MONOPREP PAP TEST
Device Facts
| Record ID | P040052 |
|---|---|
| Device Name | MONOPREP PAP TEST |
| Applicant | Hologic, Inc. |
| Product Code | MKQ · Pathology |
| Decision Date | Mar 3, 2006 |
| Decision | APWD |
| Device Class | Class 3 |
Indications for Use
The MonoPrep Pap Test (hereinafter called MPPT) is intended for use in collecting and preparing cervical-vaginal cytology specimens for Pap stained-based screening for cervical cancer, its precursor lesions and other cytologic categories and conditions defined by The 2001 Bethesda System: terminology for reporting results of cervical cytology(1). The MonoPrep Pap Test produces slides that are intended to replace conventionally prepared Pap smear slides.
Device Story
The MonoPrep Pap Test (MPPT) system is an automated platform for preparing cervical cytology slides. It processes liquid-based cervical cell suspensions collected in a buffered alcohol transport solution. The system uses a processor to uncap vials, stir samples, and aspirate cells through a disposable filter membrane. The processor then deposits the cells onto a glass slide, applies fixative, and registers the slide via barcode. It is operated by laboratory personnel in a clinical setting. The system offers two aspiration control modes: turbidity-based (for normal specimens) and flow-based (for bloody or inflammatory specimens). The resulting slides are analyzed by cytotechnologists and cytopathologists to detect cervical abnormalities. By providing a standardized, thin-layer cell preparation, the device aims to improve the detection of cervical cancer and precursor lesions compared to conventional Pap smears, potentially reducing false negatives and improving diagnostic accuracy.
Clinical Evidence
Prospective, multi-center, masked, split-sample study of 10,739 subjects. Compared MPPT to conventional Pap smear (PS). Primary endpoints: relative detection of cervical abnormalities (ASC-US+, LSIL+, etc.) confirmed by independent pathologist (IP) reference diagnosis. Results showed statistically significant increases in true positive yield for MPPT vs PS for ASC-US+ (1.15), ASC-H/AGC+ (1.23), and LSIL+ (1.26). No significant difference in false positive rates or detection of HSIL+/cancer. Bench testing confirmed cell morphology, stability, and filter transfer efficiency (>95%).
Technological Characteristics
Automated cytology processor; components include collection vial with integrated stirrer, disposable acrylic/frit filter, and fixative solution. Uses turbidity-based or flow-based aspiration control. Data management system interfaces with processor for barcode tracking and parameter selection. Software-controlled processing steps. System processes up to 324 specimens unattended. Materials include acrylic housing and frit-backed filter membrane.
Indications for Use
Indicated for female patients 18 years of age or older requiring cervical-vaginal cytology screening for cervical cancer and precursor lesions. No known contraindications.
Regulatory Classification
Identification
It is a system that is used to collect and prepare cervical cytology specimens for Pap stain-based screening for cervical cancer. The specimens are collected and rinsed into a liquid preservative fluid from which they are then automatically filtered and deposited in a thin-layer on a glass microscope slide. They are stained and screened by a cytotechnologist and cytopathologist.
Submission Summary (Full Text)
{0}
8
# SUMMARY OF SAFETY AND EFFECTIVENESS DATA
## I. GENERAL INFORMATION
Device Generic Name: Liquid-based Cervical Cytology Slide Preparation Device
Device Trade Name: MonoPrep Pap Test (MPPT)
Applicant's Name and Address: MonoGen, Inc.
2461 East Oakton Street
Arlington Heights, IL 60005
Date of Panel Recommendation: None
Premarket Approval Application Number: P040052
Date of Notice of Approval to Applicant: March 3, 2006
## II. INDICATIONS FOR USE
Intended Use
The MonoPrep Pap Test (hereinafter called MPPT) is intended for use in collecting and preparing cervical-vaginal cytology specimens for Pap stained-based screening for cervical cancer, its precursor lesions and other cytologic categories and conditions defined by The 2001 Bethesda System: terminology for reporting results of cervical cytology(1). The MonoPrep Pap Test produces slides that are intended to replace conventionally prepared Pap smear slides.
## III. CONTRAINDICATIONS
There are no known contraindications for use.
## IV. WARNINGS AND PRECAUTIONS
Warnings and precautions for use of the device are stated in the MonoGen MonoPrep Pap Test labeling (Attachment 1).
### DEVICE DESCRIPTION
The MonoGen MonoPrep Pap Test (MPPT) system is a device which converts a liquid suspension of cervical cells into a thin-layer of cells deposited on a glass microscope slide for Papanicolaou staining and analysis. The components of the MPPT system are described below.
{1}
Page 2 of 29 Summary of Safety and Effectiveness Data
## Cervical Specimen Collection Device
An FDA-approved endocervical cytobrush and a plastic cytospatula are provided for use with the MPPT. Break-away-tip collection devices may not be used with the MPPT.
## MonoPrep Pap Test Specimen Collection Vial with Integrated Cap/Stirrer containing MPPT Specimen Transport Solution
The MPPT collection vial has a detachable stirrer with vanes attached to a hollow tube which ensures specimen mixing and dispersal of mucus and loose specimen clumps as well as aspiration of the sample. The MPPT Specimen Transport Solution is a buffered alcohol preservation solution which is tinted for identification purposes.
## MonoPrep Pap Test Filters GYN
The MPPT Filter GYN is a single-use, disposable item designed for gynecological specimens. The filter consists of an acrylic housing and frit-backed filter membrane. The frit supports the filter and facilitates transfer of the sample from the filter to the slide. Fifty MPPT GYN Filters are packaged in a tube which loads directly into the MPPT Processor.
## MonoPrep Pap Test Processor
The MonoPrep Processor is an automated platform for cytology specimen processing consisting of the following stations: loading station; uncapping station; mixing station; filter dispensing station; aspiration station; fixative station; slide printing station; vial resealing station; and slide cassette elevating and holding station. Specimen vials can be loaded directly onto the conveyor belt or in autoloader trays. The processor can hold up to six trays each holding 54 specimen vials and one cleaning vial allowing 324 specimens to be processed unattended in an eight-hour run.
The MonoPrep Processor’s automated specimen processing steps include: vial uncapping; spinning the vial stirrer; dispensing the appropriate filter onto the stirrer manifold; providing aspiration vacuum for the filter; lowering the filter onto the MonoPrep slide; dispensing MonoPrep Fixative Solution onto the prepared slide; replacing the prepared slide and vial back into their respective cassette and tray; and registering the slide barcode with the data management system.
There are two types of processing parameters that can be used for GYN slides, GYN-Normal and GYN-Alternative. These parameters are entered into the laboratory’s data management system which interfaces with the MonoPrep Processor. The decision as to which processing parameters to set for a specimen is selected at vial accessioning and the parameters differ only in the method used to determine the time period for aspirating the specimen through the MPPT filter. Both methods use the same controlled parameters for stir speed, stir time, filter type, number of slides prepared, and number of fixative drops dispensed.
The GYN-Normal (turbidity-based aspiration control) was designed for processing a
9
{2}
Page 3 of 29 Summary of Safety and Effectiveness Data
range of normal specimens. The turbidity measurement estimates the concentration of cells in the specimen and calculates the time period needed to aspirate a sufficient amount of specimen through the filter membrane in order to collect the target number of cells on the filter.
The GYN-Alternative (flow-based aspiration control) is an alternate method for processing unusual specimens that are excessively bloody, have large numbers of inflammatory cells, or that require reprocessing as a result of a previously produced unsatisfactory (UNSAT) slide. In flow-based aspiration, the flow rate across the membrane decreases as the number of cells adhering to the membrane increase and block the filter pores. The aspiration of cells is ended when the reduction in flow rate indicates that the target number of cells has been collected on the filter membrane.
## MonoGen Data Management System
The Savant Data Management System (DMS) is a Laboratory Information System (LIS) that provides basic LIS data storage and transmission functions for the MonoPrep Processor. The DMS/LIS provides specimen vial accessioning with the user entering the vial number by keyboard or standard barcode scanner and then selecting the processing parameter set for the specimen.
## VI. ALTERNATIVE PRACTICES OR PROCEDURES
The conventional Papanicolaou smear (Pap smear) is the original well-established method for screening women for cervical neoplasia or its precursor lesions. It consists of scraping cells from the cervix and manually spreading them onto a glass slide for examination by a cytopathologist. Liquid-based cervical cell collection preparations are an alternative to the Pap smear method. With the liquid-based methods, the cells are scrapped from the cervix, rinsed into the collection fluid vial, and deposited in a thin-layer onto a glass slide for examination by a cytologist. There are two previously approved liquid-based slide preparation methods.
## VII. MARKETING HISTORY
The MonoPrep Pap Test system has not been marketed in the United States or any foreign country.
## VIII. POTENTIAL ADVERSE EFFECTS OF THE DEVICE ON HEALTH
Specimen preparation errors may result in false negative or false positive diagnoses. A false negative diagnosis may result when there are no abnormal cells on the slide when disease is actually present. False negative diagnoses result in delayed diagnosis and treatment for the patient. A false positive diagnosis may result when normal cells appear abnormal due to faulty slide preparation and no disease is present. As a result the patient may have an unnecessary colposcopy exam (a non-invasive procedure) or may be referred for biopsy (an invasive procedure).
3
{3}
Page 4 of 29 Summary of Safety and Effectiveness Data
# IX. SUMMARY OF PRECLINICAL STUDIES
## A. Preclinical Studies
The pre-clinical studies for the MonoGen MonoPrep Pap Test (MPPT) system were designed to assess the (1) MPPT Cell Morphology and Presentation, (2) MPPT Component Selection and Robustness, and (3) MPPT Processing Steps.
