Performance of Genotype MTBDRsl V2.0 over the Genotype MTBDRsl V1 for detection of second line drug resistance: An Indian perspective

Genotype MTBDRsl Version 1 (V1.0) was recommended as an initial test for rapid detection of pre-extensively drug resistant (pre-XDR) and extensively drug resistant tuberculosis (XDR-TB). However, in recent years a number of novel mutations are identified that confer resistance. Thus, Genotype MTBDRsl V2.0 was endorsed by WHO. Though, Genotype MTBDRsl V2.0 has been rolled out in national TB programme in 2018, there is dearth of data from India on its performance for second line drug susceptibility testing (DST). For this, performance of new version was evaluated on 113 MDR-TB isolates. The results showed that 39 (34.5%) of these isolates were resistant to FQ and 7 (6.2%) were XDR by Genotype MTBDRsl V2.0. Amongst the FQ resistant isolates most prevalent mutation was ΔWT3-D94G (17; 38.6%) and N538D (12; 85.7%). Among the AG/CP and KAN resistant isolates most common mutation in the rrs region was ΔWT1-A1401G (5; 71.4%) and C-14T (2; 28.5%) in eis gene. Second line Bactec MGIT-960 detected 40 (35.4%) isolates as resistant to FQ and 6 (5.3%) as XDR isolates, whereas Genotype MTBDRsl V1.0 also detected 39 (34.5%) as resistant to FQ but missed 2 isolates in correctly identifying as XDR (5; 4.4%). Thus, concordance of second line Bactec MGIT-960 with Genotype MTBDRsl V2.0 was similar (100%) for FQ detection but it has improvised the diagnostic sensitivity for correctly identifying XDR isolates. Nevertheless, the cost of Genotype MTBDRsl V2.0 remains an issue for screening of second line drug (SLDs) resistance from countries with high burden of MDR-TB.


Introduction
Increasing resistance towards second line drugs (SLD's) is a major global threat leading to pre-XDR and XDR-TB which has high mortality rates due to limited treatment options and lower rates of treatment success [1,2]. Timely diagnosis and treatment remains an utmost challenge in order to stop circulating XDR-TB strains and further amplification of resistance [3][4][5]. Drug susceptibility testing (DST) for SLD's is gold standard for detection of drug resistance, a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 but varying drug potency and critical concentrations of drugs leads to poor reproducibility of DST results [6]. Among molecular approaches, line probe assays have been developed for detection of second line drug resistance like Genotype MTBDRslV1.0 (Hain Lifescience GmbH Germany) which rapidly detects genotypic resistance to FQ, AG/CP and Ethambutol (EMB), within a shorter turn-around time (TAT) of 48-72 h and helps make early diagnosis of pre-XDR-TB and XDR-TB possible. Previous reports have shown very good sensitivity of this test for detection of pre-XDR and XDR TB and average sensitivity for detection of AG/CP resistance, thus in 2016 this test was recommended to be used as an initial diagnostic test along with phenotypic culture-based DST in confirmed MDR-TB cases [7][8][9][10]. With expansion of next generation sequencing methods, genes associated with resistance towards SLD's were investigated and role of gyrB gene for FQ resistance and eis gene for KAN resistance were determined [11]. Drug resistance conferring mutations in gyrB genes has been reported to occur in codons 500 and 538 and often co-occur with mutations of gyrA gene region [12]. Mutations in promotor region of eis gene for conferring low-level KAN mutations which is known to increase sensitivity of phenotypic KAN resistance by 9% has also been reported [13]. However, frequency of these mutations among pre-XDR and XDR-TB cases particularly from India remains unknown.
As probes for mutations associated with gyrB and eis genes were not included in Genotype MTBDRsl V1.0 resulting in lower sensitivity of the test. In order to overcome these issues, advanced version of Genotype MTBDRsl V2.0 was introduced to incorporate these mutations in eis promoter gene region (position -10 to -14) and gyrB region (codon 536 to 541) [8].
Though quite a few studies have reported the efficiency of Genotype MTBDRsl V2.0 for screening of pre-XDR and XDR-TB, however, more evidences are required to endorse the report especially from high TB burden countries. In 2018, Revised National Tuberculosis Control program in India has already rolled out the test for rapid detection of pre-XDR and XDR-TB however, there are scarcity of reports related to the performance of the assay. This study was thus planned in order to compare the usefulness of line probe assay (Genotype MTBDRsl V2.0) for detection of additional mutations in gyrB and eis genes over Genotype MTBDRsl V1.0 in comparison to Bactec MGIT-960 DST from India.

