Efavirenz-Based Antiretroviral Therapy but Not Pregnancy Increased Unbound Piperaquine Exposure in Women during Malaria Chemoprevention.

Dihydroartemisinin-piperaquine (DP) is highly effective for malaria chemoprevention during pregnancy, but the standard dosing of DP that is used for nonpregnant adults may not be optimal for pregnant women. We previously reported that the pharmacokinetic exposure of total piperaquine (PQ; both bound and unbound to plasma proteins) is reduced significantly in the context of pregnancy or efavirenz (EFV)-based antiretroviral therapy (ART). However, as PQ is >99% protein-bound, reduced protein binding during pregnancy may lead to an increase in the pharmacologically active unbound drug fraction (fu), relative to the total PQ. We investigated the impact of pregnancy and EFV use on the fu of PQ to inform the interpretation of pharmacokinetics. Plasma samples from 0 to 24 h after the third (final) DP dose were collected from pregnant women at 28 weeks gestation who were receiving or not receiving EFV-based ART as well as from women 34 to 54 weeks postpartum who were not receiving EFV-based ART, who served as controls. Unbound PQ was quantified via ultrafiltration and liquid chromatography-tandem mass spectrometry, with fu being calculated as PQunbound/PQtotal. The geometric mean fu did not differ between pregnant and postpartum women (P = 0.66), but it was 23% (P < 0.01) greater in pregnant women receiving EFV-based ART, compared to that in postpartum women who were not receiving EFV-based ART. The altered drug-protein binding, potentially due to the displacement of PQ from plasma proteins by EFV, resulted in only a 14% lower unbound PQ exposure (P = 0.13) in the presence of a 31% lower total PQ exposure (P < 0.01), as estimated by the area under the concentration time curve from 0 to 24 h post-last dose in pregnant women who were receiving EFV-based ART. The results suggest that the impact of pregnancy and EFV-based ART on the exposure and, in turn, the efficacy of PQ for malaria prevention may not be as significant as was suggested by the changes in the total PQ exposure. Further study during the terminal elimination phase (e.g., on day 28 post-dose) would help better characterize the unbound PQ exposure during the full dosing interval and, thus, the overall efficacy of PQ for malaria chemoprevention in this special population.

Determination of unbound piperaquine in human plasma by ultra-high performance liquid chromatography tandem mass spectrometry.

Piperaquine (PQ) is an antimalarial drug that is highly protein-bound. Variation in plasma protein contents may affect the pharmacokinetic (PK) exposure of unbound drug, leading to alteration of clinical outcomes. All published methods for determination of PQ in human plasma measure the total PQ including both bound and unbound PQ to plasma proteins. There is no published method for unbound PQ determination. Here we report an ultra-high performance liquid chromatography tandem mass spectrometry (UHPLC-MS/MS) method for determination of PQ in human plasma filtrate prepared by filtering human plasma through Millipore Microcon® centrifugal filters (10k NMWL). The filter cup had to be treated with 5% benzalkonium chloride to reduce non-specific binding to the filter devices before filtration of plasma samples. Multiple reactions monitoring (MRM) of the ion pairs m/z 535/288 for PQ and m/z 541/294 for the internal standard (IS) was selected for quantification. When electrospray ionization (ESI+) was used, paradoxical matrix effect was observed despite the structure similarity of the deuterated IS: Ion suppression for PQ versus ion enhancement for the PQ-d6, even though they were closely eluted: 0.62 min versus 0.61 min. Separation was achieved on Evo C18 column (50 × 2.1 mm, 1.7 µm, Phenomenex Inc.) eluted with 10 mM NH4OH and MeCN. When atmospheric pressure chemical ionization in positive mode (APCI+) was used for ion source, matrix effect diminished. Separation was achieved on a PFP column (30 × 2.1 mm, 1.7 µm, Waters, Corp.) eluted with aqueous 20 mM ammonium formate 0.14% trifluoroacetic acid (A) and methanol-acetonitrile (4:1, v/v) containing 0.1% trifluoroacetic acid (B) at 0.8 mL/min flow rate in a gradient mode: 30-30-80-80-30-30%B (0-0.1-1.0-1.40-1.41-1.50 min). The retention time was 0.67 min for both PQ and the IS. The method was validated with a linear calibration range from 20 to 5,000 pg/mL and applied to clinical samples.