1. The Cell Morphology and Presentation series included the following studies: Target Cellularity; Morphological Stability of Specimens, Long Term Preservation; and Specimen Stability Under Varying Conditions.
In the *Target Cellularity* study, epithelial cell nuclear morphology, endocervical component morphology, cell distribution, cell density, thinness of cell layer, presentation of diagnostic material, and abundance of cellular material were evaluated by experienced cytotechnologists and cytopathologists. The results of the study indicated that the slides conformed to the design requirements when prepared in the specified manner. In a random sample of slides from the pivotal clinical study, the number of squamous epithelial cells on a slide ranged from 27,000 to 143,000, in 90% of the slides. The average number of squamous epithelial cells was 60,000 with 95% CI: 42,000~78,000.
The *Morphological Stability of Specimens: Long Term Preservation* study evaluated 34 residual specimens from the clinical trial which were stored at ambient room temperature (15-30°C) for a period of time ranging from 9 to 14.9 months. A second slide (MP2) was prepared and compared with MP1 from the clinical trial. A cytopathologist that participated in the clinical trial read both slide pairs to evaluate them for diagnostic concordance and any other observed differences between the two slide pairs that could affect the diagnosis. The results indicate that for 34 abnormal slide pairs, 26 (76%) received the same diagnosis. In the remaining 8 cases, the diagnoses for the pair differed by one category. In three cases the MP1 diagnosis was higher and in 5 cases, MP2 was higher. This study indicates that for specimens stored at room temperature (15-30°C) for up to a year, the preservation of diagnostic morphology in abnormal cervical cytology specimens mostly permits reproducible diagnosis.
The *Specimen Stability Under Varying Conditions* study was designed to demonstrate that MPPT vials with specimens exposed to boundary (2°C, 30°C) and stress (-20°C, 55°C) conditions does not affect the ability of the GYN-Normal aspiration system to provide satisfactory cell depositions. A LSIL specimen pool was used for the incubation at these testing temperatures and times: 2-8°C overnight; 30°C overnight; -20°C eight hours; and 55°C six hours. Five replicate slides were made from each vial and were reviewed in a masked fashion. It was determined that after exposure to these temperatures there was no material effect on diagnostic morphology or specimen diagnosis.
2. The Component Selection and Robustness series includes the following studies: Filter Quality and Defect Analysis; Vial Label, Slide Mask, and Barcode
{4}
Page 5 of 29 Summary of Safety and Effectiveness Data
Robustness; Stability of Fixative and Specimen Transport Solution; and Specimen Transport Solution Anti-Microbial Effectiveness.
The Filter Quality and Defect Analysis study tested the ability of damaged MPPT filters to produce satisfactory slides and to determine the defects that should be detected in the manufacturing process to keep known deficient filters from commercial distribution. Filters were deliberately damaged before being used to process slides from pooled specimens. The study results showed that all the slides produced were rated satisfactory according to the Bethesda System 2001 criteria, but some of the slides had incomplete cell deposition with low cell densities. The conclusion is that slides with defects such as torn or split membranes, membrane folds or creases, and those with portions of the membrane missing should be rejected, but imperfections such as bumps in the frit, small pin holes, or divots in the material had a negligible impact on slide quality.
The Vial Label, Slide Mask, and Barcode Robustness study tested the labels on the vial and slide to see if they were sufficiently durable to withstand prolonged exposure to the solutions to which they are exposed during processing on the MPPT system. The vial label contains two barcodes, the lot number and expiration date, while the slide contains a barcode and mask. After exposure to MPPT Specimen Transport Solution and cleaner solution, all were successfully displayed and read on the MPPT Processor.
The Stability of Fixative and Specimen Transport Solution study tested the compositional stability of the Specimen Transport Solution, fixative, slides, vials, bags and the seal integrity of the vial and bag configuration under expected and extreme shipping and storage conditions using standard compositional assays. One lot of MPPT-STS vials was tested as capped-sealed and foil-sealed. Testing temperatures were ambient room temperature (ART) (15-30°C); lower limit and cold stress (-20°C, 2-8°C); and upper limit and heat stress (37°C, 55°C). The results of this testing indicated that the alcohol content, ratio, and pH of the MPPT-Specimen Transport Solution remain within specifications for all tested conditions and time-points.
The Specimen Transport Solution turbidity measurements at 55°C for 24 hours showed a 28-61% decrease below the baseline. Holding the Specimen Transport Solution at 55°C for 6 hours produced only a slight decrease in turbidity resulting in a maximum stability claim of 6 hours at 55°C. Thus the stability claims are as follows: Specimen Transport Solution collection vials with and without specimen is 12 months at 15-30°C; 3 weeks at 2-37°C; and 6 hours at -20 to 55°C.
In the Specimen Transport Solution Anti-Microbial Effectiveness study, the STS was tested by an accredited laboratory for anti-microbial effectiveness per the United States Pharmacopoeia 26 methodology. The five microorganisms tested were S. aureus; E. coli; P. aeruginosa; C. albicans; and A. niger. No testing or claims were made for Mycobacterium tuberculosis. The MPPT-STS met the USP requirements for all tested organisms.
3. The MPPT Processing Steps series includes the following studies: Filter to
{5}
Page 6 of 29 Summary of Safety and Effectiveness Data
Slide Transfer Efficiency; Equivalence of Cell Deposition with Two Different Types of Filters; Control of Cellular Cross Contamination; and Potential Interference by Extraneous Materials.
## Filter to Slide Transfer Efficiency
The objective of the filter transfer efficiency study was to validate the number of squamous epithelial cells transferred from a filter to a slide; the efficiency of cell transfer; and the transfer of abnormal cells using filters made with pressed frits and compared with the performance of filters made with the original machined frits. Three pools of LSIL specimens using time-based aspiration were used with the two filter types to make a total of 6 slides. Squamous and abnormal cell numbers on the slides were determined by counting the cells in 10 reticule fields-of-view across the diameter of each cell deposition as recommended by the Bethesda System 2001 method. The filters were removed so that the number of abnormal cells could be counted and the number of squamous epithelial cells estimated. Only the abnormal cells on the filter were directly counted by reviewing the entire filter area using a 20X objective. The results indicate that for these 6 slides, the transfer of cells from the filter to slide was 99% for the cut frits and 97.0% for the pressed frits. Abnormal cells were not preferentially retained on either filter type. Though only a small number of slides were tested, the performance of the two different types of filters appears to be similar.
## Equivalence of Cell Deposition with Two Different Types of Filters
The objective of this study was to use a larger sample size to further confirm the results from the previous study that there is no difference between the two different manufactured types of filters when measuring the transferred cellular material and the numbers of abnormal cells retained on the filters. This study used a sample size of 10 vials of LSIL pools. Two slides were made from each vial using the two filter types yielding two paired slides, one from a cut-frit filter and one from a pressed-frit filter. The transfer efficiency in this study was obtained by counting the numbers of squamous epithelial cells and abnormal cells remaining on the filters as well as using an automated counter, emulating the manual counting recommended by the Bethesda 2001 System, to count the cells on the prepared slides. The results from this study confirmed that for both types of filters, the cell transfer rate was >95% with a negligible difference between the two types of frits, and less than 1% of the total numbers of cells remaining on either type of filter were abnormal thus demonstrating similar performance for both filter types.
The Control of Cellular Cross Contamination study was designed to assess the risk of cellular contamination or carryover from one specimen to the next. In the validation study, two groups of specimens were run sequentially and alternated with blank vials containing only the MPPT-STS collection fluid. Group one consisted of 15 highly cellular abnormal (LSIL or HSIL) specimens and group two consisted of 10 densely cellular pooled NILM specimens. From group one, all test blanks were free of cellular material and from group two, the final test blank contained one highly degenerated cell of undetermined origin. This study demonstrated that processing cervical specimens on the MPPT system is not affected by cellular cross-contamination that can be detected by microscopic examination.
{6}
Page 7 of 29 Summary of Safety and Effectiveness Data
The objective of the Potential Interference by Extraneous Materials study was to see if any patient-introduced foreign materials interfered with the MPPT slide preparation and diagnosis. The eighteen slides in this study were prepared from specimen pools made from patient specimens previously designated as NILM. Foreign materials (douches, antifungal agent, vaginal lubricants, condom fluid and contraceptive foams) and blood were added to the specimen vials. The slides were reviewed by a cytotechnologist to determine if any of the slides were unsatisfactory using the Bethesda System 2001 guidelines or contained abnormal cells. Abnormal slides were further reviewed by a cytopathologist. The slides were also rated on the subjective slide quality metrics of nuclear and cellular morphology; fixation quality and artifact; and staining. Two of the 18 slides were called unsatisfactory due to obscuring blood and five of the slides contained abnormal cells (ASC-US and LSIL). All seven diagnoses were confirmed by the cytopathologist. It was suggested that one or more of the NILM specimens may have contained a small number of abnormal cells. For the slide quality assessment, most of the slides were rated as satisfactory in all categories. The exceptions are two slides that are rated unsatisfactory and two that are rated superior. This study demonstrated that samples containing the potentially-interfering substances yielded slides that were mostly satisfactory for diagnosis.
## B. Additional Studies
Software Verification Test. A software verification test used to test the MonoPrep™ system was submitted by MonoGen, Inc. The software test is based upon incremental phased verification and validation activities for the MonoPrep™ system. The activities consist of methods to (1) verify that the functions provided by the system have been implemented per the specifications; (2) verify the safe operation of the system within its intended use; (3) demonstrate the quality/performance characteristics of the system; and (4) verify the integrity of the data maintained or produced by the system. The Software Reviewer found the software verification test to be adequate.
