IJC International Journal of Cancer

Transcription

1 IJC International Journal of Cancer Incidence and outcome of acquisition of human papillomavirus infection in women with normal cytology a population-based cohort study from Taiwan Angel Chao 1, Chee-Jen Chang 2, Chyong-Huey Lai 1, Fang-Yu Chao 1, Yi-Hsien Hsu 2, Hung-Hsueh Chou 1, Huei-Jean Huang 1, Shih-Ming Jung 3, Cheng-Tao Lin 1, Hui-Hsin Cheng 1, Chu-Chun Huang 1, Jung-Erh Yang 1, Ting-Chang Chang 1 1 Department of Obstetrics and Gynecology, Chang Gung Memorial Hospital, Chang Gung University College of Medicine, Taoyuan, Taiwan 2 Graduate Institutes of Clinical Medical Sciences, Chang Gung University, Taoyuan, Taiwan 3 Department of Pathology, Chang Gung Memorial Hospital, Chang Gung University College of Medicine, Taoyuan, Taiwan Little is known about acquisition of human papillomavirus () and its outcome among older women with negative testing and normal cytology. A longitudinal 3-yr follow-up of nested-cohort subjects (n ) from a population-based cervical cancer screening study whose Pap and tests were negative at baseline were conducted. Every active negative (n 5 413) participant had 12-mo follow-ups of Pap smear and testing. Colposcopy was performed if either positive or cytology was abnormal. The cytology and histology information of the remaining subjects (passive -negative, n ) was obtained from national registry database. Median age of participants was 45 yr (range, yr). The incidence of new acquisition was 4.2/100 woman-years. The 3-yr cumulative total acquisition rate was 11.1% (95% confidence interval [CI]: ). Increased number of sexual partners (2 vs. 1) of the participant was associated with risk of acquisition (odds ratio [OR]: 5.0, 95% CI: ) by multivariate analysis. Three cases of cervical intraepithelial neoplasia (CIN) 2 were identified in 3-yr follow-up in active -negative subjects. genotypes in the dysplastic tissue were actually present at baseline samples after reanalysis. From the passive -negative group, only 1 case progressed to CIN2 probably after acquisition. Negative Pap and tests assured a very low risk of developing CIN2 within 3 yr despite incident infection. Human papillomavirus () is a causative factor of cervical intraepithelial neoplasia (CIN) and subsequent invasive carcinomas. 1,2 infection is common in female population, with many women at some age having been infected with. 2 It is now widely accepted that persistent infection is a prerequisite for progression into CIN and invasive carcinoma. 3,4 The time from acquisition to development of cervical neoplasia is important for selecting an appropriate screening interval. However, the time lag from infection to disease is not easily determined, because a woman whose test Key words: cervical cancer screening, Pap smear, human papillomavirus Grant sponsor: Chang Gung Memorial Hospital; Grant numbers: CMPRG32016-I-III, CMPRG ; Grant sponsor: National Science Council-Taiwan; Grant numbers: NSC B , NSC B MY2 The first three authors contributed equally to this work DOI: /ijc History: Received 10 Apr 2009; Accepted 4 Jun 2009; Online 7 Jul 2009 Correspondence to: Chyong-Huey Lai, Department of Obstetrics and Gynecology, Chang Gung Memorial Hospital, Linkou Medical Center, 5 Fu-Shin Street, Kueishan, Taoyuan 333, Taiwan, People s Republic of China, Fax: þ , sh46erry@ms6.hinet.net is positive in her first sample could have been infected long before that sample is taken. Natural history of infection can be best studied from longitudinal cohort studies starting before sexual debut. 5 Several previous studies, mostly targeted on young women, report incident infection in women with negative tests and normal Pap smears. 6 9 In a Korean study of 171 university students (mean age, 18.4 yr), the acquisition rates of those starting sexual intercourse or had a new sex partner during study were 30 and 29.4% vs. 12.7% of those remained virginal at a median follow-up of 18 mo. 6 Winer et al. 7 and Moscicki et al. 8 found that incidence and prevalence of infection among young women (mean age, 20.0 and 19.2 yr, respectively) were usually high. However, these young women usually clear incident in a short time without developing high-grade CIN. A Columbian cohort study found that incidence of infection was 6.2 cases/100 woman-years in older women (aged yr), with negative and normal cytology. The study population was still relatively young with an interquartile range of yr old (median 32.3 yr). 