Abstract
Background
Pharmacokinetic (PK) changes can affect antiretroviral (ARV) systemic exposure for critically ill patients living with HIV (CI-PLWH). Studies to guide ARV adjustments in this population are limited.
Methods
A PK analysis was conducted in a 44-year-old CI-PLWH who presented for a heart and lung transplant on veno-arterial extracorporeal membrane oxygenation (VA ECMO). Home ARV therapy (ART) of co-formulated abacavir/lamivudine/dolutegravir (ABC/3TC/DTG) was continued. ARV serum concentrations were obtained during and after VA ECMO. Two blood levels were drawn at 1 h, for maximum serum concentration (Cmax) and a serum trough (Ct). ARVs were given as a single tablet crushed via nasogastric tube.
Results
Area under the concentration–time curve (AUC0–t) was calculated using non-compartmental analysis. Cmax and AUC0–t were higher during VA ECMO compared with post-decannulation. The Cmax of ABC was >2.5-fold higher than the mean in the reference. Cmax and Ct post VA ECMO were within range of referenced literature for all ARVs. Cmax and AUC0–t of DTG post VA ECMO was approximately four- to fivefold lower than referenced literature. HIV virological suppression was maintained throughout the hospitalization.
Conclusions
ART adjustments would not be required for this patient. Additional studies are needed to assess effects of VA ECMO and crushed tube administration of ARVs in CI-PLWH.
Introduction
Pharmacokinetic (PK) changes in critically ill patients living with HIV (CI-PLWH) serve as a challenge when continuing antiretroviral (ARV) therapy (ART) in the hospital [1]. Critical illness, adjustments in medication administration and external support can lead to patient-specific PK changes, altering systemic medication exposure. This includes decreased absorption, increased volume of distribution (Vd), and increased or decreased drug metabolism or clearance depending on medication-specific properties [1–4]. Altered systemic exposure of ARVs may lead to treatment failure, resistance or medication toxicities.
Extracorporeal circuits, including extracorporeal membrane oxygenation (ECMO), can significantly alter PK, primarily through drug sequestration and increased volume of distribution [2,5–7]. To date, only one case report has described ARV PK effects of veno-venous (VV) ECMO [4,8]. In addition, PK studies evaluating administration of crushed ARVs through feeding tubes, which can alter bioavailability, are limited [3]. Drug interactions with tube feed formulations serve as an additional challenge [3,9].
Herein, a PK analysis is described for a CI-PLWH receiving a single tablet regimen (STR) of abacavir/ lamivudine/dolutegravir (ABC/3TC/DTG) administered crushed via nasogastric (NG) tube during veno-arterial (VA) ECMO and post-decannulation.
Patient case
A 44-year-old African American female with HIV was transferred from an outside hospital (OSH) to the University of Maryland Medical Center (UMMC) for heart and lung transplant evaluation. She presented to the OSH with shortness of breath and subjective fevers. Figure 1 provides a hospital course summary. Her past medical history included asthma, pulmonary arterial hypertension (PAH) and gastroparesis. PAH diagnosis was likely secondary to HIV, which were both diagnosed 21 years prior. Her most recent ART was an STR of ABC/3TC/DTG 600/300/50 mg taken by mouth daily, which she self-reported 100% adherence to. Her last HIV RNA was <20 copies/ml and CD4+ T-cell count was 764 (25%) cells/μl drawn at the OSH.
Summary of hospital course
She was transferred to UMMC on VA ECMO, vasopressors and intravenous (IV) anti-infectives. Her home ART was continued and administered orally. On hospital day 2, she experienced haemoptysis requiring intubation. Tube feeds were initiated and her medications were switched to be crushed and administered via NG tube. She was extubated on hospital day 5, but reintubated the same day due to recurrent haemoptysis. On hospital day 7 she underwent bilateral orthotopic lung and heart transplant. She received anti-thymocyte globulin for 3 days for induction immunosuppression. She was extubated and decannulated from VA ECMO on hospital day 12.
