Autologous hematopoietic stem cell transplantation for a patient with multiple autoimmune diseases.

Level of autoantibodies after autologous hematopoietic stem cell transplantation. AGA, antigliadin antibody; AHSCT, autologous hematopoietic stem cell transplantation; Anti-GAD65, Ab anti-glutamic acid decarboxylase epitope 65 antibody; Anti-TPO, anti-thyroid peroxidase antibody; CU, chemiluminescent unit.

Central Reading of Ulcerative Colitis Clinical Trial Videos Using Neural Networks.

BACKGROUND AND AIMS: Endoscopic disease activity scoring in ulcerative colitis (UC) is useful in clinical practice but done infrequently. It is required in clinical trials, where it is expensive and slow because human central readers are needed. A machine learning algorithm automating the process could elevate clinical care and facilitate clinical research. Prior work using single-institution databases and endoscopic still images has been promising. METHODS: Seven hundred and ninety-five full-length endoscopy videos were prospectively collected from a phase 2 trial of mirikizumab with 249 patients from 14 countries, totaling 19.5 million image frames. Expert central readers assigned each full-length endoscopy videos 1 endoscopic Mayo score (eMS) and 1 Ulcerative Colitis Endoscopic Index of Severity (UCEIS) score. Initially, video data were cleaned and abnormality features extracted using convolutional neural networks. Subsequently, a recurrent neural network was trained on the features to predict eMS and UCEIS from individual full-length endoscopy videos. RESULTS: The primary metric to assess the performance of the recurrent neural network model was quadratic weighted kappa (QWK) comparing the agreement of the machine-read endoscopy score with the human central reader score. QWK progressively penalizes disagreements that exceed 1 level. The model's agreement metric was excellent, with a QWK of 0.844 (95% confidence interval, 0.787-0.901) for eMS and 0.855 (95% confidence interval, 0.80-0.91) for UCEIS. CONCLUSIONS: We found that a deep learning algorithm can be trained to predict levels of UC severity from full-length endoscopy videos. Our data set was prospectively collected in a multinational clinical trial, videos rather than still images were used, UCEIS and eMS were reported, and machine learning algorithm performance metrics met or exceeded those previously published for UC severity scores.

Artificial intelligence for the assessment of bowel preparation.

BACKGROUND AND AIMS: A reliable assessment of bowel preparation is important to ensure high-quality colonoscopy. Current bowel preparation scoring systems are limited by interobserver variability. This study aimed to demonstrate objective assessment of bowel preparation adequacy using an artificial intelligence (AI)/convolutional neural network (CNN) algorithm developed from colonoscopy videos. METHODS: Two CNNs were developed using a training set of 73,304 images from 200 colonoscopies. First, a binary CNN was developed and trained to distinguish video frames that were appropriate versus inappropriate for scoring with the Boston Bowel Preparation Scale (BBPS). A second multiclass CNN was developed and trained on 26,950 appropriate frames that were expertly annotated with BBPS segment scores (0-3). We validated the algorithm using 252 10-second video clips that were assigned BBPS segment scores by 2 experts. The algorithm provided mean BBPS scores based on the algorithm (AI-BBPS) by calculating mean BBPS based on each frame's scoring. We maximized the algorithm's performance by choosing a dichotomized AI-BBPS score that closely matched dichotomized BBPS scores (ie, adequate vs inadequate). We tested the mean BBPS score based on the algorithm AI-BBPS against human rating using 30 independent 10-second video clips (test set 1) and 10 full withdrawal colonoscopy videos (test set 2). RESULTS: In the validation set, the algorithm demonstrated an area under the curve of .918 and accuracy of 85.3% for detection of inadequate bowel cleanliness. In test set 1, sensitivity for inadequate bowel preparation was 100% and agreement between raters and AI was 76.7% to 83.3%. In test set 2, sensitivity for inadequate bowel preparation for each segment was 100% and agreement between raters and AI was 68.9% to 89.7%. Agreement between raters alone versus raters and AI were similar (κ = .694 and .649, respectively). CONCLUSIONS: The algorithm assessment of bowel cleanliness as measured with the BBPS showed good performance and agreement with experts including full withdrawal colonoscopies.

A survey of colonoscopists with and without in-depth knowledge of water-aided colonoscopy.

