Home
Programs
Center for Laboratory Sciences
Infectious Diseases Laboratory
Microbial Diseases Laboratory
TB Whole Genome Sequencing (WGS) for Drug Susceptibility Testing (DST) and Genotyping FAQs

MICROBIAL DISEASES LABORATORY BRANCH

FAQs for Sequencing-based TB Drug Susceptibility Testing [Whole Genome Sequencing (WGS) and targeted Next Generation Sequencing (tNGS)] and WGS Genotyping

Q: I do not see the WGS-DST or tNGS-DST assay on the Lab Web Portal. How do I order either of those test options?

A: When selecting the “Sequencing-based DST” option on Lab Web Portal, either the WGS or tNGS assay for drug susceptibility prediction will be automatically assigned depending on the submitted material indicated by the submitter on test requisition form:

If a pure culture is submitted, WGS-DST will automatically be ordered.
If a sediment, sterile body fluid, processed tissue, or mixed/non-viable culture are submitted, tNGS-DST will automatically be ordered.

The submitter does not need to select tNGS or WGS test specifically.

For the PZA-specific testing see the MDL memo as of 04/26/2024 and corresponding FAQs.

Q: Can I still order pyrosequencing for cultures?

A: The default sequencing-based DST test for TB cultures is WGS. Starting June 10th, 2024, the tNGS assay is the primary molecular DST method for sediments, sterile body fluids, processed tissues, or mixed/non-viable cultures; pyrosequencing (PSQ) will be provided through June 9th, 2024 for these sample types. Please contact the laboratory for consultation if PSQ is deemed necessary for cultures. While the turnaround time for sequencing-based DST is longer than for PSQ, this method provides much more comprehensive molecular resistance detection than PSQ; for cultures, WGS additionally allows determination of genotype.

Q: Can I still order pyrosequencing for cultures?

Q: What is the difference between the “Sequencing-based DST” and “WGS Genotyping” test options on Lab Web Portal? Which one should I pick for my isolate if I would like to get WGS-based resistance prediction? Which one should I select if I would like to test for potential cross-contamination between two samples.

A: Order “Sequencing-based DST” if you would like to obtain WGS-based DST results for your isolate.

WGS genotyping implies WGS for the purposes of surveillance (including upload to CDC for national-level surveillance) and outbreak detection. WGS genotyping could be also used to investigate potential cross-contamination or mix-up between samples. If an isolate only needs WGS genotyping for contamination investigation, do not select “sequencing-based DST” and only select WGS genotyping.

Note that on the Lab Web Portal, when ordering “Sequencing-based DST,” WGS genotyping is pre-selected when the material type “pure culture” is chosen; however, the submitter can opt out of WGS genotyping, which would mean that MDL will not submit WGS sequences to the CDC, and the submitting lab will be responsible for ensuring that WGS is performed and data is submitted to the CDC. We recommend leaving the WGS genotyping option selected when ordering WGS-DST unless an isolate from the patient has already been sent for national WGS surveillance genotyping elsewhere.

Q: Will WGS-DST be performed on all isolates submitted for phenotypic DST?

A: Yes. The "Sequencing-based DST” test option is ordered automatically through the Lab Web Portal when phenotypic DST is selected. For submitted cultures, this means that we will perform both pDST and WGS-DST and will issue results for the corresponding tests as separate reports. This will not delay phenotypic testing since these tests are set up and reported independently, without reflex. The exception is for PZA phenotypic susceptibility which would be performed only after applicable WGS results are obtained.

Q: What is the difference between MDL TB WGS-DST and CDC's MDDR service for molecular DST?

A: The main differences are:

Testing method/Acceptable sample types: The CDC’s MDDR service is a tNGS assay that can be performed on primary sediments and cultures, while MDL performs WGS-DST when cultures are submitted and tNGS-DST for processed specimens.
Covered gene loci: MDL’s TB WGS-DST test covers all targets available from CDC’s MDDR service and includes some additional loci. MDL’s tNGS-DST assay includes additional loci for Ethionamide and Capreomycin but is limited to the Rv0678 locus for Bedaquiline and Clofazimine. The exact genetic coordinates of the regions investigated by the CDC is available on MDDR website and MDL’s WGS and tNGS reportable genomic regions are shown below.

