Abstract
Tuberculosis remains one of the leading causes of death worldwide, especially in low- and middle-income countries. Tuberculosis treatment and control efforts are hindered by the difficulty in making the diagnosis, as currently available diagnostic tests are too slow, too expensive, or not sufficiently sensitive. Recombinase polymerase amplification (RPA) is a novel technique that allows for the amplification of DNA rapidly, at constant temperature, and with minimal expense. We calculated and compared the limit of detection, sensitivity, and specificity of two RPA-based assays for the diagnosis of pulmonary tuberculosis, using two sets of published primers. We also calculated and compared the assays' limits of detection and compared their performance using two different DNA extraction methods prior to amplification (a commercially available DNA extraction kit vs. the chelex method). The RPA-lateral flow assay had a limit of detection of 5 fg/μL of DNA, a sensitivity of 53.2%, and a specificity of 93.3%, while the real time-RPA assay had a limit of detection of 25 fg/μL of DNA, a sensitivity of 85.1%, and a specificity of 93.3%. There was no difference in assay performance when DNA extraction was carried out using the commercial kit vs. the chelex method. The real-time RPA assay has adequate sensitivity and specificity for the diagnosis of pulmonary tuberculosis and could be a viable diagnostic tool in resource-limited settings, but the lateral flow assay did not perform as well, perhaps due to the fact we used stored sputum specimens from a biorepository. More work is needed to optimize the RPA-lateral flow assay, to get a more accurate estimate of its specificity and sensitivity using prospectively collected specimens, and to develop both assays into point-of-care tests that can be easily deployed in the field.