Keywords
uterine cervical neoplasm - disease management - atypical squamous cells of the cervix
- papanicolaou test - human papillomavirus
Palavras-chave
neoplasia cervical uterina - manejo de doenças - células escamosas atípicas do colo
do útero - teste de papanicolau - papilomavírus humano
Introduction
Despite being preventable, uterine cervical cancer persists as an important cause
of women's morbidity and mortality, especially in developing countries.[1] Since it was adopted as a screening method in the middle of the last century, the
Papanicolaou (PAP) test remains a preferred method of screening and has reduced the
incidence and mortality of cervical cancer;[2] however, this fundamental procedure presents a high variation of sensitivity and
reproducibility for the detection of cancer precursor lesions.[3] In addition to this limitation, the spectrum of cytological diagnoses also presents
a group of atypical squamous cell findings, whose cytomorphological changes do not
allow a definitive diagnosis. Therefore, there remains a gray area of persistent controversy
among both gynecologists and cytopathologists.[4]
[5]
The Bethesda System (TBS) recognizes two atypical diagnoses in squamous cells: atypical
squamous cells of undetermined significance (ASC-US), which occurs when there is uncertainty
between findings of reactive changes and low-grade squamous intraepithelial lesion
(LSIL), and atypical squamous cells-cannot exclude high-grade squamous intraepithelial
lesion (ASC-H), which is identified when the doubt is whether it is an immature and
reactive metaplasia or a high-grade intraepithelial lesion (HSIL).[6]
[7] In fact, these cytological abnormalities are not defined as a specific biological
entity, but as result of an interpretive uncertainty; they do not even have a corresponding
histopathological description.[8] Thus, the cytologic diagnosis remains inconclusive, and this leads to a clinical
dilemma regarding the best way of managing such cases.[5]
[9]
The number of patients lost to cervical cancer screening follow-ups is greater in
developing countries, and it has been reported that a delay in the diagnosis of this
cancer is correlated with lower survival rates. Several factors such as socioeconomic
status, access to healthcare facilities, prior partner permission and family history
may affect a screening program success, but a previous health guidance is crucial
to enable a patient's return, especially in the case of dubious results, such as ASCs.[10]
[11]
Referral to colposcopy/biopsy may burden the health system too much, since many low-grade
lesions clear up spontaneously. On the other hand, many patients with atypical findings
may correspond to high-grade lesions or even invasive cervical cancer (ICC).[2] In Brazil, the Ministry of Health (MS) and the National Cancer Institute (INCA)[12] recommend that a cytology with an ASC-US result should be repeated in a variable
time depending on the patient's age, and, in the case of ASC-H, the recommendation
is for a colposcopy to be performed, regardless of the age of the patient.
This study aims to determine what methods are chosen after an ASCs cytology and to
evaluate the outcomes of cases diagnosed with ASC-US and ASC-H over a 6-year period
from a private health service in a large city in northeastern Brazil.
Methods
This is an observational and prospective study exclusively performed on laboratory
database records of cases diagnosed as ASCs between January of 2010 and July of 2016
in the city of Fortaleza (Brazil) and, therefore, informed consent was not necessary.
It was approved by the Ethics and Research Committee of the Federal University of
Ceará (protocol number: 55957716.3.0000.5054).
The case records were scrutinized form a database to determine, for each ASCs cytology,
the respective clinical management and results. Samples from ∼ 163 gynecologists were
assessed. Patients of any age who were not currently pregnant or immunocompromised
were included. Altogether, 2,458 cervical samples with ASCs results were analyzed
at the laboratory, and only one was excluded due Cushing syndrome.
