Temporal Trends and Projection of Cancer Attributable to Human Papillomavirus Infection in China, 2007–2030

Persistent infection with oncogenic human papillomavirus (HPV) is not only responsible for cervical cancer but also related to cancer of the anus, vulva, vagina, penis, oral cavity, oropharynx, and larynx (1). It is estimated that 690,000 cancer cases worldwide per year are attributable to HPV, with 570,000 cervical cancer and 12,000 non-cervical cancer cases (2). The cervical cancer burden has largely declined in some high-income countries due to mass HPV vaccination and organized screening programs. However, it remains a high burden in low- and middle-income countries contributing to 80% of new cancer cases worldwide (3). Consistent increases of HPV-attributable non-cervical cancers have been identified during the past years in some countries (4–8). Particularly, male oropharyngeal cancer has surpassed cervical cancer to lead the HPV-attributable cancer burden in the United States (9).

World Health Organization (WHO) has initiated the call for eliminating cervical cancer in 2018 by integrating HPV vaccine, screening, and treatment (10). However, the prevention of non-cervical cancer mainly depends on the prophylactic HPV vaccine since no feasible screening methods are available at present. The HPV vaccine has been demonstrated to have an excellent protective effect against HPV infection and/or lesions for non-cervical cancers (11–13). With the introduction of the HPV vaccine into the National Immunization Program (NIP) in more than 110 countries (14), the HPV vaccine is expected to largely benefit HPV-attributable cancer prevention shortly. In response to WHO's call on cervical cancer elimination, China has determined to fully support the global strategy of accelerating cervical cancer elimination and to explore the optimal pathway to promote HPV vaccination and extend cervical screening coverage in China. However, the HPV vaccine has not been incorporated into Chinese NIP and the coverage of vaccination for adolescent girls is poor (15) and the screening coverage remains low in China (16). Thus, it remains a challenge for HPV-attributable cancer prevention in China.

Understanding the cancer burden attributable to HPV in China will lay solid foundations for cost-effectiveness evaluation of HPV vaccination programs and pave the way for establishing evidence-based policy and applicable recommendations for cancer control. However, only limited evidence on the changing patterns of HPV-attributable cancer in China (17, 18) is available at present and there is a lack of specified estimation in subgroup populations. Therefore, we conducted this study to estimate the temporal trends of HPV-attributable cancer burden during 2007 to 2015 by cancer site, sex, age, and geographic area and to predict the cancer cases and incidence attributable to HPV infection from 2016 to 2030.

HPV-attributable cancers are defined as cancers of cervix uteri (C53), anus (C21), vulva (C51), vagina (C52), penis (C60), oropharynx (C01, 09–10), oral cavity (C02-06), larynx (C32), according to the findings from International Agency for Research on Cancer (IARC) Monographs (19).

Age-specific rates of cancer cases and deaths during 2007–2015 were obtained from the Chinese Cancer Registry Annual Report (2010–2018) published by the National Central Cancer Registry of China (NCCR). The NCCR of China was founded in 2002, acting as the national bureau for the management of cancer registration (20). In 2015, 501 cancer registries in China submitted cancer registry data. After quality control, data from 388 cancer registries that covered 321 million persons (163 male and 158 female) were evaluated (21), accounting for about 24% of the national population. Age-specific population size from 2007 to 2015 was estimated using the data released by the National Bureau of Statistics of China.

• 1) HPV-associated cancer cases and deaths were calculated by multiplying age-specific rates with the estimated population size.

• 2) Cancers attributable to HPV were calculated by multiplying site-, sex-, area-, and age-specific HPV-associated cancer burden with corresponding population attributable fractions (PAF). PAFs calculations had been described in the reference (22). Briefly, high-risk HPV prevalence with the relative risk for cervical cancer was obtained from large-scale pooled analyses or high-quality meta-analyses of studies in China. For non-cervical cancer, the information was extracted from pooled or meta-analyses conducted in Asia or worldwide. In general, PAFs for HPV-attributable cancers were considered consistent across age groups, except for vulva cancer, whose PAF varied by age (48% for those younger than 55 years old, 28% for 55–64, and 15% for 65 and older; ref. 23).

• 3) The Segi's population was applied to calculate the age-standardized rate.

• 4) The corresponding estimates of 95% confidence interval (CI) of the HPV-attributable cancers were calculated.