Q: What genetic loci are investigated by MDL’s WGS and tNGS DST assays?

A: Below is the list of genes and the genomic ranges within which MDL’s sequencing-based DST assays report mutations:

Gene name	Associated with resistance to	WGS Start nt position	WGS Stop nt position	tNGS Start nt position	tNGS Start nt position
rpoB-FL*	rifampicin	759607	763525	n/a	n/a
rpoB-1*	rifampicin	n/a	n/a	760280	760812
rpoB-2*	rifampicin	n/a	n/a	760957	761355
pncA	pyrazinamide	2288481	2290323	2288672	2289301
embA	ethambutol	4243004	4246717	not tested	not tested
embB	ethambutol	4246314	4250010	4247376	4248065
katG	isoniazid	2153689	2156570	2153404	2156137
fabG1	isoniazid, ethionamide	1673148	1674383	1673321	1673755
inhA	isoniazid, ethionamide	1673848	1675211	1674287	1674880
ethA	ethionamide	4325804	4330174	4325951	4327510
gyrB	moxifloxacin, levofloxacin	5040	7467	7377	7754
gyrA	moxifloxacin, levofloxacin	7102	10018	6298	6943
rrs	kanamycin, capreomycin, amikacin	1471646	1473582	1471850	1473945
eis	kanamycin, amikacin	2713924	2715586	2715171	2715528
tlyA	capreomycin	1917740	1918946	1917811	1918750
Rv0678	clofazimine, bedaquiline	778790	779687	778990	779487
mmpL5	clofazimine, bedaquiline	775386	778680	not tested	not tested
mmpS5	clofazimine, bedaquiline	778277	779105	not tested	not tested
pepQ	clofazimine, bedaquiline	2859100	2860618	not tested	not tested
atpE	bedaquiline	1460845	1461490	not tested	not tested
rplC	linezolid	800609	801662	801108	801483
rrl	linezolid	1473458	1476995	1475923	1476625

Footnotes: Position coordinates listed in relation to the M. tuberculosis H37Rv NC_000962.3

* WGS-DST assay sequences the full-length gene and promoter region of rpoB, while tNGS-DST assay covers two fragments within rpoB (including RIF-resistance determining region or RRDR).

Q: Is there still utility in ordering phenotypic DST if I’m already getting WGS-DST for my isolate?

A: Yes. The phenotypic and WGS results are complementary. In particular, there may be cases for which a mutation is not detected in genes associated with resistance but the isolate is phenotypically resistant due to an unknown mechanism. Phenotypic DST is also important to clarify an effect of mutations of uncertain significance detected by WGS. Hence, if phenotypic DST has NOT been performed already at a clinical or public health laboratory, we recommend ordering first-line or the alternative 4-month regimen first-line drug panel for phenotypic DST, in parallel with WGS-DST. Please note that for some drugs (e.g. PZA), the WGS results may be more reliable than those of the phenotypic assay.

In the future, MDL will evaluate modifying our testing algorithm to utilize WGS DST as the primary clinical test for first line drug resistance determination. In such workflow, samples would be only reflexed to phenotypic DST for the confirmation of resistance-conferring mutations, mutations with an uncertain effect, and upon submitter request. Submitters will be notified in the event of any service changes.

Q: Our lab performs phenotypic DST in-house. Should we send all new TB isolates to MDL for WGS DST since it covers so many different AMR genes, or only those isolates that we have problems with from our culture-based DST?

A: TB isolates from all newly diagnosed patients should be submitted for national TB WGS surveillance. MDL now accepts TB isolates for WGS genotyping and submits WGS data for national TB surveillance. If submitting isolates to MDL for WGS genotyping, we recommend ordering WGS DST as well, which would provide comprehensive information on the predicted molecular DST profile.

Q: What is the typical time frame for phenotypic versus WGS DST testing if both tests are ordered? If the phenotypic result becomes available later, does that require physicians to revisit both WGS and phenotypic results?