Study setting
A retrospective study on the mycobacterial isolates was carried out in a laboratory certified for routine diagnostics in the Department of Microbiology, All India Institute of Medical Sciences, Bhopal, India. The laboratory repository characterizes stocks and maintains all the mycobacterial isolates, obtained from routine diagnostic services for patient care. This study was performed on 113 MDR-TB isolates which were already tested for Genotype MTBDRsl V1.0 and Bactec MGIT-960 as published previously [7]. The ethical clearance for evaluating the performance of Genotype MTBDRsl V2.0 on these isolates was obtained from the institutional ethics committee of the All India Institute of Medical Sciences, Bhopal (reference # IHEC-LOP/ 2018/EF0102).

Second line drug susceptibility (DST) testing using Bactec MGIT-960 system
To obtain a pure growth of MTB, 200 μL of culture suspension from the Bactec MGIT-960 tube having microscopically confirmed growth was sub-cultured on Lowenstein Jensen (LJ) medium slants and incubated at 37˚C for 21-28 days. After visible growth, a single colony was picked from LJ medium using sterile inoculating loop and inoculated in the MGIT (Mycobacteria growth indicator tube). The tube was further incubated in the Bactec MGIT-960 system until flagged positive and this growth was used for second line MGIT DST and DNA extraction for MTBDRsl V1.0 and MTBDRsl V2.0 [7,9].
The stock solution of second line drugs, OFX, AMK, KAN, and CAP (procured from St. Louis MO, USA) were dissolved, sterilized and stored at −80˚C in aliquots for further use as mentioned earlier [7,14]. The second line DST was performed using Bactec MGIT-960 as per the manufacturer's instructions. AST tubes were then set in the carrier rack, loaded in the Bactec MGIT-960 system and monitored repeatedly by BD Epicenter, as mentioned earlier [15].
Line probe assay (Genotype MTBDRsl V1.0 and V2.0) DNA extraction. DNA extraction was done as per manufacturers' instruction. Briefly, a loopful of culture isolate was suspended in 300 μl of sterile distilled water and centrifuged at 10000 x g for 15 min to pellet out sediment. Supernatant was discarded and sediment was resuspended in 100-300 μl of sterile distilled water. This was further processed with heat lysis at 95˚C for 10 min followed by sonication and used for amplification for Genotype MTBDRslV1.0 as well as for V2.0 [16].
Hybridization and detection of the amplified product. Hybridization and detection of amplified product was performed in an automated TwinCubator as per manufacturer's instructions. Briefly, denaturation of the amplification products was done by mixing 20 μL of the amplified products with 20μl of denaturing reagent (provided in the kit) for 5 min in separate troughs of a plastic well (provided in the kit). The step was followed by addition of 1 ml of pre-warmed hybridization buffer and the procedure was performed at 45˚C for 30 min followed by two steps of washing. Streptavidin conjugated with alkaline phosphatase and substrate buffer was added for colorimetric detection of hybridized amplicons. Washing was properly performed and all strips that were air dried. DNA obtained from H37RV and negative control was also tested in order to check for cross contamination during the test. Validity of the result was considered only when bands were obtained on MTB complex control (TUB) conjugate controls (CC) and amplification controls (AC) in conjunction with the target genes locus controls [7,18,19].
For Quality Control, DNA extraction was performed in the BSL-2 laboratory, master mix preparation was performed in second room with UV chamber, while PCR amplification and hybridization were performed in another room in order to avoid cross contamination.
Moreover, each batch of line probe was tested on known pan-susceptible strain H37Rv. A negative control was also added with each batch to ensure that no cross contamination occurs.

