Kisiel, John

INTRODUCTION: Colorectal cancer remains a leading cause of cancer associated death in the United States and colonoscopy the primary screening strategy for prevention. Rates of adenomatous and serrated neoplasia detection are inversely associated with postcolonoscopy colorectal cancer. This crucial quality metric depends on accurate ascertainment of colorectal neoplasia findings from both endoscopy and histopathology records. We aimed to assess the feasibility of a random forest machine learning model to rapidly and accurately categorize colorectal neoplasia from electronic health record data. METHODS: A retrospective cohort study compared neoplasia detection rates among individuals undergoing colonoscopy at a large academic institution to develop a rule-based algorithm to categorize colorectal neoplasia from endoscopy reports and pathology systematized nomenclature of medicine – clinical terms. This cohort provided a large training set to develop a natural language processing system using a random forest approach to automatically classify unstructured pathology findings into adenoma, serrated, or advanced neoplasms. This system was manually validated through an independent holdout set. RESULTS: The training set comprised 35,953 unstructured pathology reports with matched systematized nomenclature of medicine – clinical terms from 95,188 unstructured colonoscopy reports. The final model was assessed on an independent holdout set of 337 manually annotated procedures obtaining an area under the receiver operating characteristic curve of 0.997 (confidence interval [CI] 0.994–1), 0.99 (CI 0.98–1), and 0.99 (CI 0.98–0.99) for prediction of adenoma, serrated, and advanced lesions, respectively. DISCUSSION: The random forest-based hybrid natural language processing system for classification of colonoscopy results was both accurate and explainable. NLP combined with effective machine learning algorithms can provide a scalable strategy for colonoscopy quality monitoring.

Abstract Purpose We examined barriers and facilitators to colorectal cancer (CRC) screening among Alaska Native individuals who had never been screened or were not up to date with screening guidelines. Methods As part of a larger study investigating the use of the multitarget stool DNA test in rural Alaska communities, we conducted focus groups and in‐depth interviews with 28 never screened or not up‐to‐date Alaska Native people in two remote communities between November 2022 and July 2023. Participants shared their barriers to CRC screening and offered suggestions to improve programs to better reach those who are guideline discordant. Findings General screening barriers included lack of knowledge, fear of discovering you have cancer, and cultural health beliefs. Colonoscopy‐specific barriers included embarrassment from knowing the local medical staff, having rectal area viewed or touched, fear of pain and injury, difficulty scheduling a procedure due to traditional food gathering activities, lack of provider referrals/reminders, and the high cost of air travel required to get to a colonoscopy facility. Stool DNA‐specific concerns included not feeling that it was as good as colonoscopy, lack of privacy to do the test, and hesitancy collecting stool samples. Suggestions for increasing screening rates included increasing access via paid air transportation, using local indigenous languages, improving provider relationships and reminder systems, and providing CRC and screening education using trusted messengers. Conclusions This was the first exploration of the perceptions of CRC screening among guideline discordant Alaska Native people. Identified themes can be used to improve screening program outreach effectiveness in the future.

Objectives: To describe member adherence to a mail-based, health insurer-sponsored gap closure program for colorectal cancer (CRC) screening using multi-target stool DNA (mt-sDNA; Cologuard ® ) tests. Methods: Combined patient data from Exact Sciences Laboratories LLC and data from mass-mailed mt-sDNA orders placed by a large Medicare Advantage Insurance Plan were analyzed (03/01/2023-06/30/2023). Adherence and time to test return were the primary and secondary outcomes, respectively. Their association with patient characteristics was evaluated using multivariable regression. Results: Among the 3201 member-patients included (86.6% aged 65-75 years; 58.7% female), adherence rate was 49.4%, and mean time to test return was 25.5 days. After multivariable adjustment, the odds of test return were significantly higher among 65- to 75-year-olds (odds ratio [OR] = 1.59 relative to 45- to 64-year-olds), those living in small towns (OR = 1.43 relative to metropolitan-located individuals), and with digital outreach via both SMS and email (OR = 4.31 relative to no digital outreach). Time to test return was shorter in 65- to 75-year-olds than in 45- to 64-year-olds and was not associated with other patient characteristics. Conclusions: Mass-mailed mt-sDNA tests for CRC screening were associated with an overall adherence rate of about 50% in this Medicare Advantage population, with higher likelihood of test return among patients receiving digital outreach.

The Alaska Tribal Health System is working to increase colorectal cancer (CRC) screening among Alaska Native people, who experience the highest CRC rates in the world. This study examined CRC screening provider- and system-level barriers and facilitators from the perspective of healthcare providers serving Alaska Native people in rural/remote communities. A total of 28 provider (physicians, advanced practice, and Community Health Aides/Practitioners) interviews were held from 1 February to 30 November 2021. Colonoscopy provider-level barrier themes included time, competing priorities, and staffing, while system-level barriers included travel costs, weather, and the COVID-19 pandemic. Multi-target stool DNA (mt-sDNA) barrier themes included test viability and unfamiliarity, and previous stool tests experiences. For both tests, limited medical record reminders was a major barrier. Facilitator themes for both tests included community outreach, cultural competency and patient navigation, and clinic/system improvements. In-depth interviews with tribal health providers showed that adding mt-sDNA testing may help address system-level colonoscopy barriers such as waitlists and travel costs, but other barriers remain. Further research is needed into patient barriers and facilitators, as well as the effectiveness of integrating mt-sDNA into a geographically dispersed tribal health system to reduce cancer disparities and build equity in CRC prevention among Alaska Native people.

Colorectal cancer (CRC), although a significant cause of morbidity and mortality worldwide, has seen a declining incidence and mortality in countries with programmatic screening. Faecal occult blood testing and endoscopic approaches are the predominant screening methods currently. The discovery of the adenoma–carcinoma sequence and a greater understanding of the genetic and epigenetic changes that drive the formation of CRC have contributed to innovative research to identify molecular markers for highly accurate, non-invasive screening tests for CRC. DNA, proteins, messenger RNA and micro-RNA have all been evaluated. The observation of tumour cell exfoliation into the mucocellular layer of the colonic epithelium and proven stability of DNA in a harsh stool environment make stool DNA a particularly promising marker. The development of a clinically useful stool DNA test has required numerous technical advances, including optimisation in DNA stabilisation, the development of assays with high analytical sensitivity, and the identification of specific and broadly informative molecular markers. A multitarget stool DNA test, which combines mutant and methylated DNA markers and a faecal immunochemical test, recently performed favourably in a large cross-sectional validation study and has been approved by the US Food and Drug Administration for the screening of asymptomatic, average-risk individuals. The ultimate way in which molecular marker screening assays will be used in clinical practice will require additional studies to determine optimal screening intervals, factors affecting compliance, management of false-positive results, and the use of these assays in high-risk populations, as well as other considerations.