Behavioral and sociodemographic risk factors for serological and DNA evidence of HPV6, 11, 16, 18 infections

Cancer Epidemiology 36 (2012) e183–e189

Contents lists available at SciVerse ScienceDirect

Cancer Epidemiology The International Journal of Cancer Epidemiology, Detection, and Prevention journal homepage: www.cancerepidemiology.net

Behavioral and sociodemographic risk factors for serological and DNA evidence of HPV6, 11, 16, 18 infections§ Dorothy J. Wiley a,*, Emmanuel V. Masongsong a, Shuang Lu b, Heather L. Singsb, Benissa Salem a, Anna R. Giuliano c, Kevin A. Ault d, Richard M. Haupt b, Darron R. Brown e a

Translational Science Section, School of Nursing, University of California at Los Angeles (UCLA), 700 Tiverton Avenue, Rm 5-151, Los Angeles, CA 90095-6919, United States Merck Sharp & Dohme, Corp., One Merck Drive, Whitehouse Station, NJ 08889-0100, United States c H. Lee Moffit Cancer Center and Research Institute, 12902 Magnolia Drive, Tampa, FL 33612, United States d Department of Gynecology and Obstetrics, Emory University School of Medicine, Glenn Building, 69 Jesse Hill, Jr., Drive, S.E., Rm 406, Atlanta, GA 30303, United States e Department of Medicine, Indiana University School of Medicine, 545 N. Barnhill Dr., Indianapolis, IN 46202-5124, United States b

A R T I C L E I N F O

A B S T R A C T

Article history: Accepted 26 December 2011 Available online 25 January 2012

Introduction: Risk for HPV6/11/16/18 infections in young sexually active, behaviorally low-risk females is not well described and may inform public policy. Methods: To assess exposure risk for HPV/6/11/16/18 among 16–23 year old low-risk females, data for 2409 female clinical trial participants were evaluated. Baseline visit self-reported sexual, behavioral and demographic characteristics; and results from HPV genotyping and serology, and other clinical laboratory assays were analyzed. All subjects reported <5 lifetime male sexual partners and no prior abnormal cytology at baseline. Results: While 98% (2211/ 2255) were naı¨ve to HPV16 or 18 and 99.6% (2246/2255) were naı¨ve for 1–3 index HPVs, 27% (616/2255) showed antibody, DNA or both for 1 index HPV. While 18% (409/2255) tested HPV16- or -18-antibodyor -DNA-positive, only 2% (44/2255) tested positive for both types. Against this high background, other sexually transmitted infections (STIs) were uncommonly detected, suggesting low sexual risk-taking behavior. The adjusted analyses showed race, age, alcohol consumption, current Chlamydia trachomatis (chlamydia) and Trichamonas vaginalis (trichomoniasis), bacterial vaginosis (BV), number of lifetime male sex partners predicted positive index-HPV antibody test results. However, only the number of male sex partners predicted positivity for HPV6/11- and 16/18-DNA, and chlamydia infection predicted positivity for HPV6/11-DNA alone. Conclusions: Taken together, type-specific HPV-DNA and -antibody evidence of HPV6/11/16/18 infections among behaviorally low-risk 16–23 year old females is high. Since almost all participants would have benefited by either currently available bivalent or quadrivalent vaccine strategies, delaying vaccination beyond menarche may be a missed opportunity to fully protect young females against HPV6/11/16/18 infections and related dysplasias. Early diagnosis and treatment of chlamydia and trichomonas may be important in HPV pathogenesis. ß 2012 Elsevier Ltd. All rights reserved.

Keywords: HPV-infection HPV-serology HPV-DNA detection HPV behavioral risk factors Cervicovaginal infections Chlamydia and HPV Trichomonas and HPV Bacterial vaginosis and HPV

1. Introduction Based upon randomized clinical trials (RCT) data from worldwide studies, the European Medicines Agency (EMEA) and the United States Food and Drug Administration (FDA) approved two prophylactic human papillomavirus (HPV) vaccines: a bivalent virus-like particle (VLP) vaccine targeting HPV16 and 18 (Cervarix, GlaxoSmithKline) approved in 2007 and 2009, respectively, and a

§

Clinical Trial Registration # NCT00365378. * Corresponding author at: School of Nursing, University of California at Los Angeles (UCLA), 700 Tiverton Avenue, Factor Building 5-151, Los Angeles, CA 90095-6919, United States. Tel.: +1 310 825 0540; fax: +1 310 206 0606. E-mail address: [email protected] (D.J. Wiley). 1877-7821/$ – see front matter ß 2012 Elsevier Ltd. All rights reserved. doi:10.1016/j.canep.2011.12.007

quadrivalent VLP vaccine targeting HPV6, 11, 16 and 18 (Gardasil, Merck) approved by both agencies in 2006 [1]. Both vaccines were highly efficacious in large phase 3 clinical trials in preventing vaccine-type specific Grade II/III cervical intraepithelial neoplasias (CIN) [2–4] and quadrivalent vaccine is effective in preventing vaginal and vulvar intraepithelial neoplasias (VaIN and VIN, respectively), external genital warts (EGWs), and anal intraepithelial neoplasias (AIN) [5–8]. Recently published population-based analyses demonstrate that vaccination already shows evidence of reducing HPV-related genital warts and improving the public’s health [9]. While sexual activity patterns among adolescent and young adult females recently show a delay in onset, patterns continue to show a large fraction are sexually active but that risky sexual behaviors may be relatively low, overall. For example, 30% and