• 5) Temporal trends of the HPV-attributable cancer cases, incidence, deaths, and mortality during 2007 to 2015 were calculated by JoinPoint Regression Program. The analysis identifies periods with distinct linear slopes that can be connected by join-points denoting trend changes and determine the number of join-points that should be used to best describe trends in the data (24). Average annual percentage change (AAPC) and the corresponding 95% CIs were calculated. The cancer cases, incidence, deaths, and mortality were considered increasing if the 95% CI of AAPC was entirely above zero while decreasing if the 95% CI of AAPC was below zero, otherwise, they were considered stable.

• 6) Cancer cases and incidence were further projected from 2016 to 2030 by grey prediction model GM (1,1). Grey GM (1, 1) model is one of the homogeneous exponential growth models based on the accumulation generation sequence and the least-squares method (25). It applies the accumulative generation operation (AGO) to the primary data and proceeds to solve the resulting differential equation. Then it performs an inverse AGO and calculates the predicted values of the primary data. Data management and analysis were conducted on Microsoft Excel 2010, R 3.6.2, and JoinPoint Regression Program 4.7.0.

The data generated in this study are available upon request from the corresponding author.

Table 1, Supplementary Figs. S1 to S2, and Supplementary Tables S1 to S12 presented the trends of cancer cases and incidence attributable to HPV during 2007 to 2015. The total cancer cases have elevated from 85,208 to 113,656 and increased by 3.8% (95% CI, 2.8%–4.7%) annually. The age-standardized incidence rates have changed from 4.68 to 5.80 per 100,000 persons and increased by 3.0% (95% CI, 2.5%–3.5%) per year. Female cancer cases have risen from 80,006 to 106,709, with the AAPC of 3.8% (95% CI, 2.8%–4.8%), and the incidence rates have grown from 8.76 to 10.99 per 100,000 persons, with the AAPC of 3.2% (95% CI, 2.7%–3.8%). Cervical cancer burden was growing by 3.8% (95% CI, 2.8%–4.7%) for cases and 3.2% (95% CI, 2.8%–3.7%) for incidence annually and such growths were most significant among those over 50 years old. Female anal cancer cases have risen by 6.5% (95% CI, 1.2%–12.2%) per year, with the most rapid increase happening among those at 50 to 69 years, although the growth for incidence was insignificant. A rising trend of vulvar cancer was identified, with the AAPC of 3.7% (95% CI, 1.6%–5.8%) for cases and of 2.9% (95% CI, 1.1%–4.8%) for incidence, and such increases were most evident among females over 65 years old. Growing trends for vaginal and female head-and-neck cancers were insignificant. Male cancer cases and incidence have increased by 3.7% (95% CI, 2.6%–4.9%) from 5,202 to 6,947, and by 2.4% (95% CI, 1.2%–3.7%) from 0.60 to 0.70 per 100,000 persons annually. Increasing trends were identified for male anal cancer as well, with the AAPC of 7.5% (95% CI, 2.8%–12.5%) for cases and of 6.3% (95% CI, 0.8%–12.1%) for incidence and both of which have rapidly grown among males over 50 years. Male oropharyngeal cancer cases and the incidence have increased by 4.4% (95% CI, 2.4%–6.3%) and 3.1% (95% CI, 1.2%–5.1%) annually and the elevation was most significant among those younger than 70 years. The climbing trends for penile and male oral cavity cancer and the slight decreases for male laryngeal cancer were not statistically significant. Urban and rural shared similar trends on HPV-attributable cancer cases and incidence, except anal cancer, which was more frequently happened in rural areas.

Trends of cancer deaths and mortality attributable to HPV during 2007–2015 were shown in Table 2, Supplementary Figs. S3 to S4, and Supplementary Tables S1 to S12. Total cancer deaths have climbed from 25,785 to 36,909 by 4.7% (95% CI, 2.9%–6.7%) per year on average. The age-standardized mortality rates have changed from 1.42 to 1.83 per 100,000 persons by 3.3% (95% CI, 0.9%–5.8%) annually. Female cancer deaths have increased from 23,239 to 33,181 (AAPC, 4.6%; 95% CI, 2.4%–6.8%) and the mortality have elevated from 2.53 to 3.27 per 100,000 persons (AAPC, 3.3%; 95% CI, 0.4%–6.3%). Cervical cancer deaths and mortality have risen by 4.4% (95% CI, 2.0%–7.0%) and 3.3% (95% CI, 0.1%–6.5%) per year respectively and such increases were most evident among those over 50 years old. Female anal cancer has largely increased by 13.4% (95% CI, 8.4%–18.6%) for deaths and 11.9% (95% CI, 7.0%–17.1%) for mortality rates annually, particularly among those aged 65 and older. Vaginal cancer deaths have increased by 7.2% (95% CI, 2.1%–12.4%) per year and the highest increase was identified at 70 years and older. The changing patterns were not statistically significant for vulva cancer and female head-and-neck cancer, although slight increasing and decreasing trends were identified for the two cancers respectively. The total male cancer deaths and mortality rates have increased from 2,546 to 3,728 (AAPC, 5.5%; 95% CI, 1.7%–9.4%) and from 0.30 to 0.37 per 100,000 persons (AAPC, 5.3%; 95% CI, 2.0%–8.6%) respectively. Male anal cancer deaths and mortality have risen by 16.1% (95% CI, 9.1%–23.7%) and 14.1% (95% CI, 6.3%–22.5%) per year respectively and such increases were most significant among those older than 50 years. Male's Penile and oropharyngeal cancer deaths and mortality have increased, although not statistically significant. Similarly, the overall slightly decreasing trends of other male head-and-neck cancers were insignificant. Urban and rural were identified with similar trends on HPV-attributable cancer deaths and mortality.

