Pub Date : 2024-07-03DOI: 10.1101/2024.07.02.24309587
Maximilian Wolf, Julian Lange, Dirk Benndorf, Lina Welz, Susanna Nikolaus, Laura Katharina Sievers, Florian Tran, Kay Schallert, Patrick Hellwig, Stefan Schreiber, Matthias Gunzer, Philip Rosenstiel, Udo Reichl, Konrad Aden, Robert Heyer
Background: The gut microbiome is an important contributor to the development and the course of inflammatory bowel disease (IBD). While changes in the gut microbiome composition were observed in response to IBD therapy using biologics, studies elucidating human and microbial proteins and pathways in dependence on therapy success are sparse. Methods: Fecal samples of a cohort of IBD patients were collected before and after 14 weeks of treatment with three different biologics. Clinical disease activity scores were used to determine the clinical response and remission. Fecal metaproteomes of remitting patients (n=12) and of non-remitting patients (n=12) were compared before treatment and changes within both groups were assessed over sampling time to identify functional changes and potential human and microbial biomarkers. Results: The abundance of proteins associated with the intestinal barrier, neutrophilic granulocytes, and immunoglobulins significantly decreased in remitting patients. In contrast, an increase of those proteins was observed in non-remitting patients. There were significant changes in pathways of microbial metabolism in samples from patients with remission after therapy. This included, for example, an increased abundance of proteins from butyrate fermentation. Finally, new potential biomarkers for the prediction and monitoring of therapy success could be identified, e.g. human lysosome-associated membrane glycoprotein 1, a cytotoxicity marker, or microbial anthranilate synthase component 2, a part of the tryptophan metabolism. Conclusions: Distinct changes of proteins related to gut inflammation and gut microbiome metabolism showed whether IBD remission was achieved or not. This suggests that metaproteomics could be a useful tool for monitoring remission in IBD therapies.
{"title":"Fecal metaproteomics enables functional characterization of remission in patients with inflammatory bowel disease","authors":"Maximilian Wolf, Julian Lange, Dirk Benndorf, Lina Welz, Susanna Nikolaus, Laura Katharina Sievers, Florian Tran, Kay Schallert, Patrick Hellwig, Stefan Schreiber, Matthias Gunzer, Philip Rosenstiel, Udo Reichl, Konrad Aden, Robert Heyer","doi":"10.1101/2024.07.02.24309587","DOIUrl":"https://doi.org/10.1101/2024.07.02.24309587","url":null,"abstract":"Background: The gut microbiome is an important contributor to the development and the course of inflammatory bowel disease (IBD). While changes in the gut microbiome composition were observed in response to IBD therapy using biologics, studies elucidating human and microbial proteins and pathways in dependence on therapy success are sparse. Methods: Fecal samples of a cohort of IBD patients were collected before and after 14 weeks of treatment with three different biologics. Clinical disease activity scores were used to determine the clinical response and remission. Fecal metaproteomes of remitting patients (n=12) and of non-remitting patients (n=12) were compared before treatment and changes within both groups were assessed over sampling time to identify functional changes and potential human and microbial biomarkers. Results: The abundance of proteins associated with the intestinal barrier, neutrophilic granulocytes, and immunoglobulins significantly decreased in remitting patients. In contrast, an increase of those proteins was observed in non-remitting patients. There were significant changes in pathways of microbial metabolism in samples from patients with remission after therapy. This included, for example, an increased abundance of proteins from butyrate fermentation. Finally, new potential biomarkers for the prediction and monitoring of therapy success could be identified, e.g. human lysosome-associated membrane glycoprotein 1, a cytotoxicity marker, or microbial anthranilate synthase component 2, a part of the tryptophan metabolism. Conclusions: Distinct changes of proteins related to gut inflammation and gut microbiome metabolism showed whether IBD remission was achieved or not. This suggests that metaproteomics could be a useful tool for monitoring remission in IBD therapies.","PeriodicalId":501258,"journal":{"name":"medRxiv - Gastroenterology","volume":"24 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141549723","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Background: Recently, the new nomenclature metabolic dysfunction-associated steatohepatitis (MASH) was proposed to supersede non-alcoholic steatohepatitis (NASH). To optimize the management of these patients, it is crucial to comprehend the similarities and differences between individuals with NASH and MASH. Methods: We analyzed data from 13,846 participants in the third National Health and Nutrition Examination Surveys, along with their linked mortality through 2019. NASH and MASH were defined based on respective criteria. Survey-weight adjusted multivariable Cox proportional model was used to examine mortality. Results: The overall prevalence of steatohepatitis, NASH and MASH was 5.7% (n=788), 4.1% (n=564) and 5.5% (n=763), respectively. Most individuals with NASH (96.8%) could be categorized as MASH, but only 69.7% individuals with MASH qualified as NASH. During a median follow-up of 27 years, individuals with MASH exhibited a 53% higher risk of all-cause mortality (adjusted hazard ratio [aHR] 1.53, 95% CI 1.24-1.89). But individuals with NASH did not show an association with all-cause mortality after adjustment for metabolic risk factors (aHR 1.14, 95% CI 0.91-1.44). Notably, individuals who met the criteria for MASH but not NASH (NASH(-)/MASH(+)) had a higher risk of all-cause mortality (aHR 2.47, 95% CI 1.71-3.57) compared to those with NASH(+)/MASH(+) (aHR 1.22, 95% CI 0.97-1.55). Moreover, advanced fibrosis was only associated with an increased risk of all-cause mortality in individuals with MASH, not NASH. Conclusions: MASH, rather than NASH, was associated with an elevated risk of all-cause mortality after adjusting for metabolic risk factors. Well-designed prospective studies are needed to assess and validate our findings.
