Background

Human papillomaviruses (HPV) are a large family of epitheliotropic DNA tumor viruses [1]. Approximately 80% of sexually active women are infected with at least one HPV subtype at some point in their lifetimes [2]. Continuous infection with a high-risk HPV subtype is the main cause of cervical cancer [3]. The HPV high-risk subtypes include HPV 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66, and 68. Among them, HPV 16 and 18 are highly associated with cervical cancer, and cause about 75% of cervical cancers worldwide [4]. Cervical cancer has become a global public health problem. It is the fourth most common cancer in women. In 2020, an estimated 604,000 women worldwide were diagnosed with cervical cancer, and an estimated 342,000 women died. In China, 110,000 new cases of cervical cancer and 59,000 deaths were reported, therefore having the second-largest burden of cervical cancer in the world [5]. Cervical cancer in China accounts for 18% of the worldwide cervical cancer incidence and 17% of cervical cancer deaths.

As the only cancer that has clear causes, and can be prevented and treated, it is expected to be fully eradicated. Therefore, promoting cervical cancer prevention is of great importance. Vaccination is an effective measure against HPV infection and to reduce cervical cancer incidence. Three prophylactic HPV vaccines are currently available worldwide, including the bivalent, quadrivalent, and 9-valent HPV vaccines [6]. The World Health Organization (WHO) recommends using HPV vaccination as part of routine vaccination in all countries [7]. HPV vaccines are currently used in 129 countries worldwide to prevent HPV-related diseases [8] and have been introduced into the national immunization plans (NIP) of 74 countries [9]. However, the HPV vaccination is not included in China’s NIP.

China has approved HPV vaccines since 2016 [10], including the Cervarix® (GlaxoSmithKline Inc.), Cecolin® (Wantai BioPharm), and Gardasil® and Gardasil® 9 (Merck & Co., Inc.) vaccines. It has a large population and unbalanced economic development among regions. The payment methods for obtaining HPV vaccination services are different among various provinces. Currently, there are three main payment methods for HPV vaccines in China: 1) residents are vaccinated against HPV at their own expense, 2) target populations get free HPV vaccination, such as in Ordos, Inner Mongolia, and Xiamen [11, 12], and 3) medical insurance does a co-payment with residents for the vaccine. However, in Guizhou Province, HPV vaccination is paid through the balance of the employee’s personal medical insurance account [13]. The local government-led free HPV vaccination program is rare in China. Most provinces still need residents to get vaccinated at their own expense, and residents experience a heavy burden of paying for the HPV vaccine. In 2019, China’s Vaccine Management Law authorized provincial governments to increase the types of vaccines available for immunization programs in accordance with the needs of disease prevention and control in their administrative regions [14]. However, China still lacks the economic evidence for HPV vaccine cost-effectiveness at the national and provincial levels. To provide information that could affect policy-making decisions to expand the use of HPV vaccines, we evaluated the economics of all valent HPV vaccines available in the Chinese market at the national and provincial levels.

Methods

Cost-effectiveness analysis was used to assess the economics of HPV vaccination at a national level and in 31 provinces in Mainland China. From the perspective of the health system, we compared the final cost and health effects of the two strategies of vaccinating and not vaccinating HPV vaccine for women of the target age. The HPV vaccines available in the Chinese market include the domestic bivalent, imported bivalent, imported quadrivalent, and imported 9-valent HPV vaccines. Lifetime effects after vaccination were estimated using the Papillomavirus Rapid Interface for Modeling and Economics (PRIME) model. The results of the economic evaluation were expressed by incremental cost-effectiveness ratios (ICER), and ICER indicators that were constructed based on the disability-adjusted life years (DALYs) were reported. The evaluation assumed that the target population had not been infected with HPV prior to vaccination. Stata15.0 was used to draw a map of China to show the cost-effectiveness results. Sensitivity analysis was conducted to test the impact of six parameters, including target age, discounted rate, vaccine efficacy, procurement, transportation and management cost, and cervical cancer treatment cost, on the robustness of the model results; the results are shown in a tornado diagram. The cost parameters used in the model have been adjusted to 2019 according to the average exchange rate of the RMB against the US dollar and the consumer price index [15]. To eliminate the effect of the time value of money, costs were discounted at a rate of 3%, as recommended by the WHO Guidelines on Health Economics [16]. This study is reported as per the Consolidated Health Economic Evaluation Reporting Standards (CHEERS) guideline (Additional file 1: Table S1) [17].

