Nielsen, Ann MSc*; Kjaer, Susanne Krüger MD*†; Munk, Christian MD*; Iftner, Thomas PhD‡
INFECTION WITH HUMAN PAPILLOMAVIRUS (HPV) is considered a necessary cause of cervical cancer. To date more than 100 HPV types have been identified, of which around 40 are sexually transmitted. Based on molecular, biologic, and epidemiologic evidence, HPV genotypes are categorized as low risk (noncarcinogenic) or high risk (HR) (carcinogenic) due to their ability to cause high-grade cervical intraepithelial neoplasia (CIN3) and invasive cervical cancer.1,2 Most HPV infections are transient and cause no detectable lesions, but some infections become persistent and can progress to precancer and invasive cancer. The virus is considered to be the most common sexually transmitted virus with a lifetime prevalence up to 80%.3
More than 15 HPV types have been categorized as carcinogenic or probably carcinogenic HPV types.4 However, the relative importance of the respective HPV types may vary among different populations. It is important to assess the type-specific distribution of HPV in large population-based studies to yield representative and stable estimates of the prevalence of the different HPV types.
The aim of the current study is to assess the prevalence of HR HPV in different age groups and cytologic subgroups. Furthermore, it is the aim to assess the distribution of the different HR HPV types and finally to study risk factors for prevalent HPV infections and multiple HR HPV infections in 2 randomly sampled cohorts of women from the general female population aged 20–29 years and 40–50 years, respectively.
Materials and Methods
Study Population and Data Collection
In the present cross-sectional study, our study population consisted of baseline data from women in 2 prospective cohort studies conducted simultaneously at the same study clinic. The 2 studies were based on random samples of women 20–29 years of age (younger cohort) and 40–50 years of age (older cohort). All participants signed a written informed consent before inclusion. The studies were approved by the Scientific Ethical Committee of Copenhagen and Frederiksberg Municipality, Denmark.
The enrolment and data collection procedures have been reported previously.5,6 Briefly, in the younger cohort 17,949 women 20–29 years of age were randomly sampled from the general population of Copenhagen, Denmark, by means of the computerized Central Population Register. A total of 1643 women had moved out of the study area before they were contacted. The remaining 16,345 women were invited to participate, and 11,088 women (68%) were included in the study in the period from May 1991 to January 1993. We excluded women for whom an HPV test could not be performed (n = 297) (due to no sample available or sample inadequate for HPV testing), and women who reported to be virgins (n = 247), leaving 10,544 women at baseline in the young cohort for study.
In the older cohort, a total of 2200 women aged 40–50 years were randomly sampled by means of Central Population Register from the same study area as the young cohort,5 and a total of 1580 women were included. In the current study, we excluded 126 women only participating with a telephone interview and 6 women who reported to be virgins. In addition, we excluded 5 women without HPV test results. This left 1443 women with baseline data in the older cohort.
An HPV clinic specifically for the purpose of the 2 cohort studies was established at the Gynecological Department, Rigshospitalet, Copenhagen. All participants in both cohorts went through a personal interview conducted by specially trained female nurses. The interview included questions about sociodemographic variables, contraceptive use, sexual behavior, previous sexually transmitted diseases (STDs), reproductive history, cervical screening history, and smoking habits. Furthermore, the women had a gynecological examination, where material for HPV testing was obtained from the cervix using 2 cotton-tipped plastic-shafted swabs, which were subsequently placed in a tube with TE-buffer (10 mmol/L tris-HCL and 1 mmol/L EDTA, pH = 8.0). Finally, a Pap smear was obtained using a cotton-tipped swab and cytobrush. The HPV tubes were kept at −80°C until tested.
HPV DNA Analysis
Testing for HR HPV DNA has been described previously.5 Briefly, the laboratory personnel and the investigators were fully masked to cytologic and clinical diagnosis at the time for the test for HR HPV DNA. Because the specimens were collected into a medium not originally recommendable for the hybrid capture 2 (HC2) test, a conversion protocol was performed to allow HC2 (HC2; Digene Corporation, Gaithersburg, MD) testing, as described previously.5 The assay was performed with the FDA recommended cutoff of 1.0 pg/mL using only the HR probe that detects at least 13 carcinogenic types (HPV 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 68).7 HPV testing on the older cohort was done as described above except that HPV detection using HC2 with the HR probe was performed manually.5 Replicate assays were performed on a number of plates on both cohorts with virtually identical results.
