key: cord-0328024-w4vsxvtq
authors: Prem, K.; Choi, Y. H.; Bampeacutenard, ampEacutelodie; Burger, E. A.; Hadley, L.; Laprise, J.-F.; Regan, C.; Drolet, M.; Sy, S.; Abbas, K.; Kim, J. J.; Brisson, M.; Jit, M.
title: Global impact and cost-effectiveness of one-dose versus two-dose human papillomavirus vaccination schedules: a comparative modelling analysis
date: 2021-02-08
journal: nan
DOI: 10.1101/2021.02.08.21251186
sha: 0d900738fc8c85343ad6b995a98a7dd62d2bedd2
doc_id: 328024
cord_uid: w4vsxvtq

Background: To eliminate cervical cancer as a public health problem, WHO currently recommends routine vaccination of adolescent girls with two doses of the human papillomavirus (HPV) vaccine before sexual debut. However, many countries have yet to implement this because of financial or logistical barriers to delivering two doses outside the infant immunisation programme. Methods: Using three independent HPV transmission models, we estimated the long-term health benefits and cost-effectiveness of one-dose versus two-dose HPV vaccination, in 192 countries, assuming that one dose of the vaccine gives either a shorter duration of full protection (20 or 30 years) or lifelong protection but lower vaccine efficacy (e.g., 80%) compared to two doses. We simulated routine vaccination with the 9-valent HPV vaccine in 10-year-old girls at 80% coverage for the years 2021-2120, with a one-year catch-up of 80% 11-14-year-old girls on the first year of the programme. Findings: Over the next century, one-dose vaccination at 80% coverage could avert 64 million (80%UI 62.2-64.8) and 66.6 million (80%UI 63.4-69.1) cervical cancer cases should one dose of the vaccine confer 20 and 30 years of protection, respectively. Should one dose of the vaccine provide lifelong protection at 80% vaccine efficacy, 68.4 million (80%UI 63.8-69.4) cervical cancer cases could be prevented. Across all country income groups, two-dose schedules conferring lifelong protection would avert only slightly more cases (2.1-8.7 million) than the one-dose scenarios explored. Around 330 to 5230 additional girls need to be vaccinated with the second dose to prevent one cervical cancer case, depending on the epidemiological profiles of the country. Interpretation: Results were consistent across the three independent models and suggest that one-dose vaccination has similar health benefits to a two-dose programme while simplifying vaccine delivery, reducing costs and alleviating vaccine supply constraints. Funding: Bill & Melinda Gates Foundation

Email: mark.jit@lshtm.ac.uk

To eliminate cervical cancer as a public health problem, WHO currently recommends routine vaccination of adolescent girls with two doses of the human papillomavirus (HPV) vaccine before sexual debut. However, many countries have yet to implement this because of financial or logistical barriers to delivering two doses outside the infant immunisation programme.

Using three independent HPV transmission models, we estimated the long-term health benefits and cost-effectiveness of one-dose versus two-dose HPV vaccination, in 192 countries, assuming that one dose of the vaccine gives either a shorter duration of full protection (20 or 30 years) or lifelong protection but lower vaccine efficacy (e.g., 80%) compared to two doses. We simulated routine vaccination with the 9-valent HPV vaccine in 10-year-old girls at 80% coverage for the years 2021-2120, with a one-year catch-up of 80% 11-14-year-old girls on the first year of the programme.

Over the next century, one-dose vaccination at 80% coverage could avert 64 million (80%UI 62·2-64·8) and 66·6 million (80%UI 63·4-69·1) cervical cancer cases should one dose of the vaccine confer 20 and 30 years of protection, respectively. Should one dose of the vaccine provide lifelong protection at 80% vaccine efficacy, 68·4 million (80%UI 63·8-69·4) cervical cancer cases could be prevented. Across all country income groups, two-dose schedules conferring lifelong protection would avert only slightly more cases (2·1-8·7 million) than the one-dose scenarios explored. Around 330 to 5230 additional girls need to be vaccinated with the second dose to prevent one cervical cancer case, depending on the epidemiological profiles of the country. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted February 8, 2021. ; https://doi.org/10.1101/2021.02.08.21251186 doi: medRxiv preprint

Results were consistent across the three independent models and suggest that one-dose vaccination has similar health benefits to a two-dose programme while simplifying vaccine delivery, reducing costs and alleviating vaccine supply constraints.

