Simple hybrid predictive model for point-of-care pharmaceutical tablet disintegration

Area of Science:

• Pharmaceutical Sciences
• Materials Science
• Chemical Engineering

Background:

• Consistent disintegration is vital for oral solid dosage (OSD) therapeutic efficacy.
• Predicting OSD performance from manufacturing data is challenging.
• Current models lack mechanistic insight and longitudinal validity.

Purpose of the Study:

• To develop a simple, hybrid model for predicting longitudinal disintegration times (t_disint,T) of OSDs.
• To establish a framework anchored in USP <701> standards for regulatory compliance.
• To offer a mechanistically grounded, adaptable alternative to traditional testing.

Main Methods:

• Developed a hybrid model predicting t_disint,T solely from at-line porosity measurements (f_at-line).
• Ensured model architecture is agnostic to specific porosity characterization techniques.
• Utilized a five-year real-time stability dataset with diverse formulations and tablet geometries.

Main Results:

• Achieved exceptional predictive accuracy (R²>0.92) across all longitudinal time points.
• Established global models applicable within the typical OSD porosity range (f_at-line∈[0.1,0.3]).
• Demonstrated model parameters as quantitative indicators for stability assessments.

Conclusions:

• The model provides a pragmatic path to demonstrating dosage form efficacy at point-of-care.
• Facilitates Quality by Design, technical transfer, and real-time release testing.
• Ensures preservation of quality standards until patient administration.