Genotyping was only performed in HC2 positive samples. Genotyping (LiPA; Innogenetics, Gent, Belgium) was performed from 200 μL of the remaining denatured STM sample of the HC2 test, where total nucleic acid (DNA) was isolated using the MagnaPure device (Roche Systems, Indianapolis, IN). The efficiency of DNA extraction was tested by inclusion of 104 HPV16 positive SiHa cervical carcinoma cells. Five microliters of the DNA solution was used for the LiPaV2 SPF-PCR assay in a final volume of 50 μL using AmpliTaq Gold and 40 cycles of 30 seconds of denaturation at 95°C, followed by 45 seconds of 52°C annealing temperature and 45 seconds of extension at 72°C run on a MJ Thermocycler PCT 200. The PCR product was then denatured and a 10 μL aliquot hybridized to a HPV Genotype detection strip at 49°C for 60 minutes, followed by multiple washing steps. Detection was performed by using the provided substrate solution. The reading of the hybridized strips was performed using a scanner and the LiRAS prototype software (Innogenetics). This software reports the band intensities in graytone values of 0.1–1.0 and allows direct data transfer into Excel spread sheets.
We estimated the overall HPV prevalence with 95% confidence intervals (CIs) for the younger and the older cohort. Additionally, we estimated the age-specific HPV prevalence in 1-year age groups in the younger cohort and in the categories 40–41, 42–43, 44–45, 46–47, and 48–50 years in the older cohort. The type-specific HPV prevalence was assessed overall and according to cytologic status for women with an adequate Pap smear.
Using multivariate logistic regression analyses (SAS STAT version windows 8.0), we identified risk factors associated with prevalent HPV infection in women aged 20–29 and 40–50, respectively. In the younger cohort where we had the greatest statistical power, we also compared women with multiple HR HPV types detected with those who had a single type HPV type. Potential risk factors were selected according to the available literature. The measures of association were expressed as odds ratios (OR) with 95% CIs. Backward stepwise regression at a 5% significance level was used to identify variables included in the multiple regression analysis. Selection was done among variables significant at a 10% level in the age-adjusted model.
The continuous variables age, lifetime number of sexual partners, number of sexually active years, and alcohol intake were initially modeled as linear splines with knots placed according to the quartiles among cases. We tested if the model could be reduced to a model with linear associations. Age, number of sexually active years and, in the older women, number of sexual partners showed no significant deviations from linearity, whereas for alcohol intake and number of sexual partners in the younger women, significant deflection was seen. The association for these variables was modeled as a broken line with a knot at 6 units per week and 14 sexual partners, respectively.
In the younger cohort, we had information on alcohol habits on about 90% of the women. As the results of the statistical analysis with and without alcohol intake were very similar, we only present results based on the nonrestricted data set, i.e., without alcohol intake in the multivariate model.
Prevalence of HR HPV Infection
A total of 17.9% (95% CI, 17.1–18.6) of younger women (20–29 years) and 4.4% (95% CI, 3.3–5.4) of older women (40–50 years) were HR HPV positive (HC2 positive) (Table 1). Among the younger women, 10.5% (95% CI, 9.9–11.1) had multiple types, accounting for 58.6% of the HPV positive participants, whereas less than 1% (0.9%; 95% CI, 0.4–1.4) of the older women had multiple types. The most common HR HPV types detected in the younger cohort were HPV16 (overall prevalence 5.3%; 29.7% of all HR HPV positives), HPV31 (overall prevalence 4.2%; 23.3% of all HR HPV positives), and HPV52 (overall prevalence 3.3%; 18.7% of all HR HPV positives). In the older cohort HPV16 (overall prevalence 0.8%; 19.0% of all HR HPV infections) and HPV52 (overall prevalence 0.8%; 19.0% of all HR HPV positives) were the most common HPV types (Table 1).
The age-specific HR HPV prevalence is shown in Figure 1. Among the younger women, the prevalence decreased from 25.9% (95% CI, 22.8–28.9) in women aged 20 years to 12.4% (95% CI, 10.6–14.3) in women aged 29 years. In the older cohort, we observed a significantly lower age-specific prevalence ranging from 6.1% (95% CI, 6.1–9.1) in women aged 40–41 years to 3.2% (95% CI, 1.1–5.2) in women aged 48–50 years.
