Journal of Clinical Virology 82 (2016) 4650

Contents lists available at ScienceDirect

Journal of Clinical Virology

journal homepage: www.elsevier.com/locate/jcv

Direct comparison of two vaginal self-sampling devices for the

detection of human papillomavirus infections
M. Jentschke a, , K. Chen a,b , M. Arbyn c , B. Hertel a , M. Noskowicz a , P. Soergel a ,
P. Hillemanns a
a
Department of Gynaecology and Obstetrics, Hannover Medical School, Carl-Neuberg-Str. 1, 30625 Hannover, Germany
b
Department of Gynaecology and Obstetrics, Tongji Hospital, Tongji University, Xin Cun Road 389#, 200065 Shanghai, China
c
Scientic Institute of Public Health (WIV-ISP), 14, Rue Juliette Wytsman, 1050 Brussels, Belgium

a r t i c l e i n f o a b s t r a c t

Article history: Background and objectives: Two devices for vaginal self-sampling of dry cell material (Evalyn Brush, Rovers
Received 30 March 2016 Medical Devices; Qvintip, Aprovix) were compared using the Abbott RealTime High Risk HPV test.
Received in revised form 25 June 2016 Study design: Both self-sampling devices (change of order with every patient) including instructions for
Accepted 27 June 2016
use and a questionnaire were handed to 146 patients in a colposcopy clinic prior to scheduled colposcopies
with collection of cervical reference specimens by gynaecologists using a broom-like device. Matched
Keywords:
self-collected and physician collected specimens were transferred to ThinPrep medium and tested for
Self-sampling
the presence of hr-HPV. Biopsies were taken if indicated by colposcopy.
Cervical cancer screening
Human papillomavirus
Results: Evaluation of 136 patients with complete data (136/146; 93.2%) showed high agreement of overall
hr-HPV detection rates between self-collected and clinician-collected specimens (Evalyn: 91.2% [kappa
0.822]; Qvintip: 89.0% [kappa 0.779]). Colposcopy and histological evaluation revealed 55 women without
cervical intraepithelial neoplasia (CIN), 32 CIN1, 34 CIN2, 14 CIN3 and one adenocarcinoma in situ. Hr-HPV
testing detected all CIN3+ cases on the clinician-taken or Evalyn self-samples (14/14) and 93% of them
on the Qvintip samples (13/14). There was no signicant difference regarding the sensitivity for CIN2+
or CIN3+ and specicity of hr-HPV testing on self- vs. clinician samples and on Evalyn vs. Qvintip. Based
on signal intensities of -globin, the observed DNA concentration with Evalyn samples (mean CN: 22.0;
95%-CI: 21.522.6) was found to be signicantly higher compared to that of Qvintip samples (mean CN:
23.8; 95%-CI 23.224.4), regardless of the order of self-sampling (p < 0.0001). Most women considered
self-sampling easy and comfortable. Qvintip was considered easier than the Evalyn Brush to understand
(p < 0.001) and to use (p = 0.002).
Discussion: This study conrms that hr-HPV testing with a clinically validated PCR-based HPV assay is
as accurate on self-samples as on clinician-samples without signicant difference between both self-
sampling devices.
2016 Elsevier B.V. All rights reserved.

1. Background success of screening is the coverage rate of the respective target

High-risk human papillomavirus (hr-HPV) infection of the cer-
vicovaginal tract is known to be the major cause of cervical cancer
[1] and detection of the virus in physician-collected cervical scrapes
demonstrated superior efcacy for reducing the incidence of cer-
vical cancer compared to cervical cytology [2,3]. Various countries
have announced future implementation of hr-HPV testing as a pri-
mary screening method [46]. However, a major factor for the
Corresponding author.
E-mail address: Jentschke.Matthias@mh-hannover.de (M. Jentschke).
population. In the EU, coverage rates vary widely between member
states [7].
