Tim Hermanus Heupink

Background: Comprehensive knowledge of resistance to the novel all-oral regimens for treatment of rifampicin-resistant tuberculosis (RR-TB) is important, but challenging variant detection by next-generation sequencing (NGS) in combination with variant interpretation software can diagnose resistance early. We aimed to generate a consensus on strategic issues for the implementation of NGS for routine RR-TB care. Methods: We employed a stakeholder consensus process informed by Delphi principles. First, stakeholders participated in an online survey to provide feedback on statements using a 3-point Likert scale and free text. Responses were then summarised and follow-up discussion fora with stakeholders were held to revise statements until a consensus was reached. Findings: Of the 116 invited stakeholders, 58 participated, representing RR-TB research (n=28), clinical care (n=11), diagnosis (n=5), RR-TB survivors and advocacy communities (n=5) from 14 countries in five of the six WHO regions. A consensus was rapidly reached for most statements with limited editorial changes needed. The debate mostly focused on statements covering performing sequencing at centralised facilities, acting on variants without statistical association with resistance in the WHO catalogue of mutations, strengthening WHO-endorsed regimens and the need for informed consent for the use of NGS for care and surveillance. Interpretation: The consensus-building process grounded in real-world experience from a diverse group of stakeholders generated 32 consensus statements covering five strategic domains. The blueprint generated is not a definite prescriptive framework but a document that can support policymakers when developing country-specific policies for NGS-guided precision medicine and RR-TB care. Sources of funding: This work was supported by the Global Health EDCTP3 Joint Undertaking and the Bill & Melinda Gates Foundation under the GenPath Africa project (101103171), the Research Foundation Flanders (FWO) Odysseus (G0F8316N) and TBM grants (T001018N).

Background: Whole genome sequencing (WGS) can guide effective rifampicin-resistant tuberculosis (RR-TB) treatment but the cost is a perceived barrier.Methods: We performed a health-economic evaluation of culture-based WGS combined with automated recommendation for an all-oral 6-month individualized regimen. Provider, patient and household costs for RR-TB management were analyzed for 203 participants randomized to the WGS intervention or standard of care (SOC) in the South African SMARTT trial (Sept 2021-Feb 2023). Cost-effectiveness was assessed by incremental cost-effectiveness ratio (ICER); monetary value by Incremental Net Monetary Benefit (INMB) and Return on Investment (ROI).Findings: WGS-guided treatment was dominant, reducing the per patient mean total cost (-$370) and increasing the per patient health benefit (-0.05 DALYs), resulting in an INMB of $618. Society gained $3.61 for every dollar invested. Among people living with HIV, WGS-guided treatment resulted in greater cost reduction (-$582 vs -$22), greater DALY reduction (-0.08 vs -0.01) and higher INMB ($969 vs $51). The benefit was larger for women than men (cost savings $694 vs $191, DALY reduction -0.06 vs -0.03, and INMBs $997 vs $319). People with expanded resistance experienced the greatest health benefit (0.46 DALYs averted), but the intervention was not cost-effective in this population (INMB $-1231) due to higher drug ($1690 vs $1260) and inpatient costs ($4,244 vs $1,201).Interpretation: WGS-guided RR-TB treatment was cost-effective and society gained $3.6 benefit for every dollar invested. Future research should evaluate the cost-effectiveness of WGS-guided treatment in contexts where an all-oral 6-month regimen is the SOC.

Whole-genome sequencing (WGS) holds promise for accurate and comprehensive diagnosis of drug resistance in Mycobacterium tuberculosis and identification of transmission events. ‘Early positive cultures’ (EPC) are increasingly used when WGS is implemented to guide clinical care to reduce the turnaround time. We performed a systematic literature review to compare methods used for EPC-based WGS and performed an individual sample data meta-analysis to identify variables associated with bioinformatic quality measures. Of 423 studies identified, 15 met eligibility criteria. We analysed 1,065 FASTQ files from 11 studies using Illumina sequencing; 96.1% passed all quality control thresholds. Median genome coverage was 65× (IQR, 63–82), with a pooled mapping percentage of 91.2%. The meta-analysis showed that the number of sequencing cycles was significantly associated with improved sequencing quality, while other laboratory variables had no consistent effect. Based on these findings, we suggest replacing the term EPC with ‘clinical primary culture’ and propose a standardized workflow and reporting checklist for WGS on primary Mycobacteria Growth Indicator Tube (MGIT) cultures.

