optimal human papillomavirus vaccination strategies to prevent cervical cancer in CORD-Papers-2022-06-02 (Version 1)

Title: Optimal human papillomavirus vaccination strategies to prevent cervical cancer in low-income and middle-income countries in the context of limited resources: a mathematical modelling analysis
Abstract: BACKGROUND: Introduction of human papillomavirus (HPV) vaccination has been slow in low-income and middle-income countries (LMICs) because of resource constraints and worldwide shortage of vaccine supplies. To help inform WHO recommendations we modelled various HPV vaccination strategies to examine the optimal use of limited vaccine supplies and best allocation of scarce resources in LMICs in the context of the WHO global call to eliminate cervical cancer as a public health problem. METHODS: In this mathematical modelling analysis we developed HPV-ADVISE LMIC a transmission-dynamic model of HPV infection and diseases calibrated to four LMICs: India Vietnam Uganda and Nigeria. For different vaccination strategies that encompassed use of a nine-valent vaccine (or a two-valent or four-valent vaccine assuming high cross-protection) we estimated three outcomes: reduction in the age-standardised rate of cervical cancer number of doses needed to prevent one case of cervical cancer (NNV; as a measure of efficiency) and the incremental cost-effectiveness ratio (ICER; in 2017 international $ per disability-adjusted life-year [DALY] averted). We examined different vaccination strategies by varying the ages of routine HPV vaccination and number of age cohorts vaccinated the population targeted and the number of doses used. In our base case we assumed 100% lifetime protection against HPV-16 HPV-18 HPV-31 HPV-33 HPV-45 HPV-52 and HPV-58; vaccination coverage of 80%; and a time horizon of 100 years. For the cost-effectiveness analysis we used a 3% discount rate. Elimination of cervical cancer was defined as an age-standardised incidence of less than four cases per 100 000 woman-years. FINDINGS: We predicted that HPV vaccination could lead to cervical cancer elimination in Vietnam India and Nigeria but not in Uganda. Compared with no vaccination strategies that involved vaccinating girls aged 914 years with two doses were predicted to be the most efficient and cost-effective in all four LMICs. NNV ranged from 78 to 381 and ICER ranged from $28 per DALY averted to $1406 per DALY averted depending on the country. The most efficient and cost-effective strategies were routine vaccination of girls aged 14 years with or without a later switch to routine vaccination of girls aged 9 years and routine vaccination of girls aged 9 years with a 5-year extended interval between doses and a catch-up programme at age 14 years. Vaccinating boys (aged 914 years) or women aged 18 years or older resulted in substantially higher NNVs and ICERs. INTERPRETATION: We identified two strategies that could maximise efforts to prevent cervical cancer in LMICs given constraints on vaccine supplies and costs and that would allow a maximum of LMICs to introduce HPV vaccination. FUNDING: World Health Organization Canadian Institute of Health Research Fonds de recherche du QubecSant Compute Canada PATH and The Bill & Melinda Gates Foundation. TRANSLATIONS: For the French and Spanish translations of the abstract see Supplementary Materials section.
