G Model

EURO 9622 1–5

European Journal of Obstetrics & Gynecology and Reproductive Biology xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

European Journal of Obstetrics & Gynecology and

Reproductive Biology
journal homepage: www.elsevier.com/locate/ejogrb

1 Punch biopsy guided by both colposcopy and HR-HPV status is more

2 efficient for identification of immediate high-grade squamous
3 intraepithelial lesion or worse among HPV-infected women with
4 atypical squamous cells of undetermined significance
5 Q1 Z. Dinga,b , Y. Lic , A. Chenb , M. Songd, Y. Zhanga,*
6 a
Qilu Hospital, Shandong University, Jinan, Shandong, China
7 b
The Affiliated Hospital of Qingdao University, Qingdao, Shandong, China
8 c
The Second People’s Hospital of Liao Cheng, Liao Cheng, Shandong, China
9 d
Qingdao Tumour Hospital, Qingdao, Shandong, China

A R T I C L E I N F O A B S T R A C T

Article history: Objectives: To investigate the accuracy of colposcopy for diagnosing high-grade squamous intraepithelial
Received 15 May 2016 lesion (HSIL) or worse (HSIL+) in human papillomavirus (HPV)-infected patients with atypical squamous
Received in revised form 11 August 2016 cells of undetermined significance (ASCUS) cytology, and determine whether genotyping and viral load
Accepted 10 October 2016
quantitation can be useful for detecting immediate HSIL+ risk in these patients.
Available online xxx
Study design: This study included 620 cases with ASCUS and positive for high-risk (HR)-HPV within
1 month before or after cervical cytology at Qilu Hospital between February 2013 and February 2014.
Keywords:
Based on the colposcopic impression, lesion-targeted punch biopsy, endocervical curettage biopsy or
Atypical squamous cells of undetermined
significance
random cervical punch biopsy in four quadrants was performed on these patients within 1 month. The
Colposcopy accuracy of colposcopy for diagnosing HSIL+ was evaluated through comparison with the biopsy results.
HPV viral load HR-HPV status determined by Hybrid Capture 2 or HPV genotyping was analysed retrospectively as a
HPV genotype possible predictor of HSIL+.
High-grade squamous intraepithelial lesion Results: Agreement between colposcopic impression and cervical pathology was matched perfectly in
89.2% of cases (553/620), and the strength of agreement with the k statistic was 0.698 (p < 0.001).
Colposcopy had high specificity (96.9%) but low sensitivity for detecting HSIL+ (67.5%). The risk of HSIL+
was significantly higher in patients with HPV-16 infection (52.3%) than in patients infected with other
Q3 types of HPV (17.9%, p < 0.001). HSIL+ and virus load at cut-offs (CO) of 50 relative light units (RLU)/CO
and 100 RLU/CO (p = 0.024 and 0.044, respectively). If considering HPV16 infection or high virus load (at
50 RLU/CO) as a diagnostic standard of HSIL+ when colposcopic impression was negative, sensitivity was
improved to 74.7% and 81.0%, respectively.
Conclusions: Good agreement was found between colposcopic and pathologic diagnosis. HR-HPV
genotyping or virus load is relevant to the detection of HSIL+ among HPV-infected patients with ASCUS
cytology. In these patients, biopsies considering HPV-16 infection or virus load 50 RLU/CO may be
helpful for increasing the HSIL+ detection rate.
ã 2016 Elsevier Ireland Ltd. All rights reserved.

10 Introduction High-quality screening with cytology (Pap tests) has played an 15

essential role in the early detection of these lesions, leading to a 16

11 Although considered one of the main problems in female health significant decrease in squamous cell cervical cancer mortality 17
12 in most developing countries, invasive cervical cancer (ICC) is [1,2]. The most common abnormal diagnosis in cytology-based 18
13 preventable by early detection and treatment of its precursor cervical screening is atypical squamous cells of undetermined 19
14 (precancer), high-grade squamous intraepithelial lesion (HSIL). significance (ASCUS). Accurate identification of women with HSIL 20

or worse (HSIL+) among those referred because of ASCUS cervical 21

smear results has had a major impact on patient management and 22

Q2 triage. Testing for the presence of human papillomavirus (HPV) is 23

* Corresponding author.
E-mail address: yuetutu1979@163.com (Y. Zhang). widely accepted for triaging ASCUS, as it has higher accuracy than 24

http://dx.doi.org/10.1016/j.ejogrb.2016.10.005
0301-2115/ã 2016 Elsevier Ireland Ltd. All rights reserved.