Evidence before this study Primary prevention of cervical cancer is now available with human papillomavirus (HPV) vaccination. Initially administered as a three-dose regimen, the HPV vaccine schedule recommended by WHO has now switched to two doses for individuals below the age of 15 years. Although WHO recommends all countries to routinely immunise adolescent girls with two doses, many low-and middle-income countries, with high disease burden, have yet to implement national HPV vaccination programmes because of the challenges of delivering two vaccine doses to adolescent females. Recently, HPV vaccine implementation in many countries has been further delayed due to constraints in vaccine supply and difficulties in access during COVID-19 epidemics. These financial, logistical, and supply constraints have motivated research into one-dose vaccination schedules. Evidence emerging from trials and observational studies suggests that one dose may also provide a high level of protection against incident and persistent HPV infections. If proven effective, the one-dose HPV vaccination schedule would simplify vaccine delivery and lower costs of national vaccination programmes, potentially enabling more countries to implement one and as a result, facilitating global cervical cancer prevention. We searched PubMed for trials, cohort and modelling studies published in 2018 and 2020, with the terms "(health impact OR impact OR modelling OR cost-effectiveness OR CEA OR durability OR effectiveness) AND (HPV OR human papillomavirus OR cervical cancer)" and identified 151 results. Ten published articles-four trials, three cohort studies, two modelling analyses, one systematic review of trials-evaluated the population impact of one dose of the vaccine on cervical cancer disease outcome among females and all studies showed one dose of the vaccine might be as is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted February 8, 2021. ; https://doi.org/10.1101/2021.02.08.21251186 doi: medRxiv preprint effective as two doses in preventing HPV infection. However as the trials and cohorts were single-country studies in select populations, the global impact remains unknown. Both published modelling analyses only used one model to estimate the impact of one-dose vaccination, and only examined a few countries. To our knowledge, no published article has modelled the global impact of routine one-dose vaccination on cervical cancer prevention by synthesising the results from more than one model.

This study presents the first evidence on the potential global impact of a routine one-dose regimen, from a comparative modelling analysis that synthesises results from three published dynamic models calibrated to countries with varying epidemiological and demographic profiles. We found consistent results across all models suggesting that routine one-dose vaccination provides the majority of health benefits to the two-dose programme should a single dose of the vaccine confer more than 20 years of protection at full potential efficacy or 80% efficacy with lifelong protection.

Findings suggest that routine one-dose vaccinations could avert almost as many cervical cancer cases as a two-dose programme. The one-dose regimen would be cheaper and easier to implement for most countries while alleviating vaccine supply constraints. To cope with the COVID-19 pandemic, many governments have had to implement stringent physical distancing measures, which has led to the suspension of routine immunisation programmes.

Public health authorities grapple with the logistic challenges of delivering immunisation services while minimising the risk of SARS-CoV-2 transmission. Compared to the two-dose vaccination schedule, a one-dose vaccination schedule would reduce interactions between vaccinees and health workers, simplifying vaccine delivery while also decreasing SARS-CoV-2 exposure.

. CC-BY-NC 4.0 International license It is made available under a perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

Cervical cancer is the fourth leading cause of cancer mortality among women globally with an estimated 570 000 new cases and 311 000 deaths in 2018, with the majority of deaths occurring in low-and middle-income countries (LMICs) (1) . Persistent infection with highrisk genotypes of human papillomavirus (HPV) is a necessary precursor of cervical cancer.

Primary prevention of cervical cancer is now available with HPV vaccination (2) . Four highly efficacious prophylactic vaccines-two 2-valent, one 4-valent, one 9-valent-are currently licensed for protection against HPV infection (3) (4) (5) (6) . All protect against the two most carcinogenic HPV types, 16 and 18, which are responsible for 70% of cervical cancer cases globally (7) (8) (9) . Some additionally protect against HPV types 6 and 11 which do not cause cancer but are responsible for most cases of anogenital warts, and against other high-risk types such as HPV 31, 33, 45, 52 and 58 (either directly or through cross-protection), which have been linked to a further 20% of cervical cancer cases (7) (8) (9) .

