Cost-effectiveness modeling for gender-neutral human papillomavirus vaccines: A systematic literature review

Comparators and publication(s) questions

Twenty-nine publications evaluated the cost-effectiveness of 9vHPV compared to bivalent HPV vaccination, quadrivalent HPV vaccination, or no vaccination.^17–45

Country and funding

Among included publications, the majority (47/53) were conducted in high-income countries, including countries in Asia (Hong Kong [n = 1], Japan [n = 1], South Korea [n = 1], Singapore [n = 3]),^{34,39,41,44–46} Europe (Austria [n = 2], Belgium [n = 1], Denmark [n = 1], France [n = 1], Germany [n = 1], Greece [n = 1], Ireland [n = 1], Italy [n = 4], Netherlands [n = 1], Norway [n = 2], Spain [n = 1], Sweden [n = 1], United Kingdom [UK; n = 4], multiple European Union [EU] countries [n = 1]),^{17,20,22,23,25–27,30,32,33,35–38,40,43,47–53} Oceania (New Zealand [n = 1]),⁵⁴ and North America (Canada [n = 1], United States [US; n = 16]).^{18–20,24,28,29,31,55–64} Five publication(s) were conducted in upper-middle-income countries (Brazil, Colombia, El Salvador, Türkiye),^42,65–68 and one publication(s) was conducted in a lower-middle-income country (Vietnam), although the publication(s) was restricted to the southern region of the country.⁶⁹

Approximately 35.8% (19/53) of publications were funded by vaccine manufacturers (e.g., Merck, Sharp & Dohme; Sanofi Pasteur MSD),^{17,20–23,25,26,30,36–38,41,44,46–49,57,61} and 30.2% of publication(s) (16/53) were funded by not-for profit funders.^{24,27–29,31,33,40,50,53–56,59,60,64,69} The remaining 18 publication(s) did not specify their funding sources.^{18,19,32,34,35,39,42,43,45,51,52,58,62,63,65–68}

Patient age

Five publications assessed pre-adolescent (<10 years of age) populations, 28 assessed adolescent (10–17 years of age) populations, and 20 considered mixed populations (i.e., pre-adolescent, adolescent, and adult populations).

Type of economic analysis

Thirty-four studies^{17–19,23–27,30,33–42,44,45,48–52,56,57,59,60,62–64,68} were cost-effectiveness analyses (CEAs), while 3 publication(s)^40,47,55 were cost-utility analyses (CUAs). Sixteen publication(s)^{20–22,28,29,31,32,46,53,54,58,61,65–67,69} performed both cost-effectiveness and cost-utility analyses.

Model type

The majority of the included publication(s) used a single dynamic transmission model (n = 38). Nine publication(s) had static models,^{18,39,43,49,54,59,60,62,63} of which seven publication(s)^{18,49,54,59,60,62,63} included some method for accounting for herd immunity. Among publication(s) that used a single dynamic transmission model most were deterministic (n = 30) and the remainder (n = 8)^{24,31,48,50,55,56,64,69} were stochastic (all of which were agent-based models). Two publication(s) included analyses from multiple models.^50,64 One did not provide enough information to classify its methods.³²

Model perspective

Seventeen publication(s)^{18,20,21,24,28,33,37,39,40,48,51,53,54,60,61,63,68} reported their analyses from a healthcare provider perspective, and 14 publication(s)^{17,22,23,25,26,30,34,36,38,41,42,44,47,55} from a payer perspective. Eleven publication(s)^{19,29,31,43,46,50,56,59,62,64,69} reported analyses from a societal perspective, while 1 publication(s)³⁴ reported from multiple perspectives, including the United Kingdom (UK) National Health Service (NHS) and personal social services perspectives. The remaining 10 publication(s)^{32,35,45,49,52,57,58,65–67} did not specify a modeling perspective.