Please cite this article in press as: Z. Ding, et al., Punch biopsy guided by both colposcopy and HR-HPV status is more efficient for identification
of immediate high-grade squamous intraepithelial lesion or worse among HPV-infected women with atypical squamous cells of undetermined
significance, Eur J Obstet Gynecol (2016), http://dx.doi.org/10.1016/j.ejogrb.2016.10.005
 G Model
EURO 9622 1–5

2 Z. Ding et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology xxx (2016) xxx–xxx

25 repeat cytology methods and is considered to be the most cost- Colposcopy and punch biopsy 84
26 effective strategy [3–5]. Accordingly, the authors use this strategy
27 Q4 in management of women with ASCUS. All 620 patients underwent colposcopy examination using the 85
28 Among patients with a diagnosis of ASCUS with high-risk (HR)- 2011 colposcopic terminology of the International Federation for 86
29 HPV infection, colposcopy with guided pathological biopsy is the Cervical Pathology and Colposcopy (IFCPC) and punch biopsy 87
30 gold-standard strategy for determining if treatment is needed. within 1 month. All colposcopies were performed by an obstetrics 88
31 Colposcopy is useful for estimating lesion grade, particularly in and gynecology attending physician who had worked in this field 89
32 women with HSIL on a Pap smear, but its accuracy for detecting for 10 years. Colposcopy was performed routinely with 5% acetic 90
33 HSIL+ in women with ASCUS is unsatisfactory [6]. To date, however, acid and Schiller’s test. 91
34 there has been little discussion on the accuracy of colposcopy for The 2011 colposcopic terminology of the cervix table starts with 92
35 detecting HSIL+ in HPV-infected women with ASCUS. This study ‘General assessment: adequate or inadequate’; the preliminary 93
36 involved colposcopy examination and punch biopsies in women diagnostic results according to the 2011 colposcopic terminology 94
37 with ASCUS who were HR-HPV-positive. The accuracy of colposcopy of IFCPC were established and included: suspicious for invasion, 95
38 for diagnosing HSIL+ was evaluated by comparing the findings with major lesions, minor lesions, and normal colposcopic findings. 96
39 histological test results. In addition, the clinical characteristics and Preliminary results of normal colposcopic findings and minor 97
40 HR-HPV test results were analysed retrospectively. The primary lesions were grouped as negative. Results of suspicious for invasion 98
41 objective of this study was to evaluate the prevalence of HSIL+ and and major lesions were classified as positive. 99
42 the accuracy of colposcopy for detecting HSIL+. The secondary
43 objectives were to evaluate the value of HPV genotyping and viral Biopsy pathology 100
44 load quantitation for predicting immediate HSIL+.
A biopsy was performed systematically during the colposcopy. 101
45 Materials and methods In adequate examinations, a colposcopy-directed, lesion-targeting 102

punch biopsy was performed when there were abnormal 103

46 Patients colposcopic findings. More than three biopsy samples were taken 104

from the most serious to the normal area. For normal colposcopic 105
47 The pathology database was searched retrospectively for findings, random cervical punch biopsies were performed in four 106
48 women who had a positive HR-HPV result within 1 month before quadrants from the squamocolumnar junction. In inadequate 107
49 or after a novel diagnosis of ASCUS between February 2013 and examinations, endocervical curettage biopsy was added. Two 108
50 February 2014. These patients had all undergone colposcopy experienced specialist gynaecological pathologists at the study 109
51 within 1 month during this period. Patients who had undergone institution processed all biopsies routinely, and evaluated them 110
52 treatment for cervical cancer or its precursor previously, and using the standard histological criteria of the new World Health 111
53 women who were pregnant or immunosuppressed were excluded. Organization classification for cervical histology results. All 112
54 This study was conducted with the approval of the institutional positive results mentioned above were determined without any 113
55 review board of Qilu Hospital. reference to the patients’ clinical information. 114

