Preliminary results of a study conducted in Uganda by members of the HIV Prevention Trials Network (HIVNET) showed that a potent and long-lasting drug, nevirapine, was 47 percent more effective in reducing mother-to-child transmission of HIV than a short but much more expensive course of zidovudine (AZT). At about US$4, the cost of the two doses of nevirapine used in the study (one for the mother and one for the newborn) is a fraction of the costs of other antiretroviral regimens.
Ugandan officials announced the findings of the HIVNET study, which was conducted by researchers from Makerere and Johns Hopkins universities and sponsored by the U.S. National Institute of Allergy and Infectious Diseases (NIAID), in Kampala, Uganda, July 14, 1999. "This research provides real hope that we may be able to protect many of Africa's next generation from the ravages of AIDS," said Dr. Crispus Kiyonga, Uganda's Minister of Health.
Other health officials and policymakers also hailed the findings as a major breakthrough. According to Dr. Philippa Musoke, a lecturer and pediatrician at Makerere Medical School and co-investigator on the HIVNET study, this is because the two-dose course of nevirapine is the first antiretroviral regimen that is truly affordable in the resource-constrained countries where 95 percent of HIV transmission occurs.
At the same time, Dr. Musoke and others cautioned against expecting too much too soon. They pointed to the expense and logistical difficulties of getting antiretroviral treatments such as nevirapine to HIV-positive pregnant women in countries with inadequate health care systems and limited access to voluntary HIV counseling and testing.
"Even if they are cheap and affordable fore everyone, it doesn't mean that people will access them tomorrow," observed Dr. Joseph Saba, clinical research specialist at the Joint United Nations Programme on HIV/AIDS (UNAIDS).
Dr. Saba is chairman of UNAIDS' International Working Group on Mother-to-Child Transmission, which met August 31 to develop technical recommendations on the use of nevirapine and other new antiretroviral regimens. "These results are certainly very encouraging, but we need to look more carefully at the real cost of implementing them," he said.
Searching for a Solution
Because of the life-saving potential of its results, the HIVNET study (known as HIVNET 012) has been compared to the "ACTG 076" trial that in 1994 provided the first real breakthrough in prevention of mother-to-child HIV transmission. The drug regimen used in that study, which was shown to reduce perinatal HIV transmission by two-thirds (from 24 percent to 8 percent) in the United States, quickly became the standard preventive treatment in industrialized countries. Largely as a result of its widespread use among HIV-positive pregnant women, the number of U.S. infants who contract HIV from their mothers has dropped by two-thirds, from a peak of 907 in 1992 to 297 in 1997.
But the ACTG 076 regimen, which begins as early as the 14th week of pregnancy and includes intravenous administration of AZT during labor and six weeks of treatment for newborns, costs $800 to $1,000. Even if the price were not prohibitive, the requirement that a woman receive antenatal care early in her second trimester would still make this regimen impractical in most resource-constrained countries.
These countries are also home to virtually all of the infants born with HIV infection worldwide -- somewhere between 600,000 to 800,000 a year. In the hardest-hit areas of sub-Saharan Africa, one-third to one-half of pregnant women are HIV-positive, and 25 to 35 percent of them pass the virus on to their infants in the uterus, during labor and delivery, or through breast milk.
Since 1994 a number of studies have sought to identify more affordable and practical antiretroviral regimens for the resource-constrained countries where they are most needed. This research included a trial of short-course AZT therapy supported by the U.S. Centers for Disease Control and Prevention (CDC) in Thailand, a study of a similar AZT course in breast-feeding women in Côte d'Ivoire, and a UNAIDS-sponsored trial of three different regimens of AZT and lamivudine (3TC) in South Africa, Tanzania and Uganda, known as the PETRA study.
The Thai trial, in which women were randomly assigned to receive either AZT from the 36th week of pregnancy to labor and delivery or a placebo drug, found a 50 percent reduction in HIV transmission among those taking AZT after six months of follow-up. Since the release of these results in February 1998, the four-week course of AZT, which costs about $268, has been the recommended therapy for preventing mother-to-child transmission of HIV in nonindustrialized countries, though it has been implemented in only a few limited settings.
Surprising Results
The Thai findings affected ongoing HIV research protocols as well as treatment recommendations. At Mulago Hospital in Kampala, where the HIVNET 012 trial of even shorter regimens of AZT and nevirapine had been underway for a few months, the placebo arm of the study was eliminated. Enrollment of women in the placebo arms of the PETRA trials also ended.
With evidence of the efficacy of a less expensive, shorter course of AZT that could be feasible for resource-constrained countries, trials using placebo controls were no longer considered ethical. The HIVNET 012 protocol team received approval from its ethical review boards to continue enrolling women in the other two arms of the study to determine the best regimen to use in a redesigned efficacy trial.
More than 600 HIV-positive pregnant women and their infants participated in HIVNET 012. Half were randomly assigned to take a very short course of AZT -- two pills (600 milligrams) at the beginning of labor, then one pill (300 mg) every three hours until delivery, plus twice-daily doses of AZT (200 mg) syrup for their infants for one week. The other half took a single nevirapine pill (200 mg) at the beginning of labor, and their newborns were given one liquid dose (2 mg per kilogram) within 72 hours of birth. The health of each mother and infant was monitored carefully, and the infants were tested for HIV at six to eight weeks of age and again at 14 to 16 weeks.
At six to eight weeks of age, 12 percent of the infants who had received nevirapine had become infected with HIV, compared to 21 percent of those whose mothers had taken AZT. At 14 to 16 weeks, 13 percent of the infants in the nevirapine arm and 25 percent of the infants in the AZT arm were infected. Compared to the very short course of AZT, the nevirapine regimen had reduced the risk of HIV infection through 14 to 16 weeks of age by almost half.