PRIME model

The PRIME model was used to conduct an economic evaluation of different HPV vaccines in different provinces and cities in China [18]. The PRIME model is a static proportional outcome model, developed by Jit et al. with support from the WHO [19]. The model aims to aid in the health economic assessment of HPV vaccination for decision-makers and medical workers at all levels. The model considered the association between HPV infection and cervical cancer lesions. Three vaccines are protective against various high-risk HPV subtypes. The bivalent and quadrivalent HPV vaccines are protective against two high-risk HPV subtypes. The 9-valent HPV vaccine is protective against seven high-risk HPV subtypes, which include HPV 16, 18, 31, 33, 45, 52, and 58. We set model parameters based on heterogeneous data from various regions. Table 1 lists the model parameters and their sources.

Table 1 PRIME model parameters and data sources
Full size table

Demographic and regional economic data

The 2019 national and provincial statistics included the number of women born in the cohort and the number of women of target vaccination age in the region. The target age was set according to the age at which HPV vaccination was approved in Mainland China [20]. The target age of vaccination for the bivalent and quadrivalent HPV vaccines was set at 9 years and that for the 9-valent HPV vaccine was set at 16 years.

Regional per capita gross domestic product (GDP) refers to the value of all final products and services produced less the value of products and services used for immediate consumption by all residential units in a region over a period. Since the publication of the Statistical Yearbook data lags behind by 1 year, the demographic and regional economic data used were from the 2020 Statistical Yearbook of each province. Cohort size at vaccination age (female) was calculated from the national population age structure. Data and calculation formulas are shown in Additional file 2: Tables S2–S4 and Additional file 3: Table S5.

Disease burden data

The disease burden data included epidemiological and economic burden data. Data on the proportion and incidence of and mortality due to cervical cancer caused by different HPV subtypes are obtained from the International Agency for Research on Cancer (IARC) HPV Information Center; data on the DALYs lost due to cervical cancer or death are obtained from the Global Burden of Disease research [21]. The cost of treatment for cervical cancer per capita was based on the cost of treatment for cervical cancer per patient from the time of diagnosis until death [22]. Disease burden data in the model are those for the national level.

Vaccine efficacy and coverage rates

The vaccine efficacy was based on the proportion of the reduction in the risk of developing cervical cancer associated with the bivalent, quadrivalent and 9-valent HPV vaccines, which we set at 100% [23, 24]. The rollout of HPV vaccination in China is in the initial stages, with coverage data at the national and provincial levels not available. Therefore, the coverage of HPV vaccination was estimated to be 80% [25, 26].

Vaccination costs

Vaccination costs included per capita vaccine procurement, transportation and management, and service costs. Details of the data are recorded in Additional file 4: Tables S6–S8. The per capita procurement cost was based on the transaction price of the centralized purchase of HPV vaccines by various provinces and cities in China. The per capita vaccine transportation and management, and service costs were based on the transportation fee and service fee of all class II or non-immunization planning vaccines published by various provinces and cities in China. Class II and non-immunization planning vaccines refer to the vaccines received by residents voluntarily, and at their own expense [27].

Economic evaluation indicators

Costs included the direct and discounted costs, and the incremental costs incurred between receiving HPV vaccines and not receiving them. These costs included vaccination, saved treatment, and net costs. The effect of HPV vaccination was based on the number of cervical cancer cases and deaths averted before and after vaccination. Life-years saved (DALYs averted) were based on the number of life-year (DALY) losses eventually averted due to cervical cancer cases averted by vaccinating a single age cohort in 2019. We also calculated the incremental cost of preventing one case of cervical cancer after HPV vaccination, preventing one death, and of saving the unit DALY. Cost effect is the ratio of the increased cost of saving a unit DALY (cost-effectiveness ratio, CER) and the incremental CER obtained compared to existing standard strategies. ICER of each province was compared with GDP per capita of each region; ICER < 1 times GDP per capita is very cost-effective, 1 < ICER< 3 times per capita GDP is cost-effective, and ICER> 3 times GDP is not cost-effective at all. The calculation formula of the cost and effect index is in Additional file 5.