## SUMMARY OF CLINICAL STUDIES
### A. Clinical Study Design
A prospective, multi-center, masked, split-sample study was conducted in which the objective was to assess MonoPrep Pap Test (MPPT) performance as compared to the conventional Pap smear (PS) for the detection of cervical cancer, pre-cancerous lesions and atypical cells, in subjects representing a spectrum of high, intermediate, and low-risk populations. In addition, an assessment of specimen adequacy, endocervical cells and other analyses was performed. This study used a split-sample design, in which the Pap smear was collected and prepared using FDA-cleared spatula and endocervical cytobrush. The smear residuum remaining on the collection device was then rinsed in the MPPT collection vial which was used to prepare the MPPT slide by the study laboratory. Hence, each case consisted of two slides, one prepared by MPPT and one by PS. MPPT and conventional Pap smear slides were subjected to independent, masked review by the laboratory.
Both MPPT and conventional Pap smear slides of the subjects for whom either the
{7}
Page 8 of 29 Summary of Safety and Effectiveness Data
MPPT or Pap smear slides were diagnosed as Reactive/Reparative or more severe by the study laboratory, and at least 5% of all cases where both slides were diagnosed as NII.M-WNL or UNSAT were submitted to one of the five experts, board-certified cytopathologists for masked independent reference review. The review process was used to establish an independent reference diagnosis for each patient for comparing the clinical performance of MPPT to Pap smears.
## B. Study Sites
Cervical cytology specimens were collected from 11,244 women in the United States (72 sites) as well as South Africa (11 sites) and Venezuela (2 sites). All specimens were then processed at four U.S sites: CYTO Specialty Laboratories, San Antonio, TX, (Sharon Rosenthal, M.D., Principal Investigator); DCL Medical Laboratory, Indianapolis, IN, (Carol Eisenhut, M.D., Principal Investigator); Pathology Services, Cambridge, MA, (Lynda Rushing, M.D., Principal Investigator); and Universal Diagnostic Laboratories, Brooklyn, NY, (Roosevelt Torno, M.D., Principal Investigator).
The study was conducted at four regional study laboratories. Each laboratory was fully accredited, and all study personnel were required to have documented competence with screening Pap smears and liquid-based Pap tests. Each laboratory typically performs at least 100,000 Pap tests per year. Each laboratory was also required to have at least two certified cytotechnologists and at least one board certified cytopathologist to participate in the study.
A total of 11,244 subjects were enrolled in the study. Of these 11,244, the specimens from 339 (3.0%) were received after the study cutoff date and not processed or evaluated. Of 10,905 subjects whose specimens were accepted for processing and evaluation, 121 (1.1%) were excluded from the statistical analysis due to at least one major protocol violation. There were 45 additional cases in which acetic acid was used for the preparation of the MPPT slides; these cases were also excluded from the statistical analysis of effectiveness. The total number of subjects included in the statistical analysis of effectiveness was 10,739.
Table 1 provides study site demographics; laboratory annual Pap smear and liquid-based Pap test volume; and the number of subjects evaluated at each of the four study labs. In nearly all cases, the matching Pap smear and MPPT specimen were sent to the same laboratory.
Table 1. Study Site Demographics
| Site | 1 | 2 | 3 | 4 |
| --- | --- | --- | --- | --- |
| Low Risk Population | 88% | 82% | 90% | 94% |
15
{8}
Page 9 of 29 Summary of Safety and Effectiveness Data
| High Risk Population | 12% | 18% | 10% | 6% |
| --- | --- | --- | --- | --- |
| Smear-Based Pap Tests Per Year | 21,000 | 24,400 | 126,200 | 310,100 |
| Liquid-Based Pap Tests Per Year | 191,700 | 80,700 | 54,200 | 78,300 |
| Number of Cytotechnologists in Study | 8 | 5 | 3 | 3 |
| Number of Cytopathologists in Study | 2 | 2 | 1 | 1 |
| Study Participation Dates | 03/01/04 – 10/28/04 | 03/15/04 – 11/22/04 | 04/02/04 – 11/26/04 | 03/31/04 – 12/10/04 |
| Number of Subjects in Study | 3045 | 2147 | 2119 | 3428 |
## C. Study Population
Women who met the eligibility requirements were enrolled sequentially at each site. The inclusion criteria were female patients 18 years of age; presence of sufficient cervix to obtain a Pap smear; no physician’s contraindication for obtaining a Pap smear; and the ability to provide written informed consent.
Specimens were collected from gynecology medical practices, health clinics, and medical referral centers providing gynecology services to patients representing a spectrum of high to low prevalence populations and diverse ethnic and racial heritage, age and geographical location. These included 75 US and 13 international (11 South African and 2 Venezuelan) collection sites. The following tables present the laboratory and subject information. IRB approved informed consent was obtained from all evaluable subjects. The demographic characteristics of the study population are provided in Table 2 below.
| Subject Demographics | Number | Percent |
| --- | --- | --- |
| U.S. Subjects | 7,689 | 72% |
| International Subjects | 3,050 | 28% |
| Cervical Risk | | |
| High Risk Subjects | 3,513 | 33% |
| Abnormal Pap in previous five years | 1,610 | 15% |
| Race/Ethnic | | |
| White | 5,213 | 49% |
| Hispanic | 2,690 | 25% |
| Black | 1,400 | 13% |
| Other (or not provided) | 1,141 | 11% |
| Asian | 227 | 2.1% |
9
ic
{9}
Page 10 of 29 Summary of Safety and Effectiveness Data
| Indian | 37 | 0.3% |
| --- | --- | --- |
| Pacific | 31 | 0.3% |
| Age | | |
| Mean ± SD | 35.4 ± 12.2 | |
| Range | 18 to 90 | |
## D. Laboratory Cytology Review
Each laboratory had the participation of at least two screening cytotechnologists, at least one quality-control (QC) cytotechnologist, and at least one board-certified cytopathologist. Pap smear and MonoPrep slides were prepared, screened, and interpreted by the participating laboratories’ study cytotechnologists and cytopathologists in the same manner as their routine practice, except in the case of certain protocol procedures intended to maintain consistency across the laboratory sites (e.g., common definition of “high-risk” to be used for selection of cases requiring QC review). All slides were interpreted for the study in accordance with CLIA requirements using TBS2001 nomenclature, including the criteria for a satisfactory slide. All reading of MonoPrep slides was performed independently of Pap smear reviews. Tables 3 and 4 present the comparison of the TBS2001 diagnostic categories for MPPT slides versus conventional Pap smear slides obtained by laboratory cytology review (Lab MPPT vs. Lab PS) for all four sites combined (Table 3) and each site separately (Table 4).
Table 3. Laboratory MPPT Diagnosis vs. Laboratory PS Diagnosis (Combined Sites)
| Lab MPPT Dx | Lab PS Dx | | | | | | | | | | | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| | UNSAT | NILM-WNL | NILM-RR | ASC-US | ASC-H | AGC | LSIL | HSIL | AIS | SCC | AC | Total |
| UNSAT | 43 | 58 | 6 | 12 | | | 5 | 2 | | | | 126 |
| NILM-WNL | 209 | 7,744 | 198 | 459 | 16 | 35 | 55 | 15 | | | 1 | 8,732 |
| NILM-RR | 11 | 214 | 59 | 40 | 1 | 1 | 6 | 2 | | 1 | | 335 |
| ASC-US | 23 | 538 | 41 | 201 | 4 | 7 | 73 | 7 | | | | 894 |
| ASC-H | 1 | 9 | | 10 | | | 2 | 2 | | 1 | | 25 |
| AGC | 4 | 21 | 1 | 4 | 1 | 1 | 1 | | 1 | | | 34 |
| LSIL | 6 | 135 | 1 | 112 | 1 | | 176 | 27 | | 1 | | 459 |
| HSIL | 2 | 4 | | 10 | 7 | 1 | 22 | 50 | | 6 | | 102 |
| AIS | 1 | | | | | | | | 2 | | | 3 |
| SCC | 2 | | | 1 | 4 | | | 5 | | 13 | | 25 |
| AC | | | | | | | | 1 | | 2 | 1 | 4 |
| Total | 302 | 8,723 | 306 | 849 | 34 | 45 | 340 | 111 | 3 | 24 | 2 | 10,739 |
Abbreviation for Diagnoses: UNSAT = Unsatisfactory; NILM-WNL = Negative for Intraepithelial Lesions or Malignancy, Within Normal Limits; NILM-RR = Negative for Intraepithelial Lesions or Malignancy, Reparative/Reactive; ASC-US = Atypical Squamous Cells of Undetermined Significance; ASC-H = Atypical Squamous Cells, cannot exclude HSIL; AGC = Atypical Glandular Cells; LSIL = Low-grade Squamous Intraepithelial Lesion; HSIL = High-grade Squamous Intraepithelial
17
{10}
Page 11 of 29 Summary of Safety and Effectiveness Data
Lesion; AIS = Adenocarcinoma in situ; SCC = Squamous Cell Carcinoma; AC = Adenocarcinoma.