56 Cervical cytology Study design 115

57 Cervical cytologies were performed using a liquid-based The cytological test results and HR-HPV virus loads as 116
58 cervical cytology technique (ThinPrep, Cytyc Corp., Boxborough, determined by Hybrid Capture 2 or HPV genotyping were assessed. 117
59 MA, USA), and read at the routine pathology centres for diagnosis Possible predictive factors for HSIL+ lesions in the biopsy speci- 118
60 by a cytopathologist. ASCUS criteria in accordance with the mens were evaluated, such as age, Hybrid Capture 2 viral load, HR- 119
61 2001 Bethesda system classification were used [7–9]. Cytological HPV genotype and colposcopic impression. Age was divided into 120
62 diagnoses were made without knowledge of HPV status. four groups: <30, 30–40, 40–50 and 50 years. In this study, the 121

HR-HPV RLU/CO of all samples was recorded and divided into eight 122
63 Detection of HPV infection classes: (1) 1  RLU/CO < 10; (2) 10  RLU/CO < 50; (3) 50  RLU/ 123

CO < 100; (4) 100  RLU/CO < 200; (5) 200  RLU/CO < 500; (6) 124
64 Among 620 patients, 371 had HPV semiquantitative virus load 500  RLU/CO < 1000; (7) 1000  RLU/CO < 2000; and (8) 2000  125
65 results. These results were obtained using Hybrid Capture 2 RLU/CO. The HPV genotyping results were divided into single or 126
66 (Qiagen, Valencai, CA, USA) performed according to the manu- multiple infections according to number of genotypes present, and 127
67 facturer’s recommendations. The HR-HPV panel consisted of were compared according to the presence or absence of 128
68 13 HPV types (16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 68). HPV16 infection. 129
69 The results are reported as the ratio of the test sample relative light
70 unit to that of the cut-off value (i.e. RLU/CO). The manufacturer- Statistical analysis 130
71 established positive result was 1.0 RLU/CO.
72 HPV genotyping was performed in 254 patients using HybriBio Categories for continuous variables were compared using t-test. 131
73 Q5 HPV GenoArray (HybriBio Limited, China). The test is a polymer- Pearson’s x2 test and Fisher’s exact test, as appropriate, were used 132
74 ase-chain-reaction-based HPV type-specific assay that uses to evaluate each factor in relation to HSIL+ between the pairs of 133
75 L1 consensus primers to amplify 21 HPV genotypes simultaneous- factors studied. The significance of agreement between colpo- 134
76 ly: six low-risk genotypes (6, 11, 42, 43, 44, CP8304) and 15 high- scopic diagnosis and cervical pathology was determined using 135
77 risk genotypes (16, 18, 31, 33, 35, 39, 45, 51, 52, 53, 56, 58, 59, 66, weighted k statistics. Colposcopic diagnosis and cervical pathology 136
78 68), followed by flow-through hybridization with immobilized were compared based on sensitivity, specificity, positive predictive 137
79 genotype-specific probes. The assay was performed according to value (PPV), negative predictive value (NPV), false-positive results 138
80 the manufacturer’s protocol. and false negative results. The diagnostic ability of the different 139
81 Only five patients had been tested using both methods and all RLU/CO classes was assessed by comparing the area under the 140
82 had positive results. The remaining patients were tested using a curve (AUC) of the receiver operating characteristic (ROC). All 141
83 single method. analyses were performed using Statistical Package for the Social 142

Please cite this article in press as: Z. Ding, et al., Punch biopsy guided by both colposcopy and HR-HPV status is more efficient for identification
of immediate high-grade squamous intraepithelial lesion or worse among HPV-infected women with atypical squamous cells of undetermined
significance, Eur J Obstet Gynecol (2016), http://dx.doi.org/10.1016/j.ejogrb.2016.10.005
 G Model
EURO 9622 1–5

