Estimated Number of Cases of High-Grade Cervical Lesions Diagnosed Among Women — United States, 2008 and 2016

Nancy M. McClung, PhD; Julia W. Gargano, PhD; Ina U. Park, MD; Erin Whitney, MPH; Nasreen Abdullah, MD; Sara Ehlers, MPH; Nancy M. Bennett, MD; Mary Scahill; Linda M. Niccolai, PhD; Monica Brackney, MS; Marie R. Griffin, MD; Manideepthi Pemmaraju, MBBS; Troy D. Querec, PhD; Angela A. Cleveland, MPH; Elizabeth R. Unger, MD, PhD; Lauri E. Markowitz, MD; HPV-IMPACT Working Group

Disclosures

Morbidity and Mortality Weekly Report. 2019;68(15):337-343. 

In This Article

Abstract and Introduction

Introduction

Human papillomavirus (HPV) causes approximately 30,000 cancers in the United States annually.[1] HPV vaccination was introduced in 2006 to prevent HPV-associated cancers and diseases.[1] Cervical cancer is the most common HPV-associated cancer in women.[1] Whereas HPV-associated cancers typically take decades to develop, screen-detected high-grade cervical lesions (cervical intraepithelial neoplasia grades 2 [CIN2], 3 [CIN3], and adenocarcinoma in situ, collectively CIN2+) develop within a few years after infection and have been used to monitor HPV vaccine impact.[1–3] CDC analyzed data from the Human Papillomavirus Vaccine Impact Monitoring Project (HPV-IMPACT), a population-based CIN2+ surveillance system, to describe rates of CIN2+ among women aged ≥18 years during 2008–2016. Age-specific rates were applied to U.S. population data to estimate the total number of CIN2+ cases diagnosed in the United States in 2008* and in 2016. From 2008 to 2016, the rate of CIN2+ per 100,000 women declined significantly in women aged 18–19 years and 20–24 years and increased significantly in women aged 40–64 years. In the United States in 2008, an estimated 216,000 (95% confidence interval [CI] = 194,000–241,000) CIN2+ cases were diagnosed, 55% of which were in women aged 18–29 years; in 2016, an estimated 196,000 (95% CI = 176,000–221,000) CIN2+ cases were diagnosed, 36% of which were in women aged 18–29 years. During 2008 and 2016, an estimated 76% of CIN2+ cases were attributable to HPV types targeted by the vaccine currently used in the United States. These estimates of CIN2+ cases likely reflect changes in CIN2+ detection resulting from updated cervical cancer screening and management recommendations, as well as primary prevention through HPV vaccination. Increasing coverage of HPV vaccination in females at the routine age of 11 or 12 years and catch-up vaccination through age 26 years will contribute to further reduction in cervical precancers.

In 2006, HPV vaccine was licensed and recommended for routine vaccination in females aged 11 or 12 years and for catch-up vaccination through age 26 years.[1] Two vaccines primarily have been used in the United States: until 2015, the quadrivalent vaccine, which in addition to HPV 6 and 11, targets high-risk, or oncogenic, HPV 16 and 18, and since 2016, 9-valent vaccine, which also targets high-risk HPV types 31, 33, 45, 52, and 58. HPV vaccination coverage among females aged 13–17 years has increased since 2007. In 2016, coverage of ≥1 dose was 65.1% and 3 doses was 43.0%.[1]

The HPV-IMPACT sites are located in five surveillance network locations. The specific catchment areas, defined by county or zip code, were selected to provide a diverse population of women and a feasible population size and geographic area for complete case ascertainment; in total, approximately 1.5 million women reside in the catchment areas. HPV-IMPACT uses active surveillance of diagnostic pathology laboratories to collect all CIN2+ cases.[3] Site staff members routinely audit all laboratories and gynecology practices serving catchment areas to ensure complete case ascertainment. Archived diagnostic specimens for type-specific HPV DNA detection of 37 types are obtained for cases in women aged 18–39 years.[2] Age-stratified CIN2+ incidence rates per 100,000 women were calculated for each year (2008–2016)§; trends were evaluated using joinpoint models in Joinpoint software (version 4.6.0.0; National Cancer Institute) and reported as average annual percentage change (AAPC) with 95% CIs.

To estimate the number of CIN2+ cases in 2008 and 2016 by age group, the observed age-specific CIN2+ rates were applied to age-specific, annual U.S. population estimates.** HPV types were categorized as HPV16/18, HPV31/33/45/52/58, and other type/HPV-negative. To estimate the number of HPV type–specific cases, the age-specific HPV type distribution observed from typing data was applied to age-specific total CIN2+ estimates.†† Case estimates were rounded to the nearest 1,000 cases.§§ An analysis using higher and lower CIN2+ rates observed in specific HPV-IMPACT sites was performed to describe potential uncertainty in estimates.¶¶

During 2008–2016, a total of 23,489 CIN2+ cases were reported to HPV-IMPACT, and HPV DNA typing was performed for 11,581 of 16,590 (69.8%) cases in women aged 18–39 years. In 2008, HPV-IMPACT CIN2+ rates were highest in women aged 20–24 years (559 per 100,000 women [95% CI = 521–600]) and were lower in successively older age groups (Table). In 2016, CIN2+ rates were highest in women aged 25–29 years (480 [95% CI = 448–515]) and lower in each successively older age group. From 2008 to 2016, the rate of CIN2+ per 100,000 women declined significantly in women aged 18–19 years and 20–24 years and increased significantly in women aged 40–64 years.

