MGH-Cervical (MGH) Massachusetts General Hospital

Model Overview

The MGH model is an open-cohort, age-structured, compartmental, deterministic human papillomavirus (HPV) transmission model stratified by HPV-16/18 and other high-risk HPV types. It also includes HIV infection dynamics, cervical carcinogenesis, HPV vaccination and cervical cancer screening.

Dynamic HPV and HIV transmission

Our model simulates the transmission of HPV and HIV across the population, which is divided into various compartments based on health states (ex. HPV-Infected, HIV-positive, CD4 count etc.) and demographic factors (age, gender, sexual risk level). The rate of infection that a susceptible compartment is subject to depends on: a) the mixing patterns corresponding to the gender, age, and sexual risk level used to describe the compartment, and b) the prevalence of infection in other compartments that interact with the susceptible compartment.

HIV progression and ART scale up

HIV progression is simulated by modelling the rate of progression through HIV disease states as described by CD4 count and viral load level.^1-3 ART scale-up is modeled by representing the rate of ART uptake as a function of time and CD4 count to reflect the reported clinical criteria for ART uptake. In the model, ART uptake reduces the probability of HIV transmission and attenuates a HIV-positive person’s rate of progression through the HPV and cervical pre-cancer pathway.^4-6

Cervical carcinogenesis

Women and men enter the susceptible pool upon sexual debut, and with each partnership, face a risk of HPV acquisition depending on the number of new partners, the prevalence of HPV in the population (generated “dynamically” each cycle in the model), and the probability of HPV transmission per susceptible-infected partnership. Women transition from the uninfected to the HPV-infected state, to low- (cervical intraepithelial neoplasia [CIN I]) and high-grade (CIN II and III) pre-cancer and cervical cancer. The model allows for hysterectomy (including reasons other than cervical cancer treatment) at any stage. Men transition between the susceptible and the HPV-infected state.

Women undergo routine screening and a proportion of those who test positive return for treatment. Depending on treatment type and success, women may develop immunity, regress to earlier disease stages, or continue progressing. Treatment effectiveness varies by HIV status, and some women experience persistent HPV infection. Those who clear HPV are assumed to develop temporary, partial immunity. Cervical cancer can be detected via screening or symptoms. Treatment options include hysterectomy, which is assumed to be curative, or other therapies. Mortality risk increases with cancer stage and is higher for women living with HIV, particularly at lower CD4 counts, though the impact of HIV on mortality decreases as cancer progresses. Treatment reduces mortality risk, and screening and treatment effectiveness are assumed to be the same for vaccine-targeted and non-vaccine HPV types. The model does not account for the recurrence of metastatic cancer.

HPV transmission, vaccination and screening

As described previously,^7-9 the model captures the transmission of HPV by estimating the force of infection, which is a function of sexual mixing (by age and activity class), the proportion of infected individuals in the population (generated internally in the model each cycle) and the HPV transmission probability.¹⁰ Vaccination and screening decrease the incidence of HPV according to demonstrated clinical efficacy, leading to the decrease in incidence of subsequent disease states. The screening function accounts for sensitivity and specificity of the screening tool, loss-to-follow-up, and treatment success. Screening and treatment return a proportion of individuals to the HPV-negative susceptible state, but a proportion experience treatment failure and remain in the disease compartment. The model is able to explore synergies between vaccination and screening, such as a decrease in recurrence of CIN after vaccination.

Calibration and validation

The model was calibrated using data on sexual behavior, national data for HPV seropositivity,^11-13 hysterectomy rates, and uptake of screening by age.¹⁴ HPV progression and regression rates converted to transition probabilities^9,15 were based on a review the literature.^16-17 Independent data on age-specific cervical cancer incidence rates over time were used to validate the model.

