- Email: [email protected]
Analysis of discrepant results between the Genotype® MTBDRplus assay and an antimicrobial drug susceptibility test for isoniazid-resistant tuberculosis
Respiratory Medicine 122 (2017) 12e17
Contents lists available at ScienceDirect
Respiratory Medicine journal homepage: www.elsevier.com/locate/rmed
Analysis of discrepant results between the Genotype® MTBDRplus assay and an antimicrobial drug susceptibility test for isoniazidresistant tuberculosis Kyung-Wook Jo a, 1, Yoomi Yeo b, 1, Heungsup Sung c, Mi-Na Kim c, Tae Sun Shim a, * a b c
Division of Pulmonary and Critical Care Medicine, University of Ulsan College of Medicine, Asan Medical Center, Seoul, South Korea Department of Internal Medicine, Chunggu Sungshim Hospital, South Korea Department of Laboratory Medicine, University of Ulsan College of Medicine, Asan Medical Center, Seoul, South Korea
a r t i c l e i n f o
a b s t r a c t
Article history: Received 2 April 2016 Received in revised form 9 October 2016 Accepted 21 November 2016 Available online 22 November 2016
Background: We investigated discrepant results determined using the Genotype®MTBDRplus assay and a conventional antimicrobial drug susceptibility test (ADST) for isoniazid (INH) resistance using sequencing analysis and analyzed the clinical course of patients with discrepant results. Methods: Among 1373 MTBDRplus assays performed at our tertiary referral center in South Korea between August 2009 and December 2015, the results for 46 (3.4%) differed from those for ADST. KatG and inhA gene sequencing analysis results were available for 23 patients. ADSTs were carried out using the € wensteineJensen media. absolute concentration method with Lo Results: Results from 11 patients indicated INH susceptibility by MTBDRplus assay and INH resistance by ADST. For 5 of these patients, sequencing revealed no evidence of mutations, whereas specific mutations were detected in the remaining 6 patients. These should have been detected using the MTBDRplus assay. The other 12 patients had isolates with the opposite discrepancy, that is INH resistance by MTBDRplus assay but INH susceptibility by ADST. For 7 of these cases, sequencing results were consistent with those of the MTBDRplus assay. However, sequencing analysis did not explain the discrepancies in the remaining 5 patients. All 23 patients with discrepant results received individualized treatment regimens determined by the attending physician according to their test results and susceptibility to other drugs, such as rifampin. Good outcomes were reported for the majority. Conclusion: Discrepancies between test results for INH resistance on the MTBDRplus assay and ADST appear to be infrequent. Gene sequencing analysis is useful for identifying the cause of the discrepancy. © 2016 Elsevier Ltd. All rights reserved.
Keywords: Isoniazid Discrepancy MTBDRplus assay KatG gene sequencing inhA gene sequencing
1. Introduction Tuberculosis (TB) is a global health concern, affecting both developed and developing countries [1]. The incidence of TB remains high worldwide, with 9.6 million new cases and 1.5 million deaths in 2014 [2]. Although the incidence of TB is declining, it remains a major problem in South Korea, where a plateau of approximately 100 new cases per 100,000 persons has been reached in the past decade [3].
* Corresponding author. Division of Pulmonary and Critical Care Medicine, University of Ulsan College of Medicine, Asan Medical Center, 388-1 Pungnap-Dong, Songpa-gu, Seoul 138-736, South Korea. E-mail address: [email protected] (T.S. Shim). 1 Kyung-Wook Jo and Yoomi Yeo contributed equally to this work. http://dx.doi.org/10.1016/j.rmed.2016.11.016 0954-6111/© 2016 Elsevier Ltd. All rights reserved.
The control of TB has become increasingly difficult because of the rise of multidrug-resistant disease (MDR-TB). In some settings, up to one third of newly diagnosed cases result from MDR-TB [2]. In South Korea, the burden of MDR-TB remains relatively low, with only 4.6% patients with the disease having MDR-TB was in 2008 [4]. For optimal treatment and reduction of transmission, rapid identification of drug-resistant TB is vital [5]. Because the results of a solid media-based antimicrobial drug susceptibility test (ADST) can take up to 3 months to obtain in South Korea [6], rapid direct tests have been employed for prompt identification of patients infected with drug-resistant organisms. The most widely used rapid molecular drug resistance tests are the Xpert MTB/RIF (Cepheid, Sunnyvale, CA, USA) and Genotype®MTBDRplus (Hain Lifescience, Nehren, Germany) assays. Using the Genotype®MTBDRplus assay, we previously reported a median turnaround time for detection of
K.-W. Jo et al. / Respiratory Medicine 122 (2017) 12e17
MDR-TB of 5.3 days with acceptable sensitivity and specificity [7]. However, physicians occasionally encounter cases with a discrepancy between isoniazid (INH) resistance results from the MTBDRplus assay and conventional ADSTs [7e11]. Reports of the reasons for these discrepancies are limited, as is information on treatment and outcome. We designed this retrospective study to investigate a convenience sample of TB cases managed in our institution and to review the treatment and outcome of patients with discrepant results. 2. Patients and methods 2.1. Study subjects and data collection The study was conducted at Asan Medical Center, a 2700-bed referral hospital in Seoul, South Korea. Approximately 900 patients with TB receive diagnosis and treatment at this hospital annually. The medical records of patients with discrepant results, including sequencing analysis, were analyzed retrospectively in January 2016. The study protocol was approved by the Institutional Review Board of Asan Medical Center. 2.2. Bacteriological study and ADSTs Acid-fast bacillus (AFB) smears were examined by ZiehleNeelsen staining. AFB culture was carried out using both solid (Ogawa medium; Korean Institute of Tuberculosis, South Korea) and liquid (BACTEC 960 Mycobacterial Growth Indicator Tube; Becton Dickinson, Sparks, MD, USA) media. Positive liquid media cultures or colonies on solid medium were subjected to ZiehleNeelsen staining and polymerase chain reaction assay using Seeplex TB detection (Seegen, Seoul, Korea) to differentiate between Mycobacterium tuberculosis complex and non-tuberculous mycobacteria. When growth of M. tuberculosis was detected on culture, ADST was generally requested according to Korean national TB guidelines, which recommend sensitivity testing of the first positive M. tuberculosis isolate for all patients. However, for various reasons, ADST had not been performed in all cases. Conventional ADSTs were performed using the absolute concentration method with €wensteineJensen media at the Korean Institute of Tuberculosis, Lo South Korea's TB reference laboratory. Wells contained the following critical concentrations of anti-TB drugs: 0.2 mg/mL of INH and 40 mg/mL of rifampin (RIF). Growth exceeding that of the control wells by >1% was considered to indicate drug resistance. 2.3. MTBDRplus assay and DNA sequencing The MTBDRplus assay was performed in our center according to the manufacturer's instructions. The decision concerning whether to carry out an MTBDRplus assay was made by the attending physician. The MTBDRplus assay can be requested for a specimen with a positive AFB smear. If the smear is negative, the MTBDRplus assay can only be requested for a positive M. tuberculosis culture result. DNA sequencing was performed using a polymerase chain reaction-direct sequencing system (Cosmo Co., Seoul, South Korea). A 670 base-pair (bp) fragment of the katG gene (containing codon 315), 248 bp of the inhA promoter (including 8 and 15), and, beginning in July 2014, 472 bp of the oxyR-ahpC gene were sequenced. The following forward and reverse primers (50 e30 ) were used: KatGR CCAAGGTATCTCGCAACG, KatGF CACACTTTCGGTAAGACCCA, inhAR CCAGCACCTCGACCGGACC, inhAF ATTCGTAGGGCGTCAATACA, oxyR-ahpCR CGATGAGAGCGGTGAGCTG and oxyR-ahpCF AACGTCGACTGGCTCATATC. Primers were synthesized
13
