Our study revealed that 18.2% of the sexually active females with normal cervical cytology were infected by at least one HPV genotype; this result is consistent with data obtained by other authors in women without lesions [26, 28]. In 2007, de Sanjosé et al. performed a meta-analysis on worldwide HPV prevalence and genotype distribution in women with normal cytology. On examining the data on young (< 25 years old) European women, they found a prevalence of about 22%, which was similar to that reported by Bruni et al. in the same age-group of young European women [26, 28].
A prevalence study on HPV infection in women aged 18–24 years was carried out in 2006–2007 in Italy. The authors found a prevalence of HPV of about 30% and, on considering only the high-risk genotypes, a prevalence of approximately 20%. These values are higher than ours. However, their study population included women with normal cytology and women with cytological lesions. On considering only the data from the women with negative cytology, the authors found a prevalence of about 20%, which is comparable to our findings .
We recorded a prevalence of high-risk HPV genotypes of 10.1%, while Clifford et al. found a prevalence of high-risk HPV genotypes of 5.2% in Europe. However, these authors considered a very large age-range, namely 15–74 years .
Regarding the relative importance of different high-risk HPV types, the 6 most important in our study were: HPV-16, -52, -56, -31, -51, -18. Our results are similar, but not identical, to those reported by other authors. In Europe in 2005, Clifford et al. reported the following high-risk HPV types: HPV-16, -31, -18, -56, -45 and −35 ; Bruni et al. reported the types HPV-16, -31, -18, -33, -52 and -51 , and de Sanjosé, who considered the 5 most important types, reported HPV-16, -31, -18, -33 and -35 . Our results seem to confirm the growing importance of type 52 in Italy and Europe, as in other parts of the world (North America, Africa and Asia) [6, 26, 28]. Mollers et al., who assessed the prevalence of genital HPV infections in a large cohort of sexually active young women (16–29 years old) in the Netherlands, found that the most prevalent high-risk HPV types were HPV-16, -51 and -52. However, as these authors did not perform cytological analyses, their study population included healthy women and women with lesions .
Another study that provided robust baseline estimates of the prevalence and distribution of HPV types among a nationally representative sample of women (25–64 years old) in England, prior to any impact of the National HPV Immunisation Programme, indicated that the most frequent HPV genotypes among subjects with normal cytology were: HPV-16, -52, -31 and -45 . These data were confirmed by subsequent findings obtained by the same authors in young females up to 24 years old . Furthermore, in a recent Dutch study, HPV types 16, 52, 51 and 31 had the highest type-specific incidence rates among young females (18–29 years old) in the pre-vaccine era; the researchers did not consider the cytological status of the study sample .
Among the HPV genotypes recently defined as probable/possible high-risk, we most frequently identified type 66, while among low-risk viruses type 40 was most frequently identified.
Regarding multiple infections, which are associated with a higher risk of developing high-degree precancerous lesions, we found a higher prevalence (24.3%) than other authors [44, 45]. For instance, in French women under 25 years of age undergoing routine gynaecological examinations, Monsonego et al. found a prevalence of 9.3%, although these women were not selected through the filter of normal cytological findings . The lower prevalence observed by Monsonego et al. could be explained by the fact that the mean age of the French women was probably higher than that of the women in our sample .
In young women (20–29 years old) randomly selected from a population of Danish women, Nielsen et al. found a prevalence of multiple infections of 10.5% . The difference between our data and those of Nielsen et al. could be explained by the fact that our subjects were younger than those examined by the Danish authors.
Furthermore, in our study, although the subjects recruited were not distributed homogeneously by age, the highest percentage of multiple infections (11.1%, 95% CI 0–23.0) was seen among 16-17-year-old subjects. This result is consistent with the data from the literature, and could be explained by the sexual habits of younger women and by the fact that they have less immune experience than older women [48, 49].
