Secure and Transparent Lung and Colon Cancer Classification Using Blockchain and Microsoft Azure.

BACKGROUND: The global healthcare system faces challenges in diagnosing and managing lung and colon cancers, which are significant health burdens. Traditional diagnostic methods are inefficient and prone to errors, while data privacy and security concerns persist. OBJECTIVE: This study aims to develop a secure and transparent framework for remote consultation and classification of lung and colon cancer, leveraging blockchain technology and Microsoft Azure cloud services. Dataset and Features: The framework utilizes the LC25000 dataset, containing 25,000 histopathological images, for training and evaluating advanced machine learning models. Key features include secure data upload, anonymization, encryption, and controlled access via blockchain and Azure services. METHODS: The proposed framework integrates Microsoft Azure's cloud services with a permissioned blockchain network. Patients upload CT scans through a mobile app, which are then preprocessed, anonymized, and stored securely in Azure Blob Storage. Blockchain smart contracts manage data access, ensuring only authorized specialists can retrieve and analyze the scans. Azure Machine Learning is used to train and deploy state-of-the-art machine learning models for cancer classification. Evaluation Metrics: The framework's performance is evaluated using metrics such as accuracy, precision, recall, and F1-score, demonstrating the effectiveness of the integrated approach in enhancing diagnostic accuracy and data security. RESULTS: The proposed framework achieves an impressive accuracy of 100% for lung and colon cancer classification using DenseNet, ResNet50, and MobileNet models with different split ratios (70-30, 80-20, 90-10). The F1-score and k-fold cross-validation accuracy (5-fold and 10-fold) also demonstrate exceptional performance, with values exceeding 99.9%. Real-time notifications and secure remote consultations enhance the efficiency and transparency of the diagnostic process, contributing to better patient outcomes and streamlined cancer care management.

Deep learning ensemble approach with explainable AI for lung and colon cancer classification using advanced hyperparameter tuning.

Lung and colon cancers are leading contributors to cancer-related fatalities globally, distinguished by unique histopathological traits discernible through medical imaging. Effective classification of these cancers is critical for accurate diagnosis and treatment. This study addresses critical challenges in the diagnostic imaging of lung and colon cancers, which are among the leading causes of cancer-related deaths worldwide. Recognizing the limitations of existing diagnostic methods, which often suffer from overfitting and poor generalizability, our research introduces a novel deep learning framework that synergistically combines the Xception and MobileNet architectures. This innovative ensemble model aims to enhance feature extraction, improve model robustness, and reduce overfitting.Our methodology involves training the hybrid model on a comprehensive dataset of histopathological images, followed by validation against a balanced test set. The results demonstrate an impressive classification accuracy of 99.44%, with perfect precision and recall in identifying certain cancerous and non-cancerous tissues, marking a significant improvement over traditional approach.The practical implications of these findings are profound. By integrating Gradient-weighted Class Activation Mapping (Grad-CAM), the model offers enhanced interpretability, allowing clinicians to visualize the diagnostic reasoning process. This transparency is vital for clinical acceptance and enables more personalized, accurate treatment planning. Our study not only pushes the boundaries of medical imaging technology but also sets the stage for future research aimed at expanding these techniques to other types of cancer diagnostics.

Classification of healthy and cancerous colon cells by Fourier transform infrared spectroscopy.

Colorectal cancer is one of the most diagnosed types of cancer in developed countries. Current diagnostic methods are partly dependent on pathologist experience and laboratories instrumentation. In this study, we used Fourier Transform Infrared (FTIR) spectroscopy in transflection mode, combined with Principal Components Analysis followed by Linear Discriminant Analysis (PCA-LDA) and Partial Least Squares - Discriminant Analysis (PLS-DA), to build a classification algorithm to diagnose colon cancer in cell samples, based on absorption spectra measured in two spectral ranges of the mid-infrared spectrum. In particular, PCA technique highlights small biochemical differences between healthy and cancerous cells: these are related to the larger lipid content in the former compared with the latter and to the larger relative amount of protein and nucleic acid components in the cancerous cells compared with the healthy ones. Comparison of the classification accuracy of PCA-LDA and PLS-DA methods applied to FTIR spectra measured in the 1000-1800 cm-1 (low wavenumber range, LWR) and 2700-3700 cm-1 (high wavenumber range, HWR) remarks that both algorithms are able to classify hidden class FTIR spectra with excellent accuracy (100 %) in both spectral regions. This is a hopeful result for clinical translation of infrared spectroscopy: in fact, it makes reliable the predictions obtained using FTIR measurements carried out only in the HWR, in which the glass slides used in clinical laboratories are transparent to IR radiation.

