JOURNAL OF CLINICAL MICROBIOLOGY, July 2010, p. 2495–2501 Vol. 48, No.

7
0095-1137/10/$12.00 doi:10.1128/JCM.00128-10
Copyright © 2010, American Society for Microbiology. All Rights Reserved.

Evaluation of the Analytical Performance of the Xpert MTB/RIF Assay䌤†

Robert Blakemore,1 Elizabeth Story,1 Danica Helb,1‡ JoAnn Kop,2 Padmapriya Banada,1
Michelle R. Owens,2 Soumitesh Chakravorty,1 Martin Jones,2 and David Alland1*
Division of Infectious Disease, Department of Medicine, and Ruy V. Lourenço Center for the Study of Emerging and
Reemerging Pathogens, New Jersey Medical School, University of Medicine and Dentistry of New Jersey,
Newark, New Jersey,1 and Cepheid, Sunnyvale, California2
Received 20 January 2010/Returned for modification 12 April 2010/Accepted 17 May 2010

We performed the first studies of analytic sensitivity, analytic specificity, and dynamic range for the new
Xpert MTB/RIF assay, a nucleic acid amplification-based diagnostic system that detects Mycobacterium tuber-
culosis and rifampin (RIF) resistance in under 2 h. The sensitivity of the assay was tested with 79 phyloge-
netically and geographically diverse M. tuberculosis isolates, including 42 drug-susceptible isolates and 37
RIF-resistant isolates containing 13 different rpoB mutations or mutation combinations. The specificity of the
assay was tested with 89 nontuberculosis bacteria, fungi, and viruses. The Xpert MTB/RIF assay correctly
identified all 79 M. tuberculosis isolates and correctly excluded all 89 nontuberculosis isolates. RIF resistance
was correctly identified in all 37 resistant isolates and in none of the 42 susceptible isolates. Dynamic range
was assessed by adding 102 to 107 CFU of M. tuberculosis into M. tuberculosis-negative sputum samples. The
assay showed a log-linear relationship between cycle threshold and input CFU over the entire concentration
range. Resistance detection in the presence of different mixtures of RIF-resistant and RIF-susceptible DNA was
assessed. Resistance detection was dependent on the particular mutation and required between 65% and 100%
mutant DNA to be present in the sample for 95% certainty of resistance detection. Finally, we studied whether
assay specificity could be affected by cross-contaminating amplicons generated by the GenoType MTBDRplus
assay. M. tuberculosis was not detected until at least 108 copies of an MTBDRplus amplicon were spiked into
1 ml of sputum, suggesting that false-positive results would be unlikely to occur.

Conventional diagnostic methods for Mycobacterium tuber- resistant tuberculosis cases contain mutations in this 81-bp
culosis are slow and/or lack sensitivity. A number of new diag- region (16). Our previous work has established that the Xpert
nostic approaches have brought incremental improvements to MTB/RIF assay has a limit of detection (LOD), defined as the
detection and drug susceptibility testing; however, the techni- minimum number of bacilli that can be detected with 95%
cal complexity of these assays and their dependence on dedi- confidence) of 131 CFU per ml of clinical sputum (7). The
cated laboratory infrastructure have limited their adoption, assay was also able to identify RIF resistance in samples con-
especially in low-resource, high-burden settings (1, 11, 12, 21). taining 23 common clinically occurring rpoB mutations. None
The recently introduced Xpert MTB/RIF (manufactured and of the 20 nontuberculosis mycobacteria (NTM) species
marketed by Cepheid, Sunnyvale, CA) assay simultaneously tested, including the NTM species commonly described as

Downloaded from https://journals.asm.org/journal/jcm on 18 May 2024 by 41.223.51.165.

detects the presence of M. tuberculosis and its susceptibility to causing human disease were falsely identified as M. tuber-
the important first-line drug rifampin (RIF) (7). A sample culosis (7), suggesting high specificity. Several small studies
processing system and an automated heminested real-time using clinical samples demonstrated 98% to 100% sensitivity
PCR assay are integrated into a single disposable cartridge. overall, 72% sensitivity in smear-negative patients, and a
The assay can be performed directly from a clinical sputum specificity of 100% (7).
sample or from a decontaminated sputum pellet and can gen- In the present study, we expand upon our previous work and
erally be completed in less than 2 h (7). report on several critical analytical assay performance charac-
The Xpert MTB/RIF assay detects M. tuberculosis and RIF teristics, including dynamic range, sensitivity, specificity, RIF
resistance by PCR amplification of the rifampin resistance- resistance detection in heterogeneous samples, and resiliency
determining region (RRDR) of the M. tuberculosis rpoB gene against cross-contamination by other nucleic acid amplification
and subsequent probing of this region for mutations that are techniques (NAATs).
associated with RIF resistance. Approximately 95% of RIF-
MATERIALS AND METHODS
Diagnostic system. The Xpert MTB/RIF assay and the GeneXpert instrument
* Corresponding author. Mailing address: Division of Infectious have recently been described in detail (7). In brief, the assay consists of a
Disease, University of Medicine and Dentistry of New Jersey, 185 single-use multichambered plastic cartridge preloaded with the liquid buffers and
South Orange Avenue, MSB A920C, Newark, NJ 07103. Phone: (973) lyophilized reagent beads necessary for sample processing, DNA extraction, and
972-2179. Fax: (973) 972-0713. E-mail: allandda@umdnj.edu. heminested real-time PCR. Clinical sputum samples or decontaminated sputum
† Supplemental material for this article may be found at http://jcm pellets are treated with an NaOH and isopropanol-containing sample reagent
.asm.org/. (SR). The SR is added at a 2:1 ratio to the sputum sample or sputum pellet and
‡ Present address: Divisions of Infectious Diseases and Epidemiol- incubated for 15 min at room temperature. The treated sample is transferred into
ogy, School of Public Health, University of California at Berkeley, the cartridge, the cartridge is loaded into the GeneXpert instrument, and an
Berkeley, CA. automatic process completes the remaining assay steps. The assay cartridge also
䌤
Published ahead of print on 26 May 2010. contains lyophilized Bacillus globigii spores which serve as an internal sample

