Get 20M+ Full-Text Papers For Less Than $1.50/day. Subscribe now for You or Your Team.

Learn More →

HPV co-factors related to the development of cervical cancer: results from a population-based study in Costa Rica

HPV co-factors related to the development of cervical cancer: results from a population-based... British Journal of Cancer (2001) 84(9), 1219–1226 © 2001 Cancer Research Campaign doi: 10.1054/ bjoc.2001.1779, available online at http://www.idealibrary.com on http://www.bjcancer.com HPV co-factors related to the development of cervical cancer: results from a population-based study in Costa Rica 1 2 1 1 2 1,3 4 5 2 A Hildesheim , R Herrero , PE Castle , S Wacholder , MC Bratti , ME Sherman , AT Lorincz , RD Burk , J Morales , 2 6 2 7 2 8 1 AC Rodriguez , K Helgesen , M Alfaro , M Hutchinson , I Balmaceda , M Greenberg and M Schiffman 1 2 Division of Cancer Epidemiology and Genetics, National Cancer Institute, Bethesda, MD; Caja Costarricense de Seguro Social, San Jose, Costa Rica; 3 4 5 Department of Pathology, The Johns Hopkins Medical Institutions, Baltimore, MD; Digene Corporation, Silver Spring, MD; Albert Einstein College of Medicine, 6 7 8 Bronx, NY; Information Management Services, Silver Spring, MD; Womens and Infants’ Hospital, Providence, RI; OMNIA, Blue Bell, PA Summary We examined factors associated with high-grade squamous intraepithelial lesions (HSIL) and cervical cancer among human papillomavirus (HPV)-infected women in a prevalent case–control study conducted within a population-based cohort of 10 077 women in Costa Rica. We compared 146 women with HPV-positive HSIL or cancer (HSIL/CA) against 843 HPV-positive women without evidence of HSIL/CA. Subjects completed a risk factor questionnaire. We evaluated the associations between exposures and HSIL/CA among women positive for any HPV and restricted to those positive for high-risk HPV types. Risk of HSIL/CA increased with increasing number of live births (P = 0.04). Women who smoked 6+ cigarettes/day had a RR for HSIL/CA of 2.7 (95% CI = 1.1–6.7) compared to non-smokers. Current trend use of barrier contraceptives was associated with a reduction in risk of HSIL/CA (RR = 0.39; 95% CI = 0.16–0.96). Sexual behaviour and a self-reported history of sexually transmitted diseases (STDs) other than HPV were not associated with HSIL/CA. Oral contraceptive use was associated with HSIL/CA among women with <3 pregnancies. Effects were similar in analysis restricted to women positive for high-risk HPV types. Among women positive for high-risk HPV types, 44% of HSIL/CA could be attributed to multiparity (³3 pregnancies) and/or smoking. Among HPV-positive women, multiparity and smoking are risk factors for HSIL/CA. Oral contraceptive use may be associated with HSIL/CA in subgroups of women. © 2001 Cancer Research Campaign http://www.bjcancer.com Keywords: cervix; human papillomavirus; cancer; smoking; parity The vast majority of cervical cancer cases are attributable to To address this concern, we conducted a prevalent case–control human papillomavirus (HPV) infection (Bosch et al, 1995; study within a 10 077 woman population-based study in Guana- Schiffman et al, 1996; Nobbenhuis et al, 1999; Herrero et al, caste, Costa Rica. We evaluated factors associated with progression 2000). Our current challenge is to identify factors involved in the of HPV infection and its early cytologic manifestation (LSIL) by rare progression of HPV infection, which is common and usually comparing 146 HPV-positive cases of HSIL/CA against a group of benign, to cervical cancer and its immediate precursor, high-grade 843 HPV-positive cohort members who did not have concurrent squamous intraepithelial lesions (HSIL). HSIL/CA. We chose our control group to include only non-cases Results from studies of HPV ‘co-factors’ suggest that reproduc- (not diagnosed with HSIL) that were truly at risk of becoming a tive factors, contraceptive use, cigarette smoking, and correlates of case (infected with HPV). Although this study design likely atten- sexual behaviour other than HPV infection might be associated uated the associations of cofactors that are associated with the with HSIL and cervical cancer (henceforth referred to as acquisition and possible persistence of infection, it avoided poten- HSIL/CA) (Bosch et al, 1992; Munoz et al, 1993; Becker et al, 1994; tial residual confounding by HPV. Thus, by including only HPV- Moreno et al, 1995; Kjaer et al, 1996; Chaouki et al, 1998; infected women in a strictly population-based study, we were Chichareon et al, 1998; Ho et al, 1998; Kruger-Kjaer et al, 1998; better able to estimate properly the role of possible HPV co- Olsen et al, 1998; Ngelangel et al, 1998; Roteli-Martins et al, 1998). factors. In most of these studies, HPV infection was accounted for through statistical adjustment, i.e. by averaging the impact of other factors in HPV-negative and HPV-positive cases and controls. Given METHODS strong association between HPV and HSIL/CA (relative risks of ³ Cohort base 50), it is unclear whether effects observed for co-factors (relative risks usually < 3) in these previous studies are real or due to A population-based cohort was established in Guanacaste, Costa residual confounding by HPV. Rica in 1993/4 (Herrero et al, 1997, 2000). The study was conducted after approval by the NCI and local institutional review boards, and all participants provided informed consent. Cluster Received 23 November 2000 sampling was utilized to select a representative sample of the adult Revised 19 February 2001 female population of Guanacaste (n = 10 738 eligible women). Accepted 19 February 2001 10 049 women (94% of eligibles) agreed to visit one of our study clinics. Because few women were expected to have Correspondence to: A Hildesheim 1219 1220 A Hildesheim et al screen-detected invasive cancer, we supplemented our sample participants had equivocal lesions (n = 661) or were cytologically by including women from Guanacaste diagnosed with incident normal (n = 7564). cervical cancer in 1993/4. We attempted to recruit all cervical cancer cases from major centres to which Guanacaste residents are HPV DNA testing referred for diagnosis and treatment (Herrero et al, 1997). 28 of 31 additional cervical cancer cases identified in this manner were Cervical cells were tested for HPV DNA using the Hybrid Capture alive and participated. The total number of women in our cohort Tube test (Cox et al, 1995). This assay detects infection by 16 was therefore 10 077. HPV types (6, 11, 16, 18, 31, 33, 35, 39, 42, 43, 44, 45, 51, 52, 56, 58). Testing was performed for 8563 (99.8%) of the 8582 eligible women with a pelvic examination. The PCR-based L1 consensus Data and specimen collection primer HPV test was also performed on a set of 2974 women Participants responded to a risk factor questionnaire that assessed (Hildesheim et al, 1994; Herrero et al, 2000). For the present study, information on socio-demographic characteristics; sexual, repro- we utilized PCR data from 2300 women selected for PCR testing ductive, and birth control practices; cigarette smoking; and sexually due to 1) an abnormal screening test (1702), 2) HPV positivity by transmitted diseases (STDs). Sexually active women underwent a Hybrid Capture (303), or 3) selection as a random sample of the pelvic examination, at which time a Pap smear was prepared, cells remaining cohort (295). PCR results from a group of women (n = were collected for semi-automated ThinPrep cytology (Cytyc 674) selected for testing on the basis of their sexual behaviour Corp, Boxborough, MA), Cervigrams (National Testing were not utilized (except those that were selected as part of the Laboratories, Fenton, MO) were taken, and cervical cells were random sample of the remaining cohort) to avoid biasing our anal- collected for HPV DNA testing (Herrero et al, 1997, 2000). 583 ysis of co-factors. PCR testing detected type-specific infection by virgins, 291 women who refused to have a pelvic exam or for 44 HPV types (2, 6, 11, 13, 16, 18, 26, 31, 32, 33, 34, 35, 39, 40, whom physical problems prevented a pelvic, and 621 women who 42, 45, 51, 52, 53, 54, 55, 56, 57, 58, 59, 61, 62, 64, 66, 67, 68, 69, reported having a hysterectomy were excluded. Thus, 8 582 70, 72, 73, 83, AE2, AE4, AE5, AE6, AE7, AE8, W13B, PAP155) women were considered for the present study (Figure 1). and infection with uncharacterized types (Herrero et al, 2000). Women with cervical abnormalities by visual inspection, Using Hybrid Capture and PCR HPV testing results, women were cytology or cervicography, or who were in a 2% random sample of classified as 1) positive for high-risk HPV types (16, 18, 31, 33, the population were referred to colposcopy, at which time a second 35, 39, 45, 51, 52, 56, 58, 59, 68) if they tested positive for these questionnaire obtained medical history information and informa- types by either method, 2) negative for high-risk HPV types but tion on douching practices. positive for low-risk HPV types (all other HPV types and unchar- Lesions visible at colposcopy were biopsied. Based on review acterized types) if they tested negative for high-risk types by both of cytology, cervigram and histology, each woman was assigned a Hybrid Capture and PCR (when available) and positive for low- diagnosis (Herrero et al, 1997; Schiffman et al, 1999). 40 women risk types by either method, or 3) HPV DNA-negative by Hybrid (including supplemental cases) were diagnosed with cervical Capture or both HPV DNA assays. By these criteria, 760 women cancer (39 histologically confirmed and one surgically evident), were positive for high-risk HPV types, 229 were positive for low- 128 with HSIL (93% histologically confirmed as having cervical risk types only, and 7574 were negative. Among positives, 596 intraepithelial neoplasia 2 or 3 (CIN2/3); the remainder confirmed (60.3%) were positive by Hybrid Capture and PCR, 24 (2.4%) by at least 2 reviewed cytologic diagnoses), and 189 with LSIL were positive by Hybrid Capture and had missing PCR data, 129 (39% histologically confirmed as having cervical intraepithelial (13.0%) were positive by Hybrid Capture but negative by PCR, neoplasia 1 or condylomatous atypia and the remainder confirmed and 240 (24.3%) were negative by Hybrid Capture but positive by as LSIL by at least 2 screening tests). The remaining 8225 PCR, confirming the higher sensitivity of PCR. 10 769 Eligible for Study (*) 10 077 Agreed to Participate (93.6% Participation) 583 Virgins/No Pelvic 8582 Underwent Pelvic 291 Refused Pelvic 621 Hysterectomized 7574 HPV Negative 989 HPV Positive 19 Not tested for HPV 17 HSIL/CA 48 LSIL 146 HSIL/CA 843 LSIL/HPV 5 HSIL/CA 1 LSIL 116 HSIL 30 Cancers 140 LSIL 703 HPV Positive * Includes 10 738 women selected as part of our cohort and 31 supplemental invasive cancer cases (see text) Figure 1 Summary of study subject selection British Journal of Cancer (2001) 84(9), 1219–1226 © 2001 Cancer Research Campaign HPV co-factors for cervical neoplasia 1221 Case–control study subjects Because controls were defined based on prevalent infection, there was concern that risk estimates for cofactors of interest could Cases were HPV-positive women diagnosed with prevalent be biased by the overrepresentation in our control group of women HSIL/CA at entry into our cohort study (n = 146). These cases with either persistent viral infection or women with recently comprised 86.9% (146/168) of all HSIL/CA cases in our cohort. acquired infections. We therefore conducted a parallel analysis Of the 22 potential cases who were excluded, 17 (10.1%) tested comparing cases to all cohort non-case members in our population- negative for HPV and 5 (3.0%) were not tested. HSIL/CA cases based study, statistically adjusting for HPV. Results of this analysis were combined because HSIL is recognized as the immediate