2003D-0206 - Draft Guidance for Industry on Exocrine Pancreatic Insufficiency Drug Products--Submitting New Drug Application; Availability
FDA Comment Number : EC5
Submitter : Dr. Tibor Sipos Date & Time: 07/07/2004 05:07:38
Organization : Digestive Care, Inc.
Drug Industry
Category :
Issue Areas/Comments
PART 4 OF 10
DOCKET NO. 2003D-0206 (2003N-0205)

The identification of the initial 27 bands as to their catalytic activity; e.g. lipase, protease, amylase, etc. however, are not known. In my expert opinion, the identification of the protein bands after separation by isoelectric focusing, SDS-PAGE electrophoresis or HPLC for biological activity is an extremely complex, labor-intensive and expensive process and is impractical for the routine identification and characterization of pancrelipase raw material. Therefore, these separation techniques should not be required for the identification and characterization of pancrelipase.

The enzymes in question for pancrelipase, e.g. lipase, amylase and protease, are proteins that are composed of amino acids linked together by peptide bonds. Peptide bonds exhibit characteristic absorption bands in the infra-red (IR) spectrum. Thus, we recommend the use of the IR spectrophotometric method, employing the KBr pellet technique for the physico-chemical identification and characterizations of pancrelipase. The IR/KBr method is suitable for the simultaneous identification of pancrelipase whether it contains the water-soluble or the water-insoluble portion of pancrelipase. If there is further interest to characterize the
water-soluble portion of pancrelipase, one may employ UV spectrophotometric analysis at
205-210 nm, 260 nm and 280 nm for peptide bond absorption, nucleic acid, and protein concentrations, respectively (Baily JL., Techniques in Protein Chemistry, 2nd Ed., pp. 342-346, Elsevier Publishing Co., 1967). Ratios at 280/260 nm would provide indications for the presence of nucleic acid contamination in pancrelipase products that may be further used to estimate the possible purine load per dosage when the product is prescribed to a patient. Likewise, the soluble protein content of pancrelipase may also be determined by employing the Bradford colorimetric assay using Coomassie Brilliant Blue, R-210 for the assay which only takes ten (10) minutes to complete (Methods in Enzymology 182: 50-69, 1990).

Establishing enzyme ratio specifications for lipase, amylase, and protease are not practical and is unachievable for the reasons cited above; e.g. the extreme complexity of the zymogen activation process. We have reviewed our historical files of pancrelipase drug substance over a decade and find greatly varied ratios of lipase:amylase:protease of 1:4:4 to 1:8:5 for PEC and 1:3:3 to 1:5:3 for PEC-HL, respectively.

Removal/Inactivation of Viral Agents:

 There is no evidence of pig viruses being problematic for humans and this is not done for the food chain.
 Exhaustive removal will significantly modify the API, adding significant costs and increasing the potential for additional required characterizations, including a potential for new safety and efficacy studies.

Based on the historical safety of porcine-derived pancrelipase products, and the lack of any known dangers of the porcine viruses contained in pancrelipase products, leads us to believe that extra processing steps for exhaustive viral inactivation are not necessary. It is well known that lipases are very sensitive to changes resulting from heat, moisture, solvents, or pH. Use of viral inactivating agents may most likely also have undesirable effects causing net degradation of lipase or selectively and independently on any of the components of pancrelipase: lipase, amylase, or protease. The FDA has repeatedly emphasized that toxicity studies are not required for pancrelipase, changes to processing that alter the API may have unforeseen ramifications in terms of the safety/efficacy profiles of the drug substance itself.