Buprenorphine Pharmacometric Open Label Research Study of Drug Exposure

The overall rationale for this study is to explore new dose regimens in a small number of patients. There is evidence from pharmacometric models with dose simulation that suggest there is room for improvement in buprenorphine dosing. This study will explore these doses. While the endpoint will be primarily pharmacokinetic, it is likely that the revised dose regimen will be more effective and thus holds the potential for benefit for those infants participating. This information will be used to feed back to the model and generate rationally derived, optimal doses to be tested in subsequent efficacy trials.

The neonatal abstinence syndrome (NAS) is a set of signs of withdrawal in an infant with in utero exposure to opioids. Cardinal manifestations include increased muscle tone, autonomic instability, irritability, poor sucking reflex, gastrointestinal symptoms, and impaired weight gain. All infants are treated with non-pharmacologic methods such as swaddling, rooming in with mother and minimization of stimuli. Despite these measures, ~50% of infants require pharmacologic treatment to ensure proper growth and development. While the optimal pharmacologic treatment for NAS has not been identified, expert review identifies an opioid as the primary therapy. In the US 80% of infants are treated with morphine and 20% with methadone. Sublingual buprenorphine has been demonstrated to be safe and effective in an open label phase 1 clinical trial conducted by the Thomas Jefferson University Team [NCT00521248]. These data were used to plan the BBORN (Blinded Buprenorphine OR Neonatal morphine solution) clinical trial [NCT01452789] comparing buprenorphine to morphine for NAS. BBORN demonstrated a 40% reduction in length of treatment compared to morphine in a double blind fashion (New England Journal of Medicine, June 2017). The external validity of this finding has been supported by retrospective examination of buprenorphine used in a treatment paradigm at the University of Cincinnati Medical Center, with a reduction in length of treatment of ~30% in >200 infants.

Dose selection for both the phase 1 trial and the efficacy trial (BBORN) were empirically derived. A population pharmacokinetic model for buprenorphine in NAS has been published by our group. In addition a pre-specified endpoint for the BBORN trial was a pharmacokinetic analysis of buprenorphine. A pharmacokinetic/pharmacodynamic model from the BBORN study has been published (Clinical Pharmacology and Therapeutics, March 2018). The time to control of symptoms was directly tied to buprenorphine exposure, which itself appeared to be driven primarily by clearance. Among the strengths of pharmacometric models is the ability to simulate in silico many potential dose regimens. In this manner, a dose regimen can be chosen that is more likely to be in the desired range of concentrations. This approach also allows for incorporation of covariates of drug exposure or response to treatment. This is much safer and efficient than the traditional approach of choosing an empiric dose that would need to be tested in clinical trial. An ideal dose would quickly reach this exposure while maintaining a good safety margin. There was no evidence of decline in respiratory rate in infants treated with higher doses of buprenorphine compared to lower doses, or those treated with buprenorphine compared to those treated with morphine. This may allow a higher initial dose to more quickly reach therapeutic buprenorphine concentrations. This ultimately could lead to shorter lengths of treatment and stay, though achieving this goal is outside of the scope of the current proposed project.

In summary, buprenorphine at the dose and schedule used in prior clinical trials has been demonstrated to be safe and effective. The goal of the proposed study is to simulate a dose of sublingual buprenorphine for NAS using pharmacometric modelling techniques. This dose will be tested in infants requiring treatment for NAS. Pharmacokinetic samples would be collected and used to confirm and refine the pharmacokinetic model. The proposed study would allow broad examination and refinement of the exposure/response relationship. This optimized dose could later be used in an efficacy trial.