Colorectal histology classification remains challenging because subtle morphological differences, staining variability, and inappropriate model configurations can reduce generalization. This study proposes COOT-CNN, a lightweight convolutional neural network in which architectural design and training strategy are optimized jointly through the COOT metaheuristic. The search process considers network depth, convolution type, kernel size, feature width, squeeze-and-excitation attention, activation functions, pooling, regularization, loss formulation, learning-rate scheduling, data augmentation, exponential moving average, and mixed-precision training. The selected configuration is retrained independently and evaluated using accuracy, balanced accuracy, macro-F1, ROC-AUC, precision-recall analysis, confidence intervals, and paired statistical testing. Same-split comparisons are conducted against ResNet-50, EfficientNet-B0, and Swin-T under a consistent experimental protocol. COOT-CNN achieved 96.72% test accuracy and 96.71% macro-F1, while providing statistically significant improvements over the evaluated baselines. The optimized model also maintained a compact computational profile, with fewer parameters and lower inference latency than the comparison networks. These findings demonstrate that jointly optimizing architecture and training components can produce an accurate, efficient, and stable model for binary colorectal histology patch classification. The proposed framework should be regarded as a methodological proof-of-concept, with external, patient-level, and whole-slide validation required before clinical application.

This paper introduces a hybrid deep learning model combining ConvNeXt and Swin Transformer for classifying brain tumors from MRI scans. The ConvNeXt backbone is employed to obtain detailed local spatial features, whereas the Swin Transformer identifies hierarchical long-range dependencies, facilitating complementary feature representation. The proposed model is evaluated on a combined public MRI dataset of 7,023 images distributed across four categories: glioma, meningioma, pituitary, and no tumor. Experimental results demonstrate that the proposed hybrid architecture outperforms several state-of-the-art convolutional and transformer-based models, achieving an accuracy of 95.37% with competitive precision and F-score. Additionally, qualitative explainability assessment using attention-based visualization techniques offers insight into the model's decision-making by highlighting diagnostically significant regions. Furthermore, we evaluate the proposed H-ConvNeXt-Swin model on the unified dataset and also report source-stratified performance on each of the three constituent datasets: Figshare, SARTAJ, and Br35H. Future work will focus on validating the proposed framework on multi-center clinical datasets and extending it to more complex tasks such as tumor localization and segmentation.

Accurate grading of bone marrow fibrosis is essential for the diagnosis and prognostication of myeloproliferative neoplasms (MPNs). The 2017 World Health Organization (WHO) Classification recommends a trichrome stain be performed to evaluate collagen fibrosis when reticulin fibrosis grade is >1, but the inter-observer concordance in assessing collagen fibrosis in a routine diagnostic laboratory and the utility of this stain are unclear. This study assesses the inter-observer agreement in the grading of reticulin fibrosis and collagen fibrosis, and the concordance between reticulin and collagen fibrosis. Two pathologists assessed 174 consecutive bone marrow trephine biopsies with a diagnosis of an MPN at a tertiary referral centre. Reticulin fibrosis by silver impregnation and collagen fibrosis by Masson's trichrome stain were independently graded by the pathologists while blinded to clinical information. Discordant scores were reviewed to establish a final grade by consensus. Substantial inter-observer concordance was obtained for both reticulin [83%; Cohen's kappa 0.756, 95% confidence interval (CI) 0.69-0.84] and collagen fibrosis grading (81%; kappa 0.730, 95% CI 0.64-0.82). Using the final consensus scores, reticulin and collagen fibrosis grades were concordant in 60% of cases. Collagen fibrosis grade was higher than reticulin fibrosis score in 11% of cases. Inter-observer concordance in assessing the degree of collagen fibrosis using the trichrome stain is substantial. Reticulin (silver) staining alone may underestimate the degree of fibrosis. The prognostic significance where the degree of collagen fibrosis is greater than that of reticulin is uncertain and should be investigated in future studies.

We reported a case of extremely severe tuberous sclerosis complex-associated lymphangioleiomyomatosis (TSC-LAM) involving the brain, liver, kidneys, and lungs that was successfully treated with sirolimus, with a complete 17-year follow-up. The patient was diagnosed in 2008 and experienced rapid progression; by 2012, the patient had developed respiratory failure, severe anemia, renal insufficiency, and extreme abdominal distension. After 3 days of treatment with sirolimus, the aforementioned symptoms improved significantly, and respiratory failure resovled. Over the following 7 years of regular maintenance therapy, lung function declined only slightly each year, and the 6-minute walking test(6MWT) showed no decline. However, after sirolimus was discontinued from January 2020 to April 2021, the patient's lung function and 6MWT declined rapidly, with concurrent development of pneumothorax. Although sirolimus was subsequently resumed and remained effective, with no recurrence of pneumothorax or chylothorax, parameters such as lung function and 6MWT still did not recover significantly. This case suggests that TSC-LAM requires long-term maintenance therapy with sirolimus. Abrupt discontinuation of treatment may lead to rapid clinical deterioration.

