Background
Drug resistance is a major threat to global tuberculosis (TB) care and control. WHO estimates that around 480,000 new multidrug-resistant tuberculosis (MDR-TB)1 cases occured in 2013. Current treatment regimens for drug-resistant TB are complex, lengthy, toxic and expensive. Only about one half of MDR-TB patients started on treatment globally are reported to be treated successfully, largely due to a high frequency of death and loss to follow-up, commonly associated with adverse drug reactions and high costs of treatment. In addition, it is estimated that up to a third of MDR-TB cases may have strains with additional resistance to fluoroquinolones and/or injectable second-line drugs (aminoglycosides or capreomycin), rendering their treatment even more difficult, with recourse only to highly toxic drugs.
The landscape of drug development for treatment of TB has evolved over the past ten years and novel drugs are presently or soon entering Phase III trials for the treatment of MDR-TB. Considering the global MDR-TB crisis, the limited therapeutic options available for this life-threatening condition, and the need to promote safe and responsible use of TB drugs, WHO convened an Expert Group in April 2014 to review the available evidence on the efficacy, safety and effectiveness of delamanid, a new drug for the treatment of MDR-TB, with the view to issue interim recommendations on its use in conjunction with WHO-recommended MDR-TB treatment.
Evidence assessment
Data on the pre-clinical and clinical development of the drug provided by the manufacturer, as well as publicly available data were reviewed to assess efficacy, safety and tolerability of the drug. In addition, modeling work to assess the potential cost-effectiveness of programmatic implementation was commissioned to an independent expert. The comprehensive review of these data was conducted according to the GRADE process for evidence assessment, as required by WHO.
Data on delamanid efficacy and safety resulted from three studies:
Trial 204: A phase 2, multicentre, double-blind, randomized, placebo-controlled clinical trial conducted in nine countries.
2 A total of 481 patients aged 18 to 64 were randomized to receive two months of treatment with either delamanid 100mg twice daily (BD) on top of an optimized background regimen (OBR), delamanid 200mg twice daily+OBR, or placebo+OBR;
Trial 208:
2 An open-label extension of Trial 204 that allowed continued or first-time access to delamanid in combination with OBR for an additional six months for patients who completed Trial 204 and consented to participate. A total of 213 (44.2%) of the 481 patients from Trial 204 were enrolled in this study;
Study 116:
3 An observational study which captured the long-term treatment outcomes for patients who participated in Trial 204 and Trial 208. 421 patients who initially participated in Trial 204 were included, and 390 completed the 24 months follow-up.
Summary of results
Short-term efficacy: Data for efficacy analysis was provided by the manufacturer for various time periods and from a variety of patient populations. The primary efficacy endpoint (short-term efficacy) was two-month sputum culture conversion (SCC) as derived from Trial 204. Efficacy analysis was based on a modified intention to treat (MITT) population (N=402/481 patients) and involved randomized patients who had a positive sputum culture for TB and drug-resistance to isoniazid and rifampicin confirmed phenotypically at baseline, using the mycobacteria growth indicator tube (MGIT) liquid culture system. A higher proportion of patients treated with delamanid 100mg BD (the dose recommended by the manufacturer)+OBR achieved SCC at two months (Day 57) than patients treated with placebo+OBR: 64/141 (45.4%) vs. 37/125 (29.6%); p=0.008. The hazard ratio for time to conversion to a negative sputum culture was 0.58 (95% CI: 0.39 to 0.89) in the delamanid-100mg BD group.
Long-term efficacy: Data for efficacy beyond the first two months of treatment was provided by the manufacturer through analysis of combined data using solid culture from Trial 204, Trial 208 and Study 116, grouping patients according to the total duration of delamanid received in the various trials, irrespective of the dose received (100mg BD or 200mg BD). By the end of treatment, 90.9% (130/143) of patients who received delamanid+OBR for six months or more (≥6 months) achieved sustained SCC compared to 70.9% (112/158) of patients who received delamanid+OBR for two months (≤2 months) or less.