by JieLi Bio Co (China). PCR conditions were denaturation at 94 C for 10 min, followed by 35 cycles of amplification at 94 C for 30 s, 60 C for 30 s, and 72 C for 30 s. PCR mixtures were prepared using 2 GoldStar Best MasterMix (CWBio Co., China). The PCR products were used as templates for targeted DNA sequencing. To confirm the sequence amplified PCR product, direct sequencing was performed on an ABI 373 DNA analyzer according to the protocol supplied by the manufacturer using the BigDye Terminator v3.1 Cycling Sequencing kit. Sequencing data were assembled and analyzed with BioEdit software. The mutations were determined by comparing with the H37Rv sequence of katG, inhA, and oxyR-ahpC genes from the GenBank database (http://www.ncbi.nih.gov/gene). The decision regarding whether to perform DNA sequencing analysis was made by the attending physician. As noted, oxyR-ahpC gene sequencing was only available from July 2014. 2.4. Treatment outcome Treatments were not standardized. They were individualized according to the previous treatment history and ADST results for other drugs, such as RIF. Treatment outcome categories were according to the revised 2013 WHO definitions [12]. The favorable outcome was defined as the sum of patients who were cured and those who completed treatment. In addition, we also investigated the incidence of recurrence within 1 year after treatment completion. 2.5. Statistical analyses Data are presented as means for continuous variables and percentages for discrete variables. Variables were analyzed using simple descriptive statistics. The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of the MTBDRplus assay for INH resistance were calculated as compared with conventional ADST results [13]. 3. Results 3.1. Characteristics of the study subjects and test results A total of 1551 MTBDRplus assays were carried out between August 2009 and December 2015. Of 1389 MTBDRplus assays for which corresponding ADST results were available, there were 46 (3.4%) cases with discrepant results between the two tests. Table 1 compares the rates of resistance to INH and RIF detected using MTBDRplus assay vs. ADST. Among the cases with discrepant results, 23 had both katG and inhA sequencing results and were ultimately included in the analysis (Fig. 1). The mean age of the 23 patients with discrepant INH resistance results was 52.1 years, and males were predominant (78.3%, 18/23). All patients were diagnosed with culture-confirmed pulmonary TB. The isolates from 11 patients' isolates were reported to be INH susceptible using the
Table 1 Summary of isoniazid and rifampicin resistance by the MTBDRplus assay and antimicrobial-susceptibility test in 1373 isolates. Antimicrobial susceptibility test MTBDRplus assay Isoniazid Resistant Susceptible Rifampicin Resistant Susceptible
Resistant n (%)
Susceptible n (%)
170 (12.4%) 23 (1.7%)
23 (1.7%) 1157 (84.3%)
115 (8.4%) 3 (0.2%)
15 (1.1%) 1256 (91.5%)
14
K.-W. Jo et al. / Respiratory Medicine 122 (2017) 12e17
A total of 1,551 MTBDRplus assays were performed between August 2009 and December 2015 ADST not requested (n = 107) NTM growth (n = 37) No growth on AFB culture (n = 14) No amplification (n = 4)
Both MTBDRplus assays and ADST results were available for RIF susceptibility (n = 1,389)
INH-resistant on MTBDRplus assay (n = 193)
INH-resistant on ADST (n = 23)
INH-susceptible on ADST (n = 170)
INH-susceptible on MTBDRplus assay (n = 1,196)
INH-resistant on ADST (n = 23)
Sequencing results for both KatG and inhA were not available (n = 12)
Sequencing results for both KatG and inhA were not available (n = 11)
INH-susceptible on MTBDRplus assay and INH-resistant on ADST (n = 12)
INH-susceptible on ADST (n = 1,173)
INH-susceptible on MTBDRplus assay and INH-resistant on ADST (n = 11)
Fig. 1. Study flow chart.; Abbreviations: ADST, antimicrobial drug susceptibility test; NTM, non-tuberculous mycobacteria; AFB, acid-fast bacillus; INH, isoniazid.
MTBDRplus assay and INH resistant by ADST. The isolates from the other 12 patients had the opposite discrepancy; INH resistant by MTBDRplus assay and INH susceptible by ADST. 3.2. Performance of the MTBDRplus assay in detection of INH resistance The performance of the MTBDRplus assay was analyzed in comparison with the results of conventional ADST in the 1389 cases for which both tests were performed. The sensitivity, specificity, PPV, and NPV of the MTBDRplus assay for detecting INH resistance were 88.1% (95% confidence interval [CI], 82.4e92.1), 98.1% (95% CI 97.0e98.7), 88.1% (95% CI 82.4e92.1), and 98.1% (95% CI 97.0e98.7), respectively, compared to the results of conventional ADSTs. 3.3. Discrepancy analysis: INH susceptible by MTBDRplus assay and INH resistant by ADST Table 2 shows the results of sequencing analysis for the 11 patients found to be INH susceptible using the MTBDRplus assay and INH resistant by ADST. Sequencing analysis of the KatG, inhA and oxyR-ahpC genes for patients 1e5 revealed no evidence of mutations, whereas sequencing analysis for patients 6e11 revealed specific mutations that should have been detected by the MTBDRplus assay. The M. tuberculosis RIF-susceptibility results, treatment regimens and durations, and treatment outcomes for these patients are summarized in Table 3. Patients 1e6 and 11 received the usual treatment regimen for INH-resistant TB, and the remaining 4 were treated using the MDR-TB regimen because their MTBDRplus assay or ADST indicated RIF resistance. All patients evaluated had favorable outcomes.
3.4. Discrepancy analysis: INH resistant by MTBDRplus assay and INH susceptible by ADST Table 4 shows the results of sequencing analyses and the MTBDRplus assay (wild-type and mutated band pattern) for the 12 patients found to be INH resistant using the MTBDRplus assay and INH susceptible by ADST. Sequencing analysis of the KatG and inhA genes for 7 patients (patients 12e18) were in perfect agreement with the results of the MTBDRplus assay. In 4 patients (19e21 and 23), sequencing analysis revealed no evidence of mutations. The sequencing analysis results for patient 22 were inconsistent with the results of the MTBDRplus assay. The M. tuberculosis RIF-susceptibility test results as well as treatment regimen, duration and outcome, for these 12 patients are shown in Table 5. All patients were treated with reference to susceptibility to other drugs, such as RIF as decided by the attending physician. The outcomes, where evaluated, were favorable. 4. Discussion Recently, two molecular diagnostic tests, the Xpert MTB MTB/ RIF assay and the Genotype® MTBDRplus assay, were developed and endorsed by the WHO for the rapid detection of drug-resistant TB [14]. Unlike the Xpert MTB/RIF assay, which only detects genetic mutations associated with RIF resistance [15], the MTBDRplus assay has the advantage of detecting both RIF and INH resistance. In this study, we analyzed the INH-susceptibility results from 1373 MTBDRplus assays compared with conventional ADSTs. In addition, we investigated the cause of identified discrepancies by sequencing the katG, inhA and oxyR-ahpC genes and analyzed the treatment outcomes of patients with discrepant INH-susceptibility results. The most important findings of this study are that the diagnostic
K.-W. Jo et al. / Respiratory Medicine 122 (2017) 12e17
15
Table 2 Sequencing results from 11 patient samples reported as isoniazid susceptible by MTBDRplus assay and isoniazid resistant by an antimicrobial-susceptibility test. Patient no.
1 2 3 4 5 6 7 8 9 10 11
Sputum AFB smear grade
MTBDRplus assay results Wild-type loss
Mutation detected
katG mutation
inhA mutation
oxyR-aphC mutation
3þ Negative* 3þ Negative* 1þ 1þ Negative* Negative* Negative* 4þ Negative*
Not Not Not Not Not Not Not Not Not Not Not
Not Not Not Not Not Not Not Not Not Not Not
WT WT WT WT WT AGC315ACC WT WT WT WT WT
WT WT WT WT WT WT 8 T deletion 15 C deletion 15 C deletion 15 C / A 15 C / T
WT WT ND ND ND ND ND ND ND ND ND
detected detected detected detected detected detected detected detected detected detected detected
Sequencing results
detected detected detected detected detected detected detected detected detected detected detected
Abbreviations:AFB, Acid-fast bacillus; WT, wild type; ND, not done. * In cases with negative AFB smears, cultured isolates were used for MTBDRplus assay.
