This week, the final results of STREAM Stage 1 were published in the New England Journal of Medicine. Results show that the 9-11-month multidrug-resistant tuberculosis (MDR-TB) treatment regimen being tested as part of the STREAM clinical trial is non-inferior to the 20-24-month regimen recommended in the 2011 WHO guidelines, in terms of efficacy. STREAM is the world’s first multi-country randomized phase III clinical trial to test the efficacy, safety and economic impact of shortened MDR-TB treatment regimens.

STREAM pharmacist prepares medication at the Helen Jose Tuberculosis Hospital, Johannesburg, South Africa.

MDR-TB, defined as TB that is resistant to at least the two first-line antibiotics isoniazid and rifampicin, affected an estimated 485,000 people in 2017 (source WHO 2018 Global Tuberculosis report) and has been declared a public health crisis by the World Health Organization (WHO).

The 2011 WHO-recommended 20-24-month regimen evaluated in STREAM Stage 1 causes significant side effects and the length of treatment makes it hard for both patients and health systems. The regimen has an average treatment success rate of approximately 50 percent when used in many real-world treatment settings.

In response to those challenges, in 2016, the WHO updated its guidelines to recommend a shorter, 9-11-month regimen for most people with MDR-TB under specific conditions. The guidelines acknowledged, however, that the recommendation was based on very low certainty in the evidence.  The STREAM Stage 1 results address that gap.

Based on the STREAM Stage 1 results and other available evidence, in December 2018, the WHO released a further update to the treatment guidelines for multidrug- and rifampicin-resistant tuberculosis modifying their MDR-TB guidelines. These guidelines continue to advise that a 9-11-month regimen is a viable option for national TB programmes.


The results showed that the 9-11-month regimen was statistically non-inferior to the 20-24-month regimen, in terms of efficacy (78.8 percent of assessable participants had a favourable outcome, compared to 79.8 percent in the longer regimen.)

There was no evidence that efficacy results were worse in HIV-infected participants compared to HIV-negative participants.

Until now there has been a lack of strong supporting evidence to underpin MDR-TB treatment guidelines.  The results from STREAM Stage 1 help to fill that gap. The final results show that the trial setting meant more patients successfully completed treatment on the 20-24-month regimen than we know is often the case in real life settings. In routine programmes unable to achieve the high STREAM retention rates, the 9-11-month regimen may actually perform better in comparison to the longer regimen.

Dr I.D. Rusen
STREAM Trial Project Director

We know from programmatic data that the 20-24-month regimen has a number of major drawbacks, including the difficulty of completing such a long treatment, the significant side-effects of the drugs used and poor treatment outcomes. Shorter, more effective and safer regimens are urgently needed. The outcomes in patients coinfected with HIV are particularly important as they suggest that the 9-11-month regimen is no less effective in this patient group than the longer regimen.

Professor Andrew Nunn
STREAM co-Chief Investigator from the MRC Clinical Trials Unit at UCL

Overall there were similar rates of severe side effects between the 9-11 month regimen and the 20-month regimen, but there were differences in the types of side effects caused by the two regimens. The final results show that electrocardiogram (ECG) monitoring was very useful and required throughout treatment for the 9-11-month regimen. This was done throughout STREAM Stage 1, but is likely to be more challenging in routine programme settings.

We know that ECG monitoring throughout treatment is likely to be challenging in most routine programme settings, where access to ECG monitoring is limited. We have the opportunity to try to improve the regimen during the remainder of STREAM Stage 2 to see if we can reduce the need for ECG monitoring throughout treatment. This is just one reason why dynamic clinical trials of this nature are so important.

Professor Sarah Meredith
STREAM Clinical co-Chief Investigator and Professor of Clinical Trials at the Medical Research Council Clinical Trials Unit at UCL


The analysis of the economic burden of MDR-TB led by the Liverpool School of Tropical Medicine in collaboration with University of Warwick and investigators in South Africa and Ethiopia, will be published in due course. For South Africa and Ethiopia, this analysis presents a breakdown of health system data for each regimen from including costs of in-patient stay, laboratory tests, cardiac safety monitoring, medication, staff time, social support, and consumables. More detailed data are presented from Ethiopia on costs of management of Serious Adverse Events (SAE’s) and costs incurred by trial participants, which also give an insight into the financial and social well-being of participants on each regimen.

This is the first phase III trial of TB treatment that includes a within-trial economic evaluation.  The results will be useful for countries and programmes as they decide on whether and how to introduce shorter regimens for treatment of MDR-TB.

Bertie Squire
STREAM co-Investigator and Health Economic Analysis lead


There is still an urgent need to improve the efficacy, and safety of MDR-TB treatment, so research into other shorter regimens to assess their efficacy, safety and cost is vitally important.  STREAM Stage 2 is evaluating an all oral bedaquiline-containing regimen that is potentially as effective and more tolerable than the injectable-containing regimens currently in use.  Results from STREAM Stage 2 have the potential to significantly impact future policy-making around all-oral treatment regiments.

To read the results in full in the New England Journal of Medicine, click here.

For more information on the STREAM Trial, click here.