A Flexible Dynamic Transmission Model Framework for Evaluating Tuberculosis Treatments: Methodological Challenges and Applications
Author(s)
Hema K. Gandhi, PhD, MPH1, Maia Mileva-Lopez, MA2, Norbert Hittel, MD, PhD2, Karam Diaby, PhD1, Onn Min Kon, MD FRCP3, Rosanna C. Barnard, PhD, MMath4, Vasily Lukyanov, MS4, David Bointon, MSc, BSc4, Richard Pitman, PhD4;
1Otsuka US, Princeton, NJ, USA, 2Otsuka Novel Products GmbH, Munich, Germany, 3Imperial College Healthcare NHS Trust, London, United Kingdom, 4ICON Clinical Research Ltd, Dublin, Ireland
1Otsuka US, Princeton, NJ, USA, 2Otsuka Novel Products GmbH, Munich, Germany, 3Imperial College Healthcare NHS Trust, London, United Kingdom, 4ICON Clinical Research Ltd, Dublin, Ireland
Presentation Documents
OBJECTIVES: The transmissible nature of infectious diseases such as tuberculosis (TB), and the dynamic cycle of susceptibility and infection necessitate economic modeling capable of incorporating direct and indirect effects of treatments and disease control. We present a dynamic transmission model (DTM) framework for evaluating TB treatment regimens. The DTM was designed to simulate TB transmission, calibrated using notification and mortality data from 1950-2021 in England and Wales.
METHODS: The model incorporates local transmission dynamics and latent TB infections introduced to the population through net migration. Key methodological challenges included calibrating to historical trends, capturing temporal variation in imported latent infections, and integrating clinical trial data for novel treatments with endpoints such as sputum culture conversion (SCC), AEs, tolerability, and final treatment outcomes. Solutions included parameterizing disease state transitions to reflect non-inferiority scenarios and dynamic calibration to migration-adjusted prevalence.
RESULTS: The DTM successfully replicates historical trends in TB incidence and mortality while accommodating novel treatment scenarios. Using England and Wales as a case, we demonstrate that a treatment for multi-drug resistant (MDR) TB reducing time to sputum conversion by 50% can lower MDR-TB incident cases by approximately 28% over a century. Our calibrated model suggests that the TB epidemic in England and Wales is sustained by the introduction of latent TB infections acquired overseas that subsequently activate and contribute to TB transmission. Sensitivity analyses identified importations of latent infections (introduced to the population through net migration) and time to sputum conversion as key drivers of transmission dynamics. Addressing these methodological challenges has ensured the framework’s adaptability for different geographies and treatment modalities.
CONCLUSIONS: This methodological framework offers a robust approach for evaluating TB treatments in dynamic settings, accommodating both localized and global transmission drivers. Its versatility makes it suitable for modeling treatments with diverse clinical profiles, offering valuable insights for policymakers and health economists.
METHODS: The model incorporates local transmission dynamics and latent TB infections introduced to the population through net migration. Key methodological challenges included calibrating to historical trends, capturing temporal variation in imported latent infections, and integrating clinical trial data for novel treatments with endpoints such as sputum culture conversion (SCC), AEs, tolerability, and final treatment outcomes. Solutions included parameterizing disease state transitions to reflect non-inferiority scenarios and dynamic calibration to migration-adjusted prevalence.
RESULTS: The DTM successfully replicates historical trends in TB incidence and mortality while accommodating novel treatment scenarios. Using England and Wales as a case, we demonstrate that a treatment for multi-drug resistant (MDR) TB reducing time to sputum conversion by 50% can lower MDR-TB incident cases by approximately 28% over a century. Our calibrated model suggests that the TB epidemic in England and Wales is sustained by the introduction of latent TB infections acquired overseas that subsequently activate and contribute to TB transmission. Sensitivity analyses identified importations of latent infections (introduced to the population through net migration) and time to sputum conversion as key drivers of transmission dynamics. Addressing these methodological challenges has ensured the framework’s adaptability for different geographies and treatment modalities.
CONCLUSIONS: This methodological framework offers a robust approach for evaluating TB treatments in dynamic settings, accommodating both localized and global transmission drivers. Its versatility makes it suitable for modeling treatments with diverse clinical profiles, offering valuable insights for policymakers and health economists.
Conference/Value in Health Info
2025-05, ISPOR 2025, Montréal, Quebec, CA
Value in Health, Volume 28, Issue S1
Code
MSR76
Topic
Methodological & Statistical Research
Disease
No Additional Disease & Conditions/Specialized Treatment Areas, SDC: Infectious Disease (non-vaccine), STA: Multiple/Other Specialized Treatments