Table 4. Summary Laboratory Diagnosis vs. Site
| Lab Dx | | | | | | | | | | | | | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Site | Method | UNSAT | NILM-WNL | NILM-RR | ASC-US | ASC-H | AGC | LSIL | HSIL | AIS | SCC | AC | Total |
| 1 | MPPT | 61 | 2,367 | 64 | 245 | 14 | 12 | 195 | 58 | 3 | 22 | 4 | 3,045 |
| | PS | 120 | 2,283 | 45 | 298 | 21 | 13 | 163 | 77 | 3 | 21 | 1 | 3045 |
| 2 | MPPT | 21 | 1,684 | 195 | 172 | 4 | 8 | 51 | 11 | | 1 | | 2,147 |
| | PS | 74 | 1,646 | 201 | 159 | 9 | 13 | 36 | 6 | | 2 | 1 | 2,147 |
| 3 | MPPT | 33 | 1,828 | 76 | 102 | 7 | 2 | 63 | 7 | | 1 | | 2,119 |
| | PS | 80 | 1,853 | 58 | 75 | 4 | 1 | 41 | 7 | | | | 2,119 |
| 4 | MPPT | 11 | 2,853 | | 375 | | 12 | 150 | 26 | | 1 | | 3,428 |
| | PS | 28 | 294 | 2 | 317 | | 18 | 100 | 21 | | 1 | | 3,428 |
| Combined | MPPT | 126 | 8,732 | 335 | 894 | 25 | 34 | 459 | 102 | 3 | 25 | 4 | 10,739 |
| | PS | 302 | 8,723 | 306 | 849 | 34 | 45 | 340 | 111 | 3 | 24 | 2 | 10,739 |
| Grouped Diagnoses | MPPT | UNSAT/NILM (WNL/RR) | | 9,193 | ASCUS+ | 1,546 | LSIL+ | 593 | HSIL+ | 134 | Cancer | 32 | |
E. Reference Diagnosis by the Independent Pathologist
The independent pathology (IP) review panel was composed of five (5) board-certified cytopathologists. The independent pathologists were Marshall Austin, M.D., Costal Pathology Associates, Charleston, SC; David Bolick, M.D., AmeriPath Laboratories, Sandy, UT; Michael Glant, M.D., DCL Medical Laboratories, Indianapolis, IN; Michael Henry, M.D., MIAC, Cleveland Clinic Florida, Naples, FL; and Ann Moriarty, M.D., AmeriPath Laboratories, Indianapolis, IN.
The cases which had either PS or MPPT laboratory diagnoses of NILM-RR and above were designated for IP review. There were 2,690 cases in the study with laboratory
18
{11}
Page 12 of 29 Summary of Safety and Effectiveness Data
diagnoses of NILM-RR and above on PS and/or MPPT slides; 2,684 cases (99.8%) were referred to the panel. In addition, 508 cases (6.3%) randomly selected from the 8,094 cases that were diagnosed at the laboratories as NILM-WNL or UNSAT on both PS and MPPT were referred for IP review.
Each of the slides in the referred cases was separately randomized to one of the five cytopathologists for review. Randomization was independently performed for MPPT and PS, and for slides from each site to ensure a balanced random allocation of slides among the five reference cytopathologists. The two slides were reviewed by the reference pathologists for 3,192 referred cases. Each slide was masked as to the laboratory diagnosis for either slide in the case. Seven (7) cases, for which acetic acid was used to reprocess the MPPT slides, were excluded from the statistical analysis. The distribution of the 3,185 cases reviewed by an independent pathologist and available for statistical analysis is presented by Table 5. Each cell of the table presents the total number of cases and the number of cases reviewed by Independent Pathologist ("IP").
Table 5. Distribution of Cases Reviewed by Independent Pathologist
| | | PS Lab Dx | | | | |
| --- | --- | --- | --- | --- | --- | --- |
| | | UNSAT | NILM-WNL | NILM-RR | ASC-US+ | Total |
| MPPT Lab Dx | UNSAT | 43 | 58 | 6 | 19 | 126 |
| | | IP: 1 | IP: 1 | IP: 6 | IP: 19 | IP: 27 |
| | NILM-WNL | 209 | 7,744 | 198 | 581 | 8,732 |
| | | IP: 13 | IP: 491 | IP: 198 | IP: 580 | IP: 1,282 |
| | NILM-RR | 11 | 214 | 59 | 51 | 335 |
| | | IP:11 | IP: 211 | IP: 58 | IP: 51 | IP: 331 |
| | ASC-US+ | 39 | 707 | 43 | 757 | 1,546 |
| | | IP: 39 | IP: 706 | IP: 43 | IP: 757 | IP: 1,545 |
For each IP-reviewed case (3,185 in all), the reference diagnosis was recorded as the more severe diagnosis rendered from the MPPT and PS slides by an Independent Pathologist. This result was used as the cytological "truth" diagnosis for the case or Reference Diagnosis by Independent Pathologist ("Reference Diagnosis", or RDIP). To assess the performance of the MPPT relative to conventional Pap smear for each IP-reviewed case, the laboratory diagnoses made by the study site using the two methods were compared to the RDIP.
Table 6. Independent Pathologist MPPT Diagnosis vs. Independent Pathologist PS Diagnosis (Combined Sites)
12
19
{12}
Page 13 of 29 Summary of Safety and Effectiveness Data
| IP MPPT Dx | IP PS Dx | | | | | | | | | | | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| | UNSAT | NILM-WNL | NILM-RR | ASC-US | ASC-H | AGC | LSIL | HSIL | AIS | SCC | AC | Total |
| UNSAT | 26 | 24 | 8 | 11 | 4 | 1 | 5 | 3 | | | 1 | 83 |
| NILM-WNL | 100 | 568 | 174 | 162 | 17 | 3 | 36 | 14 | | | | 1,074 |
| NILM-RR | 62 | 217 | 104 | 93 | 14 | 4 | 23 | 11 | | | | 528 |
| ASC-US | 67 | 248 | 89 | 131 | 22 | 2 | 56 | 17 | | | 1 | 633 |
| ASC-H | 11 | 27 | 18 | 12 | 6 | 1 | 8 | 6 | | | | 89 |
| AGC | 1 | 13 | 3 | 3 | 2 | | 1 | 1 | | 1 | | 25 |
| LSIL | 35 | 136 | 34 | 116 | 6 | | 153 | 13 | | 1 | | 494 |
| HSIL | 8 | 38 | 18 | 50 | 8 | 1 | 28 | 66 | 1 | 5 | | 223 |
| AIS | 1 | 1 | | | | | | | 1 | | | 3 |
| SCC | 7 | | 1 | 1 | 2 | | | 9 | | 10 | 1 | 31 |
| AC | | | | | 1 | | | | | | 1 | 2 |
| Total | 318 | 1,272 | 449 | 579 | 82 | 12 | 310 | 140 | 2 | 17 | 4 | 3,185 |
In the clinical study, there were 46 cases with Reference Diagnosis of Cancer (Adenocarcinoma, Squamous Cell Carcinoma, or AIS); 328 cases with Reference Diagnosis of HSIL+; 937 cases with Reference Diagnosis of LSIL+; 1,101 cases with Reference Diagnosis of ASC-H+; and 1,902 cases with Reference Diagnosis of ASC-US+.
## F. Outcome Measures
MonoPrep Pap Test screening performance was compared to Pap smear by assessing the relative detection of cervical abnormalities and other conditions, as defined in *The Bethesda System 2001* (TBS2001). Clinical sensitivity and specificity (e.g., with reference to a histological diagnosis) cannot be measured in this study, which relied on cytological examination alone. Rendering RDIPs based on examination of each slide by only one pathologist likely increased the variability inherent in the RDIPs. Another complicating factor is that the IP diagnosis from the MPPT slide (i.e. from the device under testing), was used to establish the Reference Diagnosis for some cases.
Instead of comparing sensitivity and specificity, laboratory true positive and false positive diagnoses by both methods, MPPT and PS, were compared for the cases with a Reference Diagnosis by the Independent Pathologists (RDIP) of ASC-US+, ASC-H/AGC+, LSIL+, HSIL+ and cancer were compared. The prospectively designed primary objective was to demonstrate that MPPT provides a statistically significant improvement over screening with Pap smears for the detection of cases with RDIP-confirmed ASC-US+ and LSIL+ cases. (See Section G. for details.)
About 6% of the cases with both PS and MPPT results of NILM-WNL were referred for RDIP. A result is that the data set of the 3,185 cases with RDIP necessarily has a statistical verification bias because only random sample of cases with both PS and MPPT results of NILM-WNL are submitted for RDIP (2). Despite this verification bias, the ratio of true positive rates by the two methods and the ratio of false positive
13
20
{13}
Page 14 of 29 Summary of Safety and Effectiveness Data
rates by the two methods are unbiased (3). For the various comparisons made below, true positive results are those for which a positive laboratory diagnosis is matched by a positive RDIP. Results without such a match were false positive. The ratios of true positive rates (TPR$_{\text{MPPT}}$/TPR$_{\text{PS}}$) and ratios of false positives rates (FPR$_{\text{MPPT}}$/FPR$_{\text{PS}}$) and their 95% confidence intervals were calculated for the cases with Reference Diagnosis of ASC-US+, ASC-H/AGC+, LSIL+, HSIL+, and cancer. Because of the split-sample design, the positive rates of PS and MPPT were correlated and the false positive rates of PS and MPPT were also correlated. In order to address properly the correlation structure in the calculation of 95% confidence intervals for the ratio of positive rates of MPPT and PS and for the ratio of false positive rates of MPPT and PS, a bootstrap technique was used. The statistical significance of ratios differing from 1.0 was demonstrated when the 95% confidence interval did not include 1.0.
## G. Clinical Study Data Results and Analysis
Tables 7 through 11 present the comparison of laboratory true positive and false positive rates for ASC-US+ (Table 7); ASC-H+/AGC+ (Table 8); LSIL+ (Table 9); HSIL+ (Table 10) and Cancer (Table 11). Tables present the number of RDIP positive and negative cases for each cutoff, the number of positive and negative laboratory results, and their ratio. These data are presented for each site, and include the 95%CI of the ratio for the pooled result of all sites for each cutoff. Data for each site are presented to illustrate the degree of consistency of the results across all sites.
