Efavirenz-based ART matches lopinavir/ritonavir for perinatal HIV treatment

Pregnant women taking efavirenz-based antiretroviral therapy had significantly better virologic outcomes at the time of delivery compared to those taking lopinavir/ritonavir in a randomised study in rural Uganda, Dr Deborah Cohan, reporting on behalf of the PROMOTE study team, told attendees at the 21st Conference on Retroviruses and Opportunistic Infections (CROI) in Boston last week.

However both regimens, overall, had very high virologic suppression rates and were extremely effective in preventing HIV transmission during pregnancy and breastfeeding Dr Cohan noted.

These data provide reassurance that high levels of viral suppression are achievable, [and] demonstrate that infants have a low risk of HIV acquisition with these regimens.

The findings from this secondary analysis of a randomised open-label study evaluating efavirenz (Sustiva, Stocrin) compared to lopinavir boosted with ritonavir (Kaletra) for the prevention of placental malaria, support the 2013 World Health Organization (WHO) guidelines recommending efavirenz-based combination anitretroviral therapy (cART) for all pregnant women regardless of CD4 count (known as 'Option B+') as a first-line option with cART based on lopinavir/ritonavir as an alternative.

These new guidelines, however, were recommended within the context of significant research gaps that included maternal and infant outcomes, antiretroviral toxicity and alternative ART regimens.

PROMOTE enrolled pregnant women living with HIV who had not previously taken HIV treatment at between 12 and 28 weeks of gestation. The women were randomised to receive AZT (zidovudine, Retrovir) and 3TC (lamivudine, Epivir) with either lopinavir/ritonavir or efavirenz at enrolment up until one year of breastfeeding and followed-up for six weeks after cessation of breastfeeding. The dose of lopinavir/ritonavir increased from 400mg/100mg twice daily to 600mg/150mg twice daily at thirty weeks. All women received cotrimoxazole prophylaxis and insecticide-treated bednets, and infants received AZT or nevirapine (Viramune) prophylaxis in accordance with Ugandan guidelines. Women received nutritional counselling to breastfeed for one year after delivery.

This substudy compared virologic, immunologic and safety outcomes between the two arms of the study.

Of the 593 women screened, 389 were eligible and enrolled, of whom 348 completed the study, and data were available for 377 at delivery.

At baseline, there were no significant differences in the two arms of the study. Mean age was 29 years and median gestation at enrolment was 21 weeks. This was a first pregnancy for fewer than 10%, with over two-thirds having three or more children at home. Median CD4 cell count in both arms was over 350 cells/mm³ yet over 90% were WHO stage 1. Malnutrition was a significant issue, with a mean body mass index (BMI) at enrolment of 21.8. At baseline, median log₁₀ HIV RNA was 4.3 and 4.1 for the efavirenz and lopinavir/ritonavir arms, respectively.

Women taking efavirenz were significantly more likely to achieve viral suppression (<400 copies/ml) at delivery compared to women on lopinavir/ritonavir, 98% (166/170) and 86% (153/178), respectively (p<0.001). However, no differences were seen at eight weeks after starting ART while pregnant, with close to 90% of women achieving viral suppression in both arms of the study, and with similar findings at weeks 24 and 48 after delivery.

Looking at all viral loads measured (1335 measurements in 374 women), a 50% lower odds of viral suppression among women on lopinavir/ritonavir (OR:0.51, 95% CI: 0.31-0.82, p = 0.0062) was seen. Dr Cohan noted, however, this is within a context of overall very high levels of virologic suppression for both regimens.

Women taking lopinavir/ritonavir had greater CD4 cell count recovery compared to women taking efavirenz at delivery, (+57 and -7 cells/mm³, p = 0.002, respectively) and at 24 weeks postpartum (+178 and +109 cells/mm³, p<0.01, respectively),

WHO grade 4 adverse events were rare and did not differ by arm. Kaposi’s sarcoma was diagnosed in one woman in the efavirenz arm and pulmonary TB in one woman in the lopinavir/ritonavir arm. However, diarrhoea and nausea and vomiting (grades 1 and 2) were significantly more likely among women taking lopinavir/ritonavir both before and after delivery.

There were no differences in preterm birth, miscarriage, stillbirth or neonatal death between the arms.

The HIV transmission rate was 0.5% (2/374 live births). Both were in the lopinavir/ritonavir arm, one in utero and one during breastfeeding. HIV-free survival was high and did not differ between the efavirenz and lopinavir/ritonavir arms, 97.2% and 92.9%, p = 0.10, respectively. Grade 3 and 4 events, mostly anaemia and neutropaenia, were similar between the arms; the incidence rate ratio (IRR): 1.25; 95% CI: 0.87-1.81, p = 0.21 (lopinavir/ritonavir compared to efavirenz).

Differences at delivery may potentially be explained by efavirenz being more potent than lopinavir/ritonavir as well as by adherence and drug exposure during pregnancy Dr Cohan noted. Self-reported adherence rates were high in both arms but self-reporting is vulnerable to social desirability bias, she added.

While pharmacokinetic (PK) studies suggest lopinavir/ritonavir exposure may be inadequate before delivery, it is unclear whether this explains the differential virologic suppression Dr Cohan said.

Dr Cohan concluded “these data provide reassurance that high levels of viral suppression are achievable, demonstrate that infants have a low risk of HIV acquisition with these regimens, and show that women can successfully initiate ART when they present to the antenatal clinic and maintain therapy thereafter.”

She added “the ultimate goal of lifelong suppressive therapy is to keep these women healthy.”