The proportion of patients reported with a favourable treatment outcome at the end of the 24-month treatment period (i.e. confirmed microbiological cure on solid culture or treatment completion) was significantly higher in the ≥6 month group (74.5%, 95% CI: 67.7–80.5) than in the ≤2 month group (55.0%, 95% CI: 48.3–61.6) (p<0.0001). The reported mortality rate was lower in the ≥6 month treatment group (2 deaths, 1%) compared to the ≤2 month treatment group (19 deaths, 8.3%) (p<0.001).
Safety: Pooled data provided by the manufacturer from 12 phase I trials in healthy subjects showed that treatment-emergent adverse effects (TEAEs) with the highest incidence in persons who received delamanid were headache, nausea and dizziness.
A total of 887 individuals have been exposed to delamanid in all trials, of which 22.1% (196/887) had a cumulative exposure longer than 6 months. The only clinically relevant TEAE with a difference in incidence between the delamanid+OBR treatment groups compared to the placebo+OBR group was prolongation of QT interval. Other TEAEs varied in occurrence but were present in similar proportions in the delamanid+OBR and the placebo+OBR groups. Most frequent were nausea, vomiting, and dizziness.
A total of 74 patients in Trial 204 and Trial 208 reported severe adverse events (SAEs). QTc interval prolongation occurred more frequently in the delamanid 100mg BD+OBR group (4.3%; 7/161) and the delamanid 200mg BD+OBR group (5.6%; 9/160) than in the placebo+OBR group (1.9%; 3/160).
Drug-drug interaction studies in healthy subjects showed no clinically-significant interactions when delamanid was co-administered with tenofovir, efavirenz or lopinavir/ritonavir.
Expert Group findings and recommendations
The Expert Group felt strongly that the main data to consider for evaluation of delamanid efficacy were those collected in the randomized controlled Trial 204. They were not convinced that the prognostic surrogates for cure used (i.e. two months SCC and time to SCC) were adequate or accurate for MDR-TB. They also noted that the duration of comparison between delamanid and placebo in Trial 204 did not allow an assessment of the potential benefit of adding delamanid to OBR for six months without interruption (as recommended by the manufacturer).The Expert Group noted a number of problems in the design and conduct of Trial 208, including absence of randomization; self-selection of patients; absence of blinding for treatment allocation; variable gaps between the end of Trial 204 and the beginning of Trial 208; and lack of consistency in the allocation of drug dosages (leading to wide variations in the dose, timing and duration of exposure to the drug for various subgroups of patients). There was also concern about risk of serious bias for the data arising from the observational Study 116 due to variability in follow-up procedures in the various sites, as well as un-blinded assessment of outcome, and serious inconsistency due to the variability in duration of drug exposure. They also agreed that the post hoc analysis of efficacy at the end of treatment (Study 116) performed by the manufacturer had little value for the evaluation of effect, since this was a retrospective analysis relying on non-standardized follow-up procedures and conditions, and on artificial grouping of patients with wide variability of drug exposure in terms of timing, duration and doses of treatment.
The Expert Group therefore requested re-analyses of data for the estimate of effect beyond the two-month SCC, considering (i) only patients included in Study 116 who received delamanid from the start of treatment in Trial 204 and continued to receive delamanid during Trial 208, as this was felt to be more consistent with the manufacturer's instruction to use delamanid for 6 months (in addition to OBR) from the start of treatment; and (ii) using as controls only those patients who did
not
receive delamanid at all in either trial. The revised analyses showed a 35% increase in cure (according to WHO definition) when delamanid was added to an OBR vs. OBR alone (RR 1.35; 95% CI 1.03 – 1.63).
Of note, 4/205 patients seemed to have developed in vitro resistance to delamanid during treatment, although no discernible reasons could be established.
Final conclusion on efficacy: The Expert Group concluded that data on short-term or long-term efficacy of delamanid added to an OBR for MDR-TB were of ‘very low’ quality, i.e. the Expert Group had very low confidence in the estimate of effect of delamanid.