Economics (PRIME) model developed in collaboration with the World Health Organization (WHO) (10, 11) have found HPV vaccination to be cost-effective in almost all countries. The HPV vaccines were initially administered as a three-dose regimen over six months. In 2014, the WHO Strategic Advisory Group of Experts on Immunization (SAGE) reviewed the evidence for dose reduction and recommended a two-dose regimen for individuals below 15 years of age (12) . With the availability of vaccines and screening tests that allow detection of both high-risk HPV types and neoplasias that are precursors to cervical cancer, the Secretary-General of WHO has called for global elimination of cervical cancer as a public health problem (13) . Current WHO guidelines recommend that all countries vaccinate females aged 9-14 years against HPV (14) .

Although some of these vaccines have been licensed for more than a decade, LMICs which have the highest incidence of cervical cancer are disproportionately less likely to introduce the HPV vaccine into their routine immunisation programmes (10, (15) (16) (17) . High vaccine procurement and delivery costs coupled with logistical constraints surrounding the delivery is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted February 8, 2021. ; https://doi.org/10.1101/2021.02.08.21251186 doi: medRxiv preprint of a two-dose regimen outside the infant vaccination schedule has hampered vaccine introduction and uptake (18) . Despite the financial support of Gavi, the Vaccine Alliance, many LMICs have yet to introduce HPV vaccines into their routine programmes (19, 20) .

Since 2017, constrained supplies of the 4-valent and 9-valent HPV vaccines has further delayed vaccine introductions in many countries (21) . Moreover, physical distancing measures such as school closure or national lockdowns implemented to cope with the current COVID-19 pandemic (22) have caused eligible populations to miss doses of HPV vaccine.

These financial, logistical, and supply constraints have motivated research into one-dose vaccination schedules. If proven effective, the one-dose HPV vaccination schedule would simplify vaccine delivery and lower costs of national vaccination programmes (19) . It could also expedite the introduction of HPV vaccines into national immunisation schedules for LMICs, potentially protecting many more females against cervical cancer (20) .

Evidence is emerging from immunogenicity trials, post-hoc analyses of efficacy trials, and post-licensure observational studies to suggest that one dose of the HPV vaccine may provide a high level of protection against incident and persistent HPV infections.

A systematic review of participants in six clinical trials who received only one dose of HPV vaccination because they did not complete their allocated schedules, suggests that this schedule may be as effective as two doses in preventing HPV infection in up to seven years of follow-up (23) . However, evidence from participants randomised to receive one dose has yet to emerge. Furthermore, antibody titres in immunogenicity trials were lower than in those receiving two or three doses. While inferior antibody titres may not necessarily translate to inferior protection, at this point, there is still uncertainty about the efficacy and duration of one-dose vaccination.

Additionally, in the event that one-dose vaccination protection is slightly worse than two or three doses, populations may still be almost as well protected through indirect (herd) protection. Such effects can be examined using HPV transmission dynamic models. To date, model-based analyses set in the United Kingdom (24), the United States, and Uganda (25, 26) suggest that one-dose schedules would be cost-effective and would prevent almost is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted February 8, 2021. ; https://doi.org/10.1101/2021.02.08.21251186 doi: medRxiv preprint as many cancers as two-dose or three-dose schedules if one dose confers at least 20 years of protection or have at least 80% efficacy against HPV 16/18 infection.

In this paper, we compare the impact and cost-effectiveness of one-dose versus two-dose vaccination, in 192 countries, assuming that one dose of the vaccine gives either shorter duration of protection or lower vaccine efficacy compared to two doses. We use a hybrid approach: firstly, we consider the age-specific impact that HPV vaccines may have using the results of multiple independent HPV transmission dynamic models, and secondly, extrapolate these effects to the remaining countries in the world using data on population demographics and cervical cancer burden synthesised in a model (PRIME).

