Large Language Models in Colorectal Cancer Care and Clinical Decision Support: Systematic Review
BACKGROUND: Colorectal cancer (CRC) is a leading cause of cancer morbidity and mortality worldwide. The complexity of guideline-concordant care and unstructured clinical data has driven demand for decision-support tools. Large language models (LLMs) show promise for processing clinical data and patient-provider communication, yet evidence is fragmented, and a CRC-specific synthesis across the full care continuum is lacking. OBJECTIVE: This systematic review evaluates the current applications, performance determinants, and clinical implications of LLMs across the continuum of CRC care. METHODS: Following PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses), we searched 6 databases (PubMed, Embase, Web of Science, Scopus, CINAHL, Cochrane) through April 1, 2026. Eligible studies were peer-reviewed original investigations of LLMs on CRC tasks with extractable outcomes; reviews, editorials, and abstracts were excluded. Two reviewers assessed quality with QUADAS-2 (Quality Assessment of Diagnostic Accuracy Studies-2), PROBAST (prediction model risk of bias assessment tool), and ROBINS-I (Risk of Bias in Nonrandomized Studies - of Interventions). Data on model types, applications, prompts, input/output formats, and outcomes were analyzed descriptively, with narrative synthesis per synthesis without meta-analysis (SWiM) guidelines. RESULTS: Of 8880 records, 37 studies met inclusion criteria (2023-2026), mostly from China and the United States, with GPT series most frequently evaluated. Overall risk of bias was low in 10/37 studies (27.0%), moderate in 14/37 (37.8%), unclear in 7/37 (18.9%), and high or serious in 6/37 (16.2%). Problematic domains included outcome measurement, intervention classification, patient selection, and lack of blinded assessment. LLMs showed utility in automating data extraction from clinical texts, supporting patient education, aiding diagnosis, and assisting clinical decision-making, with emerging visual interpretation and multimodal capacities. Domain-specific and multimodal models showed advantages over general-purpose models in certain tasks. Performance was significantly influenced by prompt design, from zero-shot queries to fine-tuning. Despite efficiency and outcome benefits, challenges persist regarding methodological quality, data privacy, and generalizability. CONCLUSIONS: This review provides an integrative framework synthesizing evidence across study designs and LLM categories in CRC care. Unlike prior reviews addressing gastroenterology broadly or limited to one design, it covers the full CRC continuum and, for the first time, comparatively evaluates general-purpose, domain-specific, and multimodal LLMs, clarifying how prompt engineering and heterogeneous metrics shape outcomes. Although findings support LLMs' clinical potential, results must be interpreted cautiously, given low overall evidence quality. Most studies lacked safeguards against bias-blinded assessment, confounder adjustment, or prospective multicenter validation. Substantial heterogeneity across tasks, LLM types, prompts, reference standards, and outcomes means reported advantages cannot be generalized. Future work should prioritize real-world integration via prospective multicenter validation, robust privacy frameworks, and rigorous human oversight. Amid rising global CRC burden and health care disparities, this review informs clinical translation, equitable scaling, and policy on LLM deployment.
22 May 2026
Read appraisal →Journal of robotic surgery
A randomized controlled trial comparing robotic NOSES versus robotic TME for Mid-rectal cancer: short-term oncological and perioperative outcomes
This randomized study compared perioperative outcomes between robotic natural orifice specimen extraction surgery (R-NOSES) and robotic total mesorectal excision (R-TME) in participants with mid-rectal cancer. After exclusions, 71 participants in the R-TME cohort and 69 in the R-NOSES cohort were included in the final analysis, from an initial randomized cohort of 150 eligible participants (1:1 allocation). The two cohorts had comparable preoperative characteristics. Intraoperative and pathological outcomes, including blood loss, operative duration, and harvested lymph nodes, showed no notable differences. The rates of overall complication were comparable between the two groups (R-NOSES 13.0% vs. R-TME 15.5%, P = 0.679), and anastomotic leakage did not vary significantly. Conversely, the R-TME group experienced a higher rate of wound complications (5.6% vs. 0.0%, P = 0.045). Notably, the R-NOSES group demonstrated significant advantages in postoperative recovery, including shorter time to first flatus (48 vs. 53 h, P = 0.02), earlier resumption of a liquid diet (79 vs. 84 h, P = 0.041), and markedly reduced VAS pain ratings on postoperative days 1 and 3. In conclusion, for selected patients with middle rectal cancer, R-NOSES appears to be a safe and feasible alternative to R-TME, with similar short-term oncological outcomes. It may offer some benefits in terms of faster recovery, less postoperative pain, and fewer wound-related complications.Trial Registration: ClinicalTrials.gov Identifier NCT06454201.
22 May 2026
Read appraisal →Journal of the Egyptian National Cancer Institute
Preclinical efficacy of drug delivery systems in colon cancer therapy: a systematic review and meta-analysis of in vivo animal studies
BACKGROUND: Drug delivery systems (DDS) offer a promising strategy to enhance the therapeutic index of chemotherapeutic agents in colon cancer by improving tumor targeting, circulation time, and controlled drug release. Despite extensive preclinical research, a quantitative synthesis evaluating the efficacy of DDS and the influence of design parameters remains lacking. METHODS: We conducted a systematic review and meta-analysis of preclinical in vivo studies assessing DDS-based chemotherapy in murine models of colon cancer. A comprehensive search of PubMed, EMBASE, Scopus, Google Scholar, Cochrane CENTRAL, and ClinicalTrials.gov was performed through October 15, 2025. Outcomes included tumor growth inhibition, with subgroup analyses examining the effects of DDS platform, chemotherapeutic agent, targeting strategy, ligand type, and route of administration. RESULTS: Twenty-three studies comprising 25 experiments and 539 animals were included. Overall, DDS-based therapies significantly reduced tumor growth compared with controls ((WMD: - 557.7; 95% CI: - 716.8 to - 398.5; I²=88%; p < 0.001) and free-drug administration ((WMD: - 276.3; 95% CI: - 367.6 to - 185.1; I²=78%; p < 0.001). Both targeted and non-targeted DDS significantly reduced tumor growth compared with free drug treatment, while targeted DDS showed significantly greater efficacy than non-targeted systems (WMD: - 240.3; 95% CI: - 399.4 to - 81.25; I²=59%; p = 0.003). Although no statistically significant differences were observed between DDS platforms or chemotherapeutic agent subgroups, micelle-based systems and DDS formulations incorporating SN-38 or doxorubicin tended to show greater tumor growth reduction. Intravenous administration demonstrated significantly greater efficacy than intraperitoneal delivery, while hyaluronic acid- and aptamer-based targeting strategies achieved the largest tumor growth inhibition. Risk-of-bias assessment indicated moderate methodological quality, with variability in reporting of randomization, blinding, and sample size calculations. CONCLUSIONS: DDS-based chemotherapy consistently improves antitumor efficacy in preclinical colon cancer models, with targeting strategy, platform type, chemotherapeutic agent, and administration route influencing outcomes. These findings support the rational design of DDS platforms and underscore their translational potential. Rigorous preclinical study design and standardized reporting of efficacy and safety are essential to facilitate clinical translation.
21 May 2026
Read appraisal →Scientific reports
Quantum machine learning for predicting anastomotic leak: a clinical study
Anastomotic leak is a life-threatening complication following colorectal surgery. This study benchmarks Quantum Neural Networks (QNNs) against hyperparameter-tuned classical models (logistic regression, multi-layer perceptrons, boosting algorithms) for anastomotic leak prediction. Using a 200-patient clinical dataset strictly bounded by a priori medical constraints, we simulated QNNs with ZZFeatureMap encoding and EfficientSU2/RealAmplitudes ansatze under realistic hardware noise. To ensure statistical reliability, performance metrics were averaged across 10 independent optimization runs. The EfficientSU2-BFGS configuration achieved the highest mean AUC of [Formula: see text], while RealAmplitudes with CMA-ES maximized Average Precision ([Formula: see text]). Crucially, at a fixed, clinically necessary sensitivity of [Formula: see text], specific QNN configurations achieved significantly higher specificity (up to [Formula: see text]) and Negative Predictive Value (up to [Formula: see text]) compared to classical models (maximum [Formula: see text] and [Formula: see text], respectively), effectively minimizing false positives. However, classical models maintained superior probability calibration for continuous risk stratification. We conclude that QNNs offer robust discriminative performance for clinical screening, warranting further validation on larger, independent cohorts.
20 May 2026
Read appraisal →Journal of clinical oncology : official journal of the American Society of Clinical Oncology
Tissue-Free Circulating Tumor DNA Assay and Patient Outcome in a Phase III Trial of FOLFOX-Based Adjuvant Chemotherapy (Alliance N0147)
PURPOSE: Detection of molecular residual disease using circulating tumor DNA (ctDNA) may enable postoperative risk stratification and guide adjuvant therapy. We evaluated the prognostic value of a tissue-free, epigenomic ctDNA assay in patients with stage III colon cancer (CC) enrolled in a phase III adjuvant chemotherapy trial. METHODS: Plasma samples were collected after surgery and before adjuvant infusional fluorouracil, leucovorin, and oxaliplatin alone or combined with cetuximab. ctDNA was analyzed using a tissue-free assay; in ctDNA-positive patients, tumor fraction (TF) was quantified and genotyping was performed with a 739-gene panel. Associations with disease-free survival (DFS), time to recurrence (TTR), and overall survival (OS) were assessed using multivariable Cox models adjusted for covariates. RESULTS: Among 2,260 evaluable patients, 461 (20.4%) were ctDNA-positive with significantly higher detection in advanced T-/N-stage, high-grade, obstruction/perforation, and BRAFV600E tumors. At a median follow-up of 6.1 years, ctDNA positivity was independently associated with shorter TTR (hazard ratio [HR], 5.96 [95% CI, 5.11 to 6.96]), DFS (HR, 5.03 [95% CI, 4.36 to 5.81]), and OS (HR, 4.45 [95% CI, 3.76 to 5.27]; all P < .0001). The 5-year DFS was 27.7% (95% CI, 23.8 to 32.2) v 77.1% (95% CI, 75.1 to 79.1) in ctDNA-positive versus ctDNA-negative patients, and adverse prognostic impact was greater in lower T/N stage, low-risk, and dMMR subsets (interaction P = .0012-.041). Among ctDNA-positive patients, TF was nearly double in those who recurred or died (P = .0002) and stratified patients for TTR, DFS, and OS (all adjusted P < .002). Genotyping identified mutations in FLT1 (OR, 8.99) and PREX2 (OR, 7.73) genes that were most strongly associated with recurrence (P < .03). CONCLUSION: Evaluation of ctDNA in resected stage III CC using a tissue-free assay provided robust and independent prognostic value. Higher ctDNA burden, dMMR, and specific mutations defined poor prognostic groups among ctDNA-positive patients.