e184

D.J. Wiley et al. / Cancer Epidemiology 36 (2012) e183–e189

>70% of 15–17 and 18–24-year old U.S. females reported ever having had sexual intercourse in 2002 population surveys [10], while population surveillance data show Chlamydia trachomatis (chlamydia) and Neisseria gonorrhea (gonorrhea) are uncommonly diagnosed, overall: 3% and 0.6–0.7%, respectively [11,12]. In contrast, for 15–24 year old females evaluated at STI clinics and juvenile detention centers show chlamydia rates were >2–6 times higher: 11–15% and 11–17%, respectively, in 2000–2004 [13]. Nonetheless, population-based surveys show the majority of 9– 27 year old females would benefit by HPV vaccination using either licensed product [14,15] and that vaccination is cost effective. The reduced incidence of high-grade dysplasias, and in the case of the quadrivalent vaccine, genital warts, may provide significant healthcare cost savings when 12–24 year old females are vaccinated [16–19]. However, what may inform public discourse and patient and provider opinion about HPV vaccination in postmenarchal females is a better understanding of risk for exposure to HPV6, 11, 16 and 18 for sexually active, behaviorally low-risk late adolescent and young adult females. Evaluating prevalence and risk factors for current HPV infection and serological evidence type-specific anti-HPV antibody may inform public policy. Accordingly, data for 16–23 year old females who were enrolled in a phase 2 RCT for a prophylactic HPV vaccine candidate were analyzed. Data herein evaluate the baseline virus and immunological characteristics relative to sexual behaviors, concurrent infections, and tobacco and alcohol use of that cohort and their relationship to serological and DNA evidence of HPV 6, 11, 16 and 18 infections. 2. Methods Beginning in October 1998, 2409 females provided written informed consent and were enrolled in a randomized placebocontrolled HPV16 vaccine trial at 16 continental U.S. sites over 12 months. The study was conducted in accordance with principles of Good Clinical Practice and was approved by the appropriate institutional review boards and regulatory agencies. Results from the vaccine trial are published elsewhere [20]. Briefly, the study was limited to unmarried 16–23 year-old females reporting 5 or fewer lifetime male sexual partners at enrollment, and no history of low- or high-grade squamous intraepithelial lesions (LSIL and HSIL), in situ or invasive cancer. Self-report data for personal and family medical history were recorded, and physical examinations, including a pelvic examination and Pap test, were performed at the first study visit. Cervical swab, external genital swab, and cervicovaginal lavage (CVL) specimens were obtained for HPV6, 11, 16 and 18 DNA testing. Venous blood specimens were collected and serum was separated. Serum, swab and CVL specimens were snap frozen using dry ice and stored at 80 8C until shipped to a central laboratory. These analyses were limited to the 2255 (2255/ 2409, 94%) enrolled women showing complete data for HPV genotyping and serology assays. A sensitive and reliable competitive Luminex Immunoassay (cLIA) was developed to detect anti-HPV6, 11, 16 and 18 (index HPVs) antibody, which has been reported comprehensively elsewhere [21]. At the first study visit, 98% (2361/2409) of serum specimens were sufficient for antibody testing for index HPVs. Detection of HPV6, 11, 16, and 18 DNA in swab and lavage specimens was performed using PCR amplification for index-HPVspecific L1, E6 and E7, followed by direct hybridization using specific oligonucleotide probes as described by others [20]. At the first study visit, nearly all (99%, 2378/2409) lavage and swab samples were sufficient for DNA testing. Clinical procedures at the first study visit included a full physical examination, a vaginal exam using speculum, and specimen collection for routine diagnoses and study-related

evaluations using standardized procedures. Routinely collected genital specimens were evaluated for bacterial vaginosis (BV), Trichomonas vaginalis (Trichomoniasis), Candidal vaginitis (yeast), gonorrhea, and chlamydia and data were included in the analyses due to potential confounding and effect-modifying relationships [22–26]. BV was diagnosed clinically when three of four criteria were present: vaginal pH > 4.5, clue cells on wet mount examination using microscopy, amine odor when potassium hydroxide was applied, and homogeneous vaginal discharge. When motile T. vaginalis was observed using microscopy in combination with pH >4.5, trichomoniasis was diagnosed. Vaginal yeast, usually caused by Candida albicans, was diagnosed when KOH-treated specimens showed yeast or pseudohyphae of Candida species under microscopy [27]. Cervical swab specimens were evaluated using molecular assays for gonorrhea and chlamydia. Only 10 subjects showed missing data for chlamydia, gonorrhea, BV, trichomoniasis, or yeast. Statistical analyses included tabular and descriptive analyses that estimated the frequency of occurrences and contrasted the odds of testing positive for index HPVs or anti-HPV-specific antibody (versus not) across demographic and risk behavior strata [28]. Trends in frequencies were assessed using the CochranArmitage Trend Test [28]. Maximum likelihood logistic regression was used to summarize the data and obtain adjusted estimates [29]. Sociodemographic characteristics such as race, chronological age, age at first intercourse, tobacco, alcohol use, contraceptive use, and lifetime number of sex partners are differentially associated with one another [30–35] and sexually transmitted infections, including HPV [36–38]. Thus, to control for confounding and explore risk factors for HPV-DNA and -antibody evidence of infection, the adjusted analyses included race and age, tobacco smoking, number of alcoholic beverages consumed weekly, contraception methods, age at first intercourse, lifetime number of male sexual partners; and clinical evidence of BV or current infection with chlamydia, trichomoniasis, gonorrhea, and yeast. Because model coefficients can be easily transformed into odds ratios, a series of adjusted odds ratios were used to compare groups of women in these analyses [29]. Associations found statistically significant for any of four models are contrasted herein (Fig. 1) and detailed bivariate and adjusted summary estimates are posted as supplementary tables for the reader (Supplementary Tables 1 and 2). 3. Results Most women would have benefitted by vaccination using either the bivalent (HPV16/18) or quadrivalent (HPV6/11/16/18) vaccine. Specifically, 98% (2211/2255) were naı¨ve to HPV16 or 18 and 99.6% (2246/2255) were naı¨ve for 1–3 index HPVs (Table 1). Overall, 27.3% (616/2255) showed antibody, DNA or both for 1 index HPV (Table 1). The study group reflected common characteristics of the ‘‘catch-up’’ vaccine-targeted population of 16–23 year old females. On average, young adult, non-smoking, nulliparous, Caucasian females comprise the study group. The age range for enrollment eligibility was narrow, ranging from 16 to 23 years; and the mean age was 20.1 (1.7) years at the first study visit (Table 1). While race and ethnic minorities were represented in the cohort, the prevalence of African Americans and Hispanics was lower than might be expected when referring to the 2000 U.S. Census data for 15–24 year old females: 9% versus 15% and 8% versus 16%, respectively [39]. Current and prior tobacco use was reported by 40% of participants, approximating smoking prevalence for U.S. females >18 years in 1990–1998, 40.7% [40]. Contraception has long been linked to HPV-related cancers and currently sexually active females reporting use of effective

D.J. Wiley et al. / Cancer Epidemiology 36 (2012) e183–e189 Anti-HPV6/11 antibody in serum

e185

HPV6/11 DNA in cervicovaginal specimen

Anti-HPV16/18 antibody in serum

HPV16/18 DNA in cervicovaginal specimen

Odds ratioa (95% CI)

