1% of total samples tested), of which there were 41 discrepant κ FLC results and 20 discrepant λ FLC results. Further investigations on these discrepant samples revealed that elevated levels of FLC, or an Natural Product Library elevated FLC ratio, on the Freelite™ assay was not supported by the mAb assay or serum IFE, and may have reflected cross-reactivity on the Freelite™ assay with whole immunoglobulin paraprotein or hindrance from the Freelite ‘gaps’ (see Fig. 6); for detailed assessment, see supplementary results. Results from the mAb assay were supported by serum IFE, as well as investigations on matched urine and analysis of patient history, where available. In summary, results indicated that all serum samples with abnormal FLC levels, or an abnormal

κ:λ ratio, were detected by the mAb assay from 1000 consecutive serum samples. Further investigations

revealed that both anti-κ FLC mAbs (BUCIS 01 and BUCIS 04) were diagnostically similar, and either could be used to indicate a sample containing an abnormal κ FLC level. Similarly, both anti-λ FLC mAbs (BUCIS 03 and BUCIS 09) were diagnostically similar, and either could be used to indicate a sample containing an abnormal λ FLC level. Results from 13,090 urine samples analysed routinely by the CIS on the mAb assay and on urine IFE were compared to assess the specificity and sensitivity of the mAbs at detecting FLC in urine. Urine IFE was conducted using ABT-263 cell line antisera against κ and λ LCs that did not distinguish between free and bound LC. Flucloronide Of the 13,090 urines, 12,242 samples were from patients who had a known serum paraprotein, 641 samples had no paraprotein present, and 207 had an unknown admission diagnosis. After initial comparisons between the mAb assay and IFE, 199 discrepancies were identified (1.52% of all samples tested). 143 of these samples had polyclonal LC by IFE but < 10 mg/L of the relevant FLC on the mAb assay. The other 56 samples had monoclonal LC present in the urine, but

< 10 mg/L of the relevant FLC on the mAb assay. These samples were re-tested on the mAb assay to exclude the possibility of user error, and all samples were re-analysed by full IFE analysis to distinguish between FLC and LC bound to whole immunoglobulin. Results from this process revealed that all samples with FLC detected by IFE were also detected by the mAb assay. Complementary analyses of matched serums and patient history also supported these findings. In summary, the mAbs detected FLC in all 2995 urine samples containing monoclonal κ FLC and all 1180 urine samples containing monoclonal λ FLC, as detected by IFE specific for FLC. Assay imprecision, or CV%, was measured in pools of serum samples with low, medium and high κ or λ FLC levels. For κ FLC, at 8.00 mg/L, 16.85 mg/L, and 238.94 mg/L, the intra-assay CV% was 4.46%, 4.69%, and 4.85%, respectively; and the inter-assay CV% was 6.45%, 6.50%, and 5.31%, respectively. For λ FLC, at 7.27 mg/L, 10.38 mg/L, and 91.13 mg/L, the intra-assay CV% was 5.69%, 4.86%, and 2.