Lopinavir/ritonavir equivalent to nevirapine in Ugandan children

A ritonavir-boosted lopinavir (LPV/r)-based regimen achieved a comparable rate of virologic suppression when compared to a non-nucleoside reverse transcriptase inhibitor (NNRTI)-based regimen in HIV-infected Ugandan children at 48 weeks, with comparable immunological responses and adverse events, researchers reported at the 20^th Conference on Retroviruses and Opportunistic Infections(CROI 2013) last week in Atlanta.

These findings from the Prevention of Malaria and HIV disease in Tororo (PROMOTE) paediatrics study are part of a wider programme to establish new ways to reduce HIV and malaria burdens in sub-Saharan Africa. 

The researchers believe these results will contribute to the discussion of the potential for a wider role of LPV/r in the treatment of HIV-infected African children, particularly in areas where malaria is endemic. Findings from the original study reported at last year’s conference showed that LPV/r was associated with a lower incidence of malaria (where artemether-lumefantirine is the malaria treatment of choice).

Current guidelines restrict the use of LPV/r to infants exposed to nevirapine (NVP) in the context of PMTCT or as second-line treatment following failure. Data from the P1060 study show that LPV/r may have a greater role in the treatment of infants (LPV/r had a lower failure rate compared to nevirapine among NVP-exposed and unexposed infants).

At the same session Norah Mwebasa also from the PROMOTE-paediatrics group reported the protective benefits of a PI based-regimen (LPV/r) against malaria re-infection, and demonstrated that  an NNRTI-based regimen (efavirenz) significantly reduced exposure to antimalarial components suggesting a higher risk for failure among HIV-infected children.

Tororo, Uganda is a rural area with an HIV prevalence rate of 8% and malaria transmission intensity of 562 infective bites per personyear.

There were an estimated 219 million cases of malaria worldwide in 2010. Malaria mortality rates have fallen by more than 25% globally and by 33% in the World Health Organization (WHO) African Region since 2000. However, most deaths are in children living in Africa where a child dies every minute from malaria.

In areas of moderate or intense transmission conditions adults will develop partial immunity over years of exposure. While it does not give complete protection, partial immunity reduces the risk of malaria infection causing severe disease. This is why most deaths from malaria in Africa are in young children. In addition people with weakened immune systems as well as those from non-endemic areas will also be at increased risk. So HIV-infected children are especially vulnerable.

From October 2009 to October 2011 the researchers enrolled 185 ART-naïve and ART-suppressed (HIV RNA under 400 copies/mL) HIV-infected Ugandan children from two to under six years of age in an open-label randomised trial. 92 received LPV/r and 93 an NNRTI-based regimen with NVP or efavirenz if three years of age or younger.

Children under two years of age with perinatal exposure to NVP were excluded.

This was a planned non-inferiority (NI) analysis of the virological efficacy of LPV/r, with the primary outcome the proportion of children with HIV RNA under 400 copies/mL at 48 weeks.

The original study was powered to address the impact of the ARV treatment regimen on malaria incidence in children. So the researchers based this analysis on a pre-specified NI margin of 11%. Secondary outcomes included time to virological suppression; change in CD4 count and percentage from baseline to 48 weeks; and the proportion that experienced grade 3 and 4 adverse events (AEs).

The children in both arms had similar characteristics.49% were female, median age was 3.1 years (0.4-5.9) and 71% were ART-naïve.

CD4 counts were in the 500 and 1000 range for ART-naïve and ART-suppressed, respectively, and similarly, CD4 percentages were in the 16 and 30 range in each arm.

Median HIV RNA levels were 5.3 log₁₀ copies and 5.5 log₁₀copies among ART naïve in the LPV/r and NNRTI arms, respectively.

Of the 88% (163) of children with HIV RNA measurements available at week 48 of treatment,  80% (67/84) in the LPV/r arm and 76% (60/79)  in the NNRTI arm had viral load below 400 copies/ml, a non-significant difference.

Proportions of those virologically suppressed were comparable in both arms at 24 and 48 weeks, with slightly higher numbers among those ART-suppressed at randomisation.

Among the ART-naïve children, the mean increases in CD4 counts at 48 weeks were 394 (368) and 405 (324) (p=0.87) in the LPV/r and NNRTI arms, respectively; similarly CD4 percentage changes were 12 (8) and 13 (7) (p=0.27), respectively.

Among the ART-suppressed, surprisingly, there was a mean decrease in CD4 counts by 300 among those in the LPV/r  arm but minimal (5.8) decrease in the NNRTI arm, yet CD4 percentages remained stable in both arms, and as a consequence this difference in absolute CD4 count change was found to be statistically non-significant (p=0.24)..

Grade 3 and 4 adverse events were experienced at least once by 32% and 27% of the LPV/r and NNRTI arms, respectively, p=0.3 with neutropenia the most frequent in both arms. 

Four deaths occurred, although none were attributed to the study medication.

The researchers concluded that in this cohort of ART naïve and ART-suppressed HIV-infected Ugandan children, use of a LPV/r based regimen was non-inferior to that of an NNRTI based regimen, showing comparable immunologic responses and AE rates.

Professor Charles Gilks of the University of Queensland, an architect of WHO’s public health approach to antiretroviral therapy in resource-limited settings, noted in the subsequent question and answer session that the proposed second-line regimen in children exposed to first-line lopinavir/ritonavir, that of nevirapine plus two NRTIs, would be “very, very brittle if you already have high levels of NRTI resistance”. He added that “the real drawback of using protease inhibitor-based initial therapy wide-scale at the moment is that it is very difficult to support it with a second-line [regimen].”