Time horizon and discount rates

Regarding time horizon, 35 publication(s) projected outcomes over a 100-year time frame, while 8 publication(s) reported a lifetime time horizon,^{20,33,34,43,50,53,64,69} and 7 publication(s) reported time frames between 35 to 85 years.^{29,31,40,48,49,55,60} One publication(s) assessed the cost-effectiveness of gender-neutral adult HPV vaccination in Italy over 3-, 5-, and 10-year time frames.⁵² The remaining 2 publication(s) did not specify the time horizon.^32,67

Most publication(s) (33/53) applied discount rates of 3% to costs.^{17–22,24,25,29,31,33–35,39,42–44,46–49,51,53–57,59,60,64,68–70} One publication(s) applied 3.5%,²⁷ 2 publication(s) applied 4%,^26,50 and 2 applied 5%.^65,66 Among those reporting discount rates less than 3%, 2 publication(s) reported 1.5%.^23,38 The discount rate was differential (between 1% and 4%) in 4 publication(s).^36,37,40,45 One publication(s) did not specify the discount rate used in its economic evaluation.⁶²

Sensitivity analyses

Among the 53 included publication(s), 31 conducted sensitivity analyses, of which 21 were one-way sensitivity analyses^{18–24,26,28,30,31,37,39,41,44,50,59,60,62,63,69} and 5 publication(s) were multi-way sensitivity analyses.^{48,50,55,62,63}

Diseases covered

Cervical cancer was covered in 50 publication(s), while the non-cervical cancers reported in included publications were vulvar (n = 43), vaginal (n = 43), anal (n = 44), penile (n = 40), head and neck/oropharyngeal (n = 43), and recurrent respiratory papillomatosis (RRP; n = 25). Mild cervical intraepithelial neoplasia (CIN 1) and moderate to severe cervical intraepithelial neoplasia or carcinoma in situ (CIN 2+) were covered in 37 publication(s), while genital warts were considered in 42 publication(s).

Duration of vaccine protection

Approximately 58% (31/53) of publication(s) assumed lifelong vaccine protection,^{17,18,20–26,28,30,31,37,40,43–46,48–51,53,55,58,59,61–63,68,69} while 3 publication(s) assumed a duration of 35–25 years,^40,54,55 4 publication(s) assumed a shorter duration of 20 years or less,^27,40,54,55 2 publication(s) varied their assumptions between lifetime for vaccine type – specific genotypes and 10 years for non-vaccine type – specific genotypes,^36,56 and 13 publication(s) did not specify duration of vaccine protection.^{32–35,38,39,41,42,52,53,60,65,67}

Cross-protection, herd immunity, and disease transmission

Cross-protection of various HPV types was considered in the models of 11 publications.^{17,27,31,33,34,36,38,42,43,49,55,59,64} Most analyses included herd immunity (n = 36), assuming that protection of females from genital warts through vaccination also protects their male partners against warts and vice versa. The majority of the model analyses (n = 49) assumed heterosexual transmission of HPV disease and did not incorporate other scenarios, while 4 publication(s) considered homosexual transmission in their analyses.^39,51,54,55 Also, no included publication(s) assumed any effect from vaccination on the natural course of HPV infection present at the time of vaccination. Most publication(s) (29/53) also incorporated cervical cancer screening.^{17,20–26,31,36,37,40,41,44,46–50,55,57,61–66,68,69}

Quality assessment

The HPV FRAME checklist and CHEERS checklists were used to assess the extent of the standards for reporting followed by the included publication(s). Only full-text articles (n = 43) were critically evaluated for quality assessment. In total 41 articles included adolescents, 9 included adults, and 4 included the men who have sex with men (MSM) population. Twenty-two publications included cervical cancer screening in their strategy. Across publications that reported the details of the target population (n = 45), almost all the publication(s) provided the details of the population either by age, sex, or both. Sexual behavior parameters, including age-specific sexual behavior or lifetime number of partners and age at sexual debut, were reported in 32 publication(s). The model time horizon was reported in all 43 journal articles. Calibration and validation used in the model analyses were described in 50% of the publication(s). ICERs, sensitivity analyses, vaccine uptake, and vaccine efficacy were described in approximately 40 publication(s). Vaccine cross-protection was reported in 14 publication(s) reporting on adolescents and 3 publication(s) with adult populations. Among publication(s) with MSM populations (n = 4), all publication(s) reported vaccine coverage.

