International AIDS Society Newsletter, no. 16

IAS
INTERNATIONAL
SOCIETY
se of Nevirapine to Reduce Mother-toChild Tr ansmission of HIV (M TCT)WHO Review of Reported Drug esistance
Department of Reproductive Health and Research, World Health Organization, Geneva, Switzerland.
For further details contact Dr TMM Farley. E-mail: [email protected], Tel: [phone], Fax: [phone]

A recent early report that a single dose of
Nevirapine given to HIV-infected women
during labour to reduce the risk of mother-tochild transmission (MTCT) of HIV can lead to
the selection of Nevirapine (NVP)-resistant
virus at 6-weeks in the mother does not warrant exclusion of NVP from pilot MTCT-prevention programmes or research studies. Further information on the safety and effectiveness of NVP is required before recommendations on wide-scale use of NVP can be made.
In the past two years, evidence has accumulated
that short and cost-effective antiretroviral regimens can reduce the transmission of HIV-1
from mother to child during labour and delivery. Results from the first of these studies
demonstrated that a one-month course of
Zidovudine (ZDV) used from 36 weeks' gestation until delivery was well-tolerated, safe and
reduced the risk of transmission by half in
women who did not breastfeed. Following
these results, recommendations on the use of
such a ZDV regimen were published in the
WHO Epidemiological Record in October 1998
[1].
In 1999, clinical trials demonstrated the efficacy of 1-month ZDV regimens in predominantly breastfeeding populations, the efficacy of a
regimen combining two antiretrovirals (ZDV
and Lamivudine [3TC]) as well as the efficacy
of NVP given as a single dose to the mother at
the beginning of labour and a single dose given
to the infant within 72 hours of life. There is
little to differentiate between currently available
regimens (ZDV from 36th week to delivery,
ZDV plus 3TC from onset of labour to 1 week
post-partum, one dose of NVP at onset of
labour and one to the new-born) in terms of
efficacy. The low cost of NVP and simplicity of
administration make it a good candidate for
low-income countries where women generally
attend antenatal care irregularly or late during
pregnancy, or where many deliveries occur in
home settings [2, 3].

Safety data are mainly available for ZDV and
3TC, which have been widely used for several
years in developed countries, although for
longer therapeutic protocols. To date, experience with single dose NVP administered to
mother at onset of labour and to the neonate is
based on over 700 mother-infants pairs enrolled in two Phase I safety studies conducted in
the USA and Uganda, and an efficacy trial in
Uganda. More extensive safety data are available with use of NVP and ZDV for long-term
treatment. During use of both ZDV (alone or
in combination with 3TC) and NVP for longer
periods than that recommended to prevent
MTCT in low income countries, rare but severe
cases of toxicity have been observed (mitochondrial disease and Stevens-Johnson syndrome) [2].
The recent report from a phase II study (HIVNET006) showing that a single dose of Nevirapine led to the selection of NVP-resistant virus
in 3 out of 14 women at 6 weeks post-partum
raised concerns about the potential adverse
effects of NVP on the mother and infant health.
WHO convened a Technical Consultation of
investigators, virologists, epidemiologists,
representatives of the manufacturer, representatives from NIH, CDC, UNAIDS and UNICEF
on 16 February 2000 to explore the potential
implications of this report. A confidential copy
of the full report from the study was made
available to the meeting participants. The four
main conclusions of the Technical Consultation
were:
1. Rapid emergence of NVP resistant virus has
been previously observed in adults receiving
chronic NVP treatment as a single agent [4].
Given the pharmacokinetics of the drug
(long half-life) and its high potency, selection
of resistant virus even after a single dose was
not wholly unexpected. Indeed, the profound decrease in the quantity of drug-sensitive virus particles strongly favours replication of resistant virus present before treatment.
U

2. Since NVP is given at onset of labour, the
consultation considered it unlikely that selection of resistant virus occurred in the few
hours before delivery. It is also unlikely that
resistant virus be transmitted during labour
or delivery to infants infected despite treatment, given that resistant virus would probably not be selected until at least several days
following delivery. However, the rapid selection of resistant virus may compromise the
efficacy of the NVP prophylactic dose given
to the infant and the likelihood of transmission of resistant virus during breastfeeding is
of greater concern. In addition, the use of
NVP may be less effective in reducing the risk
of transmission during a subsequent pregnancy in women in whom resistant virus was
selected in their first pregnancy.
3. Although the infectivity and pathogenicity of
selected NVP-resistant variants is unknown,
previous studies on other drug-resistant
viruses do not suggest that such variants are
more transmissible or worsen the prognosis
of HIV infection. In addition, resistant virus
is expected to reach undetectable levels within a few weeks or months after cessation of
treatment [5]. The temporary selection of
resistant virus induced by a single dose of
NVP during labour is therefore not expected
to lead to emergence of NVP-resistant strains
in the population. Of greater concern is
potential emergence of NVP-resistant strains
in the population if NVP were used in suboptimal treatment regimens.
4. The implications for future therapeutic
options for women and children with resistant virus were discussed. The efficacy of
combination therapies not including NVP or
other Non-Nucleoside Reverse Transcriptase
Inhibitors (NNRTI) antiretrovirals should
not be altered in such patients. In theory
potent combination therapies which include
NVP or another NNRTI should not allow reemergence of resistant virus, but the use of a
NNRTI in sub-optimal combinations should