Efficient approach to setting validation acceptance criteria for a biopharmaceutical process

BIOT 163

Tom Gleason, tgleason@amgen.com1, Rick Burdick2, Steve Rausch, srausch@amgen.com1, and James E Seely, jseely@amgen.com3. (1) Process Development Dept, Amgen, 4765 E. Walnut St., Boulder, CO 80301, (2) Quality Engineering, Amgen, 4000 Nelson Road, Longmont, CO 80503, (3) Manufacturing Science and Technology, Amgen Colorado, 4000 Nelson Road, Longmont, CO 80503
To ensure that a process is working as expected, intermediate process steps typically have validation acceptance criteria (VAC) that must be met to provide assurance that the product quality attributes will meet final specifications. VAC based on only full or pilot scale manufacturing data can be problematic because operating parameters are typically run at their set points; therefore, little information can be gained regarding how the process is affected due to variations within these parameters' operating ranges. Including bench-scale characterization data from qualified bench-scale models in the analysis provides a way to better and more rapidly predict future process performance. We describe statistical approaches to analyzing bench-scale characterization data (including testing at the edges and 3X outside the normal operating ranges) combined with large-scale data, to set VAC. This allows the establishment of appropriate VAC in instances where there are limited large-scale data due to time or resource constraints.
 

Commercialization of Biologics: Impact of Design Space Philosophy & Emerging Regulations on Process Validation
3:00 PM-5:20 PM, Tuesday, August 21, 2007 BCEC -- 108, Oral

Division of Biochemical Technology

The 234th ACS National Meeting, Boston, MA, August 19-23, 2007