Levofloxacin was associated with statistically significant reduction in tuberculosis developing in household contacts of persons with multidrug- resistant TB in a meta-analysis, but not in the constituent controlled trials.
Image credit: Anna Shvets
An extended course of daily levofloxacin was found effective in preventing household contacts of persons with multidrug-resistant tuberculosis (TB) from developing the disease, in a meta-analysis of a controlled trial with adults and another with children.1-3 The statistically significant separation from placebo in the meta-analysis did not occur, however, in the respective trials.
The disparate findings prompted an accompanying editorial to propose that the adage, "we treat patients, not lab results" be augmented with "we treat patients, not P values."4
The meta-analysis lead author Trinh Duong, MSc, Medical Research Council Clinical Trials Unit, University College London, London, UK and colleagues suggest the difference was due to the trials being insufficiently powered, with a lower incidence of TB than expected.
Duong and colleagues explain that pooling data in the meta-analysis led to a more precise estimate of the treatment effect.In addition, their Bayesian method analysis "provided evidence of treatment efficacy within each trial population separately, and showed the effect was similar among adults and children."
In their editorial, Emil Hodzic-Santor, BSec, and Mike Fralick, MD, PhD, Lunenfeld-Tanenbaum Research Institute, Sinai Health System, Toronto, Ontario, Canada, point out that a P value of .05 is on a continuous measure, and should not necessarily serve as demarcation for binary clinical decisions.
"Solely applying a P value threshold or 95% CI (confidence interval) to determine if a treatment works or not is an oversimplification, Hodzic-Santor and Fralick declare."There are many factors to weigh, including clinical factors."
Among the factors that they suggest should be considered is whether other treatments exist for the condition, and they note that prior to the trials in this meta-analysis, there were no demonstrated treatment options for the prevention of multidrug-resistant TB.
A meta-analysis combining data from two trials (VQUIN and TB-CHAMP) found that a 6-month course of daily levofloxacin reduced TB incidence by 60% among household contacts of individuals with multidrug-resistant TB.
An accompanying editorial emphasized that statistical significance should not be the sole determinant of clinical effectiveness.
While levofloxacin showed promise in preventing MDR-TB, concerns remain about musculoskeletal adverse events.
The trials were the Vietnam Quinolones for MDR-TB trial (VQUIN)2 with a cohort comprised mostly of adults, and the Tuberculosis Child Multidrug-Resistant Preventive Therapy trial3 (TB-CHAMP) in South Africa.Both enrolled participants with household exposure to an individual with microbiologically confirmed pulmonary rifampin-resistant/MDR-TB. Participants were included if they showed evidence of latent infection with Mycobacterium tuberculosis (Mtb) immune sensitization, had human immunodeficiency virus (HIV), or were severely malnourished.
Participants in each were randomly assigned on 1:1 ratio to receive either daily placebo or levofloxacin (10-15mg/kg for adults for 26 weeks in VQUIN, and 15-20mg/kg for children for 24 weeks in TB-CHAMP. Follow-up in TB-CHAMP was originally up to 96 weeks, later reduced to 72 weeks; and to 134 weeks in VQUIN. The primary endpoint was microbiologically confirmed or clinically defined TB, including TB-related death, by 54 weeks following random assignment.
"This time frame was chosen based on previous studies showing that most contacts who develop TB disease after exposure to Mtb do so within 12 months, and concerns that subsequent reexposure and exogenous reinfections with longer follow-up could dilute the treatment effect," Duong and colleagues explained.
Both trials had found fewer participants developing TB disease in the levofloxacin group, but the differences from placebo were not statistically significant. In the meta-analysis, however, 6-months daily levofloxacin was associated with a 60% reduction in TB incidence among the household contacts over 1 year; albeit with an increased risk of musculoskeletal adverse events.
"Further evaluation of the risk-benefit balance, tolerability, and cost-effectiveness of MDR-TB preventive treatment in different populations is needed," Duong and colleagues recommend.
