Human papillomavirus (HPV) is the most common sexually transmitted infection in the United States, causing more than 13 million new infections per year [1,2]. Most infections are asymptomatic and resolve within 1–2 years. However, persistent infection can lead to cancer. Each year in the United States, HPV causes an estimated 37,300 cases of cervical, other anogenital, or oropharyngeal cancer [3].

HPV infection and HPV-associated cancers are vaccine-preventable. Because HPV vaccination is most beneficial before the start of sexual activity, the Advisory Committee on Immunization Practices (ACIP) recommends routine vaccination of all children aged 11–12 years; vaccination can be given starting at age 9 years [4]. Quadrivalent HPV vaccine (4vHPV), which prevents new infections with HPV types 6, 11, 16, and 18, was introduced in the United States in 2006 and was the predominant HPV vaccine administered through 2015 [4]. Since 2016, 9-valent HPV vaccine (9vHPV) has been the only HPV vaccine available in the United States. HPV vaccines have been licensed and introduced in more than 117 countries and most vaccine used has been 4vHPV [5]. Evaluating the long-term antibody response to 4vHPV can inform future recommendations for people who completed the 4vHPV series.

Historically, Alaska Native women had high rates of cervical cancer compared to non-Hispanic White females in the United States [6]. Cervical cancer rates among Alaska Native women have since declined but disparities persist [6]. HPV vaccines are highly effective for preventing new HPV infections and studies suggest long-term persistence of vaccine-induced antibodies [7,8]. However, data are lacking on long-term response to HPV vaccine among Alaska Native children. We previously reported safety and immunogenicity data up to 2 years after the third dose of 4vHPV among Alaska Native children aged 9–14 years [9]. In this analysis, we evaluate the long-term immunogenicity among these same study participants up to 5 years post-vaccination and use modeling to predict antibody levels at 10 years after vaccination.

Of 525 enrolled participants, 477 participants completed the 3-dose series and had at least one follow-up visit. Eight participants subsequently withdrew or received a dose of 9vHPV and were excluded from this analysis. Of the remaining 469 participants, 432 (92%) had antibody data available at 1 month after completing the 3-dose series, 381 (81%) at 1 year post-vaccination, 359 (77%) at 2 years, 367 (78%) at 3 years, and 344 (73%) at 5 years. Overall, 199 (42%) participants had data available at all five timepoints post-vaccination.

Most of the 469 study participants were female (n = 398, 85%) (Table 1). The median age at enrollment was 11 years (range, 9–14 years). All participants had detectable antibodies against the 4vHPV types at all timepoints post-vaccination. Antibody levels peaked by 1 month post-vaccination, declined substantially during the next year, and gradually declined in subsequent years (Fig. 1). GMC at 5 years post-vaccination were significantly lower than those at 1 month, 1 year, and 2 years after the third dose of vaccine but not significantly lower than those at 3 years post-vaccination. For anti-HPV-16, GMC declined from 1,237 IU/ml (95% CI 1,123–1,362) at 1 month post-vaccination to 230 IU/mL (95% CI 204–259) at 1 year, 171 IU/mL (95% CI 153–192) at 2 years, 137 IU/mL (95% CI 121–154) at 3 years, and 110 IU/mL (95% CI 98–124) at 5 years (Table 2). Trends in antibody levels from 1 month to 5 years post-vaccination were similar for the other three 4vHPV types.

At 5 years post-vaccination, GMC decreased with increasing age at first dose of 4vHPV (Fig. 2). For example, for anti-HPV-16 the 5-year GMC for those who received their first dose at 9 years of age was 143 IU/ml (95% CI 141–144) compared to 39 IU/ml (95% CI 37–42) among those vaccinated at 14 years of age. Findings were similar for other types. GMC were higher among females than males for all 4vHPV types; however, these differences were not significant after adjusting for age.

The log-linear piecewise model predicted durable long-term immunogenicity to all 4vHPV types. Based on the current rate of decline, the predicted GMC at 10 years post-vaccination was 9 AU/mL (95% CI 7–13) for anti-HPV-6, 12 AU/mL (95% CI 9–16) for anti-HPV-11, 47 IU/mL (95% CI 35–52) for anti-HPV-16, and 14 IU/mL (95% CI 10–19) for anti-HPV-18.

In this longitudinal study among Alaska Native children who received 4vHPV at age 9–14 years, all participants still had detectable antibodies against all 4vHPV types at 5 years post-vaccination. Anti-HPV antibody GMC were highest at 1 month after completing the 3-dose regimen, declined during subsequent years, and persisted for at least 5 years. At 5 years post-vaccination, antibody levels were higher among children who received 4vHPV at a younger age. These findings are consistent with other long-term HPV immunogenicity studies conducted among children and adults [6,7,12,13].

Similar to a previous study that followed participants up to 10 years post-vaccination, we found that antibody levels at 5 years after vaccination were inversely associated with age at first vaccination [6]. Previously published findings from our study showed that at 1 month after the third dose of 4vHPV, GMC were higher among children who initiated vaccination at age <11 years [9]. Taken together, these findings further support recommendations that HPV vaccine can be given starting at age 9 years [4]. In our study, GMC were higher among females than males at all time points; however, female participants also were younger. Similar to the previously published results from this and other studies, after adjusting for age at initial vaccination, we did not observe significant differences between the GMC of males and females at years 3 and 5 post-vaccination [6,9]. In contrast, a metanalysis of 18 studies that evaluated sex-based differences in antibody response to 4vHPV found that females tended to have higher GMC than males but these differences were only statistically significantly for anti-HPV-6 [14].

Using a piecewise linear model and observed rates of decay from this study through 5 years after vaccination, we estimate that antibody GMC will remain above the threshold of detection for all 4vHPV types at 10 years post-vaccination. Based on a prior study, the rate of decay is expected to decrease between the 5 year and 10 year follow up visits [11]. If that finding holds, our model provides a conservative estimate and GMC at 10 years may be higher than we predicted.

Our study has several limitations. We may be underpowered to detect sex-based differences due to the small sample size among male children. This study evaluates 4vHPV, which is no longer used in the United States; we expect the antibody response against HPV types 6, 11, 16, and 18 to be similar following vaccination with 4vHPV and 9vHPV [15]. Similarly, we evaluated antibody levels following a 3-dose regimen of 4vHPV but expect a similar response and rates of decline following the currently recommended 2-dose regimen for children who receive their first HPV vaccination before age 15 years [16,17]. In addition, a pre-vaccination blood sample was not collected prior to vaccination so it is not known if any participants had pre-existing antibodies. Finally, there is no immunologic correlate of protection for HPV infection, so we cannot determine if the GMC observed at 5 years post-vaccination will protect against persistent HPV infection. However, HPV infection is rare in people who have received HPV vaccine and efficacy and effectiveness studies suggest long-term protection against infection and other outcomes [2,7,18–20].

This is the first long-term study of HPV vaccine immunogenicity among Alaska Native children and confirms that the antibody response is similar to that of children included in pre-licensure trials. At 5 years post-vaccination, all participants still had detectable antibodies against all 4vHPV types and antibody levels were higher among children who received 4vHPV at a younger age. These results support other long-term immunogenicity studies indicating an HPV vaccine booster dose is not needed and reinforces the recommendation that all children, including Alaska Native children, should continue to receive HPV vaccine according to CDC recommendations and that vaccine can be given starting at age 9 years. We plan to continue to follow this cohort to evaluate the long-term antibody response up to 20 years post-vaccination.