Stratified Sampling In Blend Uniformity Testing Advocated By PQRI

Stratified sampling should be used to demonstrate blend and dosage unit content uniformity, PQRI's Blend Uniformity Working Group has concluded.

"By going to stratified sampling of dosage units, we are putting emphasis in testing where it gives you the most value, the true read of the product," Product Quality Research Institute BUWG Chair Tom Garcia, PhD, told FDA's Pharmaceutical Science Advisory Committee Nov. 28.

Stratified sampling is "the process of selecting units deliberately from various locations within a lot or batch or from various phases or periods of a process to obtain a sample," Garcia explained. Instead of only testing end-products, "stratified sampling specifically targets locations either in the blender or throughout the compression/filling operation which have a higher risk of producing failing content uniformity results."

Stratified sampling would apply only to process validation and "products where the active ingredients are introduced into the blend."

Blend uniformity "is not a value-added test in routine manufacturing," Garcia said, because "it does not represent what is really going on in the number of cases where we have sampling error....Sampling errors are common" because "sampling technology today is not capable of extracting small quantities of blend."

Pharmaceutical Science Advisory Committee member Arthur Kibbe, PhD, Wilkes University, asked if "poor uniformity in the batches that [the BUWG] had information on didn't predict poor uniformity in terms of the tablet product you made [then] why do blend uniformity?"

FDA Office of Pharmaceutical Science Deputy Director Ajaz Hussain, PhD, replied that the way blend uniformity testing is conducted "with thief samples is not adding any value." However, when done with new online technology, blend uniformity testing "could improve our understanding of the [manufacturing] processes and the quality" of the dosage units, he said.

FDA first proposed blend uniformity as a routine in-process test in a draft guidance issued in August 1999. The testing has been opposed by industry; the PQRI working group was formed to address the subject.

"The situations where non-uniform blends are made uniform" are "inherently dangerous and should be avoided," BUWG member Garth Boehm, PhD, explained. "That's why we advocate doing the blend uniformity testing in validation."

FDA's Hussain said that validation has remained a constant problem for the agency. GMP guidelines mandate that the 10 tablets used as the basis for releasing the batch "be a representative sample. In practice, we may be missing some of that."

"Is there a time or a place where we cannot do blend uniformity on every batch once we have manufactured it successfully for sometime," asked Eon Labs Manufacturing VP-Biopharmaceutics Leo Shargel, PhD, who represented the generic industry at the meeting.

Hussain pointed to the instability of magnesium stearate as an example of the need of blend uniformity testing, suggesting that the raw material attributes of the product are subject to change. If there were no blend uniformity test, "you have no way of assessing if" raw material changes "had an impact or not," he noted.

"As we go to a more complex dosage form, excipient homogeneity becomes critically important for many control-release formulations," Hussain added. "The PQRI proposal essentially addresses that in a more formal way by expanding or increasing the number of end-products" tested for content uniformity, he said.

"If your real issue is that end-product testing is inadequate...what you ought to do is simply go to more end-product testing," committee member Marvin Meyer, PhD, University of Tennessee-Memphis, said, affirming the BUWG proposal.

The BUWG proposal "is a band-aid right now," Hussain explained. "It is not a fundamental solution to the overall problem" of content uniformity.

"Online data are generally much better - way, way better - than thief data," committee consultant Stephen Byrn, PhD, Purdue University, said. By going online, "we can troubleshoot when something goes wrong."

FDA has established the Process Analytical Technology Subcommittee, which will address how to implement and accept new online technology (see ¹ (Also see "Process Subcommittee To Look To Europe For Technology Implementation" - Pink Sheet, 10 Dec, 2001.)).