In the younger cohort, a total of 10.220 women (96.9%) had normal cytology at enrolment (Table 2). Of those, 15.9% harbored HR HPV types. Atypical squamous cells of undetermined significance or low-grade squamous intraepithelial lesion were found in 244 women (2.3%), with 81.1% being HR HPV positive. Finally, high-grade squamous intraepithelial lesion (HSIL) was diagnosed in 80 women at enrolment, of whom 81.3% were positive for HR HPV. Independent of cytologic category, HPV16 was the most prevalent HPV type, ranging from 4.4% (95% CI, 4.0–4.8) in women with normal cytology to 51.3% (95% CI, 40.3–62.2) in women with HSIL. HPV31 (21.3%; 95% CI, 12.3–30.2), HPV33 (16.3%; 95% CI, 8.2–24.3), and HPV52 (12.5%; 95% CI, 5.3–19.7) were also prevalent in HSIL (Table 2). The same pattern was observed in the older cohort, where only 20 women had abnormal cytology (data not shown). In the older cohort, mainly low-grade lesions were detected and as observed in the younger cohort, HPV16 was the most common HR type, irrespective of cytologic status (data not shown).
Risk Factors for Prevalent HR HPV Infection
Table 3 shows the risk factors for prevalent HR HPV infection in the 2 cohorts. The risk of HR HPV infection decreased significantly with decreasing age in the younger cohort (adj. OR = 0.90; 95% CI, 0.88–0.91), whereas only a slight, nonsignificant decrease was observed in the older cohort. Marital status was a strong determinant for being HR HPV positive in younger women even after adjustment for number of sexual partners. Living alone was significantly associated with risk compared to being married/cohabiting (adj. OR = 1.80; 95% CI, 1.62–2.01).
Being a smoker was not associated with HR HPV infection in younger women. In older women we observed a nonsignificantly increased risk in former and current smokers. Number of cigarettes smoked per day among current smokers was weakly associated with prevalent HR HPV infection among the younger women (OR = 1.03; 95% CI, 1.00–1.06 per 5 cigarettes) (data not shown). In contrast, there was no trend observed among older current smokers (OR = 0.98; 95% CI, 0.86–1.11 per 5 cigarettes) (data not shown). No association was observed with duration of smoking in younger or older women (data not shown).
In the younger cohort, where we also had information on alcohol consumption on about 90% of the women, we found an increasing risk of being HR HPV positive with increasing alcohol intake (data not shown). For women drinking less than 6 units of alcohol per week, the risk increased with 5% per extra unit (CI, 1.02–1.08). For women drinking 6 or more units of alcohol per week, the risk did not further increase (adj. OR = 1.00; CI, 0.99–1.01).
Lifetime number of sexual partners was strongly associated with being HR HPV positive in both the younger and older cohort. Women aged 20–29 years with 14 lifetime sexual partners or less had a significantly increasing risk per extra lifetime sexual partner (adj. OR = 1.12; 95% CI, 1.11–1.14). For women with more than 14 lifetime sexual partners, the risk did not further increase. For women aged 40–50 years, the risk increased with 5% (95% CI, 1.01–1.10) per each extra lifetime sexual partner. We found a slightly although nonsignificantly decreased risk of prevalent HR HPV infection with number of sexually active years in both younger and older women. Age at first sexual intercourse was not associated with risk of HR HPV infection in any of the cohorts (data not shown).
Current oral contraceptive use implied an increased risk of prevalent HR HPV infection in the younger cohort (adj. OR = 1.27; 95% CI, 1.14–1.42), but no association was observed with duration of oral contraceptive use (data not shown). We did not observe any association with current use of condoms either in the younger or in the older cohort (Table 3).
Ever being pregnant and number of pregnancies was associated with a significantly decreased risk of prevalent HR HPV infection among the younger women (adj. OR = 0.67; 95% CI, 0.56–0.79 for ≥2 pregnancies vs. never pregnant). This association was not observed in the older cohort.