BACKGROUND AND AIM: Endoscopy featured water-aided colonoscopy (WAC) as novel in the Innovation Forum in 2011. Gastrointestinal Endoscopy published a modified Delphi consensus review (MDCR) that supports WAC for clinical practice in 2021. We tested the hypothesis that experience was an important predictor of WAC use, either as water immersion (WI), water exchange (WE), or a combination of WI and WE. METHODS: A questionnaire was sent by email to the MDCR authors with an in-depth knowledge of WAC. They responded and also invited colleagues and trainees without in-depth knowledge to respond. Logistic regression analysis was used with the reasons for WAC use treated as the primary outcome. Reports related to WAC post MDCR were identified. RESULTS: Of 100 respondents, > 80% indicated willingness to adopt and modify practice to accommodate WAC. Higher adenoma detection rate (ADR) incentivized WE use. Procedure time slots ≤ 30 and > 30 min significantly predicted WI and WE use, respectively. Co-authors of the MDCR were significantly more likely to perform WAC (odds ratio [OR] = 7.5, P = 0.037). Unfamiliarity with (OR = 0.11, P = 0.02) and absence of good experience (OR = 0.019, P = 0.002) were associated with colonoscopists less likely to perform WAC. Reports related to WAC post MDCR revealed overall and right colon WE outcomes continued to improve. Network meta-analyses showed that WE was superior to Cap and Endocuff. On-demand sedation with WE shortened nursing recovery time. CONCLUSIONS: An important predictor of WAC use was experience. Superior outcomes continued to be reported with WE.

Single-Day Low-Residue Diet Prior to Colonoscopy Demonstrates Improved Bowel Preparation Quality and Patient Tolerance over Clear Liquid Diet: A Randomized, Single-Blinded, Dual-Center Trial.

BACKGROUND AND AIMS: Patients often refer to bowel preparation and associated dietary restrictions as the greatest deterrents to having a colonoscopy completed or performed. Large studies comparing a low-residue diet (LRD) and a clear liquid diet (CLD) are still limited. The aim of this study is to compare LRD and CLD with regard to bowel preparation quality, tolerance, and satisfaction among a diverse patient population. METHODS: This study is a dual-center, randomized, single-blinded, prospective trial involving adult patients undergoing outpatient colonoscopy at the University of California Irvine Medical Center and an affiliated Veterans Administration hospital. Patients were randomized to consume either a CLD or a planned LRD for the full day prior to colonoscopy. Both groups consumed 4L split-dosed PEG-ELS. The adequacy of bowel preparation was evaluated using the Boston Bowel Preparation Score (BBPS). Adequate preparation was defined as a BBPS ≥ 6 with no individual segment less than a score of 2. Hunger and fatigue pre - and post-procedure were graded on a ten-point scale. Nausea, vomiting, bloating, abdominal cramping, overall discomfort, satisfaction with the diet, willingness to repeat the same preparation and overall experience were assessed. RESULTS: A total of 195 subjects who underwent colonoscopy from October 2014 to October 2017 were included. The mean BBPS for the LRD and CLD groups was 8.38 and 7.93, respectively (p = 0.1). There was a significantly higher number of adequate preparations in the LRD group compared to CLD (p = 0.05). Evening hunger scores just before starting the bowel preparation were significantly lower in the LRD than the CLD group, 2.81 versus 5.97, respectively (p = 0.006). Subjects in the LRD group showed significantly less nausea (p = 0.047) and bloating (p = 0.04). Symptom scores for vomiting, abdominal cramping, and overall discomfort were similar between the groups. Satisfaction with diet was significantly higher in the LRD group than CLD, 72% versus 37.66%, respectively (p < 0.001). The overall colonoscopy experience and the satisfaction with the preparation itself were also better reported in the LRD group (p < 0.001 and p = 0.002, respectively). CONCLUSIONS: This study, which included a diverse group of patients, demonstrated that patients using a LRD before colonoscopy achieve a bowel preparation quality that is superior to patients on a CLD restriction. This study shows that a low-residue diet improves patient satisfaction and results in significantly better tolerability of bowel preparation. As a less restrictive dietary regimen, the low-residue diet may help improve patient participation in colorectal cancer screening programs.

Endoscopy and central reading in inflammatory bowel disease clinical trials: achievements, challenges and future developments.