Background: Challenges in tuberculosis (TB) control have fuelled interest in routine next generation sequencing (NGS). However, the comparative performance of targeted NGS (tNGS) and whole genome sequencing (WGS) for drug resistance and genetic relatedness inference remains unclear. Methods: In this cross-sectional study, we compared WGS (MICK-MAGMA platform) and the Deeplex MycTB tNGS assay in 90 patients with rifampicin-resistant TB in South Africa. A pairwise analysis was conducted for a total of 60 isolates, for which tNGS was conducted directly on DNA from sputum, and WGS on cultured isolates from the same samples. Results: Drug resistance inference was highly concordant (≥92%) for most drugs, but lower for isoniazid (82%) and ethionamide (78%). Mixed infections were more commonly detected in tNGS (6.7%) than WGS (1.7%), though likely due to analytical errors. tNGS detected more minor variants (allelic frequency versus 32), with minimal overlap. Most minor variants were of unknown significance; some likely stemmed from contamination or sequencing errors. Heteroresistance involving minor variants was rare (4.7% in tNGS, 0% in WGS). WGS provided lineage and sublineage information for all samples, while tNGS reported lineage for 67% and sublineage for 20%. WGS classified more samples as genetically unrelated (76%) than tNGS (40%). Conclusion: In this cohort, drug resistance inference was largely concordant between tNGS and WGS. WGS offered higher resolution for genetic relatedness, while tNGS showed greater sensitivity for minor variants. Further research is needed to clarify the clinical relevance of minor variants and assess the utility of WGS for transmission control.

Background: Whole genome sequencing (WGS) holds great potential for the management and control of tuberculosis. Accurate analysis of samples with low mycobacterial burden, which are characterized by low (40%) levels of contamination, is challenging. We created the MAGMA (Maximum Accessible Genome for Mtb Analysis) bioinformatics pipeline for analysis of clinical Mtb samples. Methods and results: High accuracy variant calling is achieved by using a long seedlength during read mapping to filter out contaminants, variant quality score recalibration with machine learning to identify genuine genomic variants, and joint variant calling for low Mtb coverage genomes. MAGMA automatically generates a standardized and comprehensive output of drug resistance information and resistance classification based on the WHO catalogue of Mtb mutations. MAGMA automatically generates phylogenetic trees with drug resistance annotations and trees that visualize the presence of clusters. Drug resistance and phylogeny outputs from sequencing data of 79 primary liquid cultures were compared between the MAGMA and MTBseq pipelines. The MTBseq pipeline reported only a proportion of the variants in candidate drug resistance genes that were reported by MAGMA. Notable differences were in structural variants, variants in highly conserved rrs and rrl genes, and variants in candidate resistance genes for bedaquiline, clofazmine, and delamanid. Phylogeny results were similar between pipelines but only MAGMA visualized clusters. Conclusion: The MAGMA pipeline could facilitate the integration of WGS into clinical care as it generates clinically relevant data on drug resistance and phylogeny in an automated, standardized, and reproducible manner.

Background: Whole genome sequencing (WGS) holds great potential for the management and control of tuberculosis. Accurate analysis of samples with low mycobacterial burden, which are characterized by low (40%) levels of contamination, is challenging. We created the MAGMA (Maximum Accessible Genome for Mtb Analysis) bioinformatics pipeline for analysis of clinical Mtb samples. Methods and results: High accuracy variant calling is achieved by using a long seedlength during read mapping to filter out contaminants, variant quality score recalibration with machine learning to identify genuine genomic variants, and joint variant calling for low Mtb coverage genomes. MAGMA automatically generates a standardized and comprehensive output of drug resistance information and resistance classification based on the WHO catalogue of Mtb mutations. MAGMA automatically generates phylogenetic trees with drug resistance annotations and trees that visualize the presence of clusters. Drug resistance and phylogeny outputs from sequencing data of 79 primary liquid cultures were compared between the MAGMA and MTBseq pipelines. The MTBseq pipeline reported only a proportion of the variants in candidate drug resistance genes that were reported by MAGMA. Notable differences were in structural variants, variants in highly conserved rrs and rrl genes, and variants in candidate resistance genes for bedaquiline, clofazmine, and delamanid. Phylogeny results were similar between pipelines but only MAGMA visualized clusters. Conclusion: The MAGMA pipeline could facilitate the integration of WGS into clinical care as it generates clinically relevant data on drug resistance and phylogeny in an automated, standardized, and reproducible manner. Key points: - Accurate analysis of clinical samples is challenging when samples have high levels of contamination and low Mycobacterium tuberculosis genome coverage - When analyzing primary liquid (MGIT) cultures, the MAGMA pipeline generates clinically relevant drug resistance information (including major, minor and structural variants) and phylogeny in an automated, standardized and reproducible way. - MAGMA-generated phylogenetic trees are annotated with drug resistance information and updated with every run so that they can be used to make clinical or public health decisions - MAGMA reports drug resistance variants for all tier 1 and tier 2 candidate drug resistance conferring genes, with interpretation of their relevance to drug resistance (associated with drug resistance, not associated with drug resistance or unknown significance) based on the WHO catalogue of mutations in Mycobacterium tuberculosis .

