R-Biopharm AG - Comment

Comment against part of the proposed guidance: - Method for LoD detection for antigen immunoassays - Unit of concentration for antigen immunoassays Rationale: The detection methods described in the draft refer to antigen, antibody, and nucleic acid tests using stool samples. However, some of the requested analytical studies are only suitable for the nucleic acid tests. One of the major topics is the estimation of the analytical sensitivity. The recommended study to determine the limit of detection (LoD) includes serial dilutions of at least two strains and the specification in colony forming units. A procedure with at least two strains is reasonable for nucleic acid tests which detect toxin genes but not for antigen immunoassays. The latter detect the produced protein (either toxin A or toxin B) and a LoD determination using purified toxins would therefore be suitable. This method was in fact used and described in the most recent approved 510(k)s submissions of such immunoassays (for example: FDA 510(k) Number K033479 ProSpecT Clostridium difficile Toxin A/B Microplate Assay from Remel; FDA 510(k) Number K082499 C. DIFF QUIK CHEK COMPLETE® from Techlab; FDA 510(k) number K050891 TOX A/B QUICK CHEK® from Techlab). Furthermore, the individual strain characteristics as well as particular culture conditions may dramatically influence the secretion rates for both the toxins, thus not the colony forming units (CFU) but the amount of toxin (ng/ml) present in the sample or the reference preparation would be an adequate unit of concentration. The question of how to determine the LoD , as well as the unit of concentration, are critical issues particularly as they create a base for selection of samples used for further analytical studies described in the draft (e.g. the analytical reactivity, interference, and precision/repeatability). Submitted by: R-Biopharm AG Regulatory Affairs Department