Elucidating Cancer Subtypes by Using the Relationship between DNA Methylation and Gene Expression.

Advancements in the field of next generation sequencing (NGS) have generated vast amounts of data for the same set of subjects. The challenge that arises is how to combine and reconcile results from different omics studies, such as epigenome and transcriptome, to improve the classification of disease subtypes. In this study, we introduce sCClust (sparse canonical correlation analysis with clustering), a technique to combine high-dimensional omics data using sparse canonical correlation analysis (sCCA), such that the correlation between datasets is maximized. This stage is followed by clustering the integrated data in a lower-dimensional space. We apply sCClust to gene expression and DNA methylation data for three cancer genomics datasets from the Cancer Genome Atlas (TCGA) to distinguish between underlying subtypes. We evaluate the identified subtypes using Kaplan-Meier plots and hazard ratio analysis on the three types of cancer-GBM (glioblastoma multiform), lung cancer and colon cancer. Comparison with subtypes identified by both single- and multi-omics studies implies improved clinical association. We also perform pathway over-representation analysis in order to identify up-regulated and down-regulated genes as tentative drug targets. The main goal of the paper is twofold: the integration of epigenomic and transcriptomic datasets followed by elucidating subtypes in the latent space. The significance of this study lies in the enhanced categorization of cancer data, which is crucial to precision medicine.

Clinical Challenges of Consensus Molecular Subtype CMS4 Colon Cancer in the Era of Precision Medicine.

Over the past decade, our understanding of the diversity of colorectal cancer has expanded significantly, raising hopes of tailoring treatments more precisely for individual patients. A key achievement in this direction was the establishment of the consensus molecular classification, particularly identifying the challenging consensus molecular subtype (CMS) CMS4 associated with poor prognosis. Because of its aggressive nature, extensive research is dedicated to the CMS4 subgroup. Recent years have unveiled molecular and microenvironmental features at the tissue level specific to CMS4 colorectal cancer. This has paved the way for mechanistic studies and the development of preclinical models. Simultaneously, efforts have been made to easily identify patients with CMS4 colorectal cancer. Reassessing clinical trial results through the CMS classification lens has improved our understanding of the therapeutic challenges linked to this subtype. Exploration of the biology of CMS4 colorectal cancer is yielding potential biomarkers and novel treatment approaches. This overview aims to provide insights into the clinico-biological characteristics of the CMS4 subgroup, the molecular pathways driving this subtype, and available diagnostic options. We also emphasize the therapeutic challenges associated with this subtype, offering potential explanations. Finally, we summarize the current tailored treatments for CMS4 colorectal cancer emerging from fundamental and preclinical studies.

Identification of unique rectal cancer-specific subtypes.

BACKGROUND: Existing colorectal cancer subtyping methods were generated without much consideration of potential differences in expression profiles between colon and rectal tissues. Moreover, locally advanced rectal cancers at resection often have received neoadjuvant chemoradiotherapy which likely has a significant impact on gene expression. METHODS: We collected mRNA expression profiles for rectal and colon cancer samples (n = 2121). We observed that (i) Consensus Molecular Subtyping (CMS) had a different prognosis in treatment-naïve rectal vs. colon cancers, and (ii) that neoadjuvant chemoradiotherapy exposure produced a strong shift in CMS subtypes in rectal cancers. We therefore clustered 182 untreated rectal cancers to find rectal cancer-specific subtypes (RSSs). RESULTS: We identified three robust subtypes. We observed that RSS1 had better, and RSS2 had worse disease-free survival. RSS1 showed high expression of MYC target genes and low activity of angiogenesis genes. RSS2 exhibited low regulatory T cell abundance, strong EMT and angiogenesis signalling, and high activation of TGF-ß, NF-κB, and TNF-α signalling. RSS3 was characterised by the deactivation of EGFR, MAPK and WNT pathways. CONCLUSIONS: We conclude that RSS subtyping allows for more accurate prognosis predictions in rectal cancers than CMS subtyping and provides new insight into targetable disease pathways within these subtypes.