9 Hence, we need to know more about acquisition of and its outcome among older women. The aims of this prospective cohort study are (i) to follow up women with normal cytology and negative infection at baseline in a population-based study 10 ;(ii) to investigate the host and viral factors in relation to incidence and Int. J. Cancer: 126, (2010) VC 2009 UICC

2 192 acquisition in normal cytology women Figure 1. Flow diagram of the study. indicates human papillomavirus; F/U, follow-up; CIN, cervical intraepithelial neoplasia; Em Ca, endometrial cancer. *Active -positive subjects in previous study. 11 outcome of viral acquisition.this study will essentially provide useful information for women receiving combined testing and Pap smear as primary screening, facilitating appropriate strategies of baseline or acquired infection. Material and Methods Patient recruitment Women aged 30 yr or older who were residents of Taoyuan, Taiwan were invited from a population-based study. 10 The number of subjects of each township was proportional to its population. Participants should have an intact uterus and no history of treatment for preinvasive or invasive cervical neoplasms. After signing informed consent that was approved by Institutional Review Board of Chang Gung Memorial Hospital (CGMH), every participant completed a structured questionnaire and had 2 cervical swabs: the first for a Pap smear and a second swab for testing. Cervical samples were collected using cytobrush and transported in a storage medium (manufactured by King Car, I-Lan, Taiwan). Only those with both adequate DNA and Pap quality were eligible for analysis. Cytological reports were signed by local certified cytology lab as per routine practice or at CGMH. The 2001 Bethesda System was used for terminology for reporting results of cervical cytology. Women with abnormal Pap results (atypical squamous cells of undetermined significance [ASCUS] or worse) were designated Group I and were referred to immediate colposcopy. 10 Women with normal cytology but positive DNA (Group II) were recalled for a colposcopic examination at CGMH, and a directed biopsy was performed in case of abnormal colposcopic finding. Those who had a normal colposcopic finding or a negative colposcopic directed biopsy and participated at least 1 followup visit (n ¼ 412) 11 were designated Group II-F (active -positive; Fig. 1). Group III represented women who had both tests negative (n ¼ 8,846) being randomly invited (n ¼ 1,071) to participate in the current follow-up study. Another informed consent to participate this follow-up study (Group IIIa-R, active -negative, n ¼ 413) was obtained, and detailed questionnaire including sex history was implemented. Active -negative participants were recalled for 12-mo follow-ups of Pap smear and testing, and colposcopy was performed when cytology turned abnormal or test turned positive. Subsequent Pap and histology data were retrieved from national registry for those who did not enter the active follow-up study (Group IIIb, passive negative, n ¼ 8,412). genotying by Easychip VR Blot and E6 type-specific PCR DNA was extracted according to the protocol for isolation of total DNA from cultured animal cells (Qiagen, Valencia, CA). 12,13 Finally, 100 ll of DNA solution was eluted and 1 ll of the aliquot was used for polymerase chain reaction (PCR) amplification. The SPF1/GP6þ consensus primers were used to amplify a fragment of approximately 184 bp in the L1 open reading