In Germany only 45% of the women over 20 years have yearly
screening examinations as recommended. Two thirds are screened
at least once in two years [8]. Among all patients diagnosed with
invasive cervical cancer, 60% had not been screened for at least ve
years. Another 31% did not receive Pap smears in yearly intervals.
Additionally, most of the so called non-responders were diag-
nosed with more advanced tumors (Stage T1b = clinically visible
lesions and worse) whilst 54% of the cancers diagnosed during
routine screening were microinvasive (Stage T1a = only diagnosed

http://dx.doi.org/10.1016/j.jcv.2016.06.016
1386-6532/ 2016 Elsevier B.V. All rights reserved.
 M. Jentschke et al. / Journal of Clinical Virology 82 (2016) 4650
Table 1
Evaluation of the questionnaires. Analogue scale, 111, (95%-CI).
Was it easy (=1) or difcult (=11) to use the self-sampling device?
Was it easy (=1) or difcult (=11) to follow the instructions?
Was it uncomfortable to use the self-sampling device? (no = 1; yes = 11)
Are you sure that you took a correct sample with the self-sampling device? (no = 1; yes = 11)
Which method (clinician (=1) or self-sampling (=11)) would you prefer in the future?
Did you have any problems or discomfort using
the self-sampling device?
*
Wilcoxon signed-rank test.
microscopically) [9]. Similar data were reported from Sweden, The
Netherlands and California [1013].
One of the most promising concepts to improve the participa-
tion rate among non-responders is to provide self-sampling devices
designed for home-collection of cervicovaginal material. HPV test-
ing on self-collected and physician-collected cervical specimens
with clinically validated PCR-based tests generally shows similar
sensitivity for high-grade CIN on both sampling methods [14,15].
Numerous studies have been conducted with a diversity of indi-
vidual self-sampling devices but little attention has been given to
the direct comparison of different self-sampling devices.
2. Objectives
This pilot study was conducted to compare two dry vaginal
self-sampling devices (Evalyn Brush, Rovers Medical Devices and
Qvintip, Aprovix) in combination with the Abbott RealTime High
Risk HPV test (RealTime; Abbott GmbH & Co. KG, Wiesbaden,
Germany).
3. Study design
At the colposcopy clinic and the gynaecological outpatient clinic
of Hannover Medical School, Germany, study participants were
recruited among the patients referred for abnormal cervical screen-
ing results or general gynaecological diseases. Pregnant women
and women with hysterectomy in the past were excluded. The
study was approved by the institutional ethics review board prior to
beginning and all participants gave written consent before enroll-
ment.
At rst, all participants were given the two sampling devices
(alternating order in every patient), written and illustrated instruc-
tions as provided by the manufacturers (translated to German) and
a questionnaire addressing personal and medical history and the
acceptance of the self-sampling devices (ten point analogue scale;
see Table 1). If there were missing answers concerning one self-
sampling device on the questionnaires the answers concerning the
other device were excluded from the nal analysis to have equal
data sets for both devices.
After completion of self-sampling procedures and question-
naires in a separate room at the clinics and without assistance
by hospital staff, all women received their scheduled examina-
tion by a gynaecologist. At rst, a liquid-based cervical cytology
smear was taken with a broom-like device (Hologic, Marlborough,
MA) and immediately suspended in 20 ml of Cytyc ThinPrep Pre-
servCyt Solution (Hologic). This sample was also used as reference
HPV sample. Then colposcopy was performed using acetic acid
and taking directed biopsies and/or endocervical curettage if indi-
cated (routine cyto- and histopathological examination). In case of
unsuspicious colposcopy there were no biopsies taken. Histology
was classied into low-, moderate- and severe cervical intraep-
ithelial neoplasia (CIN 1-3) and adenocarcinoma in situ [16]. The
nal diagnosis for each woman was dened according to the worst
histological or cytological result.
