Two articles published in the August 5, 1999 issue of the New England Journal of Medicine report on two studies reporting that the level of HIV-RNA in a pregnant woman is predictive of the risk of perinatal transmission. In the first study of women who received AZT it was reported by Lynne Moffenson and others that HIV-RNA levels at delivery was the strongest predictor of the risk of transmission of HIV-1, although HIV-RNA at baseline was also a predictor of transmission. In the second study, among women who may or may not have been taking AZT, none of the 57 women with less than 1,000 copies/ml of viral load transmitted HIV to their infants. But several qualifying considerations are discussed including that several other studies have not found this correlation, and what is the long-term safety of HAART to pregnant women and their infants. Additional considerations are mentioned. In addition, a recent news release from the NIH reported that single doses of nevirapine to mother in labor and infant in first 72 hours following birth appreciably reduced transmission compared to a short-term AZT course of treatment for mother during labor and infant after birth.
Description of the Study. The study was a multi-center, randomized, controlled phase 3 trial conducted between October 1993 and March 1997 at 53 clinical sites in the US and Puerto Rico. They evaluated whether prophylaxis with AZT combined with HIV-1 hyperimmune globuline at a dose of 200 mg per kilogram of body weight administered intravenously to the women each month during pregnancy and once to the neonates at birth would lower the risk of perinatal transmission more than would AZT and intravenous infusions of immune globulin without HIV-1 antibody at a dose of 200 mg per kilogram. The results have been reported separately.
In the study which is the subject of this published article they enrolled HIV-1 infected women who were 20-30 weeks pregnant, who had CD4s of no more than 500, and who were receiving AZT as presrcibed by their physicians. Women continued the antepartum antiretroviral regimen and received intrapartum intravenous AZT (a loading dose of 2 mg per kg followed by a continuous infusion of 1 mg per kf per hour). Their infants received the standard six-week course of AZT prophylaxis (2 mg per kg orally four times per day. Nucleoside analogues other than AZT and NNRTIs were permitted with the approval of the protocol chair. Women receiving protease inhibitors were excluded from the analysis because there was no data available on their safety during pregnancy. None of the women breat-fed their babies.
Results. There was no perinatal transmission of HIV-1 among the 84 women who had HIV-1 levels below the limit of detection 500 copies/ml) at baseline or the 107 women who had undetectable levels at delivery. The authors said other studies have reported perinatal transmission among women with HIV-RNA levels below the limit of quantification who were receiving AZT. Therefore, the authors said, they could not conclude that there is a threshold for viral load below which there is no risk for perinatal transmission. As well, in the study under discussion in the NEJM, the upper limit of the 95% confidence interval for the risk of transmission was 3.5% among women who had undetectable levels at baseline and 2.8% among those with undetectable levels at delivery.
Although the HIV-RNA level at delivery was the strongest predictor of the risk of transmission in this study. They could not address whether a treatment-related reduction in HIV-RNA during pregnancy will further decrease the risk of transmission.
In sum the authors saidantiretroviral therapy that reduces HIV-RNA to below 500 copies/ml appears to minimize the risk of transmission as well as improve the health of the women. The Public Health Service Guidelines recommend that antiretroviral treatment of HIV-1 infected women should follow the same standards used for infected women who are not pregnant and for men. Thus, combination antiretroviral therapy is recommended for the treatment of HIV infected pregnant women with viral load above 10,000 to 20,000 copies/ml. Although further studies are necessary our data strongly suggest that the use of ART regimens that reduce viral load below detection may reduce the risk of transmission as well as improve the health of the woman.
In the second study Patricia Garcia and others similarly concluded that in pregnant women with HIV-1 the level of plasma viral load predicts the risk but not the timing of the transmission. They measured serially plasma viral load in 552 women with HIV and single-ton pregnancies. The status of the infection in their infants were assessed by culture of blood and further as early or late. Early was if a culture of blood obtained within the first two days of life was positive. Late was if a culture of blood obtained in the first 7 days of life was negative but subsequent cultures were positive. The rates of transmission were analyzed at various plasma viral load levels.
Among the 552 pregnant women plasma HIV viral load was measured a total of 1396 times at 4 various visits during pregnancy and delivery. The number of measurements was similar in the group with transmission and those who did not transmit. The individual viral load ranged from <400 copies/ml to 3.1 million copies/ml. Only 16 of 279 women (5.7%) with at least 3 values had persistently undetectable levels.
