CROI 2013: Researchers Discuss Advances in Tuberculosis Treatment


A new two-drug combination of rifapentine and moxifloxacin can allow tuberculosis (TB) treatment to be taken just once-weekly during the 4-month continuation phase, according to a presentation at the 20th Conference on Retroviruses and Opportunistic Infections (CROI 2013) last week in Atlanta. A related meta-analysis found that rifampicin, a key drug in TB treatment, can be tolerated at much higher levels than used in current clinical practice, suggesting that higher doses might allow the treatment course to be shortened.

[Produced in collaboration with]


Shortening the duration of TB treatment from 6 months and reducing the number of doses of medication that need to be taken during the course of treatment could have a substantial impact on the number of people who complete treatment and are cured of TB. If a new regimen also reduced the time it takes people to convert from smear-positive (infectious) to smear-negative (non-infectious), this could have an impact on onward transmission too.

A shorter duration of treatment would also reduce the time during which potential interactions with antiretroviral drugs for HIV and other medications could take place, simplifying the job of managing care for people with coinfections or other comorbid conditions, as well as reducing the potential for under-treatment or harm to the patient.

A number of studies seeking to determine whether the standard TB treatment regimen can be shortened are underway. The RIFAQUIN study, conducted by the UK’s Medical Research Council, is the first of these trials to report results.

Rifapentine and Moxifloxacin

Rifapentine is a derivative of rifampicin, but it has a much longer half-life, which means that the drug only needs to be taken once weekly. Rifapentine has been approved for use in TB treatment in the U.S., and 2003 TB treatment guidelines recommended that it can be used in combination with isoniazid during the 4-month continuation phase of treatment for HIV-uninfected TB patients without evidence of cavitation on chest X-ray who test smear-negative at 2 months. A 12-week directly-observed regimen of rifapentine and isoniazid is recommended in the U.S. as a preventive regimen for people with latent tuberculosis.

Moxifloxacin is a fluoroquinolone antibiotic that has not been used previously in TB treatment, although it was reported in 2007 that manufacturer Bayer would be prepared to make the drug available at a discounted price for resource-limited settings if it is approved for use as TB treatment.

previous study, conducted in Brazil, found that replacing ethambutol with moxifloxacin during the 2-month intensive phase of TB treatment (in combination with rifampicin, isoniazid, and pyrazinamide) resulted in a significantly higher rate of smear-negative TB by the end of month 2. This finding gave Richard Chaisson of Johns Hopkins University School of Medicine hope that a TB treatment regimen containing moxifloxacin might reduce the duration of therapy, increasing the chance that people would be able to complete the course of treatment.

A second study, by the same research group, examined the merits of replacing isoniazid with moxifloxacin during the 2-month intensive phase. That study found no difference in the proportion of patients who were smear-negative after 2 months, but did find that the proportion of patients who became smear-negative by 2 months was higher in the U.S. than at participating trial sites in Africa.

The Global Alliance for TB Drug Development and University College London, supported by Bayer Healthcare, is currently conducting another study of moxifloxacin in TB treatment, called ReMoxTB, which is investigating substituting either ethambutol or isoniazid for moxifloxacin in order to shorten treatment duration to 4 months. Results are expected by the end of 2013.


Amina Jindani of St George’s University Medical School in London and fellow investigators with the RIFAQUIN study compared 2 strategies for either reducing the length of the TB treatment course or reducing the total number of drug doses. 

The RIFAQUIN study used a complex trial design [see illustration] intended to answer 2 questions:

Is it possible to shorten the duration of TB treatment from 6 months to 4 months by giving high-dose rifapentine in a 2-month continuation phase of treatment, and moxifloxacin in place of isoniazid throughout the treatment course (both drugs dosed twice-weekly)?

Is it possible to reduce the frequency of dosing during a 4-month continuation phase by giving high-dose rifapentine and moxifloxacin in place of rifampicin and isoniazid, dosed once-weekly?