47
Evalyn Qvintip p-value*
3.13 (2.703.57) 2.42 (2.082.76) p = 0.002
2.96 (2.543.37) 2.15 (1.862.43) p < 0.001
2.84 (2.453.24) 2.55 (2.192.91) p = 0.122
6.29 (5.806.77) 6.32 (5.796.84) p = 0.853
4.30 (3.724.89) 4.16 (3.564.75) p = 0.418
Yes: 8.1% Yes: 5.9% n.a.
No: 88.2% No: 89.0%
Self-samples were stored at room temperature for 89 days on
average (range, 167 days) then they were thoroughly washed in
20 ml of PreservCyt Solution for at least ve seconds until visible cell
material was transferred to the medium. The physician-collected
reference samples were at rst used to prepare liquid based cytol-
ogy slides with the ThinPrep 2000 system (Hologic). Then 700 l
each of the self samples and residual reference sample material
were used for hr-HPV testing with the RealTime assay on the m2000
System (Abbott).
3.1. Self sampling devices
The Evalyn Brush has a broom-like collection head that can be
retracted into a tube. It is about 20 cm long and has lateral wings
indicating the necessary depth of insertion. After insertion, the
brush has to be rotated ve times, then it has to be retracted and
covered by a cap and can be sent by mail.
The Qvintip collection device has a solid white plastic head that
is about 5 cm long and 7 mm thick. The head has multiple groves
to collect the cervicovaginal cell material. The device has to be
inserted as far as possible, then it is removed and the plastic head
is transferred into a separate tube without touching and further
manipulation. This tube can also be sent by mail, the rest of the
device is disposed of.
3.2. Hr-HPV Test
The Abbott RealTime High Risk HPV test is an automated, quali-
tative multiplex assay based on real time polymerase chain reaction
(PCR) for the detection of 14 h-HPV genotypes (16, 18, 31, 33, 35, 39,
45, 51, 52, 56, 58, 59, 66 and 68) and simultaneous differentiation of
HPV 16 and HPV 18 as described before [17]. It has been clinically
validated for routine use in cervical cancer screening in different
countries [17,18] and proved to be accurate on clinician samples in
a referral population [19] and for hr-HPV testing of self-collected
lavage samples [20].
3.3. Statistical analysis
Microsoft Excel 2010 (Microsoft Corp., Redmond, WA) and IBM
SPSS Statistics 22 (IBM Corp., Armonk, NY) were used for data col-
lection and evaluation. Cohens Kappa was calculated to assess
the agreement between the different sampling methods and the
Wilcoxon signed-rank test (two-sided p value) was used compare
the mean real-time PCR cycle number (CN) values and acceptance
rates as indicated on the questionnaires [21].
4. Results
Overall, 146 women were recruited for the study. Of these, ten
cases had to be excluded from the nal analysis: two cases of
women who initially agreed to participate but who did not perform
the study procedures, three cases without reference smears and one
case with only one self-sample. Four cases with invalid RealTime
48 M. Jentschke et al. / Journal of Clinical Virology 82 (2016) 4650

Table 2 sis, where the invalid samples were counted as not detected there
Results of partial hr-HPV genotyping (no. of cases with HPV 16, 18, other hr-HPV
were also no signicant differences between the different sampling
genotypes or multiple infections).
methods (see tables s1 and s2).
N = 136 included cases Reference sample Evalyn Qvintip The signal intensity for globin (necessary number of real-
HPV 16 17 16 18 time PCR cycles (CN) until the amplied DNA fragments reach the
HPV 18 6 5 4 predened cutoff values; this number inversely correlates with
Other HR HPV 41 40 36 the amount of DNA in the sample) observed with Evalyn samples
HPV 16 & other hr-HPV only 10 11 9
(mean CN: 22.0; 95%-CI: 21.522.6) was found to be signicantly
HPV 18 & other hr-HPV only 0 0 0
HPV 16 & HPV 18 only 0 0 0 higher compared to that of Qvintip samples (mean CN: 23.8; 95%-
HPV 16 & HPV 18 & other hr-HPV only 1 1 1 CI 23.224.4), regardless of the order of self-sampling (p < 0.0001;
Overall detection of any hr-HPV type 75 73 68 Wilcoxon signed-rank test).