Published: 2021-11-03
Journal: Lancet Infect Dis
DOI: 10.1016/s1473-3099(20)30860-4
DOI_URL: http://doi.org/10.1016/s1473-3099(20)30860-4
Author Name: Drolet Mlanie
Author link: https://covid19-data.nist.gov/pid/rest/local/author/drolet_mlanie
Author Name: Laprise Jean Franois
Author link: https://covid19-data.nist.gov/pid/rest/local/author/laprise_jean_franois
Author Name: Martin Dave
Author link: https://covid19-data.nist.gov/pid/rest/local/author/martin_dave
Author Name: Jit Mark
Author link: https://covid19-data.nist.gov/pid/rest/local/author/jit_mark
Author Name: Bnard lodie
Author link: https://covid19-data.nist.gov/pid/rest/local/author/bnard_lodie
Author Name: Gingras Guillaume
Author link: https://covid19-data.nist.gov/pid/rest/local/author/gingras_guillaume
Author Name: Boily Marie Claude
Author link: https://covid19-data.nist.gov/pid/rest/local/author/boily_marie_claude
Author Name: Alary Michel
Author link: https://covid19-data.nist.gov/pid/rest/local/author/alary_michel
Author Name: Baussano Iacopo
Author link: https://covid19-data.nist.gov/pid/rest/local/author/baussano_iacopo
Author Name: Hutubessy Raymond
Author link: https://covid19-data.nist.gov/pid/rest/local/author/hutubessy_raymond
Author Name: Brisson Marc
Author link: https://covid19-data.nist.gov/pid/rest/local/author/brisson_marc
sha: 8456316972a9a81b5f60faa76f60c03b6f4ab340
license: cc-by
license_url: https://creativecommons.org/licenses/by/4.0/
source_x: PMC
source_x_url: https://www.ncbi.nlm.nih.gov/pubmed/
pubmed_id: 34245682
pubmed_id_url: https://www.ncbi.nlm.nih.gov/pubmed/34245682
pmcid: PMC8554391
pmcid_url: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8554391
url: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8554391/
has_full_text: TRUE
Keywords Extracted from Text Content: L3 (FSW precancerous lesions cervical cytology Figure A15 grade lesions cancer L3 men MeSH Martin-Hirsch Sanjosé HRNC Figure A18 L0-L2 women ... men L0=men ∈ {0 HRNC Bayo Beachler 2016 115 Figure A31 Lin invasive cancer ±10 cervical cancer Da Fomo L2=men biopsies CIN2 Statcompiler £ Figure A23 biopsy HPV-18 Figure A25 L0 women FSW Figure A1 Human papillomavirus HRNC Figure A21 HR Figure A17 B) Vietnam, cervix Health Survey HPV HPV-16 Women Figure A19 CIN3 acetic acid Figure A21 A20 CIN1 A12 cervical cancers Red women CIN Figure A18 woman ICO L1=men L1 cervical intraepithelial lesions HPV-ADVISE susceptible/immune Figure A9 10 ′ invasive cervical cancer Figure A28 Cervical Cancer HR Figure A20 Nagaddya matrix US Figure A32 HPV-ADVISE Canada 4 Figure A13 HR Figure A19 A10 ∈ {1 Figure A10 L3=men Boily 1991 7 Figure A26 LMICs Papillomavirus infections
Extracted Text Content in Record: First 5000 Characters:The model is based on a stochastic pair formation and separation process, which represents the underlying structure of the sexual contact pattern. We model sequential monogamous stable and casual (instantaneous) partnerships, as well as casual sexual partnerships between FSW and men in stable partnerships or single. The partnership formation and separation process is driven by females. Each woman has an associated age and level of sexual activity specific rate of either forming a new partnership if they are single, or separating if they are currently involved in a stable partnership. When a new partnership is formed, the male partner is selected according to an age and level of sexual activity specific mixing matrix, which reflects the preferences of a woman to form partnerships with men given their respective age and level of sexual activity (see section 1.2.3 for details on the mixing matrices). All newly formed partnerships have an age and level of sexual activity specific probability of being stable (see details in Section 2.2.2). The partnership formation rates of single females is derived from the partner acquisition rates and the age and level of sexual activity specific proportions of stable partnerships taking into account the proportions of individuals not available for partnership formation as follows: The sexual activity mixing matrix defines the probability that an individual of given gender and level of sexual activity forms a partnership with someone of the opposite gender with a given level of sexual activity. The matrix is computed as follows (Boily 1991 7 ): l l g l g g l a l l g 1991 7 ). The preference matrix is therefore defined as follows: assortative degree parameter (where > 1 represents assortative mixing, = 1 is proportionate mixing and < 1, disassortative mixing) We have 4 levels of sexual activity; the 4 th level (L3) represents women who are sex workers (FSW), and men who are their clients (L3 men