In study regimen 1, rifapentine was dosed at 900 mg twice weekly, and in study regimen 2 rifapentine was dosed at 1200 mg once weekly. Standard-of-care TB treatment was given daily.

The primary endpoints were relapse during follow-up to 18 months after treatment initiation, and occurrence of grade 3 or 4 adverse events.

The study recruited 730 people with newly diagnosed smear-positive TB. It included people with HIV not taking antiretroviral therapy who had CD4 cell counts above 150 cells/mm3; 28% of study participants were HIV positive, with a median CD4 cell count of 312 cells/mm3.

The final modified intent-to-treat analysis excluded patients who attended study visits outside the study window (n=34), or who were lost to follow up (n=42), had missing or contaminated cultures or failed sputum (n=26), non-TB deaths (n=17), women who became pregnant during the study (n=9), and laboratory-confirmed cases of reinfection (n=10). This exclusion of 19% of randomized participants left 592 evaluable patients.

No significant difference in per-protocol outcome measures was apparent when the 4-month, once-weekly group was compared to the standard-of-care control group, whether compared by the endpoint of treatment failure, relapse, or death (5% in the control group vs 4% in the once-weekly treatment group) or by the wider endpoint that also included loss to follow-up and treatment changes due to adverse events or other unspecified reasons not related to treatment failure (14% in each arm).

When the 2-month, twice-weekly regimen was compared to the standard of care, it was found to be inferior. 17% of patients in the 2-month arm reached an endpoint of treatment failure, relapse, or death, compared to 5% in the standard-of-care arm. When loss to follow-up and treatment changes for reasons other than failure were included, 28% in the 2-month arm reached an endpoint, compared to 14% in the standard-of-care arm.


"The implications are that supervised treatment can continue beyond the first 2 months of treatment," according to Jindani, who said that the continuation regimen of 16 weekly doses would make it much easier for treatment programs and community health workers to monitor adherence and ensure that patients are not lost to follow-up.

Despite the promising results, TB treatment programs will not be switching immediately to once-weekly treatment during the continuation phase. As Andreas Diacon of Stellenbosch University said, "we will need to train thousands of nurses with a new treatment protocol." Guideline panels at international and national levels will also need to review the evidence when it is published.

There is also the thorny question of drug interactions. Although there were no differences in outcomes between HIV negative and HIV positive patients in the study, none were taking antiretroviral therapy, due to a drug interaction between rifapentine and antiretroviral drugs in the non-nucleoside reverse transcriptase inhibitor (NNRTI) and protease inhibitor classes, which may reduce levels of the antiretrovirals.

Manufacturer Sanofi-Aventis is currently conducting pharmacokinetic studies to look at the impact of high-dose rifapentine on efavirenz, tenofovir, and emtricitabine drug levels. But will all physicians be satisfied with evidence from a drug-interaction study alone, or will further studies be necessary in HIV positive patients?

Finally, there is the issue of cost. Rifapentine is an expensive drug, and although Sanofi Aventis has said it will reduce the cost of rifapentine when there is widespread uptake, TB programs will be waiting to see whether the cost of offering once-weekly treatment will be affordable within current budgets for all patients, or whether it will be prioritized for particular groups of patients.

"There are certain groups in which it would be very valid [to treat with moxifloxacin in place of isoniazid], for example patients with pre-treatment isoniazid resistance," said Jindani, identifying one special need.

In the U.S. a switch in practice is likely to be more rapid. "Here in the U.S. 10 years ago we were prepared to go to a weekly regimen, but it didn't work [for all groups of patients in a trial conducted at the time], but these data have rekindled that hope," said Chaisson.

Rifampicin Drug Dosing

Rifampicin is normally used at a dose of 600 mg per day throughout the 6-month TB treatment course. Yet the maximum-tolerated dose of rifampicin has never been defined.