The response rate of the different questions ranged between
Table 3
91.1 and 98.5%. Self-sampling was generally considered easy and
Agreement of self- and clinician sampling. not uncomfortable with both devices while using the Qvintip
and following the Qvintip instruction was signicantly easier than
Evalyn Reference sample
for Evalyn (p = 0.002 and p < 0.001; Wilcoxon signed-rank test; see
hr-HPV positive hr-HPV negative Table 1). There was no correlation of the ease of using the Evalyn
Hr-HPV positive 68 5 73 Brush with the users age but using the Qvintip was considered
Hr-HPV negative 7 56 63 signicantly easier by women between 40 and 50 years of age in
75 61 136 comparison to women <30 years (p = 0.004; analysis of variance).
Most women (63%) felt sure that they were able to collect a cor-
Qvintip Reference sample
rect sample with both devices. Finally, even if self-sampling would
hr-HPV positive hr-HPV negative prove to be as accurate as clinician sampling most women (64%)
hr-HPV positive 64 4 68 would prefer clinician sampling in the future.
hr-HPV negative 11 57 68
75 61 136

5. Discussion
test results had to be excluded, too (per protocol analysis). All of
the four reference samples had valid hr-HPV test results but there
In many countries new guidelines are developed for cervical
were two Evalyn samples and two Qvintip samples with insuf-
cancer screening [4,5,22] with HPV testing being implemented
cient cell material ( globin above CN threshold). This resulted in
in primary screening programs. However, women refraining from
one HPV 16 positive CIN 3 with an inadequate Evalyn sample and
screening continue to be at a high risk for developing cervical can-
three hr-HPV negative cases with inadequate self-samples.
cer [9]. Improving their participation rate is a very important way
A total of 136 cases with complete sets of test results were
to reduce cervical cancer incidence. Self-sampling can help to reach
included in the per protocol evaluation. Mean patient age was
these screening non-responders [2325].
36 years (range; 1778). An overall hr-HPV positivity rate of 55%
This is one of the rst studies directly comparing the perfor-
(75/136) was found in the reference sample population with mostly
mance of two commercial self-sampling devices using a clinically
non-HPV16/18 genotypes detected. Multiple infections were found
validated hr-HPV DNA test. Both devices showed very similar hr-
in 11 specimens (8%; see Table 2).
HPV detection rates and a high agreement with clinician-collected
Very high agreement of hr-HPV reactivity rates was found
cervical samples among 136 patients. There was a detection rate
between self-collected samples and physician-collected refer-
of over 90% for CIN3+ and of over 83% for CIN2+ with both self-
ence samples (Evalyn: 91.2% [kappa 0.822; 95%-CI = 0.726 0.918];
sampling devices.
Qvintip: 89.0% [kappa 0.779; 95%-CI = 0.674 0.885]; see Table 3).
We also performed an intention to treat analysis where we
No signicant differences between the agreement rates for self- and
counted the four inadequate samples as hr-HPV not detected.
clinician sampling in relation to the order of self-sampling devices
Our study did not have the possibility to re-invite patients with
used were observed.
invalid test results for retesting. On average, 20% can be reached
There were 34 CIN 2, 14 CIN 3 and one adenocarcinoma in situ
by offering self-samplers [23]. In this difcult to reach population,
among the 136 included cases. All CIN 3+ cases were detected with
there is a considerable risk of non-participation after an inadequate
Evalyn, one CIN 3 was missed with Qvintip (reference and Evalyn:
self-sample and possible non-detection of high-grade dysplasia.