### ASC-US+
Table 7. Laboratory MPPT Results Versus Laboratory PS Results for the Cases with Reference Diagnosis by Independent Pathologist of ASC-US+
| Site | Cases Pos. by IP | MPPT Pos. | PS Pos. | Ratio TPR_{MPPT}/TPR_{PS} | Cases Non-Pos. by IP | MPPT Pos. | PS Pos. | Ratio FPR_{MPPT}/FPR_{PS} |
| --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Site 1 | 702 | 489 | 479 | 1.02 | 361 | 64 | 117 | 0.55 |
| Site 2 | 303 | 163 | 135 | 1.21 | 535 | 83 | 91 | 0.91 |
| Site 3 | 272 | 171 | 115 | 1.49 | 105 | 11 | 13 | 0.85 |
| Site 4 | 625 | 451 | 382 | 1.18 | 282 | 113 | 75 | 1.51 |
| Combined (95% CI) | 1,902 | 1,274 | 1,111 | 1.15 (1.09; 1.20) | 1,283 | 271 | 296 | 0.92 (0.77; 1.06) |
In this table, "Positive" means "ASC-US+" (combined ASC-US, ASC-H, AGC, LSIL, HSIL, and Cancer) and "Non-Positive" means "Non-ASC-US+" (combined NILM-RR, NILM-WNL, and UNSAT).
The results presented in Table 7 show that for the cases with a Reference Diagnosis of ASC-US+, the MPPT method detected 1.15 (1,274/1,111) times more true positive cases than the PS method detected, for all sites combined. This increase was statistically significant, with the lower limit of the 95% confidence interval at 1.09. The observed ratios of the true positive rates varied among the sites from 1.02 to 1.49.
The ratio of the false positive rates was 0.92 (271/296), for all sites combined. The observed decrease in the false positive MPPT rate relative to the false positive PS rate was not statistically significant with 95% confidence interval of 0.77 to 1.06.
{14}
Page 15 of 29 Summary of Safety and Effectiveness Data
In order to make a conclusion about the equivalence of the false positive rates, a multiple imputation (4) was performed for obtaining the unbiased estimate of the difference between MPPT and PS false positive rates. The difference between MPPT and PS false positive rates was -0.3% with 95% CI: -0.86% to 0.26%. The criteria for the equivalence of false positive rates of MPPT and PS for ASC-US+ with delta = 0.5% for the difference of MPPT and PS false positive rates was met (the upper limit of 0.26% is below 0.5%).
## ASC-H/AGC+
Table 8. Laboratory MPPT Results Versus Laboratory PS Results for the Cases with Reference Diagnosis by Independent Pathologist of ASC-H/AGC+.
| Site | Cases Pos. by IP | MPPT Pos. | PS Pos. | Ratio TPR_{MPPT}/TPR_{PS} | Cases Non-Pos. by IP | MPPT Pos. | PS Pos. | Ratio FPR_{MPPT}/FPR_{PS} |
| --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Site 1 | 444 | 274 | 247 | 1.11 | 619 | 34 | 52 | 0.65 |
| Site 2 | 131 | 49 | 43 | 1.14 | 707 | 26 | 24 | 1.08 |
| Site 3 | 159 | 75 | 45 | 1.67 | 218 | 5 | 8 | 0.63 |
| Site 4 | 367 | 139 | 103 | 1.35 | 540 | 50 | 37 | 1.35 |
| Combined (95% CI) | 1,101 | 537 | 438 | 1.23 (1.13; 1.32) | 2,084 | 115 | 121 | 0.95 (0.72; 1.18) |
In this table, "Positive" means "ASC-H/AGC+" (combined ASC-H, AGC, LSIL, HSIL, and Cancer) and "Non-Positive" means "Non-ASC-H/AGC+" (combined ASC-US, NILM-RR, NILM-WNL, and UNSAT).
The results presented in Table 8 above show that for the cases with a Reference Diagnosis of ASC-H/AGC+, the MPPT method detected 1.23 (537/438) times more true positive cases than the PS method detected, for all sites combined. This increase was statistically significant with the lower limit of the 95% confidence interval at 1.13. The observed ratios of the positive rates varied among the sites from 1.11 to 1.67.
The ratio of the false positive rates was 0.95 (115/121) for all sites combined. The observed decrease in the false positive MPPT rate relative to the false positive PS rate was not statistically significant with 95% confidence interval of 0.72 to 1.18. In order to make a conclusion about the equivalence of the false positive rates, a multiple imputation was performed for obtaining the unbiased estimate of the difference between MPPT and PS false positive rates. The difference between MPPT and PS false positive rates was -0.05% with 95% CI: -0.34% to 0.24%. The criteria for the equivalence of false positive rates of MPPT and PS for ASC-H+ with delta = 0.5% for the difference of MPPT and PS false positive rates was met (the upper limit of 0.24% is below 0.5%).
15
22
{15}
Page 16 of 29 Summary of Safety and Effectiveness Data
## LSIL+
Table 9. Laboratory MPPT Results Versus Laboratory PS Results for the Cases with Reference Diagnosis by Independent Pathologist of LSIL+.
| Site | Cases Pos. by IP | MPPT Pos. | PS Pos. | Ratio TPR_{MPPT}/TPR_{PS} | Cases Non-Pos. by IP | MPPT Pos. | PS Pos. | Ratio FPR_{MPPT}/FPR_{PS} |
| --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Site 1 | 388 | 250 | 220 | 1.14 | 675 | 32 | 45 | 0.71 |
| Site 2 | 97 | 43 | 32 | 1.34 | 741 | 20 | 13 | 1.54 |
| Site 3 | 141 | 66 | 43 | 1.53 | 236 | 5 | 5 | 1.00 |
| Site 4 | 311 | 127 | 90 | 1.41 | 596 | 50 | 32 | 1.56 |
| Combined (95% CI) | 937 | 486 | 385 | 1.26 (1.16; 1.36) | 2,248 | 107 | 95 | 1.13 (0.84; 1.41) |
In this table, "Positive" means "LSIL+" (combined LSIL, HSIL, and Cancer) and "Non-Positive" means "Non-LSIL+" (combined AGC, ASC-H, ASC-US, NILM-RR, NILM-WNL, and UNSAT).
The results presented in Table 9 show that for the cases with a Reference Diagnosis of LSIL+, the MPPT method detected 1.26 (486/385) times more true positive cases than the PS method detected, for all sites combined. This increase was statistically significant with the lower limit of the 95% confidence interval at 1.16. The observed ratios of the positive rates varied among the sites from 1.14 to 1.53.
The ratio of the false positive rates was 1.13 (107/95) for all sites combined. The observed increase in the false positive MPPT rate relative to the false positive PS rate was not statistically significant with 95% confidence interval of 0.84 to 1.41. In order to make a conclusion about the equivalence of the false positive rates, a multiple imputation was performed for obtaining the unbiased estimate of the difference between MPPT and PS false positive rates. The difference between MPPT and PS false positive rates was +0.1% with 95% CI: -0.15% to 0.35%. The criteria for the equivalence of false positive rates of MPPT and PS for ASC-H/AGC+ with delta = 0.5% for the difference of MPPT and PS false positive rates was met (the upper limit of 0.35% is below 0.5%).
## HSIL+
Table 10. Laboratory MPPT Results Versus Laboratory PS Results for the Cases with Reference Diagnosis by Independent Pathologist of HSIL+.
| Site | Cases Pos. by IP | MPPT Pos. | PS Pos. | Ratio TPR_{MPPT}/TPR_{PS} | Cases Non-Pos. by IP | MPPT Pos. | PS Pos. | Ratio FPR_{MPPT}/FPR_{PS} |
| --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Site 1 | 156 | 79 | 82 | 0.96 | 908 | 8 | 20 | 0.40 |
| Site 2 | 32 | 8 | 6 | 1.33 | 806 | 4 | 3 | 1.33 |
| Site 3 | 31 | 7 | 6 | 1.17 | 346 | 1 | 1 | 1.00 |
| Site 4 | 109 | 19 | 15 | 1.27 | 798 | 8 | 7 | 1.14 |
| Combined (95% CI) | 328 | 113 | 109 | 1.04 (0.88; 1.19) | 2,857 | 21 | 31 | 0.68 (0.33; 1.02) |
In this table, "Positive" means "HSIL+" (combined HSIL, and Cancer) and "Non-Positive" means "Non-HSIL+" (combined LSIL, AGC, ASC-H, ASC-US, NILM-RR, NILM-WNL, and UNSAT).
16
{16}
Page 17 of 29 Summary of Safety and Effectiveness Data
The results presented in Table 10 show that for the cases with a Reference Diagnosis of HSIL+, the MPPT method detected 1.04 (113/109) times more true positive cases than the PS method detected, for all sites combined. This increase was not statistically significant with the 95% confidence interval of 0.88 to 1.19. A multiple imputation technique provided the unbiased point estimate of difference MPPT and PS true positive rates of 0.9% with 95% CI: -3.7% to 5.4%. The criteria for the equivalence of true positive rates of MPPT and PS for HSIL+ with delta = -5.0% for the difference of MPPT and PS positive rates was met (the lower limit of -3.7% is above -5.0%). The observed ratios of the positive rates varied among the sites from 0.96 to 1.33.
The ratio of the false positive rates was 0.68 (21/31) for all sites combined. The observed decrease in the false positive MPPT rate relative to the false positive PS rate was not statistically significant with a 95% confidence interval of 0.33 to 1.02. In order to make a conclusion about the equivalence of the false positive rates, a multiple imputation was performed for obtaining the unbiased estimate of the difference between MPPT and PS false positive rates. The difference between MPPT and PS false positive rates was -0.08% with 95% CI: -0.20% to 0.04%. The criteria for the equivalence of false positive rates of MPPT and PS for ASC-H/AGC+ with delta = 0.5% for the difference of MPPT and PS false positive rates was met (the upper limit of 0.04% is below 0.5%).