Central reading, that is, independent, off-site, blinded review or reading of imaging endpoints, has been identified as a crucial component in the conduct and analysis of inflammatory bowel disease clinical trials. Central reading is the final step in a workflow that has many parts, all of which can be improved. Furthermore, the best reading algorithm and the most intensive central reader training cannot make up for deficiencies in the acquisition stage (clinical trial endoscopy) or improve on the limitations of the underlying score (outcome instrument). In this review, academic and industry experts review scoring systems, and propose a theoretical framework for central reading that predicts when improvements in statistical power, affecting trial size and chances of success, can be expected: Multireader models can be conceptualised as statistical or non-statistical (social). Important organisational and operational factors, such as training and retraining of readers, optimal bowel preparation for colonoscopy, video quality, optimal or at least acceptable read duration times and other quality control matters, are addressed as well. The theory and practice of central reading and the conduct of endoscopy in clinical trials are interdisciplinary topics that should be of interest to many, regulators, clinical trial experts, gastroenterology societies and those in the academic community who endeavour to develop new scoring systems using traditional and machine learning approaches.

Establishing key research questions for the implementation of artificial intelligence in colonoscopy: a modified Delphi method.

BACKGROUND : Artificial intelligence (AI) research in colonoscopy is progressing rapidly but widespread clinical implementation is not yet a reality. We aimed to identify the top implementation research priorities. METHODS : An established modified Delphi approach for research priority setting was used. Fifteen international experts, including endoscopists and translational computer scientists/engineers, from nine countries participated in an online survey over 9 months. Questions related to AI implementation in colonoscopy were generated as a long-list in the first round, and then scored in two subsequent rounds to identify the top 10 research questions. RESULTS : The top 10 ranked questions were categorized into five themes. Theme 1: clinical trial design/end points (4 questions), related to optimum trial designs for polyp detection and characterization, determining the optimal end points for evaluation of AI, and demonstrating impact on interval cancer rates. Theme 2: technological developments (3 questions), including improving detection of more challenging and advanced lesions, reduction of false-positive rates, and minimizing latency. Theme 3: clinical adoption/integration (1 question), concerning the effective combination of detection and characterization into one workflow. Theme 4: data access/annotation (1 question), concerning more efficient or automated data annotation methods to reduce the burden on human experts. Theme 5: regulatory approval (1 question), related to making regulatory approval processes more efficient. CONCLUSIONS : This is the first reported international research priority setting exercise for AI in colonoscopy. The study findings should be used as a framework to guide future research with key stakeholders to accelerate the clinical implementation of AI in endoscopy.

A Polyp Worth Removing: A Paradigm for Measuring Colonoscopy Quality and Performance of Novel Technologies for Polyp Detection.

Leaving no significant polyp behind while avoiding risks due to unnecessary resections is a commonsense strategy to safely and effectively prevent colorectal cancer (CRC) with colonoscopy. It also alludes to polyps worth removing and, therefore, worth finding. The majority of "worthy" precancerous polyps are adenomas, which for over 2 decades, have received the most attention in performance research and metrics. Consequently, the detection rate of adenomas is currently the only validated, outcome-based measure of colonoscopy demonstrated to correlate with reduced risk of postcolonoscopy CRC. However, a third or more of postcolonoscopy CRCs originate from sessile serrated polyps (SSPs), which are notoriously difficult to find, diagnose and completely resect. Among serrated polyps, the agreement among pathologists differentiating SSPs from non-neoplastic hyperplastic polyps is moderate at best. This lack of ground truth precludes SSPs from consideration in primary metrics of colonoscopy quality or performance of novel polyp detection technologies. By instead leveraging the distinct endoscopic and clinical features of serrated polyps, including those considered important due to proximal location and larger size, clinically significant serrated polyps represent serrated polyps worth removing, enriched with subtle precancerous SSPs. With the explosion of technologies to assist polyp detection, now is the time to broaden benchmarks to include clinically significant serrated polypss alongside adenomas, a measure that is relevant both for assessing the performance of endoscopists, and for assessing new polyp detection technologies.

Oral IBGard™ Before Colonoscopy: A Single-Center Double-Blinded, Randomized, Placebo-Controlled Trial.