Baseline Characteristic

Race (non-white vs. white) Ageb Number of alcoholic beverages per week (1 to 4) Number of alcoholic beverages per week (5 or more) Contraceptive use = Barrier Current bacterial vaginosis Current Chlamydia trachomatis Current Trichomonas vaginalis Lifetime number of male sexual partners at enrollment Age of sexual debut

0.1

1.0

10.0

a

Odds Ratios are adjusted for the effect of race, history of smoking, number of alcoholic beverages consumed weekly; history of bacterial vaginosis, Chlamydia, Trichomoniasis, Gonorrhea, yeast; barrier, behavioral, hormonal and other forms of contraception; age at first intercourse, chronological age, and lifetime number of male sexual partners. Only those variables which were significant for at least one category of anti-HPV antibody or DNA are plotted. d Age measured as a continuous variable. Fig. 1. Association between demographic, sexual-behavioral and infection characteristics and anti-HPV6/11 and -HPV16/18 antibody in serum and cervicovaginal HPV6/11 and HPV16/18 DNA.

contraception were over-represented by design in this study (reviewed by Wheeler and colleagues [41]). Non-users were few and were limited to otherwise eligible females whose stated intent was to become sexually active soon after enrollment. Oral, implanted, injected or indwelling hormonal contraceptives were the most common sole form of contraception (45.5%, 1026/2255), but their use alone or in combination with barrier methods was reported by 53.0% (1194/2255) of participants. Use of any barrier contraception was reported by 37.8% (852/2255) of study participants, possibly related to the newness of sexual relationships. Nearly two-thirds (64.3%, 1380/2255) of females reported no new sexual partnerships during the 6 months preceding the first study visit. The mean and median self-reported age at first intercourse was 16.9 (1.8) and 17 years, respectively, and ages ranged from 16 to 23 years (Table 1). Cervicovaginal infections, some of which are proinflammatory and possibly salient to the natural history of infection, were detected in this study group at the baseline visit. Specifically, yeast was detected in 10.4% (235/2255) of participants, and BV and chlamydia were detected among 5.6% (126/2255) and 3.6% (81/2245), respectively (Table 1). Trichomoniasis (0.4%, 8/2255), herpes simplex (0.08%, 2/2245) and gonorrhea (0.4%, 8/2245) were rarely detected among participants and only one persistent hepatitis B infection was detected using serology (Table 1). Together only 4.1% (93/2255) of females tested positive for chlamydia, trichomonas, or gonorrhea at the first study visit. From the bivariate analyses, detection of BV, chlamydia, gonorrhea, and trichamoniasis was positively associated with detection of both antibody in serum and DNA in cervicovaginal specimens for HPV6/11 or HPV16/18 to varying extents (Supplementary Tables 1 and 2). Despite infrequent detection of other sexually transmitted infections, HPV-DNA and -antibody were commonly identified in

this study group. In total, 27.3% (616/2255) of females tested positive for one or more index HPVs using antibody or DNA assay measures (Table 1). The prevalence of antibody positivity nearly approximated detection of DNA for one or more index HPVs: 18.6% (420/2255) versus 16.5% (371/2255), respectively (p > 0.05, Table 1). When evaluated alone, anti-HPV6 and 16 antibody (9–10%) and HPV16 DNA (9%) were near equally detected (Table 2). Nonetheless, while 26.0% (579/2255) tested positive for 1–2 index HPVs, rarely did females test positive for 3 or 4 of index HPVs using serology or PCR: 1.2% (28/2255) and 0.4% (9/2255), respectively (Table 1). Nearly 11% (246/2255) tested HPV DNA-positive and serologically negative for 1–4 of the same index HPVs and 6.5% (148/2255) similarly tested positive for both antibody and DNA for the same type(s) (Table 1). Antibody and PCR positivity relationships varied across HPV types. For example, among women who tested antibody-positive for HPV6, 11 16, and 18, fewer than half tested positive for viral DNA: 22% (50/231), 6% (4/62), 40% (85/ 210), and 26% (19/72), respectively (Table 2). 3.1. Multivariate analyses Four final logistic regression models summarized the important associations and showed risk factors for anti-HPV antibodies in serum for both HPV6/11 and HPV16/18 were similar, but that their predictive relationship with HPV DNA showed some differences that are contrasted in Fig. 1, and detailed in two Supplementary Tables. For example, self-reported lifetime number of male sex partners predicted risk for detecting both HPV6/11 and HPV16/18 antibody (Fig. 1, Supplementary Tables 1 and 2). Both age and chlamydia infection were positively associated with detecting both HPV6/11 and HPV16/18 antibody (Fig. 1, Supplementary Table 1) and with HPV6/11 but not HPV16/18 DNA (Fig. 1, Supplementary

D.J. Wiley et al. / Cancer Epidemiology 36 (2012) e183–e189

e186

Table 1 Baseline demographic, sexual-behavioral and infection characteristics for a sample of 2255 16–23 year-old females.

Table 1 (Continued ) Total (N = 2255) (%)

n Total (N = 2255) Age Mean Age at first sexual intercoursea Mean Median

20.1  1.7 16.9  1.8 17

(0.4) (10.9) (14.1) (6.5)

a

Among non-virgins. For Chlamydia and Gonorrhea testing, 2245 subjects have complete test results. For all others listed, N = 2255. c Hepatitis B HBsAg, anti-HBc positive. d HPV6/11: females that test positive for either HPV6 or 11. e HPV16/18: females that test positive for either HPV16 or 18. f Results pertain to results for females with >1 of same index HPV genotypes. b

Total (N = 2255) n

4 types 9 HPV 6, 11, 16, or 18 Type-specific Results: I1 typef PCR-positive, serology-negative 246 PCR-negative, serology-positive 317 148 PCR-positive, serology-positive

(%)

