NNRTIS and PIs equal in children, trials shows, but PIs may delay need for switch to second-line

Children randomly assigned to receive either a protease inhibitor (PI) or a non-nucleoside reverse transcriptase inhibitor (NNRTI) based regimen then switched to second-line at a higher or lower viral load achieved comparable good long-term clinical, immunological and virological outcomes according to findings from the PENPACT-1 Study Team reported in the February 1 online edition of The Lancet Infectious Diseases.

 The authors add that in resource-poor settings where treatment options are limited, delaying the switch from a PI-based regimen is acceptable because the risk that NRTI and PI resistance will develop appeared to be low in this study.

The number of antiretroviral drugs for children has increased since the early 2000s. However children will potentially be on ART for decades. So the long-term effectiveness of first-line treatment options and the timing of a switch to second-line treatment are critical.

The Paediatric European Network for Treatment of AIDS (PENTA) and the Pediatric AIDS Clinical Trials Group (PACTG/IMPAACT) in the United States collaborated in a randomised open-label trial to assess the long-term outcomes of PI and NNRTI first-line antiretroviral therapy and viral load switching criteria in children.

The authors compared the effectiveness of two NRTIs plus a PI with two NRTIs plus an NNRTI and of switching to second-line treatment at a viral load of 1000 copies/mL (PI and NNRTI low) with 30,000 copies/mL (PI and NNRTI higher) in treatment-naïve HIV-infected children from Europe, North and South America. The primary outcome was a change in viral load between baseline and four years.

The authors believe this is the first trial to undertake a direct comparison of the long-term outcomes of these treatment strategies in children.

266 children (133 from Europe, 77 from North America and 56 from South America) from 68 centres in 13 countries from September 25, 2002 to September 7, 2005 were randomly assigned PI or NNRTI-based treatment regimens. 263 were included in the analysis. Median follow-up was 5 years (IQR 4.2-6.0). 71% (188) of children were on first-line at the end of the trial.

In the PI group half began with lopinavir plus ritonavir and the other half on nelfinavir. (Only 4 children remained on nelfinavir at the trial’s end; 36 substituted lopinavir plus ritonavir of which 32 did so in 2007 at the time of nelfinavir recall).

Among the NNRTI group 80 (60%) started on efavirenz and the remainder on nevirapine.

The mean reduction in viral load for PIs compared to NNRTIs was -3.16 log10 copies/mL and -3.31 log10 copies/mL, respectively with a difference of -0.15 log10 copies/mL  (95% CI: -0.41-0.11, p=0.026); and -3.26 log10 copies/mL for the low viral threshold compared to -3.20log copies/mL for the higher threshold (a difference of 0.06 copies/mL 95% CI: -0.20-0.32, p=0.56).

The proportion of children with a viral load lower than 400 copies/mL at 4 years was similar for all initial PI and NNRTI regimens (80% lopinavir plus ritonavir; 84% other PIs, predominantly nelfinavir; 80% efavirenz and 84% nevirapine) .

No difference in viral suppression was found between those who switched at a low threshold compared to those who switched at a higher threshold, the authors add. Additionally no significant differences in the accumulation of major PI or NNRTI mutations between the groups were noted.

NNRTI resistance was common. Approximately 10% more children accumulated NRTI mutations in the NNRTI higher group compared to the NNRTI low group. PI resistance was rare, and there was no increase in NRTI resistance in the PI low group compared to the PI higher group.

The authors note regular viral load monitoring will not prevent development of NNRTI resistance, whereas continuing on a failing NNRTI treatment regimen will likely increase the accumulation of NRTI mutations so limiting future use.

The researchers ask whether infants started on lopinavir/ritonavir should be delayed from switching at virological failure rather than switching to an NNRTI plus two NRTI second-line regimen with a risk for the rapid development of resistance?

The authors conclude these results are of significance in resource-poor settings where treatment options are limited and viral load monitoring rarely available. Good long-term outcomes were achieved with both PI or NNRTI based treatment regimens. Delayed switching until viral load was 30000 copies/mL resulted in greater NRTI resistance mutations in NNRTI-based regimens than switching at 1000 copies/mL. Delayed switching for children on a PI-based regimen in these settings may be a viable option “because the risk of selecting for NRTI and protease-inhibitor resistance is low.”