20 May 2026
Read appraisal →Journal of clinical oncology : official journal of the American Society of Clinical Oncology
Randomized Phase II/III Trial Comparing Hepatectomy, Followed by mFOLFOX6 With Hepatectomy Alone for Liver Metastasis From Colorectal Cancer: Long-Term Results of JCOG0603
JCOG0603 demonstrated improved disease-free survival (DFS) with adjuvant mFOLFOX6 after hepatectomy compared with hepatectomy alone in colorectal liver-only metastasis (CRLM), but the overall survival (OS) data were immature. Here, we report OS after long-term follow-up. Eligible patients with colorectal adenocarcinoma and an unlimited number of liver metastases were randomized to adjuvant mFOLFOX6 or hepatectomy alone. DFS was the primary endpoint, and OS was a secondary endpoint. Between March 2007 and January 2019, 151 and 149 patients were randomized to adjuvant chemotherapy and hepatectomy alone, respectively. At a median follow-up of 7.7 years for disease-free surviving patients, 54 (35.8%) and 51 (34.2%) patients had died in the respective arms, (hazard ratio [HR], 1.07 [95% CI, 0.73 to 1.57]). Five-year OS was 73.4% (95% CI, 65.5 to 79.7) and 80.1% (95% CI, 72.6 to 85.7) and 7-year OS was 69.4% (95% CI, 61.2 to 76.2) and 72.4% (95% CI, 64.2 to 79.1), respectively. One patient in the adjuvant chemotherapy arm possibly died of protocol treatment-related toxicity, and one in the hepatectomy-alone arm died of post-protocol treatment complications. Five-year DFS was respectively 49.7% (95% CI, 41.5 to 57.3) and 40.5% (95% CI, 32.5 to 48.3) in the adjuvant chemotherapy and hepatectomy-alone arms (HR, 0.72 [95% CI, 0.54 to 0.97]). Long-term OS did not differ with adjuvant mFOLFOX6 compared with hepatectomy alone in resectable CRLM. Adjuvant mFOLFOX6 may delay recurrence but did not improve long-term survival.
20 May 2026
Read appraisal →Journal of robotic surgery
Perioperative and oncological outcomes of robotic versus laparoscopic low anterior resection in younger rectal cancer cohorts: a systematic review and meta-analysis with narrative functional assessment
Robotic low anterior resection (R-LAR) has been proposed to overcome technical limitations of laparoscopy in rectal surgery. However, previous meta-analyses have evaluated mixed-age populations, and no meta-analysis has specifically investigated younger patients (aged < 65 years) undergoing low anterior resection. This study aimed to compare operative, short-term postoperative, and oncological outcomes between robotic and laparoscopic low anterior resection in this cohort. A systematic review and meta-analysis were conducted according to PRISMA 2020 and the Cochrane Handbook, with prospective registration in Prospero. PubMed, Scopus, and the Cochrane Library were searched up to January 2026. Comparative studies evaluating R-LAR versus L-LAR in younger rectal cancer cohorts, defined as studies with mean patient age ≤ 65 years, were included. Random-effects models were used to calculate pooled mean differences (MD), odds ratios (OR), and hazard ratios (HR) with 95% confidence intervals (CI). Risk of bias was assessed using ROBINS-I and RoB2, and certainty of evidence using GRADE. A total of 16 studies were included. Compared with laparoscopy, R-LAR was associated with longer operative duration (MD 24.59 min, 95% CI 3.85-45.33, p = 0.02, I² = 96%), lower conversion to open surgery (OR 0.38, 95% CI 0.27-0.53, p < 0.0001, I² = 38%), lower overall complications (OR 0.84, 95% CI 0.73-0.97, p = 0.02, I² = 0%), lower 30-day mortality (OR 0.59, 95% CI 0.45-0.77, p = 0.006, I² = 0%), lower 30-day reoperation (OR 0.77, 95% CI 0.61-0.98, p = 0.04, I² = 0%), higher complete TME rates (OR 3.30, 95% CI 2.14-5.10, p = 0.003, I² = 0%) and shorter hospital stay (MD -0.82, 95% CI -1.50 to -0.13, p = 0.02, I²=97%). Estimated blood loss (p = 0.05) and major complications (p = 0.05) were borderline significant in favor of R-LAR. Anastomotic leakage (p = 0.17), postoperative ileus (p = 0.49), time to first flatus (p = 0.12), time to diet initiation (p = 0.10), CRM positivity (p = 0.56), lymph node yield (p = 0.09), local recurrence (p = 0.67), loop ileostomy (p = 0.56), disease-free survival (p = 0.53), and overall survival (p = 0.73) were comparable. The certainty of evidence ranged from very low to moderate. Functional outcomes were insufficient for pooling. Robotic low anterior resection in younger rectal cancer cohorts may improve several perioperative and technical outcomes without compromising oncological adequacy. Further high-quality studies are required to confirm these findings.
19 May 2026
Read appraisal →Techniques in coloproctology
Bridging the gap: a systematic literature review and meta-analysis on the management of rectal wall defect after transanal excision
BACKGROUND: Transanal resection techniques have gained considerable importance in the treatment of benign and malignant rectal neoplasms. However, there is no definitive consensus on whether the rectal defect should be closed or left open after excision. We sought to provide an updated pooled analysis of the management of rectal wall defect after transanal excision. METHODS: In accordance with PRISMA and Cochrane guidelines, this meta-analysis was performed using the PubMed (MEDLINE), Cochrane Central Register of Controlled Trials, and Google Scholar databases to identify studies comparing perioperative outcomes after rectal defect closure versus leaving the defect open following full-thickness (FT) excision. Odds ratios (ORs) and standardized mean differences (SMDs) with 95% confidence intervals (CIs) were calculated. Heterogeneity was assessed using Cochrane's Q test. Risk of bias and certainty of evidence were judged by ROBINS-I and GRADE, respectively. RESULTS: Six comparative studies meeting the inclusion criteria were included in the final analysis. Closing the rectal defect was associated with significantly reduced rectal bleeding (OR = 0.57, 95% CI: 0.35-0.94, p = 0.03; I2 = 26%) and re-admission rates (OR = 0.34, 95% CI: 0.16-0.76, p = 0.008; I2 = 0%) compared with the open group, while other outcomes were not significantly different. A prolonged operative time was noted when the rectal wall defect was sutured (SMD = 0.15, 95% CI: 0.03-0.28, p = 0.02; I2 = 44%). CONCLUSIONS: The present analysis revealed that both approaches are safe and technically feasible; however, closing the rectal wall defect after FT excision of rectal neoplasms was associated with lower bleeding and re-admission rates. Nevertheless, randomized studies with homogeneous protocols and consistent long-term outcome data are still needed to provide definitive answers.
17 May 2026
Read appraisal →Clinical cancer research : an official journal of the American Association for Cancer Research
Circulating Tumor DNA Dynamics and Clinical Outcomes in Patients with Advanced Colorectal Cancer Treated with Cetuximab-Based Induction and Maintenance Treatment
PURPOSE: We investigated whether circulating tumor DNA (ctDNA) changes may be useful to assess clinical outcomes in patients with metastatic colorectal cancer (mCRC) randomized in the TIME-PRODIGE-28 trial comparing biweekly maintenance with cetuximab alone with observation after 4-month fluorouracil, folinic acid, and irinotecan (FOLFIRI) plus cetuximab induction chemotherapy. EXPERIMENTAL DESIGN: ctDNA samples were collected at four time points from baseline until disease progression during the first chemotherapy-free interval and analyzed using next-generation sequencing and methylation marker approaches. Progression-free survival (PFS) and overall survival (OS) from randomization were analyzed according to ctDNA kinetics and EGFR-MAPK pathway alterations. RESULTS: Among 139 randomized patients, 104 (74.8%) had paired samples available. Patients with negative baseline ctDNA remaining negative after 4-month induction chemotherapy had significantly longer PFS from randomization (9.6 months) as compared with patients with a ctDNA decrease of ≥80% (3.4 months) or a ctDNA decrease of <80% (2.1 months; P = 0.013). Patients with EGFR-MAPK pathway alterations identified either in tissue or baseline ctDNA had worse PFS and OS from randomization. Acquired alterations found in 17 of 63 (26.9%) patients at disease progression during the first chemotherapy-free interval were associated with worse OS from reintroduction of the full induction chemotherapy (14.9 vs. 19.4 months; P = 0.025). CONCLUSIONS: Our findings show the prognostic impact of both ctDNA kinetics and EGFR-MAPK pathway alteration dynamics following induction chemotherapy with FOLFIRI-cetuximab in patients with mCRC. Prospective studies evaluating ctDNA-guided treatment strategies are needed to validate the clinical utility of ctDNA monitoring to improve patient selection for first-line treatment de-escalation and anti-EGFR-based maintenance regimens, including treatment adaptation over time.
16 May 2026
Read appraisal →BMJ open gastroenterology
Development and validation of multivariable prognostic machine learning models to identify patients at risk for inadequate bowel preparation when undergoing colonoscopy: a prospective, multicentre study
OBJECTIVE: Inadequate bowel preparation impairs the accuracy of colonoscopy and increases the burden on patients and healthcare systems. Consequently, the quality of bowel preparation is an important quality indicator. We aim to develop and validate multivariable prognostic models for the identification of patients at risk of inadequate bowel preparation using machine learning (ML). METHODS: Demographic and clinical data from consecutive patients ≥18 years of age who underwent colonoscopy at six centres in Germany were prospectively collected. Adequate bowel preparation was defined as Boston Bowel Preparation Scale ≥6 with a value ≥2 in each colonic segment. We used statistical and ML methods to build prognostic models and to compare them to published models. RESULTS: Overall, we analysed 2652 patients, including 699 (26.4%) inpatient procedures. The mean patient age was 57.6 years (SD 16 years), and 48.9% were women. In 1401 (52.8%) patients, the indication was screening or surveillance, and 1035 (39%) patients had a first-time colonoscopy. The rate of inadequate bowel preparation was 16%. Sensitivities, specificities and areas under the curve of predictive models obtained by generalised boosting models, Ranger, support vector machine (radial), CatBoost and Net Regularised Generalised Linear Models were 0.51-0.86, 0.52-0.83 and 0.71-0.74, respectively. They were only marginally superior to a logistic regression model. All models had high negative predictive values >0.9 for inadequate bowel preparation. To detect one patient with inadequate bowel preparation, 8-10 patients need to be evaluated. CONCLUSIONS: Predictive models to identify patients at risk for inadequate bowel preparation obtained by ML showed comparable results compared with a logistic regression model. TRIAL REGISTRATION NUMBER: DRKS00018878.