Table 3 Treatment regimen, rifampicin susceptibility, and treatment outcome of 11 patients with tuberculosis reported as isoniazid susceptible by MTBDRplus assay and isoniazid resistant by an antimicrobial-susceptibility test. Patient Age Gender RIF-susceptibility: no. MTBDRplus assay
Initial RIFsusceptibility: regimen ADST
Duration of first Subsequent Duration of subsequent Treatment regimen (months) regimen regimen (months) outcome
Recurrence within 1 year
1
71
Male
Susceptible
Susceptible
2
Cure
No recurrence
2
58
Male
Resistant*
Susceptible*
Cure
3
61
Female Susceptible
Susceptible
Less than 1 year after treatment completion No recurrence
4
40
Female Susceptible
Susceptible
5
77
Male
Susceptible
Susceptible
6
22
Female Susceptible
Susceptible
7
52
Male
Resistant*
Susceptible*
8
71
Male
Resistant
Resistant
9
30
Male
Resistant
Resistant
10
52
Male
Resistant
Resistant
11
77
Male
Susceptible
Susceptible
INH, RIF, EMB, PZA INH, HD-RIF, PZA INH, RIF, EMB, PZA INH, RIF, LVFX INH, RIF, EMB, PZA INH, RIF, EMB, PZA MXF, KM, PTH, CS, PAS LZD, MXF, PTH, CS LZD, MXF, KM, AMK, CLR LVFX, KM, PTH, PAS, CS INH, RIF
1.5 2.5 2
2.5
MXF, RIF, PZA MXF, HDRIF, EMB LVFX, RIF, PZA LVFX, RIF, CS
6
RIF, EMB, PZA
3.5
3.5
Cure
13
Completion Lost to follow-up after treatment completion Not N/A evaluated Cure No recurrence
14
Completion No recurrence
14
Completion No recurrence
4
MXF, KM, AMC, CLR
18
Cure
No recurrence
12
LVFX, PAS, AMC LVFX, RIF
11
Cure
No recurrence
8
Cure
Died 8 months after treatment completion
1.5
Abbreviations: RIF, rifampicin; INH, isoniazid; EMB, ethambutol; PZA, pyrazinamide; MXF, moxifloxacin; HD-RIF, high-dose rifampicin; LVFX, levofloxacin; CS, cycloserine; KM, kanamycin; PTH, prothionamide; PAS, p-aminoslicylic acid; LZD, linezolid; AMK, amikacin; CLR, clarithromycin; AMC, amoxicillin/clavulanate. *These strains had disputed rpoB mutations (CTG511CCG and CTG533CCG, respectively).
accuracy of the MTBDRplus assay for detection of INH resistance is satisfactory and that sequencing analysis is helpful in assessing the causes of discrepant INH-susceptibility results. There have been a limited number of reports using sequencing analysis to investigate discrepancies between INH-susceptibility results from the MTBDRplus assay and conventional ADSTs [7e9,11]. Previously, we showed that discrepant results were recorded from 10/428 (2.3%) specimens using these two tests [7]. In that study, we found that the results of the MTBDRplus assay were compatible with those obtained by sequencing analysis in all six cases for which data were available. Lacoma et al. evaluated the performance of the MTBDRplus assay using 62 clinical strains from 28 patients in Spain [9]. They found that, among 48 strains indicated to be INH resistant by ADST, as many as 13 were reported to be INH susceptible on MTBDRplus. Sequencing analysis revealed several possible reasons for these discrepancies. Other studies have also reported discrepancies in test results for INH resistance and
performed sequencing analysis for small numbers of cases [8,11]. The present study represents the largest number of discrepant INH results analyzed by gene sequencing, although sequencing was only performed in half of the cases. There were various reasons underlying the discrepancies, and gene sequencing was able to identify these in the majority of cases. Of the 11 samples reported as INH-sensitive on MTBDRplus assay but resistant by ADST (Table 2), no mutations were found in five. The discrepant results for these cases may have been because of (1) rare gene mutations (other than katG, inhA and oxyR-ahpC) causing INH resistance, such as those in kasA or mshA [16]; (2) mutations of katG, inhA or oxyR-ahpC not covered by the MTBDRplus assay probes and the outside the regions sequenced; or (3) inaccurate ADST results [17]. By contrast, the responsible mutation should have been identified by the MTBDRplus assay in the remaining six samples. Technical errors may explain the discrepancies in these cases.
16
K.-W. Jo et al. / Respiratory Medicine 122 (2017) 12e17
Table 4 Sequencing results from 12 patient samples testing reported as isoniazid-resistant using theby MTBDRplus assay and isoniazid-susceptible by an antimicrobial-drug susceptibility test. Patient no.
Sputum AFB smear grade
*
12 13 14 15 16 17 18 19 20 21 22 23
Negative 1þ 1þ Negative* 1þ Negative* Negative* 2þ Negative* Negative* 1þ Negative*
MTBDRplus assay results
Sequencing results
Wild-type loss
Mutation detected
katG mutation
inhA mutation
oxyR-ahpC mutation
katG katG inhA WT1 inhA WT1 inhA WT1 inhA WT2 Not detected Not detected Not detected inhA WT1 inhA WT1 inhA WT1
katG MUT1 katG MUT2 inhA MUT1 inhA MUT1 inhA MUT1 inhA MUT3A inhA MUT3B katG MUT1 katG MUT1 inhA MUT1 inhA MUT2 Not detected
AGC315ACC AGC315ACA WT WT WT WT WT WT WT WT WT WT
WT WT 15 C / T 15 C / T 15 C / T 8 T / C 8 T / A WT WT WT 15 C / T WT
ND WT WT WT ND ND WT WT WT ND WT ND
Abbreviations: AFB, Acid-fast bacillus; WT, wild type; ND, not done. In cases with negative AFB smears, cultured isolates were used for MTBDRplus assay.
*
Table 5 Treatment regimen, rifampicin susceptibility, and treatment outcome for 12 patients with tuberculosis reported as isoniazid resistant by MTBDRplus assay and isoniazid susceptible by an antimicrobial-susceptibility test. Patient Age Gender RIF-susceptibility: no. MTBDRplus assay
Initial RIFsusceptibility: regimen ADST
Duration of first regimen (months)
Subsequent regimen
Duration of subsequent Treatment regimen (months) outcome
1 year recurrence
12
40
Male
Resistant
Susceptible
INH, RIF, EMB, PZA
6
18
Cure
No recurrence
13
76
Male
Susceptible
Susceptible
2
6
Cure
No recurrence
14
44
Male
Susceptible
Susceptible
15
48
Male
Susceptible
Susceptible
16
49
Female Resistant
Resistant
17
56
Male
Susceptible
Susceptible
18
71
Male
Susceptible
Susceptible
19
24
Male
Susceptible
Susceptible
20
44
Male
Susceptible
Susceptible
21
69
Female Susceptible
Susceptible
22
45
Male
Susceptible
Susceptible
RIF, EMB, PZA LVFX, RIF, EMB, SM MXF, RIF, EMB MXF, KM, PZA, PTH, CS RIF, EMB, PZA MXF, EMB, CS INH, RIF, EMB, PZA RIF, EMB, PZA MXF, RIF, EMB LVFX, RIF
KM, PAS, PTH, MXF, CS INH, RIF
23
48
Male
Susceptible
Susceptible
INH, RIF, EMB, PZA
14
Completion No recurrence
9
Cure
4 2
INH, MXF, PZA, CS INH, RIF
13
Cure
Less than 1 year after treatment completion No recurrence
7
Cure
No recurrence
Cure
Less than 1 year after treatment completion No recurrence
6
Cure
9
Not evaluated Cure
9 2
INH, RIF
4
N/A No recurrence
Completion Less than 1 year after treatment completion Cure No recurrence
Abbreviations: RIF, rifampicin; INH, isoniazid; EMB, ethambutol; PZA, pyrazinamide; KM, kanamycin; PAS, p-aminoslicylic acid; PTH, prothionamide; MXF, moxifloxacin; CS, cycloserine; LVFX, levofloxacin; SM, streptomycin.