Figures 1 and 2 and Supplementary Tables S13 and S14 showed the projection of cancer attributable to HPV from the year 2016 to 2030. The projected HPV-attributable cancer cases would increase from 117,929 to 214,077, while the age-standardized incidence rates rise from 6.07 to 9.35 per 100,000 persons. In 2030, 87.7% of HPV-attributable cancer cases would be cervical cancer (187,656), following by cancers of anus (14,125), oropharynx (2,860), penis (2,813), vagina (2,625), larynx (1,378), oral cavity (1,354), and vulva (1,266). Female cancer cases and incidence would elevate from 110,620 to 198,953 and from 11.55 to 18.30 per 100,000 persons respectively. Cervical cancer cases and incidence would increase from 105,894 to 187,656 and from 11.08 to 17.52 per 100,000 persons respectively. Female anal cancers would almost triple in 2030 compared with that of 2016. Vulva and vaginal cancers are projected to be 1.5 times higher and nearly doubled respectively, while head-and-neck cancer burdens would remain stable. Male cancer cases are projected from 7,309 to 15,124 and the incidence is from 0.74 to 0.99 per 100,000 persons. Male anal cancers will increase by more than three times in the year 2030 compared with that of 2016. Oropharyngeal cancers are projected to double in 2030, while penile, oral cavity, and laryngeal cancer burdens would remain stable.

HPV-attributable cancer burdens have largely increased during 2007 to 2015 in China. Cervical and anal cancers grew most rapidly. Meanwhile, increasing trends of oropharyngeal and vulva cancers were identified as well. The cancer burden attributable to HPV is projected to increase in the next decade, with about 88% of new cases being cervical cancer. The anal cancer burden is expected to be almost tripled by 2030. Female genital and male oropharyngeal cancer burdens are expected to keep rising, although the absolute number is relatively small compared with cervical and anal cancer burdens.

Our study showed that HPV-attributable cancer incidence rates have increased by 3.2% per year in China, which was close to the estimation using Cancer Incidence in Five Continents (CI5) data (AAPC, 3.8%; 95% CI, 2.1%–5.5%; ref. 18). The cancer burden attributable to HPV was dominated by cervical cancer in the periods from 2007 to 2030 in China. In the past years, cervical cancer has increased by 3.8% in cases and 4.4% in deaths and the cancer burden is expected to keep rising in the next decade. However, a declining cervical cancer burden has been observed in the past few decades due to organized screening and HPV vaccination in some high-income countries (9, 26). The intensively increasing cervical cancer burden in China has highlighted the urgency of effective implementation of comprehensive control strategies, involving HPV vaccination, screening, and treatment. However, the HPV vaccination rate among girls aged 9 to 14 can be almost ignored (15) and the cervical screening rate is only 26.7% among women aged 35 to 64 (27) in China. China is still far away from the WHO's 2030 target of HPV vaccination for 90% of girls less than 15 years, cervical cancer screening for 70% of women aged 35 to 45, and treatment for 90% of women who have been diagnosed with cervical precancer/cancer.

For HPV-attributable non-cervical cancers, anal cancer has borne the highest burden with consistent increasing trends since 2007 and is expected to triple in 2030 in China if no feasible prevention strategies are applied. Similar to the situation in the US, anal cancer cases have increased by 6.5% and 7.5% annually among females and males respectively and the growth was most evident among those over 50 years old (9). Although anal screening by HPV testing or cytology is recommended among high-risk populations including people living with human immunodeficiency virus, men who have sex with men, and women with cervical lesions (28), there is still a lack of evidence-based and formative screening guidelines. Therefore, effective anal cancer prevention mainly depends on nationwide HPV vaccination at present, which further demonstrates the necessity of including HPV vaccination into NIP to benefit particularly the high-risk individuals. Notably, cervical and anal HPV infections or lesions have been considered highly correlated (29), and thus promoting HPV-based cervical cancer screening would benefit anal cancer prevention potentially.

