Anal High Grade Squamous Intraepithelial Lesion
Conditions
Brief summary
The investigators have developed a portable, battery-operated, mobile high-resolution microendoscope (mHRME) that provides subcellular images of the anal epithelium, delineating the cellular and morphologic changes associated with neoplasia. The investigators' central hypothesis is that this 'optical' approach will increase the efficiency, clinical impact, and cost-effectiveness of the current standard of high-resolution anoscopy(HRA)-guided biopsy, thus facilitating usage by less-experienced clinicians in community-based or low-resource settings. To validate this, the investigators will conduct a study to determine the efficiency and diagnostic characteristics of the mHRME 'optical biopsy' approach versus the current standard of HRA-based tissue biopsy. Successful results will allow for improved efficacy and resource utilization for cancer screening in people living with HIV for anal cancer and other epithelial cancers including the cervix, oral cavity, bladder, and GI tract.
Detailed description
The investigators' central hypothesis is that using mHRME plus three-dimensional (3D) mapping as a diagnostic tool will improve the accuracy and efficiency of HSIL diagnoses. Additionally, the investigators hypothesize that the sensitivity (SN) specificity (SP), positive predictive value (PPV), and negative predictive value (NPV), as well as the receiver operating characteristic (ROC) curve for the identification of neoplasia on a per biopsy and per patient basis will be high. The investigators will first compare the HRA-directed biopsy (as the gold standard) to the results of the mHRME HSIL diagnosis. The SN of mHRME diagnosis in the detection of HSIL will be estimated with the binomial proportion of study participants who are positive for HSIL on HRA-guided biopsy at two thresholds of histology thresholds which are: 1) Anal intraepithelial neoplasia (AIN) 2+ threshold, and 2) AIN3+ threshold. SP will be estimated as the proportion of study participants who are negative for HSIL on HRA-guided biopsy at both thresholds. PPV and NPV will be estimated using the binomial proportion and 95% confidence interval (CI). In addition, Cohen's kappa statistic and ROC curves will be generated if patient characteristics such as low Clusters of differentiation 4 (CD4) count, combined antiretroviral treatment (cART) utilization, or high HIV viral load impact the determination of SN and SP. SN and SP of mHRME-based HSIL diagnosis will be estimated on a per lesion and per patient basis with 95% CI and compared by McNemar's test. A generalized linear model for logistic regression with multiple correlated outcomes will compare SN and SP of each method on a per biopsy and per patient basis. Primary Objective: To determine if the mHRME plus 3D mapping improves the accuracy of anal HSIL diagnosis compared to the gold standard of histologic diagnosis of HSIL by HRA-guided biopsy. Secondary Objectives: Determination whether HRME changes the decision to perform biopsy.
Interventions
DIAGNOSTIC_TESTmHRME (Mobile High resolution microendoscope)
Standard of care (SOC) high-resolution anoscopy (HRA) with Lugol's iodine will be performed. The unstained (abnormal) area will be evaluated with mHRME for optical biopsy diagnosis: 1. contrast agent will be applied to anal epithelium (5-10 ml of proflavine hemisulfate (0.01%)), 2. the mHRME will then be inserted and imaging of abnormal tissues will be performed. This will add 2 to 6 minutes per procedure. This is a single-arm study where all subjects will receive both SOC HRA and experimental mHRME imaging.
Contrast agent will be applied to anal epithelium (5-10 ml of proflavine hemisulfate (0.01%)) to use with the mHRME
DIAGNOSTIC_TESTHigh resolution anoscopy
Standard of care (SOC) HRA with Lugol's iodine will be performed.
Sponsors
Icahn School of Medicine at Mount Sinai
William Marsh Rice University
University of California, San Francisco
The University of Texas Health Science Center, Houston
M.D. Anderson Cancer Center
Medical University of South Carolina
Baylor College of Medicine
Study design
Allocation
NA
Intervention model
SINGLE_GROUP
Primary purpose
SCREENING
Masking
NONE
Masking description
No masking will be used in this study.
Intervention model description
This is a single-arm study where all subjects will receive both standard of care HRA (High resolution anoscopy) and experimental mHRME imaging.
