HAART clinically and immunologically effective in babies but high rate of virologic failure

A HAART regimen consisting of d4T (stavudine, Zerit), ddI (didanosine, Videx / VidexEC) and nelfinavir (Viracept) is clinically and immunologically effective when initiated in babies under three months, but is associated with a high rate of virologic failure, according to a small study published in the January 23^rd edition of AIDS.

There have been no randomised studies to determine when is the best time to start HAART in asymptomatic infants who have their HIV infection diagnosed in the early months of life. There is a rationale for early treatment, however, as approximately a quarter of all babies vertically infected with HIV progress to AIDS or death during the first year of life, and very high HIV viral loads have been recorded in very young infants. US treatment guidelines currently recommend the immediate use of HAART in babies born with HIV.

The limited data to date suggest that starting anti-HIV therapy soon after birth can dramatically improve the immune function of babies, but is associated with a high rate of virologic failure. Furthermore, there are concerns about the lack of age-specific pharmacokinetic data, food restrictions, inadequate paediatric formulas and the poor ability of babies to metabloise drugs.

To assess the impact of HAART early in life, investigators from five European countries conducted a prospective non-randomised trial into the clinical, immunological and viral efficacy of a HAART regimen of d4T, ddI and nelfinavir in vertically infected infants aged under three months. Data were also gathered on adverse effects.

A total of 20 infants were recruited to the phase I/II PENTA 7 study between summer 1999 and December 2000. The infants were monitored every four weeks for a median of 96 weeks. At baseline, median HIV viral load was 320,000 copies/mL, and CD4 cell percentage was 33%. HAART was initiated at a median age of two and a half months.

One infant died at week 60 (of non HIV-related causes), and at week 72, the primary endpoint, eleven infants (55%) were still taking their original study medication. Four infants stopped therapy because of virological failure, lack of virological response or parent request, and did not recommence therapy. Four other infants switched treatment immediately or subsequently to alternative three- or four-drug regimens due to lack of virological response.

A total of four serious adverse events were recorded, two requiring hospitalisation, but none of the these events was considered related to the study medication. A total of 48 minor adverse events were recorded in 15 infants, and eleven were thought to be related to at least one of the study drugs. However, in only three instances did side-effects lead to the modification of study medication. In one case rash led to ddI and d4T being stopped for a day and nelfinavir for ten days; diarrhoea led to ddI being permanently stopped and being replaced with 3TC; and neutropenia led to a permanent d4T dose-reduction.

Seven infants had to change from the powder formula of nelfinavir to crushed tablets because of difficulties administering the powder.

All but one baby experienced at least one instance of grade one triglyceridaemia, and nine of the twelve infants for whom baseline cholesterol values were available experienced at least one episode of hypercholesterolaemia. No grade two events were recorded and none of the babies developed symptoms of lipodystrophy.

Therapy proved to have clinical efficacy, as none of the infants progressed to AIDS and the only death recorded was not related to HIV. The investigators also noted significant height and weight gains from baseline to week 72. There were, however, two cases of HIV disease progression involving a persistent fungal infection and recurrent diarrhoea.

Although CD4 cell percentages did not increase significantly from baseline, all but two infants had a CD4 cell percentage above 25% at week 72, including two infants who had a baseline percentage below 15%, indicative of a severe risk of HIV disease progression.

Although median viral loads fell by a median of 2.06 log₁₀ in the first four weeks of therapy, only 37% of infants had a viral load below 400 copies/mL by week 48, at which point only 21% had a viral load below 50 copies/mL. A total of fourteen infants had virological failure by week 72, of whom six (30%) never achieved a viral load below 400 copies/mL. In four infants viral load fell to below 400 copies/mL and then rebounded, and a further four only achieved a viral load below 400 copies/mL after week 72. Only five babies had no viral load rebound and remained below 400 copies/mL.

Of the 18 infants with wildtype HIV at baseline, five developed a resistance-conferring mutation during the course of the study.

Commenting on these data, the investigators note that “in PENTA 7, we have shown that the early initiation of HAART…is feasible, well tolerated and results in improved immunological status and stable clinical status over a period of 72 weeks. However, the presence of incomplete viral suppression in 70% of the infants, associated with genotypic resistance…to antiretroviral drugs in six (30%), is much higher than observed in adults following treatment for primary infection.”

The investigators suggest that the lack of virological response seen in their study could have been due to the poor potency of the drugs used in the study, difficulty administering the drugs, poor adherence, or poor drug absorption. Indeed, they note that more than two-thirds of measured nelfinavir trough levels were below the recommended therapeutic levels, even when the daily dose of the drug was increased to 150mg/kg twice daily.

An editorial accompanying this study notes the progress that has been made in treating HIV infection in very young infants, and highlights the safety and immunological efficacy indicated in the PENTA 7 study. However, it notes with concern the high rate of virological failure. How could this be overcome? The editorial suggests that results might be improved by the use of a more potent protease inhibitor than nelfinavir, noting that 79% of infants treated with Kaletra achieved a viral load below 400 copies/mL in a recent study. The editorial also highlights the need for better paediatric formulations of drugs, and emphasises that infants and children need to be regarded as pharmacokinetically distinct from adults.