Discussion
TB being a stern medical threat that affects around 10 million people worldwide [20]. In addition, the steady increase in the stretch of pre-XDR/XDR-TB over the past decade has raised

PLOS ONE
concerns and created an alarming situation for development of reliable, accurate and rapid diagnostic tests for detection of second line drug resistance [21].
With advancement of sequencing approaches, more data has been reported to identify the mutations in the genes associated with second line drug resistance [22]. Further, advances in the nucleic acid amplification and hybridization methods, molecular tests for rapid detection of MDR-TB have been developed which have much reduced turn-around time (TAT).

PLOS ONE
However, for detection of second line drug resistance, not enough diagnostic tests are available except few in-house diagnostic methods and reverse hybridization assays such as Genotype MTBDRsl [16,18,23,24]. Yet, the second line BactecMGIT-960 remains gold standard [6].
In order to enhance the sensitivity of test a new version known as Genotype MTBDRsl V2.0 was introduced recently which incorporates mutations (Asn538Asp, Glu540Val) in gyrB and mutations in eis genes (G−37T, C−14T, C−12T, G−10A, C−2A) to detect low-level KAN resistance [18]. Several studies have reported performance of Genotype MTBDRsl V2.0 with enhances sensitivity ranging from 83.6-100% and 94.4-100% and specificity of 89.2-100% and 91.4-98.5% for FQ and AG/CP drugs, respectively [17,18,36]. Only one study from India has been reported on the efficacy of Genotype MTBDRsl V2.0 [36]. Hence, more studies were required in order to evaluate the performance of the test and prevalence of mutations especially in gyrB and eis genes.
This study was thus performed to evaluate the improved performance of Genotype MTBDRsl V2.0 over the Genotype MTBDRsl V1.0 while considering the phenotypic second line Bactec MGIT-960 as standard. In our study, the Genotype MTBDRsl V2.0 detected additionally 2 (2.94%) FQ mono-resistant isolates over Genotype MTBDRsl V1.0 due to presence of mutations in gyrB gene region which were also detected phenotypically by second line Bactec MGIT-960. Moreover, Genotype MTBDRsl V2.0 efficiently detected mutation in eis gene region in 2 (28.6%) isolates of which second line Bactec MGIT -960 detected 1 (14.3%) isolate as KAN resistant, and none of these isolates was detected by Genotype MTBDRsl V1.0. Thus overall, Genotype MTBDRsl V2.0 showed increased sensitivity for detection of FQ and XDR-TB isolates over Genotype MTBDRsl V1.0 and even the phenotypic second line Bactec MGIT-960.
In India the first version (V1.0) which was initially available before the launch of Genotype MTBDRsl V2.0 cost ranging from (1389 to 1527) US dollar each kit for 96 test, (14.4 to 15.9) US dollar each test while the cost of new version (V2.0) is approximately (1666 to 1944) US dollar each kit for 96 test (17.3 to 20.2) US dollar each test. The cost-benefit analyses for performance of Genotype MTBDRsl V2.0 for detection of second line drug resistance where higher rates of FQ resistance and XDR-TB is suspected should essentially be carried out whether the cost of millions of dollars, is justified or not. Because of high cost, even we wanted to test all 359 MDR isolate we had, we could test only 113 with version 2 while all were tested by version 1.0. Therefore, the national TB control programs and the manufactures must sit together and with the help of WHO they must work for reducing the prices of the kit not only for the FIND TB sites but for overall government as well as private hospitals who have setup for LPA testing services in India.
We thus conclude that Genotype MTBDRsl V2.0 should be used for rapid detection of second line drug resistance for FQ, AMK and KAN in order to initiate treatment among the patients. However, high cost per test of Genotype MTBDRsl V2.0 remains an issue. Moreover, WGS studies should be performed on large data sets to scrutinize more mutations conferring or associated with second line drug resistance in order to develop new molecular tests for rapid and accurate detection of XDR-TB.