Understanding Stratified Sampling in Pharmaceutical Industry

by

FDA recommends the necessity of statistically sound sampling plans [Intra-Location (within a single location) Variability and Inter-Location (between locations) Variability] and acceptance criteria [acceptable range of deviation from the mean content, the variability of measurements, and the percentage of units falling within predetermined limits is based on statistical principles] to assess Blend and Content Uniformity highlights the significance of rigorous quality control processes to align with the cGMP outlined in 21 CFR 211.110 to ensure that the drug products meet the required standards for quality and efficacy.

  • Stratified sampling = probability sampling method that involves dividing into distinct subgroups (=strata), and then taking a sample from each stratum.

Sampling and Assessment of Blend uniformity and content uniformity during Process Design and Process Qualification Batches –


BLEND UNIFORMITY TESTING

At minimum, 3 blend samples (3 x 1) need to be taken from at least 10 distinct locations within the blender.

Stage 1 – Assay on one sample from each location (1 x 10), Calculate SD.

If the SD ≤ to 3.0% of the target, the batch moves forward to Stage 1 dosage unit testing.

If the SD exceeds 3.0% of the target, the process must undergo a more detailed evaluation through Stage 2 blend testing.

Stage 2 – Assay replicate blend samples from each location i.e., remaining (2 x 10)], Calculate SD.

If the SD is found to be 3.0% or less, the process can advance to Stage 1 dosage unit testing.

If the SD is 3.1 to 5.0 % of target, blend uniformity is acceptable. However, CONTENT UNIFORMITY TESTING should be conducted at the STAGE 2 to provide increased assurance that the blend produces acceptable dosage forms.

If the SD exceeds 5.0%, INVESTIGATE.


CONTENT UNIFORMITY TESTING

At minimum, 3 in-process dosage units (3 x 1) need to be taken from at least 40 equally spaced predetermined locations throughout the batch (including the beginning and end of the run)

Stage 1 – Assay at least 3 dosage units per location from at least 20 (3 x 20) locations. Calculate SD.

All individual values should be within 75.0 – 125.0 % and complies with statistical test to provide an appropriate level of assurance to comply with USP <905> for 20 x 3.

If not meeting, proceed to STAGE 2.

Stage 2 – Assay at least 3 dosage units per location from the remaining 20 (3 x 20) locations. Calculate SD.

All individual values should be within 75.0 – 125.0 % and complies with statistical test to provide an appropriate level of assurance to comply with USP <905> for 40 x 3.

If not meeting, dosage units are not uniform. INVESTIGATE.

For advanced statistical analysis ASTM E2709 and E2810 is recommended.


Read also:


Resource Person: Bharathi Kodali

Leave a Comment