Race and ethnicity Asian 132 (5.9) African American 191 (8.5) Hispanic American 175 (7.8) Native American 22 (1.0) 44 (2.0) Other White 1691 (75.0) Smoking status Ever smoker 890 (39.5) Current smoker 564 (25.0) Ex-smoker 326 (14.5) Never smoked 1352 (60.0) 13 (0.6) Unknown Contraception methods Single method Barrier only 683 (30.3) Behavioral only 353 (15.7) Hormonal only 1026 (45.5) ‘‘Other’’ only 9 (0.4) Multiple methods Barrier and hormonal 153 (6.8) Other combination 31 (1.4) Lifetime number of male sexual partners at enrollmenta 0 108 (4.8) 1 521 (23.1) 2 450 (20.0) 3 419 (18.6) 4 432 (19.2) 5 325 (14.4) Number of new male sexual partners in the 6 months prior to study starta 0 1380 (64.3) 1 643 (29.9) 2 106 (4.9) 3 15 (0.7) 4 2 (0.1) 5 1 (<0.1) Infectionsb Chlamydia 81 (3.6) Gonorrhea 8 (0.4) Bacterial Vaginosis 126 (5.6) c Hepatitis B (serology) 1 (<0.0) Herpes 2 (0.1) Trichomonas 8 (0.4) Yeast 235 (10.4) HPV 6, 11, 16, or 18 positive by Serology 420 (18.6) High-risk versus low-risk HPV genotype groupings d e HPV6/11 and HPV16/18 92 (4.1) HPV16 or HPV18 only 163 (7.2) HPV6 or HPV11 only 165 (7.5) PCR 371 (16.5) High-risk versus low-risk hpv genotype groupings d e HPV6/11 and HPV16/18 30 (1.3) HPV16 or HPV18 only 234 (10.4) HPV6 or HPV11 only 107 (4.7) Serology or PCR 616 (27.3) High-risk versus low-risk HPV genotype groupings HPV6/11d and HPV16/18e 133 (5.9) HPV16 or HPV18 276 (12.2) HPV16 and HPV18 44 (1.9) HPV16 or HPV18, not both 232 (10.2) HPV6 or HPV11 only 207 (9.2) Number of index Hpv genotypes detected 1 type 470 (20.8) 2 types 109 (4.8) 3 types 28 (1.2)

Tables 1 and 2). Risk for testing HPV6/11 or HPV16/18 antibody positive increased 19% with each additional year of age and risk was nearly double for chlamydia-infected over -uninfected subjects (Fig. 1, Supplementary Table 1). However, while chlamydia-infected females were 2.2 times more likely than unaffected females to test positive for HPV6/11 DNA, they were no more likely to test similarly positive for HPV16/18 DNA, i.e., OR = 2.16 (1.17, 3.98) versus OR = 1.24 (0.73, 2.09), respectively. BV diagnosis carried a two-fold greater risk of testing positive for HPV6/11 antibody, but did not show greater odds for detecting HPV16/18 serology or DNA or HPV6/11 DNA (Fig. 1, Supplementary Tables 1 and 2). Nonetheless, some behavioral characteristics differentially influenced positive serology over detection of type-specific DNA in these analyses. For example, alcohol decreased the prevalence of antibody detected in serum but had no influence over detection of type-specific DNA for index HPVs (Fig. 1, Supplementary Table 2). Specifically, females who reported drinking 1–5 alcoholic beverages weekly were 1.39–2.04 times less likely than non-drinkers to test positive for anti-HPV16/18 antibodies (p < 0.05) and although estimates for serum anti-HPV6/11 antibody did not reach statistical significance, the effects were similar and a test for trend was statistically significant, p < 0.0001 and p < 0.0001 for

Table 2 Composite summary of HPV genotyping and type-specific serology characteristics for 2255 16–23 year-old females.a HPV Type-specific DNA by PCR

HPV Type-specific serology

Total (%)

Positive n (column %)

Negative n (column %)

50 (22) 181 (78) 231 (10)

61 (3) 1963 (97) 2024 (90)

111 (5) 2144 (95) 2255

HPV 11 HPV 11 PCR (+) HPV 11 PCR ( ) Total (row%)

4 (6) 58 (94) 62 (3)

25 (1) 2168 (99) 2193 (97)

29 (1) 2226 (99) 2255

HPV 16 HPV 16 PCR (+) HPV 16 PCR ( ) Total

85 (40) 125 (60) 210 (9)

114 (6) 1931 (94) 2045 (91)

199 (9) 2056 (91) 2255

HPV 18 HPV 18 PCR (+) HPV 18 PCR ( ) Total

19 (26) 53 (74) 72 (3)

65 (3) 2118 (97) 2183 (97)

84 (4) 2171 (96) 2255

HPV 6 HPV 6 PCR (+) HPV 6 PCR ( ) Total (row%)

n, number affected in each Type-specific category. a Paired serology and CVL and swab specimens for 2255; individual women counted more than once in the categories summarizing serology- and swabspecimens results.

D.J. Wiley et al. / Cancer Epidemiology 36 (2012) e183–e189

HPV6/11 and HPV16/18, respectively (Fig. 1, Supplementary Table 2). However, HPV DNA measures were not associated with alcohol consumption (Fig. 1, Supplementary Table 2). Further, barriercontraception use alone influenced HPV16/18 serology only. Specifically, barrier-method users were 39% less likely than hormone-contraceptive users to show type-specific antibody in serum. Otherwise, contraceptive methods were not associated with detecting either anti-HPV6/11 antibody in serum or HPV6/11 or HPV16/18 DNA. 4. Discussion Almost all females in this study group would have benefitted by HPV vaccination, even if they tested positive for 1 virus type. The majority of females studied showed naı¨vety to 1–4 index HPVs and would benefit from HPV vaccination strategies currently approved for administration in the U.S. and worldwide. More important, analyses suggest that 98% of these 16–23 year olds would be protected against the high-risk HPVs 16/18 using either the bivalent or quadrivalent vaccine strategies. Additionally, 99.6% of women studied would be protected against 1–3 of 4 index HPVs using the quadrivalent HPV vaccine. However, these data underscore the commonness of these infections: more than onequarter showed antibody or DNA evidence of HPV6, 11, 16, or 18 infections. Although our estimates varied across types and within high- and low-risk HPV (hr-HPV, lrHVP) groupings, they are broadly consistent with published data. For example, a range of observational studies, clinical trials and worldwide estimates suggest that 10–42% of cytologically normal females may test positive for HPV DNA [42–49]. Further, age-adjusted estimates from meta-analyses among North American females suggest 11.3% of cytologically normal females test positive for HPV-DNA, onethird of which are attributable to HPV16 alone [49]. NHANES data show 27% of 14–59 year old U.S. Females tested positive for any HPV DNA type using PCR, however the specific prevalence of the four index HPVs was substantially lower than observed in this study group [50]. Sexual partnership characteristics influence risk for detecting HPV antibody and DNA. Like others, our analyses confirm that lifetime number of male sexual partners increases the risk for detecting antibody in serum and HPV-DNA in CVL specimens, even after we controlled for the effect of other covariates. Similar reported positive associations between HPV infection and lifetime number of male sex partners are numerous [33,43,35,51–55]. However, the clinical trial population is not entirely representative of the general population of women aged 16–26 due to the exclusion/inclusion criteria of the trials (e.g., low lifetime number of sex partners and no past history of abnormal Pap test or external genital abnormality) and our results cannot be generalized to other populations with different background risk. For example, chlamydia, gonorrhea, and trichomonas were infrequently detected as compared to population-based surveillance data for gonorrhea (0.66–0.72%) and chlamydia (5.2%) taken from among family planning clinic attendees during this period [56]. Nonetheless, 16.5% tested positive for HPV6/11/16/18 DNA using PCR and when evaluated jointly with serology, 27.3% tested similarly positive. Together, these data suggest that risk for HPV6/11/16/18 infection is high, even in a relatively low-risk adolescent and young-adult group of sexually active females. Other cervicovaginal infections and conditions influence HPV antibody and DNA positivity. Chlamydia infections are relevant to cervical cancer beyond risk-taking behaviors and characteristics of sexual networks. In our study, chlamydia-infected females were 2.2 times more likely than unaffected females to test positive for HPV6/11 DNA and to test similarly positive for HPV16 or -18 DNA [57]. Recently, the large phase 3 clinical trials of the quadrivalent