{"title":"Metabolic dysfunction-associated steatohepatitis is associated with increased all-cause mortality","authors":"Zhao Li, Rui Song, Yingzhi Zhang, Jiahe Tan, Zhiwei Chen","doi":"10.1101/2024.06.28.24309687","DOIUrl":"https://doi.org/10.1101/2024.06.28.24309687","url":null,"abstract":"Background: Recently, the new nomenclature metabolic dysfunction-associated steatohepatitis (MASH) was proposed to supersede non-alcoholic steatohepatitis (NASH). To optimize the management of these patients, it is crucial to comprehend the similarities and differences between individuals with NASH and MASH.\u0000Methods: We analyzed data from 13,846 participants in the third National Health and Nutrition Examination Surveys, along with their linked mortality through 2019. NASH and MASH were defined based on respective criteria. Survey-weight adjusted multivariable Cox proportional model was used to examine mortality.\u0000Results: The overall prevalence of steatohepatitis, NASH and MASH was 5.7% (n=788), 4.1% (n=564) and 5.5% (n=763), respectively. Most individuals with NASH (96.8%) could be categorized as MASH, but only 69.7% individuals with MASH qualified as NASH. During a median follow-up of 27 years, individuals with MASH exhibited a 53% higher risk of all-cause mortality (adjusted hazard ratio [aHR] 1.53, 95% CI 1.24-1.89). But individuals with NASH did not show an association with all-cause mortality after adjustment for metabolic risk factors (aHR 1.14, 95% CI 0.91-1.44). Notably, individuals who met the criteria for MASH but not NASH (NASH(-)/MASH(+)) had a higher risk of all-cause mortality (aHR 2.47, 95% CI 1.71-3.57) compared to those with NASH(+)/MASH(+) (aHR 1.22, 95% CI 0.97-1.55). Moreover, advanced fibrosis was only associated with an increased risk of all-cause mortality in individuals with MASH, not NASH.\u0000Conclusions: MASH, rather than NASH, was associated with an elevated risk of all-cause mortality after adjusting for metabolic risk factors. Well-designed prospective studies are needed to assess and validate our findings.","PeriodicalId":501258,"journal":{"name":"medRxiv - Gastroenterology","volume":"58 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141504106","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-28DOI: 10.1101/2024.06.27.24309484
Friederike Klein, Freya Wellhoener, Anika Freise, Kristina M Niculovic, Howard Junca, Manuel Vicente, Elina Katz, Luiz G dos Anjos Borges, Leonard Knegendorf, Karsten Cirksena, Antonia M Triefenbach, Franziska Woelfl, Helin F Abdullah, Meike Schulz, Iris Plumeier, Silke Kahl, Iris Albers, Martijn Zoodsma, Marius Vital, Torsten Voigtlaender, Henrike Lenzen, Jessica Schmitz, Anna Saborowski, Michael P Manns, Philipp Solbach, Jan H Braesen, Gisa Gerold, Cheng-Jian Xu, Heiner Wedemeyer, Anja K Muenster-Kuehnel, Dietmar H Pieper, Benjamin Heidrich
Primary sclerosing cholangitis (PSC) is an inflammatory disease of the biliary tract eventually leading to bile duct destruction, liver failure, cholangiocellular adenocarcinoma and/or death. No disease modifying treatments are available. Especially cytotoxicity of bile acids, are discussed as potential driver of disease progression. Cholangiocytes are protected by a bicarbonate umbrella formed by the glycocalyx, a dense layer of membrane bound polyglycans extending into the extracellular space. Bile of PSC patients harbors a unique microbiome. Here we identified a new factor in the pathogenesis of PSC. The bacterial degradation of sialic acid and galactose are associated with a poor event free survival of PSC patients and could identify bacterial liberation of sialic acid as crucial element in cholangiocyte damage using cell culture experiments, individualized organoid models and liver biopsies. With this study the view on bacteria-host interactions in bile duct associated diseases is widened. Functional patterns of the bacterial community are crucial for bile duct destruction in PSC patients. This opens a new field of diagnostic tools, disease modifying treatment options and identification of patients at risk.