Z. Ding et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology xxx (2016) xxx–xxx 3

143 Sciences Version 21.0 (IBM Corp., Armonk, NY, USA). All p-values (443/457, 96.9%), and there were 3.1% false-positive and 32.5% 167
144 were two-sided and p < 0.05 was considered statistically signifi- false-negative results. 168
145 cant.
Predictors of HSIL+ 169
146 Results
Table 2 shows the final pathological diagnosis for the different 170
147 Patient characteristics and prevalence of HSIL+ risk factor levels and the predictors of HSIL+. There was no 171

difference in the age distribution of women with and without HSIL 172
148 Among the 620 women (age range 16–68 years; mean age +. There were no significant differences between the two groups in 173
149 38.8  9.7 years), 457 (73.7%) did not have HSIL+ [335 and terms of number of HR-HPV types. Colposcopic impression was 174
150 122 women with normal and low-grade squamous intraepithelial significantly associated with the presence of HSIL+ in the biopsy 175
151 lesion (LSIL) findings, respectively]. There were 163 women (26.3%) specimen (p < 0.001). 176
152 with HSIL+ [HSIL, 151; squamous carcinoma of the cervix, 11; Seventy-five of 254 HR-HPV-positive patients (29.5%) had a 177
153 adenocarcinoma in situ (AIS), 1]. There was no difference between final diagnosis of HSIL+. HSIL+ risk was significantly higher in 178
154 the mean age of the HSIL-free and HSIL+ groups (p = 0.426). Among patients with HPV16 infection (52.3%) than in patients with HPV 179
155 the 274 women with positive HPV genotyping results, 75 (29.5%) infection other than HPV16 (17.9%, p < 0.001). In HSIL+ missed by 180
156 had HSIL+. Of the 371 women with positive Hybrid Capture colposcopy, 36.7% of patients (11/30) were HPV16-positive. If 181
157 2 results, 89 had HSIL+ (24.0%). Table 1 summarizes the patient positive colposcopic impression and HPV16 infection status were 182
158 characteristics. considered predictors, the sensitivity of HSIL+ diagnosis was 183

improved from 60.0% (45/75) to 74.7% (56/75). Among Hybrid 184

159 Accuracy of colposcopy Capture 2-positive cases, 24% (89/371) had a final diagnosis of HSIL 185

+. The correlation between increased RLU/CO of HPV DNA testing 186

160 Based on the terminology used, agreement between colpo- and worse diagnosis was not significant (ROC = 0.542, p > 0.05), but 187
161 scopic diagnosis and cervical pathology was matched perfectly in at cut-offs of 50 RLU/CO and 100 RLU/CO, HSIL+ and virus load were 188
162 89.2% of cases (553/620), and the strength of agreement with the k correlated when this group was compared with women without 189
163 statistic was 0.698 (p < 0.001). The PPV of a positive colposcopic HSIL (p = 0.024 and 0.044, respectively). As shown in Fig. 1, the HSIL 190
164 impression was 88.7%, and the NPV of a negative colposcopic + outcome rates showed a progressive increase from Class 1 191
165 impression was 89.5%. However, the sensitivity of colposcopic (14.25%) to Class 5 (41.67%), but the rate decreased when virus load 192
166 impression was low (110/163, 67.5%) compared with specificity >500 RLU/CO. With the exception of Classes 1–2, all HSIL+ outcome 193

rates were >20%. Among HSIL+ cases with Hybrid Capture 2- 194

positive results but negative colposcopic impression, 26.1% (6/23) 195

had >50 RLU/CO. In addition, 70 patients (78.7%) with HSIL+ had 196
Table 1
Patient characteristics (n = 620).