Sensitivity analysis

Univariate sensitivity analysis was performed on the target age, vaccine efficacy, procurement cost, transportation and management cost, discount rate, and cervical cancer treatment cost in the model at the national level. The values of three cost-related parameters, including vaccine procurement, transportation and management costs, and cervical cancer treatment cost were adjusted by ±20%; vaccine efficacy was adjusted by − 10% and − 20%; discount rate was adjusted by ±2%; and target age for vaccination was adjusted to 13 and 26 years. We compared the changes in results caused by index changes when the target population was vaccinated with an HPV vaccine, analyzed the robustness of the model, and found the index that had the greatest impact on the results. The uncertainties of the model are summarized by a tornado diagram.

Results

Direct cost of HPV vaccines

From the perspective of vaccine valence types, the cost of the domestic bivalent HPV vaccine was the lowest, and the cost of the imported 9-valent HPV vaccine was the highest. The avoidable cost of treatment was highest for the imported 9-valent HPV vaccine, with the cost of treatment being the same for the domestic bivalent, imported bivalent and imported quadrivalent HPV vaccines. This is because the three kinds of vaccines (domestic and imported bivalent and imported quadrivalent) are effective against the same HPV subtypes, and they prevent the same number of cervical cancer cases. In terms of the net cost of prevention and treatment of cervical cancer, the net cost of the domestic bivalent HPV vaccine was the lowest, followed by the imported bivalent, imported quadrivalent and imported 9-valent HPV vaccines. The net cost at the national and provincial levels are shown in Table 2.

Table 2 Discounted cost of HPV vaccination for target population at the national and provincial levels (US$)
Full size table

Effect of HPV vaccination on cervical cancer morbidity and mortality

At the national level, the lifetime incidence of cervical cancer for females of all ages, in the same birth cohort, who completed the full courses of the domestic bivalent, imported bivalent, imported quadrivalent, and imported 9-valent HPV vaccines at the target age was significantly lower than that of females in the cohort who did not receive HPV vaccination (Fig. 1). A total of 12,545 cervical cancer cases, and 5109 deaths were averted by the domestic bivalent, imported bivalent and imported quadrivalent HPV vaccines. For the 9-valent HPV vaccine, 28,140 cervical cancer cases and 11,459 deaths were averted.

Fig. 1
figure 1

The effect of vaccination on the incidence of cervical cancer by age in 2019

Full size image

In 31 provinces, the incidence of cervical cancer decreased with age in the target population after HPV vaccination. The results showed that the number of cervical cancer cases, and deaths averted, life-years saved and DALYs prevented by the imported 9-valent HPV vaccine was higher than the other vaccines. Guangdong province, China’s most populous province, had the most prevented cases, deaths, DALYs, and saved the most life-years, followed by Shandong and Henan provinces. Figure 2 shows the number of cervical cancer cases and deaths adverted by province. The number of cervical cancer cases and deaths averted, DALYs prevented, and life-years saved at a national and provincial level are shown in Additional file 6: Tables S9–S12.

Fig. 2
figure 2

Cervical cancer cases and deaths averted by HPV vaccination at the provincial level in 2019

Full size image

The incremental cost per cervical cancer death averted after HPV vaccination in the target population was much higher than the incremental cost per cervical cancer prevented. The domestic bivalent HPV vaccine was the most cost-effective. This was followed by the imported bivalent, imported 9-valent and imported quadrivalent HPV vaccines. Nationally, the incremental cost of the bivalent HPV vaccine needed to avert cervical cancer cases was US$ 58,779, the incremental cost needed to avert cervical cancer deaths was US$ 144,329, the incremental cost to save one life-year was US$ 7879, and the incremental cost to prevent one DALY was US$ 7213. The incremental costs of cervical cancer cases and deaths averted, saving one life-year, and preventing one DALY with the imported quadrivalent HPV vaccine were US$ 148,034, US$ 363,490, US$ 19,843, and US$ 18,165, respectively (Table 3).