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
Yes
Inclusion criteria
* Consentable patients with documented HIV disease * Either: 1) previously documented HSIL or 2) abnormal anal cytology within the past 2 years * Ages 18 years and older * Seen at the Baylor-affiliated Thomas Street Clinic (TSC), Mount Sinai Hospital and affiliated clinics
Exclusion criteria
* Unable to undergo routine anoscopy * Allergy or prior reaction to the fluorescent contrast agent Proflavine or Iodine * Unable to give informed consent * Current or prior history of Invasive Anal Cancer * Known permanent or irreversible bleeding disorder, or other hematologic disorder that in the opinion of the investigator would place the patient at increased risk for adverse outcome from the procedure * Pregnancy
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Performance Characteristics: Sensitivity (SN), Specificity (SP), Positive Predictive Value (PPV) and Negative Predictive Values (NPV) | 1 day | The primary outcome of this study is to measure the operating characteristics including SN, SP, PPV and NPV comparing the physician- and algorithm- guided HRME-based image compared to the Lugol's- guided physician diagnosis of HSIL during HRA. Gold standard consensus pathology was used and pathology data needs to be obtained, verified, and entered. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Procedure Efficiency | 1 day | Diagnostic yield: The number of neoplastic biopsies/total number of biopsies obtained in patients who received biopsies. |
| Procedure Time | 1 day | Total procedure time (HRA+HRME+biopsies) vs HRME time alone |
Countries
United States
Participant flow
Recruitment details
Participants were recruited at two anal dysplasia clinics: Tisch Cancer Institute Mount Sinai in New York, NY, and Thomas Street Health Center in Houston, TX. At Tisch Cancer Institute Mount Sinai, participants were recruited between July 30, 2019, and September 28, 2021. At Thomas Street Health Center, participants were recruited between December 8, 2021, and March 28, 2023. Participants were consented and screened for eligibility at time of presentation to the two recruitment sites.
Participants by arm
| Arm | Count |
|---|---|
| Tisch Cancer Institute Mount Sinai Participants enrolled in the Tisch Cancer Institute Mount Sinai. | 79 |
| Thomas Street Health Center Participants enrolled in the Thomas Street Health Center, excluding the first 18 participants used as test subjects. | 66 |
| Training Set Cohort The first 18 participants enrolled in the Thomas Street Health Center as test subjects. | 18 |
| Total | 163 |
Withdrawals & dropouts
| Period | Reason | FG000 | FG001 | FG002 |
|---|---|---|---|---|
| Follow up | Death | 0 | 2 | 0 |
| Follow up | Lost to Follow-up | 61 | 23 | 5 |
| Follow up | Ongoing | 0 | 28 | 0 |
| HRME Procedure | No HRA and/or HRME data due to technical difficulties or quality control issues with HRME imaging | 11 | 6 | 0 |
| HRME Procedure | Optimization/validation of 3D imaging and HRME | 0 | 0 | 18 |
Baseline characteristics
| Characteristic | Tisch Cancer Institute Mount Sinai | Thomas Street Health Center | Training Set Cohort | Total |
|---|---|---|---|---|
| Age, Continuous | 45.0 years | 48.0 years | 50.5 years | 47.0 years |
| Ethnicity (NIH/OMB) Hispanic or Latino | 34 Participants | 36 Participants | 12 Participants | 82 Participants |
| Ethnicity (NIH/OMB) Not Hispanic or Latino | 45 Participants | 30 Participants | 6 Participants | 81 Participants |
| Ethnicity (NIH/OMB) Unknown or Not Reported | 0 Participants | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) American Indian or Alaska Native | 0 Participants | 1 Participants | 0 Participants | 1 Participants |
| Race (NIH/OMB) Asian | 2 Participants | 2 Participants | 0 Participants | 4 Participants |
| Race (NIH/OMB) Black or African American | 24 Participants | 26 Participants | 4 Participants | 54 Participants |
| Race (NIH/OMB) More than one race | 0 Participants | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) Native Hawaiian or Other Pacific Islander | 8 Participants | 0 Participants | 0 Participants | 8 Participants |
| Race (NIH/OMB) Unknown or Not Reported | 12 Participants | 25 Participants | 8 Participants | 45 Participants |
| Race (NIH/OMB) White | 33 Participants | 12 Participants | 6 Participants | 51 Participants |
| Sex: Female, Male Female | 4 Participants | 5 Participants | 2 Participants | 11 Participants |
| Sex: Female, Male Male | 75 Participants | 61 Participants | 16 Participants | 152 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk | EG002 affected / at risk |
|---|---|---|---|
| deaths Total, all-cause mortality | 0 / 79 | 0 / 66 | 0 / 18 |
| other Total, other adverse events | 0 / 79 | 0 / 66 | 0 / 18 |
| serious Total, serious adverse events | 0 / 79 | 0 / 66 | 0 / 18 |
Outcome results
Primary
Performance Characteristics: Sensitivity (SN), Specificity (SP), Positive Predictive Value (PPV) and Negative Predictive Values (NPV)
The primary outcome of this study is to measure the operating characteristics including SN, SP, PPV and NPV comparing the physician- and algorithm- guided HRME-based image compared to the Lugol's- guided physician diagnosis of HSIL during HRA. Gold standard consensus pathology was used and pathology data needs to be obtained, verified, and entered.