Development and validation of a novel classification scheme for combining pathological T stage and log odds of positive lymph nodes for colon cancer.

AIM: Log Odds of Positive Lymph Nodes (LODDS) have a better predictive ability than N stage for colon cancer. However, the prognostic value of developing a novel prognostic classification by combining T stage and LODDS (TLODDS) for colon cancer remains unknown. Therefore, in the present study, we aimed to develop a TLODDS classification for colon cancer, and assess whether or not the novel TLODDS classification could improve survival stratification by comparing its discrimination, model-fitting, and net benefits, with the American Joint Committee on Cancer (AJCC) Tumor/Node/Metastasis (TNM) classification. METHODS: 45,558 Western colon cancers were identified in the Surveillance, Epidemiology, and End Results database as a training set. A novel LODDS stage was established and patients with similar survival rates were grouped by combining T and LODDS stages to develop a novel TLODDS classification. The TLODDS classification was further assessed in a Chinese validation set of 3,515 colon cancers and an application set of 3,053 rectal cancers. RESULTS: We developed a novel TLODDS classification that incorporated 7 stages: stage I (T1LODDS1), IIA (T2LODDS1, T1LODDS2, T1LODDS3), IIB (T2LODDS2-3, T3LODDS1, T1LODDS4), IIC (T3LODDS2, T2LODDS4, T4aLODDS1), IIIA (T3LODDS3, T1-2LODDS5, T4bLODDS1, T4aLODDS2), IIIB (T3LODDS4-5, T4aLODDS3-4, T4bLODDS2) and IIIC (T4bLODDS3-5, T4aLODDS5). In the training set, it showed significantly better discrimination (area under the receiver operating characteristic (ROC) curve, 0.691 vs. 0.664, P < 0.001), better model-fitting (Akaike information criteria, 265,644 vs. 267,410), and superior net benefits, than the latest AJCC TNM classification. The predictive performance of the TLODDS classification was further validated in colon cancers and was successfully applied in rectal cancers with regards to both overall and disease-free survival. CONCLUSIONS: The TLODDS classification has better discriminatory ability, model-fitting, and net benefits than the existing TNM classification, and represents an alternative to the current TNM classifications for colon and rectal cancers.

Evaluation of a 55-gene classifier as a prognostic biomarker for adjuvant chemotherapy in stage III colon cancer patients.

BACKGROUND: Adjuvant chemotherapy reduces the risk of recurrence of stage III colon cancer (CC). However, more effective prognostic and predictive biomarkers are needed for better treatment stratification of affected patients. Here, we constructed a 55-gene classifier (55GC) and investigated its utility for classifying patients with stage III CC. METHODS: We retrospectively identified patients aged 20-79 years, with stage III CC, who received adjuvant chemotherapy with or without oxaliplatin, between the years 2009 and 2012. RESULTS: Among 938 eligible patients, 203 and 201 patients who received adjuvant chemotherapy with and without oxaliplatin, respectively, were selected by propensity score matching. Of these, 95 patients from each group were analyzed, and their 5-year relapse-free survival (RFS) rates with and without oxaliplatin were 73.7 and 77.1%, respectively. The hazard ratios for 5-year RFS following adjuvant chemotherapy (fluoropyrimidine), with and without oxaliplatin, were 1.241 (95% CI, 0.465-3.308; P = 0.67) and 0.791 (95% CI, 0.329-1.901; P = 0.60), respectively. Stratification using the 55GC revealed that 52 (27.3%), 78 (41.1%), and 60 (31.6%) patients had microsatellite instability (MSI)-like, chromosomal instability (CIN)-like, and stromal subtypes, respectively. The 5-year RFS rates were 84.3 and 72.0% in patients treated with and without oxaliplatin, respectively, for the MSI-like subtype (HR, 0.495; 95% CI, 0.145-1.692; P = 0.25). No differences in RFS rates were noted in the CIN-like or stromal subtypes. Stratification by cancer sidedness for each subtype showed improved RFS only in patients with left-sided primary cancer treated with oxaliplatin for the MSI-like subtype (P = 0.007). The 5-year RFS rates of the MSI-like subtype in left-sided cancer patients were 100 and 53.9% with and without oxaliplatin, respectively. CONCLUSIONS: Subclassification using 55GC and tumor sidedness revealed increased RFS in patients within the MSI-like subtype with stage III left-sided CC treated with fluoropyrimidine and oxaliplatin compared to those treated without oxaliplatin. However, the predictive power of 55GC subtyping alone did not reach statistical significance in this cohort, warranting larger prospective studies. TRIAL REGISTRATION: The study protocol was registered in the University Hospital Medical Education Network (UMIN) clinical trial registry (UMIN study ID: 000023879 ).