Table 3 Incremental costs of various HPV vaccines at the national level (US$)
Full size table

If the domestic bivalent vaccine with the best performance was used as an intervention for cervical cancer prevention, Heilongjiang province had the lowest incremental cost per case of cervical cancer averted, only requiring US$ 57,410, while Inner Mongolia had the highest incremental cost of US$ 61,977. The minimum cost per cervical cancer death averted in Henan province was US$ 140,032, and the maximum incremental cost per cervical cancer death averted in Inner Mongolia was US$ 152,182. Henan province had the lowest incremental cost per life-year saved, requiring US$ 7644; Inner Mongolia had the highest incremental cost at US$ 8308. When the imported quadrivalent vaccine was used as an intervention to prevent cervical cancer, Henan province had the lowest cost per case of cervical cancer morbidity and mortality, which was US$ 146,284 and US$ 359,193, respectively. Inner Mongolia had the highest incremental costs, with US$ 151,232 for morbidity prevention and US$ 371,343 for mortality prevention. Henan had the lowest incremental cost per life-year saved, at US $19,609, and Inner Mongolia had the highest, at US$ 20,272 (Additional file 7: Tables S13–S16).

Cost-effectiveness analysis of HPV vaccination

The four HPV vaccines were cost-effective at the national level compared to no HPV vaccination; Among the four vaccines, in 31 provinces, the domestic and imported bivalent HPV vaccines were cost-effective (< 3 times GDP per capita). The domestic bivalent vaccine was very cost-effective (<GDP per capita) in 22 regions including Liaoning, Beijing, Shanghai, Shandong, and Jiangsu. The provinces where the bivalent imported HPV vaccine was most cost-effective were mainly in the eastern coastal cities, including Beijing, Jiangsu, Shanghai, Zhejiang, Fujian, and Guangdong. The imported quadrivalent and imported 9-valent HPV vaccines were cost-effective in 29 provinces. The imported quadrivalent HPV vaccine was very cost-effective in Beijing and Shanghai, and the imported 9-valent HPV vaccine was very cost-effective in Beijing, Shanghai, and Jiangsu. There was no cost-effectiveness for the imported quadrivalent and imported 9-valent HPV vaccines in Heilongjiang and Gansu provinces (>3 times GDP per capita) (Fig. 3). The cost-effectiveness data are shown in Additional file 8: Table S17–S20.

Fig. 3
figure 3

Cost-effectiveness of HPV vaccination in 31 provinces in the target population in 2019

Full size image

The bivalent and quadrivalent HPV vaccines target the same HPV subtypes, both can protect against HPV 16 and 18, therefore they have the same protective effect against cervical cancer. The 9-valent HPV vaccine had the highest per capita vaccination cost (US$ 574.71), followed by the imported quadrivalent (US$ 357.21), imported bivalent (US$ 262.38), and domestic bivalent HPV vaccines (US$ 153.2). At both the national and provincial levels, the imported bivalent HPV vaccine was approximately 1.7 times the price of the domestic bivalent HPV vaccine, and the imported quadrivalent HPV vaccine was approximately 2.3 times the price of the domestic bivalent vaccine. Therefore, the imported bivalent and imported quadrivalent HPV vaccines were less cost-effective than the domestic bivalent HPV vaccine. The price of the imported quadrivalent HPV vaccine was approximately 1.3 times that of the imported bivalent HPV vaccine, thus making the former less cost-effective. Compared with the bivalent and quadrivalent HPV vaccines, the 9-valent HPV vaccine provides more protection against cervical cancer. Although the price of the 9-valent HPV vaccine is higher, it provides higher protection against HPV. The 9-valent is more cost-effective than the quadrivalent HPV.

Sensitivity analysis

The model results were robust, and the discount rate was the main factor affecting the baseline results. When the discount rate was adjusted by + 2%, the imported bivalent, imported quadrivalent and imported 9-valent HPV vaccines went from being cost-effective to being cost-ineffective. In the sensitivity analysis of all valent HPV vaccines, the adjustment in discount rate caused the biggest change in the ICER value. After the discount rate was adjusted (±2%), the ICER value ranged from US$ 2311 to US$ 18,959 for the domestic bivalent HPV vaccine. With the imported bivalent HPV vaccine, the ICER value ranged from US$ 4523 to US$ 33,358. ICER values ranged from US$ 6444 to US$ 45,865 after vaccination with the quadrivalent HPV vaccine. ICER values ranged from US$ 7257 to US$ 35,869 after vaccination with the 9-valent HPV vaccine (Fig. 4). The data for the sensitivity analysis are shown in Additional file 9: Table S21.