To assess the extent to which one-dose HPV vaccine schedules will provide similar protection and be cost-effective, we compared the impact of three different vaccine is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted February 8, 2021. ; https://doi.org/10.1101/2021.02.08.21251186 doi: medRxiv preprint Vaccination and Immunisation (32), the World Health Organization's Strategic Advisory Group of Experts on Immunization (12, 33, 34) and the US Advisory Committee on Immunization Practice (35) (36) (37) (38) . The models stratify population by age, gender and sexual activity-based risk group, and screening behaviour-based risk group in the HPV-ADVISE and Harvard models. They capture HPV natural history and disease, as well as HPV transmission as informed by country-specific sexual behaviour surveys. For the scenarios where one dose confers a shorter duration of protection (i.e., 20 or 30 years), we assume 100% VE (or take), as suggested by clinical trial populations (23) . We modelled routine annual vaccination with the 9-valent vaccine in 10-year-old girls to begin in 2021 and run uninterrupted until 2120.

We also included a catch-up of girls aged 11-14-year-old in the first year of the programme.

Throughout, vaccine coverage was assumed to be 80%.

Using PRIME, we then estimated the primary impact of a two-dose vaccination schedule, without herd effects and waning immunity, in 192 countries. Full details of PRIME, including model equations and updates, are available at (10, 11) . As PRIME is a static model, it cannot estimate herd effects, nor can it capture the effect of waning vaccine immunity. Here, we introduced a novel method which compares results from PRIME and the three dynamic models-PHE, HPV-ADVISE, Harvard-set in nine countries-UK, US, Canada, Nigeria, Uganda, India, Vietnam, El Salvador and Nicaragua. We calculated the difference between cervical cancer incidence predicted by PRIME and each of the dynamic models to derive the secondary effects of vaccination, which is a combination of waning immunity (20/30-year vs lifetime protection or lower vaccine take) and herd effects at every age and time-point. We then calculated the ratio of secondary to primary vaccine impact. By assuming that the primary impact of a vaccine (i.e. vaccine with lifetime protection and no herd effects) is different in every country as estimated by PRIME, we extrapolated the ratio (secondary to primary) to other countries to project the secondary effects of vaccination.

Uncertainty in predictions was captured by generating multiple simulations from the three dynamic models representing different plausible parameter sets. For the PHE model, 100 runs were simulated from the best fitting parameter sets to capture uncertainty in the duration of infection, duration of natural immunity, screening accuracy, the progression of cervical cancer, age-specific prevalence, and the number of sexual partners. For HPV- is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted February 8, 2021. ; https://doi.org/10.1101/2021.02.08.21251186 doi: medRxiv preprint ADVISE, 1000 runs were simulated from 50 parameter sets that simultaneously fit countryspecific behavioural and epidemiological data. These 50 parameter sets illustrate the uncertainty in sexual behaviour, HPV transmission, the natural history of HPV-related diseases and screening. For the Harvard model that reflect two sexual behaviour settings (low and high HPV prevalence), 50 best-fitting dynamic transmission model parameter sets, capturing variations in genotype-and sex-specific transmission probability (per month of a partnership duration), and genotype-and sex-specific natural immunity, were propagated through five cervical carcinogenesis models that have been previously calibrated (i.e., fit) to the United States, India, Uganda, El Salvador or Nicaragua (15, 31) .

For each country, we estimated the number of cervical cancer cases, deaths and disabilityadjusted life years (DALYs) -caused by HPV 16, 18, 31, 33, 45, 52, and 58-occurring under each scenario by time since vaccination and age. We then compared the impact of a onedose schedule (giving 20/30 years protection or lifelong protection but at 80% VE) with no vaccination, and a two-dose schedule (giving lifetime protection at 100% VE) with a onedose schedule. We estimated the number of females needed to vaccinate with one dose, and the number of females needed to give an additional (i.e. second) dose, to avert one cervical cancer case, death or DALY. We also projected the threshold cost to pay for the first and second dose of vaccine, which is the maximum that could be paid for the first dose (compared to no vaccination) and second dose (compared to one dose only) for the incremental cost-effectiveness ratio to remain below country-specific gross domestic product (GDP) per capita (in 2017 USD). We used the GDP per capita estimates by the World Bank (39), but also considered a lower threshold (than one GDP per capita) (40, 41) . The time horizon of the analysis was from 2020 to 2120; we accrued all health benefits of vaccination up to the end of the routine vaccination programme (i.e., the year 2120) or age 100 of all vaccinated cohorts. After projecting the various measures of effectiveness and cost-effectiveness under the several vaccination scenarios, we compared the outcomes generated with results from the 11 model-country pairs. After projecting the various measures of effectiveness and cost-effectiveness under the several vaccination scenarios, we compared the outcomes generated with results from the 11 model-country scenarios.

