Author ORCID Identifier

Date Available


Year of Publication


Degree Name

Doctor of Philosophy (PhD)

Document Type

Doctoral Dissertation




Pharmaceutical Sciences

First Advisor

Dr. Markos Leggas

Second Advisor

Dr. Chee M. Ng


Neonatal abstinence syndrome, or NAS, is a postnatal opioid withdrawal syndrome occurring in 55% to 94% of neonates as a result of in utero exposure to opioids. It has emerged as a significant public health issue, as its incidence more than quadrupled in the past decade. There is significant variability in disease severity and treatment outcomes in neonates with NAS due to patient-specific factors and treatment- or site-specific factors. To understand what contributes to variability in length of hospital stay and other outcomes in neonates with NAS, we assessed population pharmacokinetics (PK) of clonidine and morphine, and we investigated potential associations between pre-specified single nucleotide polymorphisms (SNPs) and PK or disease severity/outcome measures.

Samples collected from neonates enrolled in the No-POPPY trial (NCT03396588) up to early 2020 were used for analysis. PK samples from treated subjects on oral morphine or oral clonidine were analyzed for morphine, morphine-3-glucuronide, morphine-6-glucuronide and clonidine concentrations by LC-MS/MS. DNA was isolated from lysed whole blood or buccal swabs from treated and non-treated subjects and analyzed by TaqMan SNP genotyping assays using real-time PCR. Genetic association analysis was performed on COMT 472G>A and OPRM1 118A>G in all trial subjects, and on GNB3 825C>T and ADRA2A -1291G>C in clonidine-treated subjects. NONMEM (ver 7.3) was used to build population PK models and identify covariates associated with PK variability, including SNPs potentially associated with PK of morphine/clonidine (OCT1*2-*5, ABCC3 -211C>T, and CYP2D6 metabolizer class).

In the genetic association analysis, in contrast to a previous report, the relationship between COMT 472G>A or OPRM1 118A>G and hospital length of stay or other outcome measures was inconclusive. Clonidine-treated patients with TT genotype in GNB3 825C>T had an average length of stay that was 3.7 days shorter than those with CC/CT genotypes (p=0.045), which did not meet experiment-wise significance. Significant associations between several clinical factors and one or more measures of NAS severity were identified, including breastfeeding and maternal use of benzodiazepines and gabapentin during pregnancy. In the population PK analysis, one-compartment models, with allometric scaling incorporated a priori, were used for clonidine and morphine. Age was a significant covariate in both models, and a sigmoidal maturation model incorporating postnatal age on clearance provided a good fit to clonidine concentration data. The clonidine PK model was successfully used to simulate alternative initial dosing regimens that were more likely to achieve earlier symptom stabilization. Genetic factors evaluated in the population PK analysis did not significantly affect the disposition of clonidine or morphine. Within the scope of this project, clinical factors appeared to be more important factors affecting disposition of treatment agents and NAS severity/treatment outcomes than the genetic factors evaluated. As the next step, efforts will be made to extend current analysis as the clinical trial continues to enroll subjects.

Digital Object Identifier (DOI)

Funding Information

This research was supported by National Institutes of Health grant R01DA043519 from 2017-2020.