Near-infrared spectroscopy for the determination of testosterone in thin-film composites

• Published in: Journal of pharmaceutical and biomedical analysis 2003-09, Vol.33 (2), p.181-189
• Main Authors: Fountain, William, Dumstorf, Karen, Lowell, Amanda E., Lodder, Robert A., Mumper, Russell J.
• Format: Article
• Language: English
• Subjects: Acrylic Resins; Algorithms; Analytical method; Analytical, structural and metabolic biochemistry; Biological and medical sciences; Buccal; Cost-Benefit Analysis; Dosage Forms; Excipients; Fundamental and applied biological sciences. Psychology; Mucoadhesive; Nondestructive; Principal component regression; Protein hormones. Growth factors. Cytokines; Proteins; Quality Control; Reference Standards; Reproducibility of Results; Scattering, Radiation; Spectrophotometry, Ultraviolet; Spectroscopy, Near-Infrared; Testosterone - analysis; United States; United States Food and Drug Administration
• ISSN: 0731-7085; 1873-264X
• DOI: 10.1016/S0731-7085(03)00345-5

More rapid, reproducible, and cost-effective methods to control product quality in the pharmaceutical industry continue to be a major emphasis, particularly with the FDA through its recent process analytical technologies (PAT) initiative. Many different methods have been used to determine the stability and content uniformity of a drug in various dosage forms; however, most of these methods include the destruction of the sample. Therefore, the development of nondestructive methods that allow the analysis of each individual dosage form has become the basis of much research. A new assay for the nondestructive determination of testosterone content in mucoadhesive bi-layer thin-film composites (TFCs) using near-infrared spectroscopy (NIR) was developed. Five sets of the circular films (n=5) with theoretical testosterone content of 0, 1, 2, 3, and 4 mg per 3/8th in. diameter disks were scanned in the near-infrared region of 1100–2500 nm to determine testosterone content. The NIR results were directly compared with those obtained using a previously developed ultraviolet assay for testosterone at 240 nm. Principal component regression (PCR) was performed to calibrate the NIR assay. This correlation produced r 2=0.99 with a standard error of estimate (SEE)=0.18 mg, and a standard error of performance (SEP)=0.18 on cross validation with an equal number of samples ( F test passed at P=0.05). Though the UV assay showed a slightly better r 2 value, the NIR assay was much quicker, easier, and nondestructive. Therefore, the NIR assay may have significant potential for use in the quality control of pharmaceutical films containing drugs.