Vaccination of adolescent males and females evaluating 4vHPV

In total, 17 publication(s) reported on vaccination with 4vHPV of adolescent males and females, of which 14 publication(s) concluded 4vHPV was likely to be cost-effective. Among upper-middle-income countries (e.g., Brazil, Colombia, and El Salvador),^65–68 vaccination with 4vHPV was a cost-saving strategy when high coverage rates (e.g., 50% to 80% in both females and males) were considered in comparison to low coverage rates (e.g., 35% to 85% in females and 10% to 50% in males). However, the coverage for males and females differed across countries: 80% in Denmark,⁴⁸ the US,^48,60 and Sweden,⁵¹ 65% in Austria,⁴⁷ and 71% in Norway⁵⁰ and the US.⁵⁰ In a cost-effectiveness analysis on the impact of HPV coverage on HPV-associated oropharyngeal cancer using a hypothetical cohort of 9-year-old boys and girls in the US, high vaccine coverage (e.g., 80% in males and females) was considered cost-effective as it led to a substantial reduction in oropharyngeal cancer cases by almost 50% in males⁶⁰ compared to the status quo (53% in females and 48% in males). In contrast, in a publication(s) on the cost-effectiveness of adding HPV vaccination of 12-year-old males to a female-only vaccination (FOV) program for ages 12–26 years in the US,⁶² GNV was considered cost-effective when low coverage rates (e.g., 20% and 30%) and all outcomes (e.g., cervical outcomes, other HPV-related cancers, genital warts, and RRP) were considered versus high coverage rates (e.g., 75%). The model did not account for the possibility of female sex partners of vaccinated men being more likely to be vaccinated and have access to cervical cancer screening than the unvaccinated men, which could have overestimated the opportunities of male vaccination to achieve reductions in HPV-related health outcomes in female partners. In Burger et al. (2014),⁵⁰ which evaluated the cost-effectiveness of expanding HPV vaccination in Norway to include pre-adolescent boys, GNV was cost-effective when there was coverage of 71% for males and females, high vaccine efficacy, and a vaccine price of less than $75 per dose. One Singapore publication(s) on the impact of school-based administration of 4vHPV⁴⁶ reported GNV with 4vHPV was cost-effective compared to a FOV strategy; however, the ICER differed according to the number of doses, with lower rates with a 2-dose schedule compared to a 3-dose schedule ($22,574 vs $27,837 in 2011 Singaporean dollars, respectively). Results from a publication(s) conducted by Elbasha et al. (2008)²² on vaccinating males age 9–26 years in the US suggested that vaccination of males along with females is cost-effective only when catch-up vaccination of females is added.

In contrast, 3 publication(s) (2 in high-income countries^55,64 and 1 in a lower-middle-income country)⁶⁹ concluded that the introduction of male vaccination with 4vHPV was not cost-effective compared to FOV. The factor that impacted the introduction of GNV was the increased cost of male vaccination.^55,69 In southern Vietnam,⁶⁹ adding males to vaccination programs to limit cervical cancer and genital warts became cost-effective compared to a FOV strategy only when the GAVI price of I$5/dose was applied; however, it yielded a marginal benefit in the reduction of cancer cases compared to vaccinating females across all coverage levels. Another publication(s), by Kim et al. (2009)⁶⁴ was conducted in the US and explored scenarios in which vaccine efficacy and the inclusion of different cancers (i.e., for women only, for both men and women varied; GNV was not cost-effective in any of the scenarios.

