Challenges in treating infectious disease – linking epidemiology with pharmacometrics

Aims: In pharmacometric analyses, a goal is to quantify the impact of a therapeutic agent on a population of patients with a disease who will take the drug, with a particular focus on understanding the factors that contribute to individual variability that exists across patients within the population. In epidemiologic analyses, a goal is to quantify the primary risk factors that contribute to the spread of a disease across a population, similarly with a focus on understanding individual factors that may enhance the susceptibility or protection of members of the population to the disease. Combining these two methods presents an opportunity to understand how to get the right treatment intervention to the right patients at the right time to be able to impact the spread of infectious disease and potentially lead towards disease elimination.[1] In essence, this is the basis for the emerging area called Precision Public Health.

In Precision Public Health, the right treatment intervention may not only include using therapeutics but could also include other interventions that in total contribute to effective treatment of infectious disease. An illustrative example comes from efforts to treat malaria infection. A meta-analysis of individual subject level data from children who had participated in studies with the anti-malarial combination agent Arthemther-Lumafantrine uncovered an interesting and perhaps counterintuitive result.[2] It was noted that patients who were underweight for their age – a surrogate indicator of possible malnutrition- showed a higher rate of treatment failure than matched children who were the appropriate body weight for their age. These underweight children received the same mg dose as their normal weight counterparts and thus had a larger mg/kg dosage. Traditional pharmacometric analyses identified the primary covariate to identify which children more likely failed therapy. Epidemiological analyses pointed to factors that could be measured to address this problem using a simple method for assessing malnutrition in the field. A subsequent study using an intervention to address acute malnutrition then showed the ability to correct this problem which was ultimately due to poor absorption of the drug dose.

A second illustrative example is an effort by Bershteyn and Eckhoff to assess the impact of variation in host immunity along with patient adherence to drug regimens as levers that impact the likelihood of a developing drug resistance in a population receiving prophylaxis for prevention of HIV infection.[4] Coupling an epidemiologically based within-host viral dynamics model with individual subject PK/PD model that is affected by dosing adherence, the relative contribution of host specific biological factors and behavioral factors related to adherence to the development of resistant virus could be estimated. This allows the opportunity to plan an intervention for a particular community of patients and develop effect means for assessment of treatment to intervene before inadequate dosing can contribute to the development of resistance. 

A final example from treatment of tuberculosis will be discussed. In a meta-analysis review of clinical trials to explore the use of treatment shortening regimens for TB therapy, Wallis et al explored sentinel indicators that could be used to explore the likelihood that patients would successfully be treated based on initial results after 2 months of therapy.[5] This review identified factors related to treatment regimen and duration that increased the probability of success of clinical trials to show non-inferiority to standard of care therapy. The analysis is being extended to explore the individual subject level covariates that can predict who among a treated population is more likely to succeed with a particular regimen of a different durations. These results will allow the population tailoring of tuberculosis therapy to maximize success across a range of patients. 

Conclusion: Pharmacometric analyses of individual subject level data from infectious disease treatment studies can provide valuable information for identifying key factors that will contribute to successful treatment of a maximum number of patients. When coupled with epidemiological models to assess factors that impact the spread of disease, effective strategies for treating patients emerge that form the basis for Precision Public Health interventions. 

References:

1. Khoury MJ, Iademarco MF, Riley WT: Am J Prev Med 2015.

2. Worldwide Antimalarial Resistance Network (WWARN); BMC Medicine 2015; 13:227

3. Tarning J: Personal communiction 2016.

4. Bershteyn A, Eckhoff PA; BMC Systems Boilogy 2013; 7:11

5. Wallis RA, et al: PLoS One 2013; 8; e71116