Pregnancies among high-risk women in the first ever HIV vaccine efficacy trial.

Abstract

Concerns about the unknown effects of an investigational agent on a fetus, potential risks to the future reproductive capacity of female participants, and the fear of legal repercussions [1] have resulted in caution when enrolling women into clinical trials. HIV vaccine trials are no exception. As a result of similar concerns, pregnant women are excluded from enrolling, and women of reproductive potential are expected to use contraception for the duration of the trial. Although HIV vaccine trials have been ongoing since 1988, most of these have been small phase I and II trials involving women and men at low risk, and pregnancies occurred infrequently. In a review of safety data from 25 trials, pregnancy prevalence was 2.9% for all studies combined [2]. The enrollment of a larger than ever sample of high-risk women in the first ever HIV vaccine efficacy trial – the results of which are expected in 2003 – provides a unique opportunity to assess the incidence of pregnancy in this cohort. Sponsored by VaxGen, Inc., Vax004 began in 1998 and enrolled over 5300 high-risk volunteers. Most were men who have sex with men, but a small proportion were women at risk through sex with men [3]. Women were eligible to enrol if they tested HIV antibody negative and reported one or more of the following: a current HIV-infected male sex partner or, in the past 12 months, two or more male sex partners and a sexually transmitted disease, five or more male sex partners, or crack use. Participants received injections seven times over 3 years. For women, before each injection, urine pregnancy tests were conducted in the absence of a tubal ligation or hysterectomy. When pregnancies occurred, injections were discontinued (with the exception of one miscarriage and two terminations), but all participants were followed for the study duration. The monitoring of pregnancy outcomes is ongoing. Pregnancy incidence was calculated on the basis of person-years of follow-up using the last menstrual period plus one month. Only women aged 18–45 years and followed for 18 months are included here. Over half (57%) of the 268 women in the sample were African American, 28% were white and 12% were Latina. A third (32%) had less than a high school diploma and 39% were younger than 35 years. At baseline, over half (52%) reported using crack and 38% reported five or more sex partners in the previous 6 months. Thirty-seven per cent reported a current HIV-infected partner, and 5% reported two or more partners and a sexually transmitted disease in the previous year. Fifteen pregnancies occurred over 18 months. In the first 6 months, there were eight pregnancies; from 6 to 12 months, six more pregnancies; and between 12 and 18 months, there was one additional pregnancy. The incidence of pregnancy was 4.6 per 100 person-years (95% confidence interval 2.8; 7.6). Two-thirds who became pregnant were African American (67%), a fifth were Latina (20%), and 13% were white. Approximately half (47%) had a high school diploma or its equivalent. Those who became pregnant were more likely than others to be less than 35 years of age (80 to 39%) and at baseline to report unprotected vaginal sex (80 to 55%) in the previous 6 months, but because of the small number of pregnancies, tests of statistical significance were not possible. As the number of HIV infections among young women in the United States infected through sex with men continues to rise [4], pregnancies among young women who participate in trials are also likely to increase. Furthermore, the pregnancy incidence rate of 4.6% found here is considerably higher than the HIV incidence rate of 1.13% reported among women at heterosexual risk of HIV infection who participated in a US vaccine preparedness study [5]. This suggests that in future trials with high-risk women, pregnancy incidence should be taken into consideration when the sample size is being determined. Regarding safety, there is no reason to believe that HIV vaccines pose any health threat to pregnant women, but sufficient data to ascertain this are not yet available. Furthermore, given the small number of pregnancies that occurred in this trial, results from the pregnancy outcome data may be limited. Prioritizing the pregnancy issue in future HIV vaccine trials by involving greater numbers of high-risk women, retaining them over time, and measuring tubal ligations and hysterectomies will contribute to our being better able to assess the impact of HIV vaccines on pregnant women and their unborn children.