BACKGROUND: Peppermint oil is well known to inhibit smooth muscle contractions, and its topical administration during colonoscopy is reported to reduce colonic spasms. AIMS: We aimed to assess whether oral administration of IBGard™, a sustained-release peppermint oil formulation, before colonoscopy reduces spasms and improves adenoma detection rate (ADR).  METHODS: We performed a single-center randomized, double-blinded, placebo-controlled trial. Patients undergoing screening or surveillance colonoscopies were randomized to receive IBGard™ or placebo. The endoscopist graded spasms during insertion, inspection, and polypectomy. Bowel preparation, procedure time, and time of drug administration were documented. Statistical analysis was performed using the Student's t test and Wilcoxon rank-sum test. RESULTS: There was no significant difference in baseline characteristics or dose-timing distribution between IBGard™ and placebo groups. Similarly, there was no difference in ADR (IBGard™ = 47.8%, placebo = 43.1%, p = 0.51), intubation spasm score (1.23 vs 1.2, p = 0.9), withdrawal spasm score (1.3 vs 1.23, p = 0.72), or polypectomy spasm score (0.52 vs 0.46, p = 0.69). Limiting the analysis to patients who received the drug more than 60 min prior to the start of the procedure did not produce any significant differences in these endpoints. CONCLUSIONS: This randomized controlled trial failed to show benefit of orally administered IBGard™ prior to colonoscopy on the presence of colonic spasms or ADR. Because of its low barrier to widespread adoption, the use of appropriately formulated and timed oral peppermint oil warrants further study to determine its efficacy in reducing colonic spasms and improving colonoscopy quality.

Prediction of Polyp Pathology Using Convolutional Neural Networks Achieves "Resect and Discard" Thresholds.

OBJECTIVES: Reliable in situ diagnosis of diminutive (≤5 mm) colorectal polyps could allow for "resect and discard" and "diagnose and leave" strategies, resulting in $1 billion cost savings per year in the United States alone. Current methodologies have failed to consistently meet the Preservation and Incorporation of Valuable endoscopic Innovations (PIVIs) initiative thresholds. Convolutional neural networks (CNNs) have the potential to predict polyp pathology and achieve PIVI thresholds in real time. METHODS: We developed a CNN-based optical pathology (OP) model using Tensorflow and pretrained on ImageNet, capable of operating at 77 frames per second. A total of 6,223 images of unique colorectal polyps of known pathology, location, size, and light source (white light or narrow band imaging [NBI]) underwent 5-fold cross-training (80%) and validation (20%). Separate fresh validation was performed on 634 polyp images. Surveillance intervals were calculated, comparing OP with true pathology. RESULTS: In the original validation set, the negative predictive value for adenomas was 97% among diminutive rectum/rectosigmoid polyps. Results were independent of use of NBI or white light. Surveillance interval concordance comparing OP and true pathology was 93%. In the fresh validation set, the negative predictive value was 97% among diminutive polyps in the rectum and rectosigmoid and surveillance concordance was 94%. DISCUSSION: This study demonstrates the feasibility of in situ diagnosis of colorectal polyps using CNN. Our model exceeds PIVI thresholds for both "resect and discard" and "diagnose and leave" strategies independent of NBI use. Point-of-care adenoma detection rate and surveillance recommendations are potential added benefits.

Traditional serrated adenomas (TSA) and the company they keep: TSA presence predicts advanced neoplasm states.

BACKGROUND: The association of TSAs with metachronous neoplasms is well established and suggests that TSAs would also have an association with synchronous neoplasms. METHODS: We compared odds ratios and rates of synchronous neoplasms found in colonoscopies with and without TSAs. RESULTS: There was a mean of 2.44 neoplasms among TSA cases in comparison with 1.72 in non-TSA cases. The odds ratio for advanced neoplasia was highest among cases with one or more TSAs relative to cases with one or more HPs (7.54 [CI, 4.23-13.44]) when compared with adenomas (1.95 [CI, 1.75-2.17]) and SSPs (2.98 [CI, 2.54-3.5]). CONCLUSIONS: In this study population, there is a 7-fold higher risk of synchronous advanced neoplasms among cases with one or more TSAs.

Deep Learning Localizes and Identifies Polyps in Real Time With 96% Accuracy in Screening Colonoscopy.

