(W-079) Characterizing Pharmacokinetics Effects on Measures of Overall Drug Exposure during Pregnancy

Yuhan Long, BS, BSChE – Graduate Student, Experimental and Clinical Pharmacology, University of Minnesota; Jessica Barry, BS – PhD Candidate, University of Minnesota; Catherine Sherwin, PhD – Vice President of Clinical Pharmacology and Pharmacometrics, Differentia Bio; Angela Birnbaum, PhD, FAES, MLS(ASCP) – Professor, Experimental and Clinical Pharmacology, University of Minnesota

Objectives: Pregnancy is a complex time when drug pharmacokinetics change rapidly. Therapeutic drug monitoring during pregnancy can help determine exposure despite sparse sampling. This study investigated the differences in pharmacokinetic parameters when calculated via multiple methods to characterize drug exposure in pregnant patients to inform the design of pregnancy outcomes studies.

Methods: A generic one-compartment model with first order absorption and 1000 mg twice daily dosing was used to simulate predicted concentrations (Julia v.1.9.2). Volume and clearance were set to 32 L and 4 L/h preconception increasing linearly by 0.12 L/wk and 0.032 L/h/wk, respectively, for 40-week gestation. Mid-gestational month trough samples (~10 samples/patient from 2-38 weeks) were simulated to determine 4 exposure metrics: 1) lowest concentration, 2) average concentration, 3) highest concentration, 4) trapezoidal AUC0-40weeks. To reduce duration bias, methods 1, 2, and 3 were time-weighted by multiplying by gestational age. For method 4, the first and last simulated trough concentrations were used as surrogates for 0 and 40 weeks before estimation with trapezoidal rule. Sensitivity analyses of gestational week, preconception volume and clearance, and rate of volume and clearance increase were performed.

Results: For sensitivity on gestational weeks, monthly samples introduced with additional months of gestation led to drop in exposure estimates for methods 1 and 2. Otherwise, all exposures increased proportionally with gestational age. For higher preconception clearances, exposure measures converged to 0 mg/L*wk due to complete elimination of drug. For higher preconception volumes, the exposure measures increased but started decreasing at nonphysiologically relevant volumes (37,000 L for highest, 17,000 L for lowest). Increasing the rate of clearance change decreased all exposures, converging to 0 mg/L*wk at high rates of clearance change with method 3 taking the longest to converge. Increasing the rate of change in volume increased all exposure estimates. The rate of volume increase during pregnancy affected which sample is the lowest concentration for calculation with method 1 due to its relation with changing clearance. At rates below 0.7 L/wk, the last sample was the minimum, but the first sample was the lower at rates above.

Conclusions: Although additional samples provide more data, additional sparse sampling from longer pregnancies can bias exposure estimates for child outcomes. Differences in exposure measures need to be considered across and within methods due to biases arising from changes in pharmacokinetics during pregnancy. Therefore, choice of exposure method and sampling strategy are important considerations for pregnancy outcome studies.

Citations: none