16 May 2026
Read appraisal →Supportive care in cancer : official journal of the Multinational Association of Supportive Care in Cancer
Effectiveness of comprehensive interventions including exercise and exercise-only interventions on postoperative ileus and recovery of intestinal function in patients with colorectal cancer: a systematic review and meta-analysis
PURPOSE: This systematic review and meta-analysis aimed to evaluate the effectiveness of postoperative or preoperative comprehensive interventions including exercise and exercise-only interventions on postoperative ileus (POI) incidence and the recovery of intestinal function in patients with colorectal cancer (CRC). METHODS: Electronic searches were conducted in 6 databases. Cohort and randomized-controlled trial (RCT) studies investigating the effects of postoperative or preoperative comprehensive interventions including exercise or exercise-only interventions on POI and the recovery of intestinal function in patients with CRC were included. RESULTS: Thirty-eight studies involving 13,124 patients were included. Postoperative comprehensive interventions including exercise significantly shortened the time to first flatus (standardized mean difference [SMD] = -1.38, 95% confidence interval [CI]: -1.87 to -0.89) and time to first defecation (SMD = -1.56, 95% CI: -2.20 to -0.92), but did not reduce POI incidence (odds ratio [OR] = 1.38, 95% CI: 0.43 to 4.47). While, postoperative exercise-only interventions showed no significant effect on intestinal recovery (SMD = -0.31, 95% CI: -1.21 to 0.58). Preoperative comprehensive interventions including exercise shortened the time to first flatus (SMD = -0.41, 95% CI: -0.66 to -0.17). Also, preoperative exercise-only interventions significantly reduced the incidence of POI (OR = 0.56, 95% CI: 0.46 to 0.67). CONCLUSIONS: Postoperative comprehensive interventions effectively promote intestinal recovery; however, postoperative exercise-only interventions may not promote the recovery of intestinal function. On the other hand, preoperative comprehensive interventions or exercise-only interventions may promote the recovery of intestinal function and prevent POI onset; however, the limited number of included studies prevented definitive conclusions.
15 May 2026
Read appraisal →Journal of robotic surgery
Robotic versus laparoscopic low anterior resection for rectal cancer: an updated systematic review and meta-analysis
Robotic surgery has increasingly been adopted for the treatment of rectal cancer. However, most previous meta-analyses included heterogeneous rectal procedures, limiting conclusions specific to low anterior resection. The present study aimed to provide an updated systematic review and meta-analysis comparing robotic versus laparoscopic low anterior resection for rectal cancer. This systematic review and meta-analysis was conducted according to PRISMA guidelines and recommendations from the Cochrane Handbook. PubMed, Scopus, and the Cochrane Library were systematically searched from inception to the most recent date. Comparative studies evaluating robotic low anterior resection (R-LAR) versus laparoscopic low anterior resection (L-LAR) for rectal cancer were included. Risk of bias was assessed using RoB 2 for randomized trials and ROBINS-I for non-randomized studies. Certainty of evidence was evaluated using the GRADE approach. Thirty-three studies including 82,149 patients were analyzed (R-LAR: 56,290; L-LAR: 25,859). R-LAR was associated with a significantly lower conversion rate (OR 0.45, 95% CI 0.40-0.51, p < 0.00001, I²=28%), lower 30-day reoperation rate (OR 0.86, 95% CI 0.77-0.96, p = 0.01, I²=0%), reduced 30-day mortality (OR 0.65, 95% CI 0.52-0.82, p = 0.002, I²=0%), and higher rate of complete total mesorectal excision (OR 2.71, 95% CI 1.38-5.33, p = 0.01, I²=72%). Operative time was significantly longer in the robotic group (MD + 30.58 min, 95% CI 13.71-47.45, p = 0.001, I²=97%). No significant differences were observed for estimated blood loss, overall complications, major complications, anastomotic leakage, postoperative ileus, recovery parameters, length of hospital stay, readmission, circumferential margin positivity, lymph node yield, local recurrence, diverting ileostomy formation, disease-free survival, or overall survival. Robotic LAR may reduce conversion, reoperation, and short-term mortality and may improve completeness of total mesorectal excision, although operative time is longer. These findings suggest potential technical advantages of R-LAR, but high-quality randomized trials are required to determine whether these translate into meaningful clinical benefit.PROSPERO: CRD420261326600.
15 May 2026
Read appraisal →Journal of robotic surgery
Robotic versus laparoscopic intracorporeal anastomosis learning curve: a systematic review
The learning curve for robotic colorectal surgery is increasingly characterised. However, the specific learning curve for intracorporeal anastomosis (IA) remains poorly defined. This review compares robotic and laparoscopic IA with respect to operative time, cases required for proficiency, complication rates, and conversion to extracorporeal anastomosis (EA). A systematic review was conducted in accordance with PRISMA guidelines. The protocol was registered on Open science framework ( https://doi.org/10.17605/OSF.IO/K4MYZ ). On March 18, 2025, OVID and PubMed were searched using the search terms; "laparoscopic," "robotic," "anastomosis," "learning," "education," and "training." Yielding a total of 273 results from PubMed (179) and OVID (94). After screening a total of thirteen studies were included. Thirteen studies met inclusion criteria, including eight comparative analyses and six directly comparing robotic and laparoscopic techniques. Four studies evaluated robotic learning curves without a comparator group, while one was a literature review incorporating institutional experience. Most studies were retrospective (n = 8), with two prospective clinical studies and two experimental simulation-based studies. Hybrid approaches (robotic mobilisation with laparoscopic EA) contributed to heterogeneity (Reitz et al. 2018, https://doi.org/10.1007/s00464-018-6074-7 ). Learning curve assessment was heterogeneous, with most studies using cumulative sum (CUSUM) or risk-adjusted CUSUM (RA-CUSUM) analyses based on operative time or technical performance metrics. Proficiency in robotic IA was reported after approximately 11-29 cases in some series and around 20 cases in others, although larger studies of robotic colorectal surgery demonstrated broader ranges of up to 100 cases. Robotic right hemicolectomy with IA demonstrated a shorter learning curve and reduced anastomosis time compared with laparoscopic IA. Complication rates, conversion rates, and oncological outcomes were comparable between approaches (Gachabayov et al. 2019, Surg Technol Int, 34:163-168). Robotic intracorporeal anastomosis, particularly in right hemicolectomy, appears to have a shorter and less technically demanding learning curve compared with laparoscopic techniques, without compromising safety or oncological outcomes. These findings support a potential role for robotic platforms in facilitating adoption of intracorporeal anastomosis. However, current evidence is limited by retrospective design, small cohorts, and inconsistent definitions of proficiency (Gachabayov et al. 2019, Surg Technol Int, 34:163-168; Van Eetvelde et al. 2022, https://doi.org/10.1007/s11701-022-01514-6 ).
14 May 2026
Read appraisal →Hernia : the journal of hernias and abdominal wall surgery
Obturator hernia in adults: a systematic review of CT diagnosis and surgical management
PURPOSE: Obturator hernia is a rare but life-threatening pelvic floor hernia occurring predominantly in elderly, underweight women and is associated with delayed diagnosis and high perioperative mortality. This systematic review evaluated the diagnostic performance of computed tomography (CT), factors associated with bowel resection, and short-term outcomes of open versus laparoscopic repair in adults. METHODS: A systematic review was conducted per PRISMA 2020. PubMed/MEDLINE, Scopus, and Web of Science were searched from January 2004 to January 2026. Studies reporting at least three surgically confirmed adult cases were eligible. Three consecutive institutional cases were incorporated as an illustrative clinical adjunct. RESULTS: Twenty-two studies including 876 patients were analysed. The mean age across included studies was 79.7 years; 91.5% were female. Reported preoperative CT sensitivity across included studies was 93.2% (range 66.7-100%). Bowel resection was required in 31.6% and was associated with higher 30-day mortality (19.6% vs. 3.9%). Laparoscopic repair appeared to be associated with lower wound morbidity, shorter hospital stay (5.3 vs. 12.6 days), and lower recurrence (0% vs. 2.8%) in selected patient cohorts, although case-mix differences preclude direct comparison. Mesh was used in 44.3% overall. In three institutional cases (all elderly cachectic women presenting emergently), one underwent TAPP and two underwent open midline laparotomy; primary suture repair was performed in all owing to contamination, bowel compromise, or extreme frailty. CONCLUSION: CT remains the diagnostic cornerstone in obturator hernia. Surgical strategy should be individualised based on haemodynamic status, contamination, and frailty. Reported outcomes represent descriptive trends rather than pooled statistical estimates due to heterogeneity of available data.
14 May 2026
Read appraisal →Journal of investigative surgery : the official journal of the Academy of Surgical Research
Vascular and Wound-Healing Implications of Single-Stage Versus Multistage Repair of Anorectal Malformations: A Systematic Review and Meta-Analysis
BACKGROUND: Anorectal malformations (ARMs) are managed using either single-stage or multistage surgical repair. However, concerns persist regarding wound-related complications and long-term functional outcomes, which remain inconsistently reported across surgical approaches. OBJECTIVE: To compare wound-related complications and long-term functional outcomes following single-stage versus multistage repair of ARMs, with emphasis on wound perfusion, dehiscence, and vascular healing dynamics. METHODS: A systematic review and meta-analysis were conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and registered in the International Prospective Register of Systematic Reviews (PROSPERO; CRD420251006245). PubMed and the Cochrane Central Register of Controlled Trials (CENTRAL) were searched for studies published between 2000 and 2024 comparing single-stage and staged ARM repair in neonates and infants. Outcomes included wound dehiscence, surgical site infection (SSI), stenosis, prolapse, redo surgery, constipation, soiling, and voluntary bowel movements. Pooled odds ratios (ORs) with 95% confidence intervals (CIs) were calculated using a random-effects model. Study quality was assessed using the Methodological Index for Non-Randomized Studies (MINORS). RESULTS: Nineteen studies involving 1,972 patients were included. Single-stage repair was associated with a significantly higher risk of wound dehiscence (OR 3.31; 95% CI 1.90-5.79; p < 0.0001), while no significant differences were observed for SSI, stenosis, prolapse, redo surgery, constipation, soiling, or voluntary bowel movements. Moderate to high heterogeneity was noted for selected outcomes. CONCLUSIONS: Single-stage repair provides comparable infection rates and long-term functional outcomes to multistage surgery but carries an increased risk of wound dehiscence. Optimizing tissue perfusion and minimizing tension are essential to improve outcomes. This study highlights the important role of vascular and wound-healing mechanisms in determining outcomes after anorectal malformation (ARM) repair. While single-stage surgery reduces the burden of multiple procedures, it exposes perineal tissues to fecal contamination and potential microvascular stress, which may increase the risk of wound complications.An understanding of tissue perfusion, venous drainage, and local tissue tension is critical during neonatal reconstructive surgery. Incorporating techniques that preserve vascular integrity, minimize tension, and protect the surgical site can help optimize postoperative healing and reduce the risk of wound dehiscence.
