Oral Drug Pharmacokinetics and Treatment Effects in Populations Affected by Diarrheal Diseases

Abstract

Diarrheal diseases are a major global health concern. Globally, over 4.5 billion episodes of diarrhea occur among all ages every year. Diarrhea disproportionally affects young children and those living in low- and middle-income countries. In 2019 alone, diarrhea was responsible for approximately 1.5 million deaths among people of all ages, including over 500,000 deaths in children under 5 years of age (Dadonaite et al., 2024). Moreover, around 90% of diarrheal mortalities occur in sub-Saharan Africa and South Asia, where undernutrition, poor sanitation, and limited access to medical care are common (GBD 2016 Diarrhoeal Disease Collaborators, 2018). The high prevalence of diarrheal diseases creates a challenge for oral drug administration, as physiological changes associated with diarrhea can impact key processes involved in oral drug absorption, thus leading to alterations in oral drug pharmacokinetics (PK). As a result, the safety and effectiveness of oral drugs may be altered when taken during episodes of diarrhea. The understanding of how diarrheal diseases affect the oral drug PK, whether this impact influences treatment outcomes, and if dosage adjustments are necessary is essential for making accurate dosing decisions and optimizing the likelihood of treatment success. To this end, we leveraged PK modeling approaches to characterize and simulate oral drug PK and make predictions about treatment efficacy for populations affected by diarrheal diseases.In Chapter 2, to estimate the magnitude of diarrhea-associated impact on oral clofazimine PK, we developed a population PK model using clofazimine PK data collected from a phase 2a clinical trial of HIV-infected adults with/without cryptosporidiosis and diarrhea. Among potential covariates of cryptosporidiosis-associated diarrhea severity, HIV infection burden, baseline demographics, and study assignment, maximum diarrhea grade over the study duration was significantly associated with clofazimine bioavailability. Our model quantified a 6- and 22-fold reduction in clofazimine bioavailability associated with mild and severe diarrhea, respectively. Moreover, the model also estimated clofazimine PK parameters apparent clofazimine clearance (3.71 L/h), intercompartmental clearance (18.2 L/h; inter-individual variability [IIV] 45.0%), central volume of distribution (473 L; IIV 3.46%), peripheral volume of distribution (3434 L), absorption rate constant (0.625 h-1; IIV 149%), and absorption lag time (1.83 h), all of which were aligned with literature values. To determine whether cryptosporidiosis-associated diarrhea reduced clofazimine levels below levels associated with efficacy against Cryptosporidium infection, a pharmacokinetic/pharmacodynamic (PK/PD) modeling approach was undertaken in Chapter 3. Exposure-response relationships were determined using data collected in a preclinical mouse model of Cryptosporidium infection. Emax and logistic models were constructed, supporting predictions of efficacious clofazimine concentrations. By comparing the observed clofazimine concentrations from an unsuccessful phase 2a clinical trial to our predicted efficacious target, it was shown that the observed concentrations were well below concentrations associated with anti-Cryptosporidium efficacy. Thus, the prescribed doses were inadequate, even without any impact of diarrhea. Since the effect of diarrhea on oral drug absorption is likely disease- and drug-specific, tebipenem for shigellosis treatment in Bangladeshi pediatric patients was studied in Chapter 4. By validating an existing tebipenem pediatric population PK model with observed tebipenem PK data collected from the pilot study of an ongoing clinical trial, we first demonstrated that the existing model adequately predicted tebipenem PK in a population with diarrhea. Then, we performed population PK simulations using the validated model, incorporating demographic characteristics of the target Bangladeshi pediatric population. The simulated tebipenem PK profiles were used to predict the probability of achieving maximum treatment effect in the target population and the main trial’s chance of success, comparing two different dosing regimens. The findings of Chapter 4 showed that the effect of shigellosis-associated diarrhea on tebipenem PK was minimal. Moreover, Chapter 4 demonstrated an effective modeling workflow that can be adapted to support dose evaluation and efficacy predictions for other drugs and disease areas.