6.3.1 Choice of components for the longer MDR-TB regimens

A stepwise approach guides the design of longer MDR-TB regimens (Table 6.1 and Table 6.5). The treatment of patients with rifampicin mono-resistant TB, as well as those who have resistance to second-line agents in addition to MDR-TB (including XDR-TB), follows the same principles. The selection of agents follows a priority order based on the revised classification of regimen components, and a fully oral regimen is preferred

The analyses conducted for the WHO consolidated guidelines support the current recommendation that most patients can be successfully treated with a regimen starting with four agents that are likely or confirmed to be effective. If bedaquiline is stopped at month 6, the regimen will still have three effective agents for the rest of the treatment duration. However, if another agent needs to be stopped because of toxicity, then that medicine would need to be replaced by another one, or bedaquiline could be continued throughout the treatment under “off-label” use. If the choice is to replace a medicine, instead of prolonging the use of bedaquiline, the replacement medicine would be chosen either from Group B (unless both clofazimine and cycloserine/terizidone are already included) or from Group C. The choice from Group C is usually determined by the order in which the medicines are ranked, and by the individual circumstances of the patient and setting. Recent review of the observational data has shown no additional safety concerns when bedaquiline was used for longer than 6 months; however, no solid evidence was available to indicate whether longer use added efficacy. The clinicians may therefore consider continuing bedaquiline for longer than 6 months, and adding some flexibility for regimen design and the number of effective drugs (1).

To minimize the need to replace agents in the regimen, in addition to the option of prolonging the use of bedaquiline beyond 6 months, it is possible to start the regimen with five agents instead of four. This increases the pill burden and chances of adverse reactions, but may be justified, particularly when:

• two of the four agents are likely to be stopped before the end of treatment (e.g. if bedaquiline is stopped at month 6 and linezolid is stopped early because of toxicity);

• reliable DST is not available for one or more of the agents in the regimen but background resistance to the agent is known to be high; and

• the regimen cannot be constructed from at least four effective agents from Groups A and B.

Regimen composition may often need to be adjusted after the start of treatment once additional information from the clinical history or DST results emerges. However, if signs of non-response or impending treatment failure emerge, then the regimen should be reviewed completely rather than adjusted. A medicine may be avoided if there is high likelihood that the patient has developed, or will develop, a contraindication to it. Contraindications may depend on a history of severe reactions to the medicine or an allied substance, pregnancy or breastfeeding, co-administration of medicines that may cause interactions or have overlapping toxicities (e.g. QT interval prolongation) and problems with end-organ function (e.g. kidney or liver dysfunction). Box 6.2 outlines some of the key considerations when choosing individual medicines for a longer MDR-TB regimen.

Other tests for resistance to agents such as pyrazinamide, and for mutation patterns commonly associated with resistance to isoniazid and the thioamides, may help to inform the composition of the regimen. Currently, there is no validated rapid test for pyrazinamide susceptibility, and phenotypic testing may require several weeks to produce a reliable result; a decision to include or replace pyrazinamide should not delay the start of treatment.

Resistance beyond MDR-TB, including MDR-TB with fluoroquinolone resistance, reduces the likelihood of treatment success (79, 80). The risk of MDR-TB with fluoroquinolone resistance is much higher in some settings (e.g. in Eastern European countries). Untreated, MDR-TB with fluoroquinolone resistance has a high mortality, particularly in PLHIV, and strains that have acquired resistance have shown little if any attenuation in transmissibility; such strains have been implicated in extensive outbreaks, with much of the global burden attributed to primary transmission (81–83).

With the lowered importance of aminoglycosides in MDR-TB treatment regimens, the utility of diagnosis of an MDR-TB with injectable resistance, as currently defined, has lost its value. However, resistance to fluoroquinolones remains an important finding for all regimens. The steps in Table 6.1 apply also to the design of a fluoroquinolone resistance regimen for MDR-TB; in addition, they cover MDR-TB with fluoroquinolone resistance among other resistance patterns.

 choosing individual medicines

Many of these patients may have comorbidities and adverse events that need to be addressed separately. Hospitalization, surgery and other adjuvant treatment may be needed at certain stages of treatment. Comprehensive monitoring and psychosocial support are important to ensure a more favourable treatment experience. Access to palliative and end-of-life care services may be needed, with a patient-centred approach to relieve the suffering of the disease and its treatment. Rigorous respiratory infection control measures at the sites where the patient is being treated, contact tracing and counselling are important accompanying measures for clinical care and public health.

Table 6.5 summarizes some common situations that a clinician may face, and the decisions that could be taken to adjust the treatment regimen accordingly. The suggested regimens may vary based on the individual clinical circumstances and the availability of medicines. Table 6.5 is not exhaustive. Although it is recommended to use at least four effective agents initially, not all the regimens composed using this algorithm have been tested directly in either research or field conditions. Moreover, when Group C agents are included, the number of medicines in the regimen may exceed four, to reflect the uncertainty about the efficacy of some of these medicines. In such situations, the advice of a specialist is important to ensure the safest and most effective possible regimen (note: content is sometimes repeated in the table given an overlap between the different scenarios).

 algorithm for longer MDR-TB regimen


 algorithm for longer MDR-TB regimen

Bdq: bedaquiline; Cfz: clofazimine; Cs: cycloserine; Dlm: delamanid; E: ethambutol; Lfx: linezolid; m: months; Mfx: moxifloxacin; MIC: minimum inhibitory concentration; TB: tuberculosis; Z: pyrazinamide.

Situations shown are not exhaustive. Other factors may influence choice, such as patient risk for poor outcome or drug–drug interactions, clinician and patient preference and availability of a medicine. More medicines may be added than the recommended minimum if there is limited confidence in the effectiveness of regimen components, if the patient was exposed in a setting where second-line TB drug resistance is frequent and longer MDR-TB regimens perform poorly despite good programmatic management of MDR/RR-TB. For MDR-TB with confirmed fluoroquinolone resistance no fluoroquinolone is used and, if Group C agents are needed, the recommended WHO grouping will be followed based on benefit versus risk and individual circumstances.

The choice and number of Group C medicines to include depends upon the confidence in the effectiveness of medicines in this group and the other components of the regimen, thus:

If 4 Group A and B agents are included and there is confidence in all of them then Group C agents are not needed.

If 3 Group A and B agents are included and there is confidence in all of them then at least one Group C agent is added.

If 2 Group A and B agents are included and there is confidence in all of them then at least three Group C agents are added.

Regardless of resistance or contraindication for Group C medicines.

Moxifloxacin, a later-generation fluoroquinolone, may still be effective at high dose when the fluoroquinolone MIC is below the clinical breakpoint. If the MIC is elevated, then fluoroquinolones are not used, and additional Group C agents will be needed.

Book navigation