(M1430-03-18) Volumetric Printing and Non-destructive Dose Verification of Water-Soluble Vat Photopolymerized Tablets Based on Supramolecular Chemistry

Purpose: Volumetric 3D printing (3DP) is a novel vat photopolymerization-based additive manufacturing technology that affords high dimensional accuracy and rapid printing speeds . By irradiating a photocurable resin with light patterns of an object viewed at different angles, the entire 3D geometry may be produced simultaneously and within seconds. However, vat photopolymerized drug delivery systems reported thus far have been insoluble, limiting applications to sustained-release systems or insoluble drug-eluting medical devices. Resins that produce water-soluble matrices will eliminate the risk of gastrointestinal obstructions and expand applications of the technology in the pharmaceutical space. In this study, a novel photocurable drug-loaded formulation based on supramolecular chemistry was investigated for producing water-soluble 3D-printed tablets (printlets) via volumetric 3DP. Additionally, for the first time, near-infrared spectroscopy (NIRS) was used for non-destructive dose verification of vat photopolymerized printlets.

Methods: Formulation preparation: Eight formulations comprising lithium phenyl-2,4,6-trimethylbenzoylphosphinate (LAP), [2-(acryloyloxy)ethyl]trimethylammonium chloride solution (TMAEA), red food colourant, water, and different amounts of warfarin sodium (0.5 – 6.0 wt%) were prepared. The formulations were mixed at room temperature until all components dissolved. Tomographic volumetric printing process: Two different sized torus shapes were created using Microsoft Paint and uploaded onto an in-house software. The software projects the image onto the resin container located 23 cm away from the 385 nm digital light projector. Before UV light exposure, the formulations were loaded into the resin container and rotated at 30 RPM. After light irradiation, the printlets were washed with isopropanol for 1 min and then post-cured under UV light at room temperature for 30 mins. Characterization of printlets: The printlets were characterized by DSC, TGA, FTIR, and NMR. Printlet and resin drug loading was obtained by dissolving the printlets and resins overnight and quantifying with HPLC. In vitro drug release studies were performed using a USP-II apparatus, with paddle speed set at 50 rpm and temperature set at 37 ± 0.5 ºC. Printlets were dropped in 750 mL of 0.1 M HCl for 2 h, then 250 mL of 0.2 M trisodium phosphate solution was added to adjust the pH to 6.8. Non-destructive dose verification: The NIR reflectance spectra of all printlets were measured using a MicroNIR 1700ES NIR spectrometer. Seven formulations were used to build the calibration model, and the remaining (3.5 wt% warfarin sodium) was used for external validation. Ten spectra were obtained for each printlet, and the raw spectral data was pre-processed by standard normal variate (SNV) scatter correction and taking the first derivative spectra by application of Savitzky-Golay filter (window size of five, second order polynomial). Five latent variables were used to construct the partial least square (PLS) regression model.

Results: All formulations were successful printed with uniform weight and size (Figure 1, scale in cm). Printing times ranged between 6.5 to 11.1 s, increasing proportionally with the formulation’s drug load. As the warfarin sodium loading increased, the printlets became visibly more opaque, likely due to the precipitation of warfarin sodium in printlets with higher drug loading after post-curing. DSC and TGA analysis confirmed the evolution of water from the printlets, characterised by the broad endotherm stretching from 60 to 150 ºC in the printlets’ DSC thermogram. The endothermic peak corresponding to the melting of warfarin sodium at 195 ºC was not detected in the printlets’ DSC thermograms, likely because the amount of crystalline warfarin sodium in these samples fell below the calorimeter’s detection limit. FTIR and NMR confirmed the incorporation of warfarin sodium and that radical polymerization had reached near completion. Two separate warfarin sodium loading PLS regression models based on NIR spectra of large and small printlets, respectively, were built (Figure 2). Both models showed strong linearity (R² (10-fold cross-validation) = 0.969 and 0.970 for large and small printlets, respectively), and high accuracy in predicting the drug loading of the test sample (VOL035) (RMSEP = 0.368% and 0.202% for large and small printlets, respectively). All printlets completely dissolved during the in vitro drug release study, and 100% drug release was achieved between 2.5 – 7 hrs (Figure 3). The small-sized torus printlets released their drug load faster than their larger equivalents, as they possessed a higher surface area to volume ratio. Drug release rate also generally increased as the warfarin sodium loading in the formulations increased. This is because warfarin sodium is a non-reactive component in the formulation and does not participate in photopolymerization, resulting in less dense polymer networks that facilitates faster sorption of dissolution media into the matrix nucleus.

Conclusion: Water-soluble printlets were successfully fabricated, for the first time, via volumetric 3DP within 6.5 – 11.1 s. NIRS was successful used to non-destructively verify the amount of warfarin sodium loaded into each printlet. The dose and release profile of the printlets could be personalized by altering the relative amount of warfarin sodium in the formulation and the size of the torus-shaped printlet.

References: 1. Rodríguez-Pombo, L., et al., Volumetric 3D printing for rapid production of medicines. Additive Manufacturing, 2022. 52: p. 102673.
2. Kelly, B.E., et al., Volumetric additive manufacturing via tomographic reconstruction. Science, 2019. 363(6431): p. 1075-1079.

Acknowledgements: A.K.J. acknowledges the UK Engineering and Physical Sciences Research Council (EPSRC) for financial support (EP/S023054/1).
A.W.B. and A.G. are founders of the pharmaceutical company FabRx. The company had no role in the writing of the abstract. The other authors declare no conflicts of interest.

Pictures of (from left to right) VOL005, VOL010, VOL020, VOL030, VOL035, VOL040, VOL050, and VOL060 printlets. Small-sized printlets are positioned on the top row, while large-sized printlets are positioned on the bottom row. Scale in cm.

Partial least squares regression models for predicting warfarin sodium loading in (A) large and (B) small printlets.

In vitro drug release data of large printlets (left) and small printlets (right).