A: When WGS and pDST are requested at the same time, both will be set up in parallel. WGS results will normally be available and reported to the submitters before phenotypic results (regularly, WGS TAT 10-21 days, phenotypic DST TAT 19-45 days). We recommend reviewing and evaluating all molecular and phenotypic DST results, in addition to other clinical and laboratory data as results become available. Discrepancies between WGS and phenotypic results are possible and consultation is available by contacting MDL (Matthew.Sylvester@cdph.ca.gov, Varvara.Kozyreva@cdph.ca.gov, CDPHTBDST@cdph.ca.gov) and MDR services (MDRTBService@cdph.ca.gov ).

Q: If a culture is determined to be mixed by WGS, will it automatically be reflexed for tNGS?

A: Yes, cultures for which WGS results indicate contamination will be reflexed to tNGS automatically.

Q: Is false-resistance possible in either WGS-DST or tNGS assays if an MTBC sample is contaminated with a nontuberculous mycobacteria (NTM)?

A: In our validation studies we saw no evidence of this being a concern. We evaluated the specificity of both WGS-DST and tNGS-DST assays by analyzing sequences of various NTM species with our bioinformatics pipeline for resistance prediction and for those sample the pipeline did not return any results that could be mistaken for MTBC. Additionally, we performed an in silico contamination study by mixing MTBC and NTM DNA in different proportions and did not observe any interference with WGS-DST or tNGS-DST from NTM contamination that would result in false-positive or false-negative resistance detection; however, for WGS-DST, above a certain percentage, contamination resulted in assay failure. Therefore, even though the WGS-DST assay is tolerant to some level of contamination, we still require a pure culture. The tNGS-DST assay is better suited for detection of TB resistance targets from mixed cultures.

Q: Will the WGS DST results be reported to the submitting labs, CalREDIE, or both?

A: Results will be reported to the submitting public health laboratories at this time. MDL is working to implement CalREDIE reporting in the near future; however, it is not available currently.

Q: Can the WGS and/or tNGS assays provide identification (ID) for all MTBC species?

A: Even though WGS technology has a potential to identify all MTBC species, the WGS DST assay validated by MDL only provides identification for Mycobacterium tuberculosis and M. bovis species and further differentiates BCG strain of M. bovis. In cases when DNA of MTBC organism is detected but cannot be identified as either M. tuberculosis or M. bovis, it will be reported as “DNA of Mycobacterium tuberculosis complex detected.” If no DNA of MTBC is detected, this will be also reflected on the report.

The tNGS method has been validated by MDL for MTBC ID confirmation as well, but unlike WGS, tNGS only allows confirmation of MTBC DNA presence and differentiation of M. bovis from non-M. bovis MTBC. M. tuberculosis species or BCG strain of M. bovis are not specifically differentiated by tNGS. Please note that the tNGS assay is not intended for diagnosis of TB disease and negative results for the detection of MTBC DNA in the specimen do not rule out presence of TB.

Q: How do I order MTBC species identification on the Lab Web Portal?

A: MTBC ID confirmation is performed as a part of Sequencing-based DST. If the submitted material is a pure culture, MTBC ID confirmation will be performed by WGS. If the submitted material is a processed specimen or mixed culture, MTBC ID confirmation will be performed by the tNGS method.

Q: If I request WGS-genotyping, should I expect a report with genotyping results from MDL?

A: For submitted pure cultures, WGS-genotyping data will be uploaded to the CDC for national surveillance. A report will be issued to the submitter only when MDL is unable to perform WGS genotyping (e.g. due to culture appearing mixed upon sequencing) and will include a request for the submitter to send an isolate. All genotyping results will be available in TB GIMS.

Q: If we submit an isolate for genotyping only, will we be notified of MDR cases?

A: If submitter did not request TB WGS DST on the requisition form and only requested WGS genotyping, we will not automatically perform analysis for resistance detection from the WGS data. However, once MDL starts submitting WGS data for national TB surveillance, we do recommend the following:

If you are submitting an isolate from a new patient, or if the previous tested sample for a patient was collected 3 months ago or more, request both WGS DST and WGS genotyping. This will allow you to receive a clinical report with predicted resistance results.
If clinician has concerns that justify WGS for an isolate from a patient who had previous testing on a sample collected <3 months ago (e.g., concern for acquired resistance, etc.), contact MDL for pre-approval, and WGS DST and/or WGS genotyping may be provided on case-by-case basis.