2495
2496 BLAKEMORE ET AL. J. CLIN. MICROBIOL.

processing and PCR control. The spores are automatically resuspended and DNA into cartridge chamber 10, which would normally receive the DNA-con-
processed during the sample processing step, and the resulting B. globigii DNA taining eluent of lysed bacterial cells.
is amplified during the PCR step. The standard user interface indicates the Five hundred microliters of TET buffer (50 mM Tris [pH 8.35], 0.1 mM
presence or absence of M. tuberculosis, the presence or absence of RIF resis- EDTA, and 0.1% mM Tween 20) was added to cartridge chamber five. A 100-␮l
tance, and a semiquantitative estimate of M. tuberculosis concentration (high, sample mixture containing either sensitivity or specificity panel DNA (or RNA
medium, low, and very low). Assays that are negative for M. tuberculosis and also for RNA viruses), 90 copies of genomic B. globigii DNA, and TET buffer was
negative for the B. globigii internal control are reported as invalid. added directly into chamber 10. M. tuberculosis isolates were tested once at 45
The PCR assay amplifies a 192-bp segment of the M. tuberculosis rpoB gene in genomes per reaction (10 times the Xpert MTB/RIF LOD for DNA) (7), and all
a heminested real-time PCR. The internal control heminested B. globigii assay is other organisms were tested twice at 106 genome copies per reaction. Negative
multiplexed with the M. tuberculosis assay. M. tuberculosis is detected using five controls contained no DNA. Initial DNA concentrations were determined with
overlapping molecular beacon probes (probes A to E) that are complementary to a NanoDrop ND-1000 (Thermo Scientific) and diluted in TET buffer. When
the entire 81-bp RIF resistance-determining “core” region of the wild-type rpoB reliable genome size was not available, appropriate concentrations were deter-
gene (5, 7, 14). Mutations in the rpoB gene target inhibit hybridization of one or mined by assuming genome sizes of 4.5 Mb/genome for bacteria and 22.5 Mb/
more of the rpoB-specific molecular beacons, reducing or eliminating the signal genome for fungi. Testing of mixed samples containing DNA from a wild-type M.
from the corresponding probes. M. tuberculosis is identified when at least two of tuberculosis strain plus an M. tuberculosis strain containing either rpoB gene
the five rpoB-specific molecular beacons give a positive signal with cycle thresh- mutation 533ccg (ccg mutation at codon position 533) or 531ttg (ttg mutation at
old (CT) values that are ⱕ38 and that differ by no more than two cycles. B. globigii codon position 531) was performed at a final concentration of 90 copies per
DNA is detected when the single B. globigii molecular beacon produces a CT of reaction. Strains used for mixture testing are indicated in Table S1 in the sup-
⬍38 cycles. plemental material. Preparations used for mixture testing were confirmed to be
The difference in CT between the first (early CT) and last (later CT) M. of appropriate concentration by quantitative real-time PCR of the M. tuberculosis
tuberculosis-specific molecular beacon (⌬CT Max) is the basis of rpoB mutation lsr2 gene (4). Cartridges were loaded into the GeneXpert instrument and pro-
and RIF resistance detection. RIF resistance is identified if the ⌬CT Max is ⬎3.5 cessed with a truncated version of the automated protocol that omitted the
sample processing steps, proceeding immediately as though the lysis and elution
cycles. RIF susceptibility is identified if the ⌬CT Max is ⱕ3.5 cycles. A sample is
of a normal sample had already been completed. This protocol preserved the
considered RIF indeterminate when the last probe returns a CT of ⬎38 and the
remaining assay steps without change and is equivalent to version 1 of the MTB
first probe has a CT value of ⬎34.5 cycles because the assay terminates at cycle
assay protocol, released on 8 April 2009.
38, and a ⌬CT Max of ⬎3.5 cannot be measured. For the purpose of the current
Three organisms (Table 1) were obtained as whole viral particle NATTROL
study, rpoB mutations that completely inhibit probe hybridization (and thus
diagnostic standards (Zeptometrix), 100 ␮l of the viral particle stock was diluted
caused one or more molecular beacons to show CT values of ⬎38) were defined
1:20 in TET buffer, and the mixture was tested with the Xpert MTB/RIF assay
as causing probe “dropouts,” whereas rpoB mutations that permit partial probe
according to the manufacturer’s instructions.
hybridization and produce a measurable ⌬CT Max of ⬎3.5 cycles were consid-
Amplicon contamination studies. We created a modified MTBDRplus assay
ered “delays.”
amplicon with defined mutations in the rpoB target, but unaltered primer binding
Dynamic range testing. The laboratory M. tuberculosis strain H37Rv obtained
sites. This distinguished between unintentional amplicon contamination and
from the American Type Culture Collection (ATCC) (Manassas, VA) was cul-
deliberate experimental contamination. Three smear-positive clinical sputum
tured and quantified as previously described (7). Excess discarded sputum sam-
samples from individuals suspected to have tuberculosis were PCR amplified
ples were collected under institutional review board (IRB)-approved protocols
using the GenoType MTBDRplus kit (Hain Lifescience GmbH) as indicated in
from patients not suspected of tuberculosis as previously described (7). M. the product insert. The 168-bp DNA sequence of the full MTBDRplus am-
tuberculosis cells were spiked into 1-ml aliquots of M. tuberculosis-negative spu- plicon was derived by outward sequencing of the product of the GenoType
tum samples to final concentrations ranging from 102 to 107 CFU/ml and were MTBDRplus assay from within the rpoB core. The MTBDRplus amplicon
tested with the Xpert MTB/RIF assay according to the manufacturer’s instruc- (minus the biotinylated ends) was then reconstructed via overlap extension of
tions. two 98-bp oligonucleotides (Hain-AT-do-F [5⬘-CACGCTCACGTGACAGACC
Analytic sensitivity panel. The 79 clinical M. tuberculosis isolates analyzed in GCCGGGCCCCAGCGCCCACAGTCGGCGCTTGTCGGTCAACCCCGAC
the sensitivity study included 70 isolates from a collection of 211 highly charac- AGCGGGTTGTTCTGGACCATGAATTGGCTCAGC-3⬘] and Hain-AT-
terized M. tuberculosis strains established by the United Nations Children’s do-R [5⬘-CACGCTCACGTGACAGACCGCCGGGCCCCAGCGCCCACAG
Fund/UNDP/World Bank/WHO Special Program for Research and Training in TCGGCGCTTGTCGGTCAACCCCGACAGCGGGTTGTTCTGGACCATG
Tropical Diseases (3). Rifampin MIC was previously established by the agar