Comparison of model fit and discriminatory ability of M category as defined by the 7th and 8th editions of the tumor-node-metastasis classification of colorectal cancer and the 9th edition of the Japanese classification.

BACKGROUND: In transitioning from the 7th edition of the tumor-node-metastasis classification (TNM-7) to the 8th edition (TNM-8), colorectal cancer with peritoneal metastasis was newly categorized as M1c. In the 9th edition of the Japanese Classification of colorectal, appendiceal, and anal carcinoma (JPC-9), M1c is further subdivided into M1c1 (without other organ involvement) and M1c2 (with other organ involvement). This study aimed to compare the model fit and discriminatory ability of the M category of these three classification systems, as no study to date has made this comparison. METHODS: The study population consisted of stage IV colorectal cancer patients who were referred to the National Cancer Center Hospital from 2000 to 2017. The Akaike information criterion (AIC), Harrell's concordance index (C-index), and time-dependent receiver operating characteristic (ROC) curves were used to compare the three classification systems. Subgroup analyses, stratified by initial treatment year, were also performed. RESULTS: According to TNM-8, 670 (55%) patients had M1a, 273 (22%) had M1b, and 279 (23%) had M1c (87 M1c1 and 192 M1c2 using JPC-9) tumors. Among the three classification systems, JPC-9 had the lowest AIC value (JPC-9: 10546.3; TNM-7: 10555.9; TNM-8: 10585.5), highest C-index (JPC-9: 0.608; TNM-7: 0.598; TNM-8: 0.599), and superior time-dependent ROC curves throughout the observation period. Subgroup analyses were consistent with these results. CONCLUSIONS: While the revised M category definition did not improve model fit and discriminatory ability from TNM-7 to TNM-8, further subdivision of M1c in JPC-9 improved these parameters. These results support further revisions to M1 subcategories in future editions of the TNM classification system.

Intratumor CMS Heterogeneity Impacts Patient Prognosis in Localized Colon Cancer.

PURPOSE: The consensus molecular subtypes (CMS) represent a significant advance in the understanding of intertumor heterogeneity in colon cancer. Intratumor heterogeneity (ITH) is the new frontier for refining prognostication and understanding treatment resistance. This study aims at deciphering the transcriptomic ITH of colon cancer and understanding its potential prognostic implications. EXPERIMENTAL DESIGN: We deconvoluted the transcriptomic profiles of 1,779 tumors from the PETACC8 trial and 155 colon cancer cell lines as weighted sums of the four CMSs, using the Weighted In Silico Pathology (WISP) algorithm. We assigned to each tumor and cell line a combination of up to three CMS subtypes with a threshold above 20%. RESULTS: Over 55% of tumors corresponded to mixtures of at least two CMSs, demonstrating pervasive ITH in colon cancer. Of note, ITH was associated with shorter disease-free survival (DFS) and overall survival, [HR, 1.34; 95% confidence interval (CI; 1.12-1.59), 1.40, 95% CI (1.14-1.71), respectively]. Moreover, we uncovered specific combinations of CMS associated with dismal prognosis. In multivariate analysis, ITH represents the third parameter explaining DFS variance, after T and N stages. At a cellular level, combined WISP and single-cell transcriptomic analysis revealed that most colon cancer cell lines are a mixture of cells falling into different CMSs, indicating that ITH may correspond to distinct functional statuses of colon cancer cells. CONCLUSIONS: This study shows that CMS-based transcriptomic ITH is frequent in colon cancer and impacts its prognosis. CMS-based transcriptomic ITH may correspond to distinct functional statuses of colon cancer cells, suggesting plasticity between CMS-related cell populations. Transcriptomic ITH deserves further assessment in the context of personalized medicine.

