Category: Formulation and Quality
Purpose: Acetylcholinesterase inhibitors and NMDA antagonists (memantine) are used clinically in the management of Alzheimer’s Disease. Rivastigmine inhibits acetylcholinesterase and butyrylcholinesterase, and is available in both free base and tartrate salt formulations. Transdermal administration of rivastigmine lessens potential plasma concentration fluctuations. However, transdermal film has limited drug permeation due to variations in skin thickness, structure, and level of irritation. To overcome these drawbacks, microneedle (MN) drug delivery is an alternative. There are 5 types of MNs: solid, coated, dissolving, hollow, and hydrogel-forming. With rivastigmine possessing liquid properties at controlled room temperatures, this project focused on the formulation development using hollow MNs.
Methods: Hollow Microstructured Transdermal System is a integrated device containing actuator, glass injection cartridge, delivery spring, adhesive, hollow microstructured array and application spring which information is available at 3M, St. Paul, MN (https://www.3m.com/3M/en_US/drug-delivery-systems-us/technologies/microneedle/hollow-needle/). Each glass cartridge may house 0.5 mL to 2 mL of intradermal delivery solution. Isotonic 0.9% saline and 5% dextrose (D5W) were chosen as solutions for injection in this study. Since rivastigmine is lipophilic, four surfactants (Span 20 [HLB 8.6]; Span 80 [HLB 4.3]; Tween 40 [HLB 15.6] and Tween 80 [HLB 15.00] (Figure 1) were tested for compatibility. Additionally, being a liquid formulation, two preservatives (benzyl alcohol or chlorobutanol) were included in the formulation. Each saline formulation contained 0.9 g of NaCl, 5.0 g of surfactants and 1.0 g of benzyl alcohol (or 0.5g of chlorobutanol) per 100 mL. The dextrose formulation contained 5.36 g of surfactants, 1.0 g of benzyl alcohol (or 0.5 g of chlorobutanol) in a total of 100 mL D5W. After mixing, all formulations were visually observed to determine candidacy for further testing. Rivastigmine concentration was determined using HPLC. A mobile phase was prepared with 58% methanol and 42% of 0.05M sodium phosphate dibasic. Rivastigmine was added to the mobile phase and its detective wavelength was determined using a Cary 50 UV-visible spectrophotometer scan (n = 3; Agilent Technology, Santa Clara, CA). The AUC of each drug sample was quantified by HPLC.
Results: On visual observation of the samples in saline group, 2 of the surfactants (Span 20 and Span 80) separated into two phases and were thus removed from further formulation consideration. Furthermore, when preservatives were added into the remaining test groups, Tween 40 (with chlorobutanol) formed a white precipitation and was also eliminated from further testing. The three remaining combinations in saline group, Tween 80 (with chlorobutanol) and Tween 40 and Tween 80 (with benzyl alcohol) miscible (Figure 2a) and further tested. Rivastigmine was added to these 3 solutions and subsequently analyzed by HPLC. The LC conditions were injection volume 20 mL, flowrate 1.0 mL/min, column temperature 40 oC, detective wavelength 214 nm, and Symmetry C18 column (Waters, Milford, MA). The AUCs of these three saline sub-groups showed no differences (p < 0.05, n = 3).
On visual observation of the samples in D5W group, Span 20 (with either preservative) formed emulsion; Tween 40 and Tween 80 with benzyl alcohol formed colloid; while Tween 40 and Tween 80 with chlorobutanol formed microemulsion (Figure 2b). The AUCs of rivastigmine with Span 20 as a surfactant (with either preservative) in D5W groups were significantly lower than those (either preservative) with Tween 40 and 80, which AUCs were 14046±913, 18656±772, and 18730±1194 respectively (p < 0.05, n = 4).
Conclusion: This project focused on the formulation development of rivastigmine, a liquid drug substance, to be administered using a hollow MNs drug delivery system. Among the samples tested which included four surfactants and two preservatives, chlorobutanol with Tween 80 in saline, benzyl alcohol with either Tween 40 or Tween 80 in saline, chlorobutanol with either Tween 40 and Tween 80 in D5W were feasible formulations candidates. Chlorobutanol was superior to benzyl alcohol as a preservative in D5W solution for injection due to the formation of the microemulsion, but no significant difference was in saline.
Reina Kitagawa– 3rd year graduate student in pharmaceutics, MCPHS University, Revere, Massachusetts
Hoathixuan Le– Boston, Massachusetts
Steven Crosby– Boston, Massachusetts
Paul Kiritsy– Boston, Massachusetts
Monica Chuong– Professor of Pharmaceutical Sciences, MCPHS University, Boston, Massachusetts