A Short Two-Drug Regimen Prevents Active TB

A study that could have global implications on the prevention of tuberculosis (TB) was presented as a late-breaker at the 5th Conference on Retroviruses and Opportunistic Illnesses held in Chicago in February (late breaker 5). Conducted in the United States, Brazil, Haiti and Mexico, the trial compared a short two-month regimen of rifampin (450-600 mg/day) plus pyrazinamide (20 mg/kg daily) to a one-year regimen of daily isoniazid (300 mg/day). Twelve months of isoniazid (INH) is the present standard of care in the prevention of tuberculosis in HIV-positive individuals with a history of a positive TB skin test reaction, which marks latent as well as active disease.

Once infected with the TB bacillus, HIV-positive persons are ten times more likely to develop active TB than HIV-negative persons. It has also been shown that TB promotes HIV replication and increases viral load (Retrovirus Conference posters 259 and 731) and possibly contributes toward HIV disease progression and mortality. Thus, any intervention to prevent TB reactivation through easier regimens that promote improved compliance or lend themselves to directly observed therapy is worth investigating. Since TB is one of the most prevalent infections in the world, especially in developing countries, the importance of having a shorter preventive regimen could have far-reaching implications in terms of both TB and HIV morbidity and mortality.

The prevention study enrolled 1,583 participants, who were randomized to one of the two treatment regimens between September 1991 and May 1996 and followed through to the end of October 1997. The average entry CD4 count was 436 cells, and 7% of study subjects had a history of AIDS. Significantly more participants in the shorter regimen (80% versus 63%) completed treatment, yet more participants in the two-drug arm had to discontinue due to side effects (9.5% versus 6.1%). Confirmed tuberculosis occurred in 19 of the 791 participants in the rifampin/pyrazinamide arm as compared to 24 of 792 participants enrolled in the INH arm, a nonsignificant difference. The rates of developing drug-resistant TB were very low and not significantly different for either group, as were the proportions of participants who developed pulmonary, extrapulmonary or combined disease. The death rates also were similar: 5.7% in the rifampin/pyrazinamide group as compared to 6.6% in the INH group. The overall incidence of reportable adverse effects was also similar in both arms: 12.3% versus 10.5%, favoring the INH group slightly. There were 2 cases of drug-related hepatitis in each group. Four participants in the INH group developed peripheral neuropathy as compared to one participant in the two-drug group. Fred Gordin, M.D., the Protocol Chair and leader of the international team, concluded that the study supported the use of the short-course rifampin/pyrazinamide over yearlong INH, as the rates of tuberculosis reactivation, survival and overall side effects were similar.

Once all the data are analyzed and confirmed, it is expected that this study will lead to a change in the recommendation for the prevention of TB in HIV-positive persons. But injection drug users on methadone maintenance may not tolerate this regimen well. This is because rifampin induces the cytochrome P450 hepatic enzymes, leading to enhanced methadone metabolism and possibly to methadone withdrawal symptoms. Similarly, the short-course regimen is not likely to be useful for those on protease inhibitors, which also have drug-drug interactions with rifampin. In most cases, rifampin is contraindicated with protease inhibitors, and rifabutin is recommended as an alternate for the treatment of active TB. However, the relative effectiveness of rifabutin in preventing TB has not yet been determined.