Nine months is a long time to take daily isoniazid -- and an even longer time to go without beer. If you have latent tuberculosis infection (LTBI) diagnosed by a positive tuberculin skin test, and you make it through a year or 2 without developing active TB, your lifetime risk of reactivation TB (given a healthy immune system) drops to only about 5% or so. Those are pretty good odds, and make it highly tempting to chuck the isoniazid after a few months. That's exactly what 36 to 70% of people with LTBI do, never completing the 9 months of INH.
As for alternatives to isoniazid, weekly rifapentine and isoniazid are an established treatment for active tuberculosis after it's been beaten back to a low bacillary burden (continuation phase). Timothy Sterling et al reasoned, why not try 3 months of this kinder, possibly gentler regimen for treatment of latent tuberculosis infection?
They randomized 7731 people with positive TB skin tests (about 2% were HIV-positive) to receive either:
- Once-weekly rifapentine plus isoniazid (900 mg each) for only 3 months (directly observed therapy)
- Daily isoniazid 900 mg for 9 months (self-administered)
Granted, this wasn't a fair fight: the rifapentine + INH group had to swallow their pill every week under the watchful eye of a study monitor, while the daily isoniazid group were left to their own devices. Unsurprisingly, 82% of the combination therapy group completed treatment; only 69% of the daily INH group did.
Tuberculosis developed in 7 of the 3,986 people taking rifapentine/INH weekly, and in 15 of the 3,745 people taking daily INH, which was an absolute reduction of 0.24% (non-inferiority).
About 1% more people discontinued rifapentin due to an adverse event (4.9% vs. 3.7% in INH-only); the biggest concern among these was for suspected hypersensitivity reactions, which occurred in 3.8% of combination-takers and only 0.5% of INH-takers. Hepatotoxicity was higher in the INH-only group (2.7% vs. 0.4%). There were no reported differences in severe toxic events, and no deaths were attributed to study drugs.
Clinical Takeaway: Carefully monitored, 3 months of weekly rifapentin and isoniazid appears to be as effective as 9 months of daily isoniazid in eradicating latent tuberculosis infection. The shorter course could result in greater treatment successes in real-life practice. Increased hypersensitivity reactions (but less hepatotoxicity) may occur in people taking rifapentin + INH; this deserves further study before the therapy becomes standard care.
Sterling TR et al (the TB Trials Consortium PREVENT TB Study Team). Three Months of Rifapentine and Isoniazid for Latent Tuberculosis Infection. N Eng J Med 2011;365:2155-2166.