A framework model using multifilter feature selection to enhance colon cancer classification.

Gene expression profiles can be utilized in the diagnosis of critical diseases such as cancer. The selection of biomarker genes from these profiles is significant and crucial for cancer detection. This paper presents a framework proposing a two-stage multifilter hybrid model of feature selection for colon cancer classification. Colon cancer is being extremely common nowadays among other types of cancer. There is a need to find fast and an accurate method to detect the tissues, and enhance the diagnostic process and the drug discovery. This paper reports on a study whose objective has been to improve the diagnosis of cancer of the colon through a two-stage, multifilter model of feature selection. The model described deals with feature selection using a combination of Information Gain and a Genetic Algorithm. The next stage is to filter and rank the genes identified through this method using the minimum Redundancy Maximum Relevance (mRMR) technique. The final phase is to further analyze the data using correlated machine learning algorithms. This two-stage approach, which involves the selection of genes before classification techniques are used, improves success rates for the identification of cancer cells. It is found that Decision Tree, K-Nearest Neighbor, and Naïve Bayes classifiers had showed promising accurate results using the developed hybrid framework model. It is concluded that the performance of our proposed method has achieved a higher accuracy in comparison with the existing methods reported in the literatures. This study can be used as a clue to enhance treatment and drug discovery for the colon cancer cure.

Human polyomaviruses genomes in clinical specimens of colon cancer patients.

Colon cancer is the third cause of cancer death in the developed countries. Some environmental factors are involved in its pathogenesis, including viral infections. The possible involvement of human polyomaviruses (HPyVs) in colon cancer pathogenesis has been previously reported, leading to inconsistent conclusions. Clinical specimens were collected from 125 colon cancer patients. Specifically, 110 tumor tissues, 55 negative surgical margins, and 39 peripheral blood samples were analyzed for the presence of six HPyVs: JC polyomavirus (JCPyV), BK polyomavirus (BKPyV), Merkel cell PyV (MCPyV), HPyV -6, -7, and -9 by means of DNA isolation and subsequent duplex Real Time quantitative polymerase chain reaction. HPyVs genome was detected in 33/204 samples (16.2%): the significant higher positivity was found in tumor tissues (26/110, 23.6%), followed by negative surgical margins (3/55, 5.5%, p < .05), and peripheral blood mononuclear cells (PBMCs) (4/39; 10.3%). HPyVs load was statistically higher only in the tumor tissues compared to negative surgical margins (p < .05). Specifically, MCPyV was detected in 19.1% (21/110) of tumor tissues, 3.6% (2/55) of negative surgical margins (p < .05), and 7.7% (3/39) of PBMCs; HPyV-6 in 2.7% (3/110) of tumor tissues, and 1.8% (1/55) of negative surgical margins; one tumor tissue (1/110, 0.9%) and one PBMCs sample (1/39, 2.6%) were positive for BKPyV; JCPyV was present in 0.9% (1/110) of tumor tissues. HPyV-7 and 9 were not detected in any sample. High prevalence and load of MCPyV genome in the tumor tissues might be indicative of a relevant rather than bystander role of the virus in the colon tumorigenesis.

Reproducibility of AJCC Criteria for Classifying Deeply Invasive Colon Cancers Is Suboptimal for Consistent Cancer Staging.