Downloaded from https://journals.asm.org/journal/jcm on 18 May 2024 by 41.223.51.165.

AATTGGCTCAGC-3⬘]). The construct included four mutations in the rpoB
proportion method on Lowenstein-Jensen medium, using a breakpoint of 40 core region (511ccg, 516gtc, 526gac, and 531tgg) intended to block hybridization
␮g/ml to establish resistance (13). M. tuberculosis DNA was isolated as described of all but one of the Xpert MTB/RIF assay’s five molecular beacon probes.
previously (17). The remaining nine clinical M. tuberculosis strains were selected Statistical analysis. The minimum proportion of mutant DNA detectable in a
from a panel of pansusceptible M. tuberculosis isolates obtained in a previous wild-type mixture was determined by binary logistic regression. The data were
study (6). The phylogenetic lineage of each M. tuberculosis isolate within the converted to the percentages of RIF-resistant responses for each ratio of wild
collection was identified by testing genomic DNA for nine single-nucleotide type to mutant DNA. Binary logistic regression was fitted through the tested
polymorphisms (SNPs), which were used to classify each isolate into one of 10 concentrations, and lower and upper 95% confidence intervals (95% CIs) were
SNP cluster groups and SNP cluster subgroups (6). The isolates were selected to generated for the curve. The 95% CI for the minimum proportion was deter-
include at least five examples of any unique SNP cluster group or subgroup. The mined by where the 95% probability level crossed the upper and lower 95% CIs.
details for each isolate, including geographic source, phylogenetic group, drug In all other cases, 95% CIs were calculated using a Student t test.
susceptibility profile, and drug resistance target genotype, are listed in Table S1
in the supplemental material.
Analytic specificity panel. The 89 nontuberculosis bacteria, fungi, and viruses RESULTS
analyzed in the specificity study were obtained from the ATCC and BEI repos-
itories (NIH Biodefense and Emerging Infections Research Resources Reposi- Dynamic range. The concentration of M. tuberculosis bacte-
tory, NIAID, NIH), the Clinical Microbiology laboratory at the university hos- ria in clinical sputum samples can vary from over 107 to less
pital of the University of Medicine and Dentistry of New Jersey (UMDNJ), than 20 CFU/ml (10). We performed analytic studies to exam-
Zeptometrix Corporation (Buffalo, NY), or other collaborators (Table 1). Sam-
ples obtained from UMDNJ were verified to be of the appropriate species or
ine the CFU relationship between CT value and cell concen-
genera as described in reference 2. tration over the potential range of M. tuberculosis cell concen-
Sensitivity and specificity panel testing. The sensitivity and specificity studies trations. Clinical sputum samples were spiked with known
were performed using production-lot cartridges preloaded with production-lot numbers of M. tuberculosis bacteria ranging from 107 to 102
enzyme beads and reagent beads. However, unlike the commercially available
CFU/ml and tested in the assay. The assay correctly identified
assay, these studies used open cartridges that were not sealed with a sonically
welded plastic top and were not preloaded with liquid reagents or the sample
M. tuberculosis in all 28 sputum samples containing 107 to 103
processing control bead containing B. globigii spores. The open cartridges pre- CFU/ml and identified M. tuberculosis in 5 of the 6 samples
served access to the cartridge chambers and permitted manual placement of test containing 102 CFU/ml. The log-linear relationship between
VOL. 48, 2010 ANALYTICAL PERFORMANCE OF THE Xpert MTB/RIF ASSAY 2497

TABLE 1. Analytic specificity panel

Bacterial, fungal, or viral species Strain or sourcea Bacterial, fungal, or viral species Strain or sourcea

