Improving Pharmacokinetics of Riluzole for Spinal Cord Injury: Formulation and Combination Therapy Approaches



Journal Title

Journal ISSN

Volume Title



Acute Spinal Cord injury (SCI) refers to a traumatic insult to the spinal cord that can lead to either temporary or permanent loss of motor/sensory and autonomic functions, below the site of injury. Though extensive research is going on, there is yet no pharmacological therapy available for SCI. Riluzole, an FDA approved drug for the treatment of amyotrophic lateral sclerosis (ALS), had demonstrated enhanced tissue sparing and recovery of motor function in a rodent SCI model. Following this, a phase I trial was conducted and safety of riluzole was established in SCI patients. Preliminary assessments indicated that riluzole improved the motor outcome in cervical injured (Grade B) patients. However, high inter- subject variability in plasma concentrations (17- and 37-fold on day 3 and day 14, respectively) and decline in Cmax and Cmin (from Day 3 to Day 14) are observed during the two-week treatment regimen. Currently, a multicenter, placebo controlled, double-blinded, randomized phase II/III trial is ongoing to evaluate the therapeutic efficacy of riluzole for the treatment of acute spinal cord injury. A PK sub study in this trial (expected enrollment of 60 patients) was designed to further characterize the changes in riluzole plasma PK during the two week treatment regimen. Riluzole is a P-gp substrate and P-gp overexpression following spinal cord injury in experimental SCI models has been demonstrated that can potentially reduce the systemic and intra-spinal concentrations of orally administered riluzole.

We hypothesize that therapeutic outcome of riluzole is positively correlated to its plasma and intra-spinal exposures.

The objective of our research is to understand the pharmacokinetics of riluzole in SCI patients and improve the pharmacokinetics by overcoming the potential challenges that might limit its efficacy. To achieve the stated objective, three specific aims were set and achieved with the following studies:

  1. To develop and validate sensitive LC-MS/MS assay of riluzole in human plasma and cerebrospinal fluid (CSF) for quantification of riluzole in clinical samples. Liquid-liquid extraction using ethyl acetate was used to extract riluzole from plasma and CSF. Chromatographic separation was achieved using C18 column with an isocratic elution. Mass detection was performed using electrospray ionization (ESI) in the positive mode. Multiple reaction monitoring (MRM) was used for quantification.

Linearity of the LC-MS/MS assay was demonstrated over the range of 0.5-800 ng/ml in human plasma and 1.0-800 ng/ml in human CSF. The accuracy and precision of the assay method were within ± 15% as per the FDA guidelines. The method was used successfully to quantify riluzole in clinical plasma and CSF samples, obtained from patients enrolled in phase II/III trial.

  1. To develop a stable aqueous based liquid formulation of riluzole. Poor aqueous solubility of riluzole was increased by optimum concentrations of FDA approved co-solvents. A Central Composite Design was used to optimize the composition of co-solvents required to solubilize entire dose (50 mg) of riluzole. The optimized formulation was tested for stability, in vitro hemolytic potential. Plasma pharmacokinetics and CNS distribution were also characterized after oral and IV administrations.

An aqueous based soluble formulation of riluzole was developed using FDA approved water miscible organic solvents, namely, polyethylene glycol (PEG 400), propylene glycol (PG) and glycerin (GLY). The selected optimum composition consisted of 15%, 20% and 10% v/v of PEG 400, PG and GLY, respectively. The formulation was suitable for both oral and intravenous administrations and was well tolerated in animals. The bioavailability of the oral co-solvent formulation was 94.4% which was higher than that of the crushed tablet (64.1%). The co-solvent formulation had a faster absorption, slower elimination and higher bioavailability as compared to tablets (crushed and suspended).

  1. To determine the effect of co-administration of minocycline (neuroprotective and p-gp inhibitor) on plasma and intra-spinal availability of orally administered riluzole in uninjured and SCI rat models. Riluzole was co-administered with minocycline in uninjured and spinal cord injured rats. Plasma and spinal cords were harvested. Riluzole and minocycline concentrations were quantified using validated assay methods.

The bio-distribution profiles of riluzole and minocycline are distinctly different. Riluzole distributes more to brain and spinal cord, whereas minocycline distributes more to plasma. When riluzole (10 mg/kg, oral) was co-administered with minocycline (40 mg/kg, IV) in uninjured rats, the exposure of riluzole was significantly reduced in plasma, brain and spinal cord. Minocycline exposure was also significantly reduced in brain and spinal cord. However, when riluzole was co-administered with minocycline (25 mg/kg, ip), there was no change in the exposure of riluzole, but minocycline exposure in spinal cord was significantly reduced. Thus we observed a differential impact of dose and route of administration of minocycline on pharmacokinetics of orally delivered riluzole.

In SCI injured rats, the plasma and spinal cord concentration of riluzole were significantly higher on day 14 as compared to uninjured rats (Riluzole alone treated groups). Riluzole concentrations on day 3 was significantly higher in both plasma and spinal cord in the combination cohort [riluzole, 10 mg/kg, oral co-administered with minocycline (10 mg/kg, IV)], whereas minocycline concentrations in plasma and spinal cord were significantly higher on day 7 in the combination cohort, as compared to the alone cohorts. The time course of riluzole in plasma did not change from 24 hr to 14 days, either alone or in combination longitudinal draw groups. The spinal cord concentration was also similar on day 14 between the alone and the combination group. In case of minocycline, plasma concentrations were significantly higher at all time points (24 hr, 72 hr, 7 d and 14d) in the combination group as compared to alone group (longitudinal draw). The spinal cord levels were also higher in the combination group on day 14 as compared to minocycline alone group. The PD study is still ongoing and its outcomes may provide a better insight to interpret and correlate with the PK findings

Overall, we have successfully accomplished the goals for specific aims 1 and 2 with the development and validation of a LC-MS/MS assay of riluzole in human plasma and CSF and development of a soluble formulation of riluzole that can be administered both orally and intravenously with flexibility of dose adjustment in patients. We have also investigated the potential combination of riluzole and minocycline in both uninjured and injured rats (specific aim 3). Our observations did not support our hypothesis since minocycline co-administration did not increase plasma or spinal cord exposures of riluzole. Moreover, the uninjured model might not be suitable to evaluate the impact of combination in injured rats. There are several aspects that can be explored with further experiments such as identification of common uptake and efflux transporters that would help to get better insight into the mechanism of interaction. Also, different dose combinations might be evaluated in vivo to identify optimum concentrations of riluzole and minocycline that would improve the pharmacokinetics/exposure of both drugs in spinal cord injured rats.



Riluzole, Spinal cord injuries, LC-MS/MS, Minocycline, Formulation, Co-solvent