Characteristics n (%) Table 2

Risk factors predicting high-grade squamous intraepithelial lesion or worse in
Age (years) biopsy specimens.
<30 122 (19.7%)
30–40 206 (33.2%) Characteristic Biopsy results Total p-value
40–50 207 (33.4%)
LSIL HSIL
50 85 (13.7%)
n = 457 n = 163
HPV genotyping Age (years) 38.6  9.7 39.3  9.8 38.8  9.7 0.426a
Type 16 86 (33.9%)
Other 168 (66.1%) HPV genotyping
Type 16 41 (47.7%) 45 (52.3%) 86 0.000b
HPV load (RLU) Other types 138 (82.1%) 30 (17.9%) 168
1  HPV < 10 57 (15.4%)
10  HPV 314 (74.6%) HPV load (RLU)
1  HPV < 50 115 (31.0%) 1  HPV < 10 49 (86.0%) 8 (14.0%) 57 0.056b
50  HPV 256 (69.0%) 10  HPV 233 (74.2%) 81 (25.8%) 314
1  HPV < 100 142 (38.3%) 1  HPV < 50 96 (83.5%) 19 (16.5%) 115 0.024b
100  HPV 229 (61.7%) 50  HPV 186 (72.7%) 70 (27.3%) 256
1  HPV < 100 116 (81.7%) 26 (18.3%) 142 0.044b
Colposcopic impression 100  HPV 166 (72.5%) 63 (27.5%) 229
Negative 496 (80.0%) 1  HPV < 200 138 (79.8%) 35 (20.2%) 173 0.113b
Normal colposcopic findings 148 (23.9%) 200  HPV 144 (72.7%) 54 (27.3%) 198
Minor lesions 348 (56.1%)
Positive 124 (20.0%) High-risk HPV types (n)
Minor lesions 113 (18.2%) 1 106 (67.9%) 50 (32.1%) 156 0.266b
Suspicious for invasion 11 (1.8%) >1 73 (74.5%) 25 (25.5%) 98

Biopsy results Colposcopic impression

No lesion 335 (54.0%) Negative 443 (91.2%) 43 (8.8%) 486 0.000b
LSIL 122 (19.7%) Positive 14 (10.4%) 120 (89.6%) 134
HSIL 151 (24.3%)
HPV, human papillomavirus; RLU, relative light unit; LSIL, low-grade squamous
AIS 1 (0.2%)
intraepithelial lesion; HSIL, high-grade squamous intraepithelial lesion.
ICC 11 (1.8%)
Age is presented as mean  standard deviation.
HPV, human papillomavirus; RLU, relative light unit; LSIL, low- Preliminary results of normal colposcopic findings and minor lesions were grouped
grade squamous intraepithelial lesion; HSIL, high-grade squamous as negative. Results of suspicious for invasion and major lesions were classified as
intraepithelial lesion; AIS, adenocarcinoma in situ; ICC, invasive positive.
a
cervical cancer. Student’s t-test.
b
Values are presented as n (%). Chi-squared test.

Please cite this article in press as: Z. Ding, et al., Punch biopsy guided by both colposcopy and HR-HPV status is more efficient for identification
of immediate high-grade squamous intraepithelial lesion or worse among HPV-infected women with atypical squamous cells of undetermined
significance, Eur J Obstet Gynecol (2016), http://dx.doi.org/10.1016/j.ejogrb.2016.10.005
 G Model
EURO 9622 1–5

4 Z. Ding et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology xxx (2016) xxx–xxx

Fig. 1. Relationship between human papillomavirus viral load [relative light unit/cut-off (RLU/CO)] and prevalence of high-grade squamous intraepithelial lesion or worse
(HSIL+) (%).