BACKGROUND & AIMS: The benefit of colonoscopy for colorectal cancer prevention depends on the adenoma detection rate (ADR). The ADR should reflect the adenoma prevalence rate, which is estimated to be higher than 50% in the screening-age population. However, the ADR by colonoscopists varies from 7% to 53%. It is estimated that every 1% increase in ADR lowers the risk of interval colorectal cancers by 3%-6%. New strategies are needed to increase the ADR during colonoscopy. We tested the ability of computer-assisted image analysis using convolutional neural networks (CNNs; a deep learning model for image analysis) to improve polyp detection, a surrogate of ADR. METHODS: We designed and trained deep CNNs to detect polyps using a diverse and representative set of 8,641 hand-labeled images from screening colonoscopies collected from more than 2000 patients. We tested the models on 20 colonoscopy videos with a total duration of 5 hours. Expert colonoscopists were asked to identify all polyps in 9 de-identified colonoscopy videos, which were selected from archived video studies, with or without benefit of the CNN overlay. Their findings were compared with those of the CNN using CNN-assisted expert review as the reference. RESULTS: When tested on manually labeled images, the CNN identified polyps with an area under the receiver operating characteristic curve of 0.991 and an accuracy of 96.4%. In the analysis of colonoscopy videos in which 28 polyps were removed, 4 expert reviewers identified 8 additional polyps without CNN assistance that had not been removed and identified an additional 17 polyps with CNN assistance (45 in total). All polyps removed and identified by expert review were detected by the CNN. The CNN had a false-positive rate of 7%. CONCLUSION: In a set of 8,641 colonoscopy images containing 4,088 unique polyps, the CNN identified polyps with a cross-validation accuracy of 96.4% and an area under the receiver operating characteristic curve of 0.991. The CNN system detected and localized polyps well within real-time constraints using an ordinary desktop machine with a contemporary graphics processing unit. This system could increase the ADR and decrease interval colorectal cancers but requires validation in large multicenter trials.

Distinct colon mucosa microbiomes associated with tubular adenomas and serrated polyps.

Colorectal cancer is the second most deadly and third most common cancer in the world. Its development is heterogenous, with multiple mechanisms of carcinogenesis. Two distinct mechanisms include the adenoma-carcinoma sequence and the serrated pathway. The gut microbiome has been identified as a key player in the adenoma-carcinoma sequence, but its role in serrated carcinogenesis is less clear. In this study, we characterized the gut microbiome of 140 polyp-free and polyp-bearing individuals using colon mucosa and fecal samples to determine if microbiome composition was associated with each of the two key pathways. We discovered significant differences between the microbiomes of colon mucosa and fecal samples, with sample type explaining 10-15% of the variation observed in the microbiome. Multiple mucosal brushings were collected from each individual to investigate whether the gut microbiome differed between polyp and healthy intestinal tissue, but no differences were found. Mucosal aspirate sampling revealed that the microbiomes of individuals with tubular adenomas and serrated polyps were significantly different from each other and polyp-free individuals, explaining 1-4% of the variance in the microbiome. Microbiome composition also enabled the accurate prediction of subject polyp types using Random Forest, which produced an area under curve values of 0.87-0.99. By directly sampling the colon mucosa and distinguishing between the different developmental pathways of colorectal cancer, our study helps characterize potential mechanistic targets for serrated carcinogenesis. This research also provides insight into multiple microbiome sampling strategies by assessing each method's practicality and effect on microbial community composition.

Microbiome distinctions between the CRC carcinogenic pathways.

Colorectal cancer (CRC) is the third most commonly diagnosed cancer, the third leading cause of cancer-related deaths, and has been on the rise among young adults in the United States. Research has established that the colonic microbiome is different in patients with CRC compared to healthy controls, but few studies have investigated if and how the microbiome may relate to CRC progression through the serrated pathway versus the adenoma-carcinoma sequence.Our view is that progress in CRC microbiome research requires consideration of how the microbiome may contribute to CRC carcinogenesis through the distinct pathways that lead to CRC, which could enable the creation of novel and tailored prevention, screening, and therapeutic interventions. We first highlight the limitations in existing CRC microbiome research and offer corresponding solutions for investigating the microbiome's role in the adenoma-carcinoma sequence and serrated pathway. We then summarize the findings in the select human studies that included data points related to the two major carcinogenic pathways. These studies investigate the microbiome in CRC carcinogenesis and 1) utilize mucosal samples and 2) compare polyps or tumors by histopathologic type, molecular/genetic type, or location in the colon.Key findings from these studies include: 1) Fusobacterium is associated with right-sided, more advanced, and serrated lesions; 2) the colons of people with CRC have bacteria typically associated with normal oral flora; and 3) colons from people with CRC have more biofilms, and these biofilms are predominantly located in the proximal colon (single study).