Epstein-Barr virus-associated smooth muscle tumours (EBV-SMTs) are rare neoplasms typically affecting immunosuppressed individuals such as organ transplant recipients. This case follows a woman in her 30s who presented with multi-organ masses 5 years after a kidney transplant. While initial suspicion focused on common post-transplant infections or malignancy, a liver biopsy revealed a smooth muscle neoplasm positive for EBV by EBV-encoded RNA in-situ hybridisation, consistent with EBV-SMT. Managing EBV-SMT is challenging as no standardised treatment regimens exist. When reducing immunosuppression-the conventional first-line approach-failed, a multidisciplinary team opted for a non-standard regimen including Valganciclovir and Pazopanib, which is not a recognised therapy for this condition. The patient achieved a partial metabolic response and maintained stable disease at a 2 year follow-up. This case emphasises the importance of considering this differential when encountering spindle cell neoplasms in an immunosuppressed setting while highlighting the necessity of establishing standardised treatment guidelines for this rare disease.

The advent of potent immune checkpoint inhibitor and other targeted therapies has transformed neoadjuvant therapy into a potentially more efficacious treatment strategy overall. Administering immunotherapy before surgery can enhance the immune system's ability to target melanoma cells, utilize neoantigens, and treat clinically occult and metastatic disease. Furthermore, it provides an opportunity for assessment of treatment response based on pathologic assessment and ultimately leads to better survival rates, lower recurrence, and minimizes overuse of adjuvant treatments. Continued clinical trials will further elucidate best practices with respect to treatment duration, choice, and treatment strategies.

The treatment of locally advanced rectal cancer has been transformed over the last century. While initially treated with surgery alone, adjuvant therapy with chemotherapy and radiation became common for locally advanced disease to prevent local and distant recurrence. Several trials subsequently showed benefit to moving this therapy into the neoadjuvant setting which led to improved oncologic outcomes, greater compliance, and similar surgical morbidity. In this article, we will explore this historical transition to a total neoadjuvant approach that can lead to cure in select patients and brings into question the role of surgery altogether.

Esophagectomy remains the standard therapy for localized esophageal cancer, but postoperative function depends heavily on effective gastric conduit emptying. The aperistaltic conduit located in the negative pressure thoracic cavity is at risk of delayed gastric emptying leading to aspiration and pulmonary complications. This article summarizes historical and contemporary evidence evaluating pyloroplasty and alternative pyloric drainage strategies during esophagectomy. While earlier trials produced mixed results, recent level I evidence supports pyloroplasty as a safe adjunct that reduces gastric outlet obstruction and postoperative interventions. Meticulous conduit construction and vigilant postoperative assessment remain essential. There continues to be controversy, and an ongoing randomized trial, The PYNI-GAREREO Phase III Trial of Pyloroplasty Vs. No Pyloroplasty will be completed in the near future. (Trials.2023 Jun 19;24:412) and will add further level 1 evidence to this controversy.

The low response rate to immunotherapy in patients with lung adenocarcinoma is primarily due to tumor immune evasion. The tumor immunosuppressive microenvironment orchestrates this evasion, yet the underlying mechanisms remain elusive. Here we identify non-specific cytotoxic cell receptor protein 1 (NCCRP1) as a critical and previously uncharacterized regulator of this process.

We first analyzed publicly accessible patient-derived single-cell RNA sequencing and RNA sequencing data to investigate the expression of NCCRP1 in lung adenocarcinoma and its impact on the tumor immune microenvironment. Subsequently, the Nccrp1 gene was knocked out in mouse lung adenocarcinoma cells using CRISPR-Cas9. The effect of NCCRP1 deletion on tumor growth was then evaluated using subcutaneous transplantation models in both NSG and C57BL/6 mice. Single-cell RNA sequencing, flow cytometry and multiple targeted in vivo interventions were employed to assess the influence of NCCRP1 on the tumor immune microenvironment. To explore the underlying mechanisms by which NCCRP1 regulates the tumor immune microenvironment, we conducted co-immunoprecipitation, RNA pull-down, ubiquitination assay, mass spectrometry, isobaric tags for relative and absolute quantitation proteomics, dual-luciferase reporter gene assay, and ELISA.

Loss of NCCRP1 inhibits lung tumor growth and prolongs survival in immunocompetent C57BL/6 mouse models. NCCRP1 deficiency upregulates CX3CL1 to recruit CX3CR1⁺ antitumoral macrophages. These macrophages subsequently secrete CXCL9 and CXCL10, enhancing the infiltration of CD8⁺ T cells and NK cells into the tumor microenvironment. Mechanistically, NCCRP1 and STAU1 competitively bind to NEDD4. NCCRP1 deficiency enhances the STAU1-NEDD4 interaction, promoting ubiquitination and proteasomal degradation of STAU1, thereby increasing CX3CL1 messenger RNA stability. Remarkably, ablating NCCRP1 synergized with anti-programmed cell death protein 1 or interferon-γ therapy, leading to complete tumor eradication.

Our findings highlight NCCRP1 targeting as a promising therapeutic strategy to reprogram the immunosuppressive microenvironment and overcome the "cold tumor" phenotype in lung adenocarcinoma.