Bacteria Proteus mirabilis ....................................Clinical strain

Acinetobacter baumanii ....................................BEI NR-10146 Proteus vulgaris ......................................BEI DD-460
Acinetobacter calcoaceticus ..............................Clinical strain Providencia alcalifaciens .......................PI 368
Actinomyces israelii ...........................................ATCC 12102 Pseudomonas aeruginosa ......................ATCC 27853
Actinomyces meyeri ...........................................Clinical strain Rhodococcus equi..................................ATCC 14187
Bacillus cereus ................................................... BEI NR-4198 Salmonella enterica................................BEI NR-515
Bacillus subtilis ..................................................Clinical strain Salmonella typhi ....................................Clinical strain
Bordetella parapertussis.....................................ATCC 151311D Serratia marcescens ...............................Clinical strain
Bordetella pertussis ............................................Cepheid Shigella boydii ........................................Clinical strain
Campylobacter jejuni.........................................BEI NR-3057 Shigella flexneri ......................................Clinical strain
Chlamydia pneumoniae ....................................Cepheid Staphylococcus aureus...........................ATCC 25953
Citrobacter freundii............................................Clinical strain Staphylococcus capitis ...........................Clinical strain
Corynebacterium diptheriae ..............................PI 581 Staphylococcus epidermidis...................ATCC 12228
Corynebacterium pseudodiptheriticum .............ATCC 10700 Staphylococcus haemolyticus ................Clinical strain
Corynebacterium xerosis....................................Clinical strain Staphylococcus hominis ........................Clinical strain
Enterobacter aerogenes......................................Clinical strain Staphylococcus lugdunensis ..................Clinical strain
Enterobacter cloacae .........................................Clinical strain Stenotrophomonas maltophilia .............Clinical strain
Enterococcus avium ..........................................Clinical strain Streptococcus equi .................................Clinical strain
Enterococcus faecalis ........................................Clinical strain Streptococcus pyogenes..........................ATCC 19615
Enterococcus faecium .......................................Clinical strain Streptococcus agalactiae........................ATCC 12386
Escherichia coli..................................................Clinical strain Streptococcus constellatus .....................Clinical strain
Escherichia coli O157H7..................................ATCC 35150 Streptococcus mitis ................................Clinical strain
Fusobacterium nucleatum.................................ATCC 25586D Streptococcus mutans ............................Clinical strain
Haemophilus influenzae....................................ATCC 49247 Streptococcus pneumoniae ....................Clinical strain
Haemophilus parahemolyticus..........................ATCC 10014 Streptococcus uberis...............................Clinical strain
Haemophilus parainfluenzae ............................Clinical strain Veillonella parvula .................................ATCC 10790D-5
Klebsiella oxytoca...............................................Clinical strain Stenotrophomonas maltophilia .............Clinical strain
Klebsiella pneumoniae.......................................Clinical strain Yersinia pestis......................................... BEI NR-2644
Legionella pneumophila....................................ATCC 33152D
Leuconostoc mesenteroides...............................Clinical strain Fungi
Listeria grayi....................................................... ATCC 25401 Candida albicans ...................................ATCC 14053D
Listeria monocytogenes .....................................BEI NR-4211 Cryptococcus neoformans .....................A. Casadevall lab
Moraxella catarrhalis .........................................ATCC 8176 Histoplasma capsulatum .......................J. Nosanchuk lab
Morganella morganii .........................................Clinical strain Kingella kingae.......................................ATCC 23332D
Mycoplasma pneumoniae .................................ATCC 15531D
Neisseria gonorrhoeae .......................................ATCC 49226 Viruses
Neisseria lactamica............................................ATCC 23971 Adenovirusb ...........................................ZM NATADV1-ST
Neisseria meningitidis ........................................Clinical strain Herpes simplex virus type 1 ................Cepheid
Neisseria mucosa ............................................... ATCC 69695 Herpes simplex virus type 2 ................Cepheid
Nocardia asteroides ...........................................ATCC 19247 Influenza A virusc .................................Cepheid
Nocardia cyriageorgica ......................................ATCC BAA-1516 Influenza B virusc .................................Cepheid
Nocardia farcinica.............................................ATCC 3318 Parainfluenza virus 2b ..........................ZM NATPARA2-ST

Downloaded from https://journals.asm.org/journal/jcm on 18 May 2024 by 41.223.51.165.

Pasteurella multocida ........................................ATCC 3160 Parainfluenza virus 3b ..........................ZM NATPARA3-ST
Peptostreptococcus anaerobius .........................ATCC 49031D Respiratory syncytial virus Ac .............Cepheid
Porphyromonas gingivalis..................................ATCC 33277D-5 Respiratory syncytial virus Bc .............Cepheid
Prevotella melaninogenica.................................ATCC 25845D-5 Rhinovirus 6c .........................................Cepheid
Propionibacterium acnes...................................Clinical strain Rhinovirus 16c .......................................Cepheid
a
The 89 nontuberculosis bacteria, fungi, and viruses analyzed were obtained from the ATCC and BEI repositories (NIH Biodefense and Emerging Infections
Research Resources Repository), the University of Medicine and Dentistry of New Jersey (UMDNJ), Zeptometrix Corporation (Buffalo, NY), Presque Isle Cultures
(Erie, PA) (PI), or other collaborators. lab, laboratory. Clinical strains were from UMDNJ, and clinical strain DNA was from reference 2.
b
Whole-particle viral testing, sample from Zeptometrix Corporation.
c
The genomic source material was RNA.

CT values and the number of M. tuberculosis cells present in Sensitivity. Previous analytic studies examined assay perfor-
each sample was maintained over the entire dynamic range, mance using the laboratory M. tuberculosis strain H37Rv. We
confirming the ability of the assay to provide semiquantitative tested the ability of the assay to detect a more phylogenetically
estimates of input cell number (Fig. 1A). The B. globigii inter- and geographically diverse collection of M. tuberculosis isolates
nal control assay also functioned well over the entire range of (Table 2). The studies were performed at a relatively low target
added M. tuberculosis cells. The B. globigii assay was positive in concentration (10 times the LOD with DNA) to test sensitivity
all instances with an average cycle threshold of 29.7 (95% CI, under challenging conditions. The assay identified all 79 dif-
28.9 to 30.4) (Fig. 1B). Interestingly, the CT values of the B. ferent M. tuberculosis strains as M. tuberculosis positive. We
globigii assay did not appear to be affected by the number of M. also tested the ability of the assay to identify different rpoB
tuberculosis cells added to the sample. We suspect that this mutations and thereby RIF resistance in a variable clinical
decoupling is due to the heminested PCR design of both as- strain background. The assay detected RIF resistance in all 37
says. strains with known mutations in the rpoB core. The 42 RIF-
2498 BLAKEMORE ET AL. J. CLIN. MICROBIOL.