The median plasma viral load levels were higher among women who transmitted to their infants than among those who did not transmit (29,000 v 10,000). In general, the risk of transmission increased with increasing maternal viral loads. However, no threshold could be defined above which the risk of transmission was 100%. Women who were not treated with AZT and whose plasma viral load exceeded 100,000 copies/ml had the highest risk of transmissin (19 of 30 infants infected). Regardless of whether they received AZT or not, none of the 57 women with plasma viral loads below 1000 copies/ml transmitted to their infants. (upper limit of the 95% confidence interval, 5.1%). Women whose plasma viral load increased throughout pregnancy had the highest rate of transmission but it was not statistically significant.
Treatment with AZT was not associated with maternal viral loads but the receipt of any type of AZT therapy in general was significantly associated with a lower rate of transmission (15.2% vs 24.6%). The authors found no threshold level above which the rate of transmission was absolute. A number of women in the study had above 100,000 copies/ml of viral load and did not transmit HIV to their infant. Although there was no case of transmission of HIV when maternal viral load was <1000 copies/ml the authors cited 3 studies reporting transmission occurring when viral loads were at low levels. One study reported 16 of 132 (12%) who had viral loads less than 1000 copies/ml but transmitted HIV to their infants.
Several studies suggest that treatment with AZT during pregnancy reduces the risk of transmission, but not solely as a result of a reduction in maternal plasma viral loads. Data from several studies suggest that an AZT therapy should be included in treating a pregnant woman because it will add effectiveness in reducing the risk of transmission.
In this study, perinatal transmission occurred predomantly in the prepartum period. Although maternal plasma viral loads were closely associated with the risk of transmission, there was no discernible relation between the levels and the timing of neonatal infection. The relation between maternal plasma viral load and the risk of transmission may relate more to the mechanism of transmission (i.e., transplacental transfer of virus or exposure to virus in the genital tract) than to the timing of transmission.
These authors also suggest using Public Health Service Guidelines. They say there is general agreement that decreasing plasma viral load levels to below detection is the most appropriate goal in trying to prevent perinatal transmission. But they add that validating that goal is important, since treatment induced reduction in the viral load may not be equivalent to a spontaneously low level with respect to the risk of transmission. The authors say whether there is an additional benefit of elective cesarean delivery in decreasing the already low risk of transmission among pregnant women with low viral load has yet to be determined.
Editorial. On page 441 of the same NEJM issue Martha Rogers and Nathan Shaffer raise potential gaps they perceive in these studies. The studies by Mofenson et al. and Garcia et al. provide further evidence that one of the strongest predictors of both intrauterine and intrapartum transmission is the maternal plasma level of viral RNA. These two studies add to the evidence from studies in Thailand that the risk of transmission increases with the maternal plasma HIV RNA level. The Thai studies found that as much as two thirds of the risk of transmission was attributable to high plasma viral RNA levels. Taken together, these studies imply that therapy that maximally reduces the maternal viral level may be one of the most effective intervention strategies.
However, the data also clearly indicate that the risk of transmission is determined by multiple factors. Some transmission occurred even at low plasma levels of maternal viral RNA, and no upper threshold of maternal viremia was always associated with transmission. Likewise, there may be no lower threshold below which transmission never occurs, although in these two studies, as in the studies in Thailand, no transmission was observed among women with undetectable plasma viral RNA levels. Other studies have reported transmission by women with undetectable viral RNA levels, and the upper confidence limits for the transmission rates in women with very low or undetectable levels in the studies reported in this issue of the Journal were between 2.8 percent and 5.1 percent.
Both reports also imply that obstetrical management that emphasizes the prevention of premature birth, low birth weight, rupture of membranes more than four hours before delivery, and chorioamnionitis may further reduce the risk of transmission. Neither study was designed to assess the protective effect of elective delivery by cesarean section, which has been observed in other studies. A key research question that remains is whether there is any additional protective effect of delivery by cesarean section for women who receive potent combination antiretroviral therapy and have very low or undetectable plasma viral RNA levels.
