The synthesis of indomethacin prodrugs for the formation of nanosuspensions by emulsion templated freeze drying

Abstract

Emulsion-templated freeze drying (ETFD) is a versatile technique for producing nanosuspensions of poorly water-soluble drugs, but predicting formulation success remains a significant challenge. In this study, we investigate how structural modification of the model drug indomethacin, through esterification with a series of alkyl and aromatic groups, influences nanosuspension formation via ETFD. A panel of seven indomethacin prodrugs was synthesised and screened across binary combinations of water-soluble stabilisers. The resulting formulations were assessed based on particle diameter, polydispersity index (PDI), and visual dispersion quality. Analysis of stabiliser combinations revealed specific systems that consistently supported nanoparticle formation across multiple prodrugs. Additionally, there was a positive relationship between increased hydrophobicity, represented by the calculated log P, and the formation of viable nanosuspensions. Moreover, the stability of these nanosuspensions was assessed, revealing that esters with higher log P values exhibited better dispersion stability. The findings provide valuable insights into the selection of active pharmaceutical ingredients for nanosuspension formulation and further the understanding of the influence of drug properties on nanosuspension stability and production. This research contributes to the development of effective nanosuspension strategies for a wide range of poorly water-soluble pharmaceutical compounds.

Fig. 1. Synthesis of indomethacin prodrugs for the formation of nanosuspensions by emulsion templated freeze drying. (A) Seven hydrophobically modified indomethacin prodrugs were prepared using Steglich esterification on indomethacin using dicyclohexylcarbodiimide (DCC) as a coupling agent and 4-dimethylaminopyridine (DMAP) as a catalyst. Indomethacin or one of the indomethacin prodrugs was used as the API in the ETFD process. 42 different binary combinations of surfactant and polymer stabilisers were screened in the ETFD process. (B) The ETFD process was comprised of four stages: (i) the API was dissolved in chloroform and added to an aqueous solution of stabilisers. (ii) The sample was sonication to produce an emulsion. (iii) The sample was freeze dried resulting in a porous monolith. (iv) The nanosuspension was produced upon the addition of water to the monolith.

Fig. 2. Screening indomethacin nanosuspension formulations by the ETFD method using binary combinations of stabilisers. (A) Summary of the data obtained for 10 wt% indomethacin loading formulations, using assessment criteria of mean diameter <500 nm, a standard deviation of <10% between the triplicate diameter measurements, the absence of any visible particles in the dispersion, and a PDI <0.3. (B) Summary of the data obtained for 30 wt% indomethacin loading formulations, using assessment criteria of a mean diameter of <600 nm, a standard deviation <10%, the absence of any visible particles in the dispersion and a PDI <0.5. In both cases, the polymeric type stabilisers are found on the x-axis and the surfactant type stabilisers are those on the z-axis. Samples that met the screening criteria are shown in green, samples that did not fully disperse or had a diameter of >1 μm are represented as red circles and unsuccessful samples in red (i.e. where the diameter or PDI were above the specified assessment criteria, or standard deviation between the triplicate diameter measurements was of >10%). The full names and structures of the different stabilisers are shown in Fig. S1.

Fig. 3. Analysis of the viable indomethacin nanosuspensions producing at 30% wt loading. (A) Size distribution graphs of HPMC:Tween 20 and HPMC:Tween 80 30% wt indomethacin nanosuspension as measured by DLS. (B) Reproducibility of 30% wt indomethacin nanosuspension formulations as measured by DLS to determine the mean diameter and PDI (in blue) using HPMC as the polymeric stabiliser and either Tween 20 or Tween 80.

Fig. 4. Production of hydrophobic esterified prodrugs of indomethacin. (A) The structures of the hydrophobically modified indomethacin esters synthesised and their log P values. (B) The relationship between the mass percentage of the indomethacin API within the prodrug based on the alkyl/aryl modification.

Fig. 5. Characterisation of the ethyl ester indomethacin prodrug. (A) FTIR spectra of the indomethacin ethyl ester showing significant carbonyl stretches at 1726 (ester) and 1673 cm⁻¹ (amide). (B) ¹H NMR (CDCl₃, 400 MHz) spectrum for the indomethacin ethyl ester. (C) ¹³C NMR (CDCl₃, 400 MHz) spectrum for the indomethacin ethyl ester, the inset is focussed on the region 130–140 ppm.

Fig. 6. Mean particle properties (mean diameter and polydispersity index as measured by DLS) produced from the ETFD screening of indomethacin ester prodrugs at 30% wt. loading (benzyl, t-butyl, n-butyl hexyl and dodecyl) with binary combinations of polymer and surfactant stabilisers that produced viable formulations. Note that the ethyl and stearyl prodrugs and the polymer stabiliser PEG 1 K are omitted as they produced no viable formulations. The viable samples are shown by circles representing the successful binary combinations of stabilisers. Green boxes around the samples indicate that the samples met the more restrictive assessment criteria of a mean diameter of <350 ± 50 nm and PDI = ≤0.3 ± 0.1.

Fig. 7. A summary of the number of viable formulations produced for the different stabiliser combinations across all the indomethacin prodrugs at 30% wt. loading.

Fig. 8. The relationship between the number of viable samples achieved for each indomethacin prodrugs at 30 wt% loading and the calculated log P of the prodrug.

Fig. 9. Stability and particle properties of the 30 wt% indomethacin prodrug nanosuspension samples with NDC-PVA as the stabiliser combination. (A) Photo of the after (i) immediate dispersion when redispersed at 1 mg mL⁻¹ active in PBS (0.01 M) and (ii) after 6 hours of dispersion. The blue box highlights the sedimented solid seen with 30 wt% n-butyl indomethacin analogies. (B) Comparison of the mean diameter of the hexyl prodrug and dodecyl prodrug before and after 24 hours in the dispersed form. (C) DLS size intensity distributions of the hexyl ester and the dodecyl ester prodrugs 24 hours after dispersion.