A Pilot Study of a Low-Cost Optical Imaging Tool for Cervical Cancer Screening in Brazil
Overview
- Phase
- Not Applicable
- Intervention
- Wide-Field White Light Imaging
- Conditions
- Cervical Cancer
- Sponsor
- M.D. Anderson Cancer Center
- Enrollment
- 300
- Locations
- 1
- Primary Endpoint
- Area Under the Receiver Operating Characteristic (ROC) Curve for Classifying Cervical Tissue into One of 2 Diagnostic Categories
- Status
- Completed
- Last Updated
- 4 years ago
Overview
Brief Summary
Objectives:
Primary Objective: To perform a pilot clinical study to test multi-modal optical imaging for detection of cervical neoplasia in Brazil.
Secondary Objective: Analyze clinical data to establish the imaging modes which demonstrate the highest degree of correlation with disease state.
Detailed Description
Wide-field and high-resolution images of precancerous cervical lesions and normal uterine cervix will be obtained by imaging 357 patients, recruited from women presenting for colposcopic evaluation from the Prevention Department at Barretos Cancer Hospital, a MDACC Sister Institution, in Barretos, Brazil. Barretos Cancer Hospital will serve as the lead site for the study Initial wide-field white light images will be acquired in unpolarized and cross-polarized modes before application of 3-6% acetic acid. Routine colposcopy involves visual inspection of the cervix for aceto-whitening and will be followed by a second pair of white-light images in each polarization mode. The clinician will identify all sites required for biopsy according to clinical impression, marking each location on the recorded unpolarized white light image. Topical application of 0.01% proflavine solution will then be applied for 1 minute, after which wide-field imaging in fluorescence mode will be performed. The clinician will then select up to 2 additional sites for biopsy based on the appearance of this wide-field fluorescence image, recording the location of each. The high-resolution microendoscope will then be used to acquire images at all sites selected for biopsy, followed by collection of biopsy specimens. The entire imaging procedure should add less than 10 minutes to the standard colposcopy procedure.
Investigators
Eligibility Criteria
Inclusion Criteria
- •Patients undergoing colposcopy
- •18 years of age or older
- •Negative pregnancy test
- •Subjects must be willing and able to provide informed consent
Exclusion Criteria
- •Patients under 18 years of age
- •Positive pregnancy test
- •Subjects unwilling and unable to provide informed consent
Arms & Interventions
Multi-Modal Optical Imaging of Cervix
Initial wide-field white light images acquired of cervix in unpolarized and cross-polarized modes before application of 3-6% acetic acid. The clinician will identify all sites required for biopsy according to clinical impression, marking each location on the recorded unpolarized white light image. Topical application of 0.01% proflavine solution will then be applied for 1 minute, after which wide-field imaging in fluorescence mode will be performed. The clinician will then select up to 2 additional sites for biopsy based on the appearance of this wide-field fluorescence image, recording the location of each. The high-resolution microendoscope will then be used to acquire images at all sites selected for biopsy, followed by collection of biopsy specimens.
Intervention: Wide-Field White Light Imaging
Multi-Modal Optical Imaging of Cervix
Initial wide-field white light images acquired of cervix in unpolarized and cross-polarized modes before application of 3-6% acetic acid. The clinician will identify all sites required for biopsy according to clinical impression, marking each location on the recorded unpolarized white light image. Topical application of 0.01% proflavine solution will then be applied for 1 minute, after which wide-field imaging in fluorescence mode will be performed. The clinician will then select up to 2 additional sites for biopsy based on the appearance of this wide-field fluorescence image, recording the location of each. The high-resolution microendoscope will then be used to acquire images at all sites selected for biopsy, followed by collection of biopsy specimens.
Intervention: Acetic Acid
Multi-Modal Optical Imaging of Cervix
Initial wide-field white light images acquired of cervix in unpolarized and cross-polarized modes before application of 3-6% acetic acid. The clinician will identify all sites required for biopsy according to clinical impression, marking each location on the recorded unpolarized white light image. Topical application of 0.01% proflavine solution will then be applied for 1 minute, after which wide-field imaging in fluorescence mode will be performed. The clinician will then select up to 2 additional sites for biopsy based on the appearance of this wide-field fluorescence image, recording the location of each. The high-resolution microendoscope will then be used to acquire images at all sites selected for biopsy, followed by collection of biopsy specimens.
Intervention: Proflavine Solution
Multi-Modal Optical Imaging of Cervix
Initial wide-field white light images acquired of cervix in unpolarized and cross-polarized modes before application of 3-6% acetic acid. The clinician will identify all sites required for biopsy according to clinical impression, marking each location on the recorded unpolarized white light image. Topical application of 0.01% proflavine solution will then be applied for 1 minute, after which wide-field imaging in fluorescence mode will be performed. The clinician will then select up to 2 additional sites for biopsy based on the appearance of this wide-field fluorescence image, recording the location of each. The high-resolution microendoscope will then be used to acquire images at all sites selected for biopsy, followed by collection of biopsy specimens.
Intervention: Wide-Field Fluorescence Imaging
Outcomes
Primary Outcomes
Area Under the Receiver Operating Characteristic (ROC) Curve for Classifying Cervical Tissue into One of 2 Diagnostic Categories
Time Frame: 1 day
Data used to develop an algorithm to maximize the area under the receiver operating characteristic (ROC) curve for classifying tissue into one of 2 diagnostic categories (high grade squamous intraepithelial lesions; normal or low grade squamous intraepithelial lesions). Descriptive statistics used to summarize the demographic and clinical characteristics of patients by diagnostic category. Descriptive statistics and boxplots used to summarize nuclear size and N/C by diagnostic category. Logistic regression methods used with the training set to model the logit of the probability of being classified as a high grade squamous intraepithelial lesion and to identify the parameters and threshold values which maximize the area under the ROC curve.