ACTG Study 185 was unable to determine the efficacy of HIV hyperimmune globulin for women who did not breast-feed their infants and who were receiving antiretroviral therapy, because the rate of perinatal transmission in the control group was so low. In resource-poor countries, a yet-unanswered question is whether short courses of peripartum antiretroviral therapy followed by passive immunization with HIV hyperimmune globulin or active immunization of infants with an HIV vaccine can provide protection during lactation for infants of women who breast-feed.
Neither of the two studies in this issue of the Journal examined the role of viral levels in genital secretions. Although viral levels in genital secretions correlate with plasma levels, one study found that women with high genital viral levels are at greater risk of perinatal transmission than those with low levels after adjustments for plasma viral level are made; this finding suggests that the strategy of lowering the viral level in genital secretions may provide additional protection. This possibility should be examined further.
As was the case with ACTG 076, neither study was able to determine the relative importance of 2 strategies being usedreduction of maternal viral load and prophylactic treatment of infants. Prophylactic treatment of infants may be effective in addition to or even in the absence of maternal therapy. In an observational study, Wade et al found a lower rate of infection among infants who were treated with AZT for 6 weeks, starting within the first 48 hours of life, than among infants who did not receive such treatment (9.3% vs 26.6%).
Several questions concerning prophylactic treatment of infants remain. How effective is it for the infants of women who do not receive antenatal therapy? What additional efficacy does prophylactic therapy for infants provide when it is coupled with antenatal treatment? What is the optimal duration of therapy? How soon after birth must prophylactic therapy be initiated? What is the most effective drug? I think this is a way in which nevirapine may be used.
In the United States, the current strategies recommended to reduce perinatal HIV transmission include universal voluntary HIV counseling and testing of pregnant women; treatment with zidovudine, given to infected mothers antenatally and during the intrapartum period and to their newborns; obstetrical management that attempts to minimize the infant's exposure to maternal blood and genital secretions; and avoidance of breast-feeding. It has also been recommended that the choice of antiretroviral therapy for pregnant women should be based on the same considerations used for women who are not pregnant, with discussion of the known and unknown risks and benefits of such therapy during pregnancy. The studies reported in this issue of the Journal and those published elsewhere strongly suggest that a maximal reduction of maternal plasma viral levels will further reduce the risk of perinatal transmission, in addition to prolonging the survival of the mother.
The authors of the editorial say that a careful discussion of the risks and benefits of potent ART therapy during pregnancy should be conducted because little is known about the safety of many antiretroviral drugs during preganancy. In addition, the exposed children have not been followed long enough for any long-term risks to be measured. Many challenges remain: ensuring the availability of prenatal care for high-risk populations (such as women with substance abuse and adolescent girls), making HIV screening available or standard of care for all pregnant women, monitoring and evaluating the effect of drug resistance and transmission of drug resistance, evaluating potential toxicities of ART, improving adherence.
Nevirapine. A point I would like to add. These published articles were probably submitted before they could consider the NIH news of a joint Uganda-US study. The study reported that a single dose of nevirapine to the mother in labor and a single dose to the baby within 3 days after birth reduced the rate of transmission appreciably compared to a similar short cohort of AZT. The short course nevirapine resulted in a 47% reduction in mother-to-infant transmission compared with the short course of AZT used. Since the first data from this study was just recently reported, it remains to be seen just how researchers can translate these data into models of clinical application. Possibly a single dose of nevirapine for the mother and for the infant after birth could follow a HAART treatment for a pregnant mother.
The nevirapine regimen consisted of a single 200 mg tablet given to mothers in labor and a single 2 mg/kg dose of nevirapine oral suspension to the newborns within 72 hours after delivery. The AZT regimen was 600 mg at the onset of labor, 300 mg every 3 hours during labor, and 4 mg/kg of AZT twice daily to the newborn for the first week after delivery. The trial was conducted in Uganda by HIVNET (HIV Prevention Trials Network) and was sponsored by the US NIAID. Follow-up studies are ongoing. The NIH reported that based on average US wholesale costs, the cost of drug used in the nevirapine regimen was about 200 times cheaper than the long-course AZT used in the US and almost 20 times cheaper than a short course of AZT given to the mother during the last month of pregnancya regimen tested in Thailand by the CDC and reported effective in 1998. I think we need to conduct comparative studies of cost and effectiveness for various regimens.