3 Chao et al. 193 Table 1. Examples according to definition of acquisition in the follow-up of women with baseline negative infection Follow-up visit 1 Follow-up visit 2 Follow-up visit 3 Follow-up visit 4 Case B TS B TS B TS B TS Acquisition (A)/indeterminate (I) , 61, 18, , 18, 72 18, 51 61, 72 NA NA A of 51, 61, 18, 72 at visit A of 52 at visit 3 3, I of 18 4, I of 59 5,011 18, 58 18, 58 18, , 58 A of 18, 58 at visit 2 5, I of 16 7,860 39, 58 39, 58 39, 58 NA NA NA NA I of 39, 58 B, blot assay;, human papillomavirus; NA, not applicable because of no follow-up; TS, type-specific PCR. frame as previously described. 10,11,13 The resulting 15 ll amplimers were then hybridized with an Easychip VR Blot (King Car, I-Lan, Taiwan; hereafter Blot) membrane. Thirty-eight types of (6, 11, 16, 18, 26, 31, 32, 33, 35, 37, 39, 42, 43, 44, 45, 51, 52, 53, 54, 55, 56, 58, 59, 61, 62, 66, 67, 68, 69, 70, 71 [CP8061], 72, 74, 81 [CP8304], 82 [MM4], 83 [MM7], 84 [MM8] and L1AE5) oligonucleotide probes are immobilized on a nylon membrane for detection. E6 type-specific PCRs were performed to validate multiple types. types were divided into high-risk and low risk groups: the leading 14 types (-16, -18, -58, -33, -52, -39, -45,-31, -51, -70, -56, -59, -35 and -53) of cervical cancer in Taiwanese population were designated high-risk group, and the remaining 24 types of low prevalence were called low-risk group. 13 Type-specific PCR E6 type-specific PCR was performed for 50 cycles of 1 min at 94 C, 1 min at 59 C and 1 min at 72 C. A final extension of 5 min at 72 C was added after 50 cycles. Total yield of each PCR product was analyzed by electrophoresis on 2% agarose gels stained with ethidium bromide. Quality of isolated DNA was checked with GAPDH PCR by which a 190 bp product was amplified. Definitions of acquisition and indeterminate results Acquisition was defined as (1) the contraction of a new type (not present in the previous test) in any one of the follow-up visits detected by both the Blot and E6 type-specific PCR in a single visit (-18 in Case 5,011, -52 of Case 984); (2) a new type detected by 1 of the 2 methods and remained positive at subsequent (-51, -61 and -72 of Case 388, -52 of Case 984) or nonconsecutive visit (-58 in Case 5,011) using Blot or E6 type-specific PCR (-18 of Case 388). In case of acquisition of a new type with discrepant results in the tests of a single visit, samples of the previous and subsequent visits would be submitted to E6 type-specific PCR for cross-check (Cases 388 and 984). No cross validation by the other method or by followup (Cases 3,759, 4,591, 5,978 and 7,860) was designated as indeterminate (Table 1). According to protocol, women were advised not to receive treatment because of a CIN1 diagnosis for 2-yr observation indicated persistent and no evidence of regression of the CIN1 lesion. However, treatment did not cause a protocol violation, while these treated subjects were censored at the point of treatment. Statistical analyses Host characteristics (age at enrollment, ever pregnant, parity, Pap smear history, family history of cervical cancer, number of life-time sex partner, number of sex partner of male consort, age of first intercourse, smoking, use of oral contraceptives and use of hormone replacement therapy) and viral characteristics ( genotype, high-risk or low-risk type and multiple or single type) of incident infection were included in analyses. If 1 had bothhigh-risk and low-risk type, the individual was attributed to the high-risk group. Cumulative acquisition rates were calculated according to Kaplan Meier method. The association between factors and likelihood of total acquisition was expressed as hazard ratio (HR). The 3-yr cumulative acquisition rates were compared between women with -positive (active -positive, Group II-F) 11 and -negative (active -negative, Group IIIa-R) at baseline. Adjusted HRs of type-specific acquisition of not present at baseline and time to progression to CIN2, according to both baseline and acquisition of of the women with normal cytology were calculated from national registry database (the permit for inquiry of such information was addressed in the informed consent). The data were analyzed by the SPSS 14.0 statistical package (SPSS, Chicago, IL) and STATA 9.0 software (College Station, TX). Continuous covariates were compared between groups using both parametric and nonparametric approaches, mainly, the Student t and Mann-Whitney U tests when appropriate. All tests were 2-sided, and a p value of <0.05 was considered statistically significant.