Fig. 4
figure 4

Tornado diagram of the univariate sensitivity analyses (US$/DALY gained)

Full size image

Discussion

This study evaluated the cost-effectiveness of HPV vaccination at the national and provincial levels in China compared with no HPV vaccination. Compared with not vaccinating with the HPV vaccine, vaccinating with the HPV vaccine can reduce the incidence of cervical cancer cases in women of all ages. When HPV vaccine coverage reached 80%, for the target population in 2019, introducing bivalent, quadrivalent or 9-valent HPV vaccines into the immunization program could have averted more than 12,545–28,140 cervical cancer cases and approximately 5109–11,459 deaths. Once the HPV vaccine is included in the immunization program, 80% vaccine coverage can be expected. In 2020, Erdos, China, implemented a program for free HPV vaccination for girls aged 13–18 years, and the vaccination rate of the target population reached 85% [28]. The incremental cost of using the domestic bivalent, imported bivalent, imported quadrivalent, and imported 9-valent HPV vaccines for each DALY saved is US$ 7213, US$ 13,074, US$ 18,165, and US$ 16,939, respectively. With 3 times GDP per capita as the threshold, HPV vaccination is cost-effective nationwide. This result is consistent with the research results of HPV vaccination in Vietnam, Australia, South Africa, and other countries [29–31].

With the 3 times per capita GDP as the threshold, the usage of the domestic and imported bivalent HPV vaccines in 31 provinces is cost-effective. Among them, the domestic bivalent vaccine is very cost-effective in 22 of the 31 provinces due to its price advantage. The imported bivalent HPV vaccine is very cost-effective in six economically developed regions (per capita GDP in 2019 > US$ 13,655). Except for Heilongjiang and Gansu, usage of the imported quadrivalent and 9-valent HPV vaccines in other provinces was cost-effective. In the deterministic sensitivity analysis, when the most model parameters were changed, HPV vaccination was still cost-effective.

The net cost of the 9-valent HPV vaccine was higher. However, the 9-valent HPV vaccine protects against more HPV subtypes, prevents more morbidity, and saves more treatment costs than the other vaccines. The reduction in the cost of the 9-valent HPV vaccine can further reduce its net cost. For cervical cancer, bivalent and quadrivalent HPV vaccines have the same protective effect. Currently, the net cost depends on the vaccine price. The bivalent HPV vaccines, especially the domestic bivalent HPV vaccine, have the greatest price advantage and the lowest net cost.

There are provincial differences in the economics of HPV vaccination. The increased cost is either completely worth it, or it is accepted that it is affected by the ICER value and threshold. The ICER values for the domestic bivalent HPV vaccine in Gansu and Beijing were 7138 (US$/DALY gained) and 7254 (US$/DALY gained), respectively; the per capita GDP was US$ 4784 and US$ 23,811, respectively. In Beijing, even if the 9-valent HPV vaccine was used at its highest price, its ICER value was still less than double the per capita GDP. In Gansu, even if the domestically made bivalent HPV vaccine was used at its lowest price, its ICER value was only less than 3 times the per capita GDP. The level of economic development in each province will affect its ability to pay for HPV vaccines. Including the HPV vaccine in the scope of medical insurance payments or state subsidies can increase the availability of HPV vaccines in economically disadvantaged areas.

This study had some limitations. First, our study only considered the protective effect of HPV vaccines on cervical cancer and did not consider the protective effect of HPV vaccines on genital warts, oral cancer, and other diseases, which may have caused the ICER to be overestimated. Second, we assumed that the target age population had not been infected with HPV when entering the model, but there may have been people who were infected with HPV. Third, the impact of cervical cancer screening and HPV transmission on the incidence of cervical cancer was not considered, and the herd immune response was also not considered. Fourth, the vaccine dropout rate, due to side effects after vaccination, was not considered.

Conclusions

The HPV vaccine being included in the immunization program can reduce the burden of cervical cancer. As a country with a large population, to help accelerate the elimination of cervical cancer, China should include the HPV vaccine in its immunization program as soon as possible. From a provincial perspective, Guangdong, Shandong, Henan, Sichuan, and Jiangsu have benefited from the preventable cervical cancer incidence and avoidable cervical cancer deaths after HPV vaccination, and consideration should be given to including the HPV vaccine in their immunization programs as soon as possible. In various provinces, there is a large gap in the ability to pay for HPV vaccines. To improve the accessibility of HPV vaccines, more attention should be given to economically disadvantaged areas.