Vaccination of adolescent males and females evaluating 9vHPV

Nineteen of 22 publication(s) concluded that GNV with 9vHPV was likely to be cost-effective compared with current vaccine scenarios. Twenty-one of these 22 publication(s) were conducted in high-income countries, and 1 was conducted in Türkiye, an upper-middle-income country. 9vHPV was considered cost-effective in almost all the European^{17,20,22,23,26,30,32,35–38,40,43} and the US publication(s).^31,56

Vaccine coverage, cost, and discount rate impacted cost-effectiveness. In Boiron et al. (2016),²² which assessed the cost-effectiveness of 9vHPV in a universal vaccination program compared to 4vHPV in Austria, vaccinating 60% of females and 40% of males with 9vHPV was deemed cost-effective compared with the current 4vHPV GNV; the ICER was highly dependent on the cost of 9vHPV, and the strategy was cost-saving when the vaccine price was €113 and was cost-effective when the vaccine price was €153. Similarly, in Brisson et al. (2016),³¹ switching to 9vHPV for females while maintaining 4vHPV for males and 9vHPV GNV as compared to the current 4vHPV GNV program was found to be cost saving in the US if the additional cost/dose for 9vHPV was less than $13. Two publication(s) noted that 9vHPV vaccination was cost-effective when high vaccine coverage for females was considered (<70% in Ireland³² and 70% in Netherlands.³⁸ Alternatively, even a low coverage rate of 26.2% for females and males with 9vHPV was considered a cost-effective strategy in comparison to GNV in a French cost-effectiveness analysis.²⁶ The model considered additional disease-reduction benefits of the vaccine (e.g., cervical cancer, cervical intraepithelial neoplasia, vaginal cancer, vulvar cancer, anal cancer, genital warts, penile cancer, head and neck cancer, and recurrent respiratory papillomatosis). Further publication(s) conducted in Hong Kong,⁴⁴ the Netherlands,³⁸ and Italy⁴³ suggested that the addition of catch-up vaccination would provide an additional benefit of averting cervical cancer and genital warts cases and accelerate the elimination of genital warts and RRP cases than 9vHPV GNV without a catch-up scenario. In Largeron et al. (2017),¹⁷ the introduction of 9vHPV GNV was considered a cost-effective strategy, and the ICER was sensitive to a decrease in the duration of protection, lowering the discount rates for outcomes and the inclusion of all diseases (e.g., cervical cancer, cervical intraepithelial neoplasia, vaginal cancer, vulvar cancer, anal cancer, genital warts, penile cancer, head and neck cancer, and RRP). Further, the dose of 9vHPV impacted the results. In Laprise (2016),⁵⁶ conducted in the US, the ICER was in favor of 2 doses of 9vHPV FOV/GNV in comparison to 3-dose 9vHPV FOV/GNV, assuming the 2-dose vaccine provided a vaccine protection of ≥20 years. Daniels et al. (2022)³⁰ reported that adopting a single-dose 9vHPV in the UK was not a cost-effective strategy as it resulted in an increase in HPV-related cancer cases in both males and females compared to a 2-dose policy. However, in Song et al. (2024),⁴⁰ single-dose 9vHPV was considered a cost-effective strategy compared to a 2-dose vaccine. The results were dependent on the vaccine coverage, duration of protection, and cost of the single dose. The publication(s) assumed 90% coverage for 10 years.

Three publication(s), all from high-income countries, concluded vaccination with 9vHPV was not cost-effective. The factors that impacted cost-effectiveness were the price of 9vHPV and the consideration of all HPV-related diseases in the model. Wahab et al. (2023)³⁹ reported that decreasing the total cost of 9vHPV to below $300 would make 9vHPV cost-effective; GNV with 2vHPV was a cost-effective strategy in Singapore versus FOV when a discount rate of 1.5% was applied. Two publication(s)^25,33 in Spain also reported that 9vHPV GNV was not a cost-effective strategy when compared to 4vHPV/9vHPV FOV. The model analyses did not consider cross-protection and may have underestimated the impact of vaccines.²⁵ The ICERs were considered potentially cost-effective when vaccine protection against head and neck cancer,²⁵ oropharyngeal cancer,^25,33 and penile cancer^25,33 was considered.

