Old Polymorph, New Technique: Assessing Ritonavir Crystallinity Using Low-Frequency Raman Spectroscopy

Two decades ago, postmarket discovery of a second crystal form of ritonavir with lower solubility had major implications for drug manufacturers and patients. Since then, ritonavir has been reformulated via the hot–melt–extrusion process in an amorphous form. Here, quantitative low- and mid-frequency...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Analytical chemistry (Washington) 2023-10, Vol.95 (41), p.15325-15332
Hauptverfasser: Hatipoglu, Manolya K., Zaker, Yeakub, Willett, Daniel R., Gupta, Nirzari, Rodriguez, Jason D., Patankar, Suhas, Capella, Peter, Yilmaz, Huzeyfe
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Two decades ago, postmarket discovery of a second crystal form of ritonavir with lower solubility had major implications for drug manufacturers and patients. Since then, ritonavir has been reformulated via the hot–melt–extrusion process in an amorphous form. Here, quantitative low- and mid-frequency Raman spectroscopy methods were developed to characterize polymorphs, form I and form II, in commercial ritonavir 100 mg oral tablets as an alternate analysis approach compared to X-ray powder diffraction (XRPD). Crystallization in three lots of ritonavir products obtained from four separate manufacturers was assessed after storage under accelerated conditions at 40 °C and 75% relative humidity (RH). Results were compared with quantitative XRPD methods developed and validated according to ICH Q2 (R1) guidelines. In a four-week open-dish study, form I crystallization occurred in two of the four products and form II crystallization was detected in another ritonavir product. The limits of detection for XRPD, low-frequency Raman (LFR), and mid-frequency Raman (MFR) were determined to be 0.7, 0.8, and 0.5% for form I and 0.6, 0.6, and 1% for form II, respectively. Root-mean-squared-error of predictions were 0.6–1.0 and 0.6–2.5% for LFR- and MFR-based partial least-squares models. Further, ritonavir polymorphs could also be identified and detected directly from ritonavir tablets using transmission LFR. In summary, LFR was applied for the assessment of polymorphism in real-world samples. While providing analytical performance similar to conventional techniques, LFR reduced the single measurement time from 66 min (XRPD) to 10 s (LFR) without the need for tedious sample preparation procedures.
ISSN:0003-2700
1520-6882
DOI:10.1021/acs.analchem.3c02781