4 194 acquisition in normal cytology women Table 2. Human papillomavirus total acquisition according to host characteristics of active -negative group (n ¼ 413) Cumulative 3-yr total No. women acquisition Characteristics recruited (%) N (%) 95% CI HR 1 HR 2 Age (yr) (24.2) 10 (11.1) (50.4) 23 (13.1) ( ) 1.08 ( ) (21.5) 11 (34.5) ( ) 1.22 ( ) >60 16 (3.9) 0 (0) 0 NA Continuous 413 (100) 44 (NA) NA 0.99 ( ) Menopausal status No 298 (72.2) 32 (12.4) Yes 115 (27.8) 12 (11.6) ( ) Hormone replacement therapy Never 73 (17.7) 7 (10.4) Current user 10 (2.4) 1 (11.1) ( ) Past user 32 (7.7) 4 (14.0) ( ) Oral contraceptives use Never/past user 337 (81.6) 36 (12.1) Current user 76 (18.4) 8 (13.2) ( ) No. sexual partners (90.6) 33 (10.1) (6.3) 9 (40.5) ( ) 4.98 ( ) Refusal to answer 13 (3.1) 2 (15.4) ( ) 3.38 ( ) Sex partner of male consort (74.1) 30 (11.5) (3.6) 4 (24.0) ( ) 1.16 ( ) Refusal to answer 92 (22.3) 10 (12.8) ( ) 0.79 ( ) Age of first sex intercourse 18 yr 25 (6.1) 4 (17.1) ( ) >18 yr 388 (93.9) 40 (11.9) Smoking Never 395 (95.6) 42 (12.2) Smoker 17 (4.1) 2 (14.4) ( ) Pap smear history Never had one 9 (2.2) 1 (12.5) ( ) Last Pap <3 yr 14 (3.4) 3 (21.4) ( ) Last Pap 3 yr 390 (94.4) 40 (11.8) Parity 0 18 (4.4) 4 (22.5) ( ) >2 178 (43.1) 13 (10.7) ( ) (52.3) 27 (17.4) , human papillomavirus; CI, confidence interval; HR, hazard ratio; NA, not applicable. 1 Age-adjusted HR. 2 Multivariate analyses of age, number of sex partner and sex partner of male consort. Results Excluding those (n ¼ 19) who had a diagnosis of CIN or abnormal cytology within 9 mo of baseline, 10 1,017 women were randomly invited for active follow-up. Four hundred and fifteen women signed informed consent and the remaining 602 did not respond. Two women were not included

5 Chao et al. 195 Table 3. acquisition according to viral factors in active -negative group (n ¼ 413) Incidence of acquisition Viral factors N 100 women-year (95%CI) Multiple type Single type High risk Low risk Single and multiple type were counted as women based (n ¼ 44). 2 p ¼ Comparison by high-risk and low-risk infections (n ¼ 63; each type as an independent infection). 4 p ¼ because of coexisting breast cancer. Hence, 413 subjects (active -negative, Group IIIa-R) with at least 1 follow-up visit were eligible, of which 369 subjects had 2 follow-up visits and 327 had 3 follow-up visits. A total of 8,412 (passive -negative, Group IIIb) were eligible for passive followup, whose cytology and histology information was obtained from national registry database (Fig. 1). A nested cohort of 8,825 women whose Pap and tests were negative at baseline (Group IIIa-R and Group IIIb) was recruited for this longitudinal 3-yr follow-up study. The median follow-up of enrolled subjects was 34.7 mo (range, ). The median age of participating women was 45 (range, 30 3 yr) for the 413 women. Figure 2. The 3-yr cumulative total acquisition rates of active -negative (Group IIIa-R, solid line) and active -positive (Group II-F, dotted line) groups. Overall, 115 (27.8%) were postmenopausal, among whom 10 (2.4%) were current user of hormone therapy. Seventy-six (18.4%) women were current users and 337 (81.6%) were never/past users of oral contraceptives. Only 4.1% of participants ever smoked and 26 (6.3%) reported a number of lifetime sex partners 2. In Pap history, 390 (94.4%) women had their last Pap 3 years, 14 (3.4%) had Pap <3 yr, and 9 (2.2%) never had Pap before (Table 2). More than half of the women (52.3%) had 2 children. Increased number of sexual partners (2 vs. 1) of the participant in the active -negative subjects was associated with risk of acquisition (odds ratio [OR]: 5.0, 95% CI: ) by multivariate analysis, while sex partner of the male consort was significant in univariate analysis but not significant in multivariate analysis (Table 2). Forty-four women developed infection, where multiple types were found in 14 (31.8%) in active -negative women. A total of 63 incident genotype-specific infections were observed