All smears were collected using Liquid-Based Cytology by SurePath (BD, Franklin Lakes,
N.J., USA) or Pap Smear, and the interpretations followed the TBS nomenclature. The
HPV tests from the database were performed by Hybrid Capture 2 (HC2) (Qiagen AG, Hombrechtikon,
Switzerland) or real-time polymerase chain reaction (PCR). The HC2 test was performed
for high-risk (HR) HPV (genotypes 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59,
68). Expression was measured as the reactive light unit (RLU). The Cobas 4800 system
(Roche Diagnostics, Pleasanton, CA, USA) with three channels (HPV16, HPV18, or other
12 HR HPV types) was used according to the manufacturer's instructions.
The analysis of the management was based (until July 2011) on the consensus of the
Brazilian Nomenclature for Cervical Reports and Recommended Practices of 2006; after
July 2011, the 2011 Brazilian guidelines for the cervical cancer screening of INCA
were used because they were applicable at the time of data collection.
To verify if the clinical management for ASC-US followed the current protocols, the
data was separated into two groups: repeat cytology at 12 months for those under 30
years of age from August 2011 onwards and repeat cytology at 6 months for the remaining
patients.
To evaluate the absence of evolution data in the sample, the patient records for those
aged 30 years or over whose ASC-US results were collected up to 6 months before the
end date of the study were excluded. Similarly, those patients aged less than 30 years
whose ASC-US results were collected up to 12 months before the end date of the study
were excluded, and patients aged less than 30 years with ASC-H cytology whose data
were collected within 6 months before the end date of the study were also excluded.
These exclusions were performed because the absence of data in these cases would be
expected.
Statistical analyses were performed using the GraphPad Prism software, version 6.0
(Graph-Pad Software Inc., San Diego, CA, USA). The mean and standard deviations were
used for continuous variables. The Fisher exact test was applied for binomial variables,
and a one-sample Student t-test was performed for continuous variables; the 95% confidence intervals (CIs) were
used, and results were considered as statistically significant when p < 0.05.
Results
For the ASC-US group, the ages of the patients ranged from 14 to 85 years, with a
mean of 33.12 years (±11.28); for the ASC-H group, the mean age was 37.10 years (±12.86),
and they ranged from 17 to 82 years.
After applying the exclusion criteria to verify absence of evolution, the ASC-US results
were 1,998. Among these specific cases, 739 (36.97%) did not present any evolution
examination results to assess the clinical management that followed. There was also
no follow-up for ASC-H patients in 98 out of 242 records (40.50%).
Regarding the Pap smear method, the data that contained follow-up information indicated
that a repeat cytology was the most frequent management for the ASC-US group (66.72%).
Cytology was also the most frequent evolution in cases of ASC-H (85 registers; 59.02%).
The isolated biopsy was the only method that showed statistical significance between
the groups, with 17 (11.81%) ASC-H records (p = 0.0112) ([Table 1]).
Table 1
Atypical squamous cells subgroups and performed management (n = 1,403)
|
Evolution
|
ASC-US
n (%)
|
ASC-H
n (%)
|
p-value
|
Relative Risk
(CI 95%)
|
|
Cytology
|
840 (66.72)
|
85 (59.02)
|
0.0775
|
1.036 (0.9960–1.078)
|
|
Cytology + DNA-HPV
|
133 (10.56)
|
13 (9.02)
|
0.6662
|
1.017 (0.9634–1.073)
|
|
DNA-HPV
|
202 (16.04)
|
28 (19.44)
|
0.2868
|
1.231 (0.8353–1.814)
|
|
Biopsy
|
74 (5.88)
|
17 (11.81)
|
0.0112
|
1.930 (1.219–3.056)
|
|
Biopsy + DNA-HPV
|
10 (0.79)
|
1 (0.69)
|
1.0000
|
1.018 (0.8397–1.222)
|
|
Total
|
1,259 (100.00)
|
144 (100.00)
|
|
|
Abbreviations: ASC-US, atypical squamous cells of undetermined significance; ASC-H,
atypical squamous cells cannot exclude high grade squamous intraepithelial lesion;
CI, confidence interval.
Fisher exact test. Statistical significance - p < 0.05.