were identical to the results presented and therefore are not precursor to cervical cancer. Although cancer cases were older shown. than HSIL cases (median age = 47 years vs 34 years, respectively), comparison of HSIL (n = 116) and cancer (n = 30) cases control- RESULTS ling for age revealed no notable differences between the two groups with respect to the many risk factors examined. HSIL/CA cases were, on average, 4 years older than controls HPV-positive women with and without a cytological diagnosis (median age = 36 for HSIL/CA and 32 for LSIL/HPV). After of LSIL at entry into our cohort were included in a single control controlling for age, no association was seen between level of group (n = 843) since LSIL is understood to be the cytological education and risk of HSIL/CA (data not shown). The Pap smear manifestation of HPV infection (Schiffman and Brinton, 1995). screening behaviour of cases and controls was very similar; 85.6% The 843 controls included 140 HPV DNA-positive women with of cases and 82.8% of controls reported ever having had a smear LSIL and 703 women with HPV DNA positivity only. Those with and no association was observed between the interval since last LSIL comprised 74.1% (140/189) of all LSILs in our cohort. Of smear or the number of smears and HSIL/CA (age-adjusted RR for the 49 potential controls who were excluded, 48 (25.4%) tested 5+ Paps compared to none = 1.1; 95% CI = 0.62–2.1). The age- negative for HPV and one (0.5%) was not tested. Women with adjusted risk of HSIL/CA was 5.2 times higher (95% CI = LSIL were slightly younger than those without (median age = 29 2.8–10.1) in women with high-risk HPV infection than those with years for the LSIL vs 32 years for the HPV positive only group). low-risk types. Only 10 HSIL/CA cases were positive for low-risk After control for age, the only notable differences observed HPV types alone, 5 of whom were positive for uncharacterized between controls with and without LSIL were a higher likelihood types (Herrero et al, 2000). of those with LSIL to test positive for high-risk HPV types (age- Age at first intercourse, interval between menarche and sexual adjusted RR = 2.0; 95% CI = 1.2–3.3) and a marginally greater debut, and total number of sexual partners were not associated likelihood of these women to have had sexual intercourse >1/week with disease (Table 1). There was a suggestion that number of in the past year (age-adjusted RR = 1.5; 95% CI = 1.0–2.3). sexual partners in the preceding year was negatively associated with HSIL/CA (P = 0.14) and that an increased frequency of trend sexual intercourse was associated with HSIL/CA (P = 0.15). trend Statistical analysis Women who reported ever being pregnant were at increased risk of The relative risk (RR), as estimated by the odds ratio, was the disease (RR = 4.5, 95% CI = 1.1–19), and among ever-pregnant measure of association between exposures and disease. 95% confi- women risk increased with increasing number of pregnancies dence intervals (CI) determined the statistical significance of find- (P = 0.10) and with live births (P = 0.04). Similar trends trend trend ings. Logistic regression analysis permitted adjustment for age and were observed with the number of full-term (P = 0.06) and with trend confounding co-factors (Breslow and Day, 1980). Dose–response number of vaginal births (P = 0.05) (data not shown). Having trend relationships were tested for statistical significance by treating had one miscarriage/abortion (RR = 0.71, 95% CI = 0.42–1.2) or 2 the categorical variable as continuous and evaluating whether the or more miscarriages/abortions (RR = 0.60, 95% CI = 0.31–1.2) resultant beta coefficient departed from zero. The fraction of was not significantly associated with risk of HSIL/CA. HSIL/CA among women infected with high-risk HPVs that is No significant associations were observed for age at first preg- attributable to exposure to co-factors was estimated by the method nancy for women who had their first pregnancy between the ages of Benichou (Benichou, 2000). 20–24 (RR = 1.2, 95% CI = 0.76–1.2) and 25+ (RR = 0.77, 95% Analysis was conducted among all HPV-positive subjects and CI = 0.37–1.6) compared to women who had their first pregnancy also restricted to women positive for high-risk HPV types (136 when they were under 20 years of age. No significant association cases and 624 controls). Both sets of results are presented in the was observed with any occurrence of stillbirths (RR = 0.53, 95% tables. HSIL cases comprise women with CIN2 and CIN3. To alle- CI = 0.23–1.2) or tubal/ectopic pregnancies (RR = 1.9, 95% CI = viate concerns that CIN2 cases might be false positive cases (i.e. 0.37–10). No significant association were observed for women LSILs rather than HSILs) or that they may represent a different who had one caesarean section (RR = 1.2; 95% CI = 0.61–2.4) or biological entity, analyses were also repeated after exclusion of 2+ caesarean sections (RR = 0.70; 95% CI = 0.29–1.7). CIN2 cases. While CIN2 cases were younger than CIN3/cancer Overall, no significant association was observed between oral cases (median age = 31 and 38, respectively), results from the contraceptive use and risk of HSIL/CA (Table 2). However, a analysis restricted to cases with CIN3 or cancer yielded similar significant association between oral contraceptives and HSIL/CA results (data not shown). The few differences noted in this was evident among women with <3 pregnancies (P = 0.02; trend restricted analysis are presented in the text. Analyses adjusted for Table 3), an effect that persisted after further adjustment for barrier age (18–24, 25–29, 30–44, 45–64, 65+), HPV type (high-risk vs contraceptive use. Current use of barrier contraceptives (condoms low-risk), number of pregnancies (0/1, 2/3, 4/5, 6+), and number or diaphragms) was associated with a reduced risk of HSIL/CA; of cigarettes smoked per day (0, 1–5, 6+) are presented. Finer further adjustment for oral contraceptive use did not affect the adjustments for age did not significantly alter the estimates. Other pattern observed. Use of intrauterine devices and tubal ligations factors did not confound associations presented. were not related to risk (data not shown). There was a suggestion © 2001 Cancer Research Campaign British Journal of Cancer (2001)84(9), 1219–1226 1222 A Hildesheim et al Table 1 Distribution and risk of HSIL/CA associated with sexual and reproductive practices Analysis restricted to women positive for Overall analysis high risk HPV types a c d Variable % Cases % Controls RR 95% CI RR 95% CI (n = 146) (n = 843) Age at first intercourse <16 26.7 24.3 1.0 1.0 16–19 49.3 52.5 1.1 [0.69–1.7] 1.2 [0.72–1.9] 20+ 24.0 23.2 1.2 [0.69–2.1] 1.2 [0.65–2.1] Years between menarche and sexual debut <2 14.4 14.6 1.0 1.0 2–3 34.9 27.2 1.3 [0.72–2.3] 1.3 [0.69–2.3] 4–5 25.3 25.2 1.3 [0.69–2.4] 1.3 [0.67–2.4] 6+ 25.4 33.0 0.93 [0.50–1.7] 0.86 [0.45–1.6] Number of sexual partners in lifetime 1 34.9 41.5 1.0 1.0 2–3 43.2 41.7 1.2 [0.76–1.8] 1.2 [0.78–1.9] 4+ 21.9 16.8 1.1 [0.67–1.9] 1.0 [0.59–1.8] Number of sexual partners in past year 0 22.6 17.3 1.0 1.0 1 76.0 77.9 0.87 [0.52–1.4] 0.89 [0.53–1.5] 2+ 1.4 4.8 0.23 [0.05–1.1] 0.24 [0.05–1.1] P = 0.14 P = 0.16 trend trend Frequency of sexual intercourse in past year <1/week 21.2 27.0 1.0 1.0 1/week 26.6 22.6 1.5 [0.83–2.8] 1.7 [0.91–3.2] >1 – ²2/week 21.2 24.1 1.3 [0.67–2.5] 1.4 [0.70–2.7] >2–<4/week 20.4 17.7 1.6 [0.81–3.1] 1.3 [0.66–2.7] 4+/week 10.6 8.6 1.8 [0.82–4.1] 2.1 [0.90–4.7] P = 0.15 P = 0.23 trend trend Ever pregnant No 1.4 7.8 1.0 1.0 Yes 98.6 92.2 4.5 [1.1–19] 4.6 [1.1–20] # Pregnancies 1 9.7 19.4 1.0 1.0 2 11.1 20.1 0.97 [0.44–2.1] 0.86 [0.38–1.9] 3 16.7 14.9 1.7 [0.78–3.6] 1.4 [0.65–3.2] 4–5 25.7 17.1 2.3 [1.1–4.8] 2.3 [1.1–4.9] 6–8 22.9 13.1 2.7 [1.2–5.9] 2.5 [1.1–5.6] 9+ 13.9 15.3 1.1 [0.47–2.7] 1.1 [0.43–2.6] P =0.10 P =0.12 trend trend # Live births 0–1 11.1 23.3 1.0 1.0 2 11.8 20.1 1.2 [0.56–2.5] 1.0 [0.48–2.2] 3 18.7 16.7 1.7 [0.86–3.6] 1.5 [0.73–3.2] 4–5 29.9 15.4 3.7 [1.8–7.4] 3.5 [1.7–7.2] 6–8 16.0 12.9 2.2 [1.0–5.0] 2.2 [0.98–5.0] 9+ 12.5 11.6 1.5 [0.61–3.5] 1.4 [0.56–3.4] P =0.04 P =0.04 trend trend 1 control missing age at 1st intercourse information. 3 controls missing information on years between menarche and sexual debut. 1 control missing information on number of partners in the past year. 1 control missing information on frequency of sexual intercourse in past year. High-risk HPV types include c d types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 68. RR adjusted for age, HPV type, number of pregnancies, and number of cigarettes/day. RR adjusted for e f age, number of pregnancies, and number of cigarettes/day. Restricted to those who reported being sexually active in past year. Logistic models do not include number of pregnancies. that use of injectable contraceptives (mainly Depo Provera) for ³1 of cigarettes smoked per day (P = 0.0007). No association was trend year was associated with HSIL/CA (overall RR = 4.2; 95% CI = observed between the smoking habits of the husbands/live-in part- 0.96–19; RR in analysis restricted to high-risk HPV positives = ners of our study participants and risk of HSIL/CA, in analysis 6.2; 95% CI = 1.1–34) in a small number of women (3 cases, 8 restricted to non-smoking women (data not shown). controls), consistent with previous findings (Herrero et al, 1990). No significant associations with HSIL/CA were observed for Current IUD users had a RR of 2.9 relative to non-users in analysis self-reported yeast infections, any of the sexually transmitted restricted to cases with CIN3 or cancer (95% CI = 1.3–6.5). diseases (STDs) other than HPV examined (gonorrhoea, syphilis, A> 2-fold association was observed between cigarette smoking HSV, other STDs), hand and foot warts, cold sores, other medical and HSIL/CA (Table 2) and risk increased with increasing number conditions (tuberculosis, chicken pox, shingles, asthma), and British Journal of Cancer (2001) 84(9), 1219–1226 © 2001 Cancer Research Campaign HPV co-factors for cervical neoplasia 1223 Table 2 Distribution and risk of HSIL/CA associated with birth control practices and cigarette smoking Analysis restricted to women positive for Overall analysis high risk HPV types a c d Variable % Cases % Controls RR 95% CI RR 95% CI (n = 146) (n = 843) Oral contraceptive use Never 36.3 34.8 1.0 1.0 Former use 34.9 40.4 0.90 [0.54–1.5] 0.93 [0.55–1.6] Current use 28.8 24.8 1.6 [0.89–2.9] 1.5 [0.83–2.8] <5 years of use 37.7 44.9 0.99 [0.59–1.7] 0.99 [0.58–1.7] 5+ years of use 26.0 20.3 1.2 [0.71–2.2] 1.3 [0.70–2.3] Barrier contraceptive use Never 64.4 57.5 1.0 1.0 Former use 31.5 32.1 1.0 [0.65–1.5] 0.91 [0.58–1.4] Current use 4.1 10.4 0.39 [0.16–.96] 0.39 [0.16–.97] <5 years of use 25.4 32.1 0.92 [0.57–1.5] 0.85 [0.52–1.4] 5+ years of use 2.3 10.4 0.58 [0.17–2.1] 0.40 [0.09–1.8] P = 0.44 P = 0.22 trend trend Cigarette smoking Never 79.5 90.3 1.0 1.0 Former 8.2 3.7 2.4 [1.2–5.1] 1.7 [0.76–4.0] Current 12.3 6.0 2.3 [1.3–4.3] 2.3 [1.2–4.3] <10 years of use 8.2 4.9 2.6 [1.2–5.3] 2.2 [1.0–4.8] 10 + years of use 12.3 4.8 2.2 [1.2–4.2] 2.0 [1.0–3.8] 1–5 cigarettes/day 15.1 7.5 2.3 [1.3–3.9] 1.8 [.99–3.3] 6+ cigarettes/day 5.4 2.2 2.7 [1.1–6.7] 3.1 [1.2–7.9] P = .0007 P = .003 trend trend 1 control missing information on oral contraceptive use. 1 control was missing information on ever use and currency of use of barrier contraceptives. 