We also examined the association of self-reported STDs and the risk of HR HPV infection. Previous genital chlamydia infection (adj. OR = 1.17; 95% CI, 1.03–1.34) and genital warts (adj. OR = 1.26; 95% CI, 1.11–1.44) were associated with an increased risk among the younger women after adjustment for potentially confounding factors, including number of sexual partners (Table 3).
Risk Factors for Multiple HR HPV Types
Table 4 shows risk factors for multiple HR HPV types compared to single type infection in the younger cohort, where we had statistical power to study this issue. The risk of having multiple HR HPV types was not significantly related to age (adj. OR = 0.98; 95% CI, 0.92–1.04). Women living alone had an increased risk of multiple HR HPV types compared to married or cohabiting women (adj. OR = 1.28; 95% CI, 1.05–1.57).
Smoking was not associated with risk of having multiple HR HPV types (data not shown). Regarding alcohol consumption, we found that women drinking less than 6 units of alcohol per week had a nonsignificantly increased risk of 3% per extra unit (CI, 0.98–1.09). The risk did not further increase for women drinking more than 6 units of alcohol per week (adj. OR = 1.00; 95% CI, 0.98–1.02) (data not shown).
Lifetime number of sexual partners was associated with an increased risk of multiple HR HPV types. The risk increased by 5% for each extra sexual partner in women having 14 or less lifetime sexual partners. The risk did not further increase for women with more than 14 lifetime sexual partners. The risk for multiple HR HPV types decreased significantly with years of sexual activity (adj. OR = 0.37; 95% CI, 0.23–0.60 for ≥11 sexually active years vs. ≤5 years).
Current use of oral contraceptives (adj.OR = 1.28; 95% CI, 1.05–1.55) and reporting ever having had a chlamydia infection (adj. OR = 1.31; 95% CI, 1.03–1.66) were associated with a significantly increased risk of multiple HR HPV types.
The current study confirms the highly prevalent nature of HPV infection. The overall prevalence of HR HPV infection among younger women (20–29 years) was 17.9%, whereas it was 4.4% in the older cohort (40–50 years), when using the HC2 method. It is conceivable that the prevalence would have been even higher if all samples had been tested with the more sensitive LiPA test. Nonparticipants in our study were slightly younger than participants, which may as well have lowered the HPV prevalence in our study.
We did not find a bimodal HPV age curve with increased risk in older women, which has been observed in some studies,8–12 but not in others.13–15 In the studies of Herrero et al. and Molano et al., the peak was mostly observed in women older than 65 years and with a predominance of low-risk types.10,12 In our study, the women were 50 years of age at maximum, and we only tested for HR HPV types included in the HC2 probe. Furthermore, the organized screening program in Denmark may have lowered the prevalence in the older cohort through early detection and treatment of abnormal cytology.16
HPV16 was found to be the most common HR type detected in both younger and older women. In addition, we observed that HPV16 was the most common type detected among women with abnormal cytology; it occurred increasingly frequently with increasing severity of the lesion. These findings are in line with those from some previous studies.10,17,18
As in most studies, we found age and lifetime number of sexual partners as the most important risk factors for HR HPV infection, emphasizing once more the importance of sexual transmission. We had information about number of recent sexual partners (i.e., in the last year) in 7786 women aged 20–29 years and in 1439 women from the older cohort. In an attempt to distinguish the effect of lifetime number of sexual partners and recent sexual partners, we performed a stratified analysis and found that among the younger women, recent partners was a strong predictor of HR HPV infection, whereas it was not associated with a significant increase in risk in the older cohort (data not shown). The result that recent sexual partners were of less importance in the older cohort may indicate that HR HPV infections among the older women mainly are due to persistent HR HPV infections, whereas new HR HPV infections are more predominant in the younger cohort. Our results support the findings from a cohort study by Herrero et al.10
Furthermore, we observed a strong association between marital status and HR HPV positivity in women aged 20–29 years, which remained after adjustment for potential confounding factors, including number of sexual partners. This association may be caused by a longer time since last exposure to HPV in married women than in single women. Another explanation could be that single women might have had more casual relationships where the probability of an HPV-infected sexual partner is higher compared to sexual partners in more stable relationships.