TABLE 3. Xpert MTB/RIF result for direct testing

of genomic DNA
No. of strains with the following result with
the Xpert MTB/RIF assay:

Organism M. tuberculosis positive

M. tuberculosis
Resistance Resistance negative
detected not detected

M. tuberculosis strainsa
RIF-resistant 37 0 0
RIF-susceptible 0 42 0

Bacteriab 0 0 74
Fungib 0 0 4
Virusb 0 0 11
None (negative control) 0 0 52
a
Strain characteristics are presented in Table 1. RIF, rifampin.
b
Organism characteristics are presented in Table 1.

FIG. 1. Dynamic range studies. Log dilutions of M. tuberculosis molecular beacon probes to drop out completely (CT of ⬎38).
H37Rv cells were added to 1 ml of M. tuberculosis-negative sputum to Mutations 516tac, 526cgc, and 533ccg caused a delayed CT,
final concentrations ranging from 102 to 107 CFU/ml (n ⫽ 5 or 6 per
dilution). (A) Average rpoB probe B cycle thresholds (CTs) were plot- resulting in ⌬CT Max values of 9.7 (95% CI, 9.2 to 10.2; n ⫽ 2),
ted for each M. tuberculosis concentration tested. Clinically relevant M. 7.7 (95% CI, 7.4 to 8.0; n ⫽ 3), and 4.23 (95% CI, 3.7 to 4.8;
tuberculosis concentrations all fall within the linear range of the assay. n ⫽ 3), respectively. The 509agg mutation was present only in
(B) Average B. globigii internal control probe CTs were plotted for combination with the 526cgc mutation. The sample was cor-
each concentration of M. tuberculosis tested. The B. globigii assay gave
rectly identified as RIF resistant on the basis of a delayed
an average CT of 29.7 (95% CI, 28.9 to 30.4) for all dilutions and was
not influenced by the concentration of M. tuberculosis in the sample. probe D CT corresponding to the 526cgc mutation. Interest-
Dotted lines indicate maximum valid CT values for probe positivity. ingly, the 509agg mutation did not cause a dropout or delay of
its corresponding probe.
Specificity. A previous study had shown that the Xpert
susceptible isolates without known rpoB core mutations were MTB/RIF assay did not cross-react with 20 different NTM
all identified as RIF susceptible (Table 3). The RIF-susceptible species tested at high copy numbers, suggesting that the assay
isolates had an average ⌬CT Max of 1.8 (95% CI, 1.74 to 1.92). would have high specificity (7). However, normal flora of the
The 37 RIF-resistant isolates contained 13 unique SNPs, three upper respiratory tract and respiratory tract pathogens can also
of which were present only in combination with another SNP be present in sputum samples of individuals suspected to have
(Table 2). Nine of 13 mutations caused at least one of the tuberculosis. We tested a panel of 89 organisms, including

Downloaded from https://journals.asm.org/journal/jcm on 18 May 2024 by 41.223.51.165.

TABLE 2. Characteristics of M. tuberculosis test panel isolates

No. of isolates in SNP cluster groupc Total no.

Codon of isolates
Mutation(s)b
position(s)a with the
I II IIIa IIIb IIIc IV V VIa VIb mutation(s)

511 CTG3ccg 1 1
516 GAC3gtc 1 1 2
516 GAC3tac 1 1 2
526 CAC3agc 1 1
526 CAC3cgc 1 1 2
526 CAC3gac 1 1 1 3
526 CAC3tac 1 1 1 1 4
531 TCG3tgg 1 1 1 3
531 TCG3ttg 2 2 1 1 1 4 1 1 13
533 CTG3ccg 1 1 1 3
512, 531 AGC3gcc, TCG3ttg 1 1
515, 516 ATG3att, GAC3tac 1 1
509, 526 AGC3agg, CAC3cgc 1 1

Wild type 5 4 4 4 4 4 6 7 4 42

Total no. 9 15 6 7 5 5 17 10 5 79
a
Codon position according to the convention of Telenti et al. (15).
b
The sequence of the codon for the wild type is shown first, and the sequence for the mutant codon is shown in lowercase type.
c
Single-nucleotide polymorphism cluster groups as determined in references 3 and 6.
VOL. 48, 2010 ANALYTICAL PERFORMANCE OF THE Xpert MTB/RIF ASSAY 2499

many that are found in the respiratory tract (Table 1) to fur-

ther examine the specificity of the Xpert MTB/RIF assay. Each
test was performed with 106 genomes per PCR to increase the
likelihood of a false-positive test. Our results showed that the
Xpert MTB/RIF assay did not mistakenly classify any of these
organisms as M. tuberculosis (specificity of 100%) (Table 3).
None of the organisms except Nocardia asteroides produced
any CT of ⱕ38. Nocardia cyriageorgica was detected by a single
probe with an average CT of 38.2. Partial homology between M.
tuberculosis and N. asteroides rpoB primer binding sequences
allowed for weak amplification of 106 N. asteroides genomes
and hybridization of two probes with CT values of 34.9 and
37.2. Although two of the five molecular beacons detected this
amplicon, the CT values of these molecular beacons were con-
sistently ⬎2 cycles apart from each other. This real-time PCR
signal did not fulfill the criteria for M. tuberculosis detection by
the Xpert MTB/RIF assay diagnostic algorithm and thus, were
not reported as M. tuberculosis by the assay.
Detection of RIF resistance in mixed isolate samples. Pa-
tients infected with mixtures of both RIF-susceptible and RIF-
resistant populations of M. tuberculosis could produce sputum
containing a mixture of both strains, complicating resistance
FIG. 2. Minimum detectable fraction of mutant DNA. DNA ex-
detection. We tested the ability of the assay to detect the tracted from RIF-resistant and RIF-susceptible M. tuberculosis isolates
RIF-resistant fraction of a mixed sample. M. tuberculosis DNA were mixed to six final concentrations of mutant DNA. Mutant DNA
with a wild-type rpoB sequence was mixed in various ratios with contained either an rpoB 531ttg mutation (A) or an rpoB 533ccg
M. tuberculosis DNA that contained rpoB mutations. Two dif- mutation (B). Sample mixtures equivalent to 20 times the assay limit of
detection (LOD) of 4.5 genomes per reaction were added to the
ferent rpoB mutants were studied. One mutation (rpoB 531ttg) chamber receiving eluted DNA in the full cartridge protocol and were
caused the molecular beacon complementary to the corre- processed using a PCR-only protocol. The percentage of samples (n ⫽
sponding wild-type rpoB sequence to completely “drop out.” 20) identified as RIF resistant was plotted at each concentration. As
The second mutation (rpoB 533ccg) caused the molecular bea- determined by logistic regression, there was a 95% probability of
con complementary to the corresponding wild-type rpoB se- detecting RIF resistance when the DNA mixture contained 61.2% or
100% of the total DNA in the sample mixture for a 531ttg and 533ccg
quence to have a “delayed” CT, which was sufficient for the mutant, respectively. The curves from top to bottom indicate the up-
Xpert MTB/RIF diagnostic algorithm to identify the mutant as per, middle, and lower bounds of the 95% CI. Dashed lines indicate
RIF resistant. Our mixing studies demonstrated that the pro- limit of detection.
portion of mutant DNA required for the detection of RIF
resistance was dependent on the type of mutation. For the
“dropout” 531ttg mutation, the assay could detect the presence 3A). We spiked various numbers of a modified MTBDRplus