16 cases and 119 controls were missing information on duration of barrier contraceptives. High-risk HPV types include types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, c d 59, 68. RR adjusted for age, HPV type, number of pregnancies, and number of cigarettes/day. RR adjusted for age, number of pregnancies, and number of cigarettes/day. RR adjusted for age, HPV types, and number of pregnancies. women infected with high-risk HPV types, 44% of HSIL/CA can be Table 3 RR estimates and 95% confidence intervals for the joint effect of attributed to these 2 factors. We performed our analysis restricted to pregnancy and oral contraceptive use only those women with HPV infection and chose controls from a subset of women with HPV infection and without HSIL/CA i.e., Oral contraceptive use Number of pregnancies (# Cases # Controls) (# Cases/# Controls) LSIL, ASCUS or cytologically normal. We are the first to acknowl- edge that this may not be an ideal control group. Ideally, one would <3 pregnancies 3+Pregnancies have wanted to select as controls women who had been infected (n = 32/372) (n = 114/470) with HPV in the past (at the same time that the cases were infected) Never (53/293) 1.0 4.7 [1.8–12] but who resolved their infection (in contrast to the cases, whose <5 years (55/378) 1.8 [.65–4.9] 3.7 [1.5–9.3] infection progressed). Given the lack of a reliable method for the 5+years (38/171) 3.1 [1.1–9.1] 4.0 [1.5–10] detection of past infection with one of many HPV types, selection of such an ‘ideal’ control group was not possible. In selecting HPV- RRs adjusted for age, HPV type, and number of cigarettes/day. Estimates infected women as controls, we likely biased our control group presented are from analysis including all subjects. Similar estimates were towards women who have newly acquired HPV infections in which obtained in analysis restricted to high-risk HPV positive women. there has been insufficient time for progression to HSIL/CA, or douching practices (vinegar/water and other preparations) (data women who have persistent (not transient) HPV infections. Women not shown). who have newly acquired infections are more likely to have under- We estimated the fraction of HSIL/CA that can be attributable to gone a change in sexual behaviour that has led to infection. It is exposures other than HPV among women who are infected with unclear whether the inclusion of women with newly acquired infec- high-risk HPV types. In those women with a high-risk HPV type tions would attenuate or exaggerate our risk estimates. Persistent infection, 42% of HSIL/CA (95% CI = 21–63) can be attributed to HPV infection, which occurs relatively infrequently, is an important multiparity (having ³3 pregnancies), 10% to cigarette smoking precursor to high-grade lesions. Inclusion of these women as (95% CI = 3–18), and these 2 exposures together account for 44% controls may therefore have attenuated our findings. This bias, of HSIL/CA (95% CI = 23–66) (Table 4). however, could not have explained our positive findings with smoking and reproductive practices. Also reassuring is the fact that in a parallel analysis in which all non-case cohort members were DISCUSSION included as controls and adjusting for HPV was achieved through We observed that reproductive practices and smoking are important statistical means, identical results were obtained. This suggests that risk factors for the development of HSIL/CA; in Guanacaste, among the findings presented herein are robust. © 2001 Cancer Research Campaign British Journal of Cancer (2001)84(9), 1219–1226 1224 A Hildesheim et al Table 4 Proportion of HSIL and cervical cancer cases attributable to multiparity and cigarette smoking, among women exposed to high-risk HPV Exposure % Exposed RR Attributable fraction (95% CI) >3 pregnancies 54.7% 2.2 42% (21–63) Cigarette smoking 10.4% 2.1 10% (3–18) ³ 3 pregnancies and/or cigarette smoking 59.3% 2.2 44% (23–66) RRs adjusted for age. Pregnancy RR additionally adjusted for number of cigarettes/day. Cigarette smoking RR additionally adjusted for number of pregnancies. The major strengths of our study are its population-based design Multiparity was not associated with number of lifetime or recent and high participation rate (>93%), and its ability to adequately sexual partners and thus the association with HSIL/CA was not account for confounding by HPV, a known causative agent in the confounded by these sexual parameters. pathogenesis of HSIL/CA. A further strength of the present study Various mechanisms might explain the association between was our ability to examine associations restricted to HPV types smoking and HSIL/CA. Constituents of cigarettes, including carci- linked with HSIL/CA, providing further assurance that observa- nogens, have been found in the cervical mucus of female smokers tions are not confounded by HPV type. Estimates from our anal- (Prokopczyk et al, 1997). In addition, exposure to tobacco may ysis restricted to HPV-infected women were identical to those in have a detrimental effect on the ability of the host to mount an the population-based analysis, suggesting that HPV was not a effective immune response against viral infections (Johnson et al, significant confounder of the observed associations with high- 1990; Geng et al, 1996). It is important to point out that the preva- grade lesions. Our findings substantiate earlier reports that lence of smoking among women in our study was low and that smoking (Reeves et al, 1987; Bosch et al, 1992; Munoz et al, 1993; among smokers, intensity of use was moderate. In populations Ho et al, 1998; Olsen et al, 1998; Roteli-Martins et al, 1998) and where prevalence and intensity of smoking are higher, the effect of parity (Brinton et al, 1980; Bosch et al, 1992; Munoz et al, 1993; smoking on risk of HSIL/CA might be even more dramatic. Schiffman et al, 1993) were associated with cervical cancer. Results of studies examining the association between oral Finally, since the Pap smear screening behaviour of cases and contraceptives and HSIL/CA have been conflicting (Bosch et al, controls were similar in our rural population, the observed effects 1992; Munoz et al, 1993; Becker et al, 1994; Moreno et al, 1995; cannot be explained by differential detection due to differences in Kjaer et al, 1996; Chaouki et al, 1998; Chichareon et al, 1998; Ho Pap smear screening. et al, 1998; Kruger-Kjaer et al, 1998; Ngelangel et al, 1998; Among HPV-infected women, no residual effect of lifetime Roteli-Martins et al, 1998). We failed to observe an effect of oral number of sexual partners was observed. This finding, along with contraceptives overall, but noted a significant association between the lack of disease association observed for self-reported STDs oral contraceptive use and HSIL/CA among women with <3 preg- other than HPV, suggests that STDs other than HPV are unlikely to nancies, an effect also observed with other subdivisions of preg- be important in the progression of HPV to HSIL/CA. Self- nancy numbers (e.g., <2 pregnancies and <4 pregnancies). While reporting of STDs in this population was lower than the expected one must be cautious when interpreting results from this subgroup prevalence for these infections suggesting that there was signifi- analysis, it is possible that multiparous women who use oral cant underreporting. To better address the role of non-HPV STDs contraceptives are non-compliant or that misclassification of dura- and lower reproductive tract alterations as risk factors for infected tion of use is higher among multiparous women who start and stop women to progress to high-grade lesions, future studies will use use on multiple occasions. If so, the true effect of oral contracep- biological assays, such as DNA and serological assays, to ascer- tives on disease risk might be larger than that observed in our tain exposure to other STDs, bacterial vaginosis and cervical study. inflammation. Thus, the unreliability of self-reported STDs will be It has been postulated that early initiation of sexual activity is a avoided. The tendency for women with multiple recent partners to risk factor for HSIL/CA because of metaplasia resultant from be at a non-significantly reduced risk of HSIL/CA is consistent ectopy at young ages (Harris et al, 1980). The lack of association with the natural history of cervical neoplasia, where detectable between age at first intercourse, age at first pregnancy, or years infections and LSILs occur proximal to exposure (i.e. recent between menarche and sexual debut and HSIL/CA in our study sexual behaviour is closely linked to HPV infection and LSIL) argues against this ‘vulnerability period’ theory. while HSILs do not develop until several years after initial It is unclear why barrier contraception (condoms and/or dia- exposure. phragms) was associated with reduced risk of HSIL/CA among The underlying mechanism for the observed association between already HPV-infected women. It is possible that barrier users vary pregnancies and HSIL/CA is unknown. Our finding that still-births from nonusers with regards to unmeasured lifestyle factors asso- did not increase risk and that caesarean sections did not reduce risk ciated with disease progression, or that barrier contraceptive use argues against the possibility that trauma during delivery explains modulates the degree of exposure to HPV. The protective effect the observed associations. Other possible explanations include observed among barrier contraceptive users might also reflect a influences of endogenous hormones, nutrition and modulation of protection afforded by barrier contraceptives against STDs other the immune response to HPV by pregnancy (Fife et al, 1987; than HPV. It should be noted, however, that no effect of self- Schneider et al, 1987; Pater et al, 1988; Mitrani-Rosenbaum et al, reported STDs on risk of HSIL/CA was observed in our study, 1989; Rando et al, 1989; Smith et al, 1991; Mittal et al, 1993; as discussed above. Replication of this unexpected finding is Arbeit et al, 1996; Bartholomew et al, 1998; Fife et al, 1999). required before it is accepted. British Journal of Cancer (2001) 84(9), 1219–1226 © 2001 Cancer Research Campaign HPV co-factors for cervical neoplasia 1225 We included as controls HPV-infected women with and without particular Fernando Cardenas, Manuel Barrantes, Elmer Perez, cytological evidence of LSIL. This was done because previous Lidia Ana Morera and Iris Ugarte. We also acknowledge the work has shown that LSIL is the cytological manifestation of HPV collaboration of health authorities in Costa Rica for their enthu- infection (Schiffman and Brinton, 1995). The fact that HPV- siastic support of this project and of the outreach workers of the infected women with and without LSIL were similar with respect Ministry of Health of Costa Rica who carried out the population to most factors examined further justifies combining them into a census for their dedication to the health of the people of single group. However, a few differences noted between these 2 Guanacaste. We thank Dr Sophia Wang (NCI) for her critical groups warrant mention. First, compared to HPV-positive women review of this manuscript. Dr Lorincz is scientific director of with no evidence of LSIL, those diagnosed with LSIL were twice Digene and holds Digene stock and stock options. as likely to be infected with high-risk HPV types, as noted by us in a previous publication (Herrero et al, 2000). Second, HPV-positive REFERENCES women with LSIL were younger than those without LSIL suggesting that in older women HPV infections are less likely to Arbeit JM, Howley PM and Hanahan D (1996) Chronic estrogen-induced cervical manifest themselves cytologically. and vaginal squamous carcinogenesis in human papillomavirus type 16 Our study is cross-sectional; replication of our findings in pros- transgenic mice. Proc Natl Acad Sci 93: 2930–2935 Bartholomew JS, Glenville S, Sarkar S, Burt DJ, Stanley MA, Ruiz-Cabello F, pectively followed women is needed. Follow-up of women in our Chengang J, Garrido F and Stern PL (1998) Integration of high risk human Costa Rican cohort should permit such an analysis once sufficient papillomavirus DNA is linked to the down-regulation of class I human follow-up time is accumulated. In the prospective