We found a slightly increased risk with number of cigarettes smoked per day among young women. Some previous studies have found smoking as a risk factor for HPV infection8,19 but not others.9,11,12,20,21 An effect of smoking, if any, may be caused by an increase in susceptibility to infection or it may be a proxy for unmeasured sexual behavior.22 The same mechanisms may cause the association that we observed with alcohol intake in the younger women. Previous studies conducted in Asia have not observed any association with alcohol intake.19,20 Danish women most likely have a higher and more frequent intake of alcohol than women in Asia, which may explain the differences between our study and previous findings.
Results regarding the association of oral contraceptives use with risk of HPV infection have been inconsistent.8,9,11–15,21,23,24 It has been hypothesized that oral contraceptives use may result in higher susceptibility to HPV infection through for example increased ectopy or may affect sensitivity and specificity of HPV tests.22,25 In the current study, we observed an increased risk among women who currently used oral contraceptives in the young cohort.
Previous investigations have found that number of full-term pregnancies increases the risk of cervical cancer.26 Thus, our finding of a protective effect of pregnancies on risk of HR HPV infection may seem conflicting. However, this has also been observed in other studies.10,20,24 The mechanism is not understood and further research is needed to elucidate the relationship between prevalent HPV infection, pregnancies, and cervical carcinomas.
Both self-reported genital chlamydia infection and genital warts were associated with prevalent HR HPV infection. This may be explained by higher exposure to HPV among women who previously had an STD. Alternatively, the findings may support the hypothesis that previous STDs increase women’s susceptibility to HPV either by cervical inflammation or by microabrasions.22
In the younger cohort, we were able to assess possible determinants for having multiple HR HPV types compared to having a single HR HPV type. In concordance with the findings of Rousseau et al., we found that sexual behavior is of even stronger importance among women with multiple HR HPV types compared to women with a single type HPV infection.27 The increased risk among women with more lifetime sexual partners may partly be explained by a higher exposure to different HPV types. Greater efficiency of the immune response gained by previous exposure to different HPV types may explain why women who have been sexually active for several years are at a lower risk for multiple HPV types than women with less years of sexual activity.
The strength of this study is the high number of women from the general population. Because of the size of the study, we were able to examine the association between several different risk factors and prevalent HR HPV infection with a relatively strong statistical power. However, it should be emphasized that the results in the older cohort are based on a smaller number of women than in the younger cohort.
In conclusion, we observed that HR HPV is highly prevalent among Danish women. We observed a significantly lower HR HPV prevalence in the older cohort. In addition, we found a high prevalence of multiple types in Danish women, notably in the younger cohort. HPV16, which has a strong oncogenic potential, was the most common HR HPV type, both overall and among women with abnormal cytology. We confirmed the impact sexual activity has on risk of prevalent HR HPV infection. We found that lifetime number of sexual partners was associated with increased risk of HR HPV infection in both cohorts, whereas number of recent sexual partners only was associated with HR HPV infection in the younger cohort. In addition, we found by comparing women with multiple HR HPV types with women with only 1 HR HPV type that increasing number of lifetime sexual partners increased the risk of having multiple HR HPV types detected. These data contribute important baseline knowledge of the actual burden of HR HPV before the vaccination against HPV is implemented.
1. Schiffman M, Castle PE. Human papillomavirus: Epidemiology and public health. Arch Pathol Lab Med 2003; 127:930–934.
2. Schiffman M, Kjær SK. Chapter 2: Natural history of anogenital human papillomavirus infection and neoplasia. J Natl Cancer Inst Monogr 2003; 31:14–19.
3. Syrjänen KJ. Natural history of genital human papillomavirus infections. In: Lacey C, ed. Papillomavirus Reviews: Current Research on Papillomaviruses. Leeds, United Kingdom: Leeds University Press, 1996.
4. Munoz N, Bosch FX, de SS, et al. Epidemiologic classification of human papillomavirus types associated with cervical cancer. N Engl J Med 2003; 348:518–527.
5. Kjaer SK, Hogdall E, Frederiksen K, et al. The absolute risk of cervical abnormalities in high-risk human papillomavirus-positive, cytologically normal women over a 10-year period. Cancer Res 2006; 66:10630–10636.
6. Kjaer SK, van den Brule AJ, Bock JE, et al. Human papillomavirus—The most significant risk determinant of cervical intraepithelial neoplasia. Int J Cancer 1996; 65:601–606.