{"title":"Cholangiocyte glycocalyx degradation boosts primary sclerosing cholangitis","authors":"Friederike Klein, Freya Wellhoener, Anika Freise, Kristina M Niculovic, Howard Junca, Manuel Vicente, Elina Katz, Luiz G dos Anjos Borges, Leonard Knegendorf, Karsten Cirksena, Antonia M Triefenbach, Franziska Woelfl, Helin F Abdullah, Meike Schulz, Iris Plumeier, Silke Kahl, Iris Albers, Martijn Zoodsma, Marius Vital, Torsten Voigtlaender, Henrike Lenzen, Jessica Schmitz, Anna Saborowski, Michael P Manns, Philipp Solbach, Jan H Braesen, Gisa Gerold, Cheng-Jian Xu, Heiner Wedemeyer, Anja K Muenster-Kuehnel, Dietmar H Pieper, Benjamin Heidrich","doi":"10.1101/2024.06.27.24309484","DOIUrl":"https://doi.org/10.1101/2024.06.27.24309484","url":null,"abstract":"Primary sclerosing cholangitis (PSC) is an inflammatory disease of the biliary tract eventually leading to bile duct destruction, liver failure, cholangiocellular adenocarcinoma and/or death. No disease modifying treatments are available. Especially cytotoxicity of bile acids, are discussed as potential driver of disease progression. Cholangiocytes are protected by a bicarbonate umbrella formed by the glycocalyx, a dense layer of membrane bound polyglycans extending into the extracellular space. Bile of PSC patients harbors a unique microbiome. Here we identified a new factor in the pathogenesis of PSC. The bacterial degradation of sialic acid and galactose are associated with a poor event free survival of PSC patients and could identify bacterial liberation of sialic acid as crucial element in cholangiocyte damage using cell culture experiments, individualized organoid models and liver biopsies. With this study the view on bacteria-host interactions in bile duct associated diseases is widened. Functional patterns of the bacterial community are crucial for bile duct destruction in PSC patients. This opens a new field of diagnostic tools, disease modifying treatment options and identification of patients at risk.","PeriodicalId":501258,"journal":{"name":"medRxiv - Gastroenterology","volume":"22 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141519778","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-28DOI: 10.1101/2024.06.28.24309576
Sara Samir Foad Al-Badran, Christopher Bigley, Mark Johnstone, Aula Ammar, Alexander Winton, Jennifer Hay, Jean Quinn, Jakub Jawny, Ditte Andersen, Natalie Fisher, Philip Dunne, Noori Maka, Gerard Lynch, Stephen McSorley, Joanne Edwards, on behalf of the INCISE Collaborative
Objectives Adenomas are known precursors to colorectal cancer (CRC). Current UK post-polypectomy surveillance guidelines use polyp size, numbers, and histology to stratify the risk of patients developing metachronous polyps or CRC. However, these risk guidelines suffer from poor predictive value, often leading to under/over surveillance. Design Adenomas removed from 1257 patients at bowel screening colonoscopy were retrospectively identified to investigate mutational profile and protein expression trends associated with the detection of metachronous polyps or CRC. The presence or absence of metachronous polyps or CRC was recorded 6 months to 6 years after index polypectomy. Results APC and KRAS were the most mutated genes in these patients (87% and 34% respectively), and both were significantly co-occurring with the 6th most mutated gene SOX9 (17% co-occurring with APC, p=0.047; 23% co-occurring with KRAS, p=0.012). High SOX9 cytoplasmic expression was significantly associated with the detection of metachronous polyps or CRC (HR 1.543, p=0.001) and improved high risk stratification when combined with BSG2020 guidelines versus guidelines alone (HR 2.626, p<0.0001). High cytoplasmic SOX9 alone and in combination with current guidelines was an independent predictor of metachronous polyps or CRC according to various regression models. This was validated in an independent test dataset, where high cytoplasmic expression was significantly associated with the detection of metachronous polyps or CRC (HR 1.654, p=0.012) and enhanced risk stratification when combined with BSG2020 guidelines versus guidelines alone (HR 2.473, p=0.0018). Conclusion High cytoplasmic SOX9 expression within adenomas is associated with shorter time to detection of metachronous polyps or CRC.
{"title":"SOX9 Expression in Colorectal Adenomas Improves Surveillance Colonoscopy Risk Stratification in a Bowel Screening Population","authors":"Sara Samir Foad Al-Badran, Christopher Bigley, Mark Johnstone, Aula Ammar, Alexander Winton, Jennifer Hay, Jean Quinn, Jakub Jawny, Ditte Andersen, Natalie Fisher, Philip Dunne, Noori Maka, Gerard Lynch, Stephen McSorley, Joanne Edwards, on behalf of the INCISE Collaborative","doi":"10.1101/2024.06.28.24309576","DOIUrl":"https://doi.org/10.1101/2024.06.28.24309576","url":null,"abstract":"Objectives\u0000Adenomas are known precursors to colorectal cancer (CRC). Current UK post-polypectomy surveillance guidelines use polyp size, numbers, and histology to stratify the risk of patients developing metachronous polyps or CRC. However, these risk guidelines suffer from poor predictive value, often leading to under/over surveillance. Design\u0000Adenomas removed from 1257 patients at bowel screening colonoscopy were retrospectively identified to investigate mutational profile and protein expression trends associated with the detection of metachronous polyps or CRC. The presence or absence of metachronous polyps or CRC was recorded 6 months to 6 years after index polypectomy.