How sure are we, and what does this mean for the future?
Several weeks ago many heard the report in the lay press of a baby born in Mississippi being cured of human immunodeficiency virus (HIV) infection. Although the media buzz was mostly about the drama, less attention was given to what actually happened and the controversy that followed.
What do we know about this case?
It is clear that there was a baby born to a woman who was not aware that she was infected with HIV until she went into labor; rapid HIV antibody testing detected the infection. It was ultimately confirmed that she was infected and actually had relatively low amounts of HIV circulating in her blood. When maternal infection is diagnosed during pregnancy, the woman is routinely started on combination antiretroviral therapy with a marked reduction in the risk of transmitting HIV to the newborn. In this case, the diagnosis was made late, and the mother was not able to be treated, with the focus shifting to the baby who was given a combination of therapy typically reserved for those already known to be infected, as opposed to an abbreviated regimen given primarily for prophylaxis to prevent rather than actually treat infection. The baby was confirmed to be infected based upon the standard definition of having detectable viral DNA or RNA from two separate blood samples and was continued on a treatment regimen with the viral load being repeatedly positive over the first several weeks of life, until being undetectable after approximately three weeks. After being followed for months, the child was lost to follow-up until returning to the clinic off medications for months. In anticipation of restarting therapy, a repeat viral load turned out to be negative. This prompted a more intensive investigation that demonstrated no detectable plasma HIV RNA, cellular DNA, HIV-specific immune responses, or even HIV antibodies by standard assay, leading the investigators to conclude that the child had achieved at the very least a "functional cure."
What is different about this case from most others?
Perhaps the most important message from this case was the missed opportunity to identify the mother's infection earlier so that she could be treated with antiretroviral therapy to enhance both her health and to reduce the risk of transmission of HIV to her infant. After this, the difference in the treatment of the newborn in this case from the standard of care is relatively subtle but potentially important (for example, using therapeutic doses of a full antiretroviral regimen, as used for treatment of infection rather than the recommended lesser treatment that while insufficient for treatment has been shown to prevent infection from occurring). The latter regimen is referred to as prophylaxis and includes six weeks of therapeutic doses of zidovudine (Retrovir) with the addition of nevirapine (Viramune, Viramune XR) given as a single dose three times during the first week of life. Another difference in this case from others might be that the diagnosis of perinatal transmission by the presence of two positive virologic tests from unique samples occurred within a few days of delivery, allowing for sustained therapy with antiretrovirals.
What may explain the events described in this patient?
It is clear that at least for now the baby has remained without detectable HIV after many months off of therapy. The investigators suggested that this represented at least a functional cure, which means that virus may not be completely eradicated from the body, a very difficult thing to prove, but at the very least, what virus is present, if any, is completely controlled without evidence of viral replication. Although there is no consensus on what has happened, I would suggest that the list below includes at least some of the leading possibilities that might explain the events.
- The baby was indeed infected but early treatment with therapeutic doses of antiretrovirals, plus or minus some intrinsic characteristic in newborns, led to little or no establishment of a reservoir of infection in the infant's resting memory cells and subsequent clearance with time, ultimately resulting in a true cure.
- A reservoir was established but early treatment resulted in it being extraordinarily small and perhaps not completely cleared with time, but being so small that it will take more time before viral rebound occurs.
- There was transfer of HIV-containing maternal blood, perhaps plasma and cells, that were detected during the first weeks of infection and the presence of antiretroviral therapy simply served as post-exposure prophylaxis, with the baby never truly being infected.
- There were host or virologic factors that resulted in this child being one of the uncommon individuals who completely control their infection and that such virologic control would have occurred regardless of what was done to the infant.
An argument made by the investigators against the possibility that infection did not actually occur is that it would have required substantial transfer of maternal blood, which had a viral load of ~2500 copies/mL, to result in the baby having the observed initial plasma viral load of ~20,000 copies/mL. In addition, the decay of virus over the first few weeks occurred gradually, at least partially mimicking that which is seen in infected individuals starting on antiretroviral therapy. Further support for the possibility that early treatment could result in a smaller reservoir with potential enhanced likelihood of virologic control after prolonged antiretroviral therapy comes from two other recently reported studies. One study included four other newborns started on combination therapy during the first weeks of life who were also shown to have very low levels of HIV in their blood, although treatment had not been discontinued to assess whether viral rebound would occur. The other was from a French group of adults treated during the first weeks of HIV infection. A subset of 14 had low levels of the HIV reservoir while on treatment, and upon discontinuation of therapy, showed little or no evidence of viral rebound.
Where do we go from here?
Based upon our experience in the field, I would suggest that the first thing we should do is take a few collective deep breaths and remind ourselves of the many past false starts resulting from case reports and case series. Next, we should take a few more deep breaths and think through what this new data may be telling us, look for opportunities to further study the phenomenon with existing patients, and design new clinical trials to systematically define how early treatment might be important in achieving a functional cure.