during the 3-yr follow-up period. The incidence of new acquisition was 4.2/100 woman-years. The incidence of new acquisition of high-risk was 3.6/100 woman-years, while the low-risk group was 2.2/100 womanyears. The 3-yr cumulative total acquisition rate was 11.1% (95% confidence interval [CI]: ). The leading types were -52, -18 and -58 (Table 3). When compared with active -positive women (36.4% using 12-mo followup interval data), acquisition rate was significantly lower (p < ) in active -negative women (11.1%) with normal cytology at baseline (Fig. 2). In the 413 active -negative cohort, a total of 1,098 cytology samples were collected in 3-yr follow-up, among which 29 abnormal cytology in 27 women were observed (17 ASCUS, 2 low-grade squamous intraepithelial lesion [LSIL], 3 glandular abnormalities, 5 atypical squamous cells, cannot exclude high-grade squamous intraepithelial lesion [ASC-H] and 2 high-grade squamous intraepithelial lesion [HSIL]). Meanwhile, national database recorded a total of 7,878 Pap smears, among which 134 abnormal cytology (98 ASCUS, 17

6 196 acquisition in normal cytology women Table 4. Longitudinal follow-up of those developed CIN2 lesions in the active -negative cohort (group IIIa-R) 1 or CIN2 in passive -negative cohort (group IIIb) n ¼ 6) Baseline Follow-up 1 Follow-up 2 Follow-up 3 Final Pathology Time to CIN (mo) Blot TS-PCR Blot TS-PCR Cytology Histology Blot TS-PCR Cytology Blot TS-PCR Cytology Blot TS-PCR Cytology Subject code Group 4,905 IIIa-R Normal AGUS NA NA NA NA NA NA AIS ,381 IIIa-R ,70 3 Normal 52 52,70 3 HSIL NA NA NA NA NA NA CIN ,246 IIIa-R Normal HSIL NA NA NA NA NA NA CIN ,539 5 IIIb 2 4 Normal NA NA NA NA NA NA NA NA NA EM Ca 6 NA NA NA 9,258 5 IIIb 2 4 Normal NA NA NA NA NA NA NA NA NA Miscoded CIN2 7 NA NA NA IIIb 2 8 Normal NA NA NA NA NA NA NA NA NA CIN2 9 NA 8 NA AGUS, atypical glandular cells; CIN, cervical intraepithelial neoplasia; EM CA, endometrial cancer;, human papillomavirus; HSIL, high-grade squamous intraepithelial lesion; NA, not applicable; TS-PCR, type-specific polymerase chain reaction. 1 Biopsy with CIN1 in which paraffin block not obtainable is not listed here. 2 By 100 ng input DNA. 3 Negative by E6 TS-PCR, but positive by Q-PCR. 4 By TS-PCR of the 5 most common types of invasive cervical cancer in Taiwan Group IIIb of this study; histology retrieved from national registry data under initial informed consent permission. 6 The coding for pathology was adenocarcinoma, and the original pathology report provided by the subject was endometrial carcinoma; miscoded as cervical adenocarcinoma in our previous study The coding for pathology was CIN2 by national registry but both original pathology report and histology review by our pathologist revealed no pathologic diagnosis. 8 By TS-PCR of the 14 most common types of invasive cervical cancer in Taiwan The coding for pathology was CIN2 by national registry but histology review was refused by the subject. LSIL, 5 glandular abnormalities, 9 ASC-H and 5 HSIL) in 114 women in passive -negative (n ¼ 8,412) were observed. The median number of Pap smears (1 vs. 3 smear(s)/woman, p < 0.001) and rate of abnormal cytology results (1.36 vs. 6.54, p < ) in 3-yr follow-up was significantly decreased among passive -negative than the active -negative cohort (Fig. 1). Every woman in the active -negative cohort had a colposcopy and 26 biopsies were performed upon acquisition of or cytology ASCUS (n ¼ 70), while only 30 of 134 abnormal Pap smears were examined with a biopsy in passive -negative women. There were both 3 cases of histologically CIN2 in the active -negative and passive -negative women, respectively (Table 4). However, high-risk genotypes in the histology specimens, namely -18, -70 and -52 each, were actually present at baseline in those women who developed CIN2 in the nested cohort (further verification by type-specific PCR and repeat SPF1/GP6 and Blot using 100 ng input DNA). From the national registry database, 3 women