References

1. Hubert JB, Burgard M, Dussaix E, et al. Natural history of serum HIV-1 RNA levels in 330 patients with a known date of infection. The SEROCO Study Group. AIDS 2000; 14(2): 123-31.
2. Lodi S, Phillips A, Touloumi G, et al. Time from human immunodeficiency virus seroconversion to reaching CD4+ cell count thresholds <200, <350, and <500 Cells/mm(3): assessment of need following changes in treatment guidelines. Clin Infect Dis 2011; 53(8): 817-25.
3. Lyles RH, Munoz A, Yamashita TE, et al. Natural history of human immunodeficiency virus type 1 viremia after seroconversion and proximal to AIDS in a large cohort of homosexual men. Multicenter AIDS Cohort Study. J Infect Dis 2000; 181(3): 872-80.
4. Simbayi LC ZK, Zungu N, Moyo S, Marinda E, Jooste S, Mabaso M, Ramlagan S, North A, van Zyl J, Mohlabane N, Dietrich C, Naidoo I, SABSSM V Team. The Fifth South African National HIV Prevalence, Incidence, Behaviour and Communications Survey, 2017. Cape Town, 2019.
5. Rodger AJ, Cambiano V, Bruun T, et al. Sexual Activity Without Condoms and Risk of HIV Transmission in Serodifferent Couples When the HIV-Positive Partner Is Using Suppressive Antiretroviral Therapy. JAMA 2016; 316(2): 171-81.
6. Kharsany ABM, Cawood C, Lewis L, et al. Trends in HIV Prevention, Treatment, and Incidence in a Hyperendemic Area of KwaZulu-Natal, South Africa. JAMA Netw Open 2019; 2(11): e1914378.
7. Barnabas RV, Laukkanen P, Koskela P, Kontula O, Lehtinen M, Garnett GP. Epidemiology of HPV 16 and cervical cancer in Finland and the potential impact of vaccination: mathematical modelling analyses. PLoS Med. 2006;3(5):e138. [Abstract]
8. Tan N, Sharma M, Winer R, Galloway D, Rees H, Barnabas RV. Model-estimated effectiveness of single dose 9-valent HPV vaccination for HIV-positive and HIV-negative females in South Africa. Vaccine. 2018 Aug;36(32):4830–6. [Abstract]
9. Rao DW, Bayer CJ, Liu G, Admire Chikandiwa, Sharma M, Hathaway CL, et al. Modelling cervical cancer elimination using single‐visit screening and treatment strategies in the context of high HIV prevalence: estimates for KwaZulu‐Natal, South Africa. Journal of the International AIDS Society. 2022 Oct 1;25(10). [Abstract]
10. Barnabas RV, Brown ER, Onono MA, Bukusi EA, Njoroge B, Winer RL, et al. Efficacy of single-dose HPV vaccination among young African women. NEJM evidence [Internet]. 2022 Jun 1;1(5):EVIDoa2100056. [Abstract]
11. Garnett GP, Anderson RM. Sexually transmitted diseases and sexual behavior: insights from mathematical models. J Infect.Dis. 1996;174 Suppl 2:S150-S61. [Abstract]
12. Denny L, Adewole I, Anorlu R, et al. Human papillomavirus prevalence and type distribution in invasive cervical cancer in sub-Saharan Africa. Int J Cancer 2014; 134(6): 1389-98.
13. Mbulawa ZZ, Coetzee D, Williamson AL. Human papillomavirus prevalence in South African women and men according to age and human immunodeficiency virus status. BMC Infect Dis 2015; 15: 459.
14. Liu G, Sharma M, Tan N, Barnabas RV. HIV-positive women have higher risk of human papilloma virus infection, precancerous lesions, and cervical cancer. AIDS 2018; 32(6): 795-808
15. Dillner J, Kallings I, Brihmer C, Sikstrom B, Koskela P, Lehtinen M, et al. Seropositivities to human papillomavirus types 16, 18, or 33 capsids and to Chlamydia trachomatis are markers of sexual behavior. J Infect Dis. 1996;173(6):1394-8. [Abstract]
16. Miller DK, Homan SM. Determining transition probabilities: confusion and suggestions. Med Decis Making. 1994;14(1):52-8. [Abstract]
17. Myers ER, McCrory DC, Nanda K, Bastian L, Matchar DB. Mathematical model for the natural history of human papillomavirus infection and cervical carcinogenesis. Am J Epidemiol. 2000;151(12):1158-71. [Abstract]