\u0000Results\u0000<em>APC</em> and <em>KRAS</em> were the most mutated genes in these patients (87% and 34% respectively), and both were significantly co-occurring with the 6th most mutated gene <em>SOX9</em> (17% co-occurring with <em>APC</em>, p=0.047; 23% co-occurring with <em>KRAS</em>, p=0.012). High SOX9 cytoplasmic expression was significantly associated with the detection of metachronous polyps or CRC (HR 1.543, p=0.001) and improved high risk stratification when combined with BSG2020 guidelines versus guidelines alone (HR 2.626, p<0.0001). High cytoplasmic SOX9 alone and in combination with current guidelines was an independent predictor of metachronous polyps or CRC according to various regression models. This was validated in an independent test dataset, where high cytoplasmic expression was significantly associated with the detection of metachronous polyps or CRC (HR 1.654, p=0.012) and enhanced risk stratification when combined with BSG2020 guidelines versus guidelines alone (HR 2.473, p=0.0018).\u0000Conclusion\u0000High cytoplasmic SOX9 expression within adenomas is associated with shorter time to detection of metachronous polyps or CRC.","PeriodicalId":501258,"journal":{"name":"medRxiv - Gastroenterology","volume":"158 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141504107","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-27DOI: 10.1101/2024.06.26.24309389
Brian Critelli, Amier Hassan, Ila Lahooti, Lydia Noh, Jun Sung Park, Kathleen Tong, Ali Lahooti, Nathan Matzko, Jan Niklas Adams, Lukas Liss, Justin Quion, David Restrepo, Melica Nikahd, Stacey Culp, Adam Lacy-Hulbert, Cate Speake, James Buxbaum, Jason Bischof, Cemal Yazici, Anna Evans-Phillips, Sophie Terp, Alexandra Weissman, Darwin Conwell, Philip Hart, Mitchell Ramsey, Somashekar Krishna, Samuel Han, Erica Park, Raj Shah, Venkata Akshintala, John A Windsor, Nikhil K Mull, Georgios Papachristou, Leo Anthony Celi, Peter Lee
Title: A Systematic Review of Machine Learning-based Prognostic Models for Acute Pancreatitis: Towards Improving Methods and Reporting Quality Background: An accurate prognostic tool is essential to aid clinical decision making (e.g., patient triage) and to advance personalized medicine. However, such prognostic tool is lacking for acute pancreatitis (AP). Increasingly machine learning (ML) techniques are being used to develop high-performing prognostic models in AP. However, methodologic and reporting quality has received little attention. High-quality reporting and study methodology are critical to model validity, reproducibility, and clinical implementation. In collaboration with content experts in ML methodology, we performed a systematic review critically appraising the quality of methodology and reporting of recently published ML AP prognostic models. Methods: Using a validated search strategy, we identified ML AP studies from the databases MEDLINE, PubMed, and EMBASE published between January 2021 and December 2023. Eligibility criteria included all retrospective or prospective studies that developed or validated new or existing ML models in patients with AP that predicted an outcome following an episode of AP. Meta-analysis was considered if there was homogeneity in the study design and in the type of outcome predicted. For risk of bias (ROB) assessment, we used the Prediction Model Risk of Bias Assessment Tool (PROBAST). Quality of reporting was assessed using the Transparent Reporting of a Multivariable Prediction Model of Individual Prognosis or Diagnosis – Artificial Intelligence (TRIPOD+AI) statement that defines standards for 27 items that should be reported in publications using ML prognostic models. Results: The search strategy identified 6480 publications of which 30 met the eligibility criteria. Studies originated from China (22), U.S (4), and other (4). All 30 studies developed a new ML model and none sought to validate an existing ML model, producing a total of 39 new ML models. AP severity (23/39) or mortality (6/39) were the most common outcomes predicted. The mean area-under-the-curve for all models and endpoints was 0.91 (SD 0.08). The ROB was high for at least one domain in all 39 models, particularly for the analysis domain (37/39 models). Steps were not taken to minimize over-optimistic model performance in 27/39 models. Due to heterogeneity in the study design and in how the outcomes were defined and determined, meta-analysis was not performed. Studies reported on only 15/27 items from TRIPOD+AI standards, with only 7/30 justifying sample size and 13/30 assessing data quality. Other reporting deficiencies included omissions regarding human-AI interaction (28/30), handling low-quality or incomplete data in practice (27/30), sharing analytical codes (25/30), study protocols (25/30) and reporting source data (19/30),. Discussion: There are significant deficiencies in the methodology and reporting of recently published ML
标题:基于机器学习的急性胰腺炎预后模型的系统性回顾:提高方法和报告质量 背景:准确的预后工具对于辅助临床决策(如患者分流)和推进个性化医疗至关重要。然而,急性胰腺炎(AP)尚缺乏此类预后工具。越来越多的机器学习(ML)技术被用于开发急性胰腺炎的高效预后模型。然而,方法学和报告质量却很少受到关注。高质量的报告和研究方法对模型的有效性、可重复性和临床实施至关重要。我们与 ML 方法学方面的专家合作,对最近发表的 ML AP 预后模型的方法学和报告质量进行了严格评估。方法:采用经过验证的检索策略,我们从 MEDLINE、PubMed 和 EMBASE 等数据库中确定了 2021 年 1 月至 2023 年 12 月间发表的 ML AP 研究。资格标准包括所有回顾性或前瞻性研究,这些研究针对 AP 患者开发或验证了新的或现有的 ML 模型,可预测 AP 发病后的结果。如果研究设计和预测结果类型具有同质性,则考虑进行荟萃分析。在偏倚风险 (ROB) 评估方面,我们使用了预测模型偏倚风险评估工具 (PROBAST)。报告质量采用 "个体预后或诊断的多变量预测模型的透明报告--人工智能(TRIPOD+AI)声明 "进行评估,该声明定义了在使用多变量预后模型的出版物中应报告的 27 个项目的标准。结果:搜索策略共发现 6480 篇出版物,其中 30 篇符合资格标准。研究分别来自中国(22 项)、美国(4 项)和其他国家(4 项)。所有 30 项研究都开发了一种新的 ML 模型,没有研究试图验证现有的 ML 模型,因此总共产生了 39 个新的 ML 模型。AP 严重程度(23/39)或死亡率(6/39)是最常见的预测结果。所有模型和终点的平均曲线下面积为 0.91(SD 0.08)。在所有 39 个模型中,至少有一个领域的 ROB 很高,尤其是分析领域(37/39 个模型)。在 27/39 个模型中,没有采取措施尽量降低过于乐观的模型性能。由于研究设计以及结果的定义和确定方式存在异质性,因此未进行荟萃分析。研究仅报告了 TRIPOD+AI 标准中的 15/27 个项目,其中仅 7/30 个项目说明了样本量的合理性,13/30 个项目评估了数据质量。其他报告缺陷包括遗漏了人类与人工智能之间的相互作用(28/30)、在实践中处理低质量或不完整数据(27/30)、共享分析代码(25/30)、研究方案(25/30)和报告源数据(19/30)。讨论:最近发表的基于 ML 的 AP 患者预后模型在方法和报告方面存在重大缺陷。尽管这些预后模型有望获得更高的预测准确性,但它们的有效性、可重复性和实施性都受到了影响。资助:无注册:研究注册(Reviewregistry1727)