On hospital day 10, liver enzymes and serum creatinine were elevated, but resolved after 1 day. HIV RNAs drawn on hospital days 2, 6, 11, 20, 27 and 43 were <20 copies/ml. A CD4+ T-cell count drawn on hospital day 30, 24 days post-transplantation, was 62 (23%) cells/μl. She was advanced to an oral diet on hospital day 26 and discharged on day 50.
Methods
Ethics
The protocol was approved by the University of Maryland, Baltimore Institutional Review Board. Written informed consent was obtained from the patient prior to sample collection.
Sample Collection and Preparation
Serum concentrations were obtained for ABC, 3TC and DTG on hospital days 12 and 17. Two blood levels were drawn within the same dosing interval for each set; one at 1 h to target maximum serum concentration (Cmax) and one before the next dose to target a trough (Ct). During the first set of levels, the patient was cannulated on VA ECMO for 12 days. During the second set, the patient was decannulated from VA ECMO for 5 days (Figure 1). ABC, 3TC and DTG were given as an STR, Triumeq®, which was crushed and administered via NG tube during both sets of levels. It is not clear whether the tablet was administered with tube feeds or water consistently.
Samples were centrifuged at 4,700 g for 5 min at 4°C, stored at -80°C, then shipped on dry ice to the University of North Carolina Center for AIDS Research Clinical Pharmacology and Analytical Chemistry Core Facility (Chapel Hill, NC, USA) for analysis. Plasma concentrations of ABC, 3TC and DTG were quantified using previously published analytical methods [10]. The calibration curves were linear with a lower limit of quantitation (LLOQ) at 1.00 ng/ml. Assay precision and accuracy were within 15% (20% at LLOQ).
Pharmacokinetic Analysis
Area under the concentration–time curve (AUC) was calculated using non-compartmental analysis. Reported Cmax and Ct are reported as direct measured data. Results were compared with published PK studies of non-CI-PLWH receiving each individual ARV as whole tablets [11–13].
Results
Systemic exposures and PK results of ABC, 3TC and DTG are detailed in Table 1 and Figure 2. The Cmax and AUC0–t were higher for the three ARVs during VA ECMO compared with post-decannulation. Ct was similar during ECMO and post-decannulation for all three ARVs. The ABC Cmax of 10,200 ng/ml and AUC0–t of 116,942 ng*h/ml during VA ECMO was higher compared with previously published literature [11]. The 3TC Cmax of 2,200 ng/ml during VA ECMO was within range of previous literature. The 3TC Cmax of 967 ng/ml post-decannulation was lower that published values [12]. For DTG, the Cmax during and after VA ECMO of 1,490 ng/ml and 607 ng/ml, respectively, was lower compared with the literature [13]. The DTG AUC0–t was lower during and after VA ECMO of 21,523 and 11,704 ng*h/ml, respectively, compared with the literature.
Concentration versus time post dose
Conclusions
This case report characterizes PK of ABC, 3TC and DTG administered as an STR, crushed via NG tube, in a CI-PLWH during and after VA ECMO. The Cmax and AUC0–t were higher for all three ARVs during ECMO compared with post-decannulation. ABC Cmax was >2.5-fold higher during ECMO than the Cmax seen in literature (Table 1). DTG Cmax and AUC0–t post-ECMO were approximately four- to fivefold lower than the mean observed in literature [11]. Because Ct remained above the protein-adjusted 90% inhibitory concentration (IC90) for wild-type HIV of 64 ng/ml, this decrease is likely clinically insignificant [13]. Specific to DTG, trough concentrations in the SPRING-1 trial suggest a clinical target trough of 300 ng/ml, 4x higher than the protein-adjusted IC90 [14]. DTG trough concentrations in this report of 409 ng/ml and 415 ng/ml during and post VA ECMO, respectively, exceeded this higher threshold. No events during the hospital course were attributed to ARV toxicities and the patient remained virally suppressed.
Pharmacokinetic parameters of ABC, 3TC and DTG during and after VA ECMO compared with reference studies
ABC, abacavir; AUC, area under the concentration–time curve; Cmax, maximum concentration; Ct, concentration at the end of the dosing interval; DTG, dolutegravir; VA ECMO, veno-arterial extracorporeal oxygenation; 3TC, lamivudine.
