Individualized Treatment of Multidrug-resistant Tuberculosis Using Whole-Genome Sequencing and Expanded Drug-Susceptibility Testing

A case of multidrug-resistant tuberculosis is presented. It highlights the role of whole-genome sequencing, expanded phenotypic drug susceptibility testing, and enhanced case management, offering a more complete understanding of drug susceptibility to Mycobacterium tuberculosis . This approach guides an effective individualized treatment strategy that results in rapid sustained culture conversion.

Multidrug-resistant tuberculosis (MDR-TB) treatment has been recently transformed with the introduction of the standardized short-course (SSC) regimens as the preferred option for treatment [1]. MDR-TB is a global health crisis affecting approximately 500 000 individuals annually and is considerably more difficult to treat than drug-susceptible TB disease. Given the magnitude of the disease, the recommended treatment in high-burden settings remains standardized empirical combination regimens. This has been largely driven by the need to scale-up treatment provision coupled with limited access to laboratory-based drug-susceptibility testing (DST) [1,2]. Rapid molecular-based DST such as Xpert MTB/RIF Ultra (Xpert; Cepheid, USA) and MTBDRplus and MTBDRsl line probe assays (LPAs; Hain Lifescience GmbH, Nehren, Germany) provides information on a small selection of key drug-resistance mutations. Hence, treatment regimens may contain ineffective and potentially toxic drugs [3]. The recent arrival of the new TB drugs bedaquiline (BDQ) and delamanid holds significant promise in improving drug-resistant TB (DR-TB) treatment success. Clinical trials are currently underway to evaluate the use of these new drugs in various regimens, with the goal of creating shorter, injectionfree standardized regimens. These drugs have been incorporated into current regimens since they became available [4]. As a consequence, early reports of resistance to both of these agents are emerging [5].
A compelling alternative, presented in the case reported here, is to individualize therapy based on whole-genome predictions of susceptibility. Whole-genome sequencing (WGS) technology for Mycobacterium tuberculosis has advanced significantly and has progressed from the research arena to clinical application for diagnosis and management of DR-TB. Leveraging on the advances in WGS technology coupled with enhanced case management by a team of DR-TB clinicians, we portray how the technology can be used to provide personalized care for patients with DR-TB. This approach could potentially impact the prognosis and outcomes of the disease.

CASE REPORT
A 41-year-old male was referred in March 2019 to the specialist DR-TB referral hospital in KwaZulu-Natal, South Africa, with pulmonary rifampicin-resistant TB, diagnosed with the Xpert Ultra assay. The patient was subsequently enrolled into the effectiveness of individualised multi-(extensively) drug-resistant tuberculosis treatment study (CAPRISA 020 InDEX study) and randomized to receive individualized treatment based on WGS of the cultured M. tuberculosis isolate [6]. The patient presented with classic clinical features associated with active TB, which included a 2-week history of cough, night sweats, chest pain, weight loss, and poor appetite. Chest radiography indicated consolidation in the right upper lobe and bilateral infiltration of lower zones. His past medical history was notable for a previous episode of drug-susceptible TB in 2004, for which he completed 6 months of treatment. He was diagnosed with human immunodeficiency virus in 2004 and commenced on antiretroviral treatment (ART) since his diagnosis. On presentation, his CD4 T-cell count was 172 cells/μL and viral load was <150 copies/ mL. On enrollment into the study, MDR-TB was confirmed by the MTBDRplus LPA performed on the sputum sample, demonstrating resistance mutations in rpoB and the inhA promoter region. No mutations were detected on the MTBDRsl LPA, indicating susceptibility to the fluoroquinolones and second-line injectable drugs. He was initiated on a standard, injection-free regimen that contained BDQ (400 mg; loading dose for 2 weeks, 200 mg; 3 doses per week), linezolid (LZD; 600 mg daily), isoniazid (INH) high-dose (INH-HD; 900 mg daily), levofloxacin (LFX; 1 g daily), clofazimine (CFZ; 100 mg daily), pyrazinamide (Z; 1.25 g daily), and ethambutol (E; 1.2 g daily), indicated for 9 months of treatment [7]. His initial ART regimen, which was comprised of tenofovir (300 mg),  negative. Figure 1 contains the complete laboratory profiling data (WGS profiling and extended DST) of the patient's clinical isolate and chest radiographs.

DISCUSSION
We report on a case of MDR-TB treated using an individualized treatment strategy based on WGS prediction of drug susceptibility to the infecting organism. The case highlights the challenges associated with the lack of appropriate diagnostics to guide the use of the standard MDR-TB regimen. The application of WGS, expanded phenotypic DST, and enhanced case management revealed that the patient was on a regimen that contained 3 drugs with confirmed susceptibility. This included BDQ, LFX, and CFZ, as LZD was discontinued after 1 month of treatment due to anemia. Given the high background burden of resistance to E and Z, with approximately 50% of MDR-TB patients with documented resistance to Z and 61% to E [11], replacement of these agents with similar sterilizing agents is warranted. This highlights that careful selection of effective companion drugs is warranted as these potentiate the activity of core drugs promoting relapse-free cure. Further, the use of an SSC regimen should be closely monitored, especially when 1 or more of group A drugs are clinically contraindicated, potentially compromising the regimen. In this case, modification of the drug regimen was required because the patient developed anemia. DR-TB treatment is rapidly transitioning into shorter, injection-free standard regimens that include the novel 6-month combination of BDQ, LZD, and pretomanid for extensively DR-TB and complicated cases of MDR-TB, as well as various novel combinations currently under evaluation [12]. The current case highlights that the eligibility criteria for these regimens require careful consideration and should ideally be guided by individual-level DST and clinical profiling. In this report, WGS and expanded DST were used as diagnostic adjuncts, demonstrating inadequacy of the novel injection-free BDQ-containing regimen. A significant number of DR-TB patients now receive BDQ-containing treatment regimens; however, as a result of increased use of the drug, there are emerging reports of acquired BDQ resistance with cross-resistance to CFZ [5]. This has important implications for the implementation of novel regimen combinations. At present, no rapid assay has the capability of detecting resistance to BDQ, LZD, CFZ, and cycloserine, which are key components of DR-TB [13]. While robust sequencing technology is available, elucidation of the genetic basis for resistance to the newer drugs remains limited and thus warrants confirmation by phenotypic DST in the interim, especially in cases with prior exposure to these drugs.
In conclusion, the case highlights the shortfalls of current diagnostic platforms in guiding DR-TB treatment. Initiation of treatment based on the standard diagnostic pipeline may inadvertently result in patients receiving suboptimal treatment, amplify resistance, and increase the risk of DR-TB transmission. Large-scale studies to assess the role of diagnostic adjuncts such as WGS, targeted sequencing panels, and expanded phenotypic assays are urgently needed in order to determine adequate treatment selection and personalized care approaches as highlighted in the described case. We strongly support the recommendations by Dowdy et al to address the policy gap regarding implementation of standard regimens. DST should be conducted with at least the core drugs such as BDQ and LZD, and no patient should be continued on the regimen for more than 2 months without documented susceptibility to these agents. Further, prior to the continuation phase, patients should have documented susceptibility to the fluoroquinolone and other key drugs carried into the continuation phase [14]. While standardized treatment approaches improve treatment access at a reduced cost and complexity, we cannot ignore the potential risk of resistance amplification.