can concurrently be in partnership with a FSW and a L0, L1 or L2 woman). By having such categories, we can directly parameterize the percent of the female population that are FSW, and percent of the male population that are clients based on observed data. The age mixing matrix is specific to each country. The mixing matrix, Λ , ′ , , , defines the probability that an individual of given gender ( ), age group ( ) and sexual activity level ( ) forms a partnership with someone of opposite gender ( ′ ) of a given age ( ′ ). This age mixing matrix is thus level of sexual activity-specific and was derived from observed data as explained in Section 2.2.2. Females The global mixing matrix is computed for females only, because the partnership formation and dissolution process is driven by females. The global matrix is computed as the Hadamard (elementwise) product of the mixing matrix by sexual activity level and the mixing matrix by age weighted by the male age-specific partner acquisition rates: Figure A1 ) and progress in the model to more severe stages of cervical intraepithelial lesions of grade 1 (CIN1), 2 (CIN2) or 3 (CIN3), and invasive cervical cancer (CC) of stage 1 (localized), stage 2 (regional) or stage 3 (distant). Women with CIN may also regress to a less severe stage or clear the infection and directly return to susceptible/immune status ( Figure A1 ). For transmission probabilities and clearance, progression and regression rates see Section 2.2.3. The mutually exclusive compartments represent the different HPV epidemiological states. Arrows represent the possible HPV-type, age, and gender specific transitions between these states for each individual. Each country has their own screening behavior. Upon entry in the simulated population, 10-year-old females are assigned a level of screening behavior based on the interval between two routine screening tests. Screening behavior is country specific. The levels of screening behavior range from a short interval between two routine screening tests ( = 0) to never being screened ( = 4). Please see Section 2.2.4 for the distribution of women assigned to each level of screening behavior. Different screening methods (ex., Pap or HPV testing) can be attributed to each woman (if screening is available in the population). CC screening initiation is determined by an age-specific rate (which is function of a woman's screening behavior). A screening interval is then attributed to each screening behavior level (see Section 2.2.4 for details). Depending on their true health states ( Figure A1) A calibration procedure is used to identify multiple parameter sets that simultaneously fit highlystratified sexual behavior and natural history data. We identified country-specific prior range of parameters and calibration data through a 4-step process. First, we extracted country-specific data from the Statcompiler of the DHS program to obtain standardized indicators of sexual activity across the 5 LMIC (mainly from the Demographic and Health Survey (D
Keywords Extracted from PMC Text: HPV vaccines4 52–56 appendix 3 [p 8 HPV-52 LMICs.1, 2, 3 individuals HICs Bill appendix 4 [pp 2, 25–26] 17–20 Consortium.30 HPV-33 girls-only 5–7 LMICs,5 INT$1 × GDP appendix 3 p 8) US$1 NNVs WHO.30 HPV vaccine supply.6 NNV LMICs.3 women).22 HPV-18 CIN1 non-vaccine LMICs.8, 9 appendix 4 p 22 HICs,17 people appendix 4 pp 14–16, 34–35 resolved.10 HPV Cancer.22 LMICs gender-neutral ≥15 appendix 4 p 24 Cervical Cancer appendix 4 pp 2 payer herd Vaccine IgG appendix 4 pp 25–26 appendix 4 pp 2, 25–26 appendix 3 p 21 years.31 CIN2 GDP × COVID-19 Human papillomavirus extended-interval women Papillomavirus cancer HPV-ADVISE Nigeria HPV-45 HPV vaccines appendix 4 HPV-31 cervical intraepithelial neoplasia ICER CIN3 appendix 4 p 4 HPV vaccine Uganda.12 cervical cancer positive.38 men appendix 3 HPV-16 HPV-58 appendix 4 p 2 appendix 4 pp 27–31 cervical Figure 2 HPV-related 6–7 appendix 4 p 5 USA;29 vaccinee?6 appendix 3 pp 6–51 appendix 4 pp 21–22 constraints,4 human appendix 4 pp 12–13, 34–35