Q: Do LHJs also need to submit isolates to the Michigan State Public Health Laboratory since MDL now submits WGS data from sequenced isolates for national TB surveillance?

A: No, submission of TB isolates to MDL for WGS genotyping will replace the need to send isolates to Michigan Genotyping lab.

Q: Do I need to send an Isolate Submission Form (ISF) with patient metadata to the MDL when requesting TB WGS genotyping?

A: No, all patient data for TB GIMS uploads will be collected via the MDL Lab Web Portal (LWP/ETOR). If submitting isolates to MDL, please don’t send an ISF to CDPH TB controllers to avoid confusion.

Q: What are the performance characteristics of the WGS-DST assay and tNGS-DST assays?

A: Overall MTBC WGS-DST and tNGS-DST assays performance (please note that since the validation datasets differed, a direct performance comparison between the two methods based on the values below is not possible):

Performance Characteristic	Evaluated aspect of the assay	WGS	tNGS
Accuracy	Predicted S/R profile	98.16%	98.68%
	Detection of genomic variations in targeted loci	99.73%	99.70%
	MTBC ID	100%	96.88%
Repeatability (Qualitative)	Predicted S/R profile	100%	100%
	Detection of genomic variations in targeted loci	100%	100%
	MTBC ID	100%	100%
Reproducibility (Qualitative)	Predicted S/R profile	100%	100%
	Detection of genomic variations in targeted loci	99.8%	99.50%
	MTBC ID	100%	100%
Diagnostic Sensitivity	Predicted S/R profile	91.25%	93.44%
	Detection of genomic variations in targeted loci	99%	98.90%
	MTBC ID	100%	96.36%
Diagnostic Specificity	Predicted S/R profile	99.40%	100.00%
	Detection of genomic variations in targeted loci	99.89%	99.80%
	MTBC ID	100%	100.00%

Definitions:

Predicted S/R profile- predicted susceptibility(S)/resistance(R) to drugs in MTBC organisms based on WGS-DST or tNGS.
Detection of genomic variations in targeted loci- detection of genomic variations (single nucleotide polymorphisms [SNPs], multi nucleotide polymorphisms [MNPs], and indels) throughout the established reporting range in the genetic targets included in this assay that are known or suspected of being responsible for drug resistance in MTBC.
MTBC ID- For WGS: identification of MTBC (based on lineage-specific SNPs) and differentiation of M. tuberculosis from M. bovis, and further delineation of M. bovis BCG strain. For tNGS: ability to detect MTBC DNA from clinical specimen and differentiation of M. bovis vs. MTBC not-M. bovis.

WGS-DST assay performance by drug:

Drug	TP	TN	FN	FP	Accuracy	Dx Sensitivity	Dx Specificity	PPV	NPV
INH	85	86	2	0*	98.84%	97.70%	100%	100%	97.73%
ETA	44	114	0	3*	98.14%	100%	97.44%	93.62%	100%
RIF	18	171	0	2	98.95%	100%	98.84%	90%	100%
PZA	26	147	16	0	91.53%	61.90%	100%	100%	90.18%
EMB	11	170	0	3*	98.37%	100%	98.27%	78.57%	100%
AMK	5	173	0	0	100%	100%	100%	100%	100%
KAN	4	126	0	0	100%	100%	100%	100%	100%
CAP	6	183	1	0	99.47%	85.71%	100%	100%	99.46%
MFX	17	129	2	0	98.65%	89.47%	100%	100%	98.47%
LFX	1	14	0	0	100%	100%	100%	100%	100%
BDQ	0	5	0	0	100%	N/A	100%	N/A	100%
CFZ	0	5	0	0	100%	N/A	100%	N/A	100%
LZD	2	2	0	0	100%	100%	100%	100%	100%
Overall	219	1325	21	8	98.16%	91.25%	99.40%	96.48%	98.44%

Footnotes: *Numbers after discrepancies were resolved with gold standard method.

tNGS-DST assay performance by drug:

Drug	TP	TN	FN	Accuracy	Dx Sensitivity	Dx Specificity	PPV	NPV
INH	13	16	1	96.67%	92.86%	100%	100%	94.12%
ETA	7	24	1	96.88%	87.50%	100%	100%	96%
RIF	10	27	0	100%	100%	100%	100%	100%
PZA	7	23	2	93.75%	77.78%	100%	100%	92%
EMB	8	28	0	100%	100%	100%	100%	100%
AMK	3	31	0	100%	100%	100%	100%	100%
KAN	2	19	0	100%	100%	100%	100%	100%
CAP	3	32	0	100%	100%	100%	100%	100%
MFX	4	26	0	100%	100%	100%	100%	100%
LFX	0	2	0	100%	N/A	100%	N/A	100%
BDQ	0	5	0	100%	N/A	100%	N/A	100%
CFZ	0	5	0	100%	N/A	100%	N/A	100%
LZD	0	4	0	100%	N/A	100%	N/A	100%
Overall	57	242	4	98.68%	93.44%	100%	100%	98.37%

Footnotes: *Numbers after discrepancies were resolved with gold standard method.

WGS-DST assay performance by gene target:

Performance	pncA	katG	fabG1/ inhA	rpoB	gyrA	gyrB	rrs	embB	rv0678	atpE	pepQ	mmpS5	mmpS5	rplC	rrl	Total
TP	27	48	43	30	21	0	5	5	12	3	1	1	1	2	0	199
TN	157	123	129	143	98	27	114	23	5	15	16	15	15	16	12	908
FP	0	0	0	0	0	0	0	0	0	0	1	0	0	0	0	1
FN	0	0	0	0	2	0	0	0	0	0	0	0	0	0	0	2
Accuracy	100%	100%	100%	100%	98.35%	100%	100%	100%	100%	100%	94.44%	100%	100%	100%	100%	99.73%
Specificity	100%	100%	100%	100%	100%	100%	100%	100%	100%	100%	94.12%	100%	100%	100%	100%	99.89%
Sensitivity	100%	100%	100%	100%	91.30%	N/A	100%	100%	100%	100%	100%	100%	100%	100%	N/A	99.00%
PPV	100%	100%	100%	100%	100%	N/A	100%	100%	100%	100%	50%	100%	100%	100%	N/A	99.50%
NPV	100%	100%	100%	100%	98.00%	100%	100%	100%	100%	100%	100%	100%	100%	100%	100%	99.78%

Footnote: Performance for several WGS gene targets could not be calculated directly due to lack of reference molecular data on mutations in those loci.

tNGS-DST assay performance by gene target:

Performance	pncA	katG	inhA	fabG1	rpoB	gyrA	*gyrB*	rrs	eis	tlyA	embB	rv0678	rplC	rrl	ethA	Total
TP	12	18	2	7	15	6	5	4	1	1	10	4	1	0	7	93
TN	25	31	59	45	39	36	37	34	37	36	28	34	36	36	30	543
FP	0	0	0	0	0	0	0	1	0	0	0	0	0	0	0	1
FN	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1	1
Accuracy	100%	100%	100%	100%	100%	100%	100%	97.4%	100%	100%	94.44%	100%	100%	100%	97.4%	99.73%
Specificity	100%	100%	100%	100%	100%	100%	100%	97.1%	100%	100%	94.12%	100%	100%	100%	100%	99.89%
Sensitivity	100%	100%	100%	100%	100%	100%	100%	100%	100%	100%	100%	100%	100%	N/A	87.5%	99.00%
PPV	100%	100%	100%	100%	100%	100%	100%	80%	100%	100%	50%	100%	100%	N/A	100%	99.50%
NPV	100%	100%	100%	100%	100%	100%	100%	100%	100%	100%	100%	100%	100%	100%	96.8%	99.78%

Please note that the sensitivity for PZA for both the tNGS and WGS assays is likely underestimated as the reference method (i.e. phenotypic DST by MGIT) is known to have a high rate of false resistance due to technical variation that may affect results. The sensitivity for fluoroquinolones in our study was likely impacted by the presence of heteroresistance.

Q: How should I interpret the results of the WGS-DST or tNGS-DST report?

