JCO Clinical Cancer Informatics最新文献

{"title":"Use of Patient-Reported Outcomes in Risk Prediction Model Development to Support Cancer Care Delivery: A Scoping Review.","authors":"Roshan Paudel, Samira Dias, Carrie G Wade, Christine Cronin, Michael J Hassett","doi":"10.1200/CCI-24-00145","DOIUrl":"10.1200/CCI-24-00145","url":null,"abstract":"<p><strong>Purpose: </strong>The integration of patient-reported outcomes (PROs) into electronic health records (EHRs) has enabled systematic collection of symptom data to manage post-treatment symptoms. The use and integration of PRO data into routine care are associated with overall treatment success, adherence, and satisfaction. Clinical trials have demonstrated the prognostic value of PROs including physical function and global health status in predicting survival. It is unknown to what extent routinely collected PRO data are used in the development of risk prediction models (RPMs) in oncology care. The objective of the scoping review is to assess how PROs are used to train risk RPMs to predict patient outcomes in oncology care.</p><p><strong>Methods: </strong>Using the scoping review methodology outlined in the Joanna Briggs Institute Manual for Evidence Synthesis, we searched four databases (MEDLINE, CINAHL, Embase, and Web of Science) to locate peer-reviewed oncology articles that used PROs as predictors to train models. Study characteristics including settings, clinical outcomes, and model training, testing, validation, and performance data were extracted for analyses.</p><p><strong>Results: </strong>Of the 1,254 studies identified, 18 met inclusion criteria. Most studies performed retrospective analyses of prospectively collected PRO data to build prediction models. Post-treatment survival was the most common outcome predicted. Discriminative performance of models trained using PROs was better than models trained without PROs. Most studies did not report model calibration.</p><p><strong>Conclusion: </strong>Systematic collection of PROs in routine practice provides an opportunity to use patient-reported data to develop RPMs. Model performance improves when PROs are used in combination with other comprehensive data sources.</p>","PeriodicalId":51626,"journal":{"name":"JCO Clinical Cancer Informatics","volume":"8 ","pages":"e2400145"},"PeriodicalIF":3.3,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11534280/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142562529","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Optimizing End Points for Phase III Cancer Trials.","authors":"Steven E Schild","doi":"10.1200/CCI-24-00210","DOIUrl":"https://doi.org/10.1200/CCI-24-00210","url":null,"abstract":"","PeriodicalId":51626,"journal":{"name":"JCO Clinical Cancer Informatics","volume":"8 ","pages":"e2400210"},"PeriodicalIF":3.3,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142591689","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}

{"title":"Patients Facing Large Language Models in Oncology: A Narrative Review.","authors":"Charles Raynaud, David Wu, Jarod Levy, Matteo Marengo, Jean-Emmanuel Bibault","doi":"10.1200/CCI-24-00149","DOIUrl":"https://doi.org/10.1200/CCI-24-00149","url":null,"abstract":"<p><p>The integration of large language models (LLMs) into oncology is transforming patients' journeys through education, diagnosis, treatment monitoring, and follow-up. This review examines the current landscape, potential benefits, and associated ethical and regulatory considerations of the application of LLMs for patients in the oncologic domain.</p>","PeriodicalId":51626,"journal":{"name":"JCO Clinical Cancer Informatics","volume":"8 ","pages":"e2400149"},"PeriodicalIF":3.3,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142607285","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}

{"title":"Bias in Prediction Models to Identify Patients With Colorectal Cancer at High Risk for Readmission After Resection.","authors":"Mary M Lucas, Mario Schootman, Jonathan A Laryea, Sonia T Orcutt, Chenghui Li, Jun Ying, Jennifer A Rumpel, Christopher C Yang","doi":"10.1200/CCI.23.00194","DOIUrl":"10.1200/CCI.23.00194","url":null,"abstract":"<p><strong>Purpose: </strong>Machine learning algorithms are used for predictive modeling in medicine, but studies often do not evaluate or report on the potential biases of the models. Our purpose was to develop clinical prediction models for readmission after surgery in colorectal cancer (CRC) patients and to examine their potential for racial bias.