A phase 1 dose‐finding study of intravenous L‐citrulline in sickle cell disease: a potential novel therapy for sickle cell pain crisis

Acute pain from vaso-occlusion (VOC) in sickle cell disease (SCD) is the most frequent cause of emergency room visits and hospital admissions, contributing to the high burden of health care costs (Lanzkron et al, 2010). While major advances in the care of patients with SCD have occurred over the last 30 years, very little progress has been made in the actual treatment of VOC. Nitric oxide (NO) is a powerful vasodilator that plays a fundamental role in VOC (Morris, 2008). NO is produced from L-citrulline and L-arginine, amino acids generated through the urea cycle from the NO synthase (NOS) family of enzymes (Moncada & Higgs, 1993). There are three NOS isoforms: neuronal (nNOS) found in neuronal tissue, inducible NOS (iNOS) found in cells and tissues, and endothelial NOS (eNOS) found in vascular endothelial cells. Shen et al (2005) showed that citrulline was the major supply for intracellular L-arginine and endothelial NO production in murine endothelial cells. Furthermore, Wijnands et al (2012) showed that L-citrulline supplementation restored intracellular NO production, which was related to the degree of eNOS phosphorylation. Moreover, enhanced arginase-induced arginine consumption is believed to play an integral role in the pathogenesis of sickle cell complications. In a more recent study, L-citrulline supplementation increased NO production and improved microcirculatory flow during conditions with arginase-induced arginine deficiency (Wijnands et al, 2015). However, intravenous citrulline has never been evaluated in human SCD. Hence, this study aimed to characterize the pharmacokinetic (PK) and safety profile of intravenous (IV) citrulline in this unique patient population. A single centre open label phase 1 trial of IV citrulline was performed in participants with SCD following approval from the University of Mississippi Medical Center (UMMC) institutional review board. The study was registered at ClinicalTrials.gov (NCT02314689; NCT02697240), where the inclusion and exclusion criteria are described. Th phase 1 study was performed in two steps. Step 1 included a dose escalation bolus infusion of IV citrulline in steady-state SCD to determine the PK and safety profile with a peak goal plasma citrulline concentration of 80–100 lmol/l (Barr et al, 2007). Step 2 of the study was performed to evaluate safety and PK during a vaso-occlusive crisis. The study drug, L-citrulline, was administered as open label vials of 50 mg/ml (5%) isotonic solution. Plasma sampling for PK studies were collected at specific time points. Briefly, for amino acid analysis, deproteinated plasma samples were subjected to cation exchange chromatography using a 4-component pH and ionic strength graded lithium citrate buffer system on a Beckmann 7300 amino acid analyser (Beckmann, Palo Alto, CA). Data obtained for each patient was fitted to a single-compartment PK model. The appearance of citrulline in plasma was described by a zeroorder process (rate of citrulline appearance, Rapp) to account for endogenous production, whereas the removal of citrulline was determined by a first-order process (constant of citrulline removal, krem). It was assumed that the values of all parameters remained constant for each patient during the course of plasma sampling. Scientist v2.0 (Micromath Scientific Software, St. Louis, MO) was used to fit the plasma citrulline concentration to the PK model by a weighted, least squares procedure to obtain values for Rapp, krem, and the volume of distribution (Vd). Clearance was calculated as the product of krem and Vd. For safety assessments, the Investigator determined the intensity of any adverse event (AE) according to the National Cancer Institute Common Terminology Criteria for Adverse Events Version 4.03 (https://evs.nci.nih.gov/ftp1/CTCAE/ CTCAE_4.03_2010-06-14_QuickReference_5x7.pdf) and their causal relationship. A Data Safety Monitoring Board, comprising 3 physicians who were not related to the study, reviewed every third subject. A total of 8 subjects with SCD were enrolled in this phase 1 study of IV citrulline: four participants were enrolled in Step 1 and another four participants in Step 2. Patient demographics, genotype and baseline blood counts are shown in Table SI. In the first cohort of four participants, the IV bolus infusion of 20 mg/kg of L-citrulline yielded a mean peak level of 259 lmol/l and trough level in the range of 20–40 lmol/l at 4 h after infusion (Fig 1). Citrulline PK parameters for Step 1 with bolus infusion are shown in Table SII. Pharmacokinetic model simulations indicated a 20 mg/kg bolus dose of IV citrulline followed by 7 mg/kg per hour continuous infusion was needed to maintain the target citrulline plasma concentration of 100 lmol/l. Subsequently, four subjects with VOC were enrolled to receive IV citrulline bolus and continuous infusion. Individuals with VOC showed significantly lower baseline citrulline levels compared to steady-state (mean SD: 9 37 1 43 vs. correspondence