Since 1998, pregnant women in the UK have been encouraged to take the anti-HIV treatment they require regardless of their pregnancy - with the notable exception of efavirenz-containing HAART - in order to prevent vertical (mother-to-infant) transmission of HIV.
However, zidovudine (AZT) is the only drug fully approved for use in pregnancy and the most recent British HIV Association (BHIVA) guidelines advise that women with CD4 counts that would not merit treatment for the mother (above 200-250) and low viral load (below 10,000 copies) should be offered AZT. This is according to results of the 076 protocol, which proved in 1994 that AZT could reduce mother-to-baby transmission of HIV among pregnant women with CD4 counts above 200 who had not taken AZT before. Unlike the 076 protocol, however, the BHIVA guidelines suggest deferring treatment until the third trimester of pregnancy to limit the risk of drug resistance and reduce foetal exposure.
Balancing the interests of mother and child is difficult, and there has been considerable criticism of the practice of continuing to offer AZT monotherapy during pregnancy, particularly since a 2001 study found that a single dose of nevirapine given at the onset of labour is almost 50% more effective than four weeks of AZT treatment at preventing transmission of HIV from mother to baby, and 70% cheaper. And so a small British study, comparing the efficacy of AZT monotherapy pre- and post- 1998 guidelines is welcome additional information that appears to show that when used in accordance to these guidelines, AZT monotherapy is safe and effective for both mother and infant.
In order to compare both the emergence of resistance patterns and the efficacy in preventing vertical transmission between the unselective use of AZT monotherapy prior to 1998 with selective AZT monotherapy after 1998, a group of London researchers identified 91 pregnant HIV-positive women at four HIV treatment centres. These mothers-to- be had been prescribed AZT monotherapy between January 1996 and March 2001 to prevent vertical infection, the majority of whom were African (84%) and infected with a non-B HIV subtype (82%).
The only significant difference (p=0.001) between those treated prior to 1998 (n=31) and those after 1998 (n=60) was the median baseline CD4 count, which was lower prior to 1998 (299 versus 463 cells/mm3). For the cohort as a whole, AZT monotherapy was begun at a median of 28 weeks into pregnancy (range 15-37), with a median exposure of 9.5 and 9 weeks for the pre- and post-1998 groups, respectively.
Stored blood plasma taken at any time between time of delivery and four weeks post delivery was available for 65 of the 91 women, although, since viral load was below the limit of detection in 16, only 49 samples were suitable for genotypic resistance testing, of which reverse transcriptase (RT) resistance data was available for 40.
Resistance was seen only in the pre-1998 group (n=19); no mutations were detected in any of the 21 samples post-1998.
In the pre-1998 group four had RT mutations at the time of delivery. One woman who began AZT monotherapy at 15 weeks’ gestation (25 weeks total exposure) - much earlier than currently recommended - had the T215Y mutation, which confers high-level resistance to AZT. A second woman with the T215Y mutation received AZT for 14 weeks and had a high viral load at delivery (350,000 copies/ml). An M41L mutation, associated with low-level AZT resistance, was present both at baseline and delivery in a woman with previous AZT exposure. And the fourth woman, who was NNRTI-naive, was found to have an A98G mutation, associated with reduced sensitivity to NNRTIs.
Again, transmission was only seen in the pre-1998 group. Of the three transmissions seen, two were associated with incomplete adherence and the third was seen in the mother with the T215Y mutation and high viral load, who transmitted wild-type rather than resistant virus to her infant.
Given that neither resistance nor transmission were seen in any of the women after 1998, when AZT monotherapy was restricted to women with low viral loads and better immune function, the authors argue that “zidovudine monotherapy remains an attractive intervention in this setting, as it limits foetal drug exposure and is well tolerated.”
However, they point out that since the number of women included in their study was small, and that current genotypic assays might not pick up minority HIV strains, “the upper limit of the 95% confidence interval would suggest that up to 16.1% of women in a similar situation could possess such mutations” and that women taking AZT monotherapy to prevent vertical transmission should be closely followed “if they subsequently start combination therapy to confirm they are not at a higher risk of treatment failure.”
Further information on this website
Long-term evidence for effectiveness of nevirapine in reducing mother-to-child HIV-1 transmission - news story
Mother to baby transmission (pdf file) - factsheet
Larbalestier et al. Drug resistance is uncommon in pregnant women with low viral loads taking zidovudine monotherapy to prevent perinatal HIV transmission AIDS 17: 2665-2667. 2003.