A: The most important information on the WGS-DST or tNGS-DST report for clinical use is the sequencing-based DST interpretations that are provided on the report opposite the corresponding drug name. The user may review information about detected mutations for each individual gene displayed on the report; however, all detected genomic variations and their potential effects are summarized for each drug based on the validated by MDL interpretation algorithm. Below are the possible sequencing-based DST result options for individual drugs and their meaning:

Drug interpretation result as it appears on clinical report	Explanation
Mutation(s) associated with resistance to XXX detected	There is strong evidence of association of the detected mutation with phenotypic resistance. Susceptibility is highly unlikely. This prediction is based on the data collected by WHO and the global TB community on correlation of specific mutations with resistance in phenotypic assays, and overall scientific knowledge of resistance mechanisms in MTBC.
The detected mutation(s) have uncertain significance. Resistance to XXX cannot be ruled out	Insufficient amount of evidence is available regarding the association of the detected mutation (s) with resistance. Confirmatory phenotypic DST is needed for definitive resistance determination.
No mutations associated with resistance to XXX detected	Resistance is unlikely, but cannot be ruled out due to: Mutations in loci not covered by WGS or tNGS that contribute to unknown mechanisms of resistance; or, Heteroresistance below the limit of detection (LOD) of the sequencing-based assay
Predicted resistance to rifampin OR Predicted susceptibility to rifampin	There is a high likelihood that the strain is resistant/susceptible to rifampin. Resistance/susceptibility to rifampin can be predicted with higher certainty than other drugs, and this is reflected in the reporting language.
Predicted susceptibility to rifampin. The detected synonymous mutation(s) do not confer resistance	Synonymous mutations are normally not reported for drugs other than rifampin, however, synonymous mutations in rpoB are to resolve potential discrepancies with PCR-based assays, e.g., silent mutations that cause false-positive results in the Xpert MTB-RIF PCR assay.
Predicted low-level resistance to rifampin. May test susceptible by phenotypic methods	The detected mutation is known to cause low-level, yet clinically relevant, resistance to rifampin. When strains harboring such mutations are tested by phenotypic DST at a critical concentration of 1 ug/mL, they may test susceptible.
Pending Retest	One or more gene targets had insufficient coverage, which does not allow confident susceptibility predictions for the corresponding drug. This is a preliminary report, and the sample will be re-sequenced.
Not all targets could be sequenced; resistance to XXX cannot be ruled out	One or more gene targets had insufficient coverage, which does not allow confident susceptibility predictions for the corresponding drug. Sequencing has been repeated but successful sequence could not be obtained. This is a final report. Please contact the lab to see if sample resubmission is necessary.

Q: Where can I get information on the number of isolates with a given mutation that previously tested Resistant vs. Susceptible?

A: At MDL, we utilize the WHO “Catalogue of mutations in Mycobacterium tuberculosis complex and their association with drug resistance” as a basis for our mutation interpretations. It contains information on individual mutations in the TB genome, the number of samples encountered that harbor this mutation, and its association with phenotypic resistance/susceptibility. MDL is currently using the WHO v.1 (2021) database of mutations, but for the reference purposes, we recommend referring to the WHO v.2 (2023) catalogue of mutations. MDL will revalidate our reportable interpretations based on the WHO v.2 catalogue in the near future.

The searchable tables with all described mutations are available as supplementary material (WHO-UCN-TB-2023.7-eng.xlsx [Link; click “View raw” to download]). It is recommended to review Columns “Present_SOLO_R” and “Present_SOLO_S” for the number of strains in the WHO dataset possessing a single given mutation, in the absence of other resistance-conferring mutations, that tested resistant or susceptible to the corresponding drug, respectively. One of the particularly useful values for evaluating the likelihood that a mutation causes resistance is the “PPV|SOLO_lb” value (positive predictive value for resistance); the higher the value, the more likely the presence of this mutation is associated with resistance.

Be aware that in WHO v.2 catalogue of mutations, the nomenclature for inhA / fabG1 mutations has changed and all mutation positions are now displayed in relation to inhA. For example, a mutation previously known as fabG1 c.-15C>T is now listed as inhA c.-777C>T in the WHO v.2 catalogue. Information about the former alias for such mutations is available in WHO v.2 tables in the Comments section.

Q: For rifampin, why does the WGS/tNGS-based predictions indicate “Predicted susceptibility” or “Predicted resistance,” but for other drugs it only states that mutation(s) associated with resistance are detected/not detected?