Extracted PMC Text Content in Record: First 5000 Characters:Globally, in 2020, an estimated 604 000 new cases of cervical cancer and 342 000 cervical cancer-related deaths occurred.1 Over 80% of cases of cervical cancer occur in low-income and middle-income countries (LMICs).1 These inequalities in the burden of cervical cancer are set to increase, with 88% of high-income countries (HICs) having introduced HPV vaccination in women and girls as of the end of 2019 compared with less than 40% of LMICs.1, 2, 3 Furthermore, as of the end of 2019, 44% of high-income countries also vaccinate boys, compared with only 5% of LMICs.3 The reasons for the lower uptake of HPV vaccination in LMICs, which have been exacerbated by the COVID-19 pandemic, include financial and human resource constraints,4 paucity of evidence on its population-level effectiveness and cost-effectiveness in LMICs,5 and, importantly, worldwide shortage of HPV vaccine supply.6 Research in context Evidence before this study Approximately 500 000 new cases of cervical cancer are diagnosed in low-income and middle-income countries (LMICs) each year. Human papillomavirus (HPV) vaccines are highly effective: the Papillomavirus Rapid Interface for Modelling and Economics (PRIME) model predicted that HPV vaccination of girls aged 12 years with two doses of the vaccine was cost-effective in 156 (87%) of 179 countries. However, less than 40% of LMICs have introduced HPV vaccination programmes. Key barriers to introduction include financial and human resource constraints, and, since 2019, a worldwide shortage of HPV vaccine supply that might last until 2024. These barriers might have been intensified by the COVID-19 pandemic. In parallel, the WHO director-general has issued a global call to eliminate cervical cancer as a public health problem, which will result in sustained efforts to achieve high vaccination coverage across LMICs. Added value of this study In this modelling analysis we identified two novel HPV vaccination strategies that should minimise the number of doses needed to prevent one case of cervical cancer and the cost per DALY averted: two-dose routine vaccination of girls aged 14 years with or without a later switch to routine vaccination of girls aged 9 years, and routine vaccination of girls aged 9 years with an extended interval of 5 years between doses and a catch-up programme for girls aged 14 years. These strategies would maximise prevention of cervical cancer with the fewest doses in the short term and at the lowest cost, which would allow a maximum of LMICs to introduce HPV vaccination and could reduce the effect of HPV vaccine supply shortage on efforts to eliminate cervical cancer. Implication of all the available evidence Our modelling results have directly informed WHO Strategic Advisory Group of Experts on Immunization's recommendation in October, 2019, to continue vaccination with a two-dose schedule; temporarily postpone vaccination of multiple-age cohorts, older age groups (≥15 years), and gender-neutral vaccination; and consider implementing strategies such as those identified in our study. Because of the substantial burden and inequalities in the distribution of cervical cancer across the world, the WHO Director-General Tedros Adhanom Ghebreyesus made a global call for action towards the elimination of cervical cancer as a public health problem.7 This initiative would entail efforts to achieve high coverage of routine vaccination in girls, but could also include efforts to accelerate elimination by vaccinating multiple age cohorts of women, introducing gender-neutral vaccination (ie, vaccination of both boys and girls), and increasing uptake of cervical cancer screening. In the context of the call for elimination of cervical cancer and limited resources, HPV vaccination policy decisions in LMICs will probably require trade-offs between two potentially conflicting perspectives: maximising population-level impact (eg, to reach elimination of cervical cancer) versus optimising vaccination efficiency and return on investment (eg, to minimise the number of doses needed to prevent one case of cancer and to minimise the cost-effectiveness ratio). The main HPV vaccination policy questions being examined in LMICs by their health authorities and by WHO are as follows: what are the ages and the number of age cohorts that should be vaccinated? Should only girls or girls and boys be vaccinated? And how many and when should doses be given per vaccinee?6 To date, results from modelling studies have predicted that routine and multiple-age cohort HPV vaccination of girls is likely to be highly cost-effective in most LMICs.8, 9 However, most models for LMICs have not included realistic country-specific representation of sexual behaviour and natural history of HPV infection and cervical cancer and have not been designed to examine other, more complex policy questions (eg, adding vaccination of boys and different strategies among girls aged 9–14 years). Moreover, to our knowledge, no comprehensive mod
PDF JSON Files: document_parses/pdf_json/8456316972a9a81b5f60faa76f60c03b6f4ab340.json
PMC JSON Files: document_parses/pmc_json/PMC8554391.xml.json
G_ID: optimal_human_papillomavirus_vaccination_strategies_to_prevent_cervical_cancer_in