e187

vaccine have shown that baseline chlamydia positivity was associated with CIN2 but not with CIN3 in placebo recipients, suggesting that chlamydia infection might facilitate the development of early cervical lesions [57]. Analyses also suggest that chlamydia infection upregulates VEGF-C, NF-Kb and survivin in coinfected CIN and cervical cancer specimens when compared to otherwise similar tissues testing positive for hrHPV alone [58]. However, the pathobiology of the possible role of chlamydia remains open. Further, while some studies suggest BV may result from HPV infection, a more recent study suggests causality and that BV may slow HPV clearance among the infected [22,24,59]. It may be that altered vaginal flora associated with BV modifies the microenvironment in ways that both favor primary HPV infection and impede the immune responses. Sociodemographic characteristics were differentially associated with detection of HPV antibody and DNA. Herein, age was positively and equally associated with risk for detecting antiHPV6/11 and HPV16/18 antibody in serum, and is consistent with some seroprevalence patterns previously reported [60]. However, while HPV DNA detection was not associated with age in these data, it may be more challenging and less durable than is antibody in serum. Nonetheless, some published data show cervical HPV DNA detection is highest among young adult females, especially when measured across a broader age range than in these data [61– 63]. Generally, reports consistently show the risk for single and multi-type HPV infections and persistence decreases with advancing age [43,35,51,64,65]. Additionally, while we found no significant association between race and detecting HPV6/11 or HPV16/18-DNA, we found anti-HPV6/11 antibody was more often detected in Whites than non-Whites. Published reports about these associations are mixed and some data show some differences between non-Hispanic whites and minority populations for index HPV serology [60,66]. Other data show race is unrelated to seroconversion or reversion following type-specific DNA detection in CVL [67] and findings in at least one RCT showed no differences in geometric mean (serum) antibody titers with HPV-VLP vaccination across race groups, suggesting race-based differences may be due to characteristics of natural infections rather than underlying biological variation across groups [47]. Several behavioral characteristics reported here are notable. Participants reporting use of barrier contraceptives were 39% less likely than hormone contraceptive-users to test positive for antiHPV16/18 antibody in serum, possibly consistent with the influence of endogenous and exogenous sex hormones on virus replication (as reviewed in De Villiers) [68]. However, strong independent associations between hormone contraceptives and cervical cancer are reported for women with >5 years of exposure, their initiation before age 20, and with recent use relative to diagnosis of malignancy, most of which cannot be characterized for this group [69]. Other investigators have reported a negative association between alcohol use and important humoral responses to HPVs [20,35,70,71] and our findings are similar. Conversely, while we found no association between alcohol consumption and detection of index HPV DNA, our data do suggest alcohol may blunt immune responses rather than suggest females with higher use were unexposed to the index HPVs. Experimental data are supportive of this hypothesis, suggesting lower serum IgG, IgM, and IgA, and proinflammatory cytokine levels are associated with high alcohol consumption [72–74]. Importantly, the association between alcohol consumption, number of partnerships and the risk for other STIs often confounds our understanding of underlying causal associations. While we believe these analyses contribute significantly to public discourse about vaccinating post-menarchal females, some caveats remain. For example, persistent and recrudescent HPV infections and repeated exposures to the virus cannot be

e188

D.J. Wiley et al. / Cancer Epidemiology 36 (2012) e183–e189

distinguished in these cross-sectional data. Many variables rely upon self-reported data from participants. Future research would be strengthened by widening the age range of the study cohort, and by increasing participation among additional racial, ethnic, and socioeconomic groups of women. In conclusion, even among behaviorally low-risk adolescent and young adult females, the prevalence of historical and current HPV6/11/16/18 infections in our study cohort was high. The paucity of sexual-behavioral risk factors causing disparities in antibody or DNA evidence of infection makes it clear that prevention requires strategies beyond behavior modification. Most 16–23 year olds would benefit from either currently available bivalent or quadrivalent vaccine strategies. Developing vaccines that more comprehensively protect against all of the 13 hrHPVs would further enhance prevention strategies. Although benefit would still be derived from post-menarchal vaccination, delaying vaccination beyond sexual debut, or at least menarche, is a missed opportunity to fully protect young females against HPV6/11/16/18 infections and resulting dysplasias. Further, our analyses suggest that effective prevention strategies for HPV-related infections and dysplasias should include development of better detection and treatment services for chlamydia, trichomonas, and BV. At a population level, infections prevented through sufficiently early HPV vaccination would most dramatically reduce long-term costs of Pap test abnormalities, treatment for low- and high-grade dysplasias, and invasive cancers, as well as unnecessary premature mortality. Conflict of interest statement All authors have completed the ICMJE form for disclosure of potential conflict of interest. Dr. Wiley has received grants and payment for lectures including service on speakers bureaus from Merck. Ms. Salem has no financial disclosures. Mr. Masongsong has received grant support from Merck. Dr. Brown has patents, has received royalties, owns stock/stock options, and has received travel/accommodations/meeting expenses from Merck. He also has board membership for PDS Incorporated. Dr. Ault has received grants from Merck, Gen Probe, Roche and the NIH. Dr. Giuliano has received grants, board membership, consultancy fees and payment for lectures including service on speakers bureaus from Merck. Drs. Sings, Haupt, and Mr. Lu are employees of Merck and may own stock or stock options in the company. Source of funding Merck. This post hoc analysis was designed by the named external investigators in collaboration with the named sponsor coauthors. The sponsor (S. Lu) performed the statistical analysis. All authors were actively involved in the collection, analysis or interpretation of the data, the revising of the manuscript for intellectual content, and the decision to submit for publication. Acknowledgments We thank all study participants and investigators. We also thank Ms. Margaret James, Karyn Davis (Merck), Zoe Masongsong, and Hilary Hsu for assistance in the preparation of this manuscript.