13 May 2026
Read appraisal →Lancet (London, England)
Long-term effects of colonoscopy screening on colorectal cancer incidence and mortality: a multicountry, population-based randomised controlled trial
BACKGROUND: We previously reported the 10-year effects of colonoscopy screening on colorectal cancer incidence and mortality. Here, we report the effects after 13 years of follow-up. METHODS: In this multicountry, population-based randomised controlled trial, 84 583 men and women aged 55-64 years at enrolment from Norway, Poland, and Sweden were randomly allocated (1:2) to colonoscopy screening or no screening and analysed. The primary outcomes were colorectal cancer incidence and mortality after 10-15 years of follow-up in intention-to-screen analyses, with first analysis after 10 years, and repeated every other year or at longer intervals. This trial is registered with ClinicalTrials.gov, NCT00883792, and is ongoing. FINDINGS: At 13 years of follow-up, colorectal cancer incidence was 375 colorectal cancers (1·46%) of 28 217 individuals in the screening group and 912 colorectal cancers (1·80%) of 56 366 individuals in the no-screening group. The risk ratio (RR) was 0·81 (95% CI 0·71-0·90) in intention-to-screen analyses and 0·55 (0·33-0·81) in per-protocol analyses. The risk for proximal colorectal cancer was 129 (0·51%) in the screening group versus 283 (0·56%) in the no-screening group (RR 0·91 [0·71-1·09]), and the risk for distal colorectal cancer was 224 (0·87%) in the screening group versus 563 (1·11%) in the no-screening group (RR 0·79 [0·65-0·89]; interaction p<0·0001). In men, the colorectal cancer risk was 214 (1·69%) of 14 154 in the screening group and 541 (2·19%) of 28 247 in the no-screening group (RR 0·77 [0·64 to -0·88]); in women, the risk was 161 (1·24%) of 14 063 in the screening group versus 371 (1·43%) of 28 119 in the no-screening group (RR 0·87 [0·70 to 1·02]; interaction p<0·0001). Colorectal cancer mortality was 106 (0·41%) of 28 217 in the screening group and 236 (0·47%) of 56 366 in the no-screening group (intention-to-screen RR 0·88 [0·68-1·08], per-protocol RR 0·70 [0·26-1·25]). The observed colorectal cancer mortality in the non-screening group (0·47%) was substantially lower than expected at the time of designing the trial (0·82%). INTERPRETATION: One colonoscopy significantly reduced colorectal cancer incidence but not mortality over 13 years. Colorectal cancer mortality was lower in both study groups than when the trial was designed. FUNDING: The Norwegian Research Council, the Nordic Cancer Union, the Norwegian Cancer Society, and the Health Fund of South-East Norway.
13 May 2026
Read appraisal →Journal of clinical oncology : official journal of the American Society of Clinical Oncology
Phase I Study of Mosperafenib, a Novel Paradox Breaker B-Raf Proto-Oncogene Serine/Threonine Kinase (BRAF) Inhibitor, in Patients With BRAF V600-Mutant Solid Tumors
PURPOSE: BRAF V600 mutations are tumor-agnostic oncogenic drivers whose targeted inhibition is challenged by therapeutic resistance. Mosperafenib is a novel paradox breaker and brain-penetrant BRAF inhibitor (BRAFi), tested in this study for safety, maximum tolerated dose (MTD), pharmacokinetics (PK), and preliminary clinical activity. PATIENTS AND METHODS: This phase Ia/b study was conducted in patients with advanced solid tumors harboring a BRAF V600 mutation (ISRCTN13713551). The modified continuous reassessment method guided dose escalation of mosperafenib, which was given orally up to 3,600 mg once daily in 28-day cycles. The primary objective was to estimate the MTD and/or recommended phase II dose(s). RESULTS: Eighty patients (60% BRAFi-exposed)-63 with colorectal cancer (CRC), 13 with melanoma, and 4 with other solid tumors-received ≥1 dose of mosperafenib as a single agent for a median of 3.7 months (range, 0.2-28.6). Two dose-limiting grade 3 toxicities of rash and rash maculopapular were reported. MTD was not reached. Grade 3 to 4 treatment-related adverse events (TRAEs) occurred in 13 patients (16.3%); no grade 5 TRAEs events were reported. Two patients (2.5%) discontinued study treatment due to TRAEs. There were no reports of palmar-plantar erythrodysesthesia or keratoacanthoma. Linear and time-independent PK was demonstrated across the tested dose range, achieving exposure levels with sustained PK-derived pERK inhibition ≥90%. An exposure-response relationship was observed. Overall response rate was 24.2%, including two complete responses and 14 partial responses. Median progression-free survival was 6.4 months in patients with CRC (200 mg once daily-1200 mg three times a day) and 3.5 months in patients with melanoma (200 mg once daily-800 mg twice a day). CONCLUSION: Mosperafenib demonstrated a favorable safety profile, sustained target inhibition, and early signs of clinically meaningful single-agent activity in BRAFi-naïve and BRAFi-exposed patients.
13 May 2026
Read appraisal →Cancer immunology, immunotherapy : CII
Artificial intelligence-powered H&E-based quantification of spatial tumor-infiltrating lymphocyte distribution identifies prognostic immune niches in colorectal cancer
PURPOSE: The prognostic significance of tumor-infiltrating lymphocytes (TILs) in colorectal cancer (CRC) is well established; however, existing approaches inadequately capture their spatial distribution. We investigated the prognostic implications of TIL spatial distribution in CRC using an artificial intelligence (AI)-based method. METHODS: A total of 202 patients with stage II-III CRC were included. TIL densities in intratumoral (iTIL) and stromal (sTIL) regions were quantified using AI-based analysis of hematoxylin and eosin (H&E)-stained images. Based on proximity to the tumor-stromal border (TSB), TILs were subclassified into core iTIL, bounding iTIL, bounding sTIL, and outermost sTIL. Immunoscore was calculated from CD3+ and CD8+ T-cell densities in the tumor center and invasive margin. RESULTS: Correlations between AI-based and pathologist assessments (iTIL: r = 0.57; sTIL: r = 0.70) were comparable to inter-pathologist correlations (iTIL: r = 0.47; sTIL: r = 0.70). In univariate Cox regression analysis, bounding iTIL, bounding sTIL, and outermost sTIL were significantly associated with recurrence-free survival (RFS), whereas core iTIL was not. Composite TIL and TSB scores were developed by incorporating the prognostically significant regions. In multivariable analysis, the TIL score (p = 0.001), TSB score (p < 0.001), and Immunoscore (p < 0.001) independently predicted RFS. In microsatellite instability-high tumors, only the TSB score remained prognostically significant. CONCLUSION: AI-powered spatial analysis of TILs, particularly the TSB score, demonstrated prognostic performance comparable to conventional Immunoscore, thereby supporting the value of spatial immune profiling and AI-driven analysis of H&E-stained slides for improved risk stratification in CRC.
11 May 2026
Read appraisal →Signal transduction and targeted therapy
Total neoadjuvant chemotherapy combined with PD‑1 blockade and IL‑2 in MSS/pMMR locally advanced rectal cancer: short-term results of a prospective, single-arm phase II study
Neoadjuvant therapy has become a cornerstone in the management of locally advanced rectal cancer (LARC). In this single-arm, open-label phase II study, we evaluated the efficacy and safety of total neoadjuvant chemotherapy (TNT) using a CapOX regimen combined with a programmed cell death protein 1 (PD‑1) antibody (sintilimab) and interleukin‑2 (IL‑2) in patients with microsatellite stable (MSS)/defining proficient mismatch repair (pMMR) LARC. A total of 33 patients, aged 18-75 years, with rectal tumors located within 12 cm from the anal verge and staged as cT3/4N_anyM0 or cT_anyN⁺M0, were enrolled. Patients received a regimen consisting of oxaliplatin, sintilimab, capecitabine, and IL‑2 administered in a three‑week cycle, with response evaluations performed after every two cycles. Following six cycles of treatment, 33 patients underwent radical surgery, achieving a 100% R0 resection rate. The pathological complete response (pCR) rate was 42.4% (95% CI: 25.68-59.16%) while the remaining 19 patients (57.6%, 95% CI: 41.07-74.09%) were assessed as having a partial response. The tumor regression grades (TRG) were, TRG1: 3 cases (9.1%, 95% CI: 1.90-25.97%), TRG2: 14 cases (42.4%, 95% CI: 26.27-60.38%), and TRG3: 2 cases (6.1%, 95% CI: 0.73-20.37%), respectively. Surgical safety reported as no cases of grade B/C anastomotic leakage or bowel obstruction. Adverse events (AEs) were manageable, with a 21.2% incidence of grade 3 events and no grade IV/V events or treatment‑related deaths. With a median follow‑up of 25.5 weeks, no recurrences were observed. Analysis of blood and tissue samples from patients with different treatment outcomes revealed significant activation of CD8+ T cells, NK cells, and M1 macrophage subsets in the tumor microenvironment of patients achieving pCR. These compelling data demonstrate promising efficacy with a favorable safety profile in MSS/pMMR LARC. These findings warrant studies to validate this regimen as a novel treatment paradigm for rectal cancer. ClinicalTrials.gov registration: NCT06108596.
10 May 2026
Read appraisal →European journal of cancer (Oxford, England : 1990)
Sex differences in efficacy and toxicity of first-line treatment of metastatic colorectal cancer: An analysis of 18,041 patients in the ARCAD CRC database.