We presented the results, aggregated by World Bank income groups (details in the is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted February 8, 2021. ; https://doi.org/10.1101/2021.02.08.21251186 doi: medRxiv preprint appendix), as the median (and 80% uncertainty intervals (UI)) from each of the modelcountry predictions. Both health outcomes and costs were discounted at 0% and 3% (42) .

The funder of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report.

In 192 countries over the years 2021-2120, the models projected that routine annual is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted February 8, 2021. ; https://doi.org/10.1101/2021.02.08.21251186 doi: medRxiv preprint averted by a two-schedule vaccination schedule. However, the HPV-ADVISE and Harvard models, mostly parameterised with data from LMICs, projected that 93.8% (80%UI 92·1-95·0%) could be averted (Figure 3) .

The models consistently projected that fewer girls need to be vaccinated with the first dose to prevent one cervical cancer case in low-income countries 30 (80%UI 15-64) than middle- Across all income groups, the threshold (i.e. maximum) cost a country should pay for the second dose was low-from 1·20 (80%UI 0·70-2·40) USD in low-income countries to 22·30 (80%UI 14-32·40) USD in high-income countries if one-dose confers 20 year protection-as few cancers would be averted with a longer duration of protection (>20 or 30 years) or higher vaccine efficacy (>80%). With a higher GDP per capita, middle-and high-income countries have a higher threshold cost (Figure 5 ).

In this study, three independent transmission dynamic models projected consistent results suggesting that routine one-dose HPV vaccine programmes at 80% coverage worldwide could provide a high-level of population protection and be cost-effective. We is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted February 8, 2021. ; https://doi.org/10.1101/2021.02.08.21251186 doi: medRxiv preprint significant potential public health impact of the one-dose vaccination schedule if vaccine uptake is high across all countries. Our model-based analysis predicts that routinely vaccinating 10-year-old girls at 80% coverage in LICs/LMICs could result in four times (population-adjusted) more cervical cases averted than in high-income countries. Under our one-dose assumptions, routine one-dose HPV vaccination programmes would protect up to 68·4 million females against cervical cancer over the years 2021-2120.

Our comparison of one-and two-dose vaccine schedules is motivated by several advantages of a one-dose schedule. Firstly, many LMICs have yet to implement national HPV vaccination programmes because of the challenges of delivering two vaccine doses to adolescent females (18) . Compared to the two-dose HPV vaccination schedule, a one-dose HPV vaccination schedule would be cheaper and easier to implement, potentially enabling more LMICs to introduce the HPV vaccines into national immunisation schedules. More recently, HPV vaccine implementation in LMICs has been delayed due to constraints in HPV vaccine supply.

Secondly, the COVID-19 pandemic has disrupted several routine immunisation programmes (43) (44) (45) , including HPV vaccination (46) . Abbas and colleagues predicted that the benefits of resuming routine childhood immunisation services outweigh the risk of being infected with COVID-19 during the vaccination visits (45) , reinforcing WHO's call for all countries to continue routine immunisation services safely (47) . With physical distancing measures such as school closures or national lockdowns being implemented in many countries to cope with the COVID-19 pandemic (22) , health officials grapple with reconfiguring school-based HPV vaccine delivery (46) (47) (48) . Compared to the two-dose vaccination schedule, a one-dose schedule would further minimise interactions between vaccinees and health workers, simplifying vaccine delivery while also decreasing SARS-CoV-2 exposure.

The lack of country-specific behavioural, virological and clinical data in many countries limits fitting transmission dynamic models individually to most countries (49) . However, in this comparative modelling study, we synthesised results from three published dynamic models based in nine countries, covering high-, middle-and low-income settings across three continents and a wide variety of epidemiological characteristics for HPV transmission and is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted February 8, 2021. ; https://doi.org/10.1101/2021.02.08.21251186 doi: medRxiv preprint cervical cancer (15, 49) . Our approach provides a common framework using PRIME for population demographics, cervical cancer burden and impact/cost-effectiveness calculations, while varying representation of HPV transmission and cervical cancer natural history across the three dynamic models. We then extrapolate the age-and time-dependent ratio of the secondary to primary impacts of vaccine strategies to other countries. While there may be considerable uncertainty around extrapolating this ratio to another country, the use of 11 model-country pairs lends confidence that we are likely to have captured the range of possible outcomes for most countries. More precise estimates would require fitting these models to specific countries (49, 50) .