Final conclusion on safety: The Expert Group noted that the risk of any adverse event in the delamanid treatment arms in Trial 204 was not significantly different than in the placebo arm. They noted, however, that prolongation of the QT interval was the most serious adverse event, with a significantly higher risk of QT prolongation relative to baseline in the delamanid treatment arms compared to the placebo arm. Safety risks for patients receiving at least six months of treatment with delamanid+OBR compared to OBR alone could not be assessed in Trial 208 due to the absence of proper controls. The Expert Group also noted that there was no information on the potential synergy of cardiotoxic effects if delamanid was used in combination with other drugs which also prolong the QT interval (such as moxifloxacin); therefore, the overall evidence for safety was graded as ‘low’, despite the clear difference in mortality observed when delamanid was added to an OBR.
Cost-effectiveness: The Expert Group assessed the results of a cost-effectiveness analysis (CEA) conducted to model the incremental cost-effectiveness of adding delamanid to existing WHO-recommended MDR-TB regimens. This CEA was undertaken for various settings to allow for variation among countries in income level, the model of care used for MDR-TB treatment, and background patterns of drug resistance. It focused on the direct benefits to patients, but did not attempt to assess the indirect (and acquired) transmission benefits, nor did it assess the broader economic benefits to patients or society. Since several analyses were conducted by the manufacturer to assess efficacy (see above), a sensitivity analysis was performed on the cost-effectiveness of delamanid applying the different trial data and respective assumptions. Using a conservative approach, delamanid was found to be cost-effective in most settings, but the quality of this evidence was considered ‘very low’, and the Expert Group concluded that further work was needed to evaluate cost-effectiveness once the final price of the drug is made public by the manufacturer.
Overall, balancing the potential benefits and risks of adding delamanid to an optimised MDR-TB regimen, the Expert Group concluded that the anticipated benefits probably outweighed anticipated undesirable effects. Therefore, the Expert Group recommended that delamanid (100mg BD for 6 months) may be added to a WHO recommended regimen in MDR-TB adult patients under specific conditions (conditional recommendation, very low confidence in estimates of effect).
WHO interim policy recommendations
Available data on delamanid efficacy and safety is very limited as assessed by the GRADE process; however, the overall benefits of the inclusion of delamanid in a WHO-recommended MDR-TB regimen appear to outweigh the observed harms. Therefore, considering the global MDR-TB crisis, the limited therapeutic options available for this life-threatening condition, and the need to promote safe and responsible use of TB drugs, WHO is making the following interim policy recommendation for the use of delamanid in the treatment of MDR-TB:
WHO recommends that delamanid may be added to a WHO-recommended regimen in adult patients with pulmonary MDR-TB (conditional recommendation; very low confidence in estimates of effect).
In view of the insufficient experience with the use of delamanid under the different conditions that may be expected in treatment programmes, and the uncertainty about its overall added value in the treatment of MDR-TB patients, WHO recommends that the use of delamanid in the treatment regimen of MDR-TB be made subject to the following five conditions:
1. Proper patient inclusion
The current recommendation for the use of delamanid applies to adults (≥18yrs) with pulmonary MDR-TB disease, including people living with HIV. Special caution and proper clinical judgment should be applied when delamanid is used in persons 65 years and older, or in those with diabetes, hepatic or severe renal impairment, or those who use alcohol or substances, given that data on efficacy and safety under such conditions are extremely limited or unavailable.
Use of the drug in children and in pregnant and breastfeeding women is not currently advised due to a lack of evidence on safety, efficacy and proper dosing in these groups.
Because delamanid is shown to cause prolongation of the QT interval, patients with a QTcF>500ms should not receive the drug.
When an effective and reasonably well-tolerated MDR-TB regimen can be composed with conventional second-line drugs, the routine addition of delamanid may not be warranted and the implications of additional health service costs should be considered. MDR-TB patients in whom delamanid may have a particular role include those with:
higher risk for poor outcomes (eg. drug intolerance or contraindication, extensive or advanced disease);
additional resistance to fluoroquinolones or injectable drugs;
XDR-TB (see 3.b for additional measures to apply when the drug is used in XDR-TB patients).
While patients with exclusive extrapulmonary disease were not included in the delamanid trials, there is no absolute contraindication for its use in such patients and inclusion may be considered where any potential harm that delamanid may cause is offset by the benefit expected.
