Jukka Rantanen

Three-dimensional (3D) printing can produce tablets for individual patients at the point-of-care, e.g. in a hospital pharmacy. The quality of these tablets needs to be verified before they are dispensed. However, conventional quality control is not feasible for point-of-care manufacturing. In this study, the suitability of near-infrared (NIR) and Raman spectroscopies was explored for the assessment of furosemide drug content in 3D-printed tablets. Tablets of three different diameters (2.7, 4.5 and 6.2 mm) containing furosemide 10 – 25% w/w, with a dose range of 3.5 – 35 mg furosemide per tablet, in a polymer matrix of Gelucire 48/16 and glycerol, were produced using a semi-solid extrusion 3D printer. Weight distribution was determined and content uniformity was assessed using HPLC-UV. Content data were linked to NIR and Raman spectra. Subsequent partial least squares regression (PLS) models encompassed good predictive values (root mean square error of prediction (RMSEP) = 0.8242%; RMSEP = 1.307%, respectively). NIR and Raman spectroscopies are suitable for non-destructive analysis of personalized 3D-printed furosemide tablets. NIR provides robust bulk quantification, while Raman spectroscopy gives detailed chemical insight, especially including the low-frequency region. Subsequent PLS regression provides an easy approach for clinical practice, due to its transparency, robustness and ease of validation.