Our model projections of vaccine impact also involve other sources of uncertainty that we did not explicitly quantify. The PRIME model uses country-specific cervical cancer burden from the Global Cancer Incidence, Mortality and Prevalence (GLOBOCAN) database (51), which may underestimate the full burden of HPV-related disease, and thus vaccine impact, in LMICs (15) . In this study, we only assessed the effect of HPV vaccination on cervical cancers. If we also accounted for the vaccine impact on other HPV-related cancers, we would anticipate a greater value of HPV vaccination programmes (25, 52) . However, the paucity of data on the efficacy of one-dose on non-cervical cancers complicates the analysis evaluating their vaccine impact. Finally, we project the impact of HPV vaccination on cervical cancers over the next century. Over the past decades, we have witnessed substantial demographic (53) and behavioural changes (54, 55) with extraordinary improvements in public health (56) . In 2020, the COVID-19 pandemic has caused substantial disruptions to population demography (57) and sexual behaviour (58) , with uncertainty around the longerterm consequences of such disruption. Moreover, over the next century, we may see substantial advancements to pre-cancer screening and treatment services which will further decreases cervical cancer incidence. Such uncertainties in life expectancy, population and economic forecasts have significant implications for our predictions. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted February 8, 2021. ; https://doi.org/10.1101/2021.02.08.21251186 doi: medRxiv preprint

Under the scenarios where a single HPV vaccine dose confers more than 20 years of protection or 80% efficacy with lifelong protection, routine one-dose HPV vaccination provides the majority of health benefits to the two-dose programme while simplifying vaccine delivery, reducing costs and alleviating vaccine supply constraints. These results are fairly consistent when projected from three independent transmission dynamic models used in nine countries. The outcomes of our comparative modelling analysis contribute to the extensive evidence base, including emerging evidence from the single-dose HPV vaccine trials and observational studies, which would be beneficial to policymakers when they consider HPV vaccination in their populations. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted February 8, 2021. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint To compare the impact and costeffectiveness of one-dose versus two-dose vaccination in 192 countries, we adopted a hybrid approach. First, we synthesised the age-specific impact of HPV vaccines of three published transmission dynamic models-PHE, HPV-ADVISE, Harvard-from 11 model-country settings. Second, we derive the primary impact of vaccination using a static model (PRIME). Third, we extrapolate the primary and secondary effects to the remaining countries in the world. Fourth, we measure and compare population-level impact (e.g., cervical cancers averted, number of females needed to be vaccinated, threshold costs of the first and second dose of the vaccine) for three vaccine strategies: the counterfactual, no HPV vaccination; a one-dose HPV vaccination schedule in which we assume that one dose of the vaccine gives either a shorter duration of protection (20 or 30 years) or lower vaccine efficacy (e.g., 80%) compared to two doses; and a two-dose HPV vaccination schedule in which two doses of the vaccine would provide lifetime protection. The threshold cost is the maximum that could be paid for the first dose (compared to no vaccination) and second dose (compared to one dose only) for the incremental cost-effectiveness ratio to remain below the costeffectiveness threshold. Two cost-effectiveness thresholds are presented: country gross domestic product (GDP) per capita (in 2017 USD) costs in panels A-D and a lower threshold as suggested by Jit (2020). The lower cost-effectiveness threshold presented in panels E-H is 30-40% and 60-65% of GDP per capita in low-income and middle-to high-income countries, respectively. Cost and health outcomes are discounted at 3% and 0%, respectively.