A case report describing effects of ECMO on ARV PK assessed serum levels of ritonavir (RTV), darunavir (DRV), 3TC and tenofovir (TFV) in a CI-PLWH undergoing VV ECMO. Apart from 3TC, the other ARVs were given at empiric double doses. RTV and TFV serum levels were lower than expected, whereas DRV levels were higher, although the PK fell within the expected range of population PK models. 3TC was observed to exhibit a PK profile out of expected range compared with a population PK model [8].
Studies evaluating effects of crushing and administering ARVs through enteral feeding tubes are limited. Although ABC and 3TC have commercially available solutions, DTG is currently only available as a tablet, which must be crushed for tube administration. A prospective study evaluating PK after a single dose of ABC/3TC/DTG taken crushed by mouth in 22 healthy volunteers found variability among ARV systemic concentrations. Compared with population PK, ABC was bioequivalent when administered with water, 3TC was bioequivalent when administered with water and the enteral feed, and DTG was not bioequivalent when mixed with either. Specifically, the DTG AUC0–t and Cmax was 26% and 30% higher compared with references [15]. Few case reports assess crushed DTG in PLWH [16–19]. One case report described a CI-PLWH where DTG was increased to 50 mg twice daily given crushed via gastrostomy tube. The DTG Ct of 820 ng/ml following the higher 50 mg twice daily dosing was similar to the reference of 830 ng/ml with 50 mg daily dosing. Another case report in a CI-PLWH found that DTG exposure was similar when administered oral versus jejunostomy tube, but the AUC exposure was 76% and 75% lower compared with the literature, respectively [17]. Similar to the current report, all previous reports describe maintenance of viral suppression with altered DTG administration [16–19].
In the current report, it is unclear why increased Cmax and AUC0–t were observed for ABC, 3TC and DTG during ECMO compared with post-decannulation. This observation is not thought to be due to ECMO alone. ECMO mainly increases medication Vd through drug sequestration in the circuit. Lipophilic and highly protein bound medications are thought to be the most affected [7]. DTG exhibits the highest lipophilicity and protein binding compared with 3TC, which exhibits the lowest. Critical illness and subsequent hypoperfusion of liver and kidney is thought to most directly affect medication metabolism and clearance [7]. The patient had a slight increase in liver enzymes and serum creatinine around the time of blood collection. However, this is thought to be non-contributory given the change persisted for <2 days. It is also possible the tube formulation switch and increase in total percentage of tube feed received post-ECMO could have altered ARV absorption.
Limitations of this study include the critical status of this patient. There are many factors that could affect the PK profile of ARVs outside of NG tube administration and VA ECMO. Additionally, only two concentrations were obtained, limiting the ability to draw conclusions over the entirety of the dosing interval. Also, the Cmax was drawn at 1 h, which was selected based on Cmax of the three different ARVs, although the exact Cmax of each individual ARV differs. Lastly, there was a delay in obtaining plasma concentrations, which were completed and obtained once the patient was taking medications orally; therefore, results were not used to adjust the patient's ART.
Future studies are required to characterize ARV PK during ECMO and when administered crushed for enteral tube feed for CI-PLWH. Based on our findings, alterations in PK with crushed ABC/3TC/DTG in a patient on VA ECMO did not result in a change in HIV viral suppression.
Footnotes
Acknowledgements
We would like to thank Amanda P Schauer at the Center for AIDS Research Clinical Pharmacology & Analytical Core, Chapel Hill, NC, USA, for her analysis of the anti-retroviral plasma samples. We would also like to thank Mackenzie Cottrell at the Center for AIDS Research Clinical Pharmacology & Analytical Core, Chapel Hill, NC, USA, for her compilation of the TDM report. ASD is supported by the National Institute of General Medical Sciences of the NIH under Award Number T32GM086330. The content is solely the responsibility of the authors and does not necessarily represent the official views of National Institutes of Health.
The authors declare no competing interests.