from passive -negative (Group IIIb) whose histology results were identified and coded as carcinoma in 1 and CIN2 in 2, respectively. After consent was obtained from 2 individuals to provide pathology reports, 1 of which turned out to be endometrial cancer (Subject 2,539), and the other with a histology code of CIN2 (Subject 9,258) was free of dysplasia in the original pathology report and after histology review. However, the remaining 1 (the incident CIN2þ case in Subject 196) refused to provide the pathology report. Further verification of the baseline samples by type-specific PCR and repeat SPF1/GP6 PCR plus Blot failed to detect any sequences (Table 4). Taken together, only a single case progressedto CIN2 at 10 mo, probably because of incident infection, although the histology has not been reviewed for confirmation. When compared with active -positive women (Group II-F, 15 CIN2 cases in 3 yr), 11 the rate of progression to CIN2 was significantly lower (p ¼ 0.004) in active -negative women (Group IIIa-R, n ¼ 3) regardless of undetected infections at baseline or incident infections. Discussion In this nested cohort study, the 3-yr cumulative total acquisition rate was 11.1% among 413 women who had both negative cytology and at baseline. Data on acquisition in older women worldwide are limited. 9,14 Ho et al. 15 noted a 3-yr cumulative acquisition rate of 43% in 608 college women (mean age ) at 6-mo follow-up interval. Goodman et al. 14 observed that acquisition rate was 13.0/1000 woman-mo (follow-up at 4-mo interval) in the clinic-based Hawaiian cohort with a median age of 33 yr. Franco et al. 16 reported 13.4/1000 woman-months (median age: 33, first and third quartiles, 26 and 39 yr) in Brazil (every 4 mo in the first year and twice yearly thereafter), while Richardson et al. 17 found 19.0/1000 woman-months (median

7 Chao et al. 197 age: 21, range, yr) in Canada with follow-up intervals of 5 7 mo. In our cohort (median age: 45; range, yr), the incidence of acquisition was 4.2/100 woman-years. To our knowledge, our study population has been the oldest in the literature. Similar to other investigators, 7,8,14,15,18,19 increased number of sexual partners was a predictor of acquisition in our study. The risk of incident infection increased significantly in the participants (OR: 5.0, CI: ) when comparing 2 vs. 1 male partner. Increase number of sex partners of the male consort increased the risk of infection in univariate analysis is consistent with other reports. Winer et al. 7 concluded that the actual risk may be greater than the observed data, for women could underestimate their partners previous sexual experience. Other risk factors such as postmenopausal status, current users of oral contraceptives and never user of hormone replacement therapy were positively correlated with prevalence in our cross-sectional population-based study. 10 Nevertheless, we found that these factors were not related to the risk of contracting a new type of. No association of oral pills on infection, either past or current users, was consistent with other investigators. 14 Other studies found that women with -positive at study entry were more likely to contract other infections during follow-up than were women who were free of infection. 20,21 Furthermore, Rousseau et al. 20 reported that acquisition occurred with a significant frequency in coinfections at study entry, where -16 and -52 were the most common types. In our study, 3-yr cumulative rate was 48.2% for Group II-F (active -positive) with a follow-up interval of 6 mo as the original design. The 3-yr cumulative rate of active -positive subjects became 36.4%, using the same 12-mo interval data. Nevertheless, the difference between active -negative and active -positive women was significant based on the same follow-up interval (Fig. 2). Although three cases with confirmed CIN2 in Group IIIa-R, -18, 70 and 52, each in the tissues, were actually present at baseline by further verifying methods. No unequivocal acquisition caused a lesion CIN2 within 3 yr, despite 44 of them acquired infections. On the national database, the only incident CIN2 case (though not available for histology review) in passive -negative subjects was negative at baseline when verified by type-specific PCR of the 14 leading types. 