We found statistically significant differences in the HPV prevalence rates observed at different ages, a particularly high prevalence being noted in the first years of sexual activity. In the subsequent age-classes, the prevalence declined and stabilized. These results are supported by substantial data that indicate that the first HPV infections often occur soon after the first sexual intercourse [50–53]. In our adolescent group (16–17 years old), the prevalence of low-risk HPV infections was higher than that of probable/possible risk and high-risk HPV genotypes. The only multiple infection caused exclusively by HPV low-risk genotypes (HPV-6 and HPV-11 together) was detected in a 16-year-old. We observed a decrease in low-risk HPV infections with age, a finding that has been confirmed by other studies [54–56]. This decrease is consistent with the scant capacity of these genotypes to circumvent human defences. Several studies have reported a lower rate of persistence of infections by low-risk genotypes than of infections by high-risk types, suggesting that high-risk genotypes may more effectively evade the immune response and persist for a longer time in the human body [57–59]. Indeed, Franco et al. found that the mean duration of infections was 8.2 and 13.5 months for non-oncogenic and oncogenic types, respectively . Moreover, in a study conducted on female university students, Richardson et al. found that infection by HPV-16 was the most persistent (mean duration 18.3 months), while HPV-6 displayed the shortest mean duration (< 10 months) . The ability of high-risk HPV types to persist for a longer time in the host could be linked to the increase in the prevalence of high-risk HPV genotypes with age in our study.
Regarding the estimated cumulative proportions of young females without cervical lesions who were free from any HPV infection at each age, it is interesting that 49.8% of subjects were free from any HPV infection at the age of 26 years (Figure 4). Furthermore, if we consider the estimated cumulative proportions of females free from high-risk vaccine HPV infection at each age, 93.3% of 26-year-old women proved to be free from HPV-16 and/or HPV-18 (Figure 5A). Since the currently available HPV vaccines have shown partial cross-protection against high-risk oncogenic types [24, 25], we also calculated the cumulative proportion of women free from infection by apha7/apha9 high-risk HPV species. At the age of 26 years, 92.2% and 84.7% of women, respectively, were also free from these infections (Figure 5B and 5C). In addition, at the age of 26 years, 97.1% of the women studied were free from infection by types 6 and 11 (Figure 5D).
In Italy, HPV vaccination has been offered free of charge to 12-year-old females since 2008, and both the quadrivalent and the bivalent vaccine are available. More recently, some Italian Regions have extended free vaccination to other female cohorts. Although the Italian Ministry of Health initially recommended vaccination for women aged 25–26 years, too , most Italian Regions offer vaccination free of charge only to younger women, with the aim of obtaining the maximum cost/effect benefit of vaccination in terms of public health . The Italian strategy is similar to the strategies adopted by other developed countries and the member states of the European Union, despite obvious differences [30, 64].
In this perspective, our study suggests that vaccinating females without cervical lesions up to 26 years of age appears to be useful both from the point of view of public health and from the point of view of the individual health of the women. Furthermore, studies on HPV vaccines have demonstrated that vaccination is effective, albeit to different degrees, both in naïve females and in women with evidence of prior HPV exposure .
Our considerations regarding the need to extend free vaccination to women up to 26 years of age are supported by epidemiological and economic studies. Epidemiological studies have shown that the risk of HPV infection remains in older age-groups; indeed, all sexually active women are exposed to the risk of infection at any age [27, 65]. A meta-analysis on worldwide HPV prevalence in women with normal cytology has shown that HPV prevalence is high up to 34 years of age, subsequently decreases up to the age of 44 years, and then tends to rise again in older age-groups . Moreover, economic evaluations also support the utility of extending free vaccination to women up to 26 years of age. Indeed, a study conducted in Italy has reported that vaccinating 25-year-old women is cost-effective, and that this strategy, in comparison with screening alone, could avoid 696 cases of cervical cancer, 11,000 cases of CIN1 and 1,500 cases of CIN2/3 . Furthermore, Westra et al. developed a Markov model to estimate the age-specific health benefits and cost-effectiveness of vaccination in the Netherlands; they conclude not only that vaccinating girls against HPV before their sexual debut is a highly effective and cost-effective strategy for the prevention of cervical cancer, but also that vaccination of women up to the age of 25 years is generally cost-effective . Finally, a review that reports the evidence for the dual approach of HPV vaccination and HPV-based cervical screening has reported that HPV vaccination is likely to provide additional protection to women under the age of 30 years .