{"title":"A Systematic Review of Machine Learning-based Prognostic Models for Acute Pancreatitis: Towards Improving Methods and Reporting Quality","authors":"Brian Critelli, Amier Hassan, Ila Lahooti, Lydia Noh, Jun Sung Park, Kathleen Tong, Ali Lahooti, Nathan Matzko, Jan Niklas Adams, Lukas Liss, Justin Quion, David Restrepo, Melica Nikahd, Stacey Culp, Adam Lacy-Hulbert, Cate Speake, James Buxbaum, Jason Bischof, Cemal Yazici, Anna Evans-Phillips, Sophie Terp, Alexandra Weissman, Darwin Conwell, Philip Hart, Mitchell Ramsey, Somashekar Krishna, Samuel Han, Erica Park, Raj Shah, Venkata Akshintala, John A Windsor, Nikhil K Mull, Georgios Papachristou, Leo Anthony Celi, Peter Lee","doi":"10.1101/2024.06.26.24309389","DOIUrl":"https://doi.org/10.1101/2024.06.26.24309389","url":null,"abstract":"Title: A Systematic Review of Machine Learning-based Prognostic Models for Acute Pancreatitis: Towards Improving Methods and Reporting Quality Background: An accurate prognostic tool is essential to aid clinical decision making (e.g., patient triage) and to advance personalized medicine. However, such prognostic tool is lacking for acute pancreatitis (AP). Increasingly machine learning (ML) techniques are being used to develop high-performing prognostic models in AP. However, methodologic and reporting quality has received little attention. High-quality reporting and study methodology are critical to model validity, reproducibility, and clinical implementation. In collaboration with content experts in ML methodology, we performed a systematic review critically appraising the quality of methodology and reporting of recently published ML AP prognostic models. Methods: Using a validated search strategy, we identified ML AP studies from the databases MEDLINE, PubMed, and EMBASE published between January 2021 and December 2023. Eligibility criteria included all retrospective or prospective studies that developed or validated new or existing ML models in patients with AP that predicted an outcome following an episode of AP. Meta-analysis was considered if there was homogeneity in the study design and in the type of outcome predicted. For risk of bias (ROB) assessment, we used the Prediction Model Risk of Bias Assessment Tool (PROBAST). Quality of reporting was assessed using the Transparent Reporting of a Multivariable Prediction Model of Individual Prognosis or Diagnosis – Artificial Intelligence (TRIPOD+AI) statement that defines standards for 27 items that should be reported in publications using ML prognostic models. Results: The search strategy identified 6480 publications of which 30 met the eligibility criteria. Studies originated from China (22), U.S (4), and other (4). All 30 studies developed a new ML model and none sought to validate an existing ML model, producing a total of 39 new ML models. AP severity (23/39) or mortality (6/39) were the most common outcomes predicted. The mean area-under-the-curve for all models and endpoints was 0.91 (SD 0.08). The ROB was high for at least one domain in all 39 models, particularly for the analysis domain (37/39 models). Steps were not taken to minimize over-optimistic model performance in 27/39 models. Due to heterogeneity in the study design and in how the outcomes were defined and determined, meta-analysis was not performed.\u0000Studies reported on only 15/27 items from TRIPOD+AI standards, with only 7/30 justifying sample size and 13/30 assessing data quality. Other reporting deficiencies included omissions regarding human-AI interaction (28/30), handling low-quality or incomplete data in practice (27/30), sharing analytical codes (25/30), study protocols (25/30) and reporting source data (19/30),. Discussion: There are significant deficiencies in the methodology and reporting of recently published ML ","PeriodicalId":501258,"journal":{"name":"medRxiv - Gastroenterology","volume":"12 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141519779","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-27DOI: 10.1101/2024.06.26.24309567
Mahmud Omar, Kassem Sharif, Benjamin S Glicksberg, Girish Nadkarni, Eyal Klang
Background and Aim: In the last two years, natural language processing (NLP) has transformed significantly with the introduction of large language models (LLM). This review updates on NLP and LLM applications and challenges in gastroenterology and hepatology. Methods: Registered with PROSPERO (CRD42024542275) and adhering to PRISMA guidelines, we searched six databases for relevant studies published from 2003 to 2024, ultimately including 57 studies. Results: Our review notes an increase in relevant publications in 2023-2024 compared to previous years, reflecting growing interest in newer models such as GPT-3 and GPT-4. The results demonstrate that NLP models have enhanced data extraction from electronic health records and other unstructured medical data sources. Key findings include high precision in identifying disease characteristics from unstructured reports and ongoing improvement in clinical decision-making. Risk of bias assessments using ROBINS-I, QUADAS-2, and PROBAST tools confirmed the methodological robustness of the included studies. Conclusion: NLP and LLMs can enhance diagnosis and treatment in gastroenterology and hepatology. They enable extraction of data from unstructured medical records, such as endoscopy reports and patient notes, and for enhancing clinical decision-making. Despite these advancements, integrating these tools into routine practice is still challenging. Future work should prospectively demonstrate real-world value. Keywords: Natural Language Processing, Large Language Models, Gastroenterology, Hepatology, Electronic Health Records.