2. Adherence to the principles of designing a WHO-recommended MDR-TB regimen
Delamanid is intended to be introduced alongside other anti-TB drugs in composing an effective second-line regimen based on WHO guidelines; the cardinal rules governing the general composition and duration of MDR-TB regimens remain the same:
The WHO-recommended MDR-TB treatment regimen (
1) is typically composed of at least pyrazinamide and four second-line drugs considered to be effective (based on drug susceptibility testing (DST) and/or previous use and/or drug resistance surveillance data): a fluoroquinolone (preferably later-generation), a second-line injectable agent, and two bacteriostatic drugs, preferably prothionamide or ethionamide plus cycloserine or
p-aminosalicylic acid.
MDR-TB patients with confirmed resistance or intolerance to either fluoroquinolones or the second-line injectable drugs represent a particular treatment challenge. In such cases, delamanid may have a crucial role to play in strengthening a regimen, bringing the number of drugs likely to be effective to a minimum of four, and reducing the risk of acquisition of additional resistance and progression towards XDR-TB.
There is as yet no standardized DST method for delamanid, nor a commercially available test. DSTs for second-line drugs other than fluoroquinolones and injectables (kanamycin, amikacin, capreomycin) are not accurate or reproducible, and MDR-TB patients may respond poorly to treatment for reasons other than drug resistance. A change in medication may, therefore, have to be based on persistence of positive sputum culture, or reversion to positive following initial culture conversion rather than DST.
While experience in the use of delamanid in the management of XDR-TB is very limited, there may be a benefit given the limitations in designing an effective regimen. In such patients, delamanid may lower the need to include other drugs belonging to Group 5 which have unproven anti-tuberculosis activity or a lower safety profile. However, special caution is necessary when delamanid is used with a fluoroquinolone or a Group 5 drug given the potential for synergistic drug-drug interactions effects, particularly on QT prolongation.
4There are currently no data on the simultaneous use of bedaquiline and delamanid in the same patient. Until such data become available, no recommendation on the joint administration of these two medicines is possible within the scope of this interim guidance.
In line with general principles of TB therapeutics, delamanid should not be introduced into a regimen in which the other companion drugs are known or believed to be ineffective, or are failing to show effectiveness. This means that delamanid should not be added alone to a failing regimen. Given the emergence of resistance to delamanid observed in the available data, all possible measures should be taken to protect the efficacy of the drug.
The recommended dose of delamanid in adults is 100mg twice a day, irrespective of body-weight, for a period of six months. As bioavailability was higher when given after a standard meal, delamanid should preferably be delivered after a meal. There was no evidence that delamanid 200mg twice a day was more effective than the 100mg dose and the higher dose was associated with higher rates of adverse events including QT interval prolongation. It should be particularly noted that supervision of delamanid intake should be adapted to twice a day.
3. Treatment is closely monitored
Adherence to best practices when administering treatment is imperative to ensure optimal drug effectiveness and safety. It is therefore recommended that the following measures are in place:
Sound treatment and management protocols, including clear patient eligibility criteria, locally appropriate procedures for informed consent (
see
5), and defined roles and responsibilities of all professionals involved. Safety concerns are best addressed through active pharmacovigilance (
2).
The treatment protocols should allow for the prospective capture of data on key variables for both effectiveness and safety, making sure that the good practices, such as those applied in the conduct of observational studies, are adhered to (
3,
4).
Treatment protocols are preferably submitted to, and approved by the relevant national ethics authority in the country prior to patient enrolment on treatment.
Preferably, oversight of treatment programmes is provided by an independent group of experts in clinical management and public health (e.g. a national MDR-TB advisory group).
The potential for emergence of delamanid resistance during the course of therapy requires that all measures to enable patient's adherence are in place before starting treatment.
4. Active pharmacovigilance and proper management of adverse drug reactions and prevention of drug–drug interactions
Alongside the measures in 3. above to monitor treatment adherence and effectiveness, special vigilance is needed for adverse events, including potential reactions to delamanid which are as yet undescribed.
