Barrett Esophagus, Esophageal Adenocarcinoma
Conditions
Brief summary
This clinical trial studies how well minimally invasive approaches (an artificial intelligence \[AI\] powered risk tool, nurse navigation, and a sponge on a string \[SOS\] test) work in diagnosing patients with Barrett's esophagus (BE) and esophageal cancer. Esophageal cancer has a poor 5-year survival rate when diagnosed after onset of symptoms. While rising, incidence of esophageal cancer remains too low to screen the entire population. BE is a condition in which the cells lining the lower part of the esophagus have changed or been replaced with abnormal cells that could lead to esophageal cancer. Currently, patients are screened for BE based on certain risk factors (reflux, age \> 50 years, White race, family history of esophageal cancer, obesity, male sex, and smoking), followed by endoscopies and surgery for treatment. These standard procedures may result in under-recognition of BE risk due to inaccurate and difficult to use risk assessment tools, high cost, invasiveness, low access to endoscopy, and sub-optimal recognition of abnormal cells during routing endoscopy. An AI powered risk tool that integrates symptoms, health history, and laboratory values from electronic health record data may more accurately assess BE and esophageal cancer risk that manual assessment. The BE-SOS screening test combines a swallowable cell collection device with assessment of DNA, which may more accurately diagnose abnormal cells. Nurse navigation involves trained personnel assisting individuals through the screening process and completing the follow-up diagnostic test if the screening test is positive. Navigators address cultural, social, access, and logistical barriers to screening. Nurse navigation may increase completion rates of diagnostic procedures following a positive screening test. These minimally invasive approaches may enable higher rates of BE screening than currently being accomplished.
Interventions
OTHERCancer Screening
Undergo BE-SOS test
Undergo BE risk assessment using AI powered BE Risk Tool
PROCEDUREEndoscopic Procedure
Undergo endoscopy
BEHAVIORALHealth Risk Assessment
Undergo manual BE risk assessment
OTHERNavigation
Receive RN navigation
OTHERQuestionnaire Administration
Ancillary studies
Sponsors
Mayo Clinic
National Cancer Institute (NCI)
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
SCREENING
Masking
NONE
Eligibility
Sex/Gender
ALL
Age
18 Years to 85 Years
Healthy volunteers
Yes
Inclusion criteria
* SPECIFIC AIM 1A: Adult patients 18-85 years old * SPECIFIC AIM 1A INTERVENTION CLUSTERS: BE Risk Tool Score \> 0.087, indicating a higher risk for BE/esophageal adenocarcinoma (EAC). This will be run as a web-based application integrating data from several domains in the electronic health record (EPIC). Output score will range from 0-1. * SPECIFIC AIM 1A CONTROL CLUSTERS: Meeting American College of Gastroenterology (ACG) screening criteria (gastroesophageal reflux disease \[GERD\]+ \> 2 BE risk factors: Age ≥ 50 years, Male sex, Caucasian race, obesity \[body mass index (BMI) \> 30\], ever smoker, family history of BE/EAC) * SPECIFIC AIM 1B: A "low risk" BE risk tool score (\< 0.0897)
Exclusion criteria
* SPECIFIC AIM 1A: History of Barrett's esophagus or esophageal adenocarcinoma * SPECIFIC AIM 1A: Prior endoscopy in the last 10 years * SPECIFIC AIM 1A: Patients who are unable to consent * SPECIFIC AIM 1A: Patients with a current history of uninvestigated dysphagia * SPECIFIC AIM 1A: History of eosinophilic esophagitis, achalasia * SPECIFIC AIM 1A: Patients on oral anticoagulation including Coumadin, Warfarin unless discontinued for five days prior to the sponge procedure * SPECIFIC AIM 1A: Patients on antiplatelet agents including Clopidogrel, unless discontinued for five days prior to the sponge procedure * SPECIFIC AIM 1A: Patients on oral thrombin inhibitors including Dabigatran and oral factor Xa inhibitors such as rivaroxaban, apixaban, and edoxaban, unless discontinued for five days prior to the sponge procedure * SPECIFIC AIM 1A: Patients with a history of known varices or cirrhosis * SPECIFIC AIM 1A: Patients with a history of esophageal or gastric resection for esophageal or gastric carcinoma * SPECIFIC AIM 1A: Patients with congenital or acquired bleeding diatheses * SPECIFIC AIM 1A: Patients with a history of esophageal squamous dysplasia or esophageal squamous carcinoma * SPECIFIC AIM 1A: Patients with limited life expectancy (\< 2 years): per provider judgement * SPECIFIC AIM 1B: History of Barrett's esophagus or esophageal adenocarcinoma * SPECIFIC AIM 1B: Prior endoscopy in the last 10 years * SPECIFIC AIM 1B: Patients who are unable to consent * SPECIFIC AIM 1B: Patients with a current history of uninvestigated dysphagia * SPECIFIC AIM 1B: History of eosinophilic esophagitis, achalasia * SPECIFIC AIM 1B: Patients on oral anticoagulation including Coumadin, Warfarin unless discontinued for five days prior to procedure * SPECIFIC AIM 1B: Patients on antiplatelet agents including Clopidogrel, unless discontinued for five days prior to procedure * SPECIFIC AIM 1B: Patients on oral thrombin inhibitors including Dabigatran and oral factor Xa inhibitors such as rivaroxaban, apixaban, and edoxaban, unless discontinued for five days prior to procedure * SPECIFIC AIM 1B: Patients with a history of known varices or cirrhosis * SPECIFIC AIM 1B: Patients with a history of esophageal or gastric resection for esophageal or gastric carcinoma * SPECIFIC AIM 1B: Patients with congenital or acquired bleeding diatheses * SPECIFIC AIM 1B: Patients with a history of esophageal squamous dysplasia or esophageal squamous carcinoma * SPECIFIC AIM 1B: Patients with limited life expectancy (\< 2 years): per provider judgement
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Rate of completed Barrett's esophagus (BE) screening | Baseline (completion of endoscopy) | Assessed as the proportion of patients who complete BE screening. This is determined as the number of patients who complete a sponge on a string (SOS) BE screening test or endoscopy screen, divided by the number of screen eligible patients in each group. |
| Negative predictive value of the AI powered BE Risk Tool (Aim 1b) | Baseline (completion of endoscopy) | A randomly selected sample of patients (N=50) with a negative BE risk tool score will undergo a research upper endoscopy to confirm the high estimated negative predicted value of the BE risk tool. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| BE screening ordered (completed + not completed) | Baseline (completion of endoscopy) | Assessed as the proportion of patients who complete BE screening (SOS BE screen test or endoscopy) for patients where BE screening has been ordered. |
| Successful completion of the SOS BE screen test | Baseline (completion of endoscopy) | Assessed as the number that had successful test completion (a test result was received) divided by the number of patients where a BE-SOS test/endoscopy order was placed |
| Completion of diagnostic endoscopy following a positive SOS BE screen test | Baseline (completion of endoscopy) | Assessed as the proportion of patients who complete diagnostic endoscopy following a positive BE-SOS test. |
| Rates of BE [with or without dysplasia/esophageal adenocarcinoma (EAC)] | Baseline (completion of endoscopy) | The proportion of patients with BE/EAC diagnoses will be determined as the number with a BE/EAC diagnosis divided by the number of patients who underwent BE-SOS or endoscopy. |
| Patients flagged for screening by the BE risk tool in the intervention clusters who also meet American College of Gastroenterology (ACG) BE screening guidelines | Baseline (completion of endoscopy) | The proportion of screen eligible patients in the intervention arm who meet ACG screening criteria will be determined as the number meeting screening criteria divided by the number in the group. |
Countries
United States
Contacts
CONTACTClinical Trials Referral Office
CONTACTRamona Lansing
PRINCIPAL_INVESTIGATORPrasad G Iyer, MD, MS
Mayo Clinic
Outcome results
None listed