A: Because we have much more extensive data on the effect of rpoB mutations on RIF resistance and mutations in the hot-spot (rifampin resistance determining region) of the rpoB gene are a predominant mechanism of RIF resistance in clinical strains, we have more certainty in WGS/tNGS predictions. Consequently, we give a more definitive statement about predicted resistance or susceptibility. We also have more knowledge of correlation of different mutations with different levels of resistance to RIF; therefore, mutations that are known to result in low-level resistance to RIF are reported with the corresponding comments, warning submitters that such strains may test susceptible by phenotypic assays.

Q: What does it mean when a mutation with “uncertain significance” is detected?

A: A mutation could be categorized as uncertain significance in two cases:

The mutation has been previously seen in the WHO catalogue (v.1) with some pre-existing data on correlation with the phenotype available; however, it is insufficient to make a statistically significant prediction of either resistance or susceptibility.
The mutation has not been seen previously, but it is covered by an expert rule that allows the interpretation of mutations that are found within genomic regions for which mechanisms of resistance are well understood.
The mutation has not been seen previously, but it is a non-synonymous mutation, i.e. leads to an amino acid change; therefore, its effect on drug resistance cannot be ruled out. For any questions regarding the WGS-DST or tNGS-DST results and the interpretation logic applied in each individual case, please reach out to MDL (Matthew.Sylvester@cdph.ca.gov, Varvara.Kozyreva@cdph.ca.gov, CDPHTBDST@cdph.ca.gov).

We recommend that cases of “uncertain” mutations undergo confirmatory phenotypic DST for the corresponding drug, either at MDL or another laboratory. With the current testing algorithm, when an “uncertain” mutation is detected by WGS or tNGS, and prior pDST has not been previously performed, MDL will automatically reflex the isolate for pDST confirmation for the potentially affected drug only, even if pDST was not originally requested by the submitter. For the drugs for which pDST is not available at MDL, we can assist with referral to other labs.

The determination of treatment regimens and/or discontinuation of treatment should be made based on a combination of available molecular and phenotypic DST data, as well as other laboratory and clinical data. Please contact CDPH MDR services (MDRTBService@cdph.ca.gov) for the consultation if you have any questions about clinical interpretation of your laboratory results and treatment decision-making.

Q: How do I interpret the results if an isolate has discrepant phenotypic and sequencing-based DST results? Specifically, what do I do if phenotypic result for pyrazinamide (PZA) is resistant, but no mutation was detected?

A: Discrepancies between sequencing-based DST (WGS/tNGS) and phenotypic results are possible due to a number of factors, including but not limited to:

Resistance-conferring mutations in genomic regions not covered by the sequencing-based DST assay.
Unknown mechanisms of resistance.
Low level heteroresistance that is below the sequencing assay limit of detection (LOD).
Propensity of phenotypic DST for some drugs for false-resistance or false-susceptibility.

Specifically, phenotypic testing of PZA using the BACTEC MGIT system is prone to poor reproducibility and false-positive results; this may lead to discrepancies such as when no pncA mutations are detected by sequencing but the strain tests phenotypically resistant. There are a number of high-confidence mutations in the pncA gene that confer resistance to PZA. PZA resistance can be also conferred by mutations in secondary (non-pncA) targets, but their clinical significance is less clearly understood. False resistance is particularly likely for isolates that test phenotypically PZA-monoresistant without pncA mutations. Due to the issues with the phenotypic method for PZA DST and a low pre-test probability, predicted resistance based on pncA sequencing results are more reliable in cases of phenotypic PZA monoresistance. However, it is important to take into account clinical data and other laboratory results.

In case of discrepant phenotypic and sequencing-based results for any drugs, please contact MDL (Matthew.Sylvester@cdph.ca.gov, Varvara.Kozyreva@cdph.ca.gov, CDPHTBDST@cdph.ca.gov) and MDR services (MDRTBService@cdph.ca.gov) for consultation and resolution of the discrepancy.

Q: What is the difference between the “No high confidence mutations detected” and “No mutations detected” results that appear on the report?