Extrapolating age-specific HPV-IMPACT rates to the U.S. population, an estimated 216,000 (95% CI = 194,000–241,000) CIN2+ cases were diagnosed in the United States in 2008 (Figure 1), including 119,000 (55%) in women aged 18–29 years, 57,000 (26%) in women aged 30–39 years, and 40,000 (18%) in women aged ≥40 years. Among the estimated 216,000 cases, 165,000 (76%) were attributable to 9-valent vaccine types (111,000 [52%] to HPV16/18 and 54,000 [25%] to HPV31/33/45/52/58) (Figure 2). Among women aged 18–24 years, 52% of CIN2+ cases were HPV16/18-attributable. Of the 165,000 CIN2+ cases attributable to 9-valent vaccine types, 91,000 (55%), 43,000 (26%), and 31,000 (19%) occurred in women aged 18–29, 30–39, and ≥40 years, respectively.

Figure 1.

Estimated number of diagnosed CIN2+ cases,* by age group — United States, 2008 and 2016
Abbreviation: CIN2+ = cervical intraepithelial neoplasia grades 2, 3, and adenocarcinoma in situ.
*Error bars indicate 95% confidence intervals, which were calculated by applying the upper and lower limits of CIN2+ rates to the age-specific U.S. population.

Figure 2.

Estimated number of diagnosed CIN2+ cases, by human papillomavirus (HPV) type* and age group — United States, 2008 and 2016
Abbreviation: CIN2+ = cervical intraepithelial neoplasia grades 2, 3 and adenocarcinoma in situ.
* Type-specificity for 2008 was based on typing data from 2008, and for 2016, was based on typing data from 2015 (most recently available) applied to 2016 case counts by diagnosis grade. HPV type group "other HPV" includes HPV-negative cases.

In 2016, an estimated 196,000 (95% CI = 176,000–221,000) CIN2+ cases were diagnosed in the United States, including 71,000 (36%) in women aged 18–29 years, 74,000 (38%) in women aged 30–39 years and 51,000 (26%) in women aged ≥40 years (Figure 1). Among the 196,000 total cases, 150,000 (76%) were attributable to 9-valent vaccine types, including 84,000 (43%) to HPV16/18 and 66,000 (34%) to HPV31/33/45/52/58 (Figure 2). Among women aged 18–24 years, 30% of CIN2+ cases were HPV 16/18-attributable. Of the 150,000 CIN2+ cases attributable to 9-valent vaccine types, 53,000 (35%), 57,000 (38%), and 40,000 (27%) occurred in women aged 18–29, 30–39, and ≥40 years, respectively.

*Surveillance through HPV-IMPACT began in 2008. Although the vaccination program began in 2006, because of the time it takes for cervical precancers to develop and be detected through screening, rates in 2008 should not be affected by vaccination and should therefore represent prevaccine era rates.
The HPV-IMPACT sites are located in five of the 10 Emerging Infections Programs (EIP) surveillance network locations (https://www.cdc.gov/ncezid/dpei/eip/eip-sites.html); the specific catchment areas within the EIP sites are New Haven County, Connecticut; Monroe County, New York; Davidson County, Tennessee; portions of Alameda County, California; and portions of Washington and Multnomah Counties, Oregon. (https://www.cdc.gov/ncird/surveillance/hpvimpact/index.html).
§Population denominators were based on county-level data from CDC's National Center for Health Statistics (https://www.cdc.gov/nchs/nvss/bridged_race/data_documentation.htm); for California and Oregon sites, county estimates were adjusted for the specific catchment area using American Community Survey data.
Trends were measured with AAPC in age-stratified rates and were considered to increase (AAPC>0) or decrease (AAPC<0) if the 95% CI did not include 0; otherwise trends were considered stable.
**The total number of U.S. CIN2+ cases in each age group and 95% CIs were estimated by multiplying the age-specific CIN2+ rates and upper and lower CIs by the age-specific U.S. population estimates. For example, in women aged 20–24 years, the observed HPV-IMPACT CIN2+ rate in 2008 was 559 (95% CI = 521–600) per 100,000 women, and there were 10,339,566 women aged 20–24 years in the United States; therefore, the estimated number of U.S. CIN2+ cases in this age group was 559 x 10,339,566/100,000 = 58,000 CIN2+ cases (rounded to the nearest 1,000). The total cases were estimated by summing the age-specific estimates.
††HPV typing data were available for approximately 70% of cases in women aged 18–39 years, and the proportion positive for each HPV type group varied by lesion grade. Therefore, to estimate the number of cases in each age group attributable to each HPV type group, the proportion positive for each HPV type group within each lesion grade was multiplied by the number of cases having each lesion grade within each age group. For age groups <40 years, the observed HPV type proportion within each lesion grade for each age group was applied. For age groups ≥40 years, the observed HPV type proportion within each lesion grade for cases among women aged 30–39 years was applied.
§§Components might not sum to totals because of rounding.
¶¶A range of lower and higher national estimates were developed based on data from two sites with historically lower CIN2+ rates (California and Tennessee) and from two sites with historically higher CIN2+ rates (Connecticut and New York). The low and high estimates of the number of prevaccine era CIN2+ cases were 159,000 (95% CI = 131,000–197,000) and 283,000 (95% CI = 246,000–328,000); the low and high estimates of the number of 2016 CIN2+ cases were 179,000 (95% CI = 150,000–217,000) and 210,000 (95% CI = 177,000–253,000).

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....