Automated PET-RAFT Polymerization Towards Pharmaceutical Amorphous Solid Dispersion Development

Abstract

In pharmaceutical oral drug delivery development, about 90% of drugs in the pipeline have poor aqueous solubility leading to severe challenges with oral bioavailability and translation to effective and safe drug products. Amorphous solid dispersions (ASDs) have been utilized to enhance the oral bioavailability of poorly soluble active pharmaceutical ingredients (APIs). However, a limited selection of regulatory-approved polymer excipients exists for the development and further understanding of tailor-made ASDs. Thus, a significant need exists to better understand how polymers can be designed to interact with specific API moieties. Here, we demonstrate how an automated combinatorial library approach can be applied to the synthesis and screening of polymer excipients for the model drug probucol. We synthesized a library of 25 random heteropolymers containing one hydrophilic monomer (2-hydroxypropyl acrylate (HPA)) and four hydrophobic monomers at varied incorporation. The performance of ASDs made by a rapid film casting method was evaluated by dissolution using ultra-performance liquid chromatography (UPLC) sampling at various time points. This combinatorial library and rapid screening strategy enabled us to identify a relationship between polymer hydrophobicity, monomer hydrophobic side group geometry, and API dissolution performance. Remarkably, the most effective synthesized polymers displayed slower drug release kinetics compared to industry standard polymer excipients, showing the ability to modulate the drug release profile. Future coupling of high throughput polymer synthesis, high throughput screening (HTS), and quantitative modeling would enable specification of designer polymer excipients for specific API functionalities.

Keywords: amorphous solid dispersions; combinatorial polymer synthesis; drug precipitation inhibition; oral drug delivery; polymer excipients.

Figure 1.

Experimental design schematic. Polymers were designed to participate in polar and non-polar interactions with the model drug probucol. In these random heteropolymers, a hydrophilic monomer (HPA) was copolymerized with 0–60 mol% of hydrophobic monomers (MA, BA, HA, and CHA). Automated polymer synthesis was conducted with liquid handling robotics to dispense correct volume of monomer, CTA, and initiator into 96-well plates. Formulation of polymer/API was assessed by a film casting screening experiment. Further characterization of polymer/API miscibility and crystallization was completed using mDSC, XRD, and polarized light microscopy.

Figure 2.

Schematic of polymers and model API. (A) Structures of all random heteropolymers which contained a hydrophilic monomer (HPA) with hydrophobic monomers (MA, BA, HA, and CHA) incorporated at 0–60 mol%. (B) Structure of model API probucol. Note that the log P of monomers and probucol are included above the structures.

Figure 3.

Dissolution plots demonstrating the effect of hydrophobic monomer incorporation. Plots contain hydrophilic monomer HPA along with 10–60 mol% of hydrophobic monomers (A) MA, (B) CHA, (C) BA, and (D) HA. As shown, the API was also film casted in the same manner but without polymer excipient. Drug release was quantified by UPLC and obtained by comparison to API standard injections (final concentration of 834 μg/mL). Film casting was completed in triplicate and the error bars represent the standard error.

Figure 4.

Comparison of API formulation ability of synthesized polymers and inclusion of conventional polymers. (A) Comparison of dissolution performance of HPA copolymers incorporated with HA and CHA monomers at t = 330 min. Drug release was quantified by UPLC and obtained by running standard injections of API. Statistical significance was determined by student’s t-test where * is p < 0.05 and ** is p < 0.01. (B) HPA homopolymer and two of the top random heteropolymers synthesized (HPA-30% BA and HPA-20% HA) are displayed alongside standard polymer excipients PVP/VA and HPMCAS-MF. Concentration was quantified by UPLC by running standard injections of the API probucol. Experiments were completed in triplicate as error bars represent standard error.

Figure 5.

Transmission XRD profiles of film casted polymer/drug mixtures. XRD profiles are grouped by hydrophobic monomer incorporation: (A) MA, (B) BA, (C) HA, and (D) CHA at 20 wt% drug loading. At the time of collection, amorphous to crystalline form conversion of polymer/drug films was not observed.

Figure 6.

mDSC thermograms (exotherm up) for selected polymer/probucol films. mDSC first heat cycle (A) reversing heat flow and (B) total heat flow plots along with second heat cycle (C) reversing heat flow and (D) total heat flow plots for polymer loaded with 20 wt% API. The representative polymers displayed are HPA homopolymer, HPA-30% BA, HPA-20% HA, HPA-30% MA, and HPA-30% CHA. Respective T_g values are displayed for the second heat cycle (C). There is a varying degree of crystallinity and miscibility of polymer and drug exhibited.

Figure 7.

Polarized light microscopy contrasting crystalline API probucol to amorphous polymer/drug formulations. (A) Probucol was prepared in acetone and diluted into PBS (2 vol% acetone) and subsequently film casted onto a glass slide. Polymer/drug mixtures prepared in acetone and film casted directly include (B) HPA homopolymer, (C) HPA-30% BA, and (D) HPA-20% HA with 20 wt% probucol. The scale bar represents 200 μm.