According to national guidelines for ASC-US management, a repeat cytology is required
within 6 or 12 months. The protocol for 712 patients demanded a 6-month cytology repeat,
and the mean gap between both cytologies was 13.03 months (±11.26), ranging from 1
to 73 months. By applying a one-sample Student t-test to compare with the 6-month parameter, a statistically significant difference
was observed between cytology means and the means expected for the age group (p < 0.0001). Data from 303 patients with ASC-US cytology results whose recommendation
was to repeat the test at 12 months revealed that the mean cytology repeat time was
11.89 months (±8,68), ranging from 2 to 52 months. There is no statistically significant
difference between the mean cytology repeat time obtained and the recommended by the
national guidelines (p = 0.8324).
For ASC-US cases, out of the 973 patients followed-up with a repeat cytology, 856
(87.98%) were negative for intraepithelial lesion or malignancy (NILM), 50 remained
ASC-US (5.14%), 57 (5.86%) presented a cytological diagnosis of LSIL and 5 (0,51%)
presented HSIL in the follow-up cytology. For the ASC-H group, there were 98 cases
out of which 75 (76.23%) were NILM, and 9 (9.18%) were HSILs. The NILM result in cytological
follow-up for ASC-US is significantly more frequent than for ASC-H. It was significantly
more frequent for ASC-H that the cytology remained ASC-H or progressed to HSIL ([Table 2]).
Table 2
Cytology results as a follow-up from a prior atypical squamous cells result (n = 1,071)
|
ASC-US
n (%)
|
ASC-H
n (%)
|
p-value
|
Relative Risk
(CI 95%)
|
|
NILM
|
856 (87.98)
|
75 (76.53)
|
0.0026
|
1.100 (1.020–1.187)
|
|
ASC-US
|
50 (5.14)
|
4 (4.08)
|
0.8110
|
1.020 (0.9437–1.103)
|
|
ASC-H
|
2 (0.21)
|
3 (3.06)
|
0.0065
|
6.733 (3.209–14.13)
|
|
AGCs
|
3 (0.31)
|
2 (2.04)
|
0.0690
|
4.442 (1.492–13.22)
|
|
LSILs
|
57 (5.86)
|
5 (5.10)
|
1.0000
|
1.013 (0.9383–1.093)
|
|
HSILs
|
5 (0.51)
|
9 (9.18)
|
< 0.0001
|
7.635 (4.926–11.83)
|
|
Total
|
973 (100)
|
98 (100)
|
|
|
Abbreviations: AGCs, atypical glandular cells; ASC-H, atypical squamous cells cannot
exclude high grade squamous intraepithelial lesion; ASC-US, atypical squamous cells
of undetermined significance; CI, confidence interval; HSILs, High-grade squamous
intraepithelial lesions; LSILs, low-grade squamous intraepithelial lesions; NILM,
negative for intraepithelial lesion or malignancy.
Fisher exact test. Statistical significance - p < 0.05.
In ∼ 148 biopsies performed after an ASC-US cytology, a predominance of NILM (65.54%)
and LSIL (22.30%) was observed, and NILM was found to be more significant in this
group than in the ASC-H group. The gap between the cytology and the biopsy was on
average 4.2 (±4.4) months, ranging from 1 to 26 months. The biopsy results, such as
the ones performed as follow-up of ASC-H cytology, showed that in half of the cases,
the histopathological examination resulted in HSIL or higher. The histopathological
finding of HSIL was significantly more frequent in the ASC-H group. The results are
shown in [Table 3].
