obÌåÓý

Abstract

In 2006, human papillomavirus (HPV) vaccine was first recommended in the United States to prevent cancers and other diseases caused by HPV; vaccination coverage increased steadily through 2021, and increasing numbers of young women had received HPV vaccine as children or adolescents. Since 2008, obÌåÓý has monitored incidence of precancerous lesions (cervical intraepithelial neoplasia [CIN] grades 2–3 and adenocarcinoma in situ [AIS], collectively CIN2+), which are detected through cervical cancer screening and can be used as an intermediate outcome for monitoring vaccination impact, via the five-site Human Papillomavirus Vaccine Impact Monitoring Project. This analysis describes trends in incidence of CIN2+ and CIN3+ (i.e., CIN grade 3 and AIS) lesions during 2008–2022. Among women aged 20–24 years who were screened for cervical cancer, rates during 2008–2022 decreased for CIN2+ by 79%, and for CIN3+ by 80%. In the same period, CIN3+ rates among screened women aged 25–29 years decreased by 37%. These data are consistent with considerable impact of HPV vaccination for preventing cervical precancers among women in the age groups most likely to have been vaccinated, and support existing recommendations to vaccinate children at the routinely recommended ages as a cancer prevention measure.

Introduction

Human papillomavirus (HPV) causes approximately 10,800 cervical cancers in the United States each year; cervical cancer is the most common HPV-attributable cancer among women (1). HPV-attributable cancers take many years to develop (median age at diagnosis = 50 years),* whereas screen-detected cervical precancers (cervical intraepithelial neoplasia [CIN] grades 2–3 and adenocarcinoma in situ [AIS], collectively CIN2+) can develop within a few years after infection (2). Screening with the Papanicolaou (Pap) test and treatment of precancerous abnormalities have been the mainstay of secondary prevention of cervical cancer for decades; screening with a test for high-risk HPV, usually as a co-test with a Pap test, has been increasingly used during the past decade (2,3). Recommended screening intervals have increased, from annually to every 3 years (cytology only) or every 5 years (incorporating an HPV test). In 2006, HPV vaccine was first recommended in the United States by obÌåÓý’s Advisory Committee on Immunization Practices to prevent cancers and other diseases caused by HPV^† (2). Routine vaccination was first recommended for girls and women; in 2011, boys and men were included in the vaccination program. Currently, routine vaccination is recommended for all children at age 11–12 years (may commence at age 9 years), with catch-up vaccination through age 26 years (2). Since 2019, shared clinical decision-making has been recommended for consideration of vaccination of adults aged 27–45 years. Two doses of HPV vaccine are recommended if the series is started at age <15 years; otherwise, 3 doses are recommended. Coverage with ≥1 HPV vaccine dose among adolescents aged 13–17 years steadily increased through 2021 then plateaued, with most recent coverage of 76.8% in 2023 (4). obÌåÓý has monitored CIN2+ incidence since 2008 via the Human Papillomavirus Vaccine Impact Monitoring Project (HPV-IMPACT) (5,6). This analysis describes trends in CIN2+ and CIN3+ (i.e., CIN grade 3 and AIS) incidence during 2008–2022.

Methods

Surveillance System

HPV-IMPACT has conducted population-based CIN2+ surveillance in five sites since 2008.^§ Participating sites conduct active surveillance of all histopathology laboratories serving catchment area residents to identify histologically confirmed CIN2+ diagnoses.

Populations and Screening Estimation

The number of women aged 20–64 years in the HPV-IMPACT catchment areas was obtained from U.S. Census Bureau data, stratified by 5-year age group.^¶ The proportions of women screened (i.e., who had a Pap or HPV test in the preceding year), by 5-year age group, were estimated** using standardized data sources and methods for all sites^†† (5,6). To account for variations in screening by insurance coverage, estimates are weighted averages of proportions screened among privately insured,^§§ publicly insured,^¶¶ and uninsured women.*** Annual age-specific numbers of women screened were estimated by multiplying age-specific proportions screened by age-specific populations of women.

Statistical Analysis

For the years 2008–2022, annual age-specific CIN2+ and CIN3+ incidence and 95% CIs were calculated using the estimated number of women screened as the denominator to control for changes in screening frequency; incidence is still affected by changes in screening test sensitivity and vaccination coverage. Incidence calculations were performed using SAS (version 9.4; SAS Institute). Age-specific trend analyses were conducted separately for CIN2+ and CIN3+ using Joinpoint software (version 5.3.0.0; National Cancer Institute).^††† Overall trends for 2008–2022 were reported as average annual percent change (AAPC), and segment-specific trends were reported as annual percent change (APC).^§§§ Because health care disruptions due to the COVID-19 pandemic might have resulted in deviations in incidence in 2020 from the underlying trend, data from 2020 were excluded from trend analyses.^¶¶¶ This activity was reviewed by obÌåÓý, deemed not research, and was conducted consistent with applicable federal law and obÌåÓý policy.****

Results

During 2008–2022, a total of 39,977 CIN2+ cases were reported; 13,027 (32.6%) were CIN3+. CIN2+ cases per 100,000 screened women decreased 11.0% annually for women aged 20–24 years; the incidence in 2022 was 79.5% lower than that in 2008 (Table) (Figure) (Supplementary Table, ). Among screened women aged 25–29 years, CIN2+ incidence increased 3.1% annually during 2008–2016, and decreased 4.3% annually during 2016–2022, but AAPC was stable. Among women aged 30–34 years and 35–39 years, CIN2+ incidence trended upward during 2008–2016 and then downward during 2016–2022; AAPCs for the entire period were positive but small. Among those in age groups 40–49 and 50–64 years, CIN2+ incidence increased significantly during 2008–2022.