The findings surprised even the members of the HIV Trials Network, who had hoped that nevirapine would perform as well or better than AZT because it is cheaper and easier to administer.
"Based on the pharmacokinetic data from our Phase I trial, we thought nevirapine would provide some protection," said Melissa Allen, a clinical research specialist at Family Health International, which has managed HIVNET's vaccine preparation studies and trials of other prevention methods in 12 international sites since 1993. "But we had no idea how dramatic the results would be."
The HIVNET results are not directly comparable with those of the CDC-sponsored trial in Thailand, which tested a longer AZT regimen among women who were not breast-feeding. Nevertheless, the HIVNET team believes that its data are strong enough to serve as the basis for several recommendations.
"The short-course regimen of nevirapine may be appropriate for women in both developed and developing countries who are first diagnosed with HIV very near to delivery and have received no other antiretroviral therapy," said Allen, a member of the HIVNET 012 protocol team. "In the context of future studies of other preventive interventions in settings where the longer-course antiretroviral regimens are not the standard of care, the nevirapine regimen could be offered to all participating women."
New Agenda
Because safe alternatives to breast-feeding are not widely available, affordable or culturally acceptable in many countries, one of the top priorities is to identify an intervention that will provide protection to infants during breast-feeding. The HIVNET is likely to undertake a trial to determine whether giving nevirapine to infants for the first few months of life reduces HIV transmission through breast-feeding.
Researchers will continue to follow the infants in HIVNET 012 for 18 months after delivery to monitor safety and also to determine the relative impact of the nevirapine and AZT regiments on long-term survival and on HIV transmission through breast-feeding after 14 to 16 weeks. Researchers say that a prolonged effect on transmission through breast-feeding is unlikely without additional doses of the drug.
Once the research protocol for an efficacy trial of nevirapine therapy during breast-feeding has been developed and approved by various technical and ethical review boards, Allen estimates that the actual research will take at least three years -- 18 months for enrollment and 18 months for follow-up. The length of the study will depend, she explained, on the number of mother-infant pairs needed to detect statistically significant differences between groups, the study design and the time it takes to recruit participants.
In the meantime, nevirapine is likely to be combined with efforts to offer HIV-positive women safe alternatives to breast-feeding. For example, as part of a pilot program sponsored by UNAIDS and UNICEF in Uganda, UNICEF will offer counseling and free formula to HIV-positive women who take antiretroviral drugs to prevent transmission of the virus to their infants.
UNAIDS, UNICEF and the World Health Organization will support pilot programs in 11 countries. They hope to include a total of 30,000 women a year in programs that provide early access to antenatal care, voluntary and confidential HIV counseling and testing, antiretroviral treatment for preventing perinatal transmission, counseling on infant feeding options, and support for women who decide not to breast-feed.
In Uganda the pilot program will soon begin in three sites, with plans to expand to an additional five sites, each serving 1,000 women a year. "The UNAIDS initiative is a good starting point to see whether what we're planning to do can actually be done in a non-study setting," Dr. Musoke said.
Catalyst for Change
For now, the Ugandan pilot program will use supplies of AZT donated by Glaxo Wellcome, the company that makes the drug. Once recommendations on the use of new antiretroviral regimens are in place, UNAIDS will begin negotiating with the manufacturer of nevirapine, Boehringer Ingelheim Pharmaceuticals, and other companies on ways to expand access to the drugs.
When they met in August, members of UNAIDS' International Working Group on Mother-to-Child Transmission and other experts on perinatal transmission examined the results of the Thai and Côte d'Ivoire studies, HIVNET 012 and the PETRA trials. The PETRA study found that 3TC added to the standard four-weeks of antenatal AZT treatment for mothers cut perinatal HIV transmission by half, while giving the two drugs to both mothers and newborns for a week reduced transmission by more than one-third.
Given the margins of statistical error for each study, however, it was impossible to say that one regimen was more effective than another. The group concluded that all five were effective in preventing mother-to-child transmission, and that decisions about which regimen to use or to make available in a country will depend on the setting, the circumstances and the resources at hand.
The group recommended that policymakers and health care providers consider all five regimens for different settings or patients, but recognized that short-course nevirapine may be the most feasible option in many countries. "We did say that nevirapine appears to be by far the least costly and most practical regimen, and that we need to make it available," Dr. Saba said.
For areas with high HIV prevalence and no access to voluntary HIV counseling and testing, some have even suggested that nevirapine could be offered to all pregnant women.
"This is something that theoretically could be feasible if we have enough safety data," Dr. Saba said. "But the benefits of voluntary counseling and testing go far beyond the prevention of mother-child transmission. If we are to really make a difference in this epidemic, we need to let people know whether they are infected."
Dr. Musoke agrees, saying that to offer nevirapine without voluntary counseling and testing would be to avoid "the real issue, which is to find those women who are HIV-positive and support them and find those women who are not infected and help them stay negative."
Presumptive nevirapine treatment might be cost-effective, she adds -- particularly in areas where every other client at antenatal clinics is HIV-positive -- "but I think it's a short-term investment."
Instead, both Dr. Musoke and Dr. Saba believe that nevirapine could serve as a catalyst for making desperately needed improvements in health infrastructure and services in sub-Saharan Africa and other areas.
"The fact that the drug is cheap makes it something we can strive for, but we should also be striving to improve infrastructure," Dr. Musoke said. "We should use this opportunity to improve the overall health care of all pregnant women."