## Cancer
Table 11. Laboratory MPPT Results Versus Laboratory PS Results for the Cases with Reference Diagnosis by Independent Pathologist of Cancer.
| Site | Cases Pos. by IP | MPPT Pos. | PS Pos. | Ratio TPR_{MPPT}/TPR_{PS} | Cases Non-Pos. by IP | MPPT Pos. | PS Pos. | Ratio FPR_{MPPT}/FPR_{PS} |
| --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Site 1 | 40 | 26 | 21 | 1.24 | 1,023 | 3 | 4 | 0.75 |
| Site 2 | 1 | 0 | 1 | 0.0 | 837 | 1 | 2 | 0.5 |
| Site 3 | 1 | 1 | 0 | n/a | 376 | 0 | 0 | n/a |
| Site 4 | 4 | 1 | 1 | 1.0 | 903 | 0 | 0 | n/a |
| Combined (95% CI) | 46 | 28 | 23 | 1.22 (0.87; 1.75) | 3,139 | 4 | 6 | 0.66 |
17
{17}
Page 18 of 29 Summary of Safety and Effectiveness Data
In this table, "Positive" means "Cancer" (combined AIS, Squamous Cell Carcinoma, and Adenocarcinoma) and "Non-Positive" means "Non-Cancer" (combined HSIL, LSIL, AGC, ASC-H, ASC-US, NILM-RR, NILM-WNL, and UNSAT).
The results presented in Table 11 show that for the cases with a Reference Diagnosis of Cancer, the MPPT method detected 1.22 (28/23) times more true positive cases than the PS method detected, for all sites combined. This increase was not statistically significant with the 95% confidence interval of 0.87 to 1.75. The ratio of the false positive rates was 0.66 (4/6) for all sites combined. The observed decrease in the false positive MPPT rate relative to the false positive PS rate was not statistically significant.
## H. LABORATORY MPPT VERSUS PAP SMEAR RESULTS FOR INDIVIDUAL REFERENCE DIAGNOSIS BY INDEPENDENT PATHOLOGIST CATEGORY
Tables 12-19 show the comparison of the laboratory MPPT diagnosis and laboratory PS diagnosis for the cases with the following Reference Diagnoses: NILM-WNL, NILM-RR, ASC-US, ASC-H, AGC, LSIL, HSIL and Cancer (Adenocarcinoma, Squamous Cell Carcinoma, or AIS) separately. This comparison illustrates the diversity of laboratory results with MPPT and Pap smear method for each Reference Diagnosis. An IP diagnosis was made for each slide, and may or may not be the same within a case. The Reference Diagnosis by Independent Pathologist was the most severe of the two IP diagnoses.
NILM-WNL
18
{18}
Page 19 of 29 Summary of Safety and Effectiveness Data
Table 12. Cases with Reference Diagnosis of NILM-WNL
| Lab PS Dx | | | | | | | | | | | | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Lab MPPT Dx | UNSAT | NILM-WNL | NILM-RR | ASC-US | ASC-H | AGC | LSIL | HSIL | AIS | SCC | AC | Total |
| UNSAT | | 1 | 1 | 3 | | | | | | | | 5 |
| NILM-WNL | 5 | 310 | 69 | 93 | 3 | 7 | 6 | | | | | 493 |
| NILM-RR | 4 | 58 | 18 | 8 | | | 1 | | | | | 89 |
| ASC-US | 1 | 82 | 3 | 10 | | 1 | | | | | | 97 |
| ASC-H | | 2 | | | | | | | | | | 2 |
| AGC | | 2 | | | | | | | | | | 2 |
| LSIL | | 4 | | | | | | | | | | 4 |
| HSIL | | | | | | | | | | | | |
| AIS | | | | | | | | | | | | |
| SCC | | | | | | | | | | | | |
| AC | | | | | | | | | | | | |
| Total | 10 | 459 | 91 | 114 | 3 | 8 | 7 | | | | | 692 |
Among the 692 cases with a Reference Diagnosis of NILM-WNL, 493 (71.2%) cases had a laboratory MPPT diagnosis of NILM-WNL and 459 cases (66.3%) had a laboratory PS diagnosis of NILM-WNL; 4 cases (0.6%) had a laboratory MPPT diagnosis of LSIL+ and 7 (1.0%) cases had a laboratory PS diagnosis of LSIL+.
NILM-RR
Table 13. Cases with Reference Diagnosis of NILM-RR
| Lab PS Dx | | | | | | | | | | | | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Lab MPPT Dx | UNSAT | NILM-WNL | NILM-RR | ASC-US | ASC-H | AGC | LSIL | HSIL | AIS | SCC | AC | Total |
| UNSAT | | | 3 | | | | 2 | | | | | 5 |
| NILM-WNL | 3 | 95 | 75 | 102 | 4 | 10 | 8 | 1 | | | | 298 |
| NILM-RR | 5 | 72 | 20 | 10 | | | | | | | | 107 |
| ASC-US | 5 | 105 | 10 | 9 | 1 | 1 | 3 | | | | | 134 |
| ASC-H | | | | 1 | | | | | | | | 1 |
| AGC | | 5 | 1 | 1 | | | | | | | | 7 |
| LSIL | 1 | 10 | | 1 | | | | | | | | 12 |
| HSIL | | | | | 1 | | | | | | | 1 |
| AIS | | | | | | | | | | | | |
| SCC | | | | | | | | | | | | |
| AC | | | | | | | | | | | | |
| Total | 14 | 287 | 109 | 124 | 6 | 11 | 13 | 1 | | | | 565 |
Among the 565 cases with a Reference Diagnosis of NILM-RR, 107 (18.9%) cases had a laboratory MPPT diagnosis of NILM-RR and 109 cases (19.3%) had a laboratory PS diagnosis of NILM-RR; 13 cases (2.3%) had a laboratory MPPT diagnosis of LSIL+ and 14 (2.5%) cases had a laboratory PS diagnosis of LSIL+.
19
{19}
29 Summary of Safety and Effectiveness Data
# ASC-US
Table 14. Cases with Reference Diagnosis of ASC-US
| Lab PS Dx | | | | | | | | | | | | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Lab MPPT Dx | UNSAT | NILM-WNL | NILM-RR | ASC-US | ASC-H | AGC | LSIL | HSIL | AIS | SCC | AC | Total |
| UNSAT | 1 | | | 2 | | | 2 | 1 | | | | 6 |
| NILM-WNL | | 58 | 45 | 163 | 7 | 6 | 20 | | | | | 299 |
| NILM-RR | 2 | 53 | 11 | 12 | | | 2 | | | | | 80 |
| ASC-US | 7 | 211 | 15 | 79 | 1 | 2 | 15 | 1 | | | | 331 |
| ASC-H | | 1 | | 3 | | | | | | | | 4 |
| AGC | 1 | 5 | | | | | | | | | | 6 |
| LSIL | | 34 | 1 | 25 | | | 9 | 3 | | | | 72 |
| HSIL | | 1 | | 1 | 1 | | | | | | | 3 |
| AIS | | | | | | | | | | | | |
| SCC | | | | | | | | | | | | |
| AC | | | | | | | | | | | | |
| Total | 11 | 363 | 72 | 285 | 9 | 8 | 48 | 5 | | | | 801 |
Among the 801 cases with a Reference Diagnosis of ASC-US, 416 (51.9%) cases had a laboratory MPPT diagnosis of ASC-US+ and 355 cases (44.3%) had a laboratory PS diagnosis of ASC-US+; 379 cases (47.3%) had a laboratory MPPT diagnosis of NILM and 435 (54.3%) cases had a laboratory PS diagnosis of NILM.
# ASC-H
Table 15. Cases with Reference Diagnosis of ASC-H
| Lab PS Dx | | | | | | | | | | | | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Lab MPPT Dx | UNSAT | NILM-WNL | NILM-RR | ASC-US | ASC-H | AGC | LSIL | HSIL | AIS | SCC | AC | Total |
| UNSAT | | | 2 | 2 | | | | | | | | 4 |
| NILM-WNL | 1 | 13 | 8 | 21 | 2 | 2 | 3 | 5 | | | | 55 |
| NILM-RR | | 8 | 5 | 1 | 1 | | 1 | | | | | 16 |
| ASC-US | | 21 | 2 | 10 | | 1 | 4 | 1 | | | | 39 |
| ASC-H | | 1 | | | | | | 1 | | | | 2 |
| AGC | | | | 1 | | | | | | | | 1 |
| LSIL | | 6 | | 4 | | | | 2 | | | | 12 |
| HSIL | | | | | | | | 1 | | | | 1 |
| AIS | | | | | | | | | 1 | | | 1 |
| SCC | | | | | | | | | | | | |
| AC | | | | | | | | | | | | |
| Total | 1 | 49 | 17 | 39 | 3 | 3 | 8 | 10 | 1 | | | 131 |
Among the 131 cases with a Reference Diagnosis of ASC-H, 17 (13.0%) cases had a laboratory MPPT diagnosis of ASC-H+ and 25 cases (19.1%) had a laboratory PS diagnosis of ASC-H+; 71 cases (54.2%) had a laboratory MPPT diagnosis of NILM
27
{20}
Page 21 of 29 Summary of Safety and Effectiveness Data
and 66 (50.4%) cases had a laboratory PS diagnosis of NILM.