The eighth edition of the American Joint Committee on Cancer (AJCC) Staging Manual attempts to address ambiguity in the pT category assignment for colon cancer from prior editions. Despite modifications, the distinction between the pT3 and pT4a categories continues to be a source of diagnostic confusion. In this study, we assessed interobserver agreement among pathologists from different institutions in the application of AJCC eighth edition criteria for categorizing deeply invasive colonic adenocarcinomas. We identified morphologic patterns that produce diagnostic confusion. We assessed 47 colon cancers that closely approached the serosal surface. Six pathologists with interest in gastrointestinal pathology and 4 focused in other subspecialties classified each case as pT3 or pT4a, based on examination of low-magnification and high-magnification images of the most deeply invasive area. Interobserver agreement was assessed using Fleiss' κ. Cases displayed 3 morphologic patterns at the advancing tumor edge, namely, (1) continuous invasion through an inflammatory focus, (2) pushing border, and (3) infiltrative glands and cell clusters with serosal reaction. Gastrointestinal pathologists achieved slight (κ=0.21) or moderate (κ=0.46) and (κ=0.51) agreement in each category, whereas agreement among nongastrointestinal pathologist was fair (0.31) and (0.39), or moderate (0.57) for each category, respectively. In 10 (21%) cases, the distinction between pT3 and pT4a would have changed the overall clinical stage. We conclude that histologic criteria for serosal penetration is a persistent source of diagnostic ambiguity for gastrointestinal and general pathologists in the pT categorization of colon cancers. Clarification of these criteria will help ensure uniform reporting of pathologic and clinical stage.

Proposal of a Classification System for Adenocarcinoma Involving the Ileocecal Valve (Bauhin's Valve): A Retrospective Study of 689 Right Hemicolectomized Patients.

BACKGROUND/AIM: As of 2020, adenocarcinoma arising in the ileocecal valve (ICV-A) has been examined along with cecal and right colon cancer (RCC) under the collective heading "ileocecal" tumor. We propose a new classification system for this cancer. PATIENTS AND METHODS: We retrospectively analyzed RCC patients from 2003 to 2019. The scheme was: i) Type I cancer for adenocarcinomas residing in ICV; ii) Type II, if they reside 1 to 5 mm from ICV; iii) Type III, 6 mm to 10 mm from ICV; iv) Type IV, at 1,1 to 5 cm; v) Type V, at more than 5 cm (ascending colon cancer). RESULTS: Of 689 hemicolectomized patients, there were 91 (13.2%) Type I, 87 Type II (12.6%), 38 (5.5%) Type III, 157 (22.8%) Type IV and 314 (45.6%) Type V. Each type was associated with at least one clinicopathologic feature. CONCLUSION: ICV-A was classified into five types (I-V) according to the distance from ICV. Further studies are needed in order to corroborate our findings.

Deep Learning Models for Histopathological Classification of Gastric and Colonic Epithelial Tumours.

Histopathological classification of gastric and colonic epithelial tumours is one of the routine pathological diagnosis tasks for pathologists. Computational pathology techniques based on Artificial intelligence (AI) would be of high benefit in easing the ever increasing workloads on pathologists, especially in regions that have shortages in access to pathological diagnosis services. In this study, we trained convolutional neural networks (CNNs) and recurrent neural networks (RNNs) on biopsy histopathology whole-slide images (WSIs) of stomach and colon. The models were trained to classify WSI into adenocarcinoma, adenoma, and non-neoplastic. We evaluated our models on three independent test sets each, achieving area under the curves (AUCs) up to 0.97 and 0.99 for gastric adenocarcinoma and adenoma, respectively, and 0.96 and 0.99 for colonic adenocarcinoma and adenoma respectively. The results demonstrate the generalisation ability of our models and the high promising potential of deployment in a practical histopathological diagnostic workflow system.

Cosecha ganglionar en cáncer de colon. Mitos y magia / Lymph node harvest in colon cancer. Myths and magic / Contagem de linfonodos no câncer de cólon. Mitos e magia

Resumen: El compromiso ganglionar es crítico en la estadificación del cáncer de colon como factor pronóstico y como determinante de tratamiento adyuvante. Se sigue discutiendo el número de ganglios adecuados a resecar, cuáles son los factores que inciden en la cosecha ganglionar y el significado biológico de ésta. Se revisan las variables clínicas y de la propia biología tumoral que hacen que la definición de un número determinado de ganglios, como gold standard de cosecha ganglionar adecuada, sea controversial. El número 12 no necesariamente es un número "mágico" marcador de calidad. Extender la resección para aumentar la cosecha ganglionar no mejora la estadificación, expone al paciente a riesgos innecesarios, sin efecto terapéutico comprobado. La "magia" sigue siendo realizar resecciones regladas, que incluyan el pedículo vascular y el meso satélite al tumor, ajustando la resección a las características del paciente. Menos no es más, pero más no es necesariamente mejor.