The sequencing analysis also revealed a variety of reasons for the discrepancies identified in 12 samples reported as INH resistant by MTBDRplus assay but sensitive on ADST. For seven of these, the sequencing analysis detected exactly the same mutation as that detected by the MTBDRplus assay. The discrepant results in these cases could be related to the reliability of ADSTs using the absolute concentration method, because the power of this method to discriminate between INH-susceptible and INH-resistant strains is imperfect and can occasionally lead to mislabeling of INH-resistant strains as INH susceptible and vice versa [17]. There may be a number of reasons for the discrepancies observed in the remaining five samples, including the following possibilities: (1) heteroresistance (patients 19e21) [10], (2) technical error in operation of the MTBDRplus assay, (3) co-infection (for patient 22), or (4) mutation of the inhA gene in a region outside that sequenced (for
patient 23). The interpretation of discrepant results and the choice of treatment were solely the responsibility of the attending physician. Later-generation fluoroquinolones (levofloxacin or moxifloxacin) were widely used in our study population. In cases resistant to INH, but not to RIF, a standard 6-month regimen in which INH is replaced by a later-generation fluoroquinolone is likely to lead to similar treatment outcomes [18]. A number of patients were treated with other drugs, including ethambutol or cycloserine, in addition to a combination of a later-generation fluoroquinolone and RIF, because of concerns regarding drug interactions between the latter two [19]. In patients whose MTBDRplus assays and ADST revealed RIF-monoresistance, MDR-TB regimens were usually prescribed for initial treatment (patients 8e10, Table 3) irrespective of the results of INH resistance tests. This is in accordance with the recent WHO
K.-W. Jo et al. / Respiratory Medicine 122 (2017) 12e17
recommendation to prescribe an MDR-TB regimen for RIFmonoresistant TB [20]. Previously, we reported that the sensitivity, specificity, positive predictive value, and negative predictive value of the MTBDRplus assay for detecting INH resistance were 93.8%, 98.3%, 92.7%, and 98.6%, respectively. Higher sensitivity and/or positive predictive values were reported in South Africa [21] and Thailand [11], whereas, one study in Spain [9] reported a much lower sensitivity (73%) for the detection of INH resistance. Hence, the performance of the MTBDRplus assay for detection of INH resistance appears to be dependent on regional differences in the frequency of specific mutations causing the resistance phenotype [22]. The current study has several limitations. Most significantly, it was conducted at a single referral center and included a relatively small number of cases. In addition, because requesting an MTBDRplus assay was at the sole discretion of the attending physician, it was not performed routinely for all patients diagnosed with TB in our center. Moreover, because the attending physician also decided whether or not to conduct sequencing analysis, gene sequencing was not performed in half of the cases with discrepant results. Sequencing analysis of the oxyR-ahpC gene was only in use later in the study period and so was conducted in only about a third of the cases. However, our results bring into question the value of oxyR-ahpC gene sequencing, because no mutations of this gene were found in the present study. Unlike other comparable studies, we did not use the Mycobacterial Growth Indicator Tube for Antimycobacterial-Susceptibility Testing or the proportion method to detect INH resistance. Because we did not carry out genotyping, we were unable to determine whether mixed infections were present in some cases where this was potentially a factor. Finally, isolates yielding discrepant results were not re-tested by all methods to ascertain the reproducibility of the initial testing result. In conclusion, discrepancies between the results of tests for INH resistance with the MTBDRplus assay or ADST using the absolute concentration method appear to be infrequent. As gene sequencing was useful to identify the cause of discrepancies in many cases, we propose that it should be performed in all cases with discrepant results. Despite a variety of possible reasons for the discrepancies, patients generally achieved good outcomes with individualized treatment regimens. Funding This work was supported by the Asan Institute for Life Sciences (grant number: 2015-617), Asan Medical Center, Seoul, Korea. Conflicts of interest None declared. Acknowledgments The authors would like to thank the Asan Institute for Life Sciences, Asan Medical Center, Seoul, Korea for financial support for
17
this project. References [1] Y.J. Ryu, Diagnosis of pulmonary tuberculosis: recent advances and diagnostic algorithms, Tuberc. Respir. Dis. 78 (2015) 64e71. [2] Word Health Organization, Global Tuberculosis Report 2015, World Health Organization, Geneva, 2015. Avaiable at, http://apps.who.int/iris/bitstream/ 10665/191102/1/9789241565059_eng.pdf. accessed on September 2016. [3] Y.K. Park, Y.S. Park, K.I. Na, et al., Increased tuberculosis burden due to demographic transition in Korea from 2001 to 2010, Tuberc. Respir. Dis. 74 (2013) 104e110. [4] Y.S. Park, S.J. Hong, Y.K. Boo, et al., The national status of tuberculosis using nationwide medical records survey of patients with tuberculosis in Korea, Tuberc. Respir. Dis. 73 (2012) 48e55. [5] E.A. Kendall, M.O. Fofana, D.W. Dowdy, Burden of transmitted multidrug resistance in epidemics of tuberculosis: a transmission modelling analysis, Lancet. Respir. Med. 3 (2015) 963e972. [6] J.S. Joh, C.H. Lee, J.E. Lee, et al., The interval between initiation of antituberculosis treatment in patients with culture-positive pulmonary tuberculosis and receipt of drug-susceptibility test results, J. Korean. Med. Sci. 22 (2007) 26e29. [7] J. Lyu, M.N. Kim, J.W. Song, et al., GenoType® MTBDRplus assay detection of drug-resistant tuberculosis in routine practice in Korea, Int. J. Tuberc. Lung. Dis. 17 (2013) 120e124. [8] D. Hillemann, S. Rüsch-Gerdes, E. Richter, Evaluation of the GenoType MTBDRplus assay for rifampin and isoniazid susceptibility testing of Mycobacterium tuberculosis strains and clinical specimens, J. Clin. Microbiol. 45 (2007) 2635e2640. [9] A. Lacoma, N. Garcia-Sierra, C. Prat, et al., GenoType MTBDRplus assay for molecular detection of rifampin and isoniazid resistance in Mycobacterium tuberculosis strains and clinical samples, J. Clin. Microbiol. 46 (2008) 3660e3667. [10] S. Hofmann-Thiel, J. van Ingen, K. Feldmann, et al., Mechanisms of heteroresistance to isoniazid and rifampin of Mycobacterium tuberculosis in Tashkent, Uzbekistan, Eur. Respir. J. 33 (2009) 368e374. [11] R. Anek-Vorapong, C. Sinthuwattanawibool, L.J. Podewils, et al., Validation of the GenoType MTBDRplus assay for detection of MDR-TB in a public health laboratory in Thailand, BMC. Infect. Dis. 10 (2010) 123. [12] Definitions and Reporting Framework for Tuberculosis e 2013 Revision. Available at: http://apps.who.int/iris/bitstream/10665/79199/1/ 9789241505345_eng.pdf (accessed on September 2016). [13] Leon Gordis, Epidemiology, fifth ed., Saunders, 2013. [14] J. Domínguez, E.C. Boettger, D. Cirillo, et al., TBNET and RESIST-TB networks. Clinical implications of molecular drug resistance testing for Mycobacterium tuberculosis: a TBNET/RESIST-TB consensus statement, Int. J. Tuberc. Lung. Dis. 20 (2016) 24e42. [15] S.D. Lawn, P. Mwaba, M. Bates, et al., Advances in tuberculosis diagnostics: the Xpert MTB/RIF assay and future prospects for a point-of-care test, Lancet. Infect. Dis. 13 (2013) 349e361. [16] T. Jagielski, Z. Bakuła, K. Roeske, et al., Detection of mutations associated with isoniazid resistance in multidrug-resistant Mycobacterium tuberculosis clinical isolates, J. Antimicrob. Chemother. 69 (2014) 2369e2375. [17] S.J. Kim, Drug-susceptibility testing in tuberculosis: methods and reliability of results, Eur. Respir. J. 25 (2005) 564e569. [18] C.R. Horsburgh Jr., C.E. Barry 3rd, C. Lange, Treatment of tuberculosis, N. Engl. J. Med. 373 (2015) 2149e2160. [19] H.M. Nijland, R. Ruslami, A.J. Suroto, et al., Rifampicin reduces plasma concentrations of moxifloxacin in patients with tuberculosis, Clin. Infect. Dis. 45 (2007) 1001e1007. [20] Word Health Organization, WHO Treatment Guidelines for Drug-resistant Tuberculosis, 2016 update. Avaiable at, http://www.who.int/tb/ MDRTBguidelines2016.pdf. accessed on September 2016. [21] M. Barnard, H. Albert, G. Coetzee, et al., Rapid molecular screening for multidrug-resistant tuberculosis in a high-volume public health laboratory in South Africa, Am. J. Respir. Crit. Care. Med. 177 (2008) 787e792. [22] M. Seifert, D. Catanzaro, A. Catanzaro, et al., Genetic mutations associated with isoniazid resistance in Mycobacterium tuberculosis: a systematic review, PLoS One 10 (2015) e0119628.