Downloaded from https://journals.asm.org/journal/jcm on 18 May 2024 by 41.223.51.165.

of resistance with 95% certainty when the mixture contained at amplicon into M. tuberculosis-negative sputum samples
least 65.6% 531ttg DNA (95% CI, 56.9 to 88.4) (Fig. 2A). For (treated with SR), simulating amplicon cross-contamination.
the “delay” 533ccg mutation, the assay could not detect the Xpert MTB/RIF assays were then performed to test whether
presence of resistance with 95% certainty unless 100% of the the spiked amplicon was falsely identified as tuberculosis. We
DNA was mutant (95% CI, 96.8 to 100%) (Fig. 2B). found that concentrations of modified MTBDRplus less than
Assay response to amplicon contamination. The GenoType approximately 108 copies per ml of sputum did not produce
MTBDRplus assay is a reverse-hybridization line probe assay that any false-positive results (Fig. 3B). We postulated that much of
is increasingly used to test for mutations conferring isoniazid the assay specificity could be due to the efficient wash steps
and/or rifampin resistance (8). The GenoType MTBDRplus pro- occurring within the Xpert MTB/RIF cartridge. Therefore, we
tocol requires substantial manipulation of the assay’s PCR prod- also added modified MTBDRplus amplicon directly to the
uct outside closed reaction tubes, increasing the risk for amplicon cartridge DNA chamber 10. At least 104 copies of
cross-contamination. The GenoType MTBDRplus and the Xpert MTBDRplus amplicon copies needed to be added to each
MTB/RIF assays amplify a similar region of the rpoB gene. Thus, reaction to cause a false-positive result (CT ⬍ 38), even with
it was theoretically possible that MTBDRplus amplicon contam- the filter and wash steps removed (Fig. 3C). These results
ination of the Xpert MTB/RIF sample could adversely affect the suggest that MTBDRplus amplicon cross-contamination is
specificity of the Xpert MTB/RIF assay. We sequenced the unlikely to adversely affect Xpert MTB/RIF assay specificity.
MTBDRplus amplicon and identified a 168-bp consensus se-
quence consistent with the approximately 166-bp amplicon length DISCUSSION
indicated by the GenoType MTBDRplus product insert. We de-
termined that the MTBDRplus amplicon overlapped completely Near-patient molecular assays must be technically simple
with the Xpert MTB/RIF assay’s inner forward primer. However, and robust compared to assays that are performed in special-
the amplicon only overlapped the Xpert MTB/RIF assay’s outer ized laboratories. These assays must also have a clinically rel-
forward and reverse primers by 13 and 11 bp, respectively (Fig. evant dynamic range and show high levels of sensitivity and
2500 BLAKEMORE ET AL. J. CLIN. MICROBIOL.

FIG. 3. Potential risk from amplicon contamination. (A) Alignment demonstrating the overlap between the GenoType MTBDRplus PCR
amplicon (determined by sequencing) and the priming regions of the Xpert MTB/RIF assay. (B) MTBDRplus simulated amplicon (1010 to 105
copies per ml) was spiked into a mixture of sputum and SR prior to transfer to the cartridge sample chamber (n ⫽ 3 per dilution). At least 108
copies per ml were required to induce a false-positive result. Sputum spiked with 103 CFU of Mycobacterium bovis BCG per ml was processed
according to the package insert as a control for sputum inhibition (open circle). (C) MTBDRplus amplicon was spiked into TET buffer that also
contained B. globigii DNA added directly to the elution receiving chamber of an open cartridge and then processed with a PCR-only protocol (n ⫽
2 per dilution). At least 104 copies per PCR were required to induce a false-positive result. The control (open circle) was 45 copies of M. tuberculosis
DNA per reaction. Dotted lines indicate maximum valid CT values for probe positivity.

specificity. Previous tests using clinical samples had suggested mutant target hybrid itself is a substantial contributor to total
that the Xpert MTB/RIF assay would have an excellent dy- fluorescence. Under these circumstances, only a small amount
namic range and that it would perform well in sensitivity and of wild-type DNA appears to be required to overcome the
specificity studies (7); however, these properties had not been delay in CT that occurs with only mutant DNA present in the
systematically examined. The analytical testing that we per- sample. Fortunately, most of the common rpoB mutations that
formed in the current study is a useful complement to clinical cause RIF resistance produce probe dropouts rather than de-
investigations because it allows for the simulation and testing lays in the Xpert MTB/RIF assay (7). However, our results
of low-probability events in a controlled environment and at suggest that the Xpert MTB/RIF assay should not be used to
defined target concentrations. The Xpert MTB/RIF assay monitor patients for the emergence of RIF resistance during
functioned over a wide dynamic range. A wide variety of M. treatment. Approximately 5% of all RIF resistance is not en-
tuberculosis strains and common rpoB mutations were reliably coded by mutations within the rpoB core region (16), and this
detected. The specificity against NTM has been established last subset of RIF-resistant isolates will not be detectable ei-
previously (7). Our study shows that specificity is also unlikely ther in pure culture or in mixed samples by the Xpert MTB/

