Is infant exposure to antiretroviral drugs during breastfeeding quantitatively important? A systematic review and meta-analysis of pharmacokinetic studies.

OBJECTIVES: The objectives of this study were to summarize antiretroviral drug concentrations in breast milk (BM) and exposure of breast-fed infants. METHODS: This was a systematic review of pharmacokinetic studies of HIV-positive women taking antiretrovirals that measured drugs in BM. The quality of pharmacokinetic and laboratory methods was assessed using pre-defined criteria. Pooled ratios and 95% CIs were calculated using the generalized inverse variance method and heterogeneity was estimated by the I(2) statistic. PubMed Central, SCOPUS and LactMed databases were searched. No date or language restrictions were applied. Searches were conducted up to 10 November 2014. Clinical relevance was estimated by comparing ingested dose with the recommended therapeutic dose for each drug. RESULTS: Twenty-four studies were included. There was substantial variability in the clinical and laboratory methods used and in reported results. Relative to maternal plasma (MP), NRTIs accumulate in BM, with BM : MP ratios (95% CI estimates) from 0.89 to 1.21 (14 studies, 1159 paired BM and MP samples). NNRTI estimates were from 0.71 to 0.94 (17 studies, 965 paired samples) and PI estimates were from 0.17 to 0.21 (8 studies, 477 paired samples). Relative to the recommended paediatric doses, a breast-fed infant may ingest 8.4% (95% CI 1.9-15.0), 12.5% (95% CI 2.6-22.3) and 1.1% (95% CI 0-3.6) of lamivudine, nevirapine and efavirenz, respectively, via BM. CONCLUSIONS: Transfer to untreated infants appears quantitatively important for some NRTIs and NNRTIs. The pharmacokinetic methods varied widely and we propose standards for the design, analysis and reporting of future pharmacokinetic studies of drug transfer during breastfeeding.

DRV concentrations and viral load in CSF in patients on DRV/r 600/100 or 800/100mg once daily plus two NRTI.

INTRODUCTION: Darunavir/r (DRV/r) is currently used at a dose of 800/100 mg once daily (OD) in a high proportion of patients. Pharmacokinetic data suggest that 600/100 OD may be effective, reducing toxicity and cost. However, drug concentrations in reservoirs such as cerebrospinal fluid (CSF) might not be adequate to inhibit viral replication. We aimed to evaluate concentrations of DRV and HIV-1 viral load (VL) in CSF patients receiving DRV 600/100 mg OD. METHODS: DRV600 is an ongoing randomized open study comparing DRV/r 800/100 mg (DRV800) vs 600/100 mg (DRV600) OD plus TDF/FTC or ABC/3TC in 100 virologically suppressed patients (eudraCT 2011-006272-39). Here we present the results of a CSF sub-study. A lumbar puncture (LP) was performed in participating patients after at least six months of inclusion in the study, 20-28 hours after a dose of DRV/r. VL (PCR, LOD 40 copies/mL) was determined in CSF and in plasma. DRV concentrations were quantified in CSF by liquid chromatography mass spectrometry (LC/MS/MS) and in plasma using high-performance liquid chromatography (HPLC). RESULTS: Sixteen patients were included (eight in each arm). All DRV600 patients and four out of eight DRV800 patients received TDF/FTC, and the other four ABC/3TC. 75% were males, median (range) age was 48 (17-71) years, CD4 cell count 532 cells/mL (190-1,394). Median total time on DRV/r was 30 (11-57) months, and since the beginning of the study 8 (6-12) months in DRV800 and 10 (7-12) months in DRV600 patients. LP was performed a median of 26 (24-28) hours after the last DRV/r+TVD or KVX dose. In DRV600 patients the median DRV plasma levels were 1,674 (326-3,742) ng/mL, CSF levels 17.08 (5.79-30.19) ng/mL and DRV CSF:plasma ratio 0.0084 (0.0028-0.0388), while in the DRV800 arm, median DRV plasma levels were 1,707 (958-3,910) ng/mL, CSF levels 13.23 (3.47-32.98) ng/mL and DRV CSF:plasma ratio 0.0104 (0.0018-0.0262). All patients had VL<40 copies/mL in plasma and 14 patients VL<40 copies/mL in CSF. Two patients (1 in each arm, and taking TDF/FTC) had detectable VL in CSF (280 and 159 c/mL). These patients had the lowest CSF DRV concentrations (5.47 and 3.47 ng/mL), with plasma DRV concentrations of 802 and 958 ng/mL respectively. CONCLUSIONS: CSF DRV concentrations and CSF VL were similar between patients receiving DRV/r 800/100 mg or 600/100 mg OD. Low CSF DRV concentrations might be associated with viral escape in CNS. This may be taken into account in patients receiving OD DRV/r. Larger studies should confirm these findings.