Pregnancy common in ART trials in sub-Saharan Africa despite exclusion criteria
27 February 2017. Related: PMTCT and maternal health.
Polly Clayden, HIV i-Base
A new analysis from two ANRS studies reports that HIV positive women in clinical trials in sub-Saharan Africa are likely to get pregnant as often as those receiving care outside of research settings and questions why pregnancy is still commonly an exclusion criterion.
Instead, the investigators say it is essential to take a pragmatic approach and revisit the relevance of the criteria for exclusion of pregnant women in such trials.
An increasing number of ART clinical trials are now conducted in sub-Saharan Africa, a region with the highest fertility rate in the world and a very high social value associated with child bearing.
Pregnancy and breast feeding are exclusion criteria in the majority of antiretroviral therapy (ART) clinical trials. Women of child bearing age agree to defer pregnancy for the duration of the study and are counselled before enrolment on using dual protection (condoms and a non-barrier contraceptive method), which are usually provided free.
A previous study conducted in Botswana reported a pregnancy rate of 7.9 per 100 person years (PY), despite women’s verbal agreement to defer childbearing until after the study was closed.
Investigators from ANRS12169-2LADY and ANRS12286-MOBIDIP trials – comparing three boosted PI regimens, and the efficacy of a mono or dual-therapy of PI with or without 3TC respectively, conducted in Cameroon, Senegal and Burkina Faso – described their experience of pregnancies among women participants. These findings were published online in HIV Clinical Trials, 1 November 2017.
The goal of the study was to describe the reproductive behaviour and pregnancy outcomes among women on second-line ART enrolled in the trials and compare them with those of HIV positive women in non-research settings.
The investigators reported 66 women had 84 pregnancies between January 2010 and July 2015 (1046.3 PY of follow up). The majority (51) of women had one pregnancy during follow up, 12 had two and three had three.
Compared with the other women participants in 2LADY, women who became pregnant were: younger (31 vs 36 years, p<0.001), less likely to be single (24.2% vs 40.1%, p=0.005), were more likely to have disclosed their HIV status to their partner (95.5 vs 31.7, p<0.001), and had been receiving ART for a shorter time before enrollment (3.5 vs 4.2 years, p<0.001). There were no differences in CD4 or viral load at enrollment.
Sixty pregnancies (71.4%) were in women receiving lopinavir and 27 (32.1%) in those receiving darunavir.
The investigators noted that 13 of 66 women who became pregnant had received medroxyprogesterone before pregnancy. But these women received a median of only one dose. Seven women received medroxyprogesterone or a levonorgestrel-releasing implant after pregnancy.
The overall pregnancy rate (per 100 PY) was 8.03 (95%CI: 6.5 to 9.9). Twenty women (7.1%), including two pregnant women lost to follow up at time of analysis. The pregnancy rate was highest in Senegal, 10.0 (95%CI: 6.4 to 15.7) vs Cameroon, 7.5 (95%CI: 5.6 to 9.9) and Burkina Faso, 7.9 (95%CI: 4.9 to 12.7).
The median time from enrollment to first pregnancy was 1.7 months. The investigators found that the incidence of pregnancy was stable during the first two years, then declined, and peaked again after four years of follow up – partially because of recurrent pregnancies.
The 84 pregnancies resulted in: 60 (73.2%) live births, 13 (15.8%) miscarriages, three (3.6%) stillbirths, two (2.4%) extra-uterine pregnancies and four (4.8%) voluntary abortions. Two women were lost to follow up. The overall fertility rate was 5.73 live births per 100 PY (95%CI: 4.45 to 7.39).
The median gestation was 38 weeks (IQR 37 to 40) and median birth weight 2.9 kg (IQR 2.6 to 3.2). Nine (15%) infants had low birth weight, four of them were also premature. The investigators only recorded one birth defect (ankyloglossia). Four infants needed resuscitation at birth; one died from respiratory disease. One woman died of bleeding at delivery.
In multivariate analysis, miscarriages/stillbirths were not associated either with age, CD4 nadir, duration of ART, CD4 count, or viral load at the start of pregnancy. The investigators noted a trend in an increased proportion of miscarriages/stillbirths with darunavir vs lopinavir exposure: OR 3.1 (95%CI: 0.9 to 10.1).
The investigators looked at the scarce published data on the reproductive behavior of HIV positive women in West and Central Africa, and found the few studies available to be very heterogeneous in terms of population sample, types of data collected, study period, and ART availability, making comparisons hard. But the reproductive behaviour of HIV positive women enrolled in clinical trials in sub-Saharan Africa appeared to be similar to that of women in non-research settings.
They concluded that this finding needs to be taken into account when planning trials in such regions with high fertility rates. The option to get pregnant in a trial should be discussed in ethics committees in the context of the level of risk associated with investigational antiretrovirals during pregnancy. Family planning counselling and contraception options need to improve.
The investigators also noted that the reported rate of miscarriage/stillbirth in this study is not different from those reported in HIV negative women, which is reassuring for HIV positive women considering pregnancy and receiving ART.
This study was conducted in African countries with some of the highest fertility rates in the world: in Burkina Faso and Senegal women have an average of almost 6 and almost 5 children respectively. So, the finding that women of child-bearing age get pregnant in clinical trials with long follow up is unsurprising.
The more salient point to take from this article is that pregnancy is frequently an exclusion criterion for ART trials and women who conceive during this period are often switched from an investigational drug or regimen, sometimes not until later in pregnancy, when it is detected.
In some cases it might be more appropriate to continue on the study drug (in the context of level of risk/benefit) with careful monitoring, and this needs to be taken into account when designing trials that include women of child-bearing age.
If women do continue on study drug in pregnancy, it is important that the data on maternal and infant outcomes is captured. New antiretrovirals always have scant information to guide recommendations in pregnancy after approval for high-income countries and this can delay their inclusion in global guidelines.
Serris A et al. Getting pregnant in HIV clinical trials: women’s choice and safety needs. The experience from the ANRS12169-2LADY and ANRS12286-MOBIDIP trials. HIV Clinical Trials. Vol 17, 2016. Issue 6. Published online 1 November 2017.