{"title":"Emerging Applications of NLP and Large Language Models in Gastroenterology and Hepatology: A Systematic Review","authors":"Mahmud Omar, Kassem Sharif, Benjamin S Glicksberg, Girish Nadkarni, Eyal Klang","doi":"10.1101/2024.06.26.24309567","DOIUrl":"https://doi.org/10.1101/2024.06.26.24309567","url":null,"abstract":"Background and Aim: In the last two years, natural language processing (NLP) has transformed significantly with the introduction of large language models (LLM). This review updates on NLP and LLM applications and challenges in gastroenterology and hepatology.\u0000Methods: Registered with PROSPERO (CRD42024542275) and adhering to PRISMA guidelines, we searched six databases for relevant studies published from 2003 to 2024, ultimately including 57 studies.\u0000Results: Our review notes an increase in relevant publications in 2023-2024 compared to previous years, reflecting growing interest in newer models such as GPT-3 and GPT-4. The results demonstrate that NLP models have enhanced data extraction from electronic health records and other unstructured medical data sources. Key findings include high precision in identifying disease characteristics from unstructured reports and ongoing improvement in clinical decision-making. Risk of bias assessments using ROBINS-I, QUADAS-2, and PROBAST tools confirmed the methodological robustness of the included studies.\u0000Conclusion: NLP and LLMs can enhance diagnosis and treatment in gastroenterology and hepatology. They enable extraction of data from unstructured medical records, such as endoscopy reports and patient notes, and for enhancing clinical decision-making. Despite these advancements, integrating these tools into routine practice is still challenging. Future work should prospectively demonstrate real-world value.\u0000Keywords: Natural Language Processing, Large Language Models, Gastroenterology, Hepatology, Electronic Health Records.","PeriodicalId":501258,"journal":{"name":"medRxiv - Gastroenterology","volume":"63 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141500534","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-26DOI: 10.1101/2024.06.24.24309401
Nicholas Shaheen, Mohamed Othman, Jawar Taunk, Kenneth J Chang, Sathya Jaganmohan, Patrick S Yachimski, John C Fang, Joseph S Spataro, Suman Verma, Victoria T Lee, Brian J deGuzman, Lishan Aklog
Background and Aims: Barrett's Esophagus (BE) is the precursor to esophageal adenocarcinoma (EAC). We aimed to assess performance, safety, and tolerability of the EsoGuard (EG) assay on samples collected non-endoscopically with the EsoCheck (EC) device (EG/EC) for BE detection in the intended-use population, meeting American College of Gastroenterology (ACG) guideline criteria (chronic gastroesophageal reflux disease (GERD) and 3+ additional risk factors). Methods: We performed a prospective, multicenter study (NCT04293458) to assess EG performance (primary endpoint) on cells collected with EC, for detection of BE and EAC using esophagogastroduodenoscopy (EGD) and biopsies as the comparator. Twenty-four sites across the U.S. and Spain participated. EC safety and usability were assessed as secondary endpoints. Results: 180 male subjects aged >50 years with chronic GERD met eligibility criteria, of which 163 (90.6%) had EGD and successful EC administration. Mean age was 60.5yrs, 34.4% were obese, 56.7% had tobacco history, and 3.9% had a 1st degree relative with BE or EAC. Of 122 samples analyzed, 93 contributed to the primary endpoint analysis. About 9% of subjects in the Primary Analysis Population had BE on EGD, none with dysplasia. Sensitivity of EG for BE was 87.5% (95% CI 47.4-99.7), specificity was 81.2% (95% CI 71.2-88.8), positive predictive value was 30.4% (95% CI 13.2-52.9), and negative predictive value was 98.6% (95% CI 92.3-99.96). Mild esophageal abrasions were observed in 1.5%; no serious adverse events were reported. Conclusions: EG/EC appears effective for BE screening. This approach provides a safe, accurate, and well-tolerated non-endoscopic alternative in high-risk patients.
