Abstract
Introduction: Tuberculosis (TB) is an infectious disease caused by the group of bacterial pathogens Mycobacterium tuberculosis complex (MTBC) and is one of the leading causes of death worldwide. Timely diagnosis and treatment of drug-resistant TB is a key pillar of WHO's strategy to combat global TB. The time required to carry out drug susceptibility testing (DST) for MTBC via the classic culture method is in the range of weeks and such delays have a detrimental effect on treatment outcomes. Given that molecular testing is in the range of hours to 1 or 2 days its value in treating drug resistant TB cannot be overstated. When developing such tests, one wants to optimize each step so that tests are successful even when confronted with samples that have a low MTBC load or contain large amounts of host DNA. This could improve the performance of the popular rapid molecular tests, especially for samples with mycobacterial loads close to the limits of detection. Where optimizations could have a more significant impact is for tests based on targeted next generation sequencing (tNGS) which typically require higher quantities of DNA. This would be significant as tNGS can provide more comprehensive drug resistance profiles than the relatively limited resistance information provided by rapid tests. In this work we endeavor to optimize pre-treatment and extraction steps for molecular testing. Methods: We begin by choosing the best DNA extraction device by comparing the amount of DNA extracted by five commonly used devices from identical samples. Following this, the effect that decontamination and human DNA depletion have on extraction efficiency is explored. Results: The best results were achieved (i.e., the lowest Ct values) when neither decontamination nor human DNA depletion were used. As expected, in all tested scenarios the addition of decontamination to our workflow substantially reduced the yield of DNA extracted. This illustrates that the standard TB laboratory practice of applying decontamination, although being vital for culture-based testing, can negatively impact the performance of molecular testing. As a complement to the above experiments, we also considered the best Mycobacterium tuberculosis DNA storage method to optimize molecular testing carried out in the near- to medium-term. Comparing Ct values following three-month storage at 4 °C and at -20 °C and showed little difference between the two. Discussion: In summary, for molecular diagnostics aimed at mycobacteria this work highlights the importance of choosing the right DNA extraction device, indicates that decontamination causes significant loss of mycobacterial DNA, and shows that samples preserved for further molecular testing can be stored at 4 °C, just as well at -20 °C. Under our experimental settings, human DNA depletion gave no significant improvement in Ct values for the detection of MTBC.