10 Of 1,203 women (aged 16 23) accrued in the placebo arm of a multicenter vaccine trial with 6-mo follow-up for 4 References yr, the time from acquisition to development of CIN or clearance in -16 (n ¼ 142) and -18 (n ¼ 62) was at 13.9 and 14.9 mo, respectively. 22 In a study from United Kingdom (median age, 18 yr; range 15 19), 11.4% (28/246) high-grade CIN developed in 6 12 mo after first detection of 16 during 3-yr follow-up. 23 Another study in young women (mean age, yr) who became sexually active immediately before or during a longitudinal follow-up reported a 2-yr cumulative incidence of CINs of 15% without mentioning time from acquisition to CIN. 5 There are strengths and limitations in our study. First, both cytologyand histology endpoints were assessable by linking the national registry database, which enabled the evaluation in the 3-yr outcome of the entire cohort (Group III, Fig. 1). However, the ascertainment of disease or no disease is better by reviewing the Pap smears and biopsies per se than reviewing the cytology and histology reports from the National Registry Database. Second, detection of in women who are initially negative may not be from a new acquisition, but may be from detection of an old infection that now has a higher copy number than previous tests. Third, the active -negative cohort was followed up in a prospective study with additional tests, which prompted more intense investigations on abnormal cytology or acquisition of than passive -negative subjects. Underestimation of incident Pap abnormality and CIN in passive -negative subjects seemed unavoidable in a short followup. retention time depends on the interval between the tests. 2 The actual acquisition time and rates were underscored by a 12-mo follow-up interval in our study. Some women might be infected and cleared before the subsequent visits. Extending this longitudinal study by inviting new cases with shortened 4 6 mo follow-up intervals is warranted to answer the question of underscoring the actual acquisition time and rates. In conclusion, negative Pap and tests assured a Taiwanese woman (or women from a similar demographic population) very low risk of developing CIN2 within 3 yr despite incident acquisition based on a 3-yr outcome of 8,825 women, therefore, a screening interval of 3 yr is safe and recommendable. Acknowledgements The authors thank the Bureau of Health Promotion, Department of Health, Taiwan for providing the national registry data in Pap smears and histological results of the participants. 1. zur Hausen H. Papillomaviruses causing cancer: evasion from host-cell control in early events in carcinogenesis. J Natl Cancer Inst 2000;92: Woodman CB, Collins SI, Young LS. The natural history of cervical infection: unresolved issues. Nat Rev Cancer 2007;7: Liaw KL, Glass AG, Manos MM, Greer CE, Scott DR, Sherman M, Burk RD, Kurman RJ, Wacholder S, Rush BB, Cadell DM, Lawler P, et al. Detection of human papillomavirus DNA in cytologically normal women and subsequent cervical squamous intraepithelial lesions. J Natl Cancer Inst 1999;91: Wallin KL, Wiklund F, Angstrom T, Bergman F, Stendahl U, Wadell G,