背景与目的:巴雷特食管(Barrett's Esophagus,BE)是食管腺癌(EAC)的前兆。我们的目的是评估 EsoGuard (EG) 检测法的性能、安全性和耐受性,该检测法是在符合美国胃肠病学院 (ACG) 指南标准(慢性胃食管反流病 (GERD) 和 3 个以上其他风险因素)的预期使用人群中,使用 EsoCheck (EC) 设备(EG/EC)对非内镜采集的样本进行巴雷特食管检测:我们进行了一项前瞻性多中心研究(NCT04293458),以食管胃十二指肠镜检查(EGD)和活组织检查为参照物,评估EG对EC收集的细胞的性能(主要终点),以检测BE和EAC。美国和西班牙的 24 个研究机构参与了这项研究。EC的安全性和可用性作为次要终点进行评估:180名年龄在50岁左右、患有慢性胃食管反流病的男性受试者符合资格标准,其中163人(90.6%)接受了胃食管造影检查并成功实施了EC治疗。平均年龄为 60.5 岁,34.4% 肥胖,56.7% 有吸烟史,3.9% 的一级亲属患有 BE 或 EAC。在分析的 122 个样本中,93 个样本参与了主要终点分析。在主要分析人群中,约有 9% 的受试者在胃肠造影检查中发现了 BE,但没有发现发育不良。EG 对 BE 的敏感性为 87.5%(95% CI 47.4-99.7),特异性为 81.2%(95% CI 71.2-88.8),阳性预测值为 30.4%(95% CI 13.2-52.9),阴性预测值为 98.6%(95% CI 92.3-99.96)。1.5%的患者出现轻度食管擦伤;未报告严重不良事件。结论:EG/EC似乎对BE筛查有效。这种方法为高危患者提供了一种安全、准确、耐受性良好的非内窥镜替代方法。
{"title":"Use of the EsoGuard® Molecular Biomarker Test in Non-Endoscopic Detection of Barrett's Esophagus among High-Risk Individuals in a Screening Population","authors":"Nicholas Shaheen, Mohamed Othman, Jawar Taunk, Kenneth J Chang, Sathya Jaganmohan, Patrick S Yachimski, John C Fang, Joseph S Spataro, Suman Verma, Victoria T Lee, Brian J deGuzman, Lishan Aklog","doi":"10.1101/2024.06.24.24309401","DOIUrl":"https://doi.org/10.1101/2024.06.24.24309401","url":null,"abstract":"Background and Aims: Barrett's Esophagus (BE) is the precursor to esophageal adenocarcinoma (EAC). We aimed to assess performance, safety, and tolerability of the EsoGuard (EG) assay on samples collected non-endoscopically with the EsoCheck (EC) device (EG/EC) for BE detection in the intended-use population, meeting American College of Gastroenterology (ACG) guideline criteria (chronic gastroesophageal reflux disease (GERD) and 3+ additional risk factors).\u0000Methods: We performed a prospective, multicenter study (NCT04293458) to assess EG performance (primary endpoint) on cells collected with EC, for detection of BE and EAC using esophagogastroduodenoscopy (EGD) and biopsies as the comparator. Twenty-four sites across the U.S. and Spain participated. EC safety and usability were assessed as secondary endpoints.\u0000Results: 180 male subjects aged >50 years with chronic GERD met eligibility criteria, of which 163 (90.6%) had EGD and successful EC administration. Mean age was 60.5yrs, 34.4% were obese, 56.7% had tobacco history, and 3.9% had a 1st degree relative with BE or EAC. Of 122 samples analyzed, 93 contributed to the primary endpoint analysis. About 9% of subjects in the Primary Analysis Population had BE on EGD, none with dysplasia. Sensitivity of EG for BE was 87.5% (95% CI 47.4-99.7), specificity was 81.2% (95% CI 71.2-88.8), positive predictive value was 30.4% (95% CI 13.2-52.9), and negative predictive value was 98.6% (95% CI 92.3-99.96). Mild esophageal abrasions were observed in 1.5%; no serious adverse events were reported. Conclusions: EG/EC appears effective for BE screening. This approach provides a safe, accurate, and well-tolerated non-endoscopic alternative in high-risk patients.","PeriodicalId":501258,"journal":{"name":"medRxiv - Gastroenterology","volume":"25 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141504108","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-05-10DOI: 10.1101/2024.05.10.24307195
Femke M. Prins, Iwan J. Hidding, Marjolein A.Y. Klaassen, Valerie Collij, Johannes P.D. Schultheiss, Werna T.C. Uniken Venema, Amber Bangma, Jurne B. Aardema, Bernadien H. Jansen, Wout G.N. Mares, Ben J.M. Witteman, Eleonora A.M. Festen, Gerard Dijkstra, Marijn C. Visschedijk, Herma H. Fidder, Arnau Vich Vila, Bas Oldenburg, Ranko Gacesa, Rinse K. Weersma
Introduction The complexity of Inflammatory Bowel Diseases (IBD) presents challenges for the management of these diseases. Predicting treatment outcomes remains difficult, leading to suboptimal outcomes and high costs. Emerging evidence suggests the potential of the gut microbiome in predicting response to biologic treatments. In this prospective study we aimed to predict treatment response to vedolizumab and ustekinumab in 79 IBD patients by integrating clinical data, gut microbiome profiles and fecal metabolites and validating these findings in a replication cohort of 47 IBD patients.
