Development and analytic validation of an immunoassay for the quantification of canine S100A12 in serum and fecal samples and its biological variability in serum from healthy dogs.

S100A12 (calgranulin C) is a Ca(2+)-binding protein that has been proposed to play a central role in both innate and acquired immune responses. In humans, S100A12 has been reported to be increased in serum and/or plasma in patients with various inflammatory disorders, and this protein has been suggested to be a sensitive and specific marker for inflammatory bowel disease (IBD). An immunoassay for S100A12 is currently available for use in humans, but antibodies against the human protein do not cross-react with canine S100A12 (cS100A12). Both sensitive and specific markers for canine patients with systemic or localized inflammatory diseases are currently lacking, thus the aim of this study was to develop and analytically validate a radioimmunoassay (RIA) for the quantification of cS100A12 in serum and fecal specimens and to determine the biological variation of cS100A12 in serum from healthy dogs. A competitive liquid-phase RIA was developed and analytically validated by determining assay working range, dilutional parallelism, spiking recovery, and intra- and inter-assay variability. Reference intervals for serum and fecal concentrations of cS100A12 were established from 124 and 65 healthy dogs, respectively, and components of variation for serum cS100A12 were determined from 11 dogs over 2.6 months. The working range of the assay was 0.6-432.7 μg/L. No cross-reactivity was observed with the cS100A8/A9 protein complex, the closest structural analogues available. Observed-to-expected ratios (O/E) for the serial dilution of serum and fecal extracts ranged from 97.2 to 146.8% and from 75.3 to 129.8%, respectively. O/E for spiking recovery for serum and fecal extracts ranged from 87.8 to 130.4% and from 84.8 to 143.8%, respectively. Coefficients of variation (CV) for intra- and inter-assay variability for sera were ≤ 8.1% and ≤ 7.8%, respectively, and were ≤ 7.8% and ≤ 8.7%, respectively, for fecal extracts. Reference intervals for serum and fecal cS100A12 were 33.2-225.1 μg/L and <24-745 ng/g, respectively. For biological variability testing, analytical, intra-individual, inter-individual, and total CV were 5.7, 29.2, 31.2, and 66.0%, respectively, yielding an index of individuality of 0.95 and a minimum critical difference (p<0.05) for sequential values of 84.9%. The RIA for cS100A12 measurement described here is analytically sensitive and specific, linear, accurate, precise, and reproducible, and will facilitate further research into the clinical utility of quantifying serum and fecal cS100A12 in canine patients with inflammatory diseases. Moderate changes in serum cS100A12 concentrations may be clinically relevant; however, the use of a population-based reference interval may require caution.