study, we can leucocyte antigens by steroid hormones in cervical tumour cells. Cancer Res use more appropriate controls, women who get an HPV infection 57: 937–942 at the same time as cases but instead of progressing to high-grade Becker TM, Wheeler CM, McGough NS, Stidley CA, Parmenter CA, Dorin MH and Jordan SW (1994) Contraceptive and reproductive risks for cervical dysplasia lesions, they resolved their infection. Another limitation of our in southwestern hispanic and non-hispanic white women. Int J Epidemiol 23: study is the availability of sensitive PCR-based HPV testing for 913–922 only a subset of participants. Some cytologically normal women Benichou GJ (2000) Variance calculations and confidence intervals for estimates of with low-level HPV infections could have been missed by our the attributable risk based on logistic models. Biometrics 46: 991–1003 Bosch FX, Munoz N, De Sanjose S, izarzugaza I, Gili M, Viladiu P, Tormo MJ, testing strategy. If women with low-level infections differ from Moreo P, Ascunce N, Gonzalez LC, Tafur L, Kaldor JM, Guerrero E, those with higher level infections with respect to the risk factors Aristizabal N, Santamaria M, Alonso de Ruiz P and Shah K (1992) Risk factors we examined, our risk estimates could be biased. This bias, for cervical cancer in Colombia and Spain. Int J Cancer 52: 750–758 however, would likely have attenuated differences between our Bosch FX, Manos MM, Munoz N, Sherman ME, Jansen A, Peto J, Schiffman MH, cases and controls and should not explain away any of our find- Moreno V and Shah KV, the IBSCC study group (1995) Prevalence of HPV DNA in cervical cancer: a worldwide perspective. J Natl Cancer Inst 87: ings. Our study had limited power to evaluate CIN2 cases separ- 796–802 ately from CIN3/cancer cases. However, our findings are robust Breslow NE and Day NE (1980) Statistical methods in cancer research: the analysis since exclusion of CIN2 cases did not alter results. Another limita- of case–control studies. IARC Publ tion is the relatively small number of cancer cases, which Brinton LA, Reeves WC, Brenes MM, Herrero R, De Britton RC, Gaitan E, Tenorio F, Garcia M and Rawls WE (1980) Parity as a risk factor for cervical cancer. precluded analyses to distinguish factors associated with HSIL Am J Epidemiol 130: 486–495 from those associated with the transition from HSIL to invasive Chaouki N, Bosch FX, Munoz N, Meijer CJLM, El Gueddari B, El Ghazi A, Deacon cancer. Since HSIL is the immediate precursor of cervical cancer J, Castellsague X and Walboomers J (1998) The viral origin of cervical cancer and since the absolute risk of HSIL progression without treatment in Rabat, Morocco. Int J Cancer 75: 546–554 is high, factors linked to HSIL are also likely to be linked to inva- Chichareon S, Herrero R, Munoz N, Bosch FX, Jacobs MV, Deacon J, Santamaria M, Chongsuvivatwong V., Meijer CJLM and Walboomers JMM (1998) Risk sive cancer. In fact, excluding the cancer cases and restricting our factors for cervical cancer in Thailand: a case-control study. J Natl Cancer Inst analysis to HSIL alone did not alter our risk estimates. However, it 90: 50–57 is likely that our study would have missed factors uniquely asso- Cox JT, Lorincz AT, Schiffman MH, Sherman ME, Kurman RJ, Hussein M and ciated with the progression of HSIL to cervical cancer. Cullen A (1995) Human papillomavirus testing by Hybrid Capture appears to be useful in triaging women with a cytologic diagnosis of atypical squamous In summary, we observed in our population-based study cells of undetermined significance. Am J Obstet Gynecol 172: 946–954 restricted to HPV positive women that reproductive practices and Fife KH, Rogers RE and Zwickl BW (1987) Symptomatic and asymptomatic cigarette smoking are important co-factors for the development cervical infections with human papillomavirus during pregnancy. J Infect Dis of HSIL/CA. Barrier contraceptive use may reduce the risk of 156: 904–911 HSIL/CA while OC use may be associated with an increased risk Fife KH, Katz BP, Brizendine EJ and Brown DR (1999) Cervical human papillomavirus deoxyribonucleic acid persists throughout pregnancy and of disease in a subgroup of women. decreases in the postpartum period. Am J Obstet Gynecol 180: 1110–1114 Geng Y, Savage SM, Razanai-Boroujerdi S and Sopori ML (1996) Effects of nicotine on the immune response. II. Chronic nicotine treatment induces T cell ACKNOWLEDGEMENTS anergy. J Immunol 156: 2384–2390 This project was supported by a series of National Cancer Institute Harris RWC, Brinton LA, Cowdell RH, Skegg DCG, Smith PG, Vessey MP and Doll R (1980) Characteristics of women with dysplasia or carcinoma in situ of the contracts. We wish to acknowledge the collaboration of Julie cervix uteri Br J Cancer 42: 359–369 Buckland (IMS, Rockville, MD) for data management. We thank Herrero R, Brinton LA, Reeves WC, Brenes MM, de Britton RC, Tenorio F and Deidra Kelly (Johns Hopkins University) and Dr Laurie Mango Gaitan E (1990) Injectable contraceptives and risk of invasive cervical cancer: (Neuromedical Systems, NY) for their collaboration in the inter- evidence of an association. Int J Cancer 46: 5–7 Herrero R, Schiffman MH, Bratti C, Hildesheim A, Balmaceda I, Sherman ME, pretation of cytological specimens. Reagents and services were Greenberg M, Cardenas F, Gomez V, Helgesen K, Morales J, Hutchinson M, supplied or discounted by Cytyc Inc (Boxborough, MA), National Mango L, Alfaro M, Potischman NW, Wacholder S, Swanson C and Brinton Testing Laboratories (Fenton, MO), Utah Medical (Midvale, UT), LA (1997) Design and methods of a population-based natural history study of and Neuromedical Systems (Suffern, NY). We offer special recog- cervical neoplasia in a rural province of Costa Rica: the Guanacaste project. nition for the excellent work of the study staff in Costa Rica, in Pan Am J Public Health 1: 362–374 © 2001 Cancer Research Campaign British Journal of Cancer (2001)84(9), 1219–1226 1226 A Hildesheim et al Herrero R, Hildesheim A, Bratti C, Sherman ME, Hutchinson M, Morales J, (1999) Relation of human papillomavirus status to cervical lesions and Balmaceda I, Greenberg MD, Alfaro M, Burk RD, Wacholder S and Schiffman consequences for cervical-cancer screening: a prospective study. The Lancet M (2000) A population-based study of human papillomavirus infection and all 354: 20–25 grades of cervical neoplasia in rural Costa Rica. J Natl Cancer Inst 92: Olsen AO, Dillner J, Skrondal A and Magnus P (1998) Combined effect of smoking 464–474 and human papillomavirus type 16 infection in cervical carcinogenesis. Hildesheim A, Schiffman MH, Gravitt P, Glass AG, Greer C, Zhang T, Scott DR, Epidemiology 9: 346–349 Rush BB, Lawler P, Sherman ME, Kurman RJ and Manos MM (1994) Pater MM, Hughes GA, Hyslop DE, Nakshatri H and Pater A (1988) Persistence of type-specific human papillomavirus infection among Glucocorticoid-dependent oncogenic transformation by type 16 but not type 11 cytologically normal women in Portland, Oregon. J Infect Dis 169: 235–240 human papilloma virus DNA. Nature 27: 832–835 Ho GY, Kadish AS, Burk RD, Basu J, Palan PR, Mikhail MS and Romney SL Prokopczyk B, Cox JE, Hoffmann D and Waggoner SE (1997) Identification of (1998) HPV 16 and cigarette smoking as risk factors for high-grade cervical tobacco-specific carcinogen in the cervical mucus of smokers and nonsmokers. intra-epithelial neoplasia. Int J Cancer 78: 281–285 J Natl Cancer Inst 89: 868–873 Johnson JD, Houchens DP, Kluwe WM, Craig DK and Fisher GL (1990) Effects of Rando RF, Lindheim S, Hasty L, Sedlacek TV, Woodland M and Eder C (1989) mainstream and environmental tobacco smoke on the immune system in Increased frequency of detection of human papillomavirus deoxtribonucleic animals and humans: a review. Toxicology 20: 369–395 acid in exfoliated cervical cells during pregnancy. Am J Obstet Gynecol 161: Kjaer SK, van den Brule AJC, Bock JE, Poll PA, Engholm G, Sherman ME, 50–55 Walboomers JMM and Meijer CJLM (1996) Human papillomavirus-The most Reeves WC, Caussy D, Brinton LA, Brenes MM, Montalvan P, Gomez B, De significant risk determinant of cervical intraepithelial neoplasia. Int J Cancer Brinton RC, Morice E, Gaitan E, De Lao SL and Rawls WE (1987) Case- 65: 601–606 control study of human papillomaviruses and cervical cancer in Latin America. Kruger-Kjaer S, Van den Brande JL, Svare E, Engholm G, Sherman ME, Poll PA, Int J Cancer 40: 450–454 Walboomers JMM, Bock JE and Meijer CJLM (1998) Different risk factor Roteli-Martins CM, Panetta K, Alves VAF, Siqueira SAC, Syrjanen KJ and Derchain patterns for high-grade and low-grade intraepithelial lesions of the cervix SFM (1998) Cigarete smoking and high-risk HPV DNA as predisposing factors among HPV-positive and HPV-negative young women. Int J Cancer 76: for high-grade cervical intraepithelial neoplasia (CIN) in young Brazilian 613–619 women. Acta Obstet Gynecol Scand 77: 678–682 Mitrani-Rosenbaum S, Tsvieli R and Tur-Kaspa R (1989) Oestrogen stimulates Schiffman MH and Brinton LA (1995) The epidemiology of cervical carcinogenesis. differential transcription of human papillomavirus type 16 in SiHa cervical Cancer 76: 1888–1901 carcinoma cells. J gen Virol 70: 2227–2232 Schiffman MH, Bauer HM, Hoover RN, Glass AG, Cadell DM, Rush BB, Mittal R, Tsutsumi K, Pater A and Pater MM (1993) Human papillomavirus type 16 Scott DR, Sherman ME, Kurman RJ, Wacholder S, Stanton CK and Manos expression in cervical keratinocytes: role of progesterone and glucocorticoid MM (1993) Epidemiologic evidence showing that human papillomavirus hormones. Obstet Gynecol 81: 5–12 infection causes most cervical intraepithelial neoplasia. J Natl Cancer Inst 85: Moreno V, Munoz N, Bosch FX, De Sanjose S, Gonzalez LC, Tafur L, Gili M, 958–964 Izarzugaz I, Navarro C, Vergara A, Viladiu P, Ascunce N and Shah F (1995) Schiffman MH, Brinton LA, Devesa SS and Fraumeni JF, Jr. (1996) Cervical cancer. Risk factors for progression of cervical intraepithelial neoplasm grade III to In: Cancer epidemiology and prevention, Schottenfeld D and Fraumeni JF, Jr. invasive cervical cancer. Cancer Epidemiol Biomarkers Prev 4: 459–467 (eds) pp 1090–1116. Oxford Univ Press: New York Munoz N, Bosch FX, deSanjose S, Vergara A, delMora A, Munoz MT, Tafur L, Gili Schiffman M, Herrero R, Hildesheim A, Sherman ME, Bratti MC, Wacholder S, M, izarzugaza I, Viladiu P, Navarro C, Alonso de Ruiz P, Aristizabal N, Alfaro M, Hutchinson M, Morales J, Greenberg MD and Lorincz AT (1999) Santamaria M, Orfila J, Daniel RW, Guerrero E and Shah KV (1993) Risk HPV DNA testing in cervical cancer screening: results from women in a high- factors for cervical intraepithelial neoplasia grade III/carcinoma in situ in Spain risk province of Costa Rica. JAMA 283: 87–93 and Colombia. Cancer Epidemiol Biomarkers Prev 2: 423–431 Schneider A, Hotz M and Gissmann L (1987) Increased prevalence of human Ngelangel C, Munoz N, Bosch FX, Limson GM, Festin MR, Deacon J, Jacobs MV, papillomavirus in the lower genital tract of pregnant women. Int J Cancer 40: Santamaria M, Meijer CJLM and Walboomers JMM (1998) Causes of cervical 198–201 cancer in the Philippines: a case-control study. J Natl Cancer Inst 90: 43–49 Smith EM, Johnson SR, Jiang D, Zaleski S, Lynch CF, Brundage S, Anderson RD Nobbenhuis MAE, Walboomers JMM, Helmerhorst TJM, Rozendaal L, Remmink and Turek LP (1991) The association between pregnancy and human papilloma AJ, Risse EKJ, van der Linden HC, Voorhorst FJ, Kenemans P and Meijer CJL virus prevalence. Cancer Detect Prev 15: 397–402 British Journal of Cancer (2001) 84(9), 1219–1226 © 2001 Cancer Research Campaign http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png British Journal of Cancer Springer Journals

HPV co-factors related to the development of cervical cancer: results from a population-based study in Costa Rica

Loading next page...
