Specificity of IS6110-based amplification assays for Mycobacterium tuberculosis complex

Abstract

We have demonstrated homology between DNA from mycobacteria other than tuberculosis strains (MOTT) and a central region of IS6110 and have urged caution in using diagnostic tests based on this target (6). Hellyer and colleagues (5) have evaluated three IS6110-based amplification methods: HincII strand displacement amplification (SDA), thermophilic SDA, and a previously described PCR (11) and have observed no cross-reaction with 27 nontuberculosis mycobacteria, 26 of which had been used previously in our study (6). They conclude that these data support the use of selected regions of IS6110 as Mycobacterium tuberculosis complex-specific targets.Hellyer and colleagues suggest that cross-reactivity may be limited to the 181-bp region or alternatively was due to PCR contamination. They describe the use of dUTP/uracil DNA glycosylase to prevent false-positive results, but this strategy has been shown to have a limited impact on false-positive rates in tuberculosis PCR in a recent international collaborative study, in which 8 of 17 laboratories using it recorded falsepositive reactions (8).We have expressed the view that homology is confined to a yet-undefined central region of the insertion sequence (4). We cannot accept the contention that PCR contamination was responsible for our results even though, as the authors suggest, a relatively high concentration of DNA was used for the PCRs. Homology between the 181-bp fragment amplified from H37Rv and genomic DNA from the 26 MOTT was demonstrated by Southern blotting, excluding the possibility of PCR contamination as an explanation of our results. The results obtained in the study of Hellyer et al. may be due to the fact that the target areas used in their PCRs are outside the homologous region, or alternatively they may be due to the small and unquantitated amount of DNA used in their assay, giving false-negative results for nontuberculosis isolates where the homology is lower and consequently a lower predicted level of sensitivity.The data supporting the contention that IS6110-based methods are specific is scanty. Among the authors quoted by Hellyer in support of this idea, Eisenach and colleagues report one positive result for 42 nontuberculosis mycobacteria, a false-positive rate of at least 2.3% (3). Pietrzak and colleagues tested specimens from only 48 patients for whom a diagnosis of tuberculosis was excluded on the basis of culture and history. They state that none of these patients had a positive PCR, but the data is not shown (9). Using the same primers, Querol and colleagues did not test any patients with MOTT (10). The IS6110 PCR of Eisenach et al. (3) has been shown to have a false-positive rate of 3% with specimens containing MOTT (1). An evaluation of IS6110-based methods in nine laboratories from France demonstrated false-positive reactions with an average rate of 7%, although a clear relationship to nontuberculosis isolates or other organisms could not be drawn (2). Those authors conclude that the results of their study suggest "that PCR using IS6110 as a ...

Identification of a New DNA Region Specific for Members of Mycobacterium tuberculosis Complex

“…This may perhaps account for the rates of false-positive results that were observed in several studies that used IS6110 as a target sequence and that ranged from 2.3% (18) to 20% (34). More recently, IS6110 was shown to hybridize to PCR products amplified from Streptococcus pneumoniae, Streptococcus pyogenes, and even from Aspergillus fumigatus (19,20). In addition to being prone to false-positive results, targeting of IS6110 for PCR amplification may also yield false-negative results.…”

False positive diagnosis of meningococcal infection by the IS1106 PCR ELISA

“…False positives were also demonstrated by a nested IS6110 based PCR for the detection of Mycobacterium tuberculosis when applied to mycobacteria that did not belong to the M. tuberculosis complex [9]. More recently, ampli¢ed DNA from Aspergillus spp., Streptococcus pneumoniae and S. pyogenes was found to hybridise to a probe derived from a 181-bp fragment of M. tuberculosis IS6110 [10]. Trinker et al [11] have also reported false-positive diagnosis of tuberculosis with the insertion sequence IS6110 leading to the commencement of inappropriate treatment.…”

False positive diagnosis of meningococcal infection by the IS1106PCR ELISA

“…False positives were also demonstrated by a nested IS 6110 based PCR for the detection of Mycobacterium tuberculosis when applied to mycobacteria that did not belong to the M. tuberculosis complex [9]. More recently, amplified DNA from Aspergillus spp., Streptococcus pneumoniae and S. pyogenes was found to hybridise to a probe derived from a 181‐bp fragment of M. tuberculosis IS 6110 [10]. Trinker et al [11] have also reported false‐positive diagnosis of tuberculosis with the insertion sequence IS 6110 leading to the commencement of inappropriate treatment.…”