Persistent infection with human papillomavirus (HPV) causes several cancers, including cervical, penile, anal, and oropharyngeal cancers (1,2). HPV vaccination has been recommended in the United States since 2006 (3). Currently, vaccination is recommended for boys and girls aged 11–12 years and can be started at age 9 years (4,5). Catch-up vaccination is recommended through age 26 for those not vaccinated at the routine age (4,5). The 9-valent HPV vaccine (licensed in 2014 and the only vaccine currently available in the United States) was initially indicated for prevention of anogenital cancers attributable to HPV. In 2020, the US Food and Drug Administration expanded the indication to include prevention of oropharyngeal and other head and neck cancers caused by HPV types targeted by the 9-valent HPV vaccine (6).

The Centers for Disease Control and Prevention (CDC) estimates that over 70% of oropharyngeal cancers in the United States are caused by HPV (7). In addition to oropharyngeal cancers, there is some evidence to suggest that HPV also causes other head and neck cancers arising from the oral cavity and larynx. Based on this evidence, the International Agency for Research on Cancer (IARC) has been providing global estimates for potentially HPV-caused oral cavity and laryngeal cancers since 2017 (8,9). In 2018, an estimated 5900 oral cavity and 4100 laryngeal cancers worldwide were attributable to high-risk HPV types (8). The estimated fraction of HPV-attributable oropharyngeal cancers varies widely worldwide, with much higher estimates in the United States (9). However, analogous estimates of HPV-attributable oral cavity and laryngeal cancers in the US population do not currently exist.

Extending the work of IARC, the purpose of our study was to estimate the US burden of HPV16-attributable oral cavity and laryngeal cancers. HPV16 is a high-risk HPV type and the type most frequently detected in oral cavity and laryngeal cancers (10). Given this fact, and because limited data exist on other types, we focused on HPV16. Institutional Review Board approval was exempted under the Common Rule (45 CFR §46) because we used de-identified data.

First, we calculated HPV16-attributable fractions (HPV16-AFs) from previous case studies. To identify case studies, we referred to previous IARC reports (11), performed a narrative literature review, and consulted with experts in the field. For the narrative review, we searched for studies from the United States published during 2017–2022 that reported HPV16-attributable estimates for oral cavity and laryngeal cancers. We chose this time period because the methodology used by IARC for AF calculation was published in 2016 (11). We also wanted to capture case studies from the most recent 5 years at the time of our review.

We only included studies that used HPV E6/E7 mRNA detection methods to detect HPV16 in oral cavity and laryngeal cancer tissue, as these methods are considered the gold standard for detecting HPV oncogenesis in non-anogenital cancers (11–14). To calculate HPV16-AFs, we pooled estimates by summing the number of HPV16-positive cases across case studies and dividing by the sum of the total number of cases. We used the Clopper-Pearson Exact method (15) to calculate 95% confidence intervals (CIs) of the HPV16-AFs.

Second, we used 2016–2020 US Cancer Statistics (USCS) incidence data to estimate the burden of oral cavity and laryngeal cancers overall and by age, sex, and race and ethnicity. These demographic data are reported by cancer registries and collected through various methods, including from medical records and pathology reports. USCS combines cancer registry data from CDC’s National Program of Cancer Registries and the National Cancer Institute’s Surveillance, Epidemiology, and End Results (SEER) Program (16). Cancer registries demonstrate that data are of high quality by meeting USCS publication criteria (17). During 2016–2020, data from 49 cancer registries met these criteria, covering 97% of the US population (17).

The USCS analysis included new cases of primary invasive, histologically confirmed oral cavity cancer and laryngeal cancer diagnosed during 2016–2020. Oral cavity cancer was defined by the following International Classification of Diseases for Oncology, 3rd Edition (ICD-O-3) (18) site codes: C02.0-C02.3, C02.9, C03.0, C03.1, C03.9, C04.0, C04.1, C04.8, C04.9, C05.0, C05.8, C05.9, C06.0-C06.2, C06.8, and C06.9. Laryngeal cancer was defined by ICD-O-3 site codes C32.0-C32.3, C32.8, and C32.9. We restricted our analysis to oral and laryngeal cancers of squamous cell carcinoma histology (codes 8050–8086 and 8120–8131).

The definitions of oral cavity and laryngeal cancers used in our analysis are for sites etiologically likely to be HPV-related and in line with methods used in a previously published CDC population-based study to assess HPV types in cancers by anatomic site (10). The site codes listed above differ from SEER Site Recode definitions (19), which mix oral cavity and oropharyngeal cancer sites; for example, tongue includes base of tongue and other mouth includes soft palate and uvula—all of which are part of the oropharynx (see Supplementary Table 1). Given this, routinely used cancer reporting and data visualization tools are suboptimal for nuanced classification of oral cavity and oropharynx cases, which is essential for accurate estimation of HPV-associated fractions for these anatomic sites.

