Study of Pharmacokinetics in HIV-infected Women

Predictors of antiretroviral pharmacokinetics in HIV-infected women with virologic suppression on combination antiretroviral therapy

Background

Women in Canada constitute the fastest growing population groups at risk for infection with HIV and AIDS. Women now represent approximately 40% of all AIDS cases worldwide and approximately 20% of cases in Canada. Although the AIDS epidemic has been ongoing for more than 20 years, surprisingly little is known about the differential efficacy and toxicity of various antiretroviral (ARV) drugs in women as compared to men. This gap in knowledge is a result of the initial exclusion of and continued underrepresentation of women in ARV clinical trials. Many studies suggest that HIV-infected women taking ARV treatment (ART) have more adverse events (AEs) than men, especially in relationship to systemic symptoms like diarrhea, as well as organ toxicity, including hepatotoxicity, lactic acidosis, peripheral neuropathy and, notably, lipodystrophy. Currently, the occurrence and management of AEs is the most important issue in the treatment of HIV. Understanding the reasons for the differences of AEs in HIV-infected women is critical and has yet to be evaluated within a large cohort. It is unknown whether these differences relate to hormonal influences, drug metabolism, adherence, fat distribution, body size or other factors. Some small studies have found that drug levels (e.g. Cmin, Cmax, AUC) are higher in women and are associated with increased toxicity. Ultimately, we plan to conduct a randomized clinical trial (RCT) to assess the utility of therapeutic drug monitoring (TDM) and ARV drug dose adjustment on the frequency of AEs in women. Prior to conducting this RCT, it is critical to conduct this study on ARV pharmacokinetics (PK) in women to obtain additional information on ARV drug levels in women and their relation to AEs.

Hypothesis

The hypothesis for this study is three-fold:

1. that the mean drug levels (Cmin and Cmax) of ARVs will be significantly higher in our female population as compared to the mean drug levels in the historical HIV population (which is primarily men)

2. that ARV drug levels, particularly Cmin, are associated with body weight in women

3. that higher ARV drug levels, particularly Cmax, are associated with higher frequency and severity of AEs.

Patient population

Eighty HIV-infected women from 8 Canadian sites who have been on their first combination ART regimen containing either a PI or an NNRTI (since these are the ARV agents eligible for PK analysis) for at least three months and who have evidence of full virologic suppression (HIV RNA VL less than 50 copies/mL) on at least two occasions at least one month apart.

Objectives Primary objectives

1. To demonstrate that the Cmin and Cmax of PIs and NNRTIs are significantly higher in our female population as compared to the mean drug levels in the historical general population (which is primarily men).

2. To determine the association between PI and NNRTI Cmin and body weight in our female population.

Secondary objectives

1. To determine the association between Cmax and the frequency and severity of AEs as measured by the proportion of patients with grade 2 or higher laboratory or clinical AEs and the Symptom index score in women.

2. To determine the association between ARV drug levels and age, race, height, body mass index, adherence, hormonal levels and therapy, menstruation history, duration of HIV infection, duration on ARV therapy, baseline VL, baseline CD4 count, present CD4 count, hepatitis B or C infection, class of ARVs, presence of ritonavir and other medications.

Study design

The study will be a cross-sectional study with ARV drug levels (Cmin and Cmax) measured weekly for three weeks Three samples will be taken from each subject to reduce variability due to both technologic and biologic sources. Data will be collected on demographic characteristics, clinical disease and ARV history and on clinical toxicities at the first visit. Blood work will be carried out on the first visit to assess for laboratory toxicity and hormonal levels.

Data analysis

Drug levels will be summarized with the mean of the 3 values for Cmin and Cmax for each woman for their PI or NNRTI and compared to the mean values of the general HIV population. Cmin and Cmax will be classified into high and low levels with a high level defined as less than or equal to 1.5 X arithmetic population mean for each drug. Characteristics of patients with high drug levels will be compared to those of patients with low drug levels. Univariate and multivariate logistic regression models will be used to identify independent associations of patient characteristics with high drug levels. The proportions of patients with AEs and the median number of AEs per patient will be compared between patients with high Cmax and patients with low Cmax.