Contents lists available at ScienceDirect
Respiratory Medicine journal homepage: www.elsevier.com/locate/rmed
Analysis of discrepant results between the Genotype® MTBDRplus assay and an antimicrobial drug susceptibility test for isoniazidresistant tuberculosis Kyung-Wook Jo a, 1, Yoomi Yeo b, 1, Heungsup Sung c, Mi-Na Kim c, Tae Sun Shim a, * a b c
Division of Pulmonary and Critical Care Medicine, University of Ulsan College of Medicine, Asan Medical Center, Seoul, South Korea Department of Internal Medicine, Chunggu Sungshim Hospital, South Korea Department of Laboratory Medicine, University of Ulsan College of Medicine, Asan Medical Center, Seoul, South Korea
a r t i c l e i n f o
a b s t r a c t
Article history: Received 2 April 2016 Received in revised form 9 October 2016 Accepted 21 November 2016 Available online 22 November 2016
Background: We investigated discrepant results determined using the Genotype®MTBDRplus assay and a conventional antimicrobial drug susceptibility test (ADST) for isoniazid (INH) resistance using sequencing analysis and analyzed the clinical course of patients with discrepant results. Methods: Among 1373 MTBDRplus assays performed at our tertiary referral center in South Korea between August 2009 and December 2015, the results for 46 (3.4%) differed from those for ADST. KatG and inhA gene sequencing analysis results were available for 23 patients. ADSTs were carried out using the € wensteineJensen media. absolute concentration method with Lo Results: Results from 11 patients indicated INH susceptibility by MTBDRplus assay and INH resistance by ADST. For 5 of these patients, sequencing revealed no evidence of mutations, whereas specific mutations were detected in the remaining 6 patients. These should have been detected using the MTBDRplus assay. The other 12 patients had isolates with the opposite discrepancy, that is INH resistance by MTBDRplus assay but INH susceptibility by ADST. For 7 of these cases, sequencing results were consistent with those of the MTBDRplus assay. However, sequencing analysis did not explain the discrepancies in the remaining 5 patients. All 23 patients with discrepant results received individualized treatment regimens determined by the attending physician according to their test results and susceptibility to other drugs, such as rifampin. Good outcomes were reported for the majority. Conclusion: Discrepancies between test results for INH resistance on the MTBDRplus assay and ADST appear to be infrequent. Gene sequencing analysis is useful for identifying the cause of the discrepancy. © 2016 Elsevier Ltd. All rights reserved.
Keywords: Isoniazid Discrepancy MTBDRplus assay KatG gene sequencing inhA gene sequencing
1. Introduction Tuberculosis (TB) is a global health concern, affecting both developed and developing countries [1]. The incidence of TB remains high worldwide, with 9.6 million new cases and 1.5 million deaths in 2014 [2]. Although the incidence of TB is declining, it remains a major problem in South Korea, where a plateau of approximately 100 new cases per 100,000 persons has been reached in the past decade [3].
* Corresponding author. Division of Pulmonary and Critical Care Medicine, University of Ulsan College of Medicine, Asan Medical Center, 388-1 Pungnap-Dong, Songpa-gu, Seoul 138-736, South Korea. E-mail address: [email protected] (T.S. Shim). 1 Kyung-Wook Jo and Yoomi Yeo contributed equally to this work. http://dx.doi.org/10.1016/j.rmed.2016.11.016 0954-6111/© 2016 Elsevier Ltd. All rights reserved.
The control of TB has become increasingly difficult because of the rise of multidrug-resistant disease (MDR-TB). In some settings, up to one third of newly diagnosed cases result from MDR-TB [2]. In South Korea, the burden of MDR-TB remains relatively low, with only 4.6% patients with the disease having MDR-TB was in 2008 [4]. For optimal treatment and reduction of transmission, rapid identification of drug-resistant TB is vital [5]. Because the results of a solid media-based antimicrobial drug susceptibility test (ADST) can take up to 3 months to obtain in South Korea [6], rapid direct tests have been employed for prompt identification of patients infected with drug-resistant organisms. The most widely used rapid molecular drug resistance tests are the Xpert MTB/RIF (Cepheid, Sunnyvale, CA, USA) and Genotype®MTBDRplus (Hain Lifescience, Nehren, Germany) assays. Using the Genotype®MTBDRplus assay, we previously reported a median turnaround time for detection of
K.-W. Jo et al. / Respiratory Medicine 122 (2017) 12e17
MDR-TB of 5.3 days with acceptable sensitivity and specificity [7]. However, physicians occasionally encounter cases with a discrepancy between isoniazid (INH) resistance results from the MTBDRplus assay and conventional ADSTs [7e11]. Reports of the reasons for these discrepancies are limited, as is information on treatment and outcome. We designed this retrospective study to investigate a convenience sample of TB cases managed in our institution and to review the treatment and outcome of patients with discrepant results. 2. Patients and methods 2.1. Study subjects and data collection The study was conducted at Asan Medical Center, a 2700-bed referral hospital in Seoul, South Korea. Approximately 900 patients with TB receive diagnosis and treatment at this hospital annually. The medical records of patients with discrepant results, including sequencing analysis, were analyzed retrospectively in January 2016. The study protocol was approved by the Institutional Review Board of Asan Medical Center. 2.2. Bacteriological study and ADSTs Acid-fast bacillus (AFB) smears were examined by ZiehleNeelsen staining. AFB culture was carried out using both solid (Ogawa medium; Korean Institute of Tuberculosis, South Korea) and liquid (BACTEC 960 Mycobacterial Growth Indicator Tube; Becton Dickinson, Sparks, MD, USA) media. Positive liquid media cultures or colonies on solid medium were subjected to ZiehleNeelsen staining and polymerase chain reaction assay using Seeplex TB detection (Seegen, Seoul, Korea) to differentiate between Mycobacterium tuberculosis complex and non-tuberculous mycobacteria. When growth of M. tuberculosis was detected on culture, ADST was generally requested according to Korean national TB guidelines, which recommend sensitivity testing of the first positive M. tuberculosis isolate for all patients. However, for various reasons, ADST had not been performed in all cases. Conventional ADSTs were performed using the absolute concentration method with €wensteineJensen media at the Korean Institute of Tuberculosis, Lo South Korea's TB reference laboratory. Wells contained the following critical concentrations of anti-TB drugs: 0.2 mg/mL of INH and 40 mg/mL of rifampin (RIF). Growth exceeding that of the control wells by >1% was considered to indicate drug resistance. 2.3. MTBDRplus assay and DNA sequencing The MTBDRplus assay was performed in our center according to the manufacturer's instructions. The decision concerning whether to carry out an MTBDRplus assay was made by the attending physician. The MTBDRplus assay can be requested for a specimen with a positive AFB smear. If the smear is negative, the MTBDRplus assay can only be requested for a positive M. tuberculosis culture result. DNA sequencing was performed using a polymerase chain reaction-direct sequencing system (Cosmo Co., Seoul, South Korea). A 670 base-pair (bp) fragment of the katG gene (containing codon 315), 248 bp of the inhA promoter (including 8 and 15), and, beginning in July 2014, 472 bp of the oxyR-ahpC gene were sequenced. The following forward and reverse primers (50 e30 ) were used: KatGR CCAAGGTATCTCGCAACG, KatGF CACACTTTCGGTAAGACCCA, inhAR CCAGCACCTCGACCGGACC, inhAF ATTCGTAGGGCGTCAATACA, oxyR-ahpCR CGATGAGAGCGGTGAGCTG and oxyR-ahpCF AACGTCGACTGGCTCATATC. Primers were synthesized
13