8 198 acquisition in normal cytology women Hallmans G, Dillner J. Type-specific persistence of human papillomavirus DNA before the development of invasive cervical cancer. N Engl J Med 1999;341: Winer RL, Hughes JP, Feng Q, O Reilly S, Kiviat NB, Holmes KK, Koutsky LA. Condom use and the risk of genital human papillomavirus infection in young women. N Engl J Med 2006;354: Oh JK, Ju YH, Franceschi S, Quint W, Shin HR. Acquisition of new infection and clearance of type-specific human papillomavirus infections in female students in Busan, South Korea: a followup study. BMC Infect Dis 2008;8: Winer RL, Lee SK, Hughes JP, Adam DE, Kiviat NB, Koutsky LA. Genital human papillomavirus infection: incidence and risk factors in a cohort of female university students. Am J Epidemiol 2003;157: Moscicki AB, Hills N, Shiboski S, Powell K, Jay N, Hanson E, Miller S, Clayton L, Farhat S, Broering J, Darragh T, Palefsky J. Risks for incident human papillomavirus infection and low-grade squamous intraepithelial lesion development in young females. JAMA 2001;285: Munoz N, Mendez F, Posso H, Molano M, van den Brule AJ, Ronderos M, Meijer C, Munoz A. Incidence, duration, and determinants of cervical human papillomavirus infection in a cohort of Colombian women with normal cytological results. J Infect Dis 2004; 190: Chao A, Hsu KH, Lai CH, Huang HJ, Hsueh S, Lin SR, Jung SM, Chao FY, Huang SL, Huang CC, Yang JE, Chang TC. Cervical cancer screening program integrating Pap smear and DNA testing: a population-based study. Int J Cancer 2008;122: Lai CH, Chao A, Chang CJ, Chao FY, Huang HJ, Hsueh S, Lin CT, Cheng HH, Huang CC, Yang JE, Wu TI, Chou HH, et al. Host and viral factors in relation to clearance of human papillomavirus infection: a cohort study in Taiwan. Int J Cancer 2008;123: Huang SL, Chao A, Hsueh S, Chao FY, Huang CC, Yang JE, Lin CY, Yan CC, Chou HH, Huang KG, Huang HJ, Wu TI, et al. Comparison between the hybrid capture II test and a SPF1/GP6þ PCRbased assay for detection of human papillomavirus DNA in cervical swab samples. J Clin Microbiol 2006;44: Lai CH, Huang HJ, Hsueh S, Chao A, Lin CT, Huang SL, Chao FY, Qiu JT, Hong JH, Chou HH, Chang TC, Chang CJ. Human papillomavirus genotype in cervical cancer: a population-based study. Int J Cancer 2007;120: Goodman MT, Shvetsov YB, McDuffie K, Wilkens LR, Zhu X, Thompson PJ, Ning L, Killeen J, Kamemoto L, Hernandez BY. Prevalence, acquisition, and clearance of cervical human papillomavirus infection among women with normal cytology: Hawaii Human Papillomavirus Cohort Study. Cancer Res 2008;68: Ho GY, Bierman R, Beardsley L, Chang CJ, Burk RD. Natural history of cervicovaginal papillomavirus infection in young women. N Engl J Med 1998;338: Franco EL, Villa LL, Sobrinho JP, Prado JM, Rousseau MC, Desy M, Rohan TE. of acquisition and clearance of cervical human papillomavirus infection in women from a high-risk area for cervical cancer. J Infect Dis 1999;180: Richardson H, Kelsall G, Tellier P, Voyer H, Abrahamowicz M, Ferenczy A, Coutlee F, Franco EL. The natural history of typespecific human papillomavirus infections in female university students. Cancer Epidemiol Biomarkers Prev 2003;12: Giuliano AR, Harris R, Sedjo RL, Baldwin S, Roe D, Papenfuss MR, Abrahamsen M, Inserra P, Olvera S, Hatch K. Incidence, prevalence, and clearance of type-specific human papillomavirus infections: the Young Women s Health Study. J Infect Dis 2002;186: Safaeian M, Kiddugavu M, Gravitt PE, Gange SJ, Ssekasanvu J, Murokora D, Sklar M, Serwadda D, Wawer MJ, Shah KV, Gray R. Determinants of incidence and clearance of high-risk human papillomavirus infections in rural Rakai, Uganda. Cancer Epidemiol Biomarkers Prev 2008;17: Rousseau MC, Pereira JS, Prado JC, Villa LL, Rohan TE, Franco EL. Cervical coinfection with human papillomavirus () types as a predictor of acquisition and persistence of infection. J Infect Dis 2001;184: Thomas KK, Hughes JP, Kuypers JM, Kiviat NB, Lee SK, Adam DE, Koutsky LA. Concurrent and sequential acquisition of different genital human papillomavirus types. J Infect Dis 2000;182: Insinga RP, Dasbach EJ, Elbasha EH, Liaw KL, Barr E. Incidence and duration of cervical human papillomavirus 6, 11, 16, and 18 infections in young women: an evaluation from multiple analytic perspectives. Cancer Epidemiol Biomarkers Prev 2007;16: Woodman CB, Collins S, Winter H, Bailey A, Ellis J, Prior P, Yates M, Rollason TP, Young LS. Natural history of cervical human papillomavirus infection in young women: a longitudinal cohort study. Lancet 2001;357: