EU-wide external quality assessment study on the sensitivity and specificity of different DNA amplification protocols for the detection of Borrelia burgdorferi sensu lato

Objectives Diagnosis of Lyme borreliosis (LB) is mainly based on clinical symptoms, patient’s history, and serological testing. Despite a lack of standardization, polymerase chain reaction (PCR) has gained importance for the detection of Borrelia DNA. So far, extensive studies comparing different protocols used in molecular diagnostic or research setting are missing. Here, we describe a European-wide comparison of commercial and in-house PCR protocols using a standardized DNA panel including all relevant Borrelia burgdorferi sensu lato (Bbsl) species to explore variation in PCR results in different laboratories. Methods A DNA testing panel composed of 90 DNA samples from 14 Bbsl strains in six different dilutions, plus six specificity controls, was sent blinded to 34 laboratories (33 European plus one in the US). The results from a total of 57 different amplification protocols were collected and compared regarding their detection limits, specificity, and sensitivity over all Bbsl strains for each dilution included in the DNA panel. Results While the detection limits showed vast differences (>10 5 genome equivalents (GE)) between different amplification protocols, none of the most commonly used PCR targets ( ospA , 16S rRNA, flagellin, 5S-23S intergenic spacer) significantly outperformed other PCR targets. Interestingly, large differences in detection limits were found not only between different protocols, but also when the same protocol was used on different Bbsl species (>10 5 GE) and even on different strains of the same Bbsl species (>10 4 GE). Specificity also varied between different protocols, with many protocols recognizing relapsing fever Borreliae . Conclusion A standardization of PCR methods used for confirmation of LB diagnosis is urgently needed.