Summary: Lymph node compromise is critical in colon cancer staging, as a prognostic factor and to determine adjuvant therapy. The number of lymph nodes to be resected is still under discussion, as well as the factor that have an impact on lymph node harvest and its biological significance. We reviewed clinical variables and variables that are specific to the tumor, what results in the definition of a certain number of lymph nodes, as the adequate Gold Standard for lymph node harvest being controversial. 12 is not necessarily a "magic" number that marks quality. Extending resection to increase lymph node harvest does not improve staging, it exposes patients to unnecessary risks, there being no therapeutic effect guaranteed. The "Magic" continues to be routine resection that includes the cystic pedicle and the area around the tumour, adjusting resection to the patient's characteristics. Less is not best, but more is not necessarily better.

Resumo: O compromisso ganglionar é crítico no estadiamento do câncer de cólon, como fator prognóstico e como determinante do tratamento adjuvante. A discussão sobre o número de gânglios adequados a ressecar, quais são os fatores que incidem sobre a definição do número de linfonodos a ser retirados e seu significado biológico. Faz-se uma revisão das variáveis clínicas e da própria biologia tumoral, que fazem com que a definição de um número determinado de gânglios como Gold Standard do número adequado de linfonodos a remover seja controversa. O número 12 não é necessariamente um número "mágico", um marcador de qualidade. Ampliar a ressecção para aumentar o número de linfonodos que serão retirados não melhora o estadiamento, expõe o paciente a riscos desnecessários, sem um efeito terapêutico comprovado. A "Magia" continua sendo realizar ressecções de acordo com parâmetros definidos, que incluam o pedículo vascular e o mesocólon satélite ao tumor, ajustando a ressecção às características do paciente. Menos não é mais, porém mais não é necessariamente melhor.

Histotype influences emergency presentation and prognosis in colon cancer surgery.

AIM: To investigate whether differences in histotype in colon cancer correlate with clinical presentation and if they might influence oncological outcomes and survival. METHODS: Data regarding colon cancer patients operated both electively or in emergency between 2009 and 2014 were retrospectively collected from a prospectively maintained database and analyzed for the purpose of this study. Rectal cancer was excluded from this analysis. Statistical univariate and multivariate analyses were performed to investigate possible significant variables influencing clinical presentation, as well as oncological outcomes and survival. RESULTS: Data from 219 patients undergoing colorectal resection for cancer of the colon only were retrieved. One hundred seventy-four patients had an elective procedure and forty-five had an emergency colectomy. Emergency presentation was more likely to occur in mucinous (p < 0.05) and signet ring cell (p < 0.01) tumors. No definitive differences in 5-year overall (44.7% vs. 60.6%, p = 0.078) and disease-free (51.2% vs. 64.4%, p = 0.09) survival were found between the two groups as a whole, but the T3 emergency patients showed worse prognosis than the elective (p < 0.03). Lymph node invasion, laparoscopy, histology, and blood transfusions were independent variables found to influence survival. Distribution assessed for pTNM stage showed T3 cancers were more common in emergency (p < 0.01). CONCLUSIONS AND DISCUSSION: Mucinous and signet ring cell tumors are related to emergency presentation, pT3 stage, poorest outcomes, and survival. Disease-free survival in patients who had emergency surgery for T3 colon cancer seems related to the histotype.

Computer-assisted assessment of colonic polyp histopathology using probe-based confocal laser endomicroscopy.