non 16/18 HPV positive; see Table 4). There were no signicant dif-
However, there were no signicant differences in the accuracy for
ferences between the sensitivity and specicity for the detection
CIN 2+ between the two sampling devices in both the per protocol
of CIN2+ between clinician sampling, Evalyn and Qvintip in the per
and the intention to treat analysis.
protocol analysis (see Tables 5 and 6). In the intention to treat analy-
Evaluation of the signal intensity (CN) of Internal Control
(human globin) values reported by RealTime revealed a sig-
Table 4 nicantly higher overall globin signal with Evalyn compared to
hr-HPV positivity rates, by clinical outcome. Qvintip (p < 0.0001). This observation indicates that the amount of
Final diagnosis N= hr-HPV positive samples cellular material collected with Qvintip is lower than with Evalyn.
It is unclear whether this is driven by the design of the devices
Reference sample Evalyn Qvintip
heads or the more thorough sampling procedure (ve rotations
No evidence of 55 11 (20%) 10 (18%) 8 (15%) with Evalyn, one rotation with Qvintip). However, since RealTime
disease/dysplasia
CIN 1 32 20 (63%) 19 (59%) 19 (59%)
is designed to optimize clinical performance rather than accurate
CIN 2 34 29 (85%) 29 (85%) 27 (79%) quantication of pathogen load these results have to be inter-
CIN 3 14 14 (100%) 14 (100%) 13 (93%) preted with caution. Interestingly, the lower cellular concentration
AIS 1 1 (100%) 1 (100%) 1 (100%) of Qvintip did not inuence its clinical performance (detection of
Total 136 75 (55%) 73 (54%) 68 (50%)
high-grade CIN) with the analytical procedures used in this study.
 M. Jentschke et al. / Journal of Clinical Virology 82 (2016) 4650
Table 5
Sensitivity and specicity for CIN2+ (per protocol analysis; n = 136).
Type of sample TP FN FP TN
Clinician sample 44 5 31 56
Evalyn Brush 44 5 29 58
Qvintip 41 8 27 60
TP: true positive; FN: false negative; FP: false positive; TN: true negative.
The acceptance of both sampling devices was high among the
study population. Usage of Qvintip was considered easier than of
the Evalyn Brush. Possibly the Evalyn self-sampling process is a
little more complicated than using the Qvintip. Surprisingly, using
the Qvintip was considered signicantly easier by women between
40 and 50 years of age in comparison to women <30 years.
There have been many studies evaluating the feasibility and per-
formance of self-sampling for hr-HPV before but only few studies
compared different sampling devices with one another. Igidbashian
et al. compared the acceptability of two different self-sampling
methods but this was done in two separate groups of women
and each woman only used one of the devices [26]. The authors
report a high acceptance of self-sampling in favor of the lavage
device and in contrast to our results a higher preference of self-
sampling in comparison to clinician sampling. Recently, Bosgraaf
et al. published a large study among over 30,000 screening non-
responders that either received a lavage device or the Evalyn Brush
[27]. The detection rates for hr-HPV and CIN in this screening pop-
ulation cannot be compared to our results among women referred
for colposcopy. However, the evaluation of the questionnaires also
showed high acceptance of both devices and only little discomfort.
Both sampling devices used in our study have been evaluated in
previous studies. Evaluation of the Evalyn Brush in a similar referral
setting showed an agreement with clinician-collected HPV samples
of about 85% (91.2% in our study) [28]. There are several studies
where Qvintip was used for self-sampling [2934] but there was
not a comparison to clinician sampling in all of them. Stenvall et al.
reported a kappa value of 0.72 for hr-HPV detection using Qvintip
sampling and clinician sampling (0.779 in our study) and a high
acceptance of Qvintip [34].