by JieLi Bio Co (China). PCR conditions were denaturation at 94 C for 10 min, followed by 35 cycles of amplification at 94 C for 30 s, 60 C for 30 s, and 72 C for 30 s. PCR mixtures were prepared using 2 GoldStar Best MasterMix (CWBio Co., China). The PCR products were used as templates for targeted DNA sequencing. To confirm the sequence amplified PCR product, direct sequencing was performed on an ABI 373 DNA analyzer according to the protocol supplied by the manufacturer using the BigDye Terminator v3.1 Cycling Sequencing kit. Sequencing data were assembled and analyzed with BioEdit software. The mutations were determined by comparing with the H37Rv sequence of katG, inhA, and oxyR-ahpC genes from the GenBank database (http://www.ncbi.nih.gov/gene). The decision regarding whether to perform DNA sequencing analysis was made by the attending physician. As noted, oxyR-ahpC gene sequencing was only available from July 2014. 2.4. Treatment outcome Treatments were not standardized. They were individualized according to the previous treatment history and ADST results for other drugs, such as RIF. Treatment outcome categories were according to the revised 2013 WHO definitions [12]. The favorable outcome was defined as the sum of patients who were cured and those who completed treatment. In addition, we also investigated the incidence of recurrence within 1 year after treatment completion. 2.5. Statistical analyses Data are presented as means for continuous variables and percentages for discrete variables. Variables were analyzed using simple descriptive statistics. The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of the MTBDRplus assay for INH resistance were calculated as compared with conventional ADST results [13]. 3. Results 3.1. Characteristics of the study subjects and test results A total of 1551 MTBDRplus assays were carried out between August 2009 and December 2015. Of 1389 MTBDRplus assays for which corresponding ADST results were available, there were 46 (3.4%) cases with discrepant results between the two tests. Table 1 compares the rates of resistance to INH and RIF detected using MTBDRplus assay vs. ADST. Among the cases with discrepant results, 23 had both katG and inhA sequencing results and were ultimately included in the analysis (Fig. 1). The mean age of the 23 patients with discrepant INH resistance results was 52.1 years, and males were predominant (78.3%, 18/23). All patients were diagnosed with culture-confirmed pulmonary TB. The isolates from 11 patients' isolates were reported to be INH susceptible using the
Table 1 Summary of isoniazid and rifampicin resistance by the MTBDRplus assay and antimicrobial-susceptibility test in 1373 isolates. Antimicrobial susceptibility test MTBDRplus assay Isoniazid Resistant Susceptible Rifampicin Resistant Susceptible
Resistant n (%)
Susceptible n (%)
170 (12.4%) 23 (1.7%)
23 (1.7%) 1157 (84.3%)
115 (8.4%) 3 (0.2%)
15 (1.1%) 1256 (91.5%)
14
K.-W. Jo et al. / Respiratory Medicine 122 (2017) 12e17
A total of 1,551 MTBDRplus assays were performed between August 2009 and December 2015 ADST not requested (n = 107) NTM growth (n = 37) No growth on AFB culture (n = 14) No amplification (n = 4)
Both MTBDRplus assays and ADST results were available for RIF susceptibility (n = 1,389)
INH-resistant on MTBDRplus assay (n = 193)
INH-resistant on ADST (n = 23)
INH-susceptible on ADST (n = 170)
INH-susceptible on MTBDRplus assay (n = 1,196)
INH-resistant on ADST (n = 23)
Sequencing results for both KatG and inhA were not available (n = 12)
Sequencing results for both KatG and inhA were not available (n = 11)
INH-susceptible on MTBDRplus assay and INH-resistant on ADST (n = 12)
INH-susceptible on ADST (n = 1,173)
INH-susceptible on MTBDRplus assay and INH-resistant on ADST (n = 11)
Fig. 1. Study flow chart.; Abbreviations: ADST, antimicrobial drug susceptibility test; NTM, non-tuberculous mycobacteria; AFB, acid-fast bacillus; INH, isoniazid.
MTBDRplus assay and INH resistant by ADST. The isolates from the other 12 patients had the opposite discrepancy; INH resistant by MTBDRplus assay and INH susceptible by ADST. 3.2. Performance of the MTBDRplus assay in detection of INH resistance The performance of the MTBDRplus assay was analyzed in comparison with the results of conventional ADST in the 1389 cases for which both tests were performed. The sensitivity, specificity, PPV, and NPV of the MTBDRplus assay for detecting INH resistance were 88.1% (95% confidence interval [CI], 82.4e92.1), 98.1% (95% CI 97.0e98.7), 88.1% (95% CI 82.4e92.1), and 98.1% (95% CI 97.0e98.7), respectively, compared to the results of conventional ADSTs. 3.3. Discrepancy analysis: INH susceptible by MTBDRplus assay and INH resistant by ADST Table 2 shows the results of sequencing analysis for the 11 patients found to be INH susceptible using the MTBDRplus assay and INH resistant by ADST. Sequencing analysis of the KatG, inhA and oxyR-ahpC genes for patients 1e5 revealed no evidence of mutations, whereas sequencing analysis for patients 6e11 revealed specific mutations that should have been detected by the MTBDRplus assay. The M. tuberculosis RIF-susceptibility results, treatment regimens and durations, and treatment outcomes for these patients are summarized in Table 3. Patients 1e6 and 11 received the usual treatment regimen for INH-resistant TB, and the remaining 4 were treated using the MDR-TB regimen because their MTBDRplus assay or ADST indicated RIF resistance. All patients evaluated had favorable outcomes.
3.4. Discrepancy analysis: INH resistant by MTBDRplus assay and INH susceptible by ADST Table 4 shows the results of sequencing analyses and the MTBDRplus assay (wild-type and mutated band pattern) for the 12 patients found to be INH resistant using the MTBDRplus assay and INH susceptible by ADST. Sequencing analysis of the KatG and inhA genes for 7 patients (patients 12e18) were in perfect agreement with the results of the MTBDRplus assay. In 4 patients (19e21 and 23), sequencing analysis revealed no evidence of mutations. The sequencing analysis results for patient 22 were inconsistent with the results of the MTBDRplus assay. The M. tuberculosis RIF-susceptibility test results as well as treatment regimen, duration and outcome, for these 12 patients are shown in Table 5. All patients were treated with reference to susceptibility to other drugs, such as RIF as decided by the attending physician. The outcomes, where evaluated, were favorable. 4. Discussion Recently, two molecular diagnostic tests, the Xpert MTB MTB/ RIF assay and the Genotype® MTBDRplus assay, were developed and endorsed by the WHO for the rapid detection of drug-resistant TB [14]. Unlike the Xpert MTB/RIF assay, which only detects genetic mutations associated with RIF resistance [15], the MTBDRplus assay has the advantage of detecting both RIF and INH resistance. In this study, we analyzed the INH-susceptibility results from 1373 MTBDRplus assays compared with conventional ADSTs. In addition, we investigated the cause of identified discrepancies by sequencing the katG, inhA and oxyR-ahpC genes and analyzed the treatment outcomes of patients with discrepant INH-susceptibility results. The most important findings of this study are that the diagnostic
K.-W. Jo et al. / Respiratory Medicine 122 (2017) 12e17
15
Table 2 Sequencing results from 11 patient samples reported as isoniazid susceptible by MTBDRplus assay and isoniazid resistant by an antimicrobial-susceptibility test. Patient no.
1 2 3 4 5 6 7 8 9 10 11
Sputum AFB smear grade
MTBDRplus assay results Wild-type loss
Mutation detected
katG mutation
inhA mutation
oxyR-aphC mutation
3þ Negative* 3þ Negative* 1þ 1þ Negative* Negative* Negative* 4þ Negative*
Not Not Not Not Not Not Not Not Not Not Not
Not Not Not Not Not Not Not Not Not Not Not
WT WT WT WT WT AGC315ACC WT WT WT WT WT
WT WT WT WT WT WT 8 T deletion 15 C deletion 15 C deletion 15 C / A 15 C / T
WT WT ND ND ND ND ND ND ND ND ND
detected detected detected detected detected detected detected detected detected detected detected
Sequencing results
detected detected detected detected detected detected detected detected detected detected detected
Abbreviations:AFB, Acid-fast bacillus; WT, wild type; ND, not done. * In cases with negative AFB smears, cultured isolates were used for MTBDRplus assay.