Artificial Intelligence in Gastrointestinal Endoscopy.

Artificial intelligence (AI) is rapidly integrating into modern technology and clinical practice. Although in its nascency, AI has become a hot topic of investigation for applications in clinical practice. Multiple fields of medicine have embraced the possibility of a future with AI assisting in diagnosis and pathology applications. In the field of gastroenterology, AI has been studied as a tool to assist in risk stratification, diagnosis, and pathologic identification. Specifically, AI has become of great interest in endoscopy as a technology with substantial potential to revolutionize the practice of a modern gastroenterologist. From cancer screening to automated report generation, AI has touched upon all aspects of modern endoscopy. Here, we review landmark AI developments in endoscopy. Starting with broad definitions to develop understanding, we will summarize the current state of AI research and its potential applications. With innovation developing rapidly, this article touches upon the remarkable advances in AI-assisted endoscopy since its initial evaluation at the turn of the millennium, and the potential impact these AI models may have on the modern clinical practice. As with any discussion of new technology, its limitations must also be understood to apply clinical AI tools successfully.

Use of the Endocuff during routine colonoscopy examination improves adenoma detection: A meta-analysis.

AIM: To perform meta-analysis of the use of Endocuff during average risk screening colonoscopy. METHODS: Scopus, Cochrane databases, MEDLINE/PubMed, and CINAHL were searched in April 2016. Abstracts from Digestive Disease Week, United European Gastroenterology, and the American College of Gastroenterology meeting were also searched from 2004-2015. Studies comparing EC-assisted colonoscopy (EAC) to standard colonoscopy, for any indication, were included in the analysis. The analysis was conducted by using the Mantel-Haenszel or DerSimonian and Laird models with the odds ratio (OR) to assess adenoma detection, cecal intubation rate, and complications performed. RESULTS: Nine studies (n = 5624 patients) were included in the analysis. Compared to standard colonoscopy, procedures performed with EC had higher frequencies for adenoma (OR = 1.49, 95%CI: 1.23-1.80; P = 0.03), and sessile serrated adenomas detection (OR = 2.34 95%CI: 1.63-3.36; P < 0.001). There was no significant difference in cecal intubation rates between the EAC group and standard colonoscopy (OR = 1.26, 95%CI: 0.70-2.27, I2 = 0%; P = 0.44). EAC was associated with a higher risk of complications, most commonly being superficial mucosal injury without higher frequency for perforation. CONCLUSION: The use of an EC on colonoscopy appears to improve pre-cancerous polyp detection without any difference in cecal intubation rates compared to standard colonoscopy.

Role of primary tumor resection among chemotherapy-treated patients with synchronous stage IV colorectal cancer: a survival analysis.

INTRODUCTION: The benefit of an operation to remove the primary tumor among patients with synchronous stage IV colorectal cancer is controversial. This study analyzed the survival benefits associated with primary tumor resection among chemotherapy-treated stage IV colorectal cancer patients. METHODS: The study analyzed 11,716 chemotherapy-treated stage IV colorectal cancer patients in the California Cancer Registry between 1996 and 2007, with follow-up through 2009. Patients were stratified into operation and non-operation groups. Estimates of median overall and colorectal cancer-specific survival were generated. RESULTS: Patients undergoing operation compared to those who are not had higher median overall and colorectal cancer-specific survival, 21 versus 10 months (p < 0.0001) and 22 versus 12 months (p < 0.0001), respectively. Patients who were offered surgery but refused had decreased median overall and colorectal cancer-specific survival when compared to patients who underwent resection, 8 versus 21 months (p < 0.001) and 7 versus 22 months (p < 0.001), respectively. In multivariate regression models, patients who underwent resection of primary tumor had improved overall (hazard ratio (HR), 0.42; 95% confidence interval (CI) 0.40-0.44, p < 0.0001) and colorectal cancer-specific survival (HR, 0.43; 95% CI, 0.41-0.45; p < 0.0001). CONCLUSION: Primary tumor resection is associated with improved survival among stage IV chemotherapy-treated colorectal cancer patients.