SAN FRANCISCO – In treatment-naive HIV patients with cryptococcal meningitis, antiretroviral therapy should be delayed until patients’ cerebrospinal fluid proves cleared of cryptococcal infection, according to Dr. David Boulware, distinguished assistant professor of infectious diseases and international medicine at the University of Minnesota in Minneapolis.
The meningitis "must be treated optimally first," he said at the annual Interscience Conference on Antimicrobial Agents and Chemotherapy. For most patients, antiretroviral therapy should come perhaps 3-4 weeks after the start of induction therapy, when their cerebrospinal fluid is again sterile. ART should come later, perhaps 6 weeks or longer, in patients who don’t mount a strong CSF inflammatory response, and in those with a persistent altered mental status; both will have a harder time clearing their CSF, he said.
The recommendations come from a study led by Dr. Boulware that pitted early ART against delayed ART in ART-naive HIV patients with cryptococcal meningitis in Uganda and South Africa, where cryptococcal meningitis is a leading killer of HIV patients. It has not been clear until now when it’s best to start ART in HIV patients with the condition.
The researchers randomized 87 patients to start an efavirenz and nucleoside reverse transcriptase inhibitor ART regimen 7-11 days after starting cryptococcal meningitis induction therapy with 0.7-1.0 mg/kg per day of amphotericin and 800 mg/day of fluconazole. In 87 other patients, the ART regimen started 5 weeks or more after the start of induction, when amphotericin had been discontinued and fluconazole had been stepped down to a lower dose.
The trial was halted in April 2012 – significantly short of its enrollment target – after researchers realized that early-ART patients were 1.7 times more likely than delayed-ART patients to die within 6 months (95% confidence interval 1.03-2.87). Six-month mortality was 42.5% (37 patients) in the early arm, and 27.6% (24) in the delayed arm.
"You should not do early ART [in this group]. First, focus on the induction treatment to sterilize the CSF. There is no benefit in starting ART during cryptococcal induction therapy. It’s not going to help, and it’s likely to be harmful to a large proportion of subjects," Dr. Boulware said.
The mortality differences were driven primarily by patients who entered the trial with altered mental status – Glasgow Coma Scale scores below 15 – and by those who didn’t mount strong CSF inflammatory responses.
But even patients who were not sick showed "no benefits and no trends of benefits" with early ART, Dr. Boulware said.
Based on the results, "we aim to start ART at around 3 to 4 weeks" when "you’re confident the CSF is sterile," he said. He also stressed the importance of making sure the CSF culture is sterile before reducing the fluconazole dose.
A longer delay is warranted in patients with little CSF inflammation, because early ART in patients who have not yet cleared the cryptococcal infection could kick off an immune reconstitution syndrome. A longer wait also is the way to go for patients with altered mental status. "Get them better, get them ambulating, get them moving around" before starting ART, Dr. Boulware said.
The mean age of the trial subjects was 35 years. Patients were about evenly split between male and female.
The conference was sponsored by the American Society for Microbiology. The trial was sponsored by the National Institute of Allergy and Infectious Diseases. Dr. Boulware disclosed research support from GlaxoSmithKline.