Table 3 Treatment regimen, rifampicin susceptibility, and treatment outcome of 11 patients with tuberculosis reported as isoniazid susceptible by MTBDRplus assay and isoniazid resistant by an antimicrobial-susceptibility test. Patient Age Gender RIF-susceptibility: no. MTBDRplus assay
Initial RIFsusceptibility: regimen ADST
Duration of first Subsequent Duration of subsequent Treatment regimen (months) regimen regimen (months) outcome
Recurrence within 1 year
1
71
Male
Susceptible
Susceptible
2
Cure
No recurrence
2
58
Male
Resistant*
Susceptible*
Cure
3
61
Female Susceptible
Susceptible
Less than 1 year after treatment completion No recurrence
4
40
Female Susceptible
Susceptible
5
77
Male
Susceptible
Susceptible
6
22
Female Susceptible
Susceptible
7
52
Male
Resistant*
Susceptible*
8
71
Male
Resistant
Resistant
9
30
Male
Resistant
Resistant
10
52
Male
Resistant
Resistant
11
77
Male
Susceptible
Susceptible
INH, RIF, EMB, PZA INH, HD-RIF, PZA INH, RIF, EMB, PZA INH, RIF, LVFX INH, RIF, EMB, PZA INH, RIF, EMB, PZA MXF, KM, PTH, CS, PAS LZD, MXF, PTH, CS LZD, MXF, KM, AMK, CLR LVFX, KM, PTH, PAS, CS INH, RIF
1.5 2.5 2
2.5
MXF, RIF, PZA MXF, HDRIF, EMB LVFX, RIF, PZA LVFX, RIF, CS
6
RIF, EMB, PZA
3.5
3.5
Cure
13
Completion Lost to follow-up after treatment completion Not N/A evaluated Cure No recurrence
14
Completion No recurrence
14
Completion No recurrence
4
MXF, KM, AMC, CLR
18
Cure
No recurrence
12
LVFX, PAS, AMC LVFX, RIF
11
Cure
No recurrence
8
Cure
Died 8 months after treatment completion
1.5
Abbreviations: RIF, rifampicin; INH, isoniazid; EMB, ethambutol; PZA, pyrazinamide; MXF, moxifloxacin; HD-RIF, high-dose rifampicin; LVFX, levofloxacin; CS, cycloserine; KM, kanamycin; PTH, prothionamide; PAS, p-aminoslicylic acid; LZD, linezolid; AMK, amikacin; CLR, clarithromycin; AMC, amoxicillin/clavulanate. *These strains had disputed rpoB mutations (CTG511CCG and CTG533CCG, respectively).
accuracy of the MTBDRplus assay for detection of INH resistance is satisfactory and that sequencing analysis is helpful in assessing the causes of discrepant INH-susceptibility results. There have been a limited number of reports using sequencing analysis to investigate discrepancies between INH-susceptibility results from the MTBDRplus assay and conventional ADSTs [7e9,11]. Previously, we showed that discrepant results were recorded from 10/428 (2.3%) specimens using these two tests [7]. In that study, we found that the results of the MTBDRplus assay were compatible with those obtained by sequencing analysis in all six cases for which data were available. Lacoma et al. evaluated the performance of the MTBDRplus assay using 62 clinical strains from 28 patients in Spain [9]. They found that, among 48 strains indicated to be INH resistant by ADST, as many as 13 were reported to be INH susceptible on MTBDRplus. Sequencing analysis revealed several possible reasons for these discrepancies. Other studies have also reported discrepancies in test results for INH resistance and
performed sequencing analysis for small numbers of cases [8,11]. The present study represents the largest number of discrepant INH results analyzed by gene sequencing, although sequencing was only performed in half of the cases. There were various reasons underlying the discrepancies, and gene sequencing was able to identify these in the majority of cases. Of the 11 samples reported as INH-sensitive on MTBDRplus assay but resistant by ADST (Table 2), no mutations were found in five. The discrepant results for these cases may have been because of (1) rare gene mutations (other than katG, inhA and oxyR-ahpC) causing INH resistance, such as those in kasA or mshA [16]; (2) mutations of katG, inhA or oxyR-ahpC not covered by the MTBDRplus assay probes and the outside the regions sequenced; or (3) inaccurate ADST results [17]. By contrast, the responsible mutation should have been identified by the MTBDRplus assay in the remaining six samples. Technical errors may explain the discrepancies in these cases.
16
K.-W. Jo et al. / Respiratory Medicine 122 (2017) 12e17
Table 4 Sequencing results from 12 patient samples testing reported as isoniazid-resistant using theby MTBDRplus assay and isoniazid-susceptible by an antimicrobial-drug susceptibility test. Patient no.
Sputum AFB smear grade
*
12 13 14 15 16 17 18 19 20 21 22 23
Negative 1þ 1þ Negative* 1þ Negative* Negative* 2þ Negative* Negative* 1þ Negative*
MTBDRplus assay results
Sequencing results
Wild-type loss
Mutation detected
katG mutation
inhA mutation
oxyR-ahpC mutation
katG katG inhA WT1 inhA WT1 inhA WT1 inhA WT2 Not detected Not detected Not detected inhA WT1 inhA WT1 inhA WT1
katG MUT1 katG MUT2 inhA MUT1 inhA MUT1 inhA MUT1 inhA MUT3A inhA MUT3B katG MUT1 katG MUT1 inhA MUT1 inhA MUT2 Not detected
AGC315ACC AGC315ACA WT WT WT WT WT WT WT WT WT WT
WT WT 15 C / T 15 C / T 15 C / T 8 T / C 8 T / A WT WT WT 15 C / T WT
ND WT WT WT ND ND WT WT WT ND WT ND
Abbreviations: AFB, Acid-fast bacillus; WT, wild type; ND, not done. In cases with negative AFB smears, cultured isolates were used for MTBDRplus assay.
*
Table 5 Treatment regimen, rifampicin susceptibility, and treatment outcome for 12 patients with tuberculosis reported as isoniazid resistant by MTBDRplus assay and isoniazid susceptible by an antimicrobial-susceptibility test. Patient Age Gender RIF-susceptibility: no. MTBDRplus assay
Initial RIFsusceptibility: regimen ADST
Duration of first regimen (months)
Subsequent regimen
Duration of subsequent Treatment regimen (months) outcome
1 year recurrence
12
40
Male
Resistant
Susceptible
INH, RIF, EMB, PZA
6
18
Cure
No recurrence
13
76
Male
Susceptible
Susceptible
2
6
Cure
No recurrence
14
44
Male
Susceptible
Susceptible
15
48
Male
Susceptible
Susceptible
16
49
Female Resistant
Resistant
17
56
Male
Susceptible
Susceptible
18
71
Male
Susceptible
Susceptible
19
24
Male
Susceptible
Susceptible
20
44
Male
Susceptible
Susceptible
21
69
Female Susceptible
Susceptible
22
45
Male
Susceptible
Susceptible
RIF, EMB, PZA LVFX, RIF, EMB, SM MXF, RIF, EMB MXF, KM, PZA, PTH, CS RIF, EMB, PZA MXF, EMB, CS INH, RIF, EMB, PZA RIF, EMB, PZA MXF, RIF, EMB LVFX, RIF
KM, PAS, PTH, MXF, CS INH, RIF
23
48
Male
Susceptible
Susceptible
INH, RIF, EMB, PZA
14
Completion No recurrence
9
Cure
4 2
INH, MXF, PZA, CS INH, RIF
13
Cure
Less than 1 year after treatment completion No recurrence
7
Cure
No recurrence
Cure
Less than 1 year after treatment completion No recurrence
6
Cure
9
Not evaluated Cure
9 2
INH, RIF
4
N/A No recurrence
Completion Less than 1 year after treatment completion Cure No recurrence
Abbreviations: RIF, rifampicin; INH, isoniazid; EMB, ethambutol; PZA, pyrazinamide; KM, kanamycin; PAS, p-aminoslicylic acid; PTH, prothionamide; MXF, moxifloxacin; CS, cycloserine; LVFX, levofloxacin; SM, streptomycin.