. CC-BY-NC 4.0 International license It is made available under a perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted February 8, 2021. ; Figure 1 . Overview of the data sources and the key steps of the modelling. To compare the impact and cost-effectiveness of one-dose versus two-dose vaccination in 192 countries, we adopted a hybrid approach. First, we synthesised the age-specific impact of HPV vaccines of three published transmission dynamic models-PHE, HPV-ADVISE, Harvard-from 11 model-country settings. Second, we derive the primary impact of vaccination using a static model (PRIME). Third, we extrapolate the primary and secondary effects to the remaining countries in the world. Fourth, we measure and compare population-level impact (e.g., cervical cancers averted, number of females needed to be vaccinated, threshold costs of the first and second dose of the vaccine) for three vaccine strategies: the counterfactual, no HPV vaccination; a one-dose HPV vaccination schedule in which we assume that one dose of the vaccine gives either a shorter duration of protection (20 or 30 years) or lower vaccine efficacy (e.g., 80%) compared to two doses; and a two-dose HPV vaccination schedule in which two doses of the vaccine would provide lifetime protection. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted February 8, 2021. ; https://doi.org/10.1101/2021.02.08.21251186 doi: medRxiv preprint Figure 3 . The proportion of cervical cancers averted by routine two-dose HPV vaccination programmes with a perfect vaccine (i.e., 100% vaccine efficacy) conferring lifelong protection that may still occur under a routine one-dose schedule. The median percentage (intervals: 10-90th percentile) of cancers not averted by a one-dose schedule compared to a two-dose program of the 11 model-country settings: the PHE model in black, HPV-ADVISE model-country pairs in red, and the Harvard model-country pairs in blue. Health outcomes are discounted at 0%. Only cervical cancer caused by HPV 16, 18, 31, 33, 45, 52 and 58, which could be averted by the 9-valent HPV vaccine, were considered. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted February 8, 2021. ; https://doi.org/10.1101/2021.02.08.21251186 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted February 8, 2021. ; https://doi.org/10.1101/2021.02.08.21251186 doi: medRxiv preprint Figure 5 . Threshold cost to pay for the first and second dose of vaccine by country income groups. The threshold cost is the maximum that could be paid for the first dose (compared to no vaccination) and second dose (compared to one dose only) for the incremental cost-effectiveness ratio to remain below the cost-effectiveness threshold. Two cost-effectiveness thresholds are presented: country gross domestic product (GDP) per capita (in 2017 USD) costs in panels A-D and a lower threshold as suggested by Jit (2020). The lower cost-effectiveness threshold presented in panels E-H is 30-40% and 60-65% of GDP per capita in low-income and middle-to high-income countries, respectively. Cost and health outcomes are discounted at 3% and 0%, respectively. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted February 8, 2021. ; https://doi.org/10.1101/2021.02.08.21251186 doi: medRxiv preprint

Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries

Primary Prevention of Cervical Cancer: American Society of Clinical Oncology Resource-Stratified Guideline

Quadrivalent Vaccine against Human Papillomavirus to Prevent Anogenital Diseases

Efficacy of human papillomavirus (HPV)-16/18 AS04-adjuvanted vaccine against cervical infection and precancer caused by oncogenic HPV types (PATRICIA): final analysis of a double-blind, randomised study in young women

A 9-Valent HPV Vaccine against Infection and Intraepithelial Neoplasia in Women

Efficacy, Safety, and Immunogenicity of an Escherichia coli-Produced Bivalent Human Papillomavirus Vaccine: An Interim Analysis of a Randomized Clinical Trial

Human papillomavirus genotype attribution in invasive cervical cancer: a retrospective cross-sectional worldwide study

Potential impact of a nine-valent vaccine in human papillomavirus related cervical disease. Infect Agent Cancer

Global burden of cancer attributable to infections in 2018: a worldwide incidence analysis. Lancet Glob Heal

Cost-effectiveness of female human papillomavirus vaccination in 179 countries: A PRIME modelling study

Effects of updated demography, disability weights, and cervical cancer burden on estimates of human papillomavirus vaccination impact at the global, regional, and national levels: a PRIME modelling study

Human papillomavirus vaccines: WHO position paper

World Health Organization. WHO | WHO leads the way towards the elimination of cervical cancer as a public health concern. World Health Organization

Impact of scaled up human papillomavirus vaccination and cervical screening and the potential for global elimination of cervical cancer in 181 countries, 2020-99: a modelling study