# AGC
Table 16. Cases with Reference Diagnosis of AGC
| Lab PS Dx | | | | | | | | | | | | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Lab MPPT Dx | UNSAT | NILM-WNL | NILM-RR | ASC-US | ASC-H | AGC | LSIL | HSIL | AIS | SCC | AC | Total |
| UNSAT | | | | | | | | | | | | |
| NILM-WNL | | 8 | | 4 | | 3 | 1 | | | | | 16 |
| NILM-RR | | 3 | 2 | | | | | | | | | 5 |
| ASC-US | | 5 | | | 1 | | | | | | | 6 |
| ASC-H | | | | | | | | | | | | |
| AGC | | 2 | | 1 | 1 | | | | 1 | | | 5 |
| LSIL | | | | 1 | | | | | | | | 1 |
| HSIL | | | | | | | | | | | | |
| AIS | | | | | | | | | | | | |
| SCC | | | | | | | | | | | | |
| AC | | | | | | | | | | | | |
| Total | | 18 | 2 | 6 | 2 | 3 | 1 | | 1 | | | 33 |
Among the 33 cases with a Reference Diagnosis of AGC, 6 (18.2%) cases had a laboratory MPPT diagnosis of ASC-H+ and 7 cases (21.2%) had a laboratory PS diagnosis of ASC-H+; 21 cases (63.6%) had a laboratory MPPT diagnosis of NILM and 20 (60.6%) cases had a laboratory PS diagnosis of NILM.
# LSIL
Table 17. Cases with Reference Diagnosis of LSIL
| Lab PS Dx | | | | | | | | | | | | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Lab MPPT Dx | UNSAT | NILM-WNL | NILM-RR | ASC-US | ASC-H | AGC | LSIL | HSIL | AIS | SCC | AC | Total |
| UNSAT | | | | 1 | | | 1 | 1 | | | | 3 |
| NILM-WNL | | 3 | | 49 | | | 16 | 2 | | | | 70 |
| NILM-RR | | 7 | | 6 | | | 1 | | | | | 14 |
| ASC-US | 4 | 89 | 8 | 72 | 1 | 1 | 44 | 1 | | | | 220 |
| ASC-H | | 1 | | 2 | | | 1 | | | | | 4 |
| AGC | 1 | 1 | | | | | | | | | | 2 |
| LSIL | 3 | 69 | | 61 | 1 | | 140 | 7 | | | | 281 |
| HSIL | | | | 3 | 1 | | 9 | 2 | | | | 15 |
| AIS | | | | | | | | | | | | |
| SCC | | | | | | | | | | | | |
| AC | | | | | | | | | | | | |
| Total | 8 | 170 | 8 | 194 | 3 | 1 | 212 | 13 | | | | 609 |
Among the 609 cases with a Reference Diagnosis of LSIL, 296 (48.6%) cases had a laboratory MPPT diagnosis of LSIL+ and 225 cases (36.9%) had a laboratory PS
21
78
{21}
1
19 Summary of Safety and Effectiveness Data
diagnosis of LSIL+; 84 cases (13.8%) had a laboratory MPPT diagnosis of NILM and 178 (29.2%) cases had a laboratory PS diagnosis of NILM.
## HSIL
Table 18. Cases with Reference Diagnosis of HSIL
| Lab PS Dx | | | | | | | | | | | | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Lab MPPT Dx | UNSAT | NILM-WNL | NILM-RR | ASC-US | ASC-H | AGC | LSIL | HSIL | AIS | SCC | AC | Total |
| UNSAT | | | | 1 | | | | | | | | 1 |
| NILM-WNL | | 2 | 1 | 22 | | 5 | 1 | 7 | | | | 38 |
| NILM-RR | | 10 | 2 | 2 | | | 1 | 2 | | | | 17 |
| ASC-US | 2 | 19 | 2 | 21 | | 1 | 7 | 4 | | | | 56 |
| ASC-H | | 4 | | 4 | | | 1 | 1 | | | | 10 |
| AGC | 1 | 5 | | 1 | | | 1 | | | | | 8 |
| LSIL | 2 | 12 | | 20 | | | 27 | 15 | | | | 76 |
| HSIL | 2 | 3 | | 6 | 4 | | 13 | 42 | | 3 | | 73 |
| AIS | | | | | | | | | | | | |
| SCC | | | | | | | | 2 | | 1 | | 3 |
| AC | | | | | | | | | | | | |
| Total | 7 | 55 | 5 | 77 | 4 | 6 | 51 | 73 | | 4 | | 282 |
Among the 282 cases with a reference diagnosis of HSIL, 76 (27.0%) cases had a laboratory MPPT diagnosis of HSIL+ and 77 cases (27.3%) had a laboratory PS diagnosis of HSIL+; 55 cases (19.5%) had a laboratory MPPT diagnosis of NILM and 60 (21.3%) cases had a laboratory PS diagnosis of NILM.
## CANCER
Table 19. Cases with Reference Diagnosis of Cancer (Adenocarcinoma, Squamous Cell Carcinoma, or AIS)
22
29
{22}
Page 23 of 29 Summary of Safety and Effectiveness Data
| Lab PS Dx | | | | | | | | | | | | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Lab MPPT Dx | UNSAT | NILM-WNL | NILM-RR | ASC-US | ASC-H | AGC | LSIL | HSIL | AIS | SCC | AC | Total |
| UNSAT | | | | 1 | | 0 | | | | | | 1 |
| NILM-WNL | | | | | | 1 | | | | | 1 | 2 |
| NILM-RR | | | | | | | | | | 1 | | 1 |
| ASC-US | | | | | | | | | | | | |
| ASC-H | 1 | | | | | | | | | 1 | | 2 |
| AGC | | 1 | | | | 1 | | | | | | 2 |
| LSIL | | | | | | | | | | 1 | | 1 |
| HSIL | | | | | | 1 | | 5 | | 3 | | 9 |
| AIS | 1 | | | | | | | | 1 | | | 2 |
| SCC | 2 | | | 1 | 4 | | | 3 | | 12 | | 22 |
| AC | | | | | | | | 1 | | 2 | 1 | 4 |
| Total | 4 | 1 | | 2 | 4 | 3 | | 9 | 1 | 20 | 2 | 46 |
Among the 46 cases with a Reference Diagnosis of Cancer (Adenocarcinoma, Squamous Cell Carcinoma, or AIS), 37 (80.4%) cases had a laboratory MPPT diagnosis of HSIL+ and 32 (69.6%) cases had a laboratory PS diagnosis of HSIL+; 3 (6.5%) cases had a laboratory MPPT diagnosis of NILM, and 1 (2.2%) case had a laboratory PS diagnosis of NILM.
Twenty-eight (60.9%) of the 46 cases had a laboratory MPPT diagnosis of Cancer and 23 (50.0%) had a laboratory PS diagnosis of Cancer. None of the 46 (0.0%) cases had a MPPT IP diagnosis of NILM (WNL or RR); 2 (4.3%) had a PS IP diagnosis of NILM (WNL or RR).
For the three cases with a MPPT Laboratory diagnosis of NILM, none were NILM by IP diagnosis of that slide. In one case, the IP diagnosis for cancer was made only on the MPPT slide, with the Pap smear IP diagnosis being UNSAT. In an additional post-study review by two study cytopathologists, the MPPT slide was considered extremely difficult to diagnose because of cytolysis with poor preservation and pre-collection necrosis. There were cells suggestive of atypical repair. The PS slide was thick, air dried, and poorly preserved "except for sprinkling of well preserved atypical keratinizing cells suggestive of squamous carcinoma."
The second case was cancer by IP diagnosis for the Pap smear, though AGC by the laboratory diagnosis of that slide. The MonoPrep laboratory diagnosis was NILM, with only the primary screening cytotechnologist review, without QC review. The MPPT IP diagnosis was ASC-US. In an additional post-study review by two study cytopathologists, the abnormal cells in the Pap smear were considered diagnostically difficult, consistent with either endometrial adenocarcinoma or endometrial AGC. For the MPPT slide, the secondary reviewing cytopathologists concurred that "rare small atypical groups" were present.
The third case's IP diagnoses were cancer for the Pap smear and UNSAT for the MPPT slide. The MonoPrep laboratory diagnosis was NILM, with only primary screening cytotechnologist review and no QC review. In an additional post-study review by two
23
30
{23}
Page 24 of 29 Summary of Safety and Effectiveness Data
study cytopathologists, both slides were considered very difficult cases, with the Pap smear being UNSAT except for the identification of a "few isolated individual clearly malignant cells buried in the blood." On extensive review "some isolated but poorly preserved similar cells" were identified on the MPPT slide.
For the case with a Laboratory PS diagnosis of NILM (WNL or RR), the PS IP diagnosis was NILM (WNL or RR), while MPPT IP diagnosis was Cancer and Laboratory MPPT diagnosis was AGC. At the laboratory, the PS slide was reviewed and diagnosed as NILM-WNL by both primary and Senior (QC) Cytotechnologists. This case was not part of the post-study slides review.
## I. Specimen Adequacy
Table 20 shows results from a comparison of preparation adequacy for the conventional PS and MPPT methods as reviewed by the laboratory for all sites combined and each site separately:
{24}
Page 25 of 29 Summary of Safety and Effectiveness Data
Table 20. Specimen Adequacy Findings
| | | Lab PS | | |
| --- | --- | --- | --- | --- |
| | | UNSAT | SAT | Total |
| Lab MPPT | UNSAT | 43 | 83 | 126 |
| | SAT | 259 | 10,354 | 10,613 |
| | Total | 302 | 10,437 | 10,739 |
| | | | | |
| Lab | | | | |
| Site | Method | UNSAT | Total Number of Slides | %UNSAT |
| 1 | MPPT | 61 | 3,045 | 2.0% |
| | PS | 120 | 3,045 | 3.9% |
| 2 | MPPT | 21 | 2,147 | 1.0% |
| | PS | 74 | 2,147 | 3.4% |
| 3 | MPPT | 33 | 2,119 | 1.6% |
| | PS | 80 | 2,119 | 3.8% |
| 4 | MPPT | 11 | 3,428 | 0.3% |
| | PS | 28 | 3,428 | 0.8% |
| Combined | MPPT | 126 | 10,739 | 1.2% |
| | PS | 302 | 10,739 | 2.8% |
The estimated unsatisfactory slide rates observed in the laboratories (i.e., without confirmation by independent pathologist (IP)) for the MPPT method were lower than for the PS method (1.2% vs. 2.8%). However, these estimates take no account of MPPT slides that might not have been recognized at the laboratories as unsatisfactory. Few (15) slide pairs with laboratory diagnoses confined to UNSAT or NILM-WNL were sent for IP review (see Table 5), including 13 pairs called UNSAT by PS and NILM-WNL by MPPT. Four MPPT slides from these 13 pairs were categorized as UNSAT by the IP. The number of these slide pairs, and the even smaller number of IP-reviewed pairs called UNSAT by MPPT and NILM-WNL by PS, make evaluation of this finding inconclusive.
