Substantially lowered dolutegravir exposure in a treatment-experienced perinatally HIV-1-infected pregnant woman

To date, data on dolutegravir (Food and Drug Administration pregnancy category B) in pregnancy are limited [1]. We present the first reported case of dolutegravir pharmacokinetics during and after pregnancy. A 24-year-old perinatally HIV-1-infected woman (CDC stage B2), exposed to several antiretroviral agents, was switched to once-daily dolutegravir/abacavir/lamivudine (Triumeq®; ViiV Healthcare, Brentford, UK) because of poor adherence in 2015. At this time, her CD4+ cell count was 260 cells/μl (13%) and her HIV-RNA viral load was 7980 copies/ml. Medication was well tolerated, and blood cell counts and liver function tests were normal. After 1 month of dolutegravir treatment, the patient was found to be pregnant (conception approximately 3 days before treatment initiation). At that point, viral load was less than 20 copies/ml, and remained undetectable throughout a 12-month follow-up period. At week 35 of gestation, a plasma concentration-time curve was recorded over the 24-h dosing interval (Fig. 1; pharmacokinetic parameters were derived by noncompartmental analysis). Dolutegravir total plasma concentrations were determined by a validated liquid chromatography–mass spectrometry method (lower limit of quantification 0.01 mg/l).Fig. 1: Dolutegravir plasma concentration-time profiles at 35 weeks of gestation and 6 weeks postpartum.The steady-state AUC0–24 h, C max, C min, and half-life for dolutegravir 50 mg taken with food during pregnancy (open circles) were 48 h·mg/l, 4.0 mg/l, 0.9 mg/l, and 12 h (r 2 = 0.96), respectively. During postpartum (closed squares), the steady-state AUC0–24 h, C max, C min, and half-life were 106 h·mg/l, 5.4 mg/l, 3.3 mg/l, and 22 h (r 2 = 0.99), respectively.Six weeks postpartum, a second maternal plasma concentration-time curve was recorded as a control to assess the effect of pregnancy on dolutegravir pharmacokinetics (Fig. 1). area under the curve (AUC)0–24, Cmax, Cmin at steady state were 45%, 74%, and 27% of postpartum values, respectively. Compared with pharmacokinetic steady-state parameters reported for nonpregnant patients, the observed values postpartum were relatively high [2]. Pharmacokinetic parameters during third trimester were comparable to mean population values and the observed Cmin was above the suggested threshold for effect (0.1 mg/l) [3]. Therefore, lowered dolutegravir exposure was not expected to be of concern in this case and in agreement with continued viral suppression. However, a more than 50% reduction of dolutegravir exposure (AUC0–24 h) during pregnancy can be of concern for patients that are in the lower ranges of exposure and/or with reduced sensitivity to dolutegravir. The lowered dolutegravir exposure observed during third trimester pregnancy may be related to induction of metabolic pathways and/or drug efflux transporters, including UGT1A1 and/or CYP450 3A4 [4,5]. Other explanations for the observed variability include within-subject variability (low for dolutegravir [6]) and pregnancy-related changes in protein binding. Intake of (iron) supplements was excluded in this case [7] and food intake was standardized. At 38 + 4 weeks of gestation, a male infant was delivered by elective caesarean section (3275 g, 50 cm, head circumference of 34.5 cm). Apgar score was 10 at 5 min and umbilical artery pH was 7.28. Minor but notable elevation in bilirubin levels was observed during the first 6 weeks of life with corresponding jaundice. Total bilirubin concentrations were 144 (conjugated 8.6 μmol/l), 80, and 4.3 μmol/l at 14, 45, and 90 days of age, respectively. Anemia was observed after birth with hemoglobin levels decreasing from 9.9 (day 3) to 5.4 (day 45) mmol/l. By 3 months of age, all values had normalized and the infant developed normally. The infant received standard chemoprophylaxis [8] with zidovudine (4 mg/kg twice daily, 14 days) and was not breastfed. At 3 months of age, repeated HIV DNA PCR tests were negative. At delivery, dolutegravir concentration in cord plasma was 1.32 mg/l and concentration in maternal plasma was 1.06 mg/l (cord-to-maternal concentration ratio of 1.25). Relatively high placental transfer of dolutegravir was previously observed in an ex-vivo placenta perfusion study [9]. Amniotic fluid concentration was low (0.03 mg/l). At 3 and 14 days of age, neonatal plasma dolutegravir concentrations were 0.88 and 0.06 mg/l, respectively, resulting in a calculated neonatal half-life of dolutegravir of 73 h. In HIV-infected adults, the terminal half-life is 14 h [2]. Pain et al.[10] previously observed prolonged dolutegravir exposure and also elevated neonatal bilirubin levels, and attributed it to neonatal UGT1A1 immaturity, and/or competition between bilirubin and dolutegravir for UGT1A1. High bilirubin levels were previously observed in newborns exposed to raltegravir (another UGT1A1 substrate) [11]. Other causes of prolonged elevated bilirubin (hemolysis, neonatal cholestasis, and hypothyroidism) were excluded. In our case, bilirubin levels were below the age-adjusted threshold for phototherapy. However, this may be important in the presence of additional risk factors such as prematurity. The observed anemia was possibly related to pre and postnatal exposure to nucleoside reverse transcriptase inhibitors [12]. In conclusion, the maternal viral load remained undetectable throughout pregnancy. AUC0–24, Cmax, and Cmin were lowered in the third trimester. Dolutegravir placental transfer was relatively high. Neonatal dolutegravir concentrations were also prolonged and high, and possibly causally related to hyperbilirubinemia. Until more data on dolutegravir use during pregnancy become available, we recommend limiting the use of dolutegravir during pregnancy to clinical trials, employing therapeutic drug monitoring of dolutegravir, and monitoring of neonatal bilirubin levels. Acknowledgements S.S., A.C., and D.B. are the primary authors that designed the study. C.F. and K.W. are the primary authors that conceived the study. S.S. and A.C. are the primary authors that analyzed the data. All authors collectively contributed to interpreting results and drafting and editing of the paper. Source of support: The PANNA network is financially supported by the ‘European AIDS Treatment Network (NEAT)’, European Commission, DG Research, 6th Framework program, contract LSHP-CT-2006–037570, BMS, MSD, ViiV Healthcare, and Janssen Pharmaceuticals N.V. The PANNA network is financially supported by ‘European AIDS Treatment Network (NEAT)’, European Commission, DG Research, 6th Framework program, contract LSHP-CT-2006-037570, BMS, MSD, ViiV Healthcare, and Janssen Pharmaceuticals N.V. We thank the patient for participating in this study and the laboratory personnel at the Laboratory of the Department of Pharmacy of the Radboud University Medical Center Nijmegen for analyzing the samples. Conflicts of interest There are no conflicts of interest.