7. Iftner T, Villa LL. Chapter 12: Human papillomavirus technologies. J Natl Cancer Inst Monogr 2003; 31:80–88.
8. Ferreccio C, Prado RB, Luzoro AV, et al. Population-based prevalence and age distribution of human papillomavirus among women in Santiago, Chile. Cancer Epidemiol Biomarkers Prev 2004; 13:2271–2276.
9. de SS, Almirall R, Lloveras B, et al. Cervical human papillomavirus infection in the female population in Barcelona, Spain. Sex Transm Dis 2003; 30:788–793.
10. Herrero R, Castle PE, Schiffman M, et al. Epidemiologic profile of type-specific human papillomavirus infection and cervical neoplasia in Guanacaste, Costa Rica. J Infect Dis 2005; 191:1796–1807.
11. Lazcano-Ponce E, Herrero R, Munoz N, et al. Epidemiology of HPV infection among Mexican women with normal cervical cytology. Int J Cancer 2001; 91:412–420.
12. Molano M, Posso H, Weiderpass E, et al. Prevalence and determinants of HPV infection among Colombian women with normal cytology. Br J Cancer 2002; 87:324–333.
13. Thomas JO, Herrero R, Omigbodun AA, et al. Prevalence of papillomavirus infection in women in Ibadan, Nigeria: A population-based study. Br J Cancer 2004; 90:638–645.
14. Matos E, Loria D, Amestoy GM, et al. Prevalence of human papillomavirus infection among women in Concordia, Argentina: A population-based study. Sex Transm Dis 2003; 30:593–599.
15. Sukvirach S, Smith JS, Tunsakul S, et al. Population-based human papillomavirus prevalence in Lampang and Songkla, Thailand. J Infect Dis 2003; 187:1246–1256.
16. Franceschi S, Herrero R, Clifford GM, et al. Variations in the age-specific curves of human papillomavirus prevalence in women worldwide. Int J Cancer 2006; 119:2677–2684.
17. Clifford G, Franceschi S, Diaz M, et al. Chapter 3: HPV type-distribution in women with and without cervical neoplastic diseases. Vaccine 2006; 24(suppl 3):S26–S34.
18. Kitchener HC, Almonte M, Wheeler P, et al. HPV testing in routine cervical screening: Cross sectional data from the ARTISTIC trial. Br J Cancer 2006; 95:56–61.
19. Chan PK, Chang AR, Cheung JL, et al. Determinants of cervical human papillomavirus infection: Differences between high- and low-oncogenic risk types. J Infect Dis 2002; 185:28–35.
20. Shin HR, Franceschi S, Vaccarella S, et al. Prevalence and determinants of genital infection with papillomavirus, in female and male university students in Busan, South Korea. J Infect Dis 2004; 190:468–476.
21. Tarkowski TA, Koumans EH, Sawyer M, et al. Epidemiology of human papillomavirus infection and abnormal cytologic test results in an urban adolescent population. J Infect Dis 2004; 189:46–50.
22. Burchell AN, Winer RL, de SS, et al. Chapter 6: Epidemiology and transmission dynamics of genital HPV infection. Vaccine 2006;24(suppl 3):S52–S61.
23. Peyton CL, Gravitt PE, Hunt WC, et al. Determinants of genital human papillomavirus detection in a US population. J Infect Dis 2001; 183:1554–1564.
24. Vaccarella S, Herrero R, Dai M, et al. Reproductive factors, oral contraceptive use, and human papillomavirus infection: Pooled analysis of the IARC HPV prevalence surveys. Cancer Epidemiol Biomarkers Prev 2006; 15:2148–2153.
25. Green J, Berrington de GA, Smith JS, et al. Human papillomavirus infection and use of oral contraceptives. Br J Cancer 2003; 88:1713–1720.
26. Cervical carcinoma and reproductive factors: Collaborative reanalysis of individual data on 16,563 women with cervical carcinoma and 33,542 women without cervical carcinoma from 25 epidemiological studies. Int J Cancer 2006; 119:1108–1124.
27. Rousseau MC, Abrahamowicz M, Villa LL, et al. Predictors of cervical coinfection with multiple human papillomavirus types. Cancer Epidemiol Biomarkers Prev 2003; 12:1029–1037.