</p><p><strong>Methods: </strong>We used the 2012-2020 American College of Surgeons' National Surgical Quality Improvement Program (ACS-NSQIP) Participant Use File and Targeted Colectomy File. Patients were categorized into four race groups - White, Black or African American, Other, and Unknown/Not Reported. Potential predictive features were identified from studies of risk factors of 30-day readmission in CRC patients. We compared four machine learning-based methods - logistic regression (LR), multilayer perceptron (MLP), random forest (RF), and XGBoost (XGB). Model bias was assessed using false negative rate (FNR) difference, false positive rate (FPR) difference, and disparate impact.</p><p><strong>Results: </strong>In all, 112,077 patients were included, 67.2% of whom were White, 9.2% Black, 5.6% Other race, and 18% with race not recorded. There were significant differences in the AUROC, FPR and FNR between race groups across all models. Notably, patients in the 'Other' race category had higher FNR compared to Black patients in all but the XGB model, while Black patients had higher FPR than White patients in some models. Patients in the 'Other' category consistently had the lowest FPR. Applying the 80% rule for disparate impact, the models consistently met the threshold for unfairness for the 'Other' race category.</p><p><strong>Conclusion: </strong>Predictive models for 30-day readmission after colorectal surgery may perform unequally for different race groups, potentially propagating to inequalities in delivery of care and patient outcomes if the predictions from these models are used to direct care.</p>","PeriodicalId":51626,"journal":{"name":"JCO Clinical Cancer Informatics","volume":"8 ","pages":""},"PeriodicalIF":3.3,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11741203/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143016054","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Classifying Tumor Reportability Status From Unstructured Electronic Pathology Reports Using Language Models in a Population-Based Cancer Registry Setting.","authors":"Lovedeep Gondara, Jonathan Simkin, Gregory Arbour, Shebnum Devji, Raymond Ng","doi":"10.1200/CCI.24.00110","DOIUrl":"10.1200/CCI.24.00110","url":null,"abstract":"<p><strong>Purpose: </strong>Population-based cancer registries (PBCRs) collect data on all new cancer diagnoses in a defined population. Data are sourced from pathology reports, and the PBCRs rely on manual and rule-based solutions. This study presents a state-of-the-art natural language processing (NLP) pipeline, built by fine-tuning pretrained language models (LMs). The pipeline is deployed at the British Columbia Cancer Registry (BCCR) to detect reportable tumors from a population-based feed of electronic pathology.</p><p><strong>Methods: </strong>We fine-tune two publicly available LMs, GatorTron and BlueBERT, which are pretrained on clinical text. Fine-tuning is done using BCCR's pathology reports. For the final decision making, we combine both models' output using an OR approach. The fine-tuning data set consisted of 40,000 reports from the diagnosis year of 2021, and the test data sets consisted of 10,000 reports from the diagnosis year 2021, 20,000 reports from diagnosis year 2022, and 400 reports from diagnosis year 2023.</p><p><strong>Results: </strong>The retrospective evaluation of our proposed approach showed boosted reportable accuracy, maintaining the true reportable threshold of 98%.</p><p><strong>Conclusion: </strong>Disadvantages of rule-based NLP in cancer surveillance include manual effort in rule design and sensitivity to language change. Deep learning approaches demonstrate superior performance in classification. PBCRs distinguish reportability status of incoming electronic cancer pathology reports. Deep learning methods provide significant advantages over rule-based NLP.</p>","PeriodicalId":51626,"journal":{"name":"JCO Clinical Cancer Informatics","volume":"8 ","pages":"e2400110"},"PeriodicalIF":3.3,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11593994/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142677700","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Waiting to Exhale: The Feasibility and Appropriateness of Home Blood Oxygen Monitoring in Oncology Patients Post-Hospital Discharge.","authors":"Si-Yang Liu, Sahil D Doshi, AnnMarie Mazzella Ebstein, Jessie Holland, Ayelet Sapir, Micheal Leung, Jennie Huang, Rosanna Fahy, Rori Salvaggio, Aaron Begue, Gilad Kuperman, Fernanda G C Polubriaginof, Peter D Stetson, Jun J Mao, Katherine Panageas, Bob Li, Bobby Daly","doi":"10.1200/CCI-24-00182","DOIUrl":"10.1200/CCI-24-00182","url":null,"abstract":"<p><strong>Purpose: </strong>Pulse oximetry remote patient monitoring (RPM) post-hospital discharge increased during the COVID-19 pandemic as patients and providers sought to limit in-person encounters and provide more care in the home. However, there is limited evidence on the feasibility and appropriateness of pulse oximetry RPM in patients with cancer after hospital discharge.