TL;DR
High-throughput sequencing (HTS) offers a powerful new approach for diagnosing tuberculosis (TB) and identifying drug resistance directly from clinical samples, potentially revolutionizing patient care. This review provides a comprehensive overview of HTS workflows, platforms, and analytical tools, highlighting its potential to overcome current diagnostic limitations and accelerate the fight against TB.
What Problem Does This Paper Address?
Tuberculosis (TB) remains a significant global health challenge, with the World Health Organization aiming to end the pandemic by 2035. Current diagnostic methods, while improving, often face limitations. Phenotypic drug susceptibility testing (pDST), the gold standard, can take weeks to yield results, delaying appropriate treatment. Molecular tests like Xpert MTB/RIF offer faster results but are limited in the number of drugs they can detect. This gap leaves a critical need for diagnostic tools that are rapid, comprehensive, and can be applied directly to clinical samples, especially in resource-limited settings where TB burden is highest. High-throughput sequencing (HTS) emerges as a promising solution, capable of identifying Mycobacterium tuberculosis DNA and simultaneously assessing resistance to multiple drugs within days. This paper addresses the need for a consolidated understanding of HTS's application in clinical TB diagnostics, exploring its workflow, technological landscape, and analytical challenges.
Methods
This paper is a review, synthesizing existing research rather than presenting new experimental methods. It focuses on the workflow of HTS for TB, which generally involves:
- Sample Collection: Various clinical samples are used, including sputum, stool, bronchial aspirates, and bone marrow. Early morning sputum samples are preferred for higher yield.
- Storage/Culture: Samples can be stored at -80°C or preserved using agents like cetylpyridinium chloride (CPC) or ethanol to maintain DNA integrity and reduce contamination. Culture can be performed, but direct-to-sample sequencing is a key advantage of HTS.
- DNA Extraction: A variety of methods and kits are employed to extract DNA from samples or culture isolates, aiming to optimize DNA quantity and quality.
- Library Preparation: This step prepares the extracted DNA for sequencing. It can involve amplifying specific regions of interest (targeted HTS or tHTS) or preparing the entire genome (whole-genome sequencing or WGS). Multiplexing, using barcoding, allows multiple samples to be sequenced in a single batch, increasing cost-effectiveness.
- Sequencing: Various HTS platforms are utilized, broadly categorized into short-read (e.g., Illumina, Ion Torrent) and long-read (e.g., PacBio, Oxford Nanopore) technologies. The choice of platform depends on the specific application and desired output.
- Data Analysis: Raw sequencing data (FASTQ files) are processed through bioinformatics pipelines. This involves aligning sequences to a reference genome, quality control, variant calling, and interpreting these variants to identify drug resistance mutations.
Datasets
The paper does not present or analyze specific datasets for training or evaluation. Instead, it reviews numerous studies that have utilized various datasets (clinical samples from different geographic locations) to evaluate HTS for TB diagnosis and drug resistance profiling. Table 1 provides an overview of selected studies, detailing the sample types, locations, kits, and platforms used, effectively serving as a compilation of relevant research benchmarks.
Key Insights and Ideas
- HTS as a Paradigm Shift: HTS moves beyond single-target molecular tests to a more comprehensive approach, offering the potential to identify multiple drug resistance mutations simultaneously from a single clinical sample.
- Direct-to-Sample Advantage: The ability to sequence directly from clinical samples (like sputum) bypasses the time-consuming culture step, significantly accelerating diagnostic turnaround time.
- Whole Genome vs. Targeted Sequencing: WGS provides a complete genomic picture useful for outbreak investigations and detailed genomic epidemiology, but it is often more resource-intensive. tHTS focuses on specific genes known to be associated with drug resistance, offering a more cost-effective and streamlined approach for clinical diagnostics.
- Platform Diversity and Trade-offs: A range of sequencing platforms exist, each with its own strengths and weaknesses regarding read length, error rates, throughput, and cost. The choice of platform is critical and depends on the specific clinical question and available infrastructure.
- Bioinformatics is Crucial: The interpretation of HTS data relies heavily on sophisticated bioinformatics pipelines. Developing and maintaining these pipelines, especially in resource-limited settings, is a significant challenge.
- Potential for Global Health Equity: As HTS costs decrease and portability increases, it holds immense promise for improving TB diagnostics in high-burden, low-resource countries, potentially supplanting or augmenting existing methods.
Key Results
The paper is a review, so it doesn't present novel quantitative results. However, it synthesizes findings from numerous studies, indicating:
- High Accuracy: Studies reviewed show that HTS, particularly tHTS, can achieve high sensitivity and specificity (often >80% sensitivity and >95% specificity for common drugs) in detecting drug resistance mutations compared to gold-standard methods.
- Rapid Turnaround: HTS workflows can provide results within 1-2 days, a significant improvement over the weeks required for traditional culture-based methods.
- Comprehensive Drug Resistance Information: HTS can identify mutations associated with resistance to multiple first- and second-line TB drugs, offering a more complete picture than current rapid molecular tests.
- Geographic Coverage: Most studies reviewed have been conducted in Africa, with some in Europe and Asia. There is a notable absence of studies from Latin or South America.
Limitations and Future Work
Limitations:
- Bioinformatics Infrastructure: The need for significant bioinformatics infrastructure and expertise remains a major barrier to widespread implementation, especially in low-resource settings.
- Cost: While decreasing, the cost of HTS can still be a limiting factor compared to simpler diagnostic tests.
- Assay Updates: Targeted assays need continuous updating to incorporate newly discovered resistance mutations.
- Clinical Outcome Data: While HTS can predict resistance, more research is needed to correlate these predictions with actual clinical outcomes.
- Geographic Gaps: The limited geographic representation of studies highlights a need for more research in underrepresented regions.
Future Work:
- Automated and User-Friendly Pipelines: Development of more automated and accessible bioinformatics pipelines is crucial.
- Integration into Clinical Practice: HTS is poised to become an adjunct to, or even a replacement for, current diagnostic methods.
- Broader Drug Coverage: Assays should aim to cover all anti-TB drugs.
- Mixed Infection Detection: Improving the depth of coverage to identify mixed infections is important.
- Lineage Information and Strain Typing: Utilizing HTS for surveillance and understanding TB epidemiology.
- Exploring New Sample Types: Further investigation into the use of samples like stool for TB diagnosis.
- Cost Reduction and Portability: Continued efforts to reduce costs and improve the portability of HTS technologies for implementation in resource-limited settings.
Summary updated