STUDY DESIGN:
This is an experimental methodological study with a single-arm, unmasked, unicentric,
international diagnostic device.
OUTLINES:
1,780 women with abnormal Pap Smears (ASCUS+) or positive high-risk HPV test in the
cervix performed in the Barretos Cancer Hospital (HCB, Barretos, Brazil) screening
program will be recruited, either at Mobile Prevention Units or Fixed Units, who have
been referred for diagnostic investigation with colposcopy. During the colposcopy
examination, the HRME device will be used to check the presence of precursor lesions in
the cervix.
PRE-TESTING VERIFICATION:
As this research may include cervical biopsy in women with fertility, for safety reasons,
the possibility of pregnancy in this group of women will be verified, since biopsy
performed in the cervix of pregnant woman increases the risk of abortion. This
verification will be performed by means of clinical history (menstrual delay, signs
suggesting pregnancy, contraceptive methods, etc.) and also by means of qualitative
research on human chorionic gonadotropin in urine. This test is also known as "rapid
pregnancy test" or "pharmacy test", with an extensive list of products sold in Brazil and
authorized by "Agencia Nacional de Vigilancia Sanitária" (ANVISA, Brazilian Agency of
Health Surveillance). The principle of the test involves a single step
immunochromatographic method to detect the hormone. Reaction occurs on a reagent strip,
where a small amount of urine is deposited. Test result is provided in approximately 2
minutes. Women who are pregnant (or suspected to be pregnant) will not be included in the
study, but will be submitted to diagnostic investigation according to the institutional
protocol for pregnant women with changes in the cervix at HCB. These women will receive
medical orientation and will be referred to a specialized obstetrical service.
COLLECTION OF CERVICAL CYTOLOGY:
Before the diagnostic examinations are performed, the doctor will collect a new cervical
cytology (Papanicolaou), which is part of the routine of the Department of Colposcopy for
every woman who has an altered Pap Smear or positive HPV test. These samples will be
preserved in an ethanol-based preservative medium (SurePath™ Preservative Fluid, Becton &
Dickinson, USA) and will be sent to the Department of Pathology of HCB where they will
undergo an automated processing for the preparation of cytology slides.
DIAGNOSTIC EXAMINATIONS:
VIA will be performed using 3-5% acetic acid, applied on the cervix and to any abnormal
lesions observed. Then, iodine solution will be applied, and the colposcopy will be
performed. It is necessary to clarify that the application of acetic acid and iodine is
routinely performed during a standard colposcopy examination.
Then, Proflavine (0.01%) will be applied to the surface of the cervix. HRME will capture
images from all areas considered abnormal by VIA and/or colposcopy. In addition, all four
quadrants will be probed with HRME to ensure that any non-acetowhite lesions are also
observed. The person responsible for the colposcopy will take note of your opinion about
the lesion and HRME image at each area (normal, benign, low-grade precancerous,
high-grade precancerous or cancer). The complete HRME imaging procedure will add 5-10
minutes to the standard colposcopy examination. Observations related to VIA, colposcopy
and HRME will be registered by quadrant. Any abnormal areas detected by VIA and/or
colposcopy will undergo a biopsy. If no abnormal area is observed, but the woman has an
altered cytology test or a positive HPV test, a cervical microbiopsy will be obtained
from an apparently normal area captured by HRME imaging of that area. For all cervical
biopsies, special tweezers will be used, causing minimal tissue trauma (microbiopsy) and
reducing participant discomfort. These tweezers are different from those commonly used in
colposcopy tests, which usually perform biopsies on large areas and bring more discomfort
to women.
Two experienced HCB pathologists, ignoring all study results, will review histology and
classify areas as normal, cervical intraepithelial neoplasia grade 1 (CIN 1), CIN 2, CIN
3, adenocarcinoma in situ (AIS) or cancer according to standardized criteria. Conflicting
results will be resolved by means of consensus review among pathologists.
HIGH-RISK HPV SCREENING AND GENOTYPING:
High-risk HPV screening will be performed on aliquots of cervical cytology samples
(SurePath™) collected just prior to colposcopy in women who have not been submitted to
this test yet. HPV test will be performed on a Cobas X480 ™ device (Roche Molecular
Systems, USA), which is available at the technology park of the Molecular Oncology
Research Center of Barretos Cancer Hospital - Pio XII Institution. The test protocol will
be performed as described by the manufacturer.
COBAS system is an automated amplification device (by means of real time polymerase chain
reaction (PCR)) for the detection of 14 high-risk HPV types (16, 18 31, 33, 35, 39, 45,
51, 52, 56, 58, 59, 66 and 68), and can process up to 94 simultaneous samples. Tests
which detected high-risk non-16 and non-18 HPV will be submitted to complementary
analysis to identify genotype(s) using the linear array technique, since COBAS system
does not provide genotyping of these high-risk HPVs. For this purpose, linear array HPV
genotyping (CE-IVD) Test for HPV Genotyping kit (Roche Molecular Systems, USA) will be
used.
DATA ANALYSIS:
The primary endpoint is the diagnosis rate of CIN2+ and CIN3+. Sensitivity and diagnostic
specificity of VIA, colposcopy and HRME based on each lesion and each patient using
histological diagnosis as gold standard will be calculated assuming lesions diagnosed as
CIN2 + or CIN3 + are positive. Sensitivity and specificity of VIA and colposcopy will
also be calculated as a comparison method.
A case-by-case description will be made using descriptive statistics. Categorical
variables will be compared by means of the chi-square test or Fisher's exact test,
depending on the expected values in the contingency tables. To compare numerical
variables, the t-tests or the Mann-Whitney test will be used depending on the adherence
to normality (to be verified by the Kolmogorov-Smirnov test).
SAMPLE SIZE CALCULATION:
Sample calculation was based on information from other studies of the same group
(including data not yet published). The following premises were taken: HRME sensitivity
and specificity for CIN2+ diagnosis (93% and 48%, respectively, per biopsy area) and
prevalence of CIN2+ in the colposcopy clinic of HCB (ranging from 20 to 30%). Considering
a prevalence of disease of 25% , it is estimated that it will be necessary to include
1,424 women (625 cases for sensitivity calculation and 799 cases for specificity
calculation) to reach the expected sensitivity and specificity rates with a maximum
margin of error of 4% for a 95% confidence interval. Considering an estimated loss of
10-15% (non-attendance for colposcopy examination and HRME), the corrected sample
estimate will range from 1,566 to 1,637 women. Therefore, the study intends to include
1,600 women with abnormal cervical cytology or positive HPV test. Additionally, a group
of women with no change in cervical cytology and HPV test will be invited to participate
in the research in order to check verification bias. This bias occurs when only the
positive cases in the screening are selected to perform complementary diagnostic
examinations, overestimating the sensitivity and underestimating the specificity. It is
recommended that a percentage of negative cases be verified according to the standard
gold examination. Therefore, 10% of the study population will consist of negative cases
in both cervical cytology and the HPV test, which corresponds to nearly 180 cases. Thus,
the final sample estimate will be of 1,780 women, of whom 1,600 women will have screening
(altered cytology or positive HPV test) results reported as positive and 180 will have
screening results reported as negative (negative cytology and HPV test).
STUDY SCHEDULE:
The study will last 3 years.
