Technical Guidelines for Dissolution Testing of Oral Solid Dosage Forms (Part Two)

I. Drug Changes and Dissolution Testing Strategies

In the lifecycle management of drugs post-marketing, change control is a key process to ensure continuous product quality stability. Depending on the extent of changes, differentiated dissolution study strategies are required to assess their impact. For minor changes (e.g., non-critical process parameter adjustments, changes in excipient suppliers), single time-point dissolution measurements are usually sufficient to demonstrate that the change has not adversely affected product quality. In such cases, it is recommended to select the most discriminative time point (usually as specified in quality standards) for measurement, requiring that the difference in dissolution between pre- and post-change samples does not exceed 10%.

For major changes (e.g., formulation composition changes, critical process modifications, transfer of production sites), comprehensive comparative studies of dissolution profiles are necessary. This research requires multi-point sampling under identical experimental conditions before and after the change to establish complete dissolution curves. The criteria for judgment must meet two conditions: overall similarity in shape of the dissolution curves and acceptable differences at each time point's dissolved amount. To quantify this similarity assessment internationally accepted statistical methods that do not rely on models are commonly used; these mainly include difference factor (f1) and similarity factor (f2).

II. Evaluation Methods for Similarity of Dissolution Curves

2.1 Calculation and Application of Difference Factor (f1) The difference factor quantifies relative deviations between two curves at various time points to evaluate their degree of difference using: f1 = {[Σ|Rt - Tt|] / [ΣRt]} × 100, where n represents sample size; Rt indicates percentage dissolved from reference sample at t; Tt indicates percentage dissolved from test sample at t. Mathematically speaking, f1 reflects average absolute deviation between two curves with values ranging from 0-100; an f1 value of 0 indicates perfect agreement while higher values indicate increasing degrees of discrepancy. Regulatory practices typically set f1<15 as a standard for determining similarity.

2.2 Calculation and Application of Similarity Factor (f2) The similarity factor assesses degree similarities through root mean square error following logarithmic transformation: f2 = 50 × log{[1+(1/n)Σ(Rt-Tt)^2]^-0.5 × 100}. This formula amplifies larger discrepancies via squaring operations while reducing numerical ranges through logarithmic transformations with f2 values also ranging from 0-100 where higher numbers signify greater similarities; regulatory agencies generally require f2≥50 to determine curve likenesses. It’s noteworthy that calculations involving f2 can be sensitive regarding number/positioning among sampled points—ideally selecting at least three or four sampling points ensuring final one’s dissolving quantity doesn’t exceed 85%.

III. Operational Specifications for Dissolution Tests

... and so forth.