When the drug was being developed in the 1960s, it was extremely expensive to synthesize, and on the basis of anecdotal reports it was also assumed at the time that the toxicities of the drug were likely to be dose-related. This led investigators to settle on a dose of 600 mg as adequate to achieve minimum inhibitory concentrations for 24 hours. Indeed, early studies of short-course treatment either limited the dose and duration of rifampicin treatment, or avoided use of the drug altogether, in order to test regimens that would be affordable in low-resource settings.

literature review by Martin Boeree and colleagues at the St Radboud University in Nijmegen subsequently found that there was little evidence to support these assumptions about dose-related toxicity. Furthermore, they argued that previous investigators had made the wrong assumption about the most relevant pharmacokinetic parameter for assessment in studies of rifampicin as a TB treatment. Rather than looking at the ratio of the trough level (Cmin) to the minimum inhibitory concentration, which shows whether adequate blood levels are maintained throughout a 24-hour period, Boeree and colleagues argued that what mattered was the total drug exposure achieved during a 24-hour period. Higher total drug exposure would be more likely to reduce bacterial load, and higher peak concentrations would be needed to inhibit bacterial replication in epithelial fluids within the lungs.

The study, conducted by Boeree and colleagues in the PanACEA Consortium in South Africa, recruited people with smear-positive TB in 5 consecutive groups, each of which received a higher dose of rifampicin for the first 7 days of TB treatment, after which isoniazid, pyrazinamide, and ethambutol were added for a further 7 days. The first group received a dose of 10 mg/kg, and subsequent groups received 20, 25, 30, or 35 mg/kg. Dose increments were approved after reviewing data on adverse events in each cohort of patients. All participants received rifampicin in combination with standard doses of isoniazid, ethambutol, and pyrazinamide.

The study measured rifampicin levels at days 7 and 14 of treatment. Mycobacterium tuberculosis colony-forming units were measured by culture on solid media at baseline and days 1 to 7 and days 12 and 14. Time to culture positivity was measured in liquid media at the same time intervals.

Rifampicin was well-tolerated at higher doses, with no grade 4 adverse events and only 4 grade 3 adverse events reported, with no association between increasing dose and incidence of grade 3 events. On the basis of these results, the researchers concluded that they had failed to identify a maximum tolerated dose, and research will now go forward to test a 35 mg dose for safety and efficacy in a Phase 2 study, and to test doses of 40 mg and 45 mg for safety.

There was also a significant trend towards a reduced number of colony-forming units of M tuberculosis at higher doses, and a 7-fold increase in total rifampicin exposure at the highest dose when compared to the standard dose. This increase in exposure was disproportionate to the increase in dose. However, very substantial variations in exposure were seen between individuals at each dose, indicating that pharmacogenomic factors such as genetic markers for rapid clearance of rifampicin will require attention in future research into the optimization of TB regimens.

Another factor that will require attention is the interaction between rifampicin and the anti-HIV drug efavirenz (Sustiva or Stocrin, also in the combination drug Atripla). At a dose of 600 mg, the effect of rifampicin on efavirenz levels is enough to persuade U.S. physicians that the efavirenz dose should be increased to 800 mg. What is unclear at present is whether the increase in rifampicin exposure is accompanied by a corresponding increase in the induction of efavirenz metabolism, or whether this effect hits a plateau at some point, beyond which the efavirenz dose would not need to be increased any further.



A Jindani, M Hatherill, S Charalambous, et al. A Multicentre Randomized Clinical Trial to Evaluate High-dose Rifapentine with a Quinolone for Treatment of Pulmonary TB: The RIFAQUIN Trial. 20th Conference on Retroviruses and Opportunistic Infections. Atlanta, March 3-6, 2013. Abstract 147LB.

M Boeree, A Diacon, R Dawson, et al (PanACEA Consortium). What Is the “Right” Dose of Rifampin? 20th Conference on Retroviruses and Opportunistic Infections. Atlanta, March 3-6, 2013. Abstract 148LB.