197 RLU/CO  50, while 26 (21.3%) had RLU/CO < 50 (p < 0.05). If high RLU/CO value to 23% in Class 4 (RLU/CO = 100–1000) and to 48% in 238
198 virus load (at cut-off 50 RLU/CO) was considered a diagnostic Class 5 (RLU/CO > 1000) [19]. The present study also found 239
199 standard of HSIL+ when colposcopic impression was negative, increased prevalence of HSIL+ along with increased RLU/CO if 240
200 sensitivity was improved from 74.2% (66/89) to 81.0% (72/89). RLU/CO < 500, but the increase in prevalence did not continue if 241
201 However, if high virus load was defined at the cut-off point of RLU/CO > 500. With the exception of Classes 1–2, all HSIL+ outcome 242
202 100 RLU/CO, there were no additional patients. rates were >20%. 243

As the second stage of cervical cancer screening, colposcopy 244

203 Discussion aims to identify HSIL+. The sensitivity of colposcopy for detecting 245

HSIL+ ranges from 48% to 87% [6,21–28], which is similar to the 246
204 Large international studies have firmly established that present findings. However, in line with the research of Wentzensen 247
205 HPV16 is the most carcinogenic HPV genotype, and that it accounts et al., this study demonstrated increased sensitivity achieved by 248
206 for >50% of all HSIL or cervical cancers [10–14]. However, it has referral HPV status [29]. If HPV16 infection in HSIL+ diagnosis was 249
207 not been confirmed how the prevalence of HSIL+ differs by considered when colposcopic impression was used, sensitivity was 250
208 HPV16 status compared with other oncogenic HPV types among improved from 60.0% (45/75) to 74.7% (56/75). In patients with 251
209 women with ASCUS. This study evaluated the immediate risks for positive Hybrid Capture 2 results, the sensitivity for diagnosing 252
210 HSIL+ attributable to baseline-detected HPV16 and other onco- HSIL+ was improved from 74.2% (66/89) by colposcopy alone to 253
211 genic HPV infection in women with ASCUS. HSIL+ risk was 81.0% (72/89) by referring to RLU/CO  50 at the same time. 254
212 significantly higher in women with HPV16 infection (52.3%) than in Consequently, it is believed that HPV16 infection and RLU/ 255
213 women with other types of HPV infection (17.9%). The present CO  50 could be risk factors for predicting HSIL+, and could be 256
214 results confirm that, among women with ASCUS, HR-HPV DNA used as a cooperative diagnostic method with colposcopy. In 257
215 genotype analysis using the HybriBio GenoArray clearly defines the conclusion, colposcopy remains the reference standard for 258
216 group of women with a worse diagnosis. diagnosing cervical cancer/precancer. This analysis supports its 259
217 This study also evaluated the value of HR-HPV viral load as a use in HSIL+ identification, specifically in HPV-infected women 260
218 predictor of HSIL+ in women with ASCUS. The main weakness of with ASCUS. Colposcopy detected most HSIL+ in these patients. At 261
219 this research is that the data were retrospective. As the original the same time, the present results indicate that HPV DNA 262
220 function of HPV testing was for screening, the HPV viral load was genotyping using GenoArray or viral load detected by Hybrid 263
221 estimated using Hybrid Capture 2 in these patients. The RLU value Capture 2 can be used to stratify the risk of HSIL+ in these patients. 264
222 is a measure of the cumulative viral load of the 13 HR-HPV types If patients with negative colposcopy have risk factors such as 265
223 included in the Hybrid Capture 2 test, and is not adjusted for HPV16 infection or viral load RLU/CO  50, biopsies could increase 266
224 specimen cellularity. Nevertheless, previous studies have revealed the immediate HSIL+ detection rate. This may have a clinical 267
225 that HSIL+ frequency increases as a function of the viral load impact on patient management. 268
226 detected by HPV DNA testing (Hybrid Capture 2) [15–18]. In
227 References 269
addition, a number of studies have focused on HPV viral load in
228 cases of ASCUS cytology. Origoni et al. reported that out of 50 cases
229 [1] Gustafsson L, Ponten J, Bergstrom R, Adami HO. International incidence rates of
of cervical intraepithelial neoplasia 2/3 detected in samples with 270
230 invasive cervical cancer before cytological screening. Int J Cancer 1997;
positive HPV DNA testing (RLU/CO > 1), 40 (80%) had RLU/CO > 100 71:159–65. 271
231 [19]. Similar results were found in the present study; out of 89 such [2] Gustafsson L, Ponten J, Zack M, Adami HO. International incidence rates of
232 invasive cervical cancer after introduction of cytological screening. Cancer 272
patients, 63 (70.8%) had RLU/CO  100, while 26 (29.2%) had RLU/ 273
233 Causes Control 1997;8:755–63.
CO < 100 (p < 0.05). Jarboe et al. also reported a low prevalence [3] Arbyn M, Roelens J, Simoens C, et al. Human papillomavirus testing versus
234 (3.2%) of HSIL in cases with weakly positive HPV DNA testing repeat cytology for triage of minor cytological cervical lesions. Cochrane 274
235 Database Syst Rev 2013;3:CD008054. 275
(Hybrid Capture 2) (RLU/CO = 1–10) [20]. In the study by Origoni
236 [4] Vanni T, Legood R, Franco EL, Villa LL, Luz PM, Schwartsmann G. Economic
et al., the Hybrid Capture 2 results were classed into five increasing evaluation of strategies for managing women with equivocal cytological 276
237 classes, where HSIL prevalence increased continuously with the 277
results in Brazil. Int J Cancer 2011;129:671–9.