BACKGROUND: Greater toxicity of adjuvant chemotherapy in females has been reported previously, but the impact of patients' sex on chemotherapy (CT)+/-targeted therapies for non-resectable, metastatic colorectal cancer (mCRC) has remained unclear. We compared efficacy and toxicity of first-line chemotherapy, with or without targeted therapies, between male and female mCRC patients. METHODS: Twenty-seven randomized trials from the ARCAD CRC database were included. Primary endpoint was overall survival (OS); secondary endpoints were progression-free survival (PFS), overall response rate, toxicity (according to NCI-CTCAE). Analyses were performed within three treatment groups: A) CT alone, B) CT+bevacizumab, and C) CT+epidermal growth factor receptor inhibitor. OS and PFS were estimated using Kaplan-Meier method. Associations with sex were assessed using study-stratified Cox models adjusted for age, performance status, body mass index, and CT backbone. RESULTS: Baseline data from 18,041 patients were analysed; among 15,271 included in outcome analyses, no sex differences in OS or PFS were observed. Adjusted OS hazard ratios (95% confidence interval) for females versus (vs.) males were: A: 1.01 (0.95-1.08, P = .69), B: 0.96 (0.89-1.04, P = .36), C: 0.97 (0.87-1.10, P = .67). Females experienced higher toxicity, including grade 1-4 and 3-4 nausea (71.8% vs. 60.5% and 7.5% vs. 4.8%), vomiting (47.7% vs. 35.5% and 6.6% vs. 4.4%), and neutropenia (52.4% vs. 44.4% and 29.2% vs. 18.8%) (P < .001). Conversely, rash (grade 1-4 and 3-4) was more frequent in males (29.6% vs. 24.3% and 5.1% vs. 3.7%, P < .001). CONCLUSIONS: Despite higher chemotherapy-related toxicity in females, treatment efficacy was similar between sexes. Rash was more common in males, highlighting that patients' sex modulates toxicity profiles in both directions. Strategies to reduce adverse events in female patients are warranted.
10 May 2026
Read appraisal →Annual review of biomedical engineering
Artificial Intelligence-Based Analysis of Laparoscopic Imaging for Intraoperative Surgical Decision Support
Across surgical specialties, minimally invasive (laparoscopic) surgery has become a standard technique, as it is associated with less trauma, reduced postoperative complication rates, and quicker recovery for patients as compared with open surgery. Due to the limited field of view and limited haptic feedback, the surgical decision-making process in laparoscopic surgery is currently guided solely by surgeons' visual interpretation of the laparoscopic video stream. Modern artificial intelligence (AI) methods excel at the interpretation of visual data and find applications in clinical routine in fields such as radiology and endoscopy. AI methods could help augment laparoscopic surgery through objective real-time analysis of the laparoscopic video stream. Research studies have demonstrated the feasibility of AI-based surgical scene and process understanding. This review provides an overview of these AI applications, focusing on approaches that could, in the next decade, be translated into intraoperative surgical decision support tools for increased surgical quality and patient safety.
9 May 2026
Read appraisal →Journal of investigative surgery : the official journal of the Academy of Surgical Research
Comparison of Functional Outcomes Between Robotic and Laparoscopic Surgery in Rectal Cancer Patients: Systematic Review and Meta-Analysis
BACKGROUND: Robotic and laparoscopic approaches are widely used for rectal cancer surgery. Although robotic systems provide enhanced visualization and dexterity, it remains unclear whether these advantages improve postoperative urinary and sexual function. METHODS: MEDLINE, Embase, Web of Science, and CENTRAL (2000-2025) were searched for studies comparing robotic and laparoscopic rectal cancer surgery. Outcomes included postoperative ileus, urinary retention, and urinary and sexual function. Random-effects meta-analysis was performed using RevMan. Study quality was assessed with MINORS, and certainty of evidence with GRADE. Publication bias was evaluated using funnel plots and Egger's test. RESULTS: Forty-four observational studies (6,121 patients) were included. Robotic surgery was associated with lower urinary retention, though this may reflect publication bias. No significant difference was found in postoperative ileus. Subgroup analyses at 3, 6, and 12 months showed no significant differences in urinary or sexual function. Although pooled results slightly favored robotic surgery, these effects were inconsistent, not sustained, and clinically modest. CONCLUSION: Robotic surgery may reduce urinary retention but shows no consistent functional superiority over laparoscopy. Outcomes at key follow-up points are comparable. As all studies were observational, evidence certainty is low, and findings should be interpreted cautiously. This study provides an updated and comprehensive synthesis of functional outcomes following robotic versus laparoscopic rectal cancer surgery, with a specific focus on clinically relevant time-point analyses (3, 6, and 12 months). Unlike prior meta-analyses, which often emphasize pooled outcomes across heterogeneous follow-up periods, this study demonstrates that no consistent or durable improvements in urinary or sexual function are observed at predefined postoperative intervals. Furthermore, by integrating GRADE-based certainty assessment, this work highlights that the current evidence is predominantly observational and of low to very low certainty. Importantly, the study incorporates contemporary evidence up to 2025, including recent systematic reviews and randomized data, and provides a clinically grounded interpretation by contextualizing effect sizes against minimal clinically important difference (MCID) thresholds. Collectively, this analysis offers a more cautious and clinically meaningful appraisal of functional outcomes, challenging assumptions of functional superiority associated with robotic rectal surgery.
2 May 2026
Read appraisal →Journal of ethnopharmacology
A multicenter, placebo-controlled clinical trial and preliminary experimental study exploring the efficacy of modified Banxia Xiexin Decoction in the treatment of advanced colorectal cancer
ETHNOPHARMACOLOGICAL RELEVANCE: Recent studies in China have increasingly focused on the evidence-based evaluation of malignant tumors, including colorectal cancer (CRC), leveraging the unique properties of herbal medicine. This has led to notable progress in the development of novel therapies. Clinical observations indicate that the modified Banxia Xiexin Decoction (mBXD) exhibits significant anti-cancer effects. However, well-designed clinical trials and foundational research in this field remain insufficient. AIMS OF THE STUDY: This study aims to assess the clinical efficacy and safety of mBXD in the treatment of advanced CRC, alongside preliminary in vivo and in vitro verification of its anti-CRC effects. METHODS: A clinical trial was conducted from October 2020 to October 2024 across 12 hospitals in Southwest China. Eligible participants were randomly assigned to either the mBXD group or the placebo group. Both groups received standard chemotherapy combined with targeted therapy; the mBXD group received mBXD granules (Ready-to-use), while the placebo group received a placebo. Treatment lasted for six months (six cycles). The primary endpoint was progression-free survival (PFS), while secondary endpoints included overall survival (OS), objective response rate (ORR), disease control rate (DCR), and serum tumor markers. Efficacy was assessed every two cycles, with follow-ups conducted every three months after treatment completion until disease progression or death. This study is registered with the Chinese Clinical Trial (No. ChiCTR2100041643). Basic research included the establishment of a subcutaneous CT26 colon cancer model and the use of the HCT116 cell line, with mBXD interventions to monitor tumor growth and evaluate the viability, invasion, and migration of HCT116 cells. RESULTS: A total of 129 participants were enrolled. Compared to the placebo group, the mBXD group showed a 1.9-month and 2.8-month extension in PFS and OS, respectively, with a 17 % reduction in the risk of disease progression and a 15 % reduction in the risk of death (P = 0.341, 0.465). After two, four, and six cycles, the ORR and DCR in the mBXD group were 3.7-12.6 % and 1.2-18.5 % higher than those in the placebo group, respectively, with only the DCR after six cycles reaching statistical significance (P = 0.037). Moreover, serum levels of CEA and CA724 were significantly reduced in the mBXD group after four and six cycles (P ≤ 0.05). No severe adverse events were reported during the trial. Additionally, mBXD significantly inhibited subcutaneous tumor growth after 14 days of treatment (P < 0.05), and its drug-containing serum notably suppressed the viability, migration, and invasion of HCT116 cells. CONCLUSIONS: mBXD demonstrates clinical efficacy in prolonging PFS and OS, stabilizing tumor lesions, and reducing serum tumor marker levels in advanced CRC patients. The treatment is both safe and well-tolerated. Furthermore, mBXD significantly inhibits the growth of subcutaneous CT26 colon cancer and suppresses the viability, invasion, and migration of HCT116 cells.
28 Apr 2026
Read appraisal →Biomedical physics & engineering express
BiFUSE-UNet: a novel dual-encoder enhanced TransUNet for polyp segmentation
Polyps are abnormal protrusions that develop on the mucosal surface of the human body, arising from hyperproliferation or aberrant differentiation of mucosal tissue. While some polyps carry a risk of malignant transformation, timely intervention can halt this progression. Accurately segmenting polyp contours in medical images poses significant challenges, primarily due to the high heterogeneity in polyp appearance and morphology, blurry boundaries with background interference, and class imbalance coupled with data scarcity. To address these challenges, this paper presents BiFuse-UNet, a novel dual-encoder TransUNet architecture built upon a bidirectional fusion mechanism. Building upon the original TransUNet architecture, BiFuse-UNet incorporates several key modules. Firstly, a multi-axis attention module (MaxViT block) is integrated to form a dual-encoder structure, which extracts multi-scale local textural features across different dimensions and captures long-range dependencies through self-attention. Secondly, a Merge fusion module is designed, incorporating a Coordinate Attention module and the SaE block from SENetV2, to enhance the model's perception of local and detailed features. Finally, a Biformer module is introduced to strike an efficient balance between global semantics and local details, thereby improving the segmentation accuracy for subtle polyps and early-stage polyps with atypical morphology. Extensive evaluations on the Kvasir-SEG and CVC-ClinicDB dataset demonstrate that BiFuse-UNet achieves state-of-the-art segmentation performance, outperforming seven existing methods on key metrics such as the Dice coefficient, Recall, Precision, and IoU. The proposed model effectively mitigates the challenges of morphological heterogeneity, blurry boundaries, and background interference, leading to highly accurate pixel-level segmentation.
27 Apr 2026
Read appraisal →PloS one
DSCANet: Integrating dual encoder and spatial cross-attention for polyp segmentation
Colonoscopy is a crucial clinical procedure for detecting colorectal polyps, which are strongly associated with the development of colon cancer. This endoscopic technique plays a vital role in both cancer prevention and early diagnosis. Accurate and efficient polyp segmentation is critical for enhancing the diagnostic reliability and clinical utility of colonoscopy. However, achieving precise segmentation presents significant challenges, primarily due to the diversity of polyps in their size and shape, coupled with poorly defined boundaries between polyps and surrounding tissues. To address these challenges, we propose a novel segmentation network, named DSCANet, which is a dual-branch encoder-structured network designed to efficiently fuse body and edge features for high-precision medical image segmentation. DSCANet integrates four key modules: a dual-branch encoder, a spatial cross-attention (SCA) module, a bipolar fusion (BF) module, and a flexible axis-attention (FAA) module. The dual-branch encoder consists of separate body and edge encoders, which extract respective features independently. The SCA module bridges the semantic gap between the two encoders' features. The BF module fuses the shallowest and deepest features, while the FAA module assists the decoder in extracting semantic information from high-level features. DSCANet achieved superior performance on multiple colorectal polyp segmentation datasets. The code is available at https://github.com/Shantdst/DSCANet.