 
/lp/springer-journals/hpv-co-factors-related-to-the-development-of-cervical-cancer-results-6GW00b9iK4

References (49)

Publisher
Springer Journals
Copyright
Copyright © 2001 by The Author(s)
Subject
Biomedicine; Biomedicine, general; Cancer Research; Epidemiology; Molecular Medicine; Oncology; Drug Resistance
ISSN
0007-0920
eISSN
1532-1827
DOI
10.1054/bjoc.2001.1779
Publisher site
See Article on Publisher Site

Abstract

British Journal of Cancer (2001) 84(9), 1219–1226 © 2001 Cancer Research Campaign doi: 10.1054/ bjoc.2001.1779, available online at http://www.idealibrary.com on http://www.bjcancer.com HPV co-factors related to the development of cervical cancer: results from a population-based study in Costa Rica 1 2 1 1 2 1,3 4 5 2 A Hildesheim , R Herrero , PE Castle , S Wacholder , MC Bratti , ME Sherman , AT Lorincz , RD Burk , J Morales , 2 6 2 7 2 8 1 AC Rodriguez , K Helgesen , M Alfaro , M Hutchinson , I Balmaceda , M Greenberg and M Schiffman 1 2 Division of Cancer Epidemiology and Genetics, National Cancer Institute, Bethesda, MD; Caja Costarricense de Seguro Social, San Jose, Costa Rica; 3 4 5 Department of Pathology, The Johns Hopkins Medical Institutions, Baltimore, MD; Digene Corporation, Silver Spring, MD; Albert Einstein College of Medicine, 6 7 8 Bronx, NY; Information Management Services, Silver Spring, MD; Womens and Infants’ Hospital, Providence, RI; OMNIA, Blue Bell, PA Summary We examined factors associated with high-grade squamous intraepithelial lesions (HSIL) and cervical cancer among human papillomavirus (HPV)-infected women in a prevalent case–control study conducted within a population-based cohort of 10 077 women in Costa Rica. We compared 146 women with HPV-positive HSIL or cancer (HSIL/CA) against 843 HPV-positive women without evidence of HSIL/CA. Subjects completed a risk factor questionnaire. We evaluated the associations between exposures and HSIL/CA among women positive for any HPV and restricted to those positive for high-risk HPV types. Risk of HSIL/CA increased with increasing number of live births (P = 0.04). Women who smoked 6+ cigarettes/day had a RR for HSIL/CA of 2.7 (95% CI = 1.1–6.7) compared to non-smokers. Current trend use of barrier contraceptives was associated with a reduction in risk of HSIL/CA (RR = 0.39; 95% CI = 0.16–0.96). Sexual behaviour and a self-reported history of sexually transmitted diseases (STDs) other than HPV were not associated with HSIL/CA. Oral contraceptive use was associated with HSIL/CA among women with <3 pregnancies. Effects were similar in analysis restricted to women positive for high-risk HPV types. Among women positive for high-risk HPV types, 44% of HSIL/CA could be attributed to multiparity (³3 pregnancies) and/or smoking. Among HPV-positive women, multiparity and smoking are risk factors for HSIL/CA. Oral contraceptive use may be associated with HSIL/CA in subgroups of women. © 2001 Cancer Research Campaign http://www.bjcancer.com Keywords: cervix; human papillomavirus; cancer; smoking; parity The vast majority of cervical cancer cases are attributable to To address this concern, we conducted a prevalent case–control human papillomavirus (HPV) infection (Bosch et al, 1995; study within a 10 077 woman population-based study in Guana- Schiffman et al, 1996; Nobbenhuis et al, 1999; Herrero et al, caste, Costa Rica. We evaluated factors associated with progression 2000). Our current challenge is to identify factors involved in the of HPV infection and its early cytologic manifestation (LSIL) by rare progression of HPV infection, which is common and usually comparing 146 HPV-positive cases of HSIL/CA against a group of benign, to cervical cancer and its immediate precursor, high-grade 843 HPV-positive cohort members who did not have concurrent squamous intraepithelial lesions (HSIL). HSIL/CA. We chose our control group to include only non-cases Results from studies of HPV ‘co-factors’ suggest that reproduc- (not diagnosed with HSIL) that were truly at risk of becoming a tive factors, contraceptive use, cigarette smoking, and correlates of case (infected with HPV). Although this study design likely atten- sexual behaviour other than HPV infection might be associated uated the associations of cofactors that are associated with the with HSIL and cervical cancer (henceforth referred to as acquisition and possible persistence of infection, it avoided poten- HSIL/CA) (Bosch et al, 1992; Munoz et al, 1993; Becker et al, 1994; tial residual confounding by HPV. Thus, by including only HPV- Moreno et al, 1995; Kjaer et al, 1996; Chaouki et al, 1998; infected women in a strictly population-based study, we were Chichareon et al, 1998; Ho et al, 1998; Kruger-Kjaer et al, 1998; better able to estimate properly the role of possible HPV co- Olsen et al, 1998; Ngelangel et al, 1998; Roteli-Martins et al, 1998). factors. In most of these studies, HPV infection was accounted for through statistical adjustment, i.e. by averaging the impact of other factors in HPV-negative and HPV-positive cases and controls. Given METHODS strong association between HPV and HSIL/CA (relative risks of ³ Cohort base 50), it is unclear whether effects observed for co-factors (relative risks usually < 3) in these previous studies are real or due to A population-based cohort was established in Guanacaste, Costa residual confounding by HPV. Rica in 1993/4 (Herrero et al, 1997, 2000). The study was conducted after approval by the NCI and local institutional review boards, and all participants provided informed consent. Cluster Received 23 November 2000 sampling was utilized to select a representative sample of the adult Revised 19 February 2001 female population of Guanacaste (n = 10 738 eligible women). Accepted 19 February 2001 10 049 women (94% of eligibles) agreed to visit one of our study clinics. Because few women were expected to have Correspondence to: A Hildesheim 1219 1220 A Hildesheim et al screen-detected invasive cancer, we supplemented our sample participants had equivocal lesions (n = 661) or were cytologically by including women from Guanacaste diagnosed with incident normal (n = 7564). cervical cancer in 1993/4. We attempted to recruit all cervical cancer cases from major centres to which Guanacaste residents are HPV DNA testing referred for diagnosis and treatment (Herrero et al, 1997). 28 of 31 additional cervical cancer cases identified in this manner were Cervical cells were tested for HPV DNA using the Hybrid Capture alive and participated. The total number of women in our cohort Tube test (Cox et al, 1995). This assay detects infection by 16 was therefore 10 077. HPV types (6, 11, 16, 18, 31, 33, 35, 39, 42, 43, 44, 45, 51, 52, 56, 58). Testing was performed for 8563 (99.8%) of the 8582 eligible women with a pelvic examination. The PCR-based L1 consensus Data and specimen collection primer HPV test was also performed on a set of 2974 women Participants responded to a risk factor questionnaire that assessed (Hildesheim et al, 1994; Herrero et al, 2000). For the present study, information on socio-demographic characteristics; sexual, repro- we utilized PCR data from 2300 women selected for PCR testing ductive, and birth control practices; cigarette smoking; and sexually due to 1) an abnormal screening test (1702), 2) HPV positivity by transmitted diseases (STDs). Sexually active women underwent a Hybrid Capture (303), or 3) selection as a random sample of the pelvic examination, at which time a Pap smear was prepared, cells remaining cohort (295). PCR results from a group of women (n = were collected for semi-automated ThinPrep cytology (Cytyc 674) selected for testing on the basis of their sexual behaviour Corp, Boxborough, MA), Cervigrams (National Testing were not utilized (except those that were selected as part of the Laboratories, Fenton, MO) were taken, and cervical cells were random sample of the remaining cohort) to avoid biasing our anal- collected for HPV DNA testing (Herrero et al, 1997, 2000). 583 ysis of co-factors. PCR testing detected type-specific infection by virgins, 291 women who refused to have a pelvic exam or for 44 HPV types (2, 6, 11, 13, 16, 18, 26, 31, 32, 33, 34, 35, 39, 40, whom physical problems prevented a pelvic, and 621 women who 42, 45, 51, 52, 53, 54, 55, 56, 57, 58, 59, 61, 62, 64, 66, 67, 68, 69, reported having a hysterectomy were excluded. Thus, 8 582 70, 72, 73, 83, AE2, AE4, AE5, AE6, AE7, AE8, W13B, PAP155) women were considered for the present study (Figure 1). and infection with uncharacterized types (Herrero et al, 2000). Women with cervical abnormalities by visual inspection, Using Hybrid Capture and PCR HPV testing results, women were cytology or cervicography, or who were in a 2% random sample of classified as 1) positive for high-risk HPV types (16, 18, 31, 33, the population were referred to colposcopy, at which time a second 35, 39, 45, 51, 52, 56, 58, 59, 68) if they tested positive for these questionnaire obtained medical history information and informa- types by either method, 2) negative for high-risk HPV types but tion on douching practices. positive for low-risk HPV types (all other HPV types and unchar- Lesions visible at colposcopy were biopsied. Based on review acterized types) if they tested negative for high-risk types by both of cytology, cervigram and histology, each woman was assigned a Hybrid Capture and PCR (when available) and positive for low- diagnosis (Herrero et al, 1997; Schiffman et al, 1999). 40 women risk types by either method, or 3) HPV DNA-negative by Hybrid (including supplemental cases) were diagnosed with cervical Capture or both HPV DNA assays. By these criteria, 760 women cancer (39 histologically confirmed and one surgically evident), were positive for high-risk HPV types, 229 were positive for low- 128 with HSIL (93% histologically confirmed as having cervical risk types only, and 7574 were negative. Among positives, 596 intraepithelial neoplasia 2 or 3 (CIN2/3); the remainder confirmed (60.3%) were positive by Hybrid Capture and PCR, 24 (2.4%) by at least 2 reviewed cytologic diagnoses), and 189 with LSIL were positive by Hybrid Capture and had missing PCR data, 129 (39% histologically confirmed as having cervical intraepithelial (13.0%) were positive by Hybrid Capture but negative by PCR, neoplasia 1 or condylomatous atypia and the remainder confirmed and 240 (24.3%) were negative by Hybrid Capture but positive by as LSIL by at least 2 screening tests). The remaining 8225 PCR, confirming the higher sensitivity of PCR. 10 769 Eligible for Study (*) 10 077 Agreed to Participate (93.6% Participation) 583 Virgins/No Pelvic 8582 Underwent Pelvic 291 Refused Pelvic 621 Hysterectomized 7574 HPV Negative 989 HPV Positive 19 Not tested for HPV 17 HSIL/CA 48 LSIL 146 HSIL/CA 843 LSIL/HPV 5 HSIL/CA 1 LSIL 116 HSIL 30 Cancers 140 LSIL 703 HPV Positive * Includes 10 738 women selected as part of our cohort and 31 supplemental invasive cancer cases (see text) Figure 1 Summary of study subject selection British Journal of Cancer (2001) 84(9), 1219–1226 © 2001 Cancer Research Campaign HPV co-factors for cervical neoplasia 1221 Case–control study subjects Because controls were defined based on prevalent infection, there was concern that risk estimates for cofactors of interest could Cases were HPV-positive women diagnosed with prevalent be biased by the overrepresentation in our control group of women HSIL/CA at entry into our cohort study (n = 146). These cases with either persistent viral infection or women with recently comprised 86.9% (146/168) of all HSIL/CA cases in our cohort. acquired infections. We therefore conducted a parallel analysis Of the 22 potential cases who were excluded, 17 (10.1%) tested comparing cases to all cohort non-case members in our population- negative for HPV and 5 (3.0%) were not tested. HSIL/CA cases based