Time frame: 1 day
Population: The training set cohort (n=18) was used to optimize and develop the automated algorithms assessed for primary and secondary outcomes in the actual clinical study. In both the training set cohort and the clinical study, subjects unable to complete the HRME examination or undergo evaluation with the automated algorithm (n=26) were excluded. The remaining 137 evaluable subjects had images assessed by the algorithm and were analyzed separately by study site and the training set cohort.
| Arm | Measure | Group | Value (NUMBER) |
|---|---|---|---|
| Tisch Cancer Institute Mount Sinai | Performance Characteristics: Sensitivity (SN), Specificity (SP), Positive Predictive Value (PPV) and Negative Predictive Values (NPV) | HRA - SN | 90.0 percentage of biopsies |
| Tisch Cancer Institute Mount Sinai | Performance Characteristics: Sensitivity (SN), Specificity (SP), Positive Predictive Value (PPV) and Negative Predictive Values (NPV) | HRA - SP | 65.6 percentage of biopsies |
| Tisch Cancer Institute Mount Sinai | Performance Characteristics: Sensitivity (SN), Specificity (SP), Positive Predictive Value (PPV) and Negative Predictive Values (NPV) | HRA - PPV | 63.2 percentage of biopsies |
| Tisch Cancer Institute Mount Sinai | Performance Characteristics: Sensitivity (SN), Specificity (SP), Positive Predictive Value (PPV) and Negative Predictive Values (NPV) | HRA - NPV | 90.9 percentage of biopsies |
| Tisch Cancer Institute Mount Sinai | Performance Characteristics: Sensitivity (SN), Specificity (SP), Positive Predictive Value (PPV) and Negative Predictive Values (NPV) | HRME - SN | 90.0 percentage of biopsies |
| Tisch Cancer Institute Mount Sinai | Performance Characteristics: Sensitivity (SN), Specificity (SP), Positive Predictive Value (PPV) and Negative Predictive Values (NPV) | HRME - SP | 67.2 percentage of biopsies |
| Tisch Cancer Institute Mount Sinai | Performance Characteristics: Sensitivity (SN), Specificity (SP), Positive Predictive Value (PPV) and Negative Predictive Values (NPV) | HRME - PPV | 64.3 percentage of biopsies |
| Tisch Cancer Institute Mount Sinai | Performance Characteristics: Sensitivity (SN), Specificity (SP), Positive Predictive Value (PPV) and Negative Predictive Values (NPV) | HRME - NPV | 91.1 percentage of biopsies |
| Thomas Street Health Center | Performance Characteristics: Sensitivity (SN), Specificity (SP), Positive Predictive Value (PPV) and Negative Predictive Values (NPV) | HRA - PPV | 27.8 percentage of biopsies |
| Thomas Street Health Center | Performance Characteristics: Sensitivity (SN), Specificity (SP), Positive Predictive Value (PPV) and Negative Predictive Values (NPV) | HRME - PPV | 42.9 percentage of biopsies |
| Thomas Street Health Center | Performance Characteristics: Sensitivity (SN), Specificity (SP), Positive Predictive Value (PPV) and Negative Predictive Values (NPV) | HRA - NPV | 88.9 percentage of biopsies |
| Thomas Street Health Center | Performance Characteristics: Sensitivity (SN), Specificity (SP), Positive Predictive Value (PPV) and Negative Predictive Values (NPV) | HRME - SN | 92.3 percentage of biopsies |
| Thomas Street Health Center | Performance Characteristics: Sensitivity (SN), Specificity (SP), Positive Predictive Value (PPV) and Negative Predictive Values (NPV) | HRME - SP | 68.0 percentage of biopsies |
| Thomas Street Health Center | Performance Characteristics: Sensitivity (SN), Specificity (SP), Positive Predictive Value (PPV) and Negative Predictive Values (NPV) | HRA - SN | 76.9 percentage of biopsies |
| Thomas Street Health Center | Performance Characteristics: Sensitivity (SN), Specificity (SP), Positive Predictive Value (PPV) and Negative Predictive Values (NPV) | HRA - SP | 48.0 percentage of biopsies |
