The case was reported earlier this year but some doctors were skeptical that the baby was really infected rather than testing positive because of exposure to virus in the mom's blood.
The new report, published online Wednesday by the New England Journal of Medicine, makes clear that the girl, now 3, was infected in the womb. She was treated unusually aggressively and shows no active infection despite stopping AIDS medicines 18 months ago.
Doctors won't call it a cure because they don't know what proof or how much time is needed to declare someone free of HIV infection, long feared to be permanent.
"We want to be very cautious here. We're calling it remission because we'd like to observe the child for a longer time and be absolutely sure there's no rebound," said Dr. Katherine Luzuriaga, a University of Massachusetts AIDS expert involved in the baby's care.
The government's top AIDS scientist, Dr. Anthony Fauci, director of the National Institute of Allergy and Infectious Diseases, agreed.
"At minimum, the baby is in a clear remission. It is possible that the baby has actually been cured. We don't have a definition for cure as we do for certain cancers, where after five years or so you can be relatively certain the person is not going to go and relapse," he said. A scientist at his institute did sophisticated tests that showed no active virus in the child.
A government-sponsored international study starting in January aims to test early treatment in babies born with HIV to see if the results in this case can be reproduced.
Most HIV-infected moms in the U.S. get AIDS medicines during pregnancy, which greatly cuts the chances they will pass the virus to their babies. But the Mississippi mom got no prenatal care and her HIV was discovered during labour. Doctors considered the baby to be at such high risk that they started the child on three powerful medicines 30 hours after birth, rather than waiting for a test to confirm infection as is usually done.
Within a month, the baby's virus fell to undetectable levels. She remained on treatment until she was 18 months old when doctors lost contact with her. Ten months later when she returned, they could find no sign of infection even though the mom had stopped giving the child AIDS medicines.
Only one other person is thought to have been cured of HIV infection — a San Francisco man who had a bone marrow transplant in 2007 from a donor with natural resistance to HIV, and showed no sign of infection five years later.
In the Mississippi baby, "there's no immune mechanism we can identify that would keep the virus in check" like that bone marrow donor, said another study author, Dr. Deborah Persaud of the Johns Hopkins Children's Center, who helped investigate the case because she has researched treatment in children.
Dr. Peter Havens, pediatric HIV chief at Children's Hospital of Wisconsin and a government adviser on HIV treatment guidelines, said the child may have an undiscovered genetic trait that helped her manage the virus.
"I'm just not convinced that her dramatic response would be replicable in a large population," he said. It's too soon to recommend treating other high-risk babies so aggressively without more study, he said.
In the upcoming study, doctors plan to give AIDS medicines for at least two years and watch for signs of remission before suspending treatment and seeing whether a remission results.
The Mississippi case "did open people's eyes further" about a possible cure, Luzuriaga said. "We might be able to intervene early and spare children a lifetime of therapy. That is the potential impact of this case."