Subcutaneous infusion of high-dose benzathine penicillin G is safe, tolerable, and suitable for less-frequent dosing for rheumatic heart disease secondary prophylaxis: a phase 1 open-label population pharmacokinetic study

Abstract

Since 1955, the recommended strategy for rheumatic heart disease (RHD) secondary prophylaxis has been benzathine penicillin G [BPG; 1.2 MU (900 mg)] injections administered intramuscularly every 4 weeks. Due to dosing frequency, pain, and programmatic challenges, adherence is suboptimal. It has previously been demonstrated that BPG delivered subcutaneously at a standard dose is safe and tolerable and has favorable pharmacokinetics, setting the scene for improved regimens with less frequent administration. The safety, tolerability, and pharmacokinetics of subcutaneous infusions of high-dose BPG were assessed in 24 healthy adult volunteers assigned to receive either 3.6, 7.2, or 10.8 MU (three, six, and nine times the standard dose, respectively) as a single subcutaneous infusion. The delivery of the BPG to the subcutaneous tissue was confirmed with ultrasonography. Safety assessments, pain scores, and penicillin concentrations were measured for 16 weeks post-dose. Subcutaneous infusion of penicillin (SCIP) was generally well tolerated with all participants experiencing transient, mild infusion-site reactions. Prolonged elevated penicillin concentrations were described using a combined zero-order (44 days) and first-order (t_1/2 = 12 days) absorption pharmacokinetic model. In simulations, time above the conventionally accepted target concentration of 20 ng/mL (0.02 µg/mL) was 57 days for 10.8 MU delivered by subcutaneous infusion every 13 weeks compared with 9 days of every 4-weekly dosing interval for the standard 1.2 MU intramuscular dose (i.e., 63% and 32% of the dosing interval, respectively). High-dose SCIP (BPG) is safe, has acceptable tolerability, and may be suitable for up to 3 monthly dosing intervals for secondary prophylaxis of RHD.

Keywords: benzathine penicillin G; population pharmacokinetics; rheumatic heart disease; secondary antibiotic prophylaxis; subcutaneous infusions.

Fig 1

Consort diagram for trial participants.

Fig 2

Structure of the pharmacokinetic model. The best model consisted of two absorption paths, one a bolus into the depot compartment and the other a zero-order process into the same depot modeled as the duration. Absorption from this depot compartment was through a transit compartment with corresponding transit half-life (t_{1/2, tr}) before absorption (t_{1/2, abs}) into the main compartment (V/F). k_el represents the elimination rate constant.

Fig 3

Prediction-corrected visual predictive checks for plasma benzylpenicillin G concentrations (ng/mL on log10 scale). Observed 50th (solid line) and 10th and 90th (dotted lines) percentiles within their simulated 95% CI (gray shaded areas) are shown with overlying data points (ο, n = 400).

Fig 4

Simulations of the pharmacokinetic model comparing high-dose subcutaneous infusions of benzathine penicillin G (10.8 MIU, dark gray) with intramuscular injection (1.2 MIU, light gray). The shaded region represents median with 90% simulation interval.