study, statistically adjusting for HPV. Results of this analysis were combined because HSIL is recognized as the immediate were identical to the results presented and therefore are not precursor to cervical cancer. Although cancer cases were older shown. than HSIL cases (median age = 47 years vs 34 years, respectively), comparison of HSIL (n = 116) and cancer (n = 30) cases control- RESULTS ling for age revealed no notable differences between the two groups with respect to the many risk factors examined. HSIL/CA cases were, on average, 4 years older than controls HPV-positive women with and without a cytological diagnosis (median age = 36 for HSIL/CA and 32 for LSIL/HPV). After of LSIL at entry into our cohort were included in a single control controlling for age, no association was seen between level of group (n = 843) since LSIL is understood to be the cytological education and risk of HSIL/CA (data not shown). The Pap smear manifestation of HPV infection (Schiffman and Brinton, 1995). screening behaviour of cases and controls was very similar; 85.6% The 843 controls included 140 HPV DNA-positive women with of cases and 82.8% of controls reported ever having had a smear LSIL and 703 women with HPV DNA positivity only. Those with and no association was observed between the interval since last LSIL comprised 74.1% (140/189) of all LSILs in our cohort. Of smear or the number of smears and HSIL/CA (age-adjusted RR for the 49 potential controls who were excluded, 48 (25.4%) tested 5+ Paps compared to none = 1.1; 95% CI = 0.62–2.1). The age- negative for HPV and one (0.5%) was not tested. Women with adjusted risk of HSIL/CA was 5.2 times higher (95% CI = LSIL were slightly younger than those without (median age = 29 2.8–10.1) in women with high-risk HPV infection than those with years for the LSIL vs 32 years for the HPV positive only group). low-risk types. Only 10 HSIL/CA cases were positive for low-risk After control for age, the only notable differences observed HPV types alone, 5 of whom were positive for uncharacterized between controls with and without LSIL were a higher likelihood types (Herrero et al, 2000). of those with LSIL to test positive for high-risk HPV types (age- Age at first intercourse, interval between menarche and sexual adjusted RR = 2.0; 95% CI = 1.2–3.3) and a marginally greater debut, and total number of sexual partners were not associated likelihood of these women to have had sexual intercourse >1/week with disease (Table 1). There was a suggestion that number of in the past year (age-adjusted RR = 1.5; 95% CI = 1.0–2.3). sexual partners in the preceding year was negatively associated with HSIL/CA (P = 0.14) and that an increased frequency of trend sexual intercourse was associated with HSIL/CA (P = 0.15). trend Statistical analysis Women who reported ever being pregnant were at increased risk of The relative risk (RR), as estimated by the odds ratio, was the disease (RR = 4.5, 95% CI = 1.1–19), and among ever-pregnant measure of association between exposures and disease. 95% confi- women risk increased with increasing number of pregnancies dence intervals (CI) determined the statistical significance of find- (P = 0.10) and with live births (P = 0.04). Similar trends trend trend ings. Logistic regression analysis permitted adjustment for age and were observed with the number of full-term (P = 0.06) and with trend confounding co-factors (Breslow and Day, 1980). Dose–response number of vaginal births (P = 0.05) (data not shown). Having trend relationships were tested for statistical significance by treating had one miscarriage/abortion (RR = 0.71, 95% CI = 0.42–1.2) or 2 the categorical variable as continuous and evaluating whether the or more miscarriages/abortions (RR = 0.60, 95% CI = 0.31–1.2) resultant beta coefficient departed from zero. The fraction of was not significantly associated with risk of HSIL/CA. HSIL/CA among women infected with high-risk HPVs that is No significant associations were observed for age at first preg- attributable to exposure to co-factors was estimated by the method nancy for women who had their first pregnancy between the ages of Benichou (Benichou, 2000). 20–24 (RR = 1.2, 95% CI = 0.76–1.2) and 25+ (RR = 0.77, 95% Analysis was conducted among all HPV-positive subjects and CI = 0.37–1.6) compared to women who had their first pregnancy also restricted to women positive for high-risk HPV types (136 when they were under 20 years of age. No significant association cases and 624 controls). Both sets of results are presented in the was observed with any occurrence of stillbirths (RR = 0.53, 95% tables. HSIL cases comprise women with CIN2 and CIN3. To alle- CI = 0.23–1.2) or tubal/ectopic pregnancies (RR = 1.9, 95% CI = viate concerns that CIN2 cases might be false positive cases (i.e. 0.37–10). No significant association were observed for women LSILs rather than HSILs) or that they may represent a different who had one caesarean section (RR = 1.2; 95% CI = 0.61–2.4) or biological entity, analyses were also repeated after exclusion of 2+ caesarean sections (RR = 0.70; 95% CI = 0.29–1.7). CIN2 cases. While CIN2 cases were younger than CIN3/cancer Overall, no significant association was observed between oral cases (median age = 31 and 38, respectively), results from the contraceptive use and risk of HSIL/CA (Table 2). However, a analysis restricted to cases with CIN3 or cancer yielded similar significant association between oral contraceptives and HSIL/CA results (data not shown). The few differences noted in this was evident among women with <3 pregnancies (P = 0.02; trend restricted analysis are presented in the text. Analyses adjusted for Table 3), an effect that persisted after further adjustment for barrier age (18–24, 25–29, 30–44, 45–64, 65+), HPV type (high-risk vs contraceptive use. Current use of barrier contraceptives (condoms low-risk), number of pregnancies (0/1, 2/3, 4/5, 6+), and number or diaphragms) was associated with a reduced risk of HSIL/CA; of cigarettes smoked per day (0, 1–5, 6+) are presented. Finer further adjustment for oral contraceptive use did not affect the adjustments for age did not significantly alter the estimates. Other pattern observed. Use of intrauterine devices and tubal ligations factors did not confound associations presented. were not related to risk (data not shown). There was a suggestion © 2001 Cancer Research Campaign British Journal of Cancer (2001)84(9), 1219–1226 1222 A Hildesheim et al Table 1 Distribution and risk of HSIL/CA associated with sexual and reproductive practices Analysis restricted to women positive for Overall analysis high risk HPV types a c d Variable % Cases % Controls RR 95% CI RR 95% CI (n = 146) (n = 843) Age at first intercourse <16 26.7 24.3 1.0 1.0 16–19 49.3 52.5 1.1 [0.69–1.7] 1.2 [0.72–1.9] 20+ 24.0 23.2 1.2 [0.69–2.1] 1.2 [0.65–2.1] Years between menarche and sexual debut <2 14.4 14.6 1.0 1.0 2–3 34.9 27.2 1.3 [0.72–2.3] 1.3 [0.69–2.3] 4–5 25.3 25.2 1.3 [0.69–2.4] 1.3 [0.67–2.4] 6+ 25.4 33.0 0.93 [0.50–1.7] 0.86 [0.45–1.6] Number of sexual partners in lifetime 1 34.9 41.5 1.0 1.0 2–3 43.2 41.7 1.2 [0.76–1.8] 1.2 [0.78–1.9] 4+ 21.9 16.8 1.1 [0.67–1.9] 1.0 [0.59–1.8] Number of sexual partners in past year 0 22.6 17.3 1.0 1.0 1 76.0 77.9 0.87 [0.52–1.4] 0.89 [0.53–1.5] 2+ 1.4 4.8 0.23 [0.05–1.1] 0.24 [0.05–1.1] P = 0.14 P = 0.16 trend trend Frequency of sexual intercourse in past year <1/week 21.2 27.0 1.0 1.0 1/week 26.6 22.6 1.5 [0.83–2.8] 1.7 [0.91–3.2] >1 – ²2/week 21.2 24.1 1.3 [0.67–2.5] 1.4 [0.70–2.7] >2–<4/week 20.4 17.7 1.6 [0.81–3.1] 1.3 [0.66–2.7] 4+/week 10.6 8.6 1.8 [0.82–4.1] 2.1 [0.90–4.7] P = 0.15 P = 0.23 trend trend Ever pregnant No 1.4 7.8 1.0 1.0 Yes 98.6 92.2 4.5 [1.1–19] 4.6 [1.1–20] # Pregnancies 1 9.7 19.4 1.0 1.0 2 11.1 20.1 0.97 [0.44–2.1] 0.86 [0.38–1.9] 3 16.7 14.9 1.7 [0.78–3.6] 1.4 [0.65–3.2] 4–5 25.7 17.1 2.3 [1.1–4.8] 2.3 [1.1–4.9] 6–8 22.9 13.1 2.7 [1.2–5.9] 2.5 [1.1–5.6] 9+ 13.9 15.3 1.1 [0.47–2.7] 1.1 [0.43–2.6] P =0.10 P =0.12 trend trend # Live births 0–1 11.1 23.3 1.0 1.0 2 11.8 20.1 1.2 [0.56–2.5] 1.0 [0.48–2.2] 3 18.7 16.7 1.7 [0.86–3.6] 1.5 [0.73–3.2] 4–5 29.9 15.4 3.7 [1.8–7.4] 3.5 [1.7–7.2] 6–8 16.0 12.9 2.2 [1.0–5.0] 2.2 [0.98–5.0] 9+ 12.5 11.6 1.5 [0.61–3.5] 1.4 [0.56–3.4] P =0.04 P =0.04 trend trend 1 control missing age at 1st intercourse information. 3 controls missing information on years between menarche and sexual debut. 1 control missing information on number of partners in the past year. 1 control missing information on frequency of sexual intercourse in past year. High-risk HPV types include c d types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 68. RR adjusted for age, HPV type, number of pregnancies, and number of cigarettes/day. RR adjusted for e f age, number of pregnancies, and number of cigarettes/day. Restricted to those who reported being sexually active in past year. Logistic models do not include number of pregnancies. that use of injectable contraceptives (mainly Depo Provera) for ³1 of cigarettes smoked per day (P = 0.0007). No association was trend year was associated with HSIL/CA (overall RR = 4.2; 95% CI = observed between the smoking habits of the husbands/live-in part- 0.96–19; RR in analysis restricted to high-risk HPV positives = ners of our study participants and risk of HSIL/CA, in analysis 6.2; 95% CI = 1.1–34) in a small number of women (3 cases, 8 restricted to non-smoking women (data not shown). controls), consistent with previous findings (Herrero et al, 1990). No significant associations with HSIL/CA were observed for Current IUD users had a RR of 2.9 relative to non-users in analysis self-reported yeast infections, any of the sexually transmitted restricted to cases with CIN3 or cancer (95% CI = 1.3–6.5). diseases (STDs) other than HPV examined (gonorrhoea, syphilis, A> 2-fold association was observed between cigarette smoking HSV, other STDs), hand and foot warts, cold sores, other medical and HSIL/CA (Table 2) and risk increased with increasing number conditions (tuberculosis, chicken pox, shingles, asthma), and British Journal of Cancer (2001) 84(9), 1219–1226 © 2001 Cancer Research Campaign HPV co-factors for cervical neoplasia 1223 Table 2 Distribution and risk of HSIL/CA associated with birth control practices and cigarette smoking Analysis restricted to women positive for Overall analysis high risk HPV types a c d Variable % Cases % Controls RR 95% CI RR 95% CI (n = 146) (n = 843) Oral contraceptive use Never 36.3 34.8 1.0 1.0 Former use 34.9 40.4 0.90 [0.54–1.5] 0.93 [0.55–1.6] Current use 28.8 24.8 1.6 [0.89–2.9] 1.5 [0.83–2.8] <5 years of use 37.7 44.9 0.99 [0.59–1.7] 0.99 [0.58–1.7] 5+ years of use 26.0 20.3 1.2 [0.71–2.2] 1.3 [0.70–2.3] Barrier contraceptive use Never 64.4 57.5 1.0 1.0 Former use 31.5 32.1 1.0 [0.65–1.5] 0.91 [0.58–1.4] Current use 4.1 10.4 0.39 [0.16–.96] 0.39 [0.16–.97] <5 years of use 25.4 32.1 0.92 [0.57–1.5] 0.85 [0.52–1.4] 5+ years of use 2.3 10.4 0.58 [0.17–2.1] 0.40 [0.09–1.8] P = 0.44 P = 0.22 trend trend Cigarette smoking Never 79.5 90.3 1.0 1.0 Former 8.2 3.7 2.4 [1.2–5.1] 1.7 [0.76–4.0] Current 12.3 6.0 2.3 [1.3–4.3] 2.3 [1.2–4.3] <10 years of use 8.2 4.9 2.6 [1.2–5.3] 2.2 [1.0–4.8] 10 + years of use 12.3 4.8 2.2 [1.2–4.2] 2.0 [1.0–3.8] 1–5 cigarettes/day 15.1 7.5 2.3 [1.3–3.9] 1.8 [.99–3.3] 6+ cigarettes/day 5.4 2.2 2.7 [1.1–6.7] 3.1 [1.2–7.9] P = .0007 P = .003 trend trend 1 control missing information on oral contraceptive use. 1 control was missing information on ever use and currency of use of barrier contraceptives. 