The results of this study can only partially be transferred to a
general screening situation. The women participating in this study
mostly attended cervical screening in the recommended yearly
intervals and the education level was rather high. Their attitude
towards cervical cancer screening in general and self-sampling
might differ from screening-deniers who would be the main tar-
get population for self-sampling. Additionally, the results could be
different with other hr-HPV tests and can only be generalized for
similar hr-HPV detection assays. Furthermore, transferring every
self-sample to 20 ml of ThinPrep solution is costly, time consuming,
labor intensive and comprises risk of sample mislabeling and cross-
contamination and thus may not suit the needs of high-throughput
routine settings. However, the pre-analytic procedure applied the
for self-collected samples in this study allowed to minimize the
number of variables between the matched self- and clinician sam-
pled specimens, thus minimizing potential bias. Conducting this
study at a colposcopy clinic helped to motivate women to partici-
pate and to keep the dropout rate low. Still, sending all participants
to a separate room for self-sampling without assistance by medi-
Table 6
Relative sensitivity and specicity for CIN2+ (per protocol analysis; n = 136).
Comparison Relative Sensitivity(95% CI) Relative Specicity(95% CI)
Evalyn vs. clinician 1.00 (0.881.14) 1.04 (0.841.29)
Qvintip vs. clinician 0.93 (0.801.09) 1.07 (0.871.32)
Qvintip vs. Evalyn 0.93 (0.801.09) 1.03 (0.871.32)
49
Total CIN2+ Sensitivity(95% CI) CIN2+ Specicity(95% CI)
136 89.8% (81.398.3) 64.4% (54.374.4)
136 89.8% (81.398.3) 66.7% (56.876.6)
136 83.7% (73.394.0) 69.0% (59.278.7)
cal personnel should reected the real self-sampling process quite
well.
In summary, this comparative study of two dry self-sampling
devices showed a very good agreement of hr-HPV positivity rates
between clinician sampling and both devices and a detection rate of
CIN3+ over 90% with both devices. The acceptance of self-collection
was very high and most women experienced no difculties or
discomfort during the procedure. In the future the potential of
alternative strategies including sending self-samplers has to be
evaluated among a larger group of screening deniers to further
validate their use in a general organized screening setting.
Funding
M. Arbyn was supported by the seventh framework program of
DG Research of the European Commission, through the COHEAHR
Network (grant No 603019) and by the Joint Action CANCON which
has received funding from the European Union in the framework
of the Health Programme (2008-13). Other funding was received
from the German Guideline Program in Oncology (German Cancer
Aid project # 110163).
Conict of interest
M. Jentschke received payment for travel expenses and lecture
fees from Abbott Molecular GmbH & Co. KG, Wiesbaden, Germany.
Abbott Molecular GmbH & Co. KG, Wiesbaden, Germany pro-
vided the hr-HPV test kits at reduced cost.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.jcv.2016.06.016.
References
[1] J.M. Walboomers, M.V. Jacobs, M.M. Manos, F.X. Bosch, J.A. Kummer, K.V.
Shah, et al., Human papillomavirus is a necessary cause of invasive cervical
cancer worldwide, J. Pathol. 189 (1999) 1219.
[2] G. Ronco, J. Dillner, K.M. Elfstrom, S. Tunesi, P.J. Snijders, M. Arbyn, et al.,
Efcacy of HPV-based screening for prevention of invasive cervical cancer:
follow-up of four European randomised controlled trials, Lancet 383 (2014)
524532.
[3] M. Arbyn, G. Ronco, A. Anttila, C.J. Meijer, M. Poljak, G. Ogilvie, et al., Evidence
regarding human papillomavirus testing in secondary prevention of cervical
cancer, Vaccine 30 (Suppl 5) (2012) F8899.
[4] M. Arbyn, A. Haelens, A. Desomer, F. Verdoodt, J. Francart, N. Thiry, et al.
Cervical cancer screening program and Human Papillomavirus (HPV) testing,
part II: Update on HPV primary screening. KCE Belgian Healthcare Knowledge
Centre; 2015.
[5] D. Saslow, D. Solomon, H.W. Lawson, M. Killackey, S.L. Kulasingam, J. Cain,
et al., American Cancer Society, American Society for Colposcopy and Cervical
Pathology, and American Society for Clinical Pathology screening guidelines
for the prevention and early detection of cervical cancer, CA. Cancer J. Clin. 62
(2012) 147172.