## J. Abundance of Endocervical / Transformation Zone Component
Laboratories assessed slides for the presence of endocervical and transformation zone component. In the split-sample study, MPPT slides demonstrated no statistically significant difference in abundance of Endocervical/Transformation zone component compared with the matching Pap smear slides as shown in Table 21. ECC/Tz were absent in fewer MPPT than PS slides, but the difference was not statistically significant (-3.3%, 95%CI: -4.0% to 11.0%).
Table 21. Cross-Tabulation of Endocervical and Transformation Zone Component
{25}
Page 26 of 29 Summary of Safety and Effectiveness Data
| | Pap Smear | | | |
| --- | --- | --- | --- | --- |
| MonoPrep | Diagnosis | Absent | Detectable | Total |
| | Absent | 640 | 606 | 1,246 |
| | Detectable | 649 | 8,604 | 9,253 |
| | Total | 1,289 | 9,210 | 10,499 |
## K. Abundance of Abnormal Cells
Laboratories also were asked to assess the relative abundance of abnormal/reactive cells in cases identified as abnormal/reactive. The categories were Abundant (>25) Typical (11-25), and Detectable (1-10). Table 22 presents the comparison for cases where both slides were abnormal/reactive. As shown by the results, there were no statistically significant differences in the abundance of such cells. This demonstrates that MonoPrep presents, on average, at least as many abnormal/reactive cells as a Pap smear, even when made from a split specimen.
Table 22. Cross-Tabulation of Abnormal Cell Abundance
| | Pap Smear | | | | | |
| --- | --- | --- | --- | --- | --- | --- |
| MonoPrep | Abundance | Abundant (>25) | Typical (11-25) | Detectable (1-10) | Total | Row % Cases |
| | Abundant (>25) | 121 | 74 | 29 | 224 | 31% |
| | Typical (11-25) | 83 | 116 | 82 | 281 | 39% |
| | Detectable (1-10) | 25 | 73 | 113 | 211 | 29% |
| | Total | 229 | 263 | 224 | 716 | 100% |
| | Col. % of Cases | 32% | 32% | 31% | 100% | |
## L. Detection of Infectious Organisms, Reactive/Reparative and Other Benign Conditions
Screening with MPPT and Pap smear slides presented no statistically significant difference in detection of benign, reactive/reparative conditions and infectious agents. Table 23 shows the detection rates for these conditions and agents.
26
{26}
Page 27 of 29 Summary of Safety and Effectiveness Data
Table 23. Summary Table Summary of Benign Conditions: MPPT versus PS
| Condition | MonoPrep (n=10,739) | | Pap Smear (n=10,739) | |
| --- | --- | --- | --- | --- |
| | n | % | n | % |
| Reactive / Reparative | 335 | 3.1% | 306 | 2.8% |
| Inflammation | 249 | 2.3% | 231 | 2.2% |
| IUD | 0 | 0.0% | 4 | 0.0% |
| Atrophic Vaginitis | 0 | 0.0% | 1 | 0.0% |
| Radiation | 3 | 0.0% | 1 | 0.0% |
| Other* | 67 | 0.6% | 77 | 0.7% |
| Infectious Agent | 1,507 | 14.0% | 1,496 | 13.9% |
| Candida / Fungus | 523 | 4.8% | 426 | 4.0% |
| Trichomonas Vaginalis | 105 | 1.0% | 158 | 1.5% |
| Actinomyces | 0 | 0.0% | 0 | 0.0% |
| Bacterial Vaginosis / Coccobaccilli | 980 | 9.1% | 1,035 | 9.6% |
| Herpes Simplex | 3 | 0.0% | 9 | 0.1% |
| Other** | 0 | 0.0% | 2 | 0.0% |
| * includes unusual observations, such as those resulting from chemical irritation, drug reactions, or cervical trauma. **includes appearance of microbial infection or sequela of unidentified or unusual taxonomy | | | | |
XI. CONCLUSIONS DRAWN FROM THE STUDIES
For all sites combined, slides prepared by MPPT, compared to PS slides, yielded statistically significant increases in true positive cytological results for the following diagnostic classes: ASC-US+ (1.15, 95%CI: 1.09 to 1.20); ASC-H/AGC+ (1.23, 95%CI: 1.13 to 1.32); and LSIL+ (1.26, 95%CI: 1.16 to 1.36). Hence the increases in true positive yield were at least 9% for ASCUS+, 13% for ASC-H/AGC+, and 16% for LSIL+.
Comparisons of false positive rates did not show a statistically significant difference between MPPT and PS for ASC-US+, ASC-H/AGC+ or LSIL+.
For all sites combined, slides prepared by MPPT, compared to PS slides, did not yield statistically significant differences in true positive or false positive cytological results for the following diagnostic classes: HSIL+ (1.04, 95%CI: 0.88 to 1.19); and Cancer (1.22, 95%CI: 0.87 to 1.75).
Presentation of endocervical cell and transformation zone component, abnormal cells and benign conditions showed no statistically significant difference between MPPT and PS slides.
34
{27}
Summary of Safety and Effectiveness Data
The data from the clinical trial and clinical support studies demonstrate that the MPPT system is safe and effective for preparing gynecologic cytology slides to screen for cervical abnormalities.
## VALIDITY OF THE CLINICAL DATA
The clinical investigation constituted valid scientific evidence as defined in 21 CFR 860.7. The investigation was well-controlled in that a test article and a control article were made from each study subject’s cervical sample. This was possible by using a split-sample collection methodology in which a conventional Pap smear was made first, and then the collection devices were rinsed in the MonoGen Specimen Transport Solution. The MonoPrep Pap Test slide was then made from the sample in the Specimen Transport Solution.
The clinical investigation protocol included a statement of the objectives and hypotheses of the study. Statistical testing was based on these pre-defined hypotheses. The clinical study sites were monitored by an independent Contract Research Organization to assure adherence to the protocol.
The statistical methods used to analyze the data from this investigation were based on the estimation of the ratios of true positive rates of MPPT and PS with 95% confidence interval and estimation of the ratios of false positive rates of MPPT and PS with 95% confidence intervals. These estimations were performed for all basic cytological categories: ASC-US+, ASC-H/AGC+, LSIL+, and HSIL+.
## RISK BENEFIT ANALYSIS
Specimen preparation errors may result in false negative or false positive diagnoses. A false negative diagnosis may result when there are no abnormal cells on the slide when disease is actually present. False negative diagnoses result in delayed diagnosis and treatment for the patient. A false positive diagnosis may result when normal cells appear abnormal due to faulty slide preparation but no disease is present. As a result the patient may have an unnecessary colposcopy exam (a non-invasive procedure) or may be referred for biopsy (an invasive procedure).
Based on the information in the studies provided, the FDA has concluded that the benefits of using the MonoPrep Pap Test system for its intended use outweigh the risks associated with using it.
## SAFETY
The MonoGen MonoPrep Pap Test system is an in vitro diagnostic test and does not contact the patient. Instructions for the safe use of the product are included in the package insert.
## EFFECTIVENESS
The data from the clinical trial and clinical support studies demonstrate that the MonoGen MonoPrep Pap Test is effective for preparing gynecologic cytology slides to screen for cervical abnormalities.
35
{28}
Page 29 of 29 Summary of Safety and Effectiveness Data
## XII. PANEL RECOMMENDATION
In accordance with the provisions of section 515(c)(2) of the act as amended by the Safe Medical Devices Act of 1990, this PMA was not referred to the Hematology and Pathology Devices panel, an FDA advisory committee, for review and recommendation because the information in the PMA substantially duplicates information previously reviewed by this panel.
## XIII. CDRH DECISION
CDRH issued an approval order on March 3, 2006.
The applicant’s manufacturing and control facilities were inspected on 9/8/05 and the facilities were found to be in compliance with the Quality System Regulation (21 CFR 820).
## XIV. APPROVAL SPECIFICATIONS
Directions for use: See the labeling (Attachment 1).
Hazards to Health from Use of the Device: See Indications, Contraindications, Warnings, Precautions and Adverse Events in the labeling.
Postapproval Requirements and Restrictions: CDRH approval of this PMA is subject to full compliance with the conditions of approval and post-approval clinical studies described in the approval order (Attachment 2).
## XV. REFERENCES
1. Solomon, D et al. The 2001 Bethesda System: terminology for reporting results of cervical cytology. *JAMA*. 2002; 287(16): 2114-19.
2. Begg, C.B. and Greenes, R.A. Assessment of diagnostic rests when disease verification is subject to selection bias. *Biometrics*, 39 (1), 207-215 (1983).
3. Schatzkin, A., Connor, R.J., Taylor, P.R., and Bunnag, B. Comparing new and old screening tests when a reference procedure cannot be performed on all screeners”. *American Journal of Epidemiology*, Vol. 125, N.4, p.672-678 (1987).
4. Schafer, J.L. “Multiple Imputation: a Primer”. *Statistical Methods in Medical Research*; 8:3-15 (1999).
29
30