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.canep.2011.12.007.

References [1] Pomfret TC, Gagnon Jr JM, Gilchrist AT. Quadrivalent human papillomavirus (HPV) vaccine: a review of safety, efficacy, and pharmacoeconomics. J Clin Pharm Ther 2011;36(1):1–9. [2] Paavonen J, Naud P, Salmeron J, Wheeler CM, Chow SN, Apter D, et al. Efficacy of human papillomavirus (HPV)-16/18 AS04-adjuvanted vaccine against cervical infection and precancer caused by oncogenic HPV types (PATRICIA): final analysis of a double-blind, randomised study in young women. Lancet 2009;374(9686):301–14. [3] Garland SM, Hernandez-Avila M, Wheeler CM, Perez G, Harper DM, Leodolter S, et al. Quadrivalent vaccine against human papillomavirus to prevent anogenital diseases. N Engl J Med 2007;356(19):1928–43. [4] Quadrivalent vaccine against human papillomavirus to prevent high-grade cervical lesions. N Engl J Med 2007;356(19):1915–27. [5] Dillner J, Kjaer SK, Wheeler CM, Sigurdsson K, Iversen OE, Hernandez-Avila M, et al. Four year efficacy of prophylactic human papillomavirus quadrivalent vaccine against low grade cervical, vulvar, and vaginal intraepithelial neoplasia and anogenital warts: randomised controlled trial. BMJ 2010;341(c3493): 1–9. [6] Kjaer SK, Sigurdsson K, Iversen OE, Hernandez-Avila M, Wheeler CM, Perez G, et al. A pooled analysis of continued prophylactic efficacy of quadrivalent human papillomavirus (Types 6/11/16/18) vaccine against high-grade cervical and external genital lesions. Cancer Prev Res (Phila Pa) 2009;2(10):868–78. [7] Garland SM. Prevention strategies against human papillomavirus in males. Gynecol Oncol 2010;117(2 Suppl.):S20–5. [8] FDA licensure of quadrivalent human papillomavirus vaccine (HPV4, Gardasil) for use in males and guidance from the Advisory Committee on Immunization Practices (ACIP). MMWR Morb Mortal Wkly Rep 2010;59(20):630–2. [9] Fairley CK, Hocking JS, Gurrin LC, Chen MY, Donovan B, Bradshaw CS. Rapid decline in presentations of genital warts after the implementation of a national quadrivalent human papillomavirus vaccination programme for young women. Sex Transm Infect 2009;85(7):499–502. [10] Chandra A, Martinez G, Mosher W, Abma J, Jones J. In: National Center for Health Statistics CfDCap, ed. Fertility, family planning, and reproductive health of U.S. women: data from the 2002 National Survey of Family Growth. U.S. Department of Health and Human Services; 2005: 1–174. [11] Services. In: Administration HRaS, ed. USDoHaH. Women’s Health USA, 2010. Rockville, MD: U.S. Department of Health and Human Services, 2010: 80. [12] U.S. Center for Disease Control and Prevention. In: National Center for HIV/ AIDS VH, STD, and TB Prevention, Division of STD Prevention, ed. Sexually transmitted diseases: Figure B Data, All States: Chlamydia positivity in women 15–24 years of age by testing site, 2000–2009. Atlanta, GA: U.S. Department of Health and Human Services, 2009: 1–11. [13] Kreimer AR, Villa A, Nyitray AG, Abrahamsen M, Papenfuss M, Smith D, et al. The epidemiology of oral HPV infection among a multinational sample of healthy men. Cancer Epidemiol Biomarkers Prev 2011;20(1):172–82. [14] Olsson SE, Kjaer SK, Sigurdsson K, Iversen OE, Hernandez-Avila M, Wheeler CM, et al. Evaluation of quadrivalent HPV 6/11/16/18 vaccine efficacy against cervical and anogenital disease in subjects with serological evidence of prior vaccine type HPV infection. Hum Vaccin 2009;5(10):696–704. [15] Dempsey AF, Gebremariam A, Koutsky LA, Manhart L. Using risk factors to predict human papillomavirus infection: implications for targeted vaccination strategies in young adult women. Vaccine 2008;26(8):1111–7. [16] Dee A, Howell F. A cost-utility analysis of adding a bivalent or quadrivalent HPV vaccine to the Irish cervical screening programme. Eur J Public Health 2010;20(2):213–9. [17] Annemans L, Remy V, Oyee J, Largeron N. Cost-effectiveness evaluation of a quadrivalent human papillomavirus vaccine in Belgium. Pharmacoeconomics 2009;27(3):231–45. [18] Goldie SJ, Kohli M, Grima D, Weinstein MC, Wright TC, Bosch FX, et al. Projected clinical benefits and cost-effectiveness of a human papillomavirus 16/18 vaccine. J Natl Cancer Inst 2004;96(8):604–15. [19] Elbasha EH, Dasbach EJ, Insinga RP, Haupt RM, Barr E. Age-Based Programs for Vaccination against HPV. Value Health 2009. [20] Mao C, Koutsky LA, Ault KA, Wheeler CM, Brown DR, Wiley DJ, et al. Efficacy of human papillomavirus-16 vaccine to prevent cervical intraepithelial neoplasia: a randomized controlled trial. Obstet Gynecol 2006;107(1):18–27. [21] Dias D, Van Doren J, Schlottmann S, Kelly S, Puchalski D, Ruiz W, et al. Optimization and validation of a multiplexed luminex assay to quantify antibodies to neutralizing epitopes on human papillomaviruses 6, 11, 16, and 18. Clin Diagn Lab Immunol 2005;12(8):959–69. [22] Castle PE, Hillier SL, Rabe LK, Hildesheim A, Herrero R, Bratti MC, et al. An association of cervical inflammation with high-grade cervical neoplasia in women infected with oncogenic human papillomavirus (HPV). Cancer Epidemiol Biomarkers Prev 2001;10(10):1021–7. [23] Evans BA, Kell PD, Bond RA, MacRae KD. Racial origin, sexual lifestyle, and genital infection among women attending a genitourinary medicine clinic in London (1992). Sex Transm Infect 1998;74(1):45–9. [24] Mao C, Hughes JP, Kiviat N, Kuypers J, Lee SK, Adam DE, et al. Clinical findings among young women with genital human papillomavirus infection. Am J Obstet Gynecol 2003;188(3):677–84. [25] Marrazzo JM, Koutsky LA, Eschenbach DA, Agnew K, Stine K, Hillier SL. Characterization of vaginal flora and bacterial vaginosis in women who have sex with women. J Infect Dis 2002;185(9):1307–13.