23 Apr 2026
Read appraisal →JAMA
Tumor Debulking in Combination With Chemotherapy in Multiorgan Metastatic Colorectal Cancer: The ORCHESTRA Randomized Clinical Trial
IMPORTANCE: Local therapy, including surgery, radiation, and ablation, is increasingly used in patients with multiorgan metastatic colorectal cancer (mCRC). However, prospective evidence for a survival benefit of tumor debulking is lacking. OBJECTIVE: To investigate whether tumor debulking added to palliative chemotherapy improves survival of patients with multiorgan mCRC. DESIGN, SETTING, AND PARTICIPANTS: This investigator-initiated, open-label, multicenter, randomized clinical trial enrolled patients with multiorgan mCRC between May 2013 and May 2023. The last date of follow-up was April 4, 2024. Patients were enrolled in 27 hospitals in the Netherlands and 1 in the UK. Adult patients with multiorgan mCRC were considered eligible if more than 80% tumor debulking was deemed feasible by resection, radiotherapy, and/or thermal ablation prior to starting first-line palliative chemotherapy. INTERVENTIONS: After achieving objective tumor response or stable disease after 3 cycles of capecitabine and oxaliplatin or 4 cycles of 5-fluorouracil and oxaliplatin with or without bevacizumab, patients were randomized 1:1 to receive chemotherapy alone (standard care group) or tumor debulking followed by chemotherapy. MAIN OUTCOMES AND MEASURES: The primary end point was overall survival. Secondary end points included progression-free survival and serious adverse events. These outcomes were analyzed in the intention-to-treat population, applicable from randomization. A prespecified interim analysis performed after the initial 100 participants were enrolled revealed that the trial was both safe and feasible to proceed. RESULTS: A total of 382 of 454 enrolled patients were randomized: 192 in the chemotherapy alone group (133 [69%] male) and 190 in the chemotherapy plus tumor debulking group (127 [67%] male). The median age was 64 years in both groups. After a median follow-up of 32.3 months, median overall survival in the chemotherapy alone group was 27.5 months vs 30.0 months in the chemotherapy plus tumor debulking group (adjusted hazard ratio, 0.88 [95% CI, 0.70-1.10]; P = .26). Median progression-free survival in the chemotherapy alone group was 10.4 months vs 10.5 months in the chemotherapy plus tumor debulking group (adjusted hazard ratio, 0.83 [95% CI, 0.67-1.02]; P = .08). More patients in the chemotherapy plus tumor debulking vs chemotherapy alone group had any serious adverse events (101 [53%] vs 74 [39%]; P = .006). CONCLUSIONS AND RELEVANCE: Tumor debulking in addition to first-line palliative systemic treatment failed to improve overall survival compared with systemic treatment alone for patients with multiorgan mCRC and should not be considered standard care. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT01792934.
23 Apr 2026
Read appraisal →Spinal cord series and cases
Early transanal irrigation in spinal cord injury-related neurogenic bowel dysfunction: preliminary insights.
STUDY DESIGN: Prospective, randomized, open-label, multicenter, parallel group study. OBJECTIVES: Neurogenic bowel dysfunction (NBD) is a common complication in spinal cord injury (SCI) patients, severely affecting quality of life. While transanal irrigation (TAI) has shown superior efficacy over conservative treatments (CT), its early adoption during the first hospitalization post-injury remains unexplored. This study aimed to compare early TAI initiation versus CT in managing NBD in SCI patients. SETTING: This study involved four spinal units across Italy. METHODS: SCI patients were randomized to receive TAI or CT during a four-week hospital stay, followed by continued treatment at home for 6 months. Bowel function, constipation, and continence were assessed using the Wexner scales, while quality of life was measured using the EQ-5D-5L and Spinal Cord Independence Measure (SCIM) questionnaires. Stool transit time and patient independence were also monitored. RESULTS: Twenty-two patients were enrolled, with the TAI group showing significant improvements in constipation (Wexner score reduction from 13.43 ± 4.24 at baseline to 7.55 ± 2.38 and 8.80 ± 4.16 at 4 weeks and 6 months, p = 0.02) and a trend toward continence improvement. A significant increase in the total SCIM score was observed in the TAI group (p = 0.02), particularly in respiratory and sphincter management. Quality of life improved in both groups, with a notable enhancement in self-care activities for the TAI group (p = 0.006). CONCLUSIONS: These preliminary results suggest early TAI adoption improves bowel management, especially constipation, and enhances patient independence. Larger studies are needed to validate these findings and assess long-term benefits.
22 Apr 2026
Read appraisal →Pediatric surgery international
Anorectal malformation in adulthood: a systematic review of biological, psychological, and sociological outcomes and experiences
PURPOSE: Anorectal malformations are congenital anomalies involving the rectum, anus, genital, and urinary tracts, and occur in approximately one in 5000 live births. A lack of transition from pediatric to adult health services means little is known about how this condition is experienced in adulthood. The purpose of this review was to discover biopsychosocial challenges and supports experienced by adults living with anorectal malformation. MATERIALS AND METHODS: A systematic quantitative literature review was the preferred method to collect and analyze data. Articles included in the review were sorted into purpose-built datasets, with data coded into biopsychosocial themes and patterns. RESULTS: The most frequently reported biological challenge was surgical intervention; the most frequently reported support was positively framed continence management. The most common psychological challenges were psychosexual concerns such as sexual anxiety, and the most common supports were interventions focused on improving self-perception. The most reported sociological challenge was navigating complex health systems, and the most reported support was forming meaningful relationships. CONCLUSION: Biopsychosocial outcomes experienced by adults born with anorectal malformation remain largely unknown. Social work and psychology can provide therapeutic interventions that enhance emotional wellbeing, psychosocial functioning, and patient-led care.
17 Apr 2026
Read appraisal →Signal transduction and targeted therapy
STRATEGIC-1: multiple-line, randomized, open-label GERCOR-PRODIGE-39 phase III trial in unresectable RAS/BRAF wild-type metastatic colorectal cancer
Managing unresectable metastatic colorectal cancer (mCRC) requires a comprehensive strategy. While chemotherapy, anti-angiogenic, and anti-epidermal growth factor receptor (EGFR) agents are available, strategy trials are needed to optimize their use and sequencing. The STRATEGIC-1 phase III trial (NCT01910610) was designed to determine the optimal treatment sequence in patients with untreated, unresectable wild-type RAS/BRAFV600E mCRC. Patients were randomized (1:1) to FOLFIRI-cetuximab then mFOLFOX6-bevacizumab (arm A) or OPTIMOX-bevacizumab then FOLFIRI-bevacizumab followed by EGFR monoclonal antibody +/- irinotecan (arm B). The primary endpoint was the duration of disease control (DDC). Secondary endpoints were overall survival (OS), time to failure of strategy (TFS), progression-free survival (PFS), overall response rate (ORR), salvage surgery rate, safety, and health-related quality of life (HRQoL). Overall, 263 patients (arm A:131, arm B:132) were randomized. After 68.4 months of median follow-up (95% CI, 76.5-98.0), the median DDC was 22.8 months (95% CI, 20.4-28.8) in arm A and 23.5 months (95% CI, 17.9-26.3) in arm B (HR = 1.01, 95% CI, 0.76-1.34; log-rank P = 0.945). The median OS was 40.4 months (95% CI, 32.4-51.1) in arm A and 34.4 months (95% CI, 27.5-42.2) in arm B (HR = 1.30, 95% CI, 0.99-1.72). The ORR was higher in arm A (82.4% versus 65.4%) in the first-line group but not in the second-line group (20.7% versus 16.4%). Adverse events were consistent with the well-known safety profiles. STRATEGIC-1 did not meet its primary endpoint and was inconclusive in identifying the optimal treatment strategy in wild-type RAS/BRAFV600E mCRC.
16 Apr 2026
Read appraisal →Computers in biology and medicine
MIPHEI-ViT: Multiplex immunofluorescence prediction from H&E images using ViT foundation models
Histopathological analysis is a cornerstone of cancer diagnosis, with Hematoxylin and Eosin (H&E) staining routinely acquired for every patient to visualize cell morphology and tissue architecture. On the other hand, multiplex immunofluorescence (mIF) enables more precise cell type identification via proteomic markers, but has yet to achieve widespread clinical adoption due to cost and logistical constraints. To bridge this gap, we introduce MIPHEI (Multiplex Immunofluorescence Prediction from H&E Images), a U-Net-inspired architecture that leverages a ViT pathology foundation model as an encoder to predict mIF signals from H&E images using rich pretrained representations. MIPHEI targets a comprehensive panel of markers spanning nuclear content, immune lineages (T cells, B cells, myeloid), epithelium, stroma, vasculature, and proliferation. We train our model using the publicly available OrionCRC dataset of restained H&E and mIF images from colorectal cancer tissue, and validate it on five independent datasets: HEMIT, PathoCell, IMMUcan, Lizard and PanNuke. On OrionCRC test set, MIPHEI achieves accurate cell-type classification from H&E alone, with F1 scores of 0.93 for Pan-CK, 0.83 for α-SMA, 0.68 for CD3e, 0.36 for CD20, and 0.28 for CD68, substantially outperforming both a state-of-the-art baseline and a random classifier for most markers. Our results indicate that, for some molecular markers, our model captures the complex relationships between nuclear morphologies in their tissue context, as visible in H&E images and molecular markers defining specific cell types. MIPHEI offers a promising step toward enabling cell-type-aware analysis of large-scale H&E datasets, in view of uncovering relationships between spatial cellular organization and patient outcomes.
15 Apr 2026
Read appraisal →Biomedical physics & engineering express
A multi-task cross-attention strategy to segment and classify polyps
Polyps are the main biomarkers for diagnosing colorectal cancer. Their early detection and accurate characterization during colonoscopy procedures rely on expert observations. Nevertheless, such a task is prone to errors, particularly in morphological characterization. This work proposes a multi-task representation capable of segmenting polyps and stratifying their malignancy from individual colonoscopy frames. The approach employs a deep representation based on multi-head cross-attention, refined with morphological characterization learned from independent maps according to the degree of polyp malignancy. The proposed method was validated on the BKAI-IGH dataset, comprising 1200 samples (1000 white-light imaging and 200 NICE samples) with fine-grained segmentation masks. The results show an average IoU of 83.5% and a recall of 94%. Additionally, external dataset validation demonstrated the model's generalization capability. Inspired by conventional expert characterization, the proposed method integrates textural and morphological observations, allowing both tasks, polyp segmentation and the corresponding malignancy stratification. The proposed strategy achieves the state-of-the-art performance in public datasets, showing promising results and demonstrating its ability to generate a polyp representation suitable for multiple tasks.