</p><p><strong>Methods and materials: </strong>This feasibility study enrolled oncology patients discharged after an unexpected admission at the Memorial Sloan Kettering Cancer Center from October 2020 to July 2021. Patients were asked to measure their blood oxygen (O₂) level daily during the hours of 9 am-5 pm during a 10-day monitoring period posthospitalization. An automated system alerted clinicians to blood O₂ levels below 93.0%. We evaluated the feasibility (>50.0% of patients providing at least one measurement from home) and appropriateness (>50.0% of alerts leading to a clinically meaningful patient interaction) of pulse oximetry RPM.</p><p><strong>Results: </strong>Sixty-two patients were enrolled in the study, with 53.2% female patients and a median age of 68 years. The most prevalent malignancy was thoracic (62.9%). The feasibility metric was met, with 45 patients (72.6%, 45 of 62) providing blood O₂ levels at least once during the 10-day monitoring program. The appropriateness threshold was not met; of the 121 alerts, only 39.7% (48 alerts) was linked to a clinically meaningful interaction.</p><p><strong>Conclusion: </strong>This feasibility study showed that while patients with cancer were willing to measure blood O₂ levels at home, most alerts did not result in meaningful clinical interactions. There is a need for improved patient support systems and logistical infrastructure to support appropriate use of RPM at home.</p>","PeriodicalId":51626,"journal":{"name":"JCO Clinical Cancer Informatics","volume":"8 ","pages":"e2400182"},"PeriodicalIF":3.3,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11908708/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142683340","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Rethinking Human Abstraction as the Gold Standard.","authors":"Kirk D Wyatt, Brian T Furner, Samuel L Volchenboum","doi":"10.1200/CCI-24-00218","DOIUrl":"10.1200/CCI-24-00218","url":null,"abstract":"<p><p>@PedsDataCommons discusses automated approaches for data extraction from electronic health records.</p>","PeriodicalId":51626,"journal":{"name":"JCO Clinical Cancer Informatics","volume":"8 ","pages":"e2400218"},"PeriodicalIF":3.3,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142717766","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}

{"title":"Automated Electronic Health Record Data Extraction and Curation Using ExtractEHR.","authors":"Tamara P Miller, Kelly D Getz, Edward Krause, Yun Gun Jo, Sandhya Charapala, M Monica Gramatages, Karen Rabin, Michael E Scheurer, Jennifer J Wilkes, Brian T Fisher, Richard Aplenc","doi":"10.1200/CCI.24.00100","DOIUrl":"10.1200/CCI.24.00100","url":null,"abstract":"<p><strong>Purpose: </strong>Although the potential transformative effect of electronic health record (EHR) data on clinical research in adult patient populations has been very extensively discussed, the effect on pediatric oncology research has been limited. Multiple factors contribute to this more limited effect, including the paucity of pediatric cancer cases in commercial EHR-derived cancer data sets and phenotypic case identification challenges in pediatric federated EHR data.</p><p><strong>Methods: </strong>The ExtractEHR software package was initially developed as a tool to improve clinical trial adverse event reporting but has expanded its use cases to include the development of multisite EHR data sets and the support of cancer cohorts. ExtractEHR enables customized, automated data extraction from the EHR that, when implemented across multiple hospitals, can create pediatric cancer EHR data sets to address a very wide range of research questions in pediatric oncology. After ExtractEHR data acquisition, EHR data can be cleaned and graded using CleanEHR and GradeEHR, companion software packages.</p><p><strong>Results: </strong>ExtractEHR has been installed at four leading pediatric institutions: Children's Healthcare of Atlanta, Children's Hospital of Philadelphia, Texas Children's Hospital, and Seattle Children's Hospital.</p><p><strong>Conclusion: </strong>ExtractEHR has supported multiple use cases, including five clinical epidemiology studies, multicenter clinical trials, and cancer cohort assembly. Work is ongoing to develop Fast Health care Interoperability Resources ExtractEHR and implement other sustainability and scalability enhancements.