Children born to HIV-positive women who take efavirenz
(Sustiva or Stocrin, also a component of Atripla)
during pregnancy are at greater risk of developing neurological disorders, some
of which may not be diagnosed until years after birth, according to a
late-breaker presentation at IDWeek 2018 last week in San Francisco.

The
researchers saw only a non-significant suggestion of an association between
neurological problems and exposure to dolutegravir (Tivicay, also in Triumeq
and Juluca), and none of the four
affected children had neural tube defects, which were a subject of concern earlier
this year.

Antiretroviral therapy (ART) has dramatically lowered the risk of
mother-to-child HIV transmission worldwide. But the long-term effects in
children who were exposed to these drugs during gestation—most of whom do not
become HIV-infected themselves­—remain to be determined. Most previous studies
have focused on congenital abnormalities and other adverse birth outcomes, not
problems that may develop later on.

In May, regulatory agencies in the United States and the
European Union warned
that women with
HIV who could potentially become pregnant should not use dolutegravir, after
researchers reported a higher — though still very low — risk of neural tube defects, such as spina bifida, in infants born to women in Botswana who were
taking this drug at the time of conception.

At the
International AIDS Conference in July, the World Health Organization (WHO)
issued new treatment guidelines recommending dolutegravir-based ART
as the preferred first-line treatment option for adults, adolescents and
children, including women with access to reliable contraception. Women from
Africa made it clear that they did not want their access to this effective and
well-tolerated medication to be limited because of only four cases of
neurological birth defects.

At
IDWeek, Claudia Crowell of the University of Washington and Seattle Children’s
Hospital presented findings from an analysis of 3,747 HIV-negative children
born to HIV-positive women in SMARTT (Surveillance Monitoring for ART
Toxicities), an observational cohort study conducted by the Pediatric HIV
Cohort Study network, which includes 20 sites in the United States, between
2005 and 2017.

About 70%
of the children were black, 26% were white and 31% were Hispanic. Most were
enrolled at birth and they were followed for the first two years after birth or
study entry.

Exposure
to protease inhibitors was most common (70%), followed by non-nucleoside
reverse transcriptase inhibitors (19%, including 4.5% exposed to efavirenz) and
the newer integrase inhibitors (11%, including 2.6% exposed to dolutegravir). Most
of the mothers (87%) had viral suppression (1000 copies/ml) and
well-preserved immune function at the time of delivery. Tobacco, alcohol and
illicit drug use were reported by 17%, 8% and 9%, respectively.

The
researchers used tracking forms and head circumference measurements to identify
suspected cases of neurological disorders, which were reviewed by experts who did
not know whether the child had been exposed to antiretrovirals.

Out of
the 3,7474 evaluable children, 237 were diagnosed with 287 neurological
conditions, a prevalence rate of 6.3%. The children were two years old, on
average, when neurological problems were confirmed. Half of the conditions were
diagnosed after age two, demonstrating the importance of longer-term follow up.

Microcephaly,
or small head size, accounted for a quarter of the disorders, followed by
febrile seizures (18%), eye abnormalities (17%), non-febrile seizures or
epilepsy (13%), and various uncommon conditions occurring in 5% or less.

Overall,
the strongest risk factor for neurological abnormalities was tobacco use,
especially during the first trimester, which nearly doubled the risk of neurological
problems. Use of alcohol or illicit drugs were no longer significant after
adjusting for tobacco use.

Among the
children with neurological disorders, 16 had been exposed to efavirenz. The
prevalence in this subgroup was 9.6%, compared with 6.2% for those not exposed
to the drug, indicating a 60% higher risk. The overall association did not
quite reach statistical significance, but it did so in various sensitivity
analyses adjusting for factors such as duration of follow-up and presence of
congenital anomalies. Previous studies have not seen a link between efavirenz
use and adverse outcomes at birth, but did not look at longer-term
developments.

The study
also saw a “suggestion of an association” with dolutegravir, but this
was not statistically significant. Only 94 children were exposed to this newer drug,
of whom four were diagnosed with neurological problems — none of which were neural
tube defects of the type seen in the Botswana study.

“We need larger cohorts, with more children
exposed to newer agents, and we need to continue to follow them to find out the
long-term clinical implications,” Crowell said at an IDWeek press
briefing. “We still haven’t determined the best antiretroviral regimen in
pregnancy.”