Given that the results of Phase III trials are expected in the next few years, it is particularly important that the introduction of delamanid is accompanied by an enhanced monitoring for adverse events. For this purpose, spontaneous reporting is not expected to represent an appropriate level of care and active pharmacovigilance techniques, such as ‘cohort event monitoring’ (CEM), will be needed to improve the early detection of adverse drug reactions. Details on the methodologies for mounting CEM, particularly when new drugs are introduced, have already been published by WHO (
2).
Any adverse drug reaction attributed to delamanid should be reported to the national pharmacovigilance centre. As for any other drug in an MDR-TB regimen, the patient should be encouraged to report to the attending health worker any adverse event that occurs during the time the drug is being taken. Such occurrences should also trigger a rapid response to manage these untoward effects in the patient.
When introducing delamanid into a regimen, there is also the potential for its interaction with other medications administered concurrently, with additive or synergic adverse effects. Other second-line drugs that are likely to be administered with delamanid, notably fluoroquinolones and clofazimine, may potentially increase the risk of cardiotoxicity. Although there are data showing QT interval prolongation when delamanid is administered simultaneously with levofloxacin, no data are available on concomitant use with moxifloxacin and/or clofazimine. Also, some antiretroviral medications can cause modest QT prolongation, especially ritonavir-containing regimens. Therefore, monitoring of patients for cardiac dysrhythmias or QT interval prolongation (i.e. using ECG), and for electrolyte imbalances (especially serum potassium) that can predispose to cardiotoxicity is imperative.
5Drug-drug interaction studies of delamanid with tenofovir, efavirenz and lopinavir/ritonavir, respectively, conducted among healthy individuals who did not have HIV or TB, suggested that no dose adjustments were needed when delamanid was used with any of these anti-retroviral agents. However, there is no published evidence so far on the use of delamanid in HIV-infected MDR-TB patients on ART. Therefore, people living with HIV who will be receiving delamanid as part of MDR-TB treatment should have their ART regimens designed in close consultation with HIV clinicians and ART specialists.
Lastly, caution is advised in patients with pre-existing health conditions that may be exacerbated or worsened by delamanid. Currently there are no data on the efficacy and safety of delamanid in patients with co-morbid conditions such as diabetes, liver and/or renal dysfunction, malignancies, alcohol and substance use, and therefore careful screening for these conditions prior to treatment initiation is advised. Hypersensitivity reactions to delamanid have not yet been described, but vigilance is nevertheless required.
5. Patient informed consent obtained
Health care workers should follow a due process for informed consent by ensuring that the patient: i) is aware of the novel nature of delamanid; ii) appreciates the reason why the drug is being proposed to be included in their treatment regimen; and iii) recognizes the possible benefits and potential harms, including the uncertainties that surround outcomes. This informed consent process applies to all situations where delamanid is employed, including under compassionate use programmes. In some settings, as per national or local policy, it is required that the informed consent is made in writing for enrolment on MDR-TB treatment.
Validity of the interim policy guidance
This interim recommendation is valid for a maximum of two years and will be updated should additional data become available.
- 1
Multidrug-resistant tuberculosis: tuberculosis with resistance to, at least, isoniazid and rifampicin.
- 2
China, Egypt, Estonia, Japan, Latvia, Peru, the Philippines, Republic of Korea, and the United States of America.
- 3
Trial 208 and Study 116 are extensions of Trial 204. They are referred to by the terminology used in the published references and as provided by the manufacturer; it should be noted that these are not different/separate studies, but involved non-randomized and selected patients from the original Trial 204.
- 4
A QTcF value greater than 440ms is considered prolonged. A value greater than 480ms (or an increase of greater than 60ms from baseline) should trigger electrolyte testing and more frequent ECG monitoring. A QTcF interval of more than 500ms is considered dangerous and stopping QT-prolonging drugs is indicated.
- 5
It is imperative that ECGs are used to monitor the QT interval regularly during delamanid use. QT interval monitoring should preferably be done using ECG machines that directly report the QTc interval. A value greater than 440 ms is considered prolonged. A value greater than 480 ms (or an increase of greater than 60 ms from baseline) should trigger electrolyte testing and more frequent ECG monitoring. A QTc interval of more than 500 ms is considered dangerous and should lead to stopping of the intake of the responsible QT prolonging drug(s).