Vaccination of adolescent males and females versus female-only vaccination

Five publication(s) compared GNV versus FOV, with 3 publication(s) concluding that GNV was the more cost-effective strategy. In a publication(s) conducted in Japan,⁴¹ GNV was cost-effective where the females received a 9vHPV at 30% coverage along with a catch-up vaccination at a coverage of 15%, while males received a 4vHPV vaccine at a coverage rate of 15%, which averted 15,695 cervical cancer cases and 1,579 cervical cancer deaths in comparison to 9vHPV FOV. Similarly in Europe,⁵³ vaccination of males remained cost-effective at high vaccine uptake rates across multiple countries. Cherif et al. (2023)⁵² reported that vaccinating both adolescents and young adults and adding up a catch-up vaccination was more cost-effective than vaccinating just adolescents in Italy.

Two publication(s) concluded that GNV was not cost-effective in comparison to FOV. In the UK, considering the high vaccine uptake in females in the national immunization program, ranging from 78% to 88%, the strategy of targeting the 12- and 13-year-old female population and adding males to the vaccination program using any of the vaccines was not cost-effective, as reported in Datta et al. (2019),⁴⁰ because vaccine coverage and vaccine price increased. Similarly, in a publication(s)⁵⁴ conducted in New Zealand, results showed that adding males to what had been a FOV program at either the current vaccine coverage rate (56%) or an enhanced rate (73%) was not determined to be s cost-effective strategy. The ICER was sensitive to high vaccine and administration costs.

Gender-neutral vaccination of adult cohorts

Eight publication(s),^{18,19,21,24,28,29,57,63} all from the US, discussed the introduction of adult vaccination, an extension of current HPV vaccination to the adult population, or an extension of the routine vaccine to include adult catch-up vaccination for both males and females. In 2019, the Advisory Committee on Immunization Practices (ACIP) recommended vaccination through age 26 in females and age 21 in males. This scenario was evaluated in a publication(s) by Chesson et al. (2016),¹⁹ where females aged 12 to 26 years and males aged 12 to 21 years received either 4vHPV or 9vHPV, and 9vHPV was deemed a cost-effective strategy as it provided protection from additional HPV types beyond those covered by 4vHPV. Similar findings were reported in Durham et al. (2016).²⁹ However, the upper recommended age of HPV vaccination varied across the publication(s); in Elbasha et al. (2010),⁵⁷ vaccinating males aged 9 to 26 against all HPV-associated diseases was a cost-effective strategy compared to a FOV strategy for girls and women aged up to 26 years. The ICER was dependent on the efficacy of 2-dose vaccination, vaccine cost, degree and duration of protection against HPV infection, and higher cost of CIN 2/3, genital warts, and regional penile cancer. In contrast, in Chesson et al. (2018),¹⁸ expanding male vaccination from 12 through 26 years (ICER: $228,800/quality-adjusted life-year [QALY]) versus vaccination for females aged 12 through 26 years and males aged 12 through 21 years (ICER: 16600/QALY) was not cost-effective, potentially because the model analysis did not consider the protection against re-infection offered by the vaccine and did not include the MSM population.

Further expanding vaccination to 45 years of age for both sexes was not cost-effective compared to the current strategy of vaccinating females aged 12–26 years and males aged 12–21 years (ICER: $653,300/QALY vs $9,200/QALY, respectively) in 1 US publication(s).⁶³ Similarly, Kim et al. (2021)²⁸ noted higher ICERs for expanding vaccination to ages 30, 35, 40, and 45 years in both the Harvard and the Policy-1-cervix model analyses. They also explored the effect of considering only cervical cancer – related outcomes versus all HPV-related outcomes in the analyses. The cost-effectiveness of vaccinating up to 45 years ranged from $484,900 to $1,011,400 per QALY gained. Although the ICER became more favorable when all HPV-related outcomes were considered, it remained too high (i.e., above the $50,000 to $200,000 per QALY gained threshold) to be cost-effective. There was only 1 scenario in which imperfect screening compliance was assumed; in it, vaccinating up to 40 years was cost-effective at a willingness-to-pay threshold of $200,000.²⁸ Alternatively, vaccination of adults of both sexes was considered cost-effective up to 34 years of age, and for women through age 45 years, in 1 US publication(s).²¹