The sequencing analysis also revealed a variety of reasons for the discrepancies identified in 12 samples reported as INH resistant by MTBDRplus assay but sensitive on ADST. For seven of these, the sequencing analysis detected exactly the same mutation as that detected by the MTBDRplus assay. The discrepant results in these cases could be related to the reliability of ADSTs using the absolute concentration method, because the power of this method to discriminate between INH-susceptible and INH-resistant strains is imperfect and can occasionally lead to mislabeling of INH-resistant strains as INH susceptible and vice versa [17]. There may be a number of reasons for the discrepancies observed in the remaining five samples, including the following possibilities: (1) heteroresistance (patients 19e21) [10], (2) technical error in operation of the MTBDRplus assay, (3) co-infection (for patient 22), or (4) mutation of the inhA gene in a region outside that sequenced (for
patient 23). The interpretation of discrepant results and the choice of treatment were solely the responsibility of the attending physician. Later-generation fluoroquinolones (levofloxacin or moxifloxacin) were widely used in our study population. In cases resistant to INH, but not to RIF, a standard 6-month regimen in which INH is replaced by a later-generation fluoroquinolone is likely to lead to similar treatment outcomes [18]. A number of patients were treated with other drugs, including ethambutol or cycloserine, in addition to a combination of a later-generation fluoroquinolone and RIF, because of concerns regarding drug interactions between the latter two [19]. In patients whose MTBDRplus assays and ADST revealed RIF-monoresistance, MDR-TB regimens were usually prescribed for initial treatment (patients 8e10, Table 3) irrespective of the results of INH resistance tests. This is in accordance with the recent WHO
K.-W. Jo et al. / Respiratory Medicine 122 (2017) 12e17
recommendation to prescribe an MDR-TB regimen for RIFmonoresistant TB [20]. Previously, we reported that the sensitivity, specificity, positive predictive value, and negative predictive value of the MTBDRplus assay for detecting INH resistance were 93.8%, 98.3%, 92.7%, and 98.6%, respectively. Higher sensitivity and/or positive predictive values were reported in South Africa [21] and Thailand [11], whereas, one study in Spain [9] reported a much lower sensitivity (73%) for the detection of INH resistance. Hence, the performance of the MTBDRplus assay for detection of INH resistance appears to be dependent on regional differences in the frequency of specific mutations causing the resistance phenotype [22]. The current study has several limitations. Most significantly, it was conducted at a single referral center and included a relatively small number of cases. In addition, because requesting an MTBDRplus assay was at the sole discretion of the attending physician, it was not performed routinely for all patients diagnosed with TB in our center. Moreover, because the attending physician also decided whether or not to conduct sequencing analysis, gene sequencing was not performed in half of the cases with discrepant results. Sequencing analysis of the oxyR-ahpC gene was only in use later in the study period and so was conducted in only about a third of the cases. However, our results bring into question the value of oxyR-ahpC gene sequencing, because no mutations of this gene were found in the present study. Unlike other comparable studies, we did not use the Mycobacterial Growth Indicator Tube for Antimycobacterial-Susceptibility Testing or the proportion method to detect INH resistance. Because we did not carry out genotyping, we were unable to determine whether mixed infections were present in some cases where this was potentially a factor. Finally, isolates yielding discrepant results were not re-tested by all methods to ascertain the reproducibility of the initial testing result. In conclusion, discrepancies between the results of tests for INH resistance with the MTBDRplus assay or ADST using the absolute concentration method appear to be infrequent. As gene sequencing was useful to identify the cause of discrepancies in many cases, we propose that it should be performed in all cases with discrepant results. Despite a variety of possible reasons for the discrepancies, patients generally achieved good outcomes with individualized treatment regimens. Funding This work was supported by the Asan Institute for Life Sciences (grant number: 2015-617), Asan Medical Center, Seoul, Korea. Conflicts of interest None declared. Acknowledgments The authors would like to thank the Asan Institute for Life Sciences, Asan Medical Center, Seoul, Korea for financial support for
17
this project. References [1] Y.J. Ryu, Diagnosis of pulmonary tuberculosis: recent advances and diagnostic algorithms, Tuberc. Respir. Dis. 78 (2015) 64e71. [2] Word Health Organization, Global Tuberculosis Report 2015, World Health Organization, Geneva, 2015. Avaiable at, http://apps.who.int/iris/bitstream/ 10665/191102/1/9789241565059_eng.pdf. accessed on September 2016. [3] Y.K. Park, Y.S. Park, K.I. Na, et al., Increased tuberculosis burden due to demographic transition in Korea from 2001 to 2010, Tuberc. Respir. Dis. 74 (2013) 104e110. [4] Y.S. Park, S.J. Hong, Y.K. Boo, et al., The national status of tuberculosis using nationwide medical records survey of patients with tuberculosis in Korea, Tuberc. Respir. Dis. 73 (2012) 48e55. [5] E.A. Kendall, M.O. Fofana, D.W. Dowdy, Burden of transmitted multidrug resistance in epidemics of tuberculosis: a transmission modelling analysis, Lancet. Respir. Med. 3 (2015) 963e972. [6] J.S. Joh, C.H. Lee, J.E. Lee, et al., The interval between initiation of antituberculosis treatment in patients with culture-positive pulmonary tuberculosis and receipt of drug-susceptibility test results, J. Korean. Med. Sci. 22 (2007) 26e29. [7] J. Lyu, M.N. Kim, J.W. Song, et al., GenoType® MTBDRplus assay detection of drug-resistant tuberculosis in routine practice in Korea, Int. J. Tuberc. Lung. Dis. 17 (2013) 120e124. [8] D. Hillemann, S. Rüsch-Gerdes, E. Richter, Evaluation of the GenoType MTBDRplus assay for rifampin and isoniazid susceptibility testing of Mycobacterium tuberculosis strains and clinical specimens, J. Clin. Microbiol. 45 (2007) 2635e2640. [9] A. Lacoma, N. Garcia-Sierra, C. Prat, et al., GenoType MTBDRplus assay for molecular detection of rifampin and isoniazid resistance in Mycobacterium tuberculosis strains and clinical samples, J. Clin. Microbiol. 46 (2008) 3660e3667. [10] S. Hofmann-Thiel, J. van Ingen, K. Feldmann, et al., Mechanisms of heteroresistance to isoniazid and rifampin of Mycobacterium tuberculosis in Tashkent, Uzbekistan, Eur. Respir. J. 33 (2009) 368e374. [11] R. Anek-Vorapong, C. Sinthuwattanawibool, L.J. Podewils, et al., Validation of the GenoType MTBDRplus assay for detection of MDR-TB in a public health laboratory in Thailand, BMC. Infect. Dis. 10 (2010) 123. [12] Definitions and Reporting Framework for Tuberculosis e 2013 Revision. Available at: http://apps.who.int/iris/bitstream/10665/79199/1/ 9789241505345_eng.pdf (accessed on September 2016). [13] Leon Gordis, Epidemiology, fifth ed., Saunders, 2013. [14] J. Domínguez, E.C. Boettger, D. Cirillo, et al., TBNET and RESIST-TB networks. Clinical implications of molecular drug resistance testing for Mycobacterium tuberculosis: a TBNET/RESIST-TB consensus statement, Int. J. Tuberc. Lung. Dis. 20 (2016) 24e42. [15] S.D. Lawn, P. Mwaba, M. Bates, et al., Advances in tuberculosis diagnostics: the Xpert MTB/RIF assay and future prospects for a point-of-care test, Lancet. Infect. Dis. 13 (2013) 349e361. [16] T. Jagielski, Z. Bakuła, K. Roeske, et al., Detection of mutations associated with isoniazid resistance in multidrug-resistant Mycobacterium tuberculosis clinical isolates, J. Antimicrob. Chemother. 69 (2014) 2369e2375. [17] S.J. Kim, Drug-susceptibility testing in tuberculosis: methods and reliability of results, Eur. Respir. J. 25 (2005) 564e569. [18] C.R. Horsburgh Jr., C.E. Barry 3rd, C. Lange, Treatment of tuberculosis, N. Engl. J. Med. 373 (2015) 2149e2160. [19] H.M. Nijland, R. Ruslami, A.J. Suroto, et al., Rifampicin reduces plasma concentrations of moxifloxacin in patients with tuberculosis, Clin. Infect. Dis. 45 (2007) 1001e1007. [20] Word Health Organization, WHO Treatment Guidelines for Drug-resistant Tuberculosis, 2016 update. Avaiable at, http://www.who.int/tb/ MDRTBguidelines2016.pdf. accessed on September 2016. [21] M. Barnard, H. Albert, G. Coetzee, et al., Rapid molecular screening for multidrug-resistant tuberculosis in a high-volume public health laboratory in South Africa, Am. J. Respir. Crit. Care. Med. 177 (2008) 787e792. [22] M. Seifert, D. Catanzaro, A. Catanzaro, et al., Genetic mutations associated with isoniazid resistance in Mycobacterium tuberculosis: a systematic review, PLoS One 10 (2015) e0119628.