D.J. Wiley et al. / Cancer Epidemiology 36 (2012) e183–e189 [26] Bunnell RE, Dahlberg L, Rolfs R, Ransom R, Gershman K, Farshy C, et al. High prevalence and incidence of sexually transmitted diseases in urban adolescent females despite moderate risk behaviors. J Infect Dis 1999;180(5):1624–31. [27] Centers for Disease Control and Prevention. Sexually transmitted diseases treatment guidelines. MMWR 2002;51(No. RR-6):1–84. [28] SAS Institute Inc. SAS: The Frequency Procedure. [Online; Index] Cary: SAS Institute Inc.; 2010. Available at: http://support.sas.com/documentation/cdl/ en/statug/63347/HTML/default/viewer.htm#/documentation/cdl/en/statug/ 63347/HTML/default/freq_toc.htm. [29] SAS Institute Inc. SAS: The Logistic Procedure. [Online; Index] Cary: SAS Institute Inc.; 2010. Available at: http://support.sas.com/documentation/ cdl/en/statug/63347/HTML/default/viewer.htm#/documentation/cdl/en/statug/63347/HTML/default/logistic_toc.htm. [30] Sun CA, Hsiung CA, Lai CH, Chen CA, Chou CY, Ho CM, et al. Epidemiologic correlates of cervical human papillomavirus prevalence in women with abnormal Pap smear tests: a Taiwan Cooperative Oncology Group (TCOG) study. J Med Virol 2005;77(2):273–81. [31] Chan PK, Chang AR, Cheung JL, Chan DP, Xu LY, Tang NL, et al. Determinants of cervical human papillomavirus infection: differences between high- and lowoncogenic risk types. J Infect Dis 2002;185(1):28–35. [32] Castle PE, Shields T, Kirnbauer R, Manos MM, Burk RD, Glass AG, et al. Sexual behavior, human papillomavirus type 16 (HPV 16) infection, and HPV 16 seropositivity. Sex Transm Dis 2002;29(3):182–7. [33] Munoz N, Kato I, Bosch FX, Eluf-Neto J, De Sanjose S, Ascunce N, et al. Risk factors for HPV DNA detection in middle-aged women. Sex Transm Dis 1996;23(6):504–10. [34] Burkett BJ, Peterson CM, Birch LM, Brennan C, Nuckols ML, Ward BE, et al. The relationship between contraceptives, sexual practices, and cervical human papillomavirus infection among a college population. J Clin Epidemiol 1992;45(11):1295–302. [35] 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(7):423–8. [36] Shields TS, Brinton LA, Burk RD, Wang SS, Weinstein SJ, Ziegler RG, et al. A casecontrol study of risk factors for invasive cervical cancer among U.S. women exposed to oncogenic types of human papillomavirus. Cancer Epidemiol Biomarkers Prev 2004;13(10):1574–82. [37] Wang SS, Zuna RE, Wentzensen N, Dunn ST, Sherman ME, Gold MA, et al. Human papillomavirus cofactors by disease progression and human papillomavirus types in the study to understand cervical cancer early endpoints and determinants. Cancer Epidemiol Biomarkers Prev 2009;18(1):113–20. [38] Herrero R, Castle PE, Schiffman M, Bratti MC, Hildesheim A, Morales J, et al. Epidemiologic profile of type-specific human papillomavirus infection and cervical neoplasia in Guanacaste, Costa Rica. J Infect Dis 2005;191(11):1796– 807. [39] U.S. Bureau of the Census. In: Division P, ed. Table 3: female population by age, sex, race and hispanic or latino origin for the United States: 2000. Washington, DC: U.S. Department of Commerce Economics and Statistics Administration, 2000. [40] Office on Smoking and Health. In: National Center for Chronic Disease Prevention Health Promotion. Office on Smoking, Health, ed. Women and smoking: a report of the surgeon general. United States: Public Health Service. Office of the Surgeon General, 2001. [41] Wheeler CM. Natural history of human papillomavirus infections, cytologic and histologic abnormalities, and cancer. Obstet Gynecol Clin North Am 2008;35(4):519–36. vii. [42] Bauer HM, Hildesheim A, Schiffman MH, Glass AG, Rush BB, Scott DR, et al. Determinants of genital human papillomavirus infection in low-risk women in Portland, Oregon. Sex Transm Dis 1993;20(5):274–8. [43] Giuliano AR, Papenfuss M, Abrahamsen M, Denman C, de Zapien JG, Henze JL, et al. Human papillomavirus infection at the United States-Mexico border: implications for cervical cancer prevention and control. Cancer Epidemiol Biomarkers Prev 2001;10(11):1129–36. [44] Bae JH, Lee SJ, Kim CJ, Hur SY, Park YG, Lee WC, et al. Human papillomavirus (HPV) type distribution in Korean women: a meta-analysis. J Microbiol Biotechnol 2008;18(4):788–94. [45] Giuliano AR, Harris R, Sedjo RL, Baldwin S, Roe D, Papenfuss MR, et al. Incidence, prevalence, and clearance of type-specific human papillomavirus infections: The Young Women’s Health Study. J Infect Dis 2002;186(4):462–9. [46] Nielsen A, Kjaer SK, Munk C, Iftner T. Type-specific HPV infection and multiple HPV types: prevalence and risk factor profile in nearly 12,000 younger and older Danish women. Sex Transm Dis 2008;35(3):276–82. [47] Giuliano AR, Lazcano-Ponce E, Villa L, Nolan T, Marchant C, Radley D, et al. Impact of baseline covariates on the immunogenicity of a quadrivalent (types 6, 11, 16, and 18) human papillomavirus virus-like-particle vaccine. J Infect Dis 2007;196(8):1153–62. [48] Ley C, Bauer HM, Reingold A, Schiffman MH, Chambers JC, Tashiro CJ, et al. Determinants of genital human papillomavirus infection in young women. J Natl Cancer Inst 1991;83(14):997–1003. [49] de Sanjose S, Diaz M, Castellsague X, Clifford G, Bruni L, Munoz N, et al. Worldwide prevalence and genotype distribution of cervical human papillomavirus DNA in women with normal cytology: a meta-analysis. Lancet Infect Dis 2007;7(7):453–9.

