P1026s is an ongoing, multi-centre, prospective study of ARV pharmacokinetics among HIV-infected pregnant women receiving ARVs for routine clinical care. The current analyses address only those women using NFV. P1026s is a sub-study of P1025, a prospective cohort study of HIV-infected pregnant women receiving care at PACTG sites. Institutional review boards approved both P1025 and P1026s at all participating sites. All participants provided written informed consent prior to participation in these studies.
Participants were eligible for inclusion in the NFV arm of P1026s if they met the following criteria: they were HIV-infected pregnant women ≥ 20 weeks’ gestation who were enrolled in P1025; they initiated NFV (625 mg tablets) at a dose of 1250 mg orally twice daily by 34 6/7 weeks of gestation; they were receiving this dose of NFV for at least 2 weeks prior to pharmacokinetic sampling; and they were planning to continue NFV until at least 6 weeks postpartum. Exclusion criteria were multiple-gestation pregnancies and clinical or laboratory toxicity that, in the opinion of the site investigator, would probably require a change in the ARV regimen during the study. The participant’s clinician determined the choice of ARVs, prescribed the drugs and remained responsible for clinical management throughout the study. Participants remained on study until the completion of postpartum pharmacokinetic sampling.
As part of P1025 patients were interviewed, physical examinations were performed and laboratory studies were conducted during pregnancy and postpartum. Data from P1025 accessed for this analysis include: demographic data (maternal age, ethnicity); concomitant ARVs; laboratory results [plasma HIV-1 RNA concentrations (viral loads) and CD4 T-cell counts]; infant birth data (gestational age, weight and length); and infant HIV infection status. Information collected as part of P1026s includes the patient’s height and weight on days of sampling, and adverse events experienced by patients.
On each sampling day, a medical history was obtained from the patient (including the time of the last two doses of NFV), a physical examination was performed and laboratory studies were obtained (including measurements of alanine aminotransferase, aspartate aminotransferase, bilirubin, creatinine, blood urea nitrogen, albumin and haemoglobin). Adverse events were reported according to the Division of AIDS (DAIDS)/National Institute of Allergy and Infectious Diseases (NIAID) Toxicity Table for Grading Severity of Adult Adverse Experiences (August 1992) (http://rcc/tech-res-intl.com
). All adverse events were followed until resolution. The study team reviewed adverse event reports on monthly conference calls. The patient’s clinician was responsible for toxicity management.
Venipuncture for plasma samples for pharmacokinetic evaluation was performed at the following timepoints: second trimester (between 20 and 26 weeks of gestation), third trimester (between 30 and 36 weeks of gestation), at delivery and postpartum (between 6 and 12 weeks after delivery). Women who did not complete an antepartum evaluation were replaced in the study. Patients were receiving a stable ARV regimen for at least 2 weeks prior to pharmacokinetic sampling. Participants were instructed to take their NFV at the same times each day for the 3 days prior to and on the day of the antepartum and postpartum pharmacokinetic evaluations. Seven plasma samples were drawn at antepartum and postpartum pharmacokinetic evaluation visits, starting immediately before the morning NFV dose and at 1, 2, 4, 6, 8 and 12 h after the witnessed dose. The morning NFV dose was ingested immediately after a standardized meal of 1000 kcal with 50% fat provided at the site. To assess transplacental passage, NFV and metabolite concentrations were measured in maternal plasma and umbilical cord samples obtained at delivery.
Concentrations of NFV and its main active metabolite [hydroxyl-tert
-butylamide (M8)] were determined simultaneously by high-performance liquid chromatography (HPLC) in the Pediatric Clinical Pharmacology Laboratory at the University of California, San Diego. Samples were stored at −20°C, and were assayed within 7–10 days of receipt. Briefly, plasma proteins were precipitated using acetonitrile (ACN) and supernatant injected directly onto a LUNA C-18 reversed phase HPLC column (Phenomenex Inc., Torrance, CA, USA). Drugs were separated isocratically using a mobile phase consisting of 10 mM potassium phosphate buffer pH 4.2: ACN (62:38 v/v). The flow rate was 1.2 mL/min and ultraviolet (UV) detection was at 206 nm. The detection limit for both NFV and M8 was 0.039 mg/mL. The mean inter- and intra-assay coefficients of variation based on validation data (quality control samples were run at multiple different concentrations over the control range of 0.039–8.5 μg/mL) revealed the lowest and highest values as follows: 5.2 ± 2.3% and 3.1 ± 2.1% for NFV, and 4.3 ± 1.8% and 3.3 ± 2.8% for M8 [5
]. The laboratory is licensed by the State of California and is Clinical Laboratory Improvement Amendments (CLIA)-compliant, participating in the ACTG proficiency testing programme twice a year [6
The concentration data were analysed by direct inspection to determine pre-dose concentration (Cpre-dose
), maximum plasma concentration (Cmax
), the corresponding time (Tmax
), the minimum plasma concentration (Cmin
), the corresponding time (Tmin
) and the 12-h post-dose concentration (C12
) for both NFV and M8. For concentrations below the assay limit of detection, a value of one half of the detection limit (0.0195 μg/mL for NFV and M8) was used in calculations of various ratios. The area under the plasma concentration–time curve during the dose interval [from time 0 to 12 h post-dose (AUC0–12
)] for NFV and M8 were estimated using the trapezoidal rule. Apparent clearance (CL/F) from plasma for NFV was calculated as dose divided by AUC0–12
. Apparent volume of distribution (Vd
/F) was estimated by CL/F divided by the terminal slope of the curve (λz
), and half-life (t1/2
) was calculated as 0.693 divided by λz
. The target NFV AUC0–12
was ≥ 10th percentile NFV AUC0–12
in non-pregnant historical controls (18.5 μg h/mL) [5
]. The target Cmin
for NFV was 0.8 μg/mL [7
Both Vd/F and CL/F also were estimated using a one-compartment model in the software program WinNonlin® (version 5.0.1; Pharsight Corporation, Mountain View, CA, USA). Pharmacokinetic parameters derived from each approach were compared to assess the potential limitations of each methodology.
The difference in any pharmacokinetic parameter (third trimester vs. postpartum) was assessed using the Wilcoxon signed-rank test. For the comparison of third trimester vs. postpartum NFV and M8 pharmacokinetic parameters, we made all comparisons at the within-patient level, using 90% confidence intervals (CIs) for the geometric mean ratio of the pharmacokinetic exposure parameters during the third trimester of pregnancy vs. postpartum.
A true geometric mean ratio of 1 indicates equal geometric mean pharmacokinetic parameters for the two timepoints being compared. If the 90% CI is entirely outside the limits of (0.8, 1.25), the pharmacokinetic parameter is deemed different for the two timepoints. If, on the other hand, the 90% CI is entirely within the limits (0.8, 1.25), the parameter is not different between the two timepoints. If the 90% CI overlaps with (0.8, 1.25), these data alone do not support any conclusions regarding the pharmacokinetic parameter.