Poster J9 - Optimization of PLGA Microspheres Encapsulating Melphalan for Localized Chemotherapy Delivery

Friday, October 25, 2024

10:00 AM – 11:00 AM EST

Location: Exhibit Hall E, F & G

Presenting Author(s)

Wesley Hargrove

Student Researcher
University of Florida
Lakeland, Florida, United States

Co-Author(s)

RC

Ryan A. Clark, BS

Graduate Research Associate
University of Florida
Gainesville, Florida, United States

Primary Investigator(s)

BK

Benjamin G. Keselowsky

Professor
University of Florida, United States

Introduction: Melphalan is an established pharmaceutical chemotherapeutic drug used to treat various cancers. Melphalan is a bi-functional alkylating drug in the nitrogen mustard group and exerts its anti-cancer activity by crosslinking DNA to disrupt DNA replication and transcription. It is typically given orally or intravenously, primarily treating hematologic cancers like multiple myeloma, but also localized cancers like ovarian cancer. However, systemic delivery is associated with many severe side effects including bone marrow suppression, nausea, and gastrointestinal tract disruption. Despite these toxicities, few alternative delivery strategies have been studied to localize melphalan delivery with the goal of localizing delivery. Administering a controlled release depot of melphalan could reduce the severity of dose-limiting side effects for applications in localized cancers. While there are many drug delivery vehicles that could do this, an ideal one would be a steady release microparticle. Poly(lactic-co-glycolic acid) (PLGA) microparticles (MPs) can be used to generate a controlled release drug depot that can be injected at a desired site. Here we compare the loading efficiency of melphalan in PLGA MPs using single or double emulsion solvent evaporation techniques.

Materials and

Methods: MPs were formulated with an emulsification-solvent evaporation method. This is a low-cost, fast, and reproducible method that enables tuning of particle size and drug loading concentration. Melphalan is poorly soluble in water, chlorinated solvents, and acetonitrile, so we tested several loading buffers for single or double emulsion based on solubility and miscibility. Melphalan was dissolved in methanol, dimethyl sulfoxide, and tris-citrate. The tris-citrate solution was cloudy so we also tested if sterile filtering the solution would make a difference for drug loaded. The drug was dissolved to saturation in the solvents, added to a PLGA and methylene chloride solution, then agitated with a dremel if it was a double emulsion. This was then added to a 2.5% PVA solution and agitation with a dremel again. The microparticles were left to evaporate in a 1% PVA solution, centrifuged out, washed, and lyophilized.
Loading was measured based on absorbance after extraction from MPs. We first tried extracting melphalan with a dual-solvent system where solubility differences between melphalan and PLGA enable separation. We tested four combinations of water or methanol with dichloromethane or chloroform. We compared this to dissolving MPs in a 0.2 N NaOH solution instead of the dual solvent mixture. Mixtures were vortexed to mix, then centrifuged to pellet out PLGA. Each combination was compared against spectrum reads of blank PLGA microparticles and dissolved known quantities of melphalan. We confirmed the size distribution with DLS.

Results, Conclusions, and Discussions: Absorbance values were taken at 258 nm.
The unfiltered tris-citrate MPs had a 2.79 um mean size, a .751 polydispersity index (PDI), a 1.868 ug of melphalan per mg of PLGA loaded concentration, and a 12.44% loading efficiency. The filtered tris-citrate MPs had a 1.36 um mean, a .126 PDI, a .709 ug/mg concentration, and a 7.773% efficiency. The DMSO particles had a 1.56 um mean, a .278 PDI, a .348 ug/mg concentration, and a 1.375% efficiency. The methanol particles had a 3.27 um mean, a .936 PDI, a .288 ug/mg concentration, and a 1.9% efficiency.

The double emulsion formulation with tris-citrate appears to provide higher encapsulation efficiency for melphalan than single emulsion formulations.. We wanted a low polydispersity ( < 0.3) to generate a monodisperse size distribution of MPs. Polydispersity was low for some of the MPs but was higher than ideal (>0.7) for the unfiltered tris-citrate and DMSO. Loading measures for melphalan-loaded MPs can be performed easily with sodium hydroxide dissolution of the MPs.

Tris-citrate double emulsion provided the best loading efficiency, but melphalan loss was still a limitation. We hope to build on this progress to test alternative solvents that can provide even better loading efficiencies.
The unfiltered tris-citrate had a better loading mass of melphalan, but worse polydispersity. This could be because various small crystals of melphalan became encapsulated. This may be the preferable method if size distribution is less important, but these particles may also be less predictable. Additionally, kinetic studies will be done to determine the release profile of melphalan. With our improved melphalan loading, we hope to translate this method to similar poorly soluble alkylating agents.
Of the loading measurements, the sodium hydroxide solution was the easiest and read most clearly out of all the mixtures. Melphalan was very soluble in the basic solution and it effectively broke down the MPs to release melphalan. Although melphalan was poorly soluble in DCM and chloroform, it is also not significantly more soluble in water or methanol so our dual-solvent separation was not ideal for melphalan extraction.

Acknowledgements (Optional): I would like to acknowledge and thank Ryan Clark, my graduate student mentor, and Dr. Keselowsky for the guidance and the training they have given me. Most of this work was done in tandem with Ryan Clark under his supervision and advice. I appreciate the time they’ve sacrificed to teach me and take me in the lab.