Global HPV Vaccine Introduction Overview: projected and current national introductions, demonstration/pilot projects, gender-neutral vaccination programs, and global HPV vaccine introduction maps

WHO. Immunization, vaccines and biologicals: data, statistics and graphics. WHO [Internet

Status of HPV vaccine introduction and barriers to country uptake. Vaccine

Progress in HPV vaccination in low-and lower-middle-income countries

Prioritisation of the human papillomavirus vaccine in a time of constrained supply

World Health Organization. Global market study: HPV vaccines. World Health Organization

School closure and management practices during coronavirus outbreaks including COVID-19: a rapid systematic review. Lancet Child Adolesc Heal

Efficacy and immunogenicity of a single dose of human papillomavirus vaccine compared to no vaccination or standard three and two-dose vaccination regimens: A systematic review of evidence from clinical trials

Comparison of two dose and three dose human papillomavirus vaccine schedules: Cost effectiveness analysis based on transmission model

Health and economic benefits of singledose HPV vaccination in a Gavi-eligible country. Vaccine

Optimal human papillomavirus (HPV) vaccination strategies to prevent cervical cancer in low-and middle-income countries in the context of limited resources: A mathematical modeling analysis

Transmission dynamic modelling of the impact of human papillomavirus vaccination in the United Kingdom

Population-level impact of the bivalent, quadrivalent, and nonavalent human papillomavirus vaccines: A model-based analysis

Technical Appendix HPV-ADVISE Canada

Projected time to elimination of cervical cancer in the USA: a comparative modelling study

The United Kingdom Joint Committee on Vaccination and Immunisation. Vaccine

Weekly Epidemiological Record= Relevé épidémiologique hebdomadaire

-conclusions and recommendations

Effectiveness and cost-effectiveness of human papillomavirus vaccination through age 45 years in the United States

Health and Economic Impact of Switching from a 4-Valent to a 9-Valent HPV Vaccination Program in the United States

Impact and Cost-effectiveness of 3 Doses of 9-Valent Human Papillomavirus (HPV) vaccine among US females previously vaccinated with 4-valent hpv vaccine

Comparison of 2-dose and 3-dose 9-valent human papillomavirus vaccine schedules in the United States: A cost-effectiveness analysis

GDP per capita (current US$) | Data

Informing Global Cost-Effectiveness Thresholds Using Country Investment Decisions: Human Papillomavirus Vaccine Introductions in 2006-2018. Value Heal

Assessing the value of human papillomavirus vaccination in Gavi-eligible low-income and middle-income countries

World Health Organization. WHO Guide on Standardization of Economic Evaluations of Immunization Programmes

Pulse survey on continuity of essential health services during the COVID-19 pandemic: interim report

COVID-19 disrupts vaccine delivery

Routine childhood immunisation during the COVID-19 pandemic in Africa: a benefit-risk analysis of health benefits versus excess risk of SARS-CoV-2 infection

Human papillomavirus (HPV) vaccine: supply and demand update | UNICEF Supply Division

World Health Organization. WHO and UNICEF warn of a decline in vaccinations during COVID-19

The potential impact of COVID-19 pandemic on the immunization performance in Indonesia

Economic analyses to support decisions about HPV vaccination in low-and middle-income countries: A consensus report and guide for analysts

Available from: /pmc/articles/PMC2769256/?report=abstract 51. International Agency for Research on Cancer. Global Cancer Observatory

HPV prophylactic vaccines and the potential prevention of noncervical cancers in both men and women

United Nations Department of Economic and Social Affairs Population Division. World Population Prospects

Human papillomavirus vaccination at a time of changing sexual behavior

Sexual behaviour in context: a global perspective

The role of public health improvements in health advances: The twentieth-century United States

Demographic perspectives on the mortality of COVID-19 and other epidemics

Impact of the COVID-19 pandemic on the sexual behavior of the population. The vision of the east and the west

YHC led the PHE analysis, MB led the HPV-ADVISE analysis, JJK led the Harvard analysis, and KP led the combined analysis. CR, KP and JJK did the literature searches

Financial support for this project was provided by PATH on behalf of the Single-Dose HPV Vaccine Evaluation Consortium which includes Harvard University 

We declare no competing interests.

All analysis codes are available at https://github.com/kieshaprem/hpv-1-dose.