Growing trends on the burden of rare cancers such as oropharyngeal, vulva, and vaginal cancer have been observed during 2007 to 2030 in China. Similarly, increasing oropharyngeal cancer burden has been identified in some high-income countries (18, 26). In the US, male oropharyngeal cancer has surpassed cervical cancer and become the highest burden of HPV-attributable cancers in recent years (9). The increase of oropharyngeal cancer burden in high-income countries was reported to be associated with higher oral HPV infection, which resulted from increasing oral sex behaviors in recently born cohorts (30, 31). The increasing trends of the vulva and vaginal cancer identified in some countries are likely to link with the generational changes in sexual behavior, such as the earlier age at sexual debut, and increasing transmission of HPV among cohorts born 1940 to 1950 and thereafter (32). With the aging population and changing sexual behavior in China, attention should be paid to HPV-attributable head-and-neck cancer and genital cancer prevention in the near future. Given the fact that no feasible screening methods are available for these cancers at present, HPV vaccination would be the best way to prevent them. Thus, additional public health benefits are expected to be earned from the nationwide HPV vaccination program under the background of cervical cancer elimination.

This study estimated the temporal trends of HPV-attributable cancer burden in the past decade based on the NCCR data and further projected the cancer burden in the next decade in China. The NCCR data is one of the important sources for the CI5 databases and the data from 36 cancer registries has been selected into the latest volume of CI5 (Volume XI, covering registrations 2007–12) according to the strict quality selection process (33). With the expansion of the number of cancer registry sites and improvement of data quality, cancer registration data is considered to be representative of the national cancer burden in China and plays an increasingly critical role in cancer prevention and control (33). Our study provides a comprehensive national profile on the past and future burden of HPV-attributable cancer at a population level and supplies fundamental data and scientific evidence for policymaking on cancer control and prevention.

However, there are several limitations in this study, which may negatively impact the precision of estimation of the actual cancer burden and the predicted burden attributable to HPV in China. First, our results were estimated by international PAFs for non-cervical cancers. Therefore, the burden of oropharyngeal cancer may have been underestimated because the Chines PAF is closer to that of in high-income Asian countries, which is higher than the international level (34). Second, our study adopted consistent PAFs across all calendar years, and thus potential biases may exist when estimating the cancer burden that is not predominantly HPV-associated. For example, oral cavity and laryngeal cancers are more commonly caused by smoking and alcohol use than HPV infection. Third, the projected cancer burden in this study is based on a statistical projection model and no potential advances in the prevention, diagnosis, or therapy that may dramatically impact future incidence rates were taken into account.

In conclusion, the HPV-attributable cancer burden has increased since 2007 and is predicted to keep rising in the near future in China. Cervical cancer dominates the HPV-attributable cancers and has consistently risen from 2007 to 2030. Next to cervical cancer, the anal cancer burden has rapidly increased in the past decade and will keep growing in the next decade. Comprehensive cervical cancer control integrating HPV vaccination, cervical screening, and treatment for precancer/cancer should be prioritized in China. Meanwhile, HPV vaccination and anal cancer screening for high-risk populations are highly recommended. Further evaluation for HPV-attributable cancer burden with Chinese-specified PAFs and a scenario-based forecast model considering the possible changes are expected to provide a more precise and tailored HPV-attributable cancer burden estimation in China.

No disclosures were reported.

R. Duan: Conceptualization, resources, data curation, software, formal analysis, visualization, methodology, writing–original draft, writing–review and editing. K. Xu: Data curation, software, formal analysis, visualization, methodology, writing–original draft. L. Huang: Data curation, formal analysis, investigation, visualization, methodology, writing–original draft, writing–review and editing. M. Yuan: Formal analysis, investigation, visualization, writing–original draft. H. Wang: Formal analysis, methodology, writing–original draft. Y. Qiao: Conceptualization, resources, validation, investigation. F. Zhao: Conceptualization, resources, data curation, formal analysis, supervision, funding acquisition, validation, investigation, visualization, methodology, project administration, writing–review and editing.

We thank M. Chen from North China University of Science and Technology and W.L. Chang from Sichuan University for data collection. This study was financially supported by the National Natural Science Foundation of China [81761128006], the Chinese Academy of Medical Sciences Initiative for Innovative Medicine [2016-I2M-1-019], and a Merck grant (# 59476). However, no employee of Merck Co. has ever participated.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.