Trends for CIN3+ were generally similar to those for CIN2+; among women aged 20–24 years, the incidence in 2022 was 80.3% lower than that in 2008. One notable difference for CIN3+ compared with CIN2+ was an overall decreasing trend among screened women aged 25–29 years (AAPC = −3.5%); the CIN3+ incidence in 2022 was 37.2% lower than that in 2008.

Discussion

These data from HPV-IMPACT provide an updated view into the epidemiology of cervical precancers in the United States during an era of increasing HPV vaccination coverage among young women and changing cervical cancer screening practices among all age groups. Among women aged 20–24 years who were screened, CIN2+ incidence decreased 79% from 2008 to 2022, and CIN3+ incidence decreased 80%. Among screened women aged 25–29 years, CIN3+ incidence decreased 37%. The data are consistent with a considerable impact from the U.S. HPV vaccination program on cervical precancers, with the largest decreases in the youngest age group for which benefit of vaccination would first be observed.

Data from HPV-IMPACT previously indicated decreasing CIN2+ and CIN3+ incidence among women aged 18–20 and 21–24 years during 2008–2015 (6). As vaccinated women age into older age groups, declines in cervical precancers are expected. For example, before 2014, women aged 20–24 years could only have been vaccinated in the catch-up vaccination age range (13–26 years), whereas during 2018–2022, all women aged 20–24 years would have been eligible for vaccination at the routine age (11–12 years) in 2006. Vaccination at the routine age is more effective because vaccination is likely to occur before exposure to HPV through sexual contact. The declines in precancers mirror reported U.S. trends in vaccine-type HPV prevalence in self-collected cervicovaginal swabs, in which declines in quadrivalent HPV-type prevalence among adolescents and women aged 14–19 years were followed by declines among women aged 20–24 years (2). This report includes the first U.S. data showing significant decreases in cervical precancers in an older age group: CIN3+ incidence among screened women aged 25–29 years decreased compared with incidence during the beginning of the surveillance period. Most women aged 25–29 years in 2022 had been eligible for vaccination at age 11–12 years. The decrease in CIN3+ incidence before CIN2+ could be because CIN3+ (compared with CIN2+ lesions) are more frequently positive for HPV16 or HPV18, and therefore, a higher proportion are preventable by quadrivalent HPV vaccination (5,7).

HPV-IMPACT previously reported increasing trends in precancer incidence among women in age groups 25–39 years during 2008–2015 (6). Increases were attributed to longer screening intervals (all ages) and increasing use of HPV testing (age ≥30 years), which are more sensitive for detection of CIN2+ than Pap tests; HPV testing has increased during the surveillance period (3). Continued HPV-IMPACT surveillance indicated that increasing CIN2+ incidence among several age groups from 25 to 64 years reversed or leveled during 2016–2017. By 2022, incidence among women aged 25–29 years had decreased to near 2008 levels (CIN2+) or below (CIN3+). These findings are consistent with model predictions that changing from Pap to HPV testing would cause transient increases in detected precancer and cancer, followed by decreases (8). Although some women in age groups 30–34 years and older would have been vaccinated, less impact is expected among women in this age group at this time because they were only eligible for catch-up vaccination, at ages when many women are already sexually experienced and therefore likely to have been infected with HPV.

Limitations

The findings in this report are subject to at least four limitations. First, numbers of women screened for cervical cancer were estimated using claims and survey data; inaccuracies could lead to under- or overestimates of precancer rates among screened women. Second, changes in screening and management guidelines and uncertainty in histologic classification could have affected CIN2+ and CIN3+ case detection. Third, data are limited to ecologic trends, and interpretations inferring relationships between vaccination and precancer incidence lack causal certainty; however, trend analyses such as these are routinely used to evaluate the impact of vaccination programs,^†††† and no other plausible explanations for the decreases in precancers have been identified. Finally, because one site’s catchment area expanded and the overall program standardized its screening estimation methods, numbers are not directly comparable with those reported in previous publications from this project.

Implications for Public Health Practice

These data are consistent with continuing impact of the U.S. HPV vaccination program on reducing cervical precancers (including CIN3+, the outcome most proximal to cervical cancer), and are consistent with both declines in vaccine-type HPV prevalence and early observations of reductions in cervical cancer among young women (2,9,10). The data also suggest that precancer incidence in age groups ≥25 years, which were previously observed to increase through 2015, have begun to decrease. HPV vaccination^§§§§ and guidelines-based cervical cancer screening^¶¶¶¶ are important tools for cervical cancer prevention.

HPV-IMPACT Working Group

Deborah Adeyemi, California Emerging Infections Program; Anjola-Oluwa A. Ajayi, Vanderbilt University Medical Center; Nicole R. Andersen, Vanderbilt University Medical Center; Bradley Beauchamp, Oregon Health Authority; Sarah E. Clarke, Vanderbilt University Medical Center; Emilio DeBess, Oregon Health Authority; Kyle Higgins, Connecticut Emerging Infections Program; Tiffanie M. Markus, Vanderbilt University Medical Center; Troy D. Querec, obÌåÓý; Michael Silverberg, Kaiser Permanente Northern California.