Barrett's Esophagus, Esophageal Adenocarcinoma
Conditions
Keywords
Barrett's esophagus, Esophageal adenocarcinoma, Endoscopy, Confocal laser endomicroscopy
Brief summary
The purpose of this study is to determine if confocal laser endomicroscopy (CLE) can improve detection of Barrett's esophagus, dysplasia, and early esophageal cancer.
Detailed description
Barrett's esophagus is a leading cause of esophageal adenocarcinoma. Detection of dysplasia and early cancers in Barrett's esophagus can be challenging, time-consuming and expensive. Small lesions may be difficult to detect with standard endoscopy protocols. Confocal laser endomicroscopy (CLE) is a new type of endoscopy where a small confocal microscope is built into the tip of a standard endoscope. For this study, we are comparing confocal laser endomicroscopy (CLE) with targeted biopsies with standard endoscopy (EGD)and biopsy for Barrett's esophagus to determine if CLE is more effective for detecting dysplasia and cancer. Participants with Barrett's esophagus in this study undergo 1) CLE with targeted mucosal biopsies (biopsy only taken if CLE shows abnormal tissue) and 2) standard EGD with biopsies. The order of procedures is randomized (some patients have CLE first while others have standard EGD first).
Interventions
Confocal laser endomicroscopy is done by performing standard endoscopy, then using a microscope on the tip of the endoscope to obtain microscopic images of the mucosa. This is done by gently placing the tip of the endoscope on the lining of the esophagus.
DEVICEstandard endoscopy (EGD)
Standard upper endoscopy (EGD) is performed using a regular upper endoscope, which is used to look at the lining of the esophagus.
Sponsors
American Society for Gastrointestinal Endoscopy
Pentax, USA
Johns Hopkins University
Study design
Allocation
RANDOMIZED
Intervention model
CROSSOVER
Primary purpose
DIAGNOSTIC
Masking
NONE
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
No
Inclusion criteria
* Barrett's esophagus or suspected Barrett's-associated neoplasia * Age \> 18 * Able to give informed consent
Exclusion criteria
* Known advanced malignant disease * Allergy to the fluorescent contrast agent fluorescein sodium * Coagulopathy or bleeding disorder
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Diagnostic Yield for Neoplasia in High Risk Patients(Suspected Neoplasia) | 6 weeks | The yield for neoplasia is calculated by the number of biopsies showing neoplasia over the total number of biopsies taken (normal + neoplastic biopsies) |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Mean Number of Biopsies With Neoplasia in High Risk Patients (Suspected Neoplasia) | 6 weeks | The number of biopsies from each procedure (i.e. biopsies taken during CLE, or biopsies taken during standard EGD) that showed neoplasia. Neoplasia is high grade dysplasia or cancer. This analysis looks at patients with Barrett's suspected (but not known) neoplasia. |
| Mean Number of Biopsies Taken in High Risk Patients (Suspected Neoplasia) | 6 weeks | The number of biopsies taken during each procedure (i.e. the number of biopsies taken during CLE, or the number of biopsies taken during standard EGD). Biopsies are taken during CLE only if CLE shows that the esophageal mucosa is abnormal. Esophageal biopsies are taken during standard EGD using a standard Barrett's esophagus protocol (4 quadrants, every 1-2 cm of the Barrett's esophagus). This analysis looks at the Barrett's patients with suspected (but not known) neoplasia. |
| Diagnostic Yield for Neoplasia in Barrett's Surveillance Patients | 6 weeks | Our hypothesis was that the yield for neoplasia would be higher using confocal laser endomicroscopy compared to standard endoscopy. The null hypothesis would be that there is no difference in yield for neoplasia when CLE is used compared to standard endoscopy. This analysis looks specifically at patients who were referred for surveillance of Barrett's esophagus (no suspected neoplasia). |
| Mean Number of Biopsies With Neoplasia in Barrett's Surveillance Patients | 6 weeks | The number of biopsies from each procedure (i.e. biopsies taken during CLE, or biopsies taken during standard EGD) that showed neoplasia. Neoplasia is high grade dysplasia or cancer. This analysis looks at patients undergoing surveillance EGD for Barrett's esophagus (no suspected neoplasia). |
| Mean Number of Biopsies Taken in Barrett's Surveillance Patients | 6 weeks | The number of biopsies taken during each procedure (i.e. the number of biopsies taken during CLE, or the number of biopsies taken during standard EGD). Biopsies are taken during CLE only if CLE shows that the esophageal mucosa is abnormal. Esophageal biopsies are taken during standard EGD using a standard Barrett's esophagus protocol (4 quadrants, every 1-2 cm of the Barrett's esophagus). This analysis looks at the patients with Barrett's esophagus in the study who were undergoing surveillance EGD (no suspected neoplasia). |
Countries
United States
Participant flow
Recruitment details
Recruited from April 2007 to May 2008 from the gastroenterology clinics at Johns Hopkins University.