{"title":"Limited predictive value of the gut microbiome and metabolome for response to biological therapy in inflammatory bowel disease","authors":"Femke M. Prins, Iwan J. Hidding, Marjolein A.Y. Klaassen, Valerie Collij, Johannes P.D. Schultheiss, Werna T.C. Uniken Venema, Amber Bangma, Jurne B. Aardema, Bernadien H. Jansen, Wout G.N. Mares, Ben J.M. Witteman, Eleonora A.M. Festen, Gerard Dijkstra, Marijn C. Visschedijk, Herma H. Fidder, Arnau Vich Vila, Bas Oldenburg, Ranko Gacesa, Rinse K. Weersma","doi":"10.1101/2024.05.10.24307195","DOIUrl":"https://doi.org/10.1101/2024.05.10.24307195","url":null,"abstract":"<strong>Introduction</strong> The complexity of Inflammatory Bowel Diseases (IBD) presents challenges for the management of these diseases. Predicting treatment outcomes remains difficult, leading to suboptimal outcomes and high costs. Emerging evidence suggests the potential of the gut microbiome in predicting response to biologic treatments. In this prospective study we aimed to predict treatment response to vedolizumab and ustekinumab in 79 IBD patients by integrating clinical data, gut microbiome profiles and fecal metabolites and validating these findings in a replication cohort of 47 IBD patients.","PeriodicalId":501258,"journal":{"name":"medRxiv - Gastroenterology","volume":"1 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140930514","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-05-10DOI: 10.1101/2024.05.09.24307139
Joan B Gornals, Albert Sumalla-Garcia, Sergi Quintana, Daniel Luna-Rodriguez, Julio G Velasquez-Rodriguez, Maria Puigcerver-Mas, Julia Escuer, Sandra Maisterra, Mar Marin, Virginia Munoa, Berta Laquente, Juli Busquets
Background Endoscopic ultrasound (EUS)-guided biliary rendezvous (RV) is an EUS-assisted technique described as a rescue method in cases of failed biliary cannulation via endoscopic retrograde cholangiography (ERC). Current literature remains unclear regarding its current role. The study aim was to evaluate the effectiveness for biliary EUS-RV, and comparison between benign vs malignant biliopancreatic disorders.
背景内镜超声(EUS)引导下胆道会合(RV)是一种 EUS 辅助技术,被描述为内镜逆行胆管造影(ERC)胆道插管失败病例的一种抢救方法。目前有关该技术作用的文献仍不明确。本研究旨在评估胆道 EUS-RV 的有效性,并对良性与恶性胆胰疾病进行比较。
{"title":"Endoscopic ultrasound-guided biliary rendezvous after failed cannulation, and comparison between benign vs malignant biliopancreatic disorders: outcomes at a single tertiary-care center","authors":"Joan B Gornals, Albert Sumalla-Garcia, Sergi Quintana, Daniel Luna-Rodriguez, Julio G Velasquez-Rodriguez, Maria Puigcerver-Mas, Julia Escuer, Sandra Maisterra, Mar Marin, Virginia Munoa, Berta Laquente, Juli Busquets","doi":"10.1101/2024.05.09.24307139","DOIUrl":"https://doi.org/10.1101/2024.05.09.24307139","url":null,"abstract":"<strong>Background</strong> Endoscopic ultrasound (EUS)-guided biliary rendezvous (RV) is an EUS-assisted technique described as a rescue method in cases of failed biliary cannulation via endoscopic retrograde cholangiography (ERC). Current literature remains unclear regarding its current role. The study aim was to evaluate the effectiveness for biliary EUS-RV, and comparison between benign vs malignant biliopancreatic disorders.","PeriodicalId":501258,"journal":{"name":"medRxiv - Gastroenterology","volume":"128 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140930432","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-05-09DOI: 10.1101/2024.05.07.24306986
MiaoXin Huang, JunMiao Li, Wei Huang, YuLin Zhou, Lei Cai, Ming Liu
Background Although the use of hot compresses with the herbal medicine Evodia rutaecarpa (ER) as a complementary and alternative therapy to promote recovery of postoperative gastrointestinal function is gradually increasing in clinical practice, there is still a lack of relevant empirical studies. Particularly, the role of ER hot compress therapy on gastrointestinal recovery post-laparoscopic surgery for colorectal cancer has not been well investigated. The purpose of this study is to evaluate the efficacy and applicability of ER hot compress therapy for the recovery of postoperative gastrointestinal function.
{"title":"The effectiveness of Evodia rutaecarpa hot compress on the recovery of gastrointestinal function after laparoscopic surgery for colorectal cancer: A propensity score-matched retrospective cohort study","authors":"MiaoXin Huang, JunMiao Li, Wei Huang, YuLin Zhou, Lei Cai, Ming Liu","doi":"10.1101/2024.05.07.24306986","DOIUrl":"https://doi.org/10.1101/2024.05.07.24306986","url":null,"abstract":"<strong>Background</strong> Although the use of hot compresses with the herbal medicine Evodia rutaecarpa (ER) as a complementary and alternative therapy to promote recovery of postoperative gastrointestinal function is gradually increasing in clinical practice, there is still a lack of relevant empirical studies. Particularly, the role of ER hot compress therapy on gastrointestinal recovery post-laparoscopic surgery for colorectal cancer has not been well investigated. The purpose of this study is to evaluate the efficacy and applicability of ER hot compress therapy for the recovery of postoperative gastrointestinal function.","PeriodicalId":501258,"journal":{"name":"medRxiv - Gastroenterology","volume":"101 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140930627","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}