SEATTLE – In patients coinfected with HIV and tuberculosis, addition of prednisone to an antiretroviral therapy (ART)/TB regimen significantly reduced the incidence of tuberculosis-immune reconstitution inflammatory syndrome (TB-IRIS).
The study showed a reduction of incidence of TB-IRIS – a worsening of the inflammatory elements of tuberculosis that often occurs within a few weeks of starting ART – in the prednisone group, with no sign of adverse events associated with immunosuppression. That safety profile “gives us the reassurance that this is something that could be scaled up. It’s effective but it’s also safe,” said Graeme Meintjes, MD, PhD, professor of medicine at the University of Cape Town, South Africa.
TB-IRIS occurs in 18% of patients undergoing ART/TB regimens, and 25% of these patients wind up in the hospital ().
Ironically, TB-IRIS is of increasing concern because of improved treatment strategies. Clinical trials have shown that, in patients with low CD4 counts, TB mortality is decreased by about 20% if ART is started within 2 weeks of initiation of TB treatment. But when ART is started so soon after TB therapy, patients with low CD4 counts are at about a twofold increased risk of TB-IRIS ().
“That brought into focus the need for an intervention to prevent this complication,” said Dr. Meintjes, who presented the study at the Conference on Retroviruses & Opportunistic Infections in partnership with the International Antiviral Society.
There were concerns that prednisone could put HIV patients at greater risk of opportunistic infections such as Kaposi’s sarcoma, although Dr. Meintjes noted that those risks were seen in a context of patients who were not taking ART. His team also showed in a previous study () that prednisone reduced duration of symptoms and hospitalization in TB-IRIS, with no increase in serious infections.
The researchers randomized 240 patients to receive prednisone (40 mg/day for 2 weeks, then 20 mg/day for 2 weeks) within 48 hours of starting ART. All patients started ART within 30 days of starting TB treatment, and had a CD4 count of 100 or fewer cells/microL.
Excluded patients included those with rifampicin resistance, neurobiological tuberculosis, Kaposi’s sarcoma, or hepatitis B, and those who weren’t on the standard first line TB treatment because they couldn’t tolerate it. Patients were also excluded if they had a poor clinical response to TB treatment.
Most endpoints were followed for 12 weeks, but HIV-related cancers were monitored out to 1 year.
Forty-seven percent of patients in the placebo group experienced TB-IRIS within 12 weeks, compared with 33% of patients in the prednisone group (risk ratio, 0.70; 95% confidence interval 0.51-0.96). In an open-label extension study, the researchers noted that 28% of patients who started out in the placebo arm eventually received corticosteroids to treat TB-IRIS, compared with 13% of those who started out in the prednisone arm (RR 0.47, 95% CI 0.27-0.83).
Subjects in the prednisone arm had a lower rate of grade 3 adverse events (29% vs. 45%, P = .01).
There was a similar mortality rate in both arms, and no difference in the incidence of Kaposi’s sarcoma or new-onset AIDS-defining infections.
Dr. Meintjes pointed out that patient selection is important. The trial patients were seen in the clinical setting, not sicker, hospitalized patients, and they had to be improving with TB treatment; significant comorbidities were excluded. “In those patients, prednisone was safe.”
Dr. Meintjes reported having no relevant financial disclosures.