Downloaded from https://journals.asm.org/journal/jcm on 18 May 2024 by 41.223.51.165.

to be affected by the presence of nonmycobacterial species or RIF assay.
cross-contamination with GenoType MTBDRplus amplicons. The GenoType MTBDRplus assay was recently endorsed by
Our study also demonstrates that the identification of RIF the World Health Organization for rapid drug susceptibility
resistance in mixed samples will depend on the nature of the testing of M. tuberculosis, and it will likely be present in tertiary
rpoB mutation that is present in the RIF-resistant isolate. clinical laboratories where the Xpert MTB/RIF assay might
Several prior studies of molecular beacon-based RIF resis- also be deployed (9). We found that the Xpert MTB/RIF assay
tance testing suggest that clinical isolates do not usually con- is relatively resistant to contamination by MTBDRplus ampli-
tain mixtures of RIF-susceptible and RIF-resistant organisms cons and could safely be used in the same laboratory environ-
(5, 7, 19). However, mixed infections have been reported (18, ment. This is the result of inefficient amplification of the
20), and studies of mixed samples are warranted. We found MTBDRplus amplicon, which has only limited overlap with
that the ability of the assay to detect mixed samples is depen- the outer priming regions of the Xpert MTB/RIF assay. Ad-
dent on the particular rpoB mutation that is present in the ditionally, the most likely route for amplicon entry will be by
mixture. Mutations that effectively block the hybridization of contaminating the clinical sample or sample-SR mixture. This
the assay’s molecular beacon to the mutant rpoB sequence mixture is filtered and washed within the GeneXpert instru-
appear to be more easily discernible than those that merely ment several times prior to PCR, significantly diluting a po-
inhibit probe hybridization. For mixtures that contain mutant tential amplicon contaminant. Other commercial or in-house
species with a “dropout” mutation, total fluorescence is almost assays that amplify a larger region of the rpoB gene could
completely dependent on the fraction of molecular beacon provide a much more efficient amplification template and
population binding to the available wild-type targets. This al- could result in increased probability of false-positive results if
lows for a substantial amount of wild-type target to accumulate used without proper contamination controls. The Xpert itself
before the affected molecular beacon reports a CT within 3.5 contains amplicons within the sealed cartridge, reducing or
cycles of the other unaffected molecular beacons. When the eliminating the need for the precautions typically associated
mutant species is a “delay” mutant, the molecular beacon- with other nucleic acid amplification tests.
VOL. 48, 2010 ANALYTICAL PERFORMANCE OF THE Xpert MTB/RIF ASSAY 2501

ACKNOWLEDGMENTS 8. Hillemann, D., S. Rusch-Gerdes, and E. Richter. 2007. Evaluation of the

