Products for blood screening NAT tests
The use of external controls, calibrated in copies/ml (and IU/ml), is essential for independent monitoring of the performance and sensitivity of nucleic acid test runs and IVD reagent batches in blood banks. The viral standard concentrations in the bioQControl products have been established after characterization in multi-center validation studies, in proficiency programs and in WHO collaborative studies. We have found the bDNA 3.0 can be used to quantify standards in copies/ml. We cross calibrated by comparing the limiting dilution results found in blood screening assays to the bDNA quantification of each dilution. It was found the theoretical limitations were met by the Chiron Ultrio assay for HCV-RNA and HIV-RNA. For both markers one copy input (as defined by bDNA 3.0 assays) was detected with a 63 % hit rate (HCV, genotype 1 and HIV genotype B). For translation to IU/ml we only used bDNA 3.0 test results..
Sensitivity panels are composed of serial BioQControl standard dilutions, calibrated in copies/ml and IU/ml by bDNA 3.0 testing. The 10 member panels are intended for validation of nucleic acid test systems and batch release control. The concentration range starts at around 3000 copies/ml and ends at 0,1 copy/ml; 0.5 logs intervals. Sensitivity analysis of assays can be performed by probit analysis on response rates (available at data exchange).. We support our customers on the statistical evaluation of the results.
Controlling the performance and sensitivity of qualitative assays requires different concentrations in run controls. The sensitivity of assays is defined by testing dilution series in the dynamic range multiple times. The results are translated to hit rates using probit analysis. In many cases the 50 % and 95 % hit rate is presented to define the analytical sensitivity.
Run control concentrations are related to the 95 % hit rate. A stop -or- go control for performance evaluation has a concentration at five times the 95% hit rate of the assay. At this level a positive result is expected. This level is outside the dynamic range of the assay and cannot be used to monitor the sensitivity of the assay. For example the 50% hit rate for the Chiron Ultrio assay is around 2-3 copies/ml while the 95 % hit rate is around 12-22 copies/ml while the run control level is 125 copies/ml.
The claim on sensitivity is directly relating to the residual risk on transmission of infection by blood transfusion. The 50 % hit rates for HBV and HCV coincide with the levels that are just infectious in absence of antibodies while for HIV the 50 % hit rate is about 2 logs lower than 50 % Chimpansee infectious doses. Therefore controlling the sensivity has more added value than performance control. Trend controls have levels around the 90-95 % hit rate of hte assay.
Trend analysis is available at data exchange and support. We defined separate algorithms and/or controls for each assay to monitor sensitivity. As data exchange will include results from several laboratories external causes (like batches or equipment) may be identified.