[6] Medical Services Advisory Committee (MSAC). Application No. 1276Renewal
of the National Cervical Screening Program. Australian government,
department of health; 2014.
[7] K.M. Elfstrom, L. Arnheim-Dahlstrom, L. von Karsa, J. Dillner, Cervical cancer
screening in Europe: quality assurance and organisation of programmes, Eur.
J. Cancer 51 (2015) 950968.
[8] S. Geyer, J. Jaunzeme, P. Hillemanns, Cervical cancer screening in Germany:
group-specic participation rates in the state of Niedersachsen (Lower
50 M. Jentschke et al. / Journal of Clinical Virology 82 (2016) 4650
Saxony). A study with health insurance data, Arch. Gynecol. Obstet. 291
(2014) 623629.
[9] K. Marquardt, H.H. Buttner, U. Broschewitz, M. Barten, V. Schneider, Persistent
carcinoma in cervical cancer screening: non-participation is the most
signicant cause, Acta Cytol. 55 (2011) 433437.
[10] B. Andrae, L. Kemetli, P. Sparen, L. Silfverdal, B. Strander, W. Ryd, et al.,
Screening-preventable cervical cancer risks: evidence from a nationwide
audit in Sweden, J. Natl. Cancer Inst. 100 (2008) 622629.
[11] W.A. Leyden, M.M. Manos, A.M. Geiger, S. Weinmann, J. Mouchawar, K.
Bischoff, et al., Cervical cancer in women with comprehensive health care
access: attributable factors in the screening process, J. Natl. Cancer Inst. 97
(2005) 675683.
[12] H.Y. Sung, K.A. Kearney, M. Miller, W. Kinney, G.F. Sawaya, R.A. Hiatt,
Papanicolaou smear history and diagnosis of invasive cervical carcinoma
among members of a large prepaid health plan, Cancer 88 (2000) 22832289.
[13] A. Zaal, M.A. de Wilde, M.J. Duk, G.C. Graziosi, M. van Haaften, S. von
Mensdorff-Pouilly, et al., The diagnostic process of cervical cancer; areas of
good practice, and windows of opportunity, Gynecol. Oncol. 138 (2015)
405410.
[14] M. Arbyn, P.E. Castle, Offering self-Sampling kits for HPV testing to reach
women who do not attend in the regular cervical cancer screening program,
Cancer Epidemiol. Biomarkers Prev. 24 (2015) 769772.
[15] M. Arbyn, F. Verdoodt, P.J. Snijders, V.M. Verhoef, E. Suonio, L. Dillner, et al.,
Accuracy of human papillomavirus testing on self-collected versus
clinician-collected samples: a meta-analysis, Lancet Oncol. 15 (2014)
172183.
[16] R.M. Richart, Cervical intraepithelial neoplasia, Pathol. Annu. 8 (1973)
301328.
[17] M. Poljak, A. Ostrbenk, The Abbott RealTime High Risk HPV test is a clinically
validated human papillomavirus assay for triage in the referral population
and use in primary cervical cancer screening in women 30 years and older: a
review of validation studies, Acta Dermatovenerol. Alp Pannonica Adriat. 22
(2013) 4347.
[18] M. Arbyn, P.J. Snijders, C.J. Meijer, J. Berkhof, K. Cuschieri, B.J. Kocjan, M.
Poljak, Which high-risk HPV assays full criteria for use in primary cervical
cancer screening? Clin. Microbiol. Infect. 21 (2015) 817826.
[19] M. Jentschke, P. Soergel, V. Lange, B. Kocjan, T. Doerk, A. Luyten, et al.,
Evaluation of a new multiplex real-time polymerase chain reaction assay for
the detection of human papillomavirus infections in a referral population, Int.
