A baby born with the AIDS virus two years ago in Mississippi who was put on antiretroviral therapy within hours of birth appears to have been cured of the infection, researchers said Sunday at a scientific conference in Atlanta.

Whether the cure is complete and permanent, or only partial and long-lasting, is not certain. Either way, the highly unusual case raises hope for the more than 300,000 babies born with the infection around the world each year.

If the findings in the new case bear up under further scrutiny, it will mark the first time the infection has been cured by drugs. The only known cure of a case of HIV infection occurred in 2007. An American man living in Germany got a bone-marrow transplant from a donor who had a rare HIV-resistance mutation in his cells.

“We are calling this a ‘functional cure,’ ” said Katherine Luzuriaga, a physician at the University of Massachusetts Medical School, who was involved in the baby’s care. “Time and further investigation will tell us whether this child actually has been cured or not.”

Researchers provided no personal details about the case, not even the sex of the infant.

Word of the case was being greeted with interest and skepticism by researchers, who have seen many other reports of cure prove false when examined closely or tried again.

“It’s a hypothesis-generating case,” Anthony S. Fauci, head of the National Institute of Allergy and Infectious Diseases, said of the Mississippi case. “It will give us some food for thought about studies that need to be done to see if this is a real phenomenon.”

Treating a newborn with the life-extending combination of drugs known as “triple therapy” is almost never done.

The reason is that it is difficult to determine immediately after birth whether a baby is infected. Antibodies from an HIV-infected mother spill into the baby’s circulation, giving a positive test for weeks. Infants are not started on the drugs until at least six weeks after birth, when infection is certain.

Mother-to-child HIV transmission is extremely rare in the United States. Fewer than 200 cases occur each year. For more than 15 years, virtually all pregnant women known to be infected have been given antiretroviral drugs during pregnancy and labor. Their babies get one or two drugs for at least six weeks after birth — a strategy that cuts the risk of transmission to less than 5 percent.

In the Mississippi case, how­ever, the mother arrived at the hospital without prenatal care. When a screening test for HIV came back positive, “she was too near delivery to give even the dose of medicine that we routinely use during labor,” Hannah Gay, a pediatrician at the University of Mississippi Medical Center in Jackson, wrote in a short narrative of the events provided to a reporter.

Because of the baby’s unusually high risk of having been infected, Gay and her colleagues started full triple therapy 30 hours after birth.

HIV tests of the infant’s blood over the next three weeks were repeatedly positive. The virus could have been from the mother, although such spillover is usually cleared quickly by the baby. Instead, the amount of HIV — the “viral load” — declined steadily over time, as happens in adults with long-established infections when they are started on triple therapy. By 29 days after birth, the baby had no detectable virus, which is the goal of treatment.

The child’s mother began missing appointments after a year. At 18 months, the child was no longer on treatment. When the child was brought back to the clinic at 23 months, the viral load was still undetectable, “very much to my surprise,” Gay wrote.

The virus has remained un­detectable by conventional tests since then. Extremely sensitive testing, done only for research purposes, has detected traces of viral RNA and DNA but no virus that is able to replicate and spread.

“We don’t know if those tests are false positives or true positives. We are doing more studies to try to figure that out,” Luzuriaga said.

Numerous studies have tried to cure HIV-infected people by starting treatment as soon as a diagnosis is made. But that is almost always weeks after a person has been infected. By then, the virus has spread throughout the body, infected millions of cells and stitched its genetic message into the chromosomes. Treatment of such acute cases hasn’t produced cures.

Fauci, however, said it is possible that some babies started on triple therapy in recent years might also be cured. “Treatment even at six weeks might be able to cure some kids,” he speculated. “We need to go back and look to see if there is any residual virus in them. It may be that we have inadvertently cured them.”

More important, he said, is to find a way to test a strategy of treatment at birth in places such as sub-Saharan Africa, where mother-to-child transmission is a huge problem.

There is one other situation where immediate treatment with antiretroviral drugs may have cured incipient infections.

When health-care workers are stuck with needles containing blood from someone with AIDS, many choose to start taking the medicines within hours of the accident. They have a lower rate of infection than people who do not take the medicines — a finding that suggests that an infection that would have occurred was aborted.

In recent years, there has been a huge push to reduce mother-to-child transmission by testing all pregnant women and putting them on antiretroviral drugs — either for the duration of pregnancy or for life — if they are positive.

In 2011, 330,000 children were born with HIV infection or acquired it during breast-feeding, according to the United Nations program UNAIDS. That was 24 percent fewer such infections than in 2009.

Details of the case will be presented Monday at the 20th Conference on Retroviruses and Opportunistic Infections. A general description was given to reporters Sunday at a news conference before the meeting began.