16 cases and 119 controls were missing information on duration of barrier contraceptives. High-risk HPV types include types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, c d 59, 68. RR adjusted for age, HPV type, number of pregnancies, and number of cigarettes/day. RR adjusted for age, number of pregnancies, and number of cigarettes/day. RR adjusted for age, HPV types, and number of pregnancies. women infected with high-risk HPV types, 44% of HSIL/CA can be Table 3 RR estimates and 95% confidence intervals for the joint effect of attributed to these 2 factors. We performed our analysis restricted to pregnancy and oral contraceptive use only those women with HPV infection and chose controls from a subset of women with HPV infection and without HSIL/CA i.e., Oral contraceptive use Number of pregnancies (# Cases # Controls) (# Cases/# Controls) LSIL, ASCUS or cytologically normal. We are the first to acknowl- edge that this may not be an ideal control group. Ideally, one would <3 pregnancies 3+Pregnancies have wanted to select as controls women who had been infected (n = 32/372) (n = 114/470) with HPV in the past (at the same time that the cases were infected) Never (53/293) 1.0 4.7 [1.8–12] but who resolved their infection (in contrast to the cases, whose <5 years (55/378) 1.8 [.65–4.9] 3.7 [1.5–9.3] infection progressed). Given the lack of a reliable method for the 5+years (38/171) 3.1 [1.1–9.1] 4.0 [1.5–10] detection of past infection with one of many HPV types, selection of such an ‘ideal’ control group was not possible. In selecting HPV- RRs adjusted for age, HPV type, and number of cigarettes/day. Estimates infected women as controls, we likely biased our control group presented are from analysis including all subjects. Similar estimates were towards women who have newly acquired HPV infections in which obtained in analysis restricted to high-risk HPV positive women. there has been insufficient time for progression to HSIL/CA, or douching practices (vinegar/water and other preparations) (data women who have persistent (not transient) HPV infections. Women not shown). who have newly acquired infections are more likely to have under- We estimated the fraction of HSIL/CA that can be attributable to gone a change in sexual behaviour that has led to infection. It is exposures other than HPV among women who are infected with unclear whether the inclusion of women with newly acquired infec- high-risk HPV types. In those women with a high-risk HPV type tions would attenuate or exaggerate our risk estimates. Persistent infection, 42% of HSIL/CA (95% CI = 21–63) can be attributed to HPV infection, which occurs relatively infrequently, is an important multiparity (having ³3 pregnancies), 10% to cigarette smoking precursor to high-grade lesions. Inclusion of these women as (95% CI = 3–18), and these 2 exposures together account for 44% controls may therefore have attenuated our findings. This bias, of HSIL/CA (95% CI = 23–66) (Table 4). however, could not have explained our positive findings with smoking and reproductive practices. Also reassuring is the fact that in a parallel analysis in which all non-case cohort members were DISCUSSION included as controls and adjusting for HPV was achieved through We observed that reproductive practices and smoking are important statistical means, identical results were obtained. This suggests that risk factors for the development of HSIL/CA; in Guanacaste, among the findings presented herein are robust. © 2001 Cancer Research Campaign British Journal of Cancer (2001)84(9), 1219–1226 1224 A Hildesheim et al Table 4 Proportion of HSIL and cervical cancer cases attributable to multiparity and cigarette smoking, among women exposed to high-risk HPV Exposure % Exposed RR Attributable fraction (95% CI) >3 pregnancies 54.7% 2.2 42% (21–63) Cigarette smoking 10.4% 2.1 10% (3–18) ³ 3 pregnancies and/or cigarette smoking 59.3% 2.2 44% (23–66) RRs adjusted for age. Pregnancy RR additionally adjusted for number of cigarettes/day. Cigarette smoking RR additionally adjusted for number of pregnancies. The major strengths of our study are its population-based design Multiparity was not associated with number of lifetime or recent and high participation rate (>93%), and its ability to adequately sexual partners and thus the association with HSIL/CA was not account for confounding by HPV, a known causative agent in the confounded by these sexual parameters. pathogenesis of HSIL/CA. A further strength of the present study Various mechanisms might explain the association between was our ability to examine associations restricted to HPV types smoking and HSIL/CA. Constituents of cigarettes, including carci- linked with HSIL/CA, providing further assurance that observa- nogens, have been found in the cervical mucus of female smokers tions are not confounded by HPV type. Estimates from our anal- (Prokopczyk et al, 1997). In addition, exposure to tobacco may ysis restricted to HPV-infected women were identical to those in have a detrimental effect on the ability of the host to mount an the population-based analysis, suggesting that HPV was not a effective immune response against viral infections (Johnson et al, significant confounder of the observed associations with high- 1990; Geng et al, 1996). It is important to point out that the preva- grade lesions. Our findings substantiate earlier reports that lence of smoking among women in our study was low and that smoking (Reeves et al, 1987; Bosch et al, 1992; Munoz et al, 1993; among smokers, intensity of use was moderate. In populations Ho et al, 1998; Olsen et al, 1998; Roteli-Martins et al, 1998) and where prevalence and intensity of smoking are higher, the effect of parity (Brinton et al, 1980; Bosch et al, 1992; Munoz et al, 1993; smoking on risk of HSIL/CA might be even more dramatic. Schiffman et al, 1993) were associated with cervical cancer. Results of studies examining the association between oral Finally, since the Pap smear screening behaviour of cases and contraceptives and HSIL/CA have been conflicting (Bosch et al, controls were similar in our rural population, the observed effects 1992; Munoz et al, 1993; Becker et al, 1994; Moreno et al, 1995; cannot be explained by differential detection due to differences in Kjaer et al, 1996; Chaouki et al, 1998; Chichareon et al, 1998; Ho Pap smear screening. et al, 1998; Kruger-Kjaer et al, 1998; Ngelangel et al, 1998; Among HPV-infected women, no residual effect of lifetime Roteli-Martins et al, 1998). We failed to observe an effect of oral number of sexual partners was observed. This finding, along with contraceptives overall, but noted a significant association between the lack of disease association observed for self-reported STDs oral contraceptive use and HSIL/CA among women with <3 preg- other than HPV, suggests that STDs other than HPV are unlikely to nancies, an effect also observed with other subdivisions of preg- be important in the progression of HPV to HSIL/CA. Self- nancy numbers (e.g., <2 pregnancies and <4 pregnancies). While reporting of STDs in this population was lower than the expected one must be cautious when interpreting results from this subgroup prevalence for these infections suggesting that there was signifi- analysis, it is possible that multiparous women who use oral cant underreporting. To better address the role of non-HPV STDs contraceptives are non-compliant or that misclassification of dura- and lower reproductive tract alterations as risk factors for infected tion of use is higher among multiparous women who start and stop women to progress to high-grade lesions, future studies will use use on multiple occasions. If so, the true effect of oral contracep- biological assays, such as DNA and serological assays, to ascer- tives on disease risk might be larger than that observed in our tain exposure to other STDs, bacterial vaginosis and cervical study. inflammation. Thus, the unreliability of self-reported STDs will be It has been postulated that early initiation of sexual activity is a avoided. The tendency for women with multiple recent partners to risk factor for HSIL/CA because of metaplasia resultant from be at a non-significantly reduced risk of HSIL/CA is consistent ectopy at young ages (Harris et al, 1980). The lack of association with the natural history of cervical neoplasia, where detectable between age at first intercourse, age at first pregnancy, or years infections and LSILs occur proximal to exposure (i.e. recent between menarche and sexual debut and HSIL/CA in our study sexual behaviour is closely linked to HPV infection and LSIL) argues against this ‘vulnerability period’ theory. while HSILs do not develop until several years after initial It is unclear why barrier contraception (condoms and/or dia- exposure. phragms) was associated with reduced risk of HSIL/CA among The underlying mechanism for the observed association between already HPV-infected women. It is possible that barrier users vary pregnancies and HSIL/CA is unknown. Our finding that still-births from nonusers with regards to unmeasured lifestyle factors asso- did not increase risk and that caesarean sections did not reduce risk ciated with disease progression, or that barrier contraceptive use argues against the possibility that trauma during delivery explains modulates the degree of exposure to HPV. The protective effect the observed associations. Other possible explanations include observed among barrier contraceptive users might also reflect a influences of endogenous hormones, nutrition and modulation of protection afforded by barrier contraceptives against STDs other the immune response to HPV by pregnancy (Fife et al, 1987; than HPV. It should be noted, however, that no effect of self- Schneider et al, 1987; Pater et al, 1988; Mitrani-Rosenbaum et al, reported STDs on risk of HSIL/CA was observed in our study, 1989; Rando et al, 1989; Smith et al, 1991; Mittal et al, 1993; as discussed above. Replication of this unexpected finding is Arbeit et al, 1996; Bartholomew et al, 1998; Fife et al, 1999). required before it is accepted. British Journal of Cancer (2001) 84(9), 1219–1226 © 2001 Cancer Research Campaign HPV co-factors for cervical neoplasia 1225 We included as controls HPV-infected women with and without particular Fernando Cardenas, Manuel Barrantes, Elmer Perez, cytological evidence of LSIL. This was done because previous Lidia Ana Morera and Iris Ugarte. We also acknowledge the work has shown that LSIL is the cytological manifestation of HPV collaboration of health authorities in Costa Rica for their enthu- infection (Schiffman and Brinton, 1995). The fact that HPV- siastic support of this project and of the outreach workers of the infected women with and without LSIL were similar with respect Ministry of Health of Costa Rica who carried out the population to most factors examined further justifies combining them into a census for their dedication to the health of the people of single group. However, a few differences noted between these 2 Guanacaste. We thank Dr Sophia Wang (NCI) for her critical groups warrant mention. First, compared to HPV-positive women review of this manuscript. Dr Lorincz is scientific director of with no evidence of LSIL, those diagnosed with LSIL were twice Digene and holds Digene stock and stock options. as likely to be infected with high-risk HPV types, as noted by us in a previous publication (Herrero et al, 2000). Second, HPV-positive REFERENCES women with LSIL were younger than those without LSIL suggesting that in older women HPV infections are less likely to Arbeit JM, Howley PM and Hanahan D (1996) Chronic estrogen-induced cervical manifest themselves cytologically. and vaginal squamous carcinogenesis in human papillomavirus type 16 Our study is cross-sectional; replication of our findings in pros- transgenic mice. Proc Natl Acad Sci 93: 2930–2935 Bartholomew JS, Glenville S, Sarkar S, Burt DJ, Stanley MA, Ruiz-Cabello F, pectively followed women is needed. Follow-up of women in our Chengang J, Garrido F and Stern PL (1998) Integration of high risk human Costa Rican cohort should permit such an analysis once sufficient