14 Apr 2026
Read appraisal →Signal transduction and targeted therapy
Irinotecan with trifluridine/tipiracil and bevacizumab for second-line metastatic colorectal cancer: a phase II multicenter study
Patients diagnosed with metastatic colorectal cancer (mCRC) face a constrained therapeutic landscape following the failure of initial treatment. This multicenter, single-arm, phase II trial (NCT06202001) aimed to assess the efficacy and safety of a novel second-line regimen comprising trifluridine/tipiracil (TAS-102), irinotecan, and bevacizumab. Patients with mCRC resistant to prior fluoropyrimidine and oxaliplatin-based chemotherapy were enrolled. Based on a preceding phase I trial, patients received biweekly cycles of oral TAS-102 (30 mg/m² twice daily, days 1-5), intravenous irinotecan (150 mg/m²), and intravenous bevacizumab (5 mg/kg). The primary outcome measure was the objective response rate (ORR). From October 2023 to August 2024, 60 patients were enrolled. As of December 2024, the ORR was 18.3% (2 complete and 9 partial responses), and the disease control rate (DCR) was 83.3%. The median progression-free survival (PFS) was 6.6 months (95% CI, 4.39-8.81), and the median overall survival (OS) was 17.3 months (95% CI, 13.55-21.05). Subgroup analyses indicated that prior resection of the primary tumor was associated with significantly longer median OS (21.9 vs. 16.2 months; p = 0.048) and PFS (8.9 vs. 5.2 months; p = 0.004). The most frequently reported treatment-related adverse events (TRAEs) were nausea (100%), neutropenia (86.7%), and anemia (83.3%). The predominant grade 3/4 TRAEs included neutropenia (48.3%), febrile neutropenia (8.3%), and diarrhea (6.7%). In conclusion, the combination of irinotecan, TAS-102, and bevacizumab shows encouraging efficacy and a manageable safety profile as a second-line therapy for mCRC, meriting further investigation.
11 Apr 2026
Read appraisal →Annals of medicine
Comparison of VCV and PCV-VG modes on diaphragmatic function in diabetic patients undergoing laparoscopic colorectal surgery: a prospective randomized controlled study
BACKGROUND: Diabetic patients are prone to induce diaphragmatic weakness, which can lead to postoperative pulmonary complications (PPCs). The optimal mechanical ventilation mode may potentially improve postoperative diaphragmatic function. This study evaluates the effects of two ventilation modes under driving pressure-guided ventilation strategy on diaphragmatic function, as assessed by diaphragm thickening fraction (DTF) and diaphragm excursion (DE), in diabetic patients following laparoscopic colorectal surgery. METHODS: Eighty patients diagnosed with Type II diabetes scheduled for elective laparoscopic colorectal surgery, were randomly allocated to either the pressure-controlled volume-guaranteed ventilation (PCV-VG) group (Group P) or the volume-controlled ventilation (VCV) group (Group V) during surgery. The primary outcome was diaphragmatic function assessed during both tidal breathing and maximal inspiratory effort after surgery. Secondary outcomes included intraoperative mechanical power, PPCs, and other complications. RESULTS: A total of eighty patients were included in the final analysis. The averaged area under the curve (AUC) for mechanical power during ventilation was significantly lower in Group P than in Group V (p = 0.002). PCV-VG significantly improved both DE and DTF within the first two days post-surgery (AUCDEtidal: p = 0.088, AUCDTFtidal: p = 0.004, AUCDEmax: p = 0.029, AUCDTFmax: p = 0.017). Postoperative diaphragmatic weakness was less frequent in Group P than in Group V (p = 0.019). However, there was no difference in the incidence of PPCs between the two groups (p = 0.155). CONCLUSION: PCV-VG mode can reduce intraoperative mechanical power, better preserve postoperative diaphragmatic function. However, these improvements did not translate into clinical benefits, as evidenced by the lack of reduction in the incidence of PPCs.
10 Apr 2026
Read appraisal →Journal of clinical oncology : official journal of the American Society of Clinical Oncology
Neoadjuvant Chemotherapy With CAPOX Versus Chemoradiation for Locally Advanced Rectal Cancer With Uninvolved Mesorectal Fascia (CONVERT): Final Results of a Phase III Trial
PURPOSE: The neoadjuvant chemoradiotherapy (nCRT) might accentuate surgical complications and toxicity in the treatment of locally advanced rectal cancer (LARC) while neoadjuvant chemotherapy (nCT) alone shows promise as an alternative treatment. However, which patients deserve most from the nCT need further clarify. This trial aimed to assess the non-inferiority of nCT with capecitabine plus oxaliplatin (CAPOX) versus nCRT with capecitabine in LARC with uninvolved mesorectal fascia (MRF). METHODS: Patients with LARC within 12 cm from the anal verge and uninvolved MRF were randomly assigned to receive 4 cycles of CAPOX chemotherapy alone (nCT group) or CRT with concurrent Capecitabine (nCRT group). The primary end point is 3-year locoregional recurrence-free survival (LRRFS). Secondary end points, such as 3-year disease-free survival (DFS), 3-year overall survival (OS), and adverse events (AEs), were also reported. RESULTS: A total of 663 patients were enrolled and 589 patients received the allocated treatment (nCT, n = 300; nCRT, n = 289). LRRFS was analyzed with a median follow-up of 48 months. 3-year LRRFS was 97.4% (95% CI, 95.5 to 99.3) in the nCRT group and 96.3% (95% CI, 94.0 to 98.6) in the nCT group, resulting in a hazard ratio (HR) of 1.40 (95% CI, 0.53 to 3.68). The nCT and nCRT achieved similar 3-year DFS (89.2% v 87.9%; HR, 0.88 [95% CI, 0.54 to 1.44]) and 3-year OS (95.0% v 94.1%; HR, 0.86 [95% CI, 0.42 to 1.76]). The nCT group showed a lower incidence of grade 2 to 4 long-term AEs (16.0% v 26.3%, P = 0.002) and proctitis (33.6% v 41.7%, P = 0.049) compared with nCRT group. CONCLUSIONS: The non-inferiority of nCT was not confirmed with a very low incidence of local recurrence in both group. But nCT offers comparable DFS and OS while mitigating the burden of toxicity as compared to nCRT. These insights shed light on a potential paradigm shift in the treatment for LARC with uninvolved MRF.
10 Apr 2026
Read appraisal →Journal of Korean medical science
Effectiveness and Cost-Utility of Anti-Adhesion Agents in Intra-Abdominal Surgery Patients
BACKGROUND: Adhesive intestinal obstruction (AIO) is a common and critical complication of intra-abdominal surgery, often requiring preventative measures such as using anti-adhesion agents (AAA). However, despite their prevalence and significance, a notable lack of long-term and large-scale clinical evidence exists supporting the effectiveness and cost-utility of these agents. This study aimed to evaluate the effectiveness and cost-utility of applying AAA in patients undergoing intra-abdominal surgery. METHODS: This is a retrospective observational study utilizing the Korean National Health Insurance Database in combination with the data from the Clinical Data Warehouse collected from six university-affiliated hospitals in Korea. The study included patients who underwent intra-abdominal surgeries between April and December 2016, with 2-year follow-up. The effectiveness of AAA was assessed by the incidence rate of AIO, and cost-utility was evaluated through calculating quality-adjusted life years (QALYs) and direct costs. Subgroup and sensitivity analyses were conducted to confirm the consistency and robustness of the study findings. RESULTS: A total of 2,831 patients were included in the study. Among them, 1,365 (47.2%) received AAA while 1,466 (52.8%) did not. The group that received AAA exhibited a significantly lower incidence of AIO, with a rate of 12.4% compared to 18.5% in the non-user group. Utility scores were higher in the user group, averaging 1.992 QALYs versus 1.986 in the non-user group. Direct costs were significantly lower in the user group, totaling 3,794 USD, which is 1,330 USD less than those incurred by the non-user group. Subgroup and sensitivity analyses reinforced these findings, demonstrating consistent reductions in AIO incidence, increased QALYs, and decreased costs observed across various subgroups, particularly among older patients and those without chronic conditions. CONCLUSION: These findings highlight the clinical effectiveness and cost-utility of AAA in patients undergoing intra-abdominal surgery. By significantly reducing the incidence of AIO and lowering healthcare costs, these agents provide empirical evidence supporting their broader adoption in clinical practice. This study addresses critical gaps in the understanding of postoperative management and offers valuable insights that could inform future healthcare policies.
8 Apr 2026
Read appraisal →Medicine
Post-marketing signal detection of fruquintinib-associated adverse events using the FAERS database
Fruquintinib, a targeted therapeutic agent approved in 2018 for the treatment of metastatic colorectal cancer, is gradually being applied in an expanding range of clinical settings. Given the complexities associated with real-world dosing, a comprehensive evaluation of its safety profile is essential for optimizing clinical decision-making. We conducted a retrospective pharmacovigilance study utilizing a spontaneous reporting system. Reports related to Fruquintinib were extracted through a standardized data cleaning process and cross-validated using 3 complementary signal detection algorithms: proportional reporting ratio, Bayesian confidence propagation neural network, and reporting odds ratio (ROR). This approach was employed to systematically assess the drug's safety and identify potential adverse event signals. A total of 92 statistically significant safety signals were identified from 1188 independent cases included in the analysis, corresponding to 1836 adverse event reports. The signal distribution exhibited organ system specificity, with gastrointestinal reactions being the most prevalent category. These reactions primarily manifested as typical drug-related toxicities, including diarrhea and nausea. Other notable complications included neurological, respiratory, dermatological, renal, and cardiovascular events. Important aes not mentioned in the instructions, such as bone marrow suppression (ROR = 11.17), peripheral neuropathy (ROR = 4.24) and dehydration (ROR = 3.98), were also discovered. This study systematically characterizes the post-marketing safety profile of Fruquintinib through a multi-algorithm synergistic analysis, confirming that its safety profile aligns with the clinically expected outcomes based on its mechanism of action. Future research should focus on analyzing drug interaction mechanisms and exploring biomarkers to develop a precise risk-benefit assessment model.