We reported average annual number and age-adjusted incidence rates for oral cavity and laryngeal cancers. We multiplied our HPV16-AFs and corresponding 95% CIs by the number of cases at each cancer site to estimate the average annual number of cases attributable to HPV16.

Nine studies were used to calculate HPV16-AFs (Table 1). Seven studies reported data on oral cavity, 6 on larynx, and 4 on both sites. The total number of samples tested for HPV16 was 1894: 1065 for oral cavity and 829 for laryngeal cancers. Based on these studies, the pooled HPV16-AF was 3.9% (95% CI = 2.9 to 5.3) for oral cavity cancer and 2.8% (95% CI = 1.8 to 4.1) for laryngeal cancer. The average annual number of oral cavity cancers was 12 612, with an age-adjusted incidence rate of 3.3 per 100 000 standard population (Table 2). For laryngeal cancer, the average annual number of cases was 11 170, with an age-adjusted incidence rate of 2.8 per 100 000 standard population. Based on our HPV16-AFs, we estimate that 497 (95% CI = 361 to 667) oral cavity cancers and 309 (95% CI = 198 to 461) laryngeal cancers were potentially attributable to HPV16 each year.

In summary, our pooled analysis of the published literature indicated that 3.9% of oral cavity cancer cases and 2.8% of laryngeal cancer cases in the United States are attributable to HPV16 infection, accounting for 806 total cases annually during the 5-year study period (2016–2020). Combined, the burden of these cancers is similar to penile cancer. During 2012–2016, an average annual number of 700 penile cancers were estimated to be attributable to HPV types targeted by the 9-valent HPV vaccine (20).

Our estimates are smaller than those reported in previous HPV typing studies based on DNA detection. Notably, Saraiya et al. (10) detected HPV16 using PCR-based genotyping based on any assay format in 20.4% of oral cavity cancers and 6.1% of laryngeal cancers. The difference in findings could be explained in part by inclusion of only studies that used E6/E7 mRNA detection methods in our study. DNA PCR may overestimate the causal fraction because detection of HPV DNA does not indicate its oncogenic activity (21). In contrast, mRNA RT-PCR do indicate transcriptional activity of HPV but require preservation of RNA, which is more susceptible to degradation than DNA (22). As a result, the AFs reported here may be underestimated.

The results from our study have potential implications for estimates of HPV-attributable cancers. Current CDC estimates of potentially HPV-attributable cancers (23) have not included oral cavity and laryngeal cancers because of limited data on optimal AFs. This study provides AFs that could be applied to future data collected on incidence to estimate the burden of these cancers and, in turn, increase accuracy of the overall reported number of HPV-attributable cancers in the United States.

Our estimated HPV16-AFs also provide insight into the percentage of oral cavity and larynx cancers that might be prevented by HPV vaccination (4). While HPV vaccination might lead to fewer cancer cases at these sites, given the relatively small number of estimated HPV16-attributable cases, the maximum potential benefit from HPV vaccination is likely in the prevention of oropharynx, cervical, and anogenital cancers. Vaccination should be promoted among groups at higher risk for HPV, including people with HIV or who are immunocompromised (24).

The results of this study should be interpreted in the context of its limitations. First, we were unable to calculate the percentage of oral cavity and laryngeal cancers attributable to other HPV types, as genotyping studies that examine RNA for these cancers are lacking. Additional genotyping studies like BROADEN (25,26), which aims to provide robust estimates of HPV-attributable head and neck cancers in European and Asian counties, may be beneficial to other countries, as well, to understand the full extent to which oral cavity and laryngeal (or other head and neck) cancers are attributable to HPV, including HPV16. At the present time, our study provides the best estimates of the proportion of oral cavity and laryngeal cancers potentially attributable to HPV16 in the United States.

Second, the COVID-19 pandemic created challenges in healthcare delivery, leading to delays in cancer screening and diagnosis (27), and may have affected the average annual case counts in our USCS analysis. Although additional data may be needed to understand the impact of the COVID-19 pandemic on cancer incidence, when we excluded year 2020 from the analysis, we did not identify large differences in incidence rates.

Third, the same HPV16-AFs were applied across all subgroups. Given this, our estimates may not reflect actual differences in the number of HPV16-attributable cases across subgroups. Lastly, we excluded cancer cases with no histologic confirmation from our USCS analysis. While these cases represent a very small percentage of cases, excluding them may result in misclassification with an unknown impact on the HPV16-AFs reported here.

Despite these potential limitations, our study provides new knowledge on HPV-attributable oral cavity and laryngeal cancers. In particular, our results contribute to the data on the overall burden of HPV-attributable cancers in the United States that could be prevented by vaccination. The low AFs observed also suggest that non-HPV-based prevention strategies are needed to control oral cavity and laryngeal cancer in the United States.