Please cite this article in press as: Z. Ding, et al., Punch biopsy guided by both colposcopy and HR-HPV status is more efficient for identification
of immediate high-grade squamous intraepithelial lesion or worse among HPV-infected women with atypical squamous cells of undetermined
significance, Eur J Obstet Gynecol (2016), http://dx.doi.org/10.1016/j.ejogrb.2016.10.005
 G Model
EURO 9622 1–5

Z. Ding et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology xxx (2016) xxx–xxx 5

[5] Lytwyn A, Sellors JW, Mahony JB, et al. Comparison of human papillomavirus
278 DNA testing and repeat Papanicolaou test in women with low-grade cervical
279 cytologic abnormalities: a randomized trial. HPV Effectiveness in Lowgrade
280 Paps (HELP) Study No. 1 Group. CMAJ 2000;163:701–7.
[6] Moss EL, Hadden P, Douce G, Jones PW, Arbyn M, Redman CW. Is the
281 colposcopically directed punch biopsy a reliable diagnostic test in women with
282 minor cytological lesions? J Low Genit Tract Dis 2012;16:421–6.
Q6 [7] The Bethesda System for reporting cervical cytology. Definitions, criteria, and
283 explanatory notes. 3rd ed. Anticancer Res 2015;35:5708.
[8] Solomon D, Davey D, Kurman R, et al. The 2001 Bethesda System: terminology
284 for reporting results of cervical cytology. JAMA 2002;287:2114–9.
[9] Herbert A, Bergeron C, Wiener H, et al. European guidelines for quality
285 assurance in cervical cancer screening: recommendations for cervical cytology
286 terminology. Cytopathology 2007;18:213–9.
[10] Walboomers JM, Jacobs MV, Manos MM, et al. Human papillomavirus is a
287 necessary cause of invasive cervical cancer worldwide. J Pathol 1999;189:
12–9.
[11] de Sanjose S, Quint WG, Alemany L, et al. Human papillomavirus genotype
288 attribution in invasive cervical cancer: a retrospective cross-sectional
289 worldwide study. Lancet Oncol 2010;11:1048–56.
[12] Munoz N, Bosch FX, de Sanjose S, et al. Epidemiologic classification of human
290 papillomavirus types associated with cervical cancer. N Engl J Med
2003;348:518–27.
Q7 [13] Nakamura Y, Matsumoto K, Satoh T, et al. HPV genotyping for triage of women
291 with abnormal cervical cancer screening results: a multicenter prospective
292 study. Int J Clin Oncol 2015.
[14] Lagos M, Van De Wyngard V, Poggi H, et al. HPV16/18 genotyping for the triage
293 of HPV positive women in primary cervical cancer screening in Chile. Infect
294 Agents Cancer 2015;10:43.
[15] Xu Y, Dotto J, Hui Y, Lawton K, Schofield K, Hui P. High grade cervical
295 intraepithelial neoplasia and viral load of high-risk human papillomavirus:
296 significant correlations in patients of 22 years old or younger. Int J Clin Exp
297 Pathol 2009;2:169–75.
[16] Lorincz AT, Castle PE, Sherman ME, et al. Viral load of human papillomavirus
298 and risk of CIN3 or cervical cancer. Lancet 2002;360:228–9.
[17] Tsai HT, Wu CH, Lai HL, et al. Association between quantitative high-risk
299 human papillomavirus DNA load and cervical intraepithelial neoplasm risk.
300 Cancer Epidemiol Biomark Prev 2005;14:2544–9.
