In December 1997, the U.S. government issued guidelines for the treatment of HIV infection in children. These were revised and published in April 1998 (Guidelines for the Use of Antiretroviral Agents in Pediatric HIV Infection, MMWR, vol. 47, no. RR-4, April 17, 1998). In this document it is stated that the "preferred approach [is to] initiate [anti-HIV] therapy-regardless of age or symptom status." In other words, treat all HIV-positive children. Later on in the recommendations, it is spelled out that the "preferred regimen" for all of these kids includes three drugs: one highly active protease inhibitor (PI) plus two nucleoside analogue reverse transcriptase inhibitors (nRTIs).
Where is the evidence for this? Well, the guidelines were based on the clinical benefit seen in adults with advanced disease when treated with this drug combination. In the guidelines for adults, it is speculated that this combination will work as well, if not better, when taken by people with less advanced HIV. The pediatric guidelines further speculate that this theory can be applied to children. Therefore, it is particularly important to watch very closely when results from triple combination therapy studies in children are reported to make sure that this speculation is founded in reality.
Finally, more than two years after the first protease inhibitor, Norvir (ritonavir), was approved with a liquid formulation, a study was reported on, using a Norvir-containing triple combination therapy, in children for 48 weeks. Unfortunately, this study (ACTG 338), reported at the 12th World AIDS Conference in Geneva, Switzerland this summer, generated as many questions as it answered.
ACTG 338 is a government-sponsored clinical trial where children are given, by the luck of the draw, one of three possible medication combinations:
There were 298 children participating in the trial with ages ranging from 2 to 17. The average (median) age was 7.1 years old. All kids had at least 16 weeks experience with nRTIs (such as AZT or ddI) but none had ever taken a protease inhibitor before. At the beginning of the trial, they were either in CDC category 1 (no immune system suppression with a CD4% of at least 25%) or in CDC category 2 (moderate immune suppression with a CD4% of between 15% and 24%). In other words, these were not children with advanced immune system suppression. The average (median) viral load at the beginning of the study was approximately 25,000.
To test if therapy is a success, the researchers use the viral load test to see if the amount of virus goes down when the drugs are given. In this trial the test could not see levels of virus below 400 copies per ml. It was the goal to get the levels below 400 ("below detection") so that the protease inhibitor, Norvir, would work longer. It has been seen in adults that if viral loads don't go down to very low levels when taking a protease inhibitor, then the virus can become resistant to the drug. That is, the drug may not work for long. Furthermore, if the virus is resistant to that protease inhibitor, it could be resistant to all the protease inhibitors currently available. What the table below shows is the percentage of kids who experienced virus going down to these very low levels. As you can see, the percentage of kids with very low level virus decreases with time. Although the triple combination did the best, at 48 weeks more than half did not have viral loads at this very low level.
|Week 12||Week 24||Week 48|
|AZT + 3TC||12%||0%||0%|
|Norvir + d4T||52%||34%||27%|
|Norvir + AZT + 3TC||54%||47%||42%|
Does that mean that these kids now have virus that is resistant to Norvir and, possibly, the other protease inhibitors? Does this eliminate the possibility of using the very strong protease inhibitors sometime in the future if these kids experience advancing disease? Should we have waited to use the protease inhibitor? Would it have worked better if the kids had not taken other anti-HIV drugs before? We do not know.
CD4 increases were about the same in the Norvir + d4T and Norvir + AZT + 3TC treatment groups, and these increases were maintained despite the viral load going back up as the study went on. Why do the CD4 counts continue to go up even when the viral load goes back up? Are these CD4 cells functional? Do we need such aggressive anti-HIV therapy to see these CD4 increases? How long will the CD4 increase last after the viral load goes back up? We do not know.
There are a few other things worth mentioning. First, 30% of those children participating stopped taking the assigned drugs mostly because of nausea and vomiting or they just couldn't stand the taste of the Norvir. At the end of the 48 weeks, only 36% of the kids on Norvir were on the dose they started with.
Also, the dose of Norvir in this study, 250 mg per meter squared (m2) of body surface area, is below what is currently recommended. If children were given the recommended dose of 400 mg/m2, would more of them have experienced very low viral loads? Would there have been more drop outs because of increased nausea or vomiting? We do not know.
So, there you have it (or don't). This study places the value of early, triple-combination therapy in children, and, therefore, the federal guidelines, in question. They might be right or they might be wrong. We just don't know for sure.
Interestingly, another presentation at Geneva looked at using the protease inhibitor, Norvir (ritonavir), in children with a lot of previous treatment experience and much more advanced HIV. This was simply an observational study, presented by Dr. Nancy Hutton, where Norvir was offered to kids at Johns Hopkins in Baltimore. The children were between one and 12 years old and had CDC class 3 immune suppression (CD4% less than 15%; serious immune suppression). They had viral loads above 100,000 at baseline. The report was on 21 children and all had previous experience with AZT and three-fourths had experience with ddI and/or 3TC.
Norvir was given at 350 mg per meter squared (m2) in combination with one or two nucleoside reverse transcriptase inhibitors (nRTIs). During the first month of Norvir, viral loads decreased by about 30- fold (for example, from 150,000 to 5,000) but then started going up again. At 12 months of treatment, CD4 cell counts increased from a baseline of below 50 to a range of 450 to 540 despite the fact that the viral load went back up to between 10,000 to 100,000.
Once again, a lot of questions need to be answered here, but it is very encouraging to see such large and lasting CD4 cell increases despite the viral load going back up in kids with advanced immune system suppression. How would these kids have done if they had waited to take therapy, in general, so all the drugs they took at this time were new to them? Is there a benefit in waiting to be treated? Will the CD4 cell increases continue? Would they have done better with earlier triple combination therapy? We do not know.
Until clinical trials are conducted which compare starting therapy at different times with different types of combinations, we will not get the answers to these questions. As with adults, treatment of children with strong combinations in advanced HIV has a clear benefit. Early treatment is anyone's best guess.
With treatment decisions more complicated than ever and with the information from clinical trials generating so many questions, it's amazing that anyone believes that there is a single right answer. It was completely absurd that the State of Maine felt they had enough information to force a mother to give strong drug combinations to her four year old HIV-positive child! Thank goodness, the court disagreed. But these struggles will continue as the government guidelines, however speculative, are used to force parents to treat their children with strong combinations of drugs early in the course of HIV infection. These decisions are difficult, complex, and personal. The courage it takes to make them must be fully appreciated and respected.
Copyright ©1999 by People With AIDS Working for Health, Inc. Non-commercial reproduction is strongly encouraged.