e189

[50] Dunne EF, Unger ER, Sternberg M, McQuillan G, Swan DC, Patel SS, et al. Prevalence of HPV infection among females in the United States. JAMA 2007;297(8):813–9. [51] Giuliano AR, Papenfuss M, Schneider A, Nour M, Hatch K. Risk factors for highrisk type human papillomavirus infection among Mexican-American women. Cancer Epidemiol Biomarkers Prev 1999;8(7):615–20. [52] Ho GY, Burk RD, Klein S, Kadish AS, Chang CJ, Palan P, et al. Persistent genital human papillomavirus infection as a risk factor for persistent cervical dysplasia. J Natl Cancer Inst 1995;87(18):1365–71. [53] Dillner J, Kallings I, Brihmer C, Sikstrom B, Koskela P, Lehtinen M, et al. Seropositivities to human papillomavirus types 16, 18, or 33 capsids and to Chlamydia trachomatis are markers of sexual behavior. J Infect Dis 1996;173(6):1394–8. [54] Burk RD, Ho GY, Beardsley L, Lempa M, Peters M, Bierman R. Sexual behavior and partner characteristics are the predominant risk factors for genital human papillomavirus infection in young women. J Infect Dis 1996;174(4):679–89. [55] Tarkowski TA, Koumans EH, Sawyer M, Pierce A, Black CM, Papp JR, et al. Epidemiology of human papillomavirus infection and abnormal cytologic test results in an urban adolescent population. J Infect Dis 2004;189(1):46–50. [56] Centers for Disease Control and Prevention. Sexually Transmitted Disease Surveillance, 2000. Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, September 2001. [57] Lehtinen M, Ault KA, Lyytikainen E, Dillner J, Garland SM, Ferris DG, et al. Chlamydia trachomatis infection and risk of cervical intraepithelial neoplasia. Sex Transm Infect 2011;87(5):372–6. [58] Paba P, Bonifacio D, Di Bonito L, Ombres D, Favalli C, Syrjanen K, et al. Coexpression of HSV2 and Chlamydia trachomatis in HPV-positive cervical cancer and cervical intraepithelial neoplasia lesions is associated with aberrations in key intracellular pathways. Intervirology 2008;51(4):230–4. [59] King CC, Jamieson DJ, Wiener J, Cu-Uvin S, Klein RS, Rompalo AM, et al. Bacterial vaginosis and the natural history of human papillomavirus. Infect Dis Obstet Gynecol 2011;2011:319460. [60] Stone KM, Karem KL, Sternberg MR, McQuillan GM, Poon AD, Unger ER, et al. Seroprevalence of human papillomavirus type 16 infection in the United States. J Infect Dis 2002;186(10):1396–402. [61] Herrero R, Hildesheim A, Bratti C, Sherman ME, Hutchinson M, Morales J, et al. Population-based study of human papillomavirus infection and cervical neoplasia in rural Costa Rica. J Natl Cancer Inst 2000;92(6):464–74. [62] Schiffman M, Kjaer SK. Chapter 2: Natural history of anogenital human papillomavirus infection and neoplasia. J Natl Cancer Inst Monogr 2003;31:14–9. [63] Smith EM, Johnson SR, Ritchie JM, Feddersen D, Wang D, Turek LP, et al. Persistent HPV infection in postmenopausal age women. Int J Gynaecol Obstet 2004;87(2):131–7. [64] Rousseau MC, Pereira JS, Prado JC, Villa LL, Rohan TE, Franco EL. Cervical coinfection with human papillomavirus (HPV) types as a predictor of acquisition and persistence of HPV infection. J Infect Dis 2001;184(12):1508–17. [65] Lai CH, Chao A, Chang CJ, Chao FY, Huang HJ, Hsueh S, et al. Host and viral factors in relation to clearance of human papillomavirus infection: a cohort study in Taiwan. Int J Cancer 2008;123(7):1685–92. [66] Hariri S, Dunne EF, Sternberg M, Unger ER, Meadows KS, Karem KL, et al. Seroepidemiology of human papillomavirus type 11 in the United States: results from the third National Health and Nutrition Examination Survey, 1991–1994. Sex Transm Dis 2008;35(3):298–303. [67] Ho GY, Studentsov Y, Hall CB, Bierman R, Beardsley L, Lempa M, et al. Risk factors for subsequent cervicovaginal human papillomavirus (HPV) infection and the protective role of antibodies to HPV-16 virus-like particles. J Infect Dis 2002;186(6):737–42. [68] de Villiers EM. Relationship between steroid hormone contraceptives and HPV, cervical intraepithelial neoplasia and cervical carcinoma. Int J Cancer 2003;103(6):705–8. [69] Moreno V, Bosch FX, Munoz N, Meijer CJ, Shah KV, Walboomers JM, et al. Effect of oral contraceptives on risk of cervical cancer in women with human papillomavirus infection: the IARC multicentric case-control study. Lancet 2002;359(9312):1085–92. [70] Koutsky LA, Ault KA, Wheeler CM, Brown DR, Barr E, Alvarez FB, et al. A controlled trial of a human papillomavirus type 16 vaccine. N Engl J Med 2002;347(21):1645–51. [71] Villa LL, Costa RL, Petta CA, Andrade RP, Ault KA, Giuliano AR, et al. Prophylactic quadrivalent human papillomavirus (types 6, 11, 16, and 18) L1 virus-like particle vaccine in young women: a randomised double-blind placebo-controlled multicentre phase II efficacy trial. Lancet Oncol 2005;6(5):271–8. [72] Romeo J, Warnberg J, Nova E, Diaz LE, Gonzalez-Gross M, Marcos A. Changes in the immune system after moderate beer consumption. Ann Nutr Metab 2007;51(4):359–66. [73] Cook RT, Waldschmidt TJ, Cook BL, Labrecque DR, McLatchie K. Loss of the CD5+ and CD45RAhi B cell subsets in alcoholics. Clin Exp Immunol 1996;103(2):304–10. [74] Cook RT. Alcohol abuse, alcoholism, and damage to the immune system – a review. Alcohol Clin Exp Res 1998;22(9):1927–42. [75] Liu H, ed. Cochran-Armitage trend test using SAS. SAS conference proceedings: PharmaSUG 2007, 2007.