Pre-assignment details
None of the patients enrolled were excluded prior to group assignment. 1 patient dropped out prior to the first procedure.
Participants by arm
| Arm | Count |
|---|---|
| All Study Participants Patients received both procedures (CLE and standard EGD), so baseline characteristics are reported for the group | 46 |
| Total | 46 |
Withdrawals & dropouts
| Period | Reason | FG000 | FG001 |
|---|---|---|---|
| First Procedure | active wheezing (did not have procedure) | 1 | 0 |
| First Procedure | esophageal stricture found (excl crit) | 1 | 0 |
| First Procedure | invasive cancer on EGD (excl criteria) | 1 | 0 |
| Second Procedure | active wheezing | 0 | 1 |
| Second Procedure | Withdrawal by Subject | 1 | 2 |
Baseline characteristics
| Characteristic | All Study Participants |
|---|---|
| Age, Categorical <=18 years | 0 Participants |
| Age, Categorical >=65 years | 21 Participants |
| Age, Categorical Between 18 and 65 years | 25 Participants |
| Age, Continuous | 63.8 years STANDARD_DEVIATION 9.3 |
| Region of Enrollment United States | 46 participants |
| Sex: Female, Male Female | 12 Participants |
| Sex: Female, Male Male | 34 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk |
|---|---|---|
| deaths Total, all-cause mortality | — / — | — / — |
| other Total, other adverse events | 0 / 39 | 0 / 39 |
| serious Total, serious adverse events | 1 / 39 | 0 / 39 |
Outcome results
Primary
Diagnostic Yield for Neoplasia in High Risk Patients(Suspected Neoplasia)
The yield for neoplasia is calculated by the number of biopsies showing neoplasia over the total number of biopsies taken (normal + neoplastic biopsies)
Time frame: 6 weeks
Population: Analysis was per protocol as the patients who did not undergo CLE could not be analyzed at all (cannot compare 2 procedures when only one (or zero) are performed. This analysis looks specifically at patients with Barrett's and suspected (but not known) neoplasia.
| Arm | Measure | Value (NUMBER) | Dispersion |
|---|---|---|---|
| Confocal Laser Endomicroscopy | Diagnostic Yield for Neoplasia in High Risk Patients(Suspected Neoplasia) | 34 percent yield for neoplasia | 0 |
| Standard Endoscopy | Diagnostic Yield for Neoplasia in High Risk Patients(Suspected Neoplasia) | 17 percent yield for neoplasia | 0 |
Comparison: Our hypothesis was that the yield for neoplasia would be higher using confocal laser endomicroscopy compared to standard endoscopy. The null hypothesis would be that there is no difference in yield for neoplasia when CLE is used compared to standard endoscopy. We estimated that the yield for neoplasia would increase from 10% to 40% using CLE and the calculated sample size was 37. We planned to enroll 48 patients to allow for possible dropouts.p-value: 0.01Wilcoxon signed rank
Secondary
Diagnostic Yield for Neoplasia in Barrett's Surveillance Patients
Our hypothesis was that the yield for neoplasia would be higher using confocal laser endomicroscopy compared to standard endoscopy. The null hypothesis would be that there is no difference in yield for neoplasia when CLE is used compared to standard endoscopy. This analysis looks specifically at patients who were referred for surveillance of Barrett's esophagus (no suspected neoplasia).
Time frame: 6 weeks
Population: Analysis was per protocol as the patients who did not undergo CLE could not be analyzed at all (cannot compare 2 procedures when only one (or zero) are performed).
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Confocal Laser Endomicroscopy | Diagnostic Yield for Neoplasia in Barrett's Surveillance Patients | 0 percent yield for neoplasia |
| Standard Endoscopy | Diagnostic Yield for Neoplasia in Barrett's Surveillance Patients | 0 percent yield for neoplasia |
Secondary
Mean Number of Biopsies Taken in Barrett's Surveillance Patients
The number of biopsies taken during each procedure (i.e. the number of biopsies taken during CLE, or the number of biopsies taken during standard EGD). Biopsies are taken during CLE only if CLE shows that the esophageal mucosa is abnormal. Esophageal biopsies are taken during standard EGD using a standard Barrett's esophagus protocol (4 quadrants, every 1-2 cm of the Barrett's esophagus). This analysis looks at the patients with Barrett's esophagus in the study who were undergoing surveillance EGD (no suspected neoplasia).