J. Gynecol. Cancer 22 (2012) 10501056.
[20] M. Jentschke, P. Soergel, P. Hillemanns, Evaluation of a multiplex real time PCR
assay for the detection of human papillomavirus infections on self-collected
cervicovaginal lavage samples, J. Virol Methods 193 (2013) 131134.
[21] J.L. Fleiss, Statistical Methods for Rates and Proportions, 2 ed., John Wiley &
Sons, New York, 1981.
[22] M. Jentschke, P. Hillemanns, Prvention des Zervixkarzinoms. Das wird die
neue S3-Leitlinie bercksichtigen, gynkologie + geburtshilfe 17 (2012)
2225.
[23] F. Verdoodt, M. Jentschke, P. Hillemanns, C.S. Racey, P.J. Snijders, M. Arbyn,
Reaching women who do not participate in the regular cervical cancer
screening programme by offering self-sampling kits: a systematic review and
meta-analysis of randomised trials, Eur. J. Cancer 51 (2015) 23752385.
[24] M. Gk, D.A. Heideman, F.J. van Kemenade, J. Berkhof, L. Rozendaal, J.W.
Spruyt, et al., HPV testing on self collected cervicovaginal lavage specimens as
screening method for women who do not attend cervical screening: cohort
study, BMJ 340 (2010) c1040.
[25] P. Hillemanns, R. Kimmig, U. Huttemann, C. Dannecker, C.J. Thaler, Screening
for cervical neoplasia by self-assessment for human papillomavirus DNA,
Lancet 354 (1999) 1970.
[26] S. Igidbashian, S. Boveri, N. Spolti, D. Radice, M.T. Sandri, M. Sideri,
Self-collected human papillomavirus testing acceptability: comparison of two
self-sampling modalities, J. Womens Health (Larchmt) 20 (2011) 397402.
[27] R.P. Bosgraaf, V.M. Verhoef, L.F. Massuger, A.G. Siebers, J. Bulten, G.M. de
Kuyper-de Ridder, et al., Comparative performance of novel self-sampling
methods in detecting high-risk human papillomavirus in 30,130 women not
attending cervical screening, Int. J. Cancer 136 (2014) 646655.
[28] R. van Baars, R.P. Bosgraaf, B.W. ter Harmsel, W.J. Melchers, W.G. Quint, R.L.
Bekkers, Dry storage and transport of a cervicovaginal self-sample by use of
the Evalyn Brush, providing reliable human papillomavirus detection
combined with comfort for women, J. Clin. Microbiol. 50 (2012) 39373943.
[29] K. Sanner, I. Wikstrom, A. Strand, M. Lindell, E. Wilander, Self-sampling of the
vaginal uid at home combined with high-risk HPV testing, Br. J. Cancer 101
(2009) 871874.
[30] H. Stenvall, I. Wikstrom, E. Wilander, High prevalence of oncogenic human
papilloma virus in women not attending organized cytological screening, Acta
Derm. Venereol. 87 (2007) 243245.
[31] I. Wikstrom, M. Lindell, K. Sanner, E. Wilander, Self-sampling and HPV testing
or ordinary pap-smear in women not regularly attending screening: a
randomised study, Br. J. Cancer 105 (2011) 337339.
[32] U. Gyllensten, K. Sanner, I. Gustavsson, M. Lindell, I. Wikstrom, E. Wilander,
Short-time repeat high-risk HPV testing by self-sampling for screening of
cervical cancer, Br. J. Cancer 105 (2011) 694697.
[33] M. Lindell, K. Sanner, I. Wikstrom, E. Wilander, Self-sampling of vaginal uid
and high-risk human papillomavirus testing in women aged 50 years or older
not attending Papanicolaou smear screening, BJOG 119 (2012) 245248.
[34] H. Stenvall, I. Wikstrom, E. Wilander, Human papilloma virus testing of
vaginal smear obtained with a novel self-sampling device, Acta Derm.
Venereol. 86 (2006) 465467.