Table 3
Histopathological result as a follow-up from a previous atypical squamous cells cytology
(n = 182)
|
ASC-US
n (%)
|
ASC-H
n (%)
|
p-value
|
Relative Risk
(CI 95%)
|
|
NILM
|
97 (65.54)
|
12 (35.29)
|
0.0017
|
1.274 (1.081–1.502)
|
|
LSILs
|
33 (22.30)
|
5 (14.71)
|
0.4825
|
1.087 (0.9373–1.262)
|
|
HSILs
|
18 (12.16)
|
15 (44.12)
|
< 0.0001
|
3.565 (2.031–6.255)
|
|
Invasive Cancer
|
0 (0.00)
|
2 (5.88)
|
0.0341
|
5.625 (4.108–7.702)
|
|
Total
|
148 (100)
|
34 (100)
|
|
|
Abbreviations: ASC-H, atypical squamous cells cannot exclude high grade squamous intraepithelial
lesion; ASC-US, atypical squamous cells of undetermined significance; HSILs, High-grade
squamous intraepithelial lesions; LSILs, low-grade squamous intraepithelial lesions;
NILM, negative for intraepithelial lesion or malignancy.
Fisher exact test. Statistical significance - p < 0.05.
A total of 441 patients with ASC-US cytology underwent a high-risk-HPV (HR-HPV) test
for the follow-up. Of these, 219 (49.66%) had positive HR-HPV at follow-up. The 58
patients with ASC-H cytology were followed-up with a HR-HPV molecular test, and in
40 (68.97%) cases, the test was positive. The HR-HPV test positivity was significantly
higher for the ASC-H group ([Table 4]).
Table 4
High risk-HPV test result as a follow-up from a previous atypical squamous cells cytology
(n = 499)
|
HPV test
|
ASC-US
n (%)
|
ASC-H
n (%)
|
p-value
|
Relative Risk
(CI 95%)
|
|
Positive
|
219 (49.66)
|
40 (68.97)
|
0.0075
|
2.059
(1.215–3.491)
|
|
Negative
|
222 (50.34)
|
18 (31.03)
|
|
Total
|
441 (100)
|
58 (100)
|
Abbreviations: ASC-US, atypical squamous cells of undetermined significance; ASC-H,
atypical squamous cells cannot exclude high-grade squamous intraepithelial lesion;
CI, confidence interval.
Fisher exact test. Statistical significance - p < 0.05.
HPV test by HC2 (Qiagen) and Real time PCR (Cobas 4000).
Discussion
Cervical cancer is a well-known and preventable malignity,[13] and yet, failure to reach women with particular age risks as well as losses of follow-up
can lead to irreparable damage. In the United States, nearly half of the women with
invasive cervical cancer diagnoses have never been screened, and an additional 10%
of cancer cases occurred among untracked women in the last 5 years.[14] The goal of cytology management with an ASCs result is to find patients who may
present high-grade lesions or even underlying ICC. However, there is no single globally
acceptable strategy for these cases. The ASC-US management includes repeat cytology,
colposcopy and/or a HPV test,[5]
[15] while ASC-H requires a prompt colposcopy to be performed.[16]
To be effective, a screening program depends on the adequate follow-up of women with
abnormal cytopathologic findings, and the Brazilian program for the control of cervical
cancer has been described as ineffective.[17]
[18] This study showed that no clinical management was found for 36.97% of ASC-US patients
and for 40.50% ASC-H patients. This may occur for different reasons, but we believe
that it is due to the lack of clinical uniformity, and inadequate information to the
patients about the risks of this temporary diagnosis of ASCs. Other Brazilian authors
also observed that women with Pap test abnormalities are not adequately targeted or
referred according to the national guidelines.[17]
This paper sought to assess whether the cytological repeat procedure followed the
current national protocol. Until the launching of the Brazilian Guidelines for Cervical
Cancer Screening, in July 2011, the consensus was that described in the Brazilian
Nomenclature for Cervical Reports and Recommended Procedures, which recommended a
repeat cytology in 6 months, regardless of age, for the ASC-US group. The 2011 guidelines
indicated that, for women under 30 years of age, cytology should be repeated at 12
months, and the determination to repeat cytology should remain within 6 months for
all other ages.[17]
[19] For patients who were recommended a repeat cytology in 6 months, our results suggest
that this was performed, on average, within 13.03 months, exhibiting a statistically
significant difference in relation to the recommended time. No significant statistical
difference was observed for the group with a recommendation for repeat cytology at
12 months, with a mean of 11.89 months. This result may indicate that repeat cytology
management at 6 months has not occurred at the recommended time.