A: We only report mutations for which phenotypic data that is available from WHO or other reputable sources or those that are covered by “expert rules” that are based on prior knowledge of mechanisms of resistance in TB. This includes non-synonymous mutations that are NOT found in the current version of WHO catalogue, since their effect on drug resistance cannot be ruled out. However, we do differentiate the following genotypes which both are interpreted as “No mutations associated with resistance to XXX detected:”

“No mutations detected:” when the gene does not have any mutations (i.e. “wild-type,” WT)
“No high confidence mutations detected:” gene contains mutations that are not likely to cause resistance. A “no high confidence mutations detected” genotype could result from a gene containing either

Synonymous mutations;
Non-synonymous mutations that are present in the WHO catalogue and are known NOT to be associated with resistance (i.e. neutral mutations);
Mutations in promoter regions or non-protein encoding genes that are NOT in the WHO database and are NOT covered by expert rules.

Q: When the WGS/tNGS-DST report says “No sequence” for a given gene, what does it mean? Should I expect results at a later time?

A: If the interpretation for the given drug (e.g. moxifloxacin) indicates “Pending Retest” and one of the associated genes (e.g. gyrB) indicates “No sequence,” then this is a preliminary report and the sample will be resequenced to obtain the valid gyrB sequence. If there is no sequence for the gene and the interpretation for the drug states: “Not all targets could be sequenced; resistance to moxifloxacin cannot be ruled out,” this means it is a final report and no additional report will be released. In the latter case, the issues are likely caused by the contamination or suboptimal quality of the sample; submitter will have an option to resubmit.

It is also possible to have a resistance-conferring mutation detected in one of the genes (e.g. gyrA) and have unsuccessful sequencing in another gene (gyrB) associated with the same drug (moxifloxacin); in that case, the report may be finalized without repeated sequencing, since the sequence of gyrB gene in this particular example would not affect the interpretation for moxifloxacin. In this case, no follow-up reports will be issued.

Q: Are synonymous mutations and mutations known not to be associated with resistance reported on the WGS/tNGS-DST report?

A: Generally, synonymous mutations (ones that do not lead to amino acid change) and non-synonymous mutations that are known to be neutral (not associated with resistance) are not displayed on MTBC WGS/tNGS DST report since they do not have an effect on resistance. Exceptions are:

Synonymous mutations within the rifampicin resistance determining region (RRDR) of the rpoB gene (codons 426-452) are reported because they have a potential to cause false-positive results in real-time PCR assays used for RIF resistance detection such as GeneXpert MTB-RIF. Either WGS or tNGS can resolve such false-positive results; hence, we recommend submitting isolates determined to be resistant by PCR methods that cannot differentiate synonymous and nonsynonymous mutations for sequencing-based DST for confirmation.
The synonymous mutation p.Leu203Leu in fabG1, even though silent, confers resistance to isoniazid and is reported.

Q: In the case of heteroresistance, what is the minimum percent of the resistant strain subpopulation that can be detected by WGS or tNGS DST assay?

A: Due to a variety of factors including sampling bias and a possible shift in the representation of the resistant subpopulation of the strain during the culture growth in vitro for WGS or due to bias introduced by amplification in case of tNGS, it is difficult to correlate the exact percentage of the resistant subpopulation as determined by sequencing-based DST with what exists in the patient. However, in our validation study we have determined that the limit of detection (LOD) for the mutation allele frequency in TB genome that ensures high accuracy of the WGS-DST assay is around 10%. tNGS-DST is comparable to WGS, but due to the difference in amplification efficiency of different loci, it has an additional variability in detected allele frequency. Both sequencing-based DST assays are well-suited for detection of heteroresistance in MTBC but cannot be used quantitatively.

Q: What is the limit of detection for the tNGS assay? What AFB grade or Gene Xpert MTB/RIF Ct values correlate with successful sequencing by tNGS?

A: Based on our validation study, the MTBC tNGS-DST assay is expected to perform well with processed clinical specimens demonstrating a Xpert MTB/RIF PCR Ct value less than or equal to 28, or approximately AFB 1+ microscopic grading; samples with lower TB loads may or may not sequence well. Pre-approval is required for specimens with rare or no AFB observed, or with a Xpert MTB/RIF Ct value higher than 28 since the probability of successful sequencing is lower.

Page Last Updated :