</p>","PeriodicalId":51626,"journal":{"name":"JCO Clinical Cancer Informatics","volume":"8 ","pages":"e2400100"},"PeriodicalIF":3.3,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11608624/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142717764","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Usage of the National Cancer Institute Cancer Research Data Commons by Researchers: A Scoping Review of the Literature.","authors":"Zhaoyi Chen, Erika Kim, Tanja Davidsen, Jill S Barnholtz-Sloan","doi":"10.1200/CCI.24.00116","DOIUrl":"10.1200/CCI.24.00116","url":null,"abstract":"<p><strong>Purpose: </strong>Over the past decade, significant surges in cancer data of all types have happened. To promote sharing and use of these rich data, the National Cancer Institute's Cancer Research Data Commons (CRDC) was developed as a cloud-based infrastructure that provides a large, comprehensive, and expanding collection of cancer data with tools for analysis. We conducted this scoping review of articles to provide an overview of how CRDC resources are being used by cancer researchers.</p><p><strong>Methods: </strong>A thorough literature search was conducted to identify all relevant publications. We included publications that directly cited CRDC resources to specifically examine the impact and contributions of CRDC by itself. We summarized the distributions and trends of how CRDC components were used by the research community and discussed current research gaps and future opportunities.</p><p><strong>Results: </strong>In terms of CRDC resources used by the research community, encouraging trends in utilization were observed, suggesting that CRDC has become an important building block for fostering a wide range of cancer research. We also noted a few areas where current applications are rather lacking and provided insights on how improvements can be made by CRDC and research community.</p><p><strong>Conclusion: </strong>CRDC, as the foundation of a National Cancer Data Ecosystem, will continue empowering the research community to effectively leverage cancer-related data, uncover novel strategies, and address the needs of patients with cancer, ultimately combatting this disease more effectively.</p>","PeriodicalId":51626,"journal":{"name":"JCO Clinical Cancer Informatics","volume":"8 ","pages":"e2400116"},"PeriodicalIF":3.3,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11575903/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142631619","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Remote Patient Monitoring Using Mobile Health Technology in Cancer Care and Research: Patients' Views and Preferences.","authors":"Dominique G Stuijt, Eva E M van Doeveren, Milan Kos, Marijn Eversdijk, Jacobus J Bosch, Adriaan D Bins, Marieke A R Bak, Martijn G H van Oijen","doi":"10.1200/CCI.24.00092","DOIUrl":"10.1200/CCI.24.00092","url":null,"abstract":"<p><strong>Purpose: </strong>There is an increasing interest in studying the potential of mobile health (mHealth) technologies, such as smartphone apps and wearables, as monitoring tools for patients with cancer during or after their treatment. However, little research is dedicated to exploring the opinions and concerns of patients regarding the adoption of these technologies. This study aimed to gain insight into patients' perspectives and preferences for participating in mHealth-based monitoring in cancer care.</p><p><strong>Methods: </strong>A qualitative study comprising semistructured interviews was conducted in the Netherlands between April and June 2023. Participants were eligible if they were 18 years or older with a current or past diagnosis of cancer. The interview guide was developed on the basis of the technology acceptance model, with main themes being use, communication, trust, privacy, and expectations.</p><p><strong>Results: </strong>Thirteen participants with urologic primary cancer were interviewed. Most patients had already some familiarity with the use of digital monitoring devices or wearables. Main barriers included persistent reminders of the illness, receiving notifications deemed unnecessary or unwanted, and the acknowledgment that mHealth technology does not serve as a substitute for human doctors. Conversely, patients recognized the potential for time-savings through the utilization of mHealth, viewed active monitoring as nonburdensome, considered mHealth a tool for reducing the communication threshold with their doctor, and expressed willingness to adopt such a platform if they perceived personal or societal relevance.</p><p><strong>Conclusion: </strong>This study has elucidated which factors are important for successful development of mHealth for patients with cancer. While both barriers and facilitators play a role, patients' attitudes were positive toward the implementation of remote digital monitoring, showing promising prospects for future research of mHealth in oncology.</p>","PeriodicalId":51626,"journal":{"name":"JCO Clinical Cancer Informatics","volume":"8 ","pages":"e2400092"},"PeriodicalIF":3.3,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11573098/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142631048","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}