Patient completion rates for tuberculosis (TB) preventive therapy were higher with 3-month weekly rifapentine plus isoniazid (3HP) vs 4-month daily rifampicin (4R), according to study findings published in The Lancet Respiratory Medicine.
Although both 3HP and 4R have been recommended for TB prevention, head-on comparative studies of these 2 treatments are not available. Investigators therefore conducted a network meta-analysis of individual patient data to compare indirect estimates of relative treatment completion, safety, and efficacy between 3HP and 4R.
A literature search comparing either 3HP or 4R to 6 months of isoniazid (6H) or 9 months of isoniazid (9H) was conducted in PubMed for randomized controlled trials (RCTs) published from January 1, 2000, to March 1, 2019. Eligible trials were published in peer-reviewed journals and reported at least 1 of the following outcomes: treatment completion, treatment-related adverse events (AEs), or incidence of TB disease. “Treatment completion” was defined as taking more than 80% of the prescribed doses, based on pill counts, in 120% of the allowed time; using this criteria, patients were categorized by their completion status (as either “completers” or “noncompleters”).
The analysis included 6 trials — 3 trials comparing 3HP with 6-9H, and 3 trials comparing 4R with 6-9H. The trials came from 14 countries and included 17,572 participants: 4897 received 3HP, 4055 received 4R, and 8620 received 6-9H. The overall population had a mean age of 34.9 years, and 50.8% were female.
[T]his individual patient data network meta-analysis from RCTs of tuberculosis preventive treatment provides evidence that 3HP under directly observed therapy had significantly higher treatment completion but also significantly higher risks of treatment-related adverse events compared with 4R.
Regarding the indirect effect between 3HP and 4R, treatment completion was more likely with 3HP, with an adjusted risk ratio (aRR) of 1.06 (95% CI, 1.02-1.10) and an adjusted risk difference (aRD) of 0.05 (95% CI, 0.02-0.07). When only 9H was the comparator, the indirect aRR was 1.02 (0.98-1.07), and the aRD was 0.03 (0.00-0.06).
The proportion of participants who had any treatment-related adverse events that resulted in permanent drug discontinuation was slightly increased for 3HP vs 6-9H, and decreased with 4R vs 6-9H in direct comparisons. The analysis showed that 3HP was associated with a greater risk than 4R, with an aRR of 2.86 (95% CI, 2.12-4.21) and an aRD of 0.03 (95% CI, 0.02-0.05). Comparable findings were observed for treatment-related AEs of grades 3 to 4 that resulted in permanent drug discontinuation, as 3HP had a greater risk vs 4R, with an aRR of 3.46 (2.09-6.17) and an aRD of 0.02 (0.01-0.03).
For direct comparisons, the rate of TB disease was comparable between 3HP and 6-9H, as well as between 4R and 9H. For the indirect effect, the rate of TB disease with 3HP was similar to that for 4R, with an adjusted incidence rate ratio of 1.16 (95% CI, 0.40-3.58) and an adjusted incidence rate difference of 0.8 per 1000 person-years of follow-up (95% CI, –2.3 to 7.0).
Study limitations included the inability to adequately analyze safety related to HIV or other comorbidities (including drug-to-drug interactions with antiretroviral therapy) due to the small number of patients in these subgroups; difficulty in assessing relative efficacy due to the small percentage of patients who developed TB disease; and the inability to account for variation resulting from the site or country where patients were treated.
“In the absence of trials directly comparing 3HP and 4R, this individual patient data network meta-analysis from RCTs of tuberculosis preventive treatment provides evidence that 3HP under directly observed therapy had significantly higher treatment completion but also significantly higher risks of treatment-related adverse events compared with 4R,” stated the study authors. “This trade-off between completion and risk of adverse events must be considered when deciding the optimal treatment for tuberculosis infection.”