Multidrug-Resistant Tuberculosis Treatment: 6 Drugs Better than 5


A regimen containing 6 drugs was more likely to cure multidrug-resistant tuberculosis (MDR-TB) than a regimen containing 5 drugs, a 9-country observational study has shown. The study also found that each active drug in the regimen increased the likelihood of a cure by 65%, and inclusion of pyrazinamide doubled the chance of curing MDR-TB, according to findings were published in the December 29 edition ofPLoS Medicine.

[Produced in collaboration with]

The findings suggest that physicians may need to consider using more drugs than current World Health Organization (WHO) guidelines recommend to cure TB in people with multidrug-resistant infections. Current WHO guidelines, issued in 2011, recommend the use of at least 4 drugs that are likely to be effective, plus pyrazinamide, for the treatment of MDR-TB.

The current guidelines recommend that in the absence of drug susceptibility testing, physicians should make a judgment about which drugs are likely to prove active in a treatment regimen for MDR-TB by taking into account known local patterns of drug resistance, the treatment history of the patient, and their known contacts with other people diagnosed with MDR-TB.

Based on the evidence available at the time, the WHO guidelines panel concluded that selection of drugs on the basis of the drug susceptibility testing resulted in a marginally greater benefit. Subsequent studies have shown that regimens chosen based on drug susceptibility testing have improved outcomes, but have not quantified how many active drugs are needed to overcome MDR-TB.

The results of the 9-country study show that including drugs in the regimen on the basis of guesswork was only effective if at least 3 of those drugs were in fact active against the patient’s specific strain of MDR-TB.

The prospective observational study -- the Preserving Effective Tuberculosis Treatment Study (PETTS) -- was conducted between 2005 and 2010. The study recruited 1659 adults receiving MDR-TB treatment in Estonia, Latvia, Peru, Philippines, Russian Federation, South Africa, South Korea, Thailand, and Taiwan.

The analysis of drug response excluded patients with extensively drug-resistant TB (XDR-TB), those lacking drug sensitivity test results, and those with sensitivity to either isoniazid or rifampicin. The analysis comprised 1137 eligible patients, followed for a median of 20 months after beginning MDR-TB treatment.

Participants had baseline resistance to a median of 4 drugs, with 13% having resistance to a second-line injectable and 6% having resistance to a fluoroquinolone. 13% had already received 1 second-line regimen, 72% had received first-line treatment only, and 14% had received no previous treatment.

Just over half of participants received 4 or fewer potentially effective drugs in their second-line regimen (55%), while 30% received five. Only 15% received 6 or more potentially active drugs. Overall, participants received a median of 3 active drugs throughout their period of treatment (drugs could be substituted during treatment).

Results showed that 80% of patients achieved sputum culture conversion (disappearance of TB mycobacteria from sputum) after a median of 2 months of treatment. After adjusting for drug exposure, in multivariate analysis the likelihood of sputum conversion was associated with:

  • Receiving 6 or more potentially effective drugs, compared to 5 drugs (adjusted hazard ratio 1.36);
  • Receiving pyrazinamide (adjusted hazard ratio 2.00);

A previous course of second-line treatment reduced the likelihood of culture conversion by 28-38% in 2 multivariate analyses, while receiving 4 or 3 effective drugs reduced the likelihood of culture conversion by 44% and 64%, respectively, when compared to treatment with 5 effective drugs.

A multivariate analysis controlling for the presence of pyrazinamide in the regimen found that when pyrazinamide was present, 5 or 6 effective drugs in the regimen increased the likelihood of sputum culture conversion. The authors urged caution in interpreting their finding on pyrazinamide due to the lack of baseline resistance testing, and said that assuming pyrazinamide to be an active drug when designing a regimen may "put patients at risk for poorer outcomes."

The benefit of using untested drugs -- the so-called Group 5 drugs in WHO guidelines -- cycloserine, terizidone, amoxicillin/clavulanate, clarithromycin, thioacetazone, clofazimine, imipenem, and linezolid -- was dependent on the number of other effective drugs in the regimen, the analysis found.

Where pyrazinamide was not present in the regimen, the presence of untested drugs -- those newer or less frequently used drugs for which routine susceptibility testing was not carried out -- increased the likelihood of culture conversion when at least 3 or 4 effective agents were included in the regimen. Each effective drug in the regimen increased the likelihood of culture conversion by 65% (adjusted hazard ratio 1.65).

However, "without a minimum number of effective drugs present, the inclusion of an additional untested drug was not associated with any significant increase in the likelihood of sputum culture conversion," the authors concluded. "Our results suggest that it would be preferential to use the new drugs bedaquiline and delaminid […] in place of Group 5 drugs in treatment regimens."

More severe clinical disease, as shown by evidence of tuberculosis in both lungs by X-ray, was associated with an approximate 30% reduction in the likelihood of culture conversion when compared to TB in one lung only.

The researchers emphasized that as well as indicating that improved responses to MDR-TB treatment may be achieved by using more than 5 effective drugs in the regimen, their findings also indicate the importance of having rapid access to high-quality drug susceptibility testing results. This would allow more doctors to put together regimens composed of a larger number of effective drugs. But they acknowledged that randomized studies of individualized MDR-TB treatment informed by drug susceptibility testing are needed to determine whether regimens of more than 5 effective drugs result in higher cure rates than current WHO-recommended regimens.



CM Yuen, EV Kurbatova, T Tupasi, et al. Association between Regimen Composition and Treatment Response in Patients with Multidrug-Resistant Tuberculosis: A Prospective Cohort Study. PLoS Medicine 12(12):e1001932. December 29, 2015.