Good outcomes within the National Bedaquiline Clinical Access programme in South Africa

45th lung health logoPolly Clayden, HIV i-Base

The South African Bedaquiline Clinical Access Programme showed good interim outcomes in people with drug resistant tuberculosis (DR-TB), according to data presented at the 45th Union Conference on Lung Health.

Bedaquiline is a diarylquinoline compound with a novel anti-tuberculosis mechanism of action that inhibits mycobacterial ATP synthase. Janssen Therapeutics manufactures this drug. The FDA approved bedaquiline for the treatment of multi-drug resistant (MDR) TB in December 2012. It was the first new anti-TB drug to be approved by the agency in over forty years.

In December 2012 the South African Medicine Control Council (MCC) approved a national programme to treat selected DR-TB patients with bedaquiline, which began in March 2013.

Dr Ndjeka from the National Department of Health presented results on behalf of colleagues from the programme.

He began the presentation by describing the TB burden in South Africa. He noted that the number of people with TB initiated on treatment was decreasing: between 2009 and 2013 this went from 406, 082 to 332,170, with an 80.9% success rate in 2012. But numbers with MDR-TB initiated on treatment doubled between 2010 and 2013: from 5,313 to 10,719, with a 45% success rate. For extensively drug resistant (XDR) TB treatment success is only between 15% and 20%.

Laboratory-confirmed pre-XDR and XDR TB patients at five approved South African sites received bedaquiline according to predefined selection criteria: > 18 years of age, susceptible to at least three anti-TB drugs with which to construct the background regimen, negative pregnancy test and using contraception. Pregnant or breastfeeding women and patients with serum creatinine grade >1, lipase grade 2, ALT or AST grade 2 or total bilirubin grade >1 were excluded.

The cases were presented to a clinical advisory committee, after which the originator company approved bedaquiline with an optimised background regimen of at least three selected second line anti-TB drugs and the MCC approved the prescription on a named-patient basis.

400 mg bedaquiline was given once daily for two weeks followed by 200 mg three times a week for 22 weeks plus optimised background regimen. The optimised background regimen continued beyond the 22 weeks.

At base line the 91 participants in the programme, included in the interim analysis, were a median age of 34.1 years (IQR 25.7 – 40.9), 56 (60.5%) were male and 55 (60.5%) were coinfected with HIV. The HIV positive group had a median CD4 count of 249 cells mm3 (IQR 134 – 356), 19 (34.5%) received lopinavir/ritonavir-based ART regimens and 36 (65.5%) nevirapine-based.

TB drug resistance patterns were described as: XDR in 33 (36.3%) participants, pre-XDR (fluoroquinalone) in 41 (45.1%) and pre-XDR (injectable) in 17 (18.1%). Optimised background regimens included: clofazimine for 68 (74.5%), linezolid for 64 (70%) and levofloxin for 76 (83.5%).

At the time of analysis, 60 participants had > 24 weeks of follow up, 2 did not complete bedaquiline (1 died, 1 lost to follow up). Of 58 that completed 24 weeks of bedaquiline, 2 died, 1 transferred and 1 was lost to follow up. Of the remaining 54 on continuation treatment, 15 were culture negative at start, 33 culture-converted and 6 were still culture positive. All the 31 participants with <25 weeks of follow up since treatment start were still on treatment. Of this group, 6 were culture negative at start, 10 culture- converted, and 15 participants had not yet received culture results.

One participant developed atrial fibrillation on bedaquiline and was withdrawn from treatment. At start of bedaquiline, median QTcF was 408ms (IQR 390-426): there was a median increase of 8 ms (IQR -12 to 31 at 2 months; 14 had increases of >40ms and 2 had QTc >ms (bedaquiline was temporarily withdrawn in 1 and the other resolved in 24 hours).

Fifteen participants experienced serious adverse events. The general pattern and frequency were considered to be in line with the known safety profile of bedaquiline. There were 3 deaths of which none were related to bedaquiline; 3 QT prolongation probably or possibly related (all resolved); and 3 psychosis/mood disorder/delusion, none related to bedaquiline (all 3 were receiving terizidone, prodrug of cycloserine known to cause mood disorder).

The programme has since been extended to include 12 sites in South Africa.

In October 2014 the MCC approved bedaquiline and it will be used within the national TB programme under strict control.


Bedaquiline added to a background regimen was associated with earlier culture conversion and higher cure rates in its registrational trials, but there were unexplained excess deaths in the bedaquiline arms of these studies.

When the FDA approved bedaquiline for treatment of MDR-TB when an effective treatment regimen cannot otherwise be given this included a black box warning about excess deaths and a requirement to complete a phase III trial.

As a result, many programmes, including in South African, were right to approach the drug with caution, and it is reassuring that further safety events were not reported.

The French compassionate use programme has been similarly reassuring. [2] At 6 months, 28/29 (97%) participants with culture-positive TB achieved culture conversion in a median of 85 days (range 8–235 days). Seven participants (20%) experienced a >60-ms increase in QT interval. This led to bedaquiline discontinuation in 2 (6%) participants. The one death (3%) was not considered related to treatment.


  1. Ndjeka N et al. Safe and effective bedaquiline treatment of drug resistant tuberculosis (DR-TB) within the National Bedaquiline Clinical Access Programme in South Africa. 45th Union World Conference on Lung Health, 28 October – 1 November 2014, Barcelona. The Union/CDC Late breaker session on TB. (PDF)
  2. Guglielmetti L et al. Compassionate use of bedaquiline for the treatment of multidrug-resistant and extensively drug-resistant tuberculosis: interim analysis of a French cohort. Clin Infect Dis. (2015) 60 (2): 188-194, first published online October 15, 2014

Links to other websites are current at date of posting but not maintained.