Due to the design of this study, we could not evaluate the gynecologist's conduct
in the presence of an ASC-H diagnosis, which must be followed-up with a colposcopy.
However, a substantial number of these patients will undergo biopsies, which in turn
are performed in the laboratory. Selvaggi[16] found that 64% of the patients who undergo colposcopy due to an ASC-H cytology result
will require a biopsy. In our study, only 12.5% of the patients underwent biopsy,
a frequency five times lower than that expected for those who underwent colposcopy,
which suggests that the approach recommended in these cases was not followed.
Cytology was more frequently performed at follow-up for both groups. For ASC-US, it
was performed in 77.28% of all managements, showing in most cases (87.98%) no cytological
abnormalities in the follow-up. This data supports the national recommendations of
repeat cytology ASC-US cases, due to a low frequency of persistence of abnormalities.
The persistence of the ASC-US diagnosis occurred in 5.14% of the cases; this result
is similar to that obtained in another study,[9] which was slightly less frequent than the diagnosis of LSIL (5.86%). For the ASC-US
group, only 0.5% were HSILs, which is very similar to that obtained by another study[20] that showed that, in general, the most common evolution of these cases is resolution
and, exceptionally, progression. Our study pointed out that 9.18% of the ASC-H group
exhibited HSIL. In an abstraction exercise, if this same frequency was found in patients
without a follow-up in our study population, we would have 3.8 ASC-US patients and
9 ASC-H patients progressing to HSIL, completely ignoring their condition.
Biopsies may have been performed after the ASC-US cytology result as a follow-up of
an abnormal prior cytology or based upon a recommendation of the gynecologist and
indicated by colposcopic guidance, that is, subjected to a greater severity on demand
of the specific case than for the other cases; this could result in a possible selection
bias. The results of our analyses show a 12% finding of cervical intraepithelial neoplasia
2 (CIN 2) or superior; this is much more than that found in the repetition of a cytology,
which was also superior to other studies.[5]
[15] However, it is close to the results of Bountris et al[21] in their meta-analysis. For histological follow-up from ASC-H cases, we found that
CIN 2 or superior was diagnosed in half of the cases; also, in two ASC-H cases, an
ICC was found. This is in accordance with results reported by a review of six studies
(32–66%).[16] Another study regarding the follow-up of ASCs cytology found that ICC was significantly
more associated with ASC-H.[20]
In Brazil, when cytology results show ASC-US, the recommended management is to perform
a repeat cytology; however, other countries propose to perform HR-HPV testing.[22] In our study, a HR-HPV test was performed as a follow-up procedure that was isolated
or associated with another kind of follow-up in 27.39% of cases; positivity for HR-HPV
was found in 49.66% of these cases. Watson et al[23] analyzed the follow-up of 45,049 ASC-US Pap smears and reported 42% of HR-HPV positivity,
which is very close to the result found by the systematic review of Arbyn et al,[4] while the ALTS study reported a rate slightly above 50%.[24] For ASC-H follow-up, positivity for HR-HPV (68.97%) was below that reported in the
ALTS study (85%).[2]
[6]
The results of this study should be interpreted with caution. A weakness of this research
was the impossibility of detecting when a follow-up consisted of a colposcopy, since
this information was not included in the laboratory's database but only in medical
records. Our analyses also cannot determine whether failure to perform a follow-up
was due to the patients or their physician. Further studies are needed to assess the
reality of ASCs management in Brazil, especially controlled studies.
Conclusion
Our results may indicate that ASCs management does not follow the national guidelines,
and therefore, there are possible risks for the patients in our country. Also, the
cytological, histological and HR-HPV positivity outcomes are in accordance with the
values found in the literature.