papillomavirus DNA is linked to the down-regulation of class I human follow-up time is accumulated. In the prospective study, we can leucocyte antigens by steroid hormones in cervical tumour cells. Cancer Res use more appropriate controls, women who get an HPV infection 57: 937–942 at the same time as cases but instead of progressing to high-grade Becker TM, Wheeler CM, McGough NS, Stidley CA, Parmenter CA, Dorin MH and Jordan SW (1994) Contraceptive and reproductive risks for cervical dysplasia lesions, they resolved their infection. Another limitation of our in southwestern hispanic and non-hispanic white women. Int J Epidemiol 23: study is the availability of sensitive PCR-based HPV testing for 913–922 only a subset of participants. Some cytologically normal women Benichou GJ (2000) Variance calculations and confidence intervals for estimates of with low-level HPV infections could have been missed by our the attributable risk based on logistic models. Biometrics 46: 991–1003 Bosch FX, Munoz N, De Sanjose S, izarzugaza I, Gili M, Viladiu P, Tormo MJ, testing strategy. If women with low-level infections differ from Moreo P, Ascunce N, Gonzalez LC, Tafur L, Kaldor JM, Guerrero E, those with higher level infections with respect to the risk factors Aristizabal N, Santamaria M, Alonso de Ruiz P and Shah K (1992) Risk factors we examined, our risk estimates could be biased. This bias, for cervical cancer in Colombia and Spain. Int J Cancer 52: 750–758 however, would likely have attenuated differences between our Bosch FX, Manos MM, Munoz N, Sherman ME, Jansen A, Peto J, Schiffman MH, cases and controls and should not explain away any of our find- Moreno V and Shah KV, the IBSCC study group (1995) Prevalence of HPV DNA in cervical cancer: a worldwide perspective. J Natl Cancer Inst 87: ings. Our study had limited power to evaluate CIN2 cases separ- 796–802 ately from CIN3/cancer cases. However, our findings are robust Breslow NE and Day NE (1980) Statistical methods in cancer research: the analysis since exclusion of CIN2 cases did not alter results. Another limita- of case–control studies. IARC Publ tion is the relatively small number of cancer cases, which Brinton LA, Reeves WC, Brenes MM, Herrero R, De Britton RC, Gaitan E, Tenorio F, Garcia M and Rawls WE (1980) Parity as a risk factor for cervical cancer. precluded analyses to distinguish factors associated with HSIL Am J Epidemiol 130: 486–495 from those associated with the transition from HSIL to invasive Chaouki N, Bosch FX, Munoz N, Meijer CJLM, El Gueddari B, El Ghazi A, Deacon cancer. Since HSIL is the immediate precursor of cervical cancer J, Castellsague X and Walboomers J (1998) The viral origin of cervical cancer and since the absolute risk of HSIL progression without treatment in Rabat, Morocco. Int J Cancer 75: 546–554 is high, factors linked to HSIL are also likely to be linked to inva- Chichareon S, Herrero R, Munoz N, Bosch FX, Jacobs MV, Deacon J, Santamaria M, Chongsuvivatwong V., Meijer CJLM and Walboomers JMM (1998) Risk sive cancer. In fact, excluding the cancer cases and restricting our factors for cervical cancer in Thailand: a case-control study. J Natl Cancer Inst analysis to HSIL alone did not alter our risk estimates. However, it 90: 50–57 is likely that our study would have missed factors uniquely asso- Cox JT, Lorincz AT, Schiffman MH, Sherman ME, Kurman RJ, Hussein M and ciated with the progression of HSIL to cervical cancer. Cullen A (1995) Human papillomavirus testing by Hybrid Capture appears to be useful in triaging women with a cytologic diagnosis of atypical squamous In summary, we observed in our population-based study cells of undetermined significance. Am J Obstet Gynecol 172: 946–954 restricted to HPV positive women that reproductive practices and Fife KH, Rogers RE and Zwickl BW (1987) Symptomatic and asymptomatic cigarette smoking are important co-factors for the development cervical infections with human papillomavirus during pregnancy. J Infect Dis of HSIL/CA. Barrier contraceptive use may reduce the risk of 156: 904–911 HSIL/CA while OC use may be associated with an increased risk Fife KH, Katz BP, Brizendine EJ and Brown DR (1999) Cervical human papillomavirus deoxyribonucleic acid persists throughout pregnancy and of disease in a subgroup of women. decreases in the postpartum period. Am J Obstet Gynecol 180: 1110–1114 Geng Y, Savage SM, Razanai-Boroujerdi S and Sopori ML (1996) Effects of nicotine on the immune response. II. Chronic nicotine treatment induces T cell ACKNOWLEDGEMENTS anergy. J Immunol 156: 2384–2390 This project was supported by a series of National Cancer Institute Harris RWC, Brinton LA, Cowdell RH, Skegg DCG, Smith PG, Vessey MP and Doll R (1980) Characteristics of women with dysplasia or carcinoma in situ of the contracts. We wish to acknowledge the collaboration of Julie cervix uteri Br J Cancer 42: 359–369 Buckland (IMS, Rockville, MD) for data management. We thank Herrero R, Brinton LA, Reeves WC, Brenes MM, de Britton RC, Tenorio F and Deidra Kelly (Johns Hopkins University) and Dr Laurie Mango Gaitan E (1990) Injectable contraceptives and risk of invasive cervical cancer: (Neuromedical Systems, NY) for their collaboration in the inter- evidence of an association. Int J Cancer 46: 5–7 Herrero R, Schiffman MH, Bratti C, Hildesheim A, Balmaceda I, Sherman ME, pretation of cytological specimens. Reagents and services were Greenberg M, Cardenas F, Gomez V, Helgesen K, Morales J, Hutchinson M, supplied or discounted by Cytyc Inc (Boxborough, MA), National Mango L, Alfaro M, Potischman NW, Wacholder S, Swanson C and Brinton Testing Laboratories (Fenton, MO), Utah Medical (Midvale, UT), LA (1997) Design and methods of a population-based natural history study of and Neuromedical Systems (Suffern, NY). We offer special recog- cervical neoplasia in a rural province of Costa Rica: the Guanacaste project. nition for the excellent work of the study staff in Costa Rica, in Pan Am J Public Health 1: 362–374 © 2001 Cancer Research Campaign British Journal of Cancer (2001)84(9), 1219–1226 1226 A Hildesheim et al Herrero R, Hildesheim A, Bratti C, Sherman ME, Hutchinson M, Morales J, (1999) Relation of human papillomavirus status to cervical lesions and Balmaceda I, Greenberg MD, Alfaro M, Burk RD, Wacholder S and Schiffman consequences for cervical-cancer screening: a prospective study. The Lancet M (2000) A population-based study of human papillomavirus infection and all 354: 20–25 grades of cervical neoplasia in rural Costa Rica. J Natl Cancer Inst 92: Olsen AO, Dillner J, Skrondal A and Magnus P (1998) Combined effect of smoking 464–474 and human papillomavirus type 16 infection in cervical carcinogenesis. Hildesheim A, Schiffman MH, Gravitt P, Glass AG, Greer C, Zhang T, Scott DR, Epidemiology 9: 346–349 Rush BB, Lawler P, Sherman ME, Kurman RJ and Manos MM (1994) Pater MM, Hughes GA, Hyslop DE, Nakshatri H and Pater A (1988) Persistence of type-specific human papillomavirus infection among Glucocorticoid-dependent oncogenic transformation by type 16 but not type 11 cytologically normal women in Portland, Oregon. J Infect Dis 169: 235–240 human papilloma virus DNA. Nature 27: 832–835 Ho GY, Kadish AS, Burk RD, Basu J, Palan PR, Mikhail MS and Romney SL Prokopczyk B, Cox JE, Hoffmann D and Waggoner SE (1997) Identification of (1998) HPV 16 and cigarette smoking as risk factors for high-grade cervical tobacco-specific carcinogen in the cervical mucus of smokers and nonsmokers. intra-epithelial neoplasia. Int J Cancer 78: 281–285 J Natl Cancer Inst 89: 868–873 Johnson JD, Houchens DP, Kluwe WM, Craig DK and Fisher GL (1990) Effects of Rando RF, Lindheim S, Hasty L, Sedlacek TV, Woodland M and Eder C (1989) mainstream and environmental tobacco smoke on the immune system in Increased frequency of detection of human papillomavirus deoxtribonucleic animals and humans: a review. Toxicology 20: 369–395 acid in exfoliated cervical cells during pregnancy. Am J Obstet Gynecol 161: Kjaer SK, van den Brule AJC, Bock JE, Poll PA, Engholm G, Sherman ME, 50–55 Walboomers JMM and Meijer CJLM (1996) Human papillomavirus-The most Reeves WC, Caussy D, Brinton LA, Brenes MM, Montalvan P, Gomez B, De significant risk determinant of cervical intraepithelial neoplasia. Int J Cancer Brinton RC, Morice E, Gaitan E, De Lao SL and Rawls WE (1987) Case- 65: 601–606 control study of human papillomaviruses and cervical cancer in Latin America. Kruger-Kjaer S, Van den Brande JL, Svare E, Engholm G, Sherman ME, Poll PA, Int J Cancer 40: 450–454 Walboomers JMM, Bock JE and Meijer CJLM (1998) Different risk factor Roteli-Martins CM, Panetta K, Alves VAF, Siqueira SAC, Syrjanen KJ and Derchain patterns for high-grade and low-grade intraepithelial lesions of the cervix SFM (1998) Cigarete smoking and high-risk HPV DNA as predisposing factors among HPV-positive and HPV-negative young women. Int J Cancer 76: for high-grade cervical intraepithelial neoplasia (CIN) in young Brazilian 613–619 women. Acta Obstet Gynecol Scand 77: 678–682 Mitrani-Rosenbaum S, Tsvieli R and Tur-Kaspa R (1989) Oestrogen stimulates Schiffman MH and Brinton LA (1995) The epidemiology of cervical carcinogenesis. differential transcription of human papillomavirus type 16 in SiHa cervical Cancer 76: 1888–1901 carcinoma cells. J gen Virol 70: 2227–2232 Schiffman MH, Bauer HM, Hoover RN, Glass AG, Cadell DM, Rush BB, Mittal R, Tsutsumi K, Pater A and Pater MM (1993) Human papillomavirus type 16 Scott DR, Sherman ME, Kurman RJ, Wacholder S, Stanton CK and Manos expression in cervical keratinocytes: role of progesterone and glucocorticoid MM (1993) Epidemiologic evidence showing that human papillomavirus hormones. Obstet Gynecol 81: 5–12 infection causes most cervical intraepithelial neoplasia. J Natl Cancer Inst 85: Moreno V, Munoz N, Bosch FX, De Sanjose S, Gonzalez LC, Tafur L, Gili M, 958–964 Izarzugaz I, Navarro C, Vergara A, Viladiu P, Ascunce N and Shah F (1995) Schiffman MH, Brinton LA, Devesa SS and Fraumeni JF, Jr. (1996) Cervical cancer. Risk factors for progression of cervical intraepithelial neoplasm grade III to In: Cancer epidemiology and prevention, Schottenfeld D and Fraumeni JF, Jr. invasive cervical cancer. Cancer Epidemiol Biomarkers Prev 4: 459–467 (eds) pp 1090–1116. Oxford Univ Press: New York Munoz N, Bosch FX, deSanjose S, Vergara A, delMora A, Munoz MT, Tafur L, Gili Schiffman M, Herrero R, Hildesheim A, Sherman ME, Bratti MC, Wacholder S, M, izarzugaza I, Viladiu P, Navarro C, Alonso de Ruiz P, Aristizabal N, Alfaro M, Hutchinson M, Morales J, Greenberg MD and Lorincz AT (1999) Santamaria M, Orfila J, Daniel RW, Guerrero E and Shah KV (1993) Risk HPV DNA testing in cervical cancer screening: results from women in a high- factors for cervical intraepithelial neoplasia grade III/carcinoma in situ in Spain risk province of Costa Rica. JAMA 283: 87–93 and Colombia. Cancer Epidemiol Biomarkers Prev 2: 423–431 Schneider A, Hotz M and Gissmann L (1987) Increased prevalence of human Ngelangel C, Munoz N, Bosch FX, Limson GM, Festin MR, Deacon J, Jacobs MV, papillomavirus in the lower genital tract of pregnant women. Int J Cancer 40: Santamaria M, Meijer CJLM and Walboomers JMM (1998) Causes of cervical 198–201 cancer in the Philippines: a case-control study. J Natl Cancer Inst 90: 43–49 Smith EM, Johnson SR, Jiang D, Zaleski S, Lynch CF, Brundage S, Anderson RD Nobbenhuis MAE, Walboomers JMM, Helmerhorst TJM, Rozendaal L, Remmink and Turek LP (1991) The association between pregnancy and human papilloma AJ, Risse EKJ, van der Linden HC, Voorhorst FJ, Kenemans P and Meijer CJL virus prevalence. Cancer Detect Prev 15: 397–402 British Journal of Cancer (2001) 84(9), 1219–1226 © 2001 Cancer Research Campaign

Journal

British Journal of CancerSpringer Journals

Published: May 1, 2001

There are no references for this article.