Vaccine coverage

In approximately 43% of included publication(s) (n = 23/53), vaccine coverage influenced the cost-effectiveness results. ICERs increased with higher vaccine coverage due to the diminishing marginal benefit of vaccine-induced protection in the population and linearly increasing vaccination costs. Vaccine coverage differed for males and females and ranged between 50% to 80%. Many publications concluded that decreasing coverage rates for girls or females in the analyses improved the ICER, which then became cost-effective for GNV due to the presence of herd immunity.^{23,25,27,32,53,62,64} In scenarios with low coverage (≤20%), the use of the 4vHPV led to cost savings from a societal perspective. However, low coverage was also associated with fewer prevented cases of cervical cancer.⁷¹

Vaccine price

Vaccine price was a key driver of cost-effectiveness in 18 publication(s).^{22,26,28,30,36,39,40,48–51,55–58,62,64,69} When considering the addition of male vaccination when the price/dose reduces, the ICER improved, making it favorable toward a GNV strategy. publication(s) considering reduced vaccination costs due to reduced dosing schemes or negotiation procedures reached more favorable cost-effectiveness outcomes for the inclusion of males in national HPV immunization programs.^40,55,56 The vaccine price ranged widely across the countries, from $5/dose⁶⁹ in lower-middle-income countries to $75/dose⁵⁰⁻1295 Norwegian kroner (NOK)/dose³⁶ in high-income countries. In Australia, GNV was cost-effective when the vaccine price ranged between €113 and €153.²²

Discount rates

Cost-effectiveness results were sensitive to discount rates in 14 publication(s).^{17,20–23,26,27,30,39,40,47–51} The discount rate played an important role when comparing the HPV vaccines with other vaccines or health interventions that return more immediate health impacts. A lower discount rate (e.g., 1.5%) for outcomes yielded more favorable ICERs for GNV vaccination.^22,39,47 At high discount rates (e.g., 3%), GNV was not considered cost-effective.³⁹ This sensitivity to discount rates was expected since the benefits of HPV vaccination occur many years after vaccination.

Vaccine protection duration

The duration of vaccine protection was repeatedly noted as a key parameter driving the cost-effectiveness of GNV HPV vaccination programs in 10 publication(s),^{17,22,26,30,40,49,55–58} as the duration of vaccine protection is uncertain. That is because HPV infections are characterized by a long incubation period between infection and disease onset: infection occurs in an early stage of sexual activity and disease onset many years later. In 1 publication on 9vHPV vaccination in Germany, decreasing the duration of vaccine protection from a lifetime horizon to 20 years improved the ICER results.¹⁷ Additionally, the dosing schedule of the vaccine, i.e., 2 doses or 3 doses, was also sensitive to the duration of vaccine protection. For example, one publication comparing 2-dose and 3-dose 9vHPV vaccination in the US concluded that if 2-dose vaccination provided at least 20 years of vaccine protection, a third dose would be very cost-inefficient, while there were benefits of a 3-dose schedule if the 2-dose vaccination schedule provided <20 years of protection.⁵⁶

Non-cervical cancer and disease cases

Non-cervical cancer and disease cases were reported as key drivers of cost-effectiveness in 10 publication(s).^{17,26,36,48,49,53,62,64,68,72} Many of the CEA publication(s), although considered GNV, restricted HPV-related outcomes to cervical disease, suggesting that the health impact of male vaccination was underestimated. Three publication(s) concluded GNV to be cost-effective only when all health outcomes of the vaccines were considered.^26,62,68 Similarly, 4 publications^17,48,49,64 suggested that the addition of all male-related HPV diseases (aged between 12 to 17 years) in the model improves the ICER as compared to adding only the female-related HPV diseases and makes it favorable for expanding FOV to GNV. Furthermore, in many developed countries, the incidence of HPV-related non-cervical cancers has recently increased, particularly for carcinomas of the anus and oropharynx.^53,73 Therefore, publication(s) that have also considered the increased patterns in HPV-related disease burden provided results that were more favorable to the inclusion of males in national immunization programs.