INTRODUCTION: Probe-based confocal laser endomicroscopy (pCLE) is a promising modality for classifying polyp histology in vivo, but decision making in real-time is hampered by high-magnification targeting and by the learning curve for image interpretation. The aim of this study is to test the feasibility of a system combining the use of a low-magnification, wider field-of-view pCLE probe and a computer-assisted diagnosis (CAD) algorithm that automatically classifies colonic polyps. METHODS: This feasibility study utilized images of polyps from 26 patients who underwent colonoscopy with pCLE. The pCLE images were reviewed offline by two expert and five junior endoscopists blinded to index histopathology. A subset of images was used to train classification software based on the consensus of two GI histopathologists. Images were processed to extract image features as inputs to a linear support vector machine classifier. We compared the CAD algorithm's prediction accuracy against the classification accuracy of the endoscopists. RESULTS: We utilized 96 neoplastic and 93 non-neoplastic confocal images from 27 neoplastic and 20 non-neoplastic polyps. The CAD algorithm had sensitivity of 95%, specificity of 94%, and accuracy of 94%. The expert endoscopists had sensitivities of 98% and 95%, specificities of 98% and 96%, and accuracies of 98% and 96%, while the junior endoscopists had, on average, a sensitivity of 60%, specificity of 85%, and accuracy of 73%. CONCLUSION: The CAD algorithm showed comparable performance to offline review by expert endoscopists and improved performance when compared to junior endoscopists and may be useful for assisting clinical decision making in real time.

Deep transfer learning methods for colon cancer classification in confocal laser microscopy images.

PURPOSE: The gold standard for colorectal cancer metastases detection in the peritoneum is histological evaluation of a removed tissue sample. For feedback during interventions, real-time in vivo imaging with confocal laser microscopy has been proposed for differentiation of benign and malignant tissue by manual expert evaluation. Automatic image classification could improve the surgical workflow further by providing immediate feedback. METHODS: We analyze the feasibility of classifying tissue from confocal laser microscopy in the colon and peritoneum. For this purpose, we adopt both classical and state-of-the-art convolutional neural networks to directly learn from the images. As the available dataset is small, we investigate several transfer learning strategies including partial freezing variants and full fine-tuning. We address the distinction of different tissue types, as well as benign and malignant tissue. RESULTS: We present a thorough analysis of transfer learning strategies for colorectal cancer with confocal laser microscopy. In the peritoneum, metastases are classified with an AUC of 97.1, and in the colon the primarius is classified with an AUC of 73.1. In general, transfer learning substantially improves performance over training from scratch. We find that the optimal transfer learning strategy differs for models and classification tasks. CONCLUSIONS: We demonstrate that convolutional neural networks and transfer learning can be used to identify cancer tissue with confocal laser microscopy. We show that there is no generally optimal transfer learning strategy and model as well as task-specific engineering is required. Given the high performance for the peritoneum, even with a small dataset, application for intraoperative decision support could be feasible.

Colon Neuroendocrine Tumors: A New Lymph Node Staging Classification.

BACKGROUND: The American Joint Commission on Cancer, the European Neuroendocrine Tumor Society, and the North American Neuroendocrine Tumor Society all classify colon neuroendocrine tumor (NET) nodal metastasis as N0 or N1. This binary classification does not allow for further prognostication by the total number of positive lymph nodes. This study aimed to evaluate whether the total number of positive lymph nodes affects the overall survival for patients with colon NET. METHODS: The National Cancer Database was used to identify patients with colon NET. Nearest-neighborhood grouping was performed to classify patients by survival to create a new nodal staging system. The Surveillance, Epidemiology, and End Results database was used to validate the new nodal staging classification. RESULTS: Colon NETs were identified in 2472 patients. Distinct 5-year survival rates were estimated for the patients with N0 (no positive lymph nodes; 69.8%; 95% confidence interval [CI], 66.7-72.7%), N1a (1 positive lymph node; 63.9%; 95% CI, 59.6-68.0%), N1b (2-9 positive lymph nodes; 38.9%; 95% CI, 35.4-42.3%), and N2 (≥ 10 positive lymph nodes; 15.7%; 95% CI, 11.9-20.0%; p < 0.001) nodal classifications. The validation population showed distinct 5-year survival rates with the new nodal staging. In multivariable Cox regression, the new nodal stage was a significant independent predictor of overall survival. CONCLUSIONS: The number of positive locoregional lymph nodes in colon NETs is an independent prognostic factor. For patients with colon NETs, N0, N1a, N1b, and N2 classifications for nodal metastasis more accurately predict survival than current staging systems.