Endoscopic mucosal resection (EMR) is widely used for large and complex colorectal lesions and offers a less invasive alternative to surgery. Although adenoma recurrence after EMR has been described in procedural cohorts, real-world estimates of subsequent colorectal cancer (CRC) diagnoses after EMR across multicenter electronic health record (EHR) systems remain limited. We conducted a retrospective cohort study using the TriNetX US Collaborative Network. Patients aged 40 years or older who underwent colonoscopy or sigmoidoscopy with EMR between January 1, 2016 and January 1, 2017 were included. Patients with Crohn’s disease, ulcerative colitis, familial adenomatous polyposis, or CRC recorded on or before the index EMR date were excluded. The primary outcome was incident CRC, defined by ICD-10-CM codes C18, C19, and C20, assessed from January 1, 2018 through December 31, 2021 after a one-year washout period. Incidence rates are reported per 1,000 person-years with exact Poisson 95
To synthesise evidence informing risk-adapted treatment and organ preservation in locally advanced rectal cancer. This narrative review used targeted English-language searches of PubMed, Embase, and the Cochrane Library through 31 March 2026, supplemented by targeted source verification in September 2026. Magnetic resonance imaging, mismatch repair status, recurrence risk, and patient preferences inform treatment selection. Total neoadjuvant therapy is established for higher-risk mismatch repair-proficient/microsatellite-stable disease. Chemoradiation followed by consolidation chemotherapy can improve organ-preservation probability compared with induction sequencing, without a demonstrated disease-free survival advantage. Selected lower-risk surgical candidates may receive chemotherapy with selective chemoradiation. Both short-course radiotherapy and long-course chemoradiation remain evidence-based options; the long-term locoregional failure signal in RAPIDO cannot be attributed to fractionation alone. Watch-and-wait requires multimodal confirmation of clinical complete response, intensive surveillance, and access to salvage surgery; oncological equivalence to immediate resection is unproven. Prospective dostarlimab data support non-operative management in selected mismatch repair-deficient rectal cancers, while long-term durability requires further evaluation. Management should integrate tumour anatomy, biology, treatment response, and patient priorities. Organ preservation requires careful selection and structured follow-up, with treatment intensity individualised rather than applied uniformly.
To assess associations of AJCC tumor regression grade (TRG) with recurrence, disease-free survival (DFS), and overall survival (OS) after neoadjuvant therapy and surgery for rectal cancer. This retrospective cohort included 216 patients operated on during 2014–2023. Favorable regression comprised TRG 0–1. The primary Cox model adjusted for pretreatment carcinoembryonic antigen, Charlson Comorbidity Index, and neoadjuvant chemotherapy class; a sensitivity model added ypTNM stage. Time-dependent adjuvant-treatment, competing-risk, proportional-hazards, and incremental-value analyses were performed. Favorable and poor regression occurred in 106 and 110 patients. There were 49 recurrences, 61 DFS events, and 26 deaths. Crude recurrence was lower with favorable regression (16.0
Postoperative circulating tumor DNA (ctDNA) is emerging as a marker of molecular residual disease (MRD) after curative-intent treatment for esophageal squamous cell carcinoma (ESCC). This review evaluates its ability to refine recurrence-risk assessment beyond pathological response and its potential role in guiding adjuvant therapy and surveillance. A structured narrative review of studies evaluating perioperative and postoperative ctDNA in resectable ESCC was conducted using PubMed/MEDLINE through August 17, 2026, supplemented by citation tracking and clinical trial registry searches. Evidence was synthesized across treatment pathways, assay platforms, and postoperative management settings. Across ESCC cohorts using digital PCR, fixed next-generation sequencing panels, and personalized tumor-informed assays, postoperative ctDNA positivity consistently identifies patients at increased risk of recurrence, although estimates remain imprecise in small cohorts. Serial testing can detect molecular relapse months before radiographic recurrence. Integrated pathology-MRD models demonstrate additional risk stratification within both pathological complete response and non-completeresponse groups. Observational findings suggest, but do not establish, that MRD-positive patients may derive greater benefit from adjuvant immunotherapy. Conversely, MRD negativity alone does not justify omission of indicated adjuvant treatment. Postoperative ctDNA provides complementary molecular risk information beyond conventional pathology in resectable ESCC. However, prospective MRD-guided trials are required before postoperative treatment or surveillance strategies can be routinely modified on the basis of ctDNA results alone.
Gallbladder carcinoma (GBC) is the most common biliary tract malignancy and is usually diagnosed at an advanced stage. Artificial intelligence (AI), including radiomics, machine learning (ML) and deep learning (DL), is increasingly applied to its diagnosis, surgical management and prognostication, but how far the reported performance can be trusted is unclear. PubMed/MEDLINE, Embase, Web of Science, the Cochrane Library and Google Scholar were searched to 5 May 2026 using database-specific search strategies. Each primary prediction-model study was characterized separately by design, number of centers and validation type (internal, temporal or external). Of 31 primary AI studies in GBC, 29 were retrospective, 22 were single-center and 24 reported internal validation only. In independent test or external cohorts, DL and ML models discriminated benign from malignant gallbladder lesions with areas under the curve (AUCs) of approximately 0.80–0.89; a multicenter model distinguishing xanthogranulomatous cholecystitis from GBC achieved an external AUC of 0.88, and automated computed tomography (CT) segmentation an external Dice score of 0.76. The best recurrence model achieved an AUC of 0.76 on internal (random-split) validation; none of the reviewed prognostic models reported a quantified external-validation estimate. Intraoperative AI is technically feasible but has no outcome data, and federated learning has been tested only in simulation. AI could plausibly assist several steps of GBC care, but the evidence is dominated by retrospective, single-center, internally validated studies from Asia, and metrics from different studies are not directly comparable. Prospective multicenter external validation, shared annotated datasets and standardized reporting are required before clinical adoption.