Time frame: 6 weeks
Population: Analysis was per protocol. Each participant was compared to self, so if the participant did not complete, the study, no comparison could be made. The patients in this analysis were referred for surveillance of Barrett's esophagus (no suspected neoplasia).
| Arm | Measure | Value (MEAN) |
|---|---|---|
| Confocal Laser Endomicroscopy | Mean Number of Biopsies Taken in Barrett's Surveillance Patients | 1.7 mean number of biopsies |
| Standard Endoscopy | Mean Number of Biopsies Taken in Barrett's Surveillance Patients | 12.6 mean number of biopsies |
Comparison: The null hypothesis would be that CLE does not decrease the number of biopsies needed to make a diagnosis compared to standard endoscopyp-value: <0.0001Wilcoxon signed-rank
Secondary
Mean Number of Biopsies Taken in High Risk Patients (Suspected Neoplasia)
The number of biopsies taken during each procedure (i.e. the number of biopsies taken during CLE, or the number of biopsies taken during standard EGD). Biopsies are taken during CLE only if CLE shows that the esophageal mucosa is abnormal. Esophageal biopsies are taken during standard EGD using a standard Barrett's esophagus protocol (4 quadrants, every 1-2 cm of the Barrett's esophagus). This analysis looks at the Barrett's patients with suspected (but not known) neoplasia.
Time frame: 6 weeks
Population: per protocol as participants would only have data to compare if they completed both endoscopies
| Arm | Measure | Value (MEAN) |
|---|---|---|
| Confocal Laser Endomicroscopy | Mean Number of Biopsies Taken in High Risk Patients (Suspected Neoplasia) | 9.8 mean number of biopsies |
| Standard Endoscopy | Mean Number of Biopsies Taken in High Risk Patients (Suspected Neoplasia) | 23.7 mean number of biopsies |
Comparison: The null hypothesis would be that CLE does not decrease the number of biopsies needed to make a diagnosis compared to standard endoscopyp-value: 0.002wilcoxon signed rank test
Secondary
Mean Number of Biopsies With Neoplasia in Barrett's Surveillance Patients
The number of biopsies from each procedure (i.e. biopsies taken during CLE, or biopsies taken during standard EGD) that showed neoplasia. Neoplasia is high grade dysplasia or cancer. This analysis looks at patients undergoing surveillance EGD for Barrett's esophagus (no suspected neoplasia).
Time frame: 6 weeks
Population: Analysis was per protocol. Each participant was compared to self, so if the participant did not complete the study, no comparison could be made.
| Arm | Measure | Value (MEAN) |
|---|---|---|
| Confocal Laser Endomicroscopy | Mean Number of Biopsies With Neoplasia in Barrett's Surveillance Patients | 0 number of biopsies with neoplasia |
| Standard Endoscopy | Mean Number of Biopsies With Neoplasia in Barrett's Surveillance Patients | 0 number of biopsies with neoplasia |
Comparison: The number of biopsies showing neoplasia was determined for each procedure (confocal laser endomicroscopy, standard EGD). These were compared. This analysis looks at the patients referred for Barrett's surveillance EGD (no suspected neoplasia).p-value: 1Wilcoxon signed-rank
Secondary
Mean Number of Biopsies With Neoplasia in High Risk Patients (Suspected Neoplasia)
The number of biopsies from each procedure (i.e. biopsies taken during CLE, or biopsies taken during standard EGD) that showed neoplasia. Neoplasia is high grade dysplasia or cancer. This analysis looks at patients with Barrett's suspected (but not known) neoplasia.
Time frame: 6 weeks
Population: Analysis was per protocol. Each participant was compared to self, so if the participant did not complete the study, no comparison could be made.
| Arm | Measure | Value (MEAN) |
|---|---|---|
| Confocal Laser Endomicroscopy | Mean Number of Biopsies With Neoplasia in High Risk Patients (Suspected Neoplasia) | 3.1 mean number of biopsies with neoplasia |
| Standard Endoscopy | Mean Number of Biopsies With Neoplasia in High Risk Patients (Suspected Neoplasia) | 3.7 mean number of biopsies with neoplasia |
Comparison: The number of biopsies showing neoplasia was determined for each procedure (confocal laser endomicroscopy, standard EGD). These were compared.p-value: 0.89Wilcoxon signed-rank