| Thomas Street Health Center | Performance Characteristics: Sensitivity (SN), Specificity (SP), Positive Predictive Value (PPV) and Negative Predictive Values (NPV) | HRME - NPV | 97.1 percentage of biopsies |
| Training Set Cohort | Performance Characteristics: Sensitivity (SN), Specificity (SP), Positive Predictive Value (PPV) and Negative Predictive Values (NPV) | HRA - PPV | 90.0 percentage of biopsies |
| Training Set Cohort | Performance Characteristics: Sensitivity (SN), Specificity (SP), Positive Predictive Value (PPV) and Negative Predictive Values (NPV) | HRA - SP | 90.9 percentage of biopsies |
| Training Set Cohort | Performance Characteristics: Sensitivity (SN), Specificity (SP), Positive Predictive Value (PPV) and Negative Predictive Values (NPV) | HRA - SN | 100.0 percentage of biopsies |
| Training Set Cohort | Performance Characteristics: Sensitivity (SN), Specificity (SP), Positive Predictive Value (PPV) and Negative Predictive Values (NPV) | HRA - NPV | 100.0 percentage of biopsies |
| Training Set Cohort | Performance Characteristics: Sensitivity (SN), Specificity (SP), Positive Predictive Value (PPV) and Negative Predictive Values (NPV) | HRME - PPV | 100.0 percentage of biopsies |
| Training Set Cohort | Performance Characteristics: Sensitivity (SN), Specificity (SP), Positive Predictive Value (PPV) and Negative Predictive Values (NPV) | HRME - SP | 100.0 percentage of biopsies |
| Training Set Cohort | Performance Characteristics: Sensitivity (SN), Specificity (SP), Positive Predictive Value (PPV) and Negative Predictive Values (NPV) | HRME - SN | 100.0 percentage of biopsies |
| Training Set Cohort | Performance Characteristics: Sensitivity (SN), Specificity (SP), Positive Predictive Value (PPV) and Negative Predictive Values (NPV) | HRME - NPV | 100.0 percentage of biopsies |
Secondary
Procedure Efficiency
Diagnostic yield: The number of neoplastic biopsies/total number of biopsies obtained in patients who received biopsies.
Time frame: 1 day
Population: The training set cohort (n=18) was used to optimize and develop the automated algorithms assessed for primary and secondary outcomes in the actual clinical study. In both the training set cohort and the clinical study, subjects unable to complete the HRME examination or undergo evaluation with the automated algorithm (n=26) were excluded. The remaining 137 evaluable subjects had images assessed by the algorithm and were analyzed separately by study site and the training set cohort.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Tisch Cancer Institute Mount Sinai | Procedure Efficiency | 39.6 percentage of biopsies |
| Thomas Street Health Center | Procedure Efficiency | 20.6 percentage of biopsies |
| Training Set Cohort | Procedure Efficiency | 45.0 percentage of biopsies |
Secondary
Procedure Time
Total procedure time (HRA+HRME+biopsies) vs HRME time alone
Time frame: 1 day
Population: The training set cohort (n=18) was used to optimize and develop the automated algorithms assessed for primary and secondary outcomes in the actual clinical study. In both the training set cohort and the clinical study, subjects who did not have both the procedure start and end times recorded were excluded (n=10). The remaining 153 subjects were analyzed separately by study site and the training set cohort.
| Arm | Measure | Group | Value (MEDIAN) |
|---|---|---|---|
| Tisch Cancer Institute Mount Sinai | Procedure Time | Total procedure time | 8 minutes |
| Tisch Cancer Institute Mount Sinai | Procedure Time | HRME time | 2 minutes |
| Thomas Street Health Center | Procedure Time | HRME time | 4 minutes |
| Thomas Street Health Center | Procedure Time | Total procedure time | 12 minutes |
| Training Set Cohort | Procedure Time | HRME time | 5 minutes |
| Training Set Cohort | Procedure Time | Total procedure time | 17 minutes |