GenoType MTBDRplus assay for rifampin and isoniazid susceptibility test-
This work was supported by National Institutes of Health grants ing of Mycobacterium tuberculosis strains and clinical specimens. J. Clin.
R41-AI52523 and R42-AI52523 and a grant from the Foundation for Microbiol. 45:2635–2640.
Innovative New Diagnostics. 9. Huang, W.-L., H.-Y. Chen, Y.-M. Kuo, and R. Jou. 2009. Performance as-
We gratefully acknowledge the United Nation’s Children’s Fund/ sessment of the GenoType MTBDRplus test and DNA sequencing in de-
UNDP/World Bank/WHO Special Programme for Research and tection of multidrug-resistant Mycobacterium tuberculosis. J. Clin. Micro-
biol. 47:2520–2524.
Training in Tropical Diseases for supplying the banked M. tuberculosis
10. Joloba, M. L., J. L. Johnson, A. Namale, A. Morrissey, A. E. Assegghai, R. D.
strains. Fungal DNA samples were the generous gifts of Arturo Case- Mugerwa, J. J. Ellner, and K. D. Eisenach. 2000. Quantitative sputum
devall and Joshua Nosanchuk at the Albert Einstein College of Med- bacillary load during rifampin-containing short course chemotherapy in hu-
icine. man immunodeficiency virus-infected and non-infected adults with pulmo-
David Alland is one of a group of coinvestigators who invented nary tuberculosis. Int. J. Tuber. Lung Dis. 4:528–536.
molecular beacon technology and receive income from licensees. J.K., 11. Migliori, G. B., A. Matteelli, D. Cirillo, and M. Pai. 2008. Diagnosis of
M.R.O., and M.J. are employed by Cepheid, which manufactures and multidrug-resistant tuberculosis and extensively drug-resistant tuberculosis:
markets the Xpert MTB/RIF assay. current standards and challenges. Can. J. Infect. Dis. Med. Microbiol. 19:
169–172.
REFERENCES 12. Nyendak, M. R., D. A. Lewinsohn, and D. M. Lewinsohn. 2009. New diag-
1. Balasingham, S. V., T. Davidsen, I. Szpinda, S. A. Frye, and T. Tonjum. nostic methods for tuberculosis. Curr. Opin. Infect. Dis. 22:174–182.
2009. Molecular diagnostics in tuberculosis: basis and implications for ther- 13. Parsons, L. M., A. Somoskovi, R. Urbanczik, and M. Salfinger. 2004. Lab-
apy. Mol. Diagn. Ther. 13:137–151. oratory diagnostic aspects of drug resistant tuberculosis. Front Biosci.
2. Chakravorty, S., B. Aladegbami, M. Burday, M. Levi, S. A. Marras, D. Shah, 9:2086–2105.
H. H. El-Hajj, F. R. Kramer, and D. Alland. 2010. Rapid universal identifi- 14. Piatek, A. S., S. Tyagi, A. C. Pol, A. Telenti, L. P. Miller, F. R. Kramer, and
cation of bacterial pathogens from clinical cultures using a novel sloppy D. Alland. 1998. Molecular beacon sequence analysis for detecting drug
molecular beacon melting temperature signature technique. J. Clin. Micro- resistance in Mycobacterium tuberculosis. Nat. Biotechnol. 16:359–363.
biol. 48:258–267. 15. Telenti, A., P. Imboden, F. Marchesi, D. Lowrie, S. Cole, M. J. Colston, L.
3. Chakravorty, S., B. Aladegbami, A. S. Motiwala, Y. Dai, H. Safi, M. Brima- Matter, K. Schopfer, and T. Bodmer. 1993. Detection of rifampicin-resis-
combe, D. Helb, and D. Alland. 2008. Rifampin resistance, Beijing-W clade- tance mutations in Mycobacterium tuberculosis. Lancet 341:647–650.
single nucleotide polymorphism cluster group 2 phylogeny, and the Rv2629 16. Van Der Zanden, A. G., E. M. Te Koppele-Vije, N. Vijaya Bhanu, D. Van
191-C allele in Mycobacterium tuberculosis strains. J. Clin. Microbiol. 46: Soolingen, and L. M. Schouls. 2003. Use of DNA extracts from Ziehl-
2555–2560. Neelsen-stained slides for molecular detection of rifampin resistance and
4. Colangeli, R., D. Helb, C. Vilcheze, M. H. Hazbon, C. G. Lee, H. Safi, B. spoligotyping of Mycobacterium tuberculosis. J. Clin. Microbiol. 41:1101–
Sayers, I. Sardone, M. B. Jones, R. D. Fleischmann, S. N. Peterson, W. R. 1108.
Jacobs, Jr., and D. Alland. 2007. Transcriptional regulation of multi-drug 17. van Embden, J. D., M. D. Cave, J. T. Crawford, J. W. Dale, K. D. Eisenach,
tolerance and antibiotic-induced responses by the histone-like protein Lsr2 B. Gicquel, P. Hermans, C. Martin, R. McAdam, T. M. Shinnick, and P. M.
in M. tuberculosis. PLoS Pathog. 3:e87. Small. 1993. Strain identification of Mycobacterium tuberculosis by DNA
5. El-Hajj, H. H., S. A. Marras, S. Tyagi, F. R. Kramer, and D. Alland. 2001. fingerprinting: recommendations for a standardized methodology. J. Clin.
Detection of rifampin resistance in Mycobacterium tuberculosis in a single Microbiol. 31:406–409.
tube with molecular beacons. J. Clin. Microbiol. 39:4131–4137. 18. van Rie, A., T. C. Victor, M. Richardson, R. Johnson, G. D. van der Spuy,
6. Filliol, I., A. S. Motiwala, M. Cavatore, W. Qi, M. H. Hazbon, M. Bobadilla E. J. Murray, N. Beyers, N. C. G. van Pittius, P. D. van Helden, and R. M.
del Valle, J. Fyfe, L. Garcia-Garcia, N. Rastogi, C. Sola, T. Zozio, M. I. Warren. 2005. Reinfection and mixed infection cause changing Mycobacte-
Guerrero, C. I. Leon, J. Crabtree, S. Angiuoli, K. D. Eisenach, R. Durmaz, rium tuberculosis drug-resistance patterns. Am. J. Respir. Crit. Care Med.
M. L. Joloba, A. Rendon, J. Sifuentes-Osornio, A. Ponce de Leon, M. D. 172:636–642.
Cave, R. Fleischmann, T. S. Whittam, and D. Alland. 2006. Global phylogeny 19. Varma-Basil, M., H. El-Hajj, R. Colangeli, M. H. Hazbon, S. Kumar, M.
of Mycobacterium tuberculosis based on single nucleotide polymorphism Bose, M. Bobadilla-del-Valle, L. G. Garcia, A. Hernandez, F. R. Kramer,
(SNP) analysis: insights into tuberculosis evolution, phylogenetic accuracy of J. S. Osornio, A. Ponce-de-Leon, and D. Alland. 2004. Rapid detection of
other DNA fingerprinting systems, and recommendations for a minimal rifampin resistance in Mycobacterium tuberculosis isolates from India and
standard SNP set. J. Bacteriol. 188:759–772. Mexico by a molecular beacon assay. J. Clin. Microbiol. 42:5512–5516.
7. Helb, D., M. Jones, E. Story, C. Boehme, E. Wallace, K. Ho, J. Kop, M. R. 20. Victor, T. C., H. Lee, S. N. Cho, A. M. Jordaan, G. van der Spuy, P. D. van
Owens, R. Rodgers, P. Banada, H. Safi, R. Blakemore, N. T. Lan, E. C. Helden, and R. Warren. 2002. Molecular detection of early appearance of

Downloaded from https://journals.asm.org/journal/jcm on 18 May 2024 by 41.223.51.165.

Jones-Lopez, M. Levi, M. Burday, I. Ayakaka, R. D. Mugerwa, B. McMillan, drug resistance during Mycobacterium tuberculosis infection. Clin. Chem.
E. Winn-Deen, L. Christel, P. Dailey, M. D. Perkins, D. H. Persing, and D. Lab Med. 40:876–881.
Alland. 2010. Rapid detection of Mycobacterium tuberculosis and rifampin 21. Young, D. B., M. D. Perkins, K. Duncan, and C. E. Barry III. 2008. Con-
resistance by use of on-demand, near-patient technology. J. Clin. Microbiol. fronting the scientific obstacles to global control of tuberculosis. J. Clin.
48:229–237. Invest. 118:1255–1265.