[18] Sherman ME, Wang SS, Wheeler CM, et al. Determinants of human
papillomavirus load among women with histological cervical intraepithelial 301
neoplasia 3: dominant impact of surrounding low-grade lesions. Cancer 302
Epidemiol Biomark Prev 2003;12:1038–44. 303
[19] Origoni M, Carminati G, Rolla S, et al. Human papillomavirus viral load
expressed as relative light units (RLU) correlates with the presence and grade 304
of preneoplastic lesions of the uterine cervix in atypical squamous cells of 305
undetermined significance (ASCUS) cytology. Eur J Clin Microbiol Infect Dis 306
2012;31:2401–6.
[20] Jarboe EA, Venkat P, Hirsch MS, Cibas ES, Crum CP, Garner EI. A weakly positive
human papillomavirus Hybrid Capture II result correlates with a significantly 307
lower risk of cervical intraepithelial neoplasia 2,3 after atypical squamous cells 308
of undetermined significance cytology. J Low Genit Tract Dis 2010;14:174–8. 309
[21] Di Stefano L, Patacchiola F, Necozione S, Paolone G, Di Febbo G, Carta G. The
correspondence between abnormal transformation zone grade 1 and grade 310
2 colposcopic parameters and histology. Eur J Clin Microbiol 2014;35:16–9. 311
[22] Diedrich JT, Felix JC, Lonky NM. Contribution of exocervical biopsy,
endocervical curettage, and colposcopic grading in diagnosing high-grade 312
cervical intraepithelial neoplasia. J Low Genit Tract Dis 2014. 313
[23] Kaban I, Cengiz H, Kaban A, Yildiz S, Ekin M, Avci E. Agreement between
colposcopy results using the Reid Colposcopic Index and histopathology. 314
Ginekologia polska 2015;86:537–40. 315
[24] Thulaseedharan JV, Malila N, Esmy PO, Muwonge R, Hakama M, Sankaranar-
ayanan R. Risk of invasive cancer among women visually screened and 316
colposcopy triaged by trained nurses in rural South India. Int J Gynaecol Obstet 317
2015;129:104–8.
[25] Ghosh I, Mittal S, Banerjee D, et al. Study of accuracy of colposcopy in VIA and
HPV detection-based cervical cancer screening program. Aust N Z J Obstet 318
Gynaecol 2014;54:570–5. 319
[26] Massad LS, Collins YC. Strength of correlations between colposcopic
impression and biopsy histology. Gynecol Oncol 2003;89:424–8. 320
[27] Li Y, Zhang H, Zheng R, Xie F, Sui L. Agreement between colposcopic diagnosis
with 2011 international terminology of colposcopy and cervical pathology in 321
cervical lesions. Zhonghua fu chan ke za zhi 2015;50:361–6. 322
[28] Zhao J, Zhang X, Chen R, et al. Performance of the R-way colposcopic evaluation
system in cervical cancer screening. Asian Pac J Cancer Prev 2015;16:4223–8. 323
[29] Wentzensen N, Walker JL, Gold MA, et al. Multiple biopsies and detection of
cervical cancer precursors at colposcopy. J Clin Oncol 2015;33:83–9. 324

Please cite this article in press as: Z. Ding, et al., Punch biopsy guided by both colposcopy and HR-HPV status is more efficient for identification
of immediate high-grade squamous intraepithelial lesion or worse among HPV-infected women with atypical squamous cells of undetermined
significance, Eur J Obstet Gynecol (2016), http://dx.doi.org/10.1016/j.ejogrb.2016.10.005