5 Apr 2026
Read appraisal →Medicine
Neuroendocrine neoplasms diagnosed in the setting of inflammatory bowel disease: A systematic review and proposed management algorithm
BACKGROUND: Patients with inflammatory bowel disease (IBD) are at higher risk of developing malignancies, including neuroendocrine neoplasms, but the specifics of that are not well known. We reviewed the existing literature to map the evidence surrounding this phenomenon. METHODS: Following PRISMA 2020 guidelines, we searched PubMed from inception to September 2024 for patients with gastrointestinal neuroendocrine neoplasms (NENs) and either Crohn disease (CD) or ulcerative colitis (UC). Eligible studies included case reports, case series, and larger retrospective IBD cohorts. We extracted demographic and clinical data including tumor grade, location, metastatic status, and mode of detection. Comparative analyses between UC and CD were performed using Fisher exact test, chi-square test, and independent t-tests. RESULTS: We identified 67 case studies with data available for a total of 108 patients (61.1% male), ranging from 14 to 87 years of age. The median duration from IBD diagnosis to NEN detection was 13 years. Fifty-five (50.9%) patients had CD and 53 (49.1%) had UC. UC patients were older at NEN diagnosis (mean 49 vs 40 years, P = .004) and had a higher metastatic rate (39.6% vs 18.2%, P = .019). Most NENs were low-grade (69.4%), while 20.4% were high-grade and 8.3% were mixed. The most common tumor locations were the rectum (29.6%) and appendix (27.8%), with rectal and colonic NENs more prevalent in UC and small bowel or appendiceal NENs more common in CD. At diagnosis, 71.3% of patients had localized disease. For high-grade NENs, 86.4% of cases were metastatic, compared to 9.3% for low-grade tumors. Only 10.2% of NENs were detected via endoscopic surveillance. CONCLUSIONS: NENs in IBD show distinct patterns by IBD type, with significant differences in tumor location, grade, and metastatic rates between UC and CD. These hypothesis-generating observations suggest heightened vigilance after approximately 10 years of IBD or around age 44, focusing on rectal lesions in UC and small bowel/appendiceal lesions in CD. Validation in prospective, population-based studies is needed before these findings can inform clinical practice.
4 Apr 2026
Read appraisal →International journal of medical informatics
Automated extraction of fluoropyrimidine treatment and treatment-related toxicities from clinical notes using natural language processing.
OBJECTIVE: Fluoropyrimidines are widely prescribed for colorectal and breast cancers, but are associated with toxicities such as hand-foot syndrome and cardiotoxicity. Since toxicity documentation is often embedded in clinical notes, we aimed to develop and evaluate natural language processing (NLP) methods to extract treatment and toxicity information. MATERIALS AND METHODS: We constructed a gold-standard dataset of 236 clinical notes from 204,165 adult oncology patients. Domain experts annotated categories related to treatment regimens and toxicities. We developed rule-based, machine learning-based (Random Forest [RF], Support Vector Machine [SVM], Logistic Regression [LR]), deep learning-based (BERT, ClinicalBERT), and large language models (LLM)-based NLP approaches (zero-shot and error analysis prompting). A 5-fold cross validation were conducted to validate each model. RESULTS: Error analysis prompting achieved optimal precision, recall, and F1 scores for treatment (F1 = 1.000) and toxicities extraction (F1 = 0.965), whereas zero-shot perform moderately (treatment F1 = 0.889, toxicities extraction F1 = 0.854) Rule-based reached F1 = 1.000 for treatment and F1 = 0.904 for toxicities extraction. LR and SVM ranked second and fourth for toxicities extraction (LR F1 = 0.914, SVM F1 = 0.903). Deep learning and RF underperformed, with performance of BERT reached F1 = 0.792 for treatment and F1 = 0.837 for toxicities extraction.,ClinicalBERT reached F1 = 0.797 for treatment and F1 = 0.884 for toxicities extraction). RF reached F1 = 0.745 for treatment and F1 = 0.853 for toxicities extraction. DISCUSSION: LMM-based error analysis outperformed all others, followed by machine learning methods. Machine learning and deep learning methods were limited by small training data and showed limited generalizability, particularly for rare categories. CONCLUSION: LLM-based error analysis most effectively extracted fluoropyrimidine treatment and toxicity information from clinical notes, and has strong potential to support oncology research and pharmacovigilance.
4 Apr 2026
Read appraisal →Yonsei medical journal
Deep Learning for Automated Tumor Segmentation of Rectal Cancer on T2-Weighted Magnetic Resonance Images
PURPOSE: To develop a deep learning (DL)-based automated segmentation model for rectal cancer on T2-weighted (T2W) magnetic resonance (MR) images. MATERIALS AND METHODS: A total of 458 patients who underwent baseline rectal MR imaging were retrospectively enrolled. An experienced radiologist labeled the tumor on each slice of the selected T2W axial images covering the entire tumor mass, and tumor volume was measured during labeling. Another radiologist labeled the rectum on the same images. Attention U-Net was trained on the T2W images from the training dataset to classify each voxel as tumor or non-tumor. Segmentation with or without rectum guidance was compared using evaluation metrics, including the Dice similarity coefficient (DSC), in the test dataset. RESULTS: The tumor segmentation DSC without rectum guidance was 73.35% (71.97-74.71); that with the predicted rectum was 71.79% (70.53-73.05); and that with the overlaid rectum was 75.52% (74.32-76.62), which was the highest among the three segmentation models. Tumor volume showed a positive correlation with tumor segmentation accuracy with the predicted rectum (r=0.333, p=0.001; overlaid rectum, r=0.319, p=0.002) and without rectum guidance (r=0.219, p=0.036). CONCLUSION: A DL-based automated segmentation model can predict RC with over 70% accuracy, which was further improved with overlaid rectum, and larger tumor volume.
4 Apr 2026
Read appraisal →Computer methods and programs in biomedicine
Application of artificial intelligence in colonoscopy imaging for polyp analysis-A systematic review
The colon is a major component of the digestive system, so early detection of colorectal polyps is essential in preventing colorectal cancer, which is a leading cause of cancer-related death worldwide. While colonoscopy remains the gold standard for polyp detection, its diagnostic accuracy is highly operator-dependent. Recent advances in Deep Learning (DL), a branch of Artificial Intelligence (AI), have shown substantial potential to improve colonoscopy image analysis by enhancing the accuracy, consistency, and objectivity of polyp detection, segmentation, and classification. Artificial intelligence-based systems have significantly reduced inter-observer variability and increased diagnostic efficiency, ultimately transforming the landscape of colorectal lesion assessment. This survey provides a comprehensive and critical analysis of the current status of deep learning applications in colorectal polyp analysis. We systematically review state-of-the-art methodologies across various DL architectures-including Convolutional Neural Networks (CNNs), transformer-based models, and hybrid approaches-and examine their performance on publicly available benchmark datasets. Additionally, we highlight the strengths and limitations of existing techniques, explore the clinical relevance of AI-assisted tools, and identify prevailing challenges such as data imbalance, real-time deployment, and generalizability across diverse populations and colonoscopy devices. By consolidating key advances and outlining future research directions, this review aims to serve as a valuable resource for researchers, clinicians, and developers seeking to leverage deep learning to enhance colorectal polyp detection, diagnosis, and clinical decision-making.
4 Apr 2026
Read appraisal →The Journal of pathology
Integrating artificial intelligence (AI) into colorectal cancer reporting
Artificial intelligence (AI) and deep learning (DL) are transforming cancer research and clinical care, with histopathology playing a central role in this transformation. In colorectal cancer (CRC), the second leading cause of cancer mortality world-wide, multimodal and vision-language models (VLMs) hold particular promise for enhancing the standardisation of histopathology reporting, the understanding of disease biology, and the discovery of novel prognostic indicators. Despite the availability of guidelines and reporting templates for essential prognostic indicators, variability remains in how key features such as TNM staging or tumour deposits are assessed and reported in routine clinical practice. AI-based tools have the potential to support refined extraction of established and extended features directly from whole-slide images. In parallel, recent studies have shown that DL models applied to pathology slides and associated AI-based biomarkers can outperform traditional histopathological prognostic indicators and uncover novel parameters, including tumour-adipocyte interactions, tumour-stroma ratio, and immune cell patterns at the invasive margin. Here, we review recent advances in both domains: AI-assisted standardisation of CRC pathology reporting and AI-driven identification of novel prognostic biomarkers. We highlight the need to refine and standardise CRC reporting practices and propose that a harmonised approach combining established pathology features with AI-derived prognostic indicators could refine risk assessment and improve outcomes for CRC patients. © 2026 The Author(s). The Journal of Pathology published by John Wiley & Sons Ltd on behalf of The Pathological Society of Great Britain and Ireland.
4 Apr 2026
Read appraisal →J Clin Oncol
Foundation Model for Predicting Prognosis and Adjuvant Therapy Benefit From Digital Pathology in GI Cancers
PURPOSE: Artificial intelligence (AI) holds significant promise for improving cancer diagnosis and treatment. Here, we present a foundation AI model for prognosis prediction on the basis of standard hematoxylin and eosin-stained histopathology slides. METHODS: In this multinational cohort study, we developed AI models to predict prognosis from histopathology images of patients with GI cancers. First, we trained a foundation model using over 130 million patches from 104,876 whole-slide images on the basis of self-supervised learning. Second, we fine-tuned deep learning models for predicting survival outcomes and validated them across seven cohorts, including 1,619 patients with gastric and esophageal cancers and 2,594 patients with colorectal cancer. We further assessed the model for predicting survival benefit from adjuvant chemotherapy. RESULTS: The AI models predicted disease-free survival and disease-specific survival with a concordance index of 0.726-0.797 for gastric cancer and 0.714-0.757 for colorectal cancer in the validation cohorts. The models stratified patients into high-risk and low-risk groups, with 5-year survival rates of 49%-52% versus 76%-92% in gastric cancer and 43%-72% versus 81%-98% in colorectal cancer. In multivariable analysis, the AI risk scores remained an independent prognostic factor after adjusting for clinicopathologic variables. Compared with stage alone, an integrated model consisting of stage and image information improved prognosis prediction across all validation cohorts. Finally, adjuvant chemotherapy was associated with improved survival in the high-risk group but not in the low-risk group (treatment-model interaction P = .01 and .006) for stage II/III gastric and colorectal cancer, respectively. CONCLUSION: The pathology foundation model can accurately predict survival outcomes and complement clinicopathologic factors in GI cancers. Pending prospective validation, it may be used to improve risk stratification and inform personalized adjuvant therapy.
1 Nov 2025
Read appraisal →