Studies have shown that variants in bedaquiline-resistance genes can occur in isolates from bedaquiline-naive patients. We assessed the prevalence of variants in all bedaquiline-candidate-resistance genes in bedaquiline-naive patients, investigated the association between these variants and lineage, and the effect on phenotype.

Extensively drug-resistant tuberculosis (XDR-TB), defined as resistance to at least isoniazid (INH), rifampicin (RIF), a fluoroquinolone (FQ) and a second-line injectable drug (SLID), is difficult to treat and poses a major threat to TB control. The transmission dynamics and distribution of XDR Mycobacterium tuberculosis (Mtb) strains have not been thoroughly investigated. Using whole genome sequencing data on 461 XDR-Mtb strains, we aimed to investigate the geographical distribution of XDR-Mtb strains in the Western Cape Province of South Africa over a 10 year period (2006–2017) and assess the association between Mtb sub-lineage, age, gender, geographical patient location and membership or size of XDR-TB clusters. First, we identified transmission clusters by excluding drug resistance-conferring mutations and using the 5 SNP cutoff, followed by merging clusters based on their most recent common ancestor. We then consecutively included variants conferring resistance to INH, RIF, ethambutol (EMB), pyrazinamide (PZA), SLIDs and FQs in the cluster definition. Cluster sizes were classified as small (2–4 isolates), medium (5–20 isolates), large (21–100 isolates) or very large (>100 isolates) to reflect the success of individual strains. We found that most XDR-TB strains were clustered and that including variants conferring resistance to INH, RIF, EMB, PZA and SLIDs in the cluster definition did not significantly reduce the proportion of clustered isolates (85.5–82.2 %) but increased the number of patients belonging to small clusters (4.3–12.4 %, P=0.56). Inclusion of FQ resistance-conferring variants had the greatest effect, with 11 clustered isolates reclassified as unique while the number of clusters increased from 17 to 37. Lineage 2 strains (lineage 2.2.1 typical Beijing or lineage 2.2.2 atypical Beijing) showed the large clusters which were spread across all health districts of the Western Cape Province. We identified a significant association between residence in the Cape Town metropole and cluster membership (P=0.016) but no association between gender, age and cluster membership or cluster size (P=0.39). Our data suggest that the XDR-TB epidemic in South Africa probably has its origin in the endemic spread of MDR Mtb and pre-XDR Mtb strains followed by acquisition of FQ resistance, with more limited transmission of XDR Mtb strains. This only became apparent with the inclusion of drug resistance-conferring variants in the definition of a cluster. In addition to the prevention of amplification of resistance, rapid diagnosis of MDR, pre-XDR and XDR-TB and timely initiation of appropriate treatment is needed to reduce transmission of difficult-to-treat TB.

The study of genetic minority variants is fundamental to the understanding of complex processes such as evolution, fitness, transmission, virulence, heteroresistance and drug tolerance in Mycobacterium tuberculosis (Mtb). We evaluated the performance of the variant calling tool LoFreq to detect de novo as well as drug resistance conferring minor variants in both in silico and clinical Mtb next generation sequencing (NGS) data. The in silico simulations demonstrated that LoFreq is a conservative variant caller with very high precision (≥96.7%) over the entire range of depth of coverage tested (30x to1000x), independent of the type and frequency of the minor variant. Sensitivity increased with increasing depth of coverage and increasing frequency of the variant, and was higher for calling insertion and deletion (indel) variants than for single nucleotide polymorphisms (SNP). The variant frequency limit of detection was 0.5% and 3% for indel and SNP minor variants, respectively. For serial isolates from a patient with DR-TB; LoFreq successfully identified all minor Mtb variants in the Rv0678 gene (allele frequency as low as 3.22% according to targeted deep sequencing) in whole genome sequencing data (median coverage of 62X). In conclusion, LoFreq can successfully detect minor variant populations in Mtb NGS data, thus limiting the need for filtering of possible false positive variants due to sequencing error. The observed performance statistics can be used to determine the limit of detection in existing whole genome sequencing Mtb data and guide the required depth of future studies that aim to investigate the presence of minor variants.