HIV treatment in primary infection: 48 week course modestly delays CD4 drop

A 48-week course of antiretroviral treatment started within six months of becoming infected modestly delays the need for lifelong treatment, reported Dr Sarah Fidler of Imperial College, London, at the sixth International AIDS Society conference (IAS 2011) in Rome.

Patients diagnosed with HIV infection fewer than 12 weeks after exposure showed greater benefit from the 48-week course of treatment.

During the early days of highly active antiretroviral therapy (HAART), Dr David Ho argued that treatment in primary infection might offer an opportunity for eradicating HIV from the body, or at least drastically limiting the damage caused to the immune system by HIV.

Subsequent clinical studies showed that early treatment did not eradicate HIV, and that when treatment was withdrawn, HIV levels swiftly rebounded to very high levels.

However, there has been continued interest in determining whether a period of treatment soon after infection might limit damage to the immune system and so delay the need for later lifelong treatment.

The SPARTAC study, led by Dr Sarah Fidler of Imperial College, London, compared three strategies for dealing with primary infection:

• Begin treatment within six months of infection and stay on treatment for 48 weeks (ART-48 arm).
• Begin treatment within six months of infection and stay on treatment for 12 weeks (ART-12 arm).
• Do nothing except monitor the CD4 count and health of the patient (standard of care, SOC arm).

The study sought to determine whether any of the strategies reduced the risk of a CD4 decline to the level at which a person would normally need to start treatment (< 350 cells/mm³). It also sought to examine the effectiveness and tolerability of early treatment, the risk of drug resistance, and the rate of CD4 cell decline in each group.

SPARTAC recruited 371 patients in the United Kingdom (40%), Australia, Brazil, Ireland, Italy, South Africa (33%), Spain and Uganda.

All participants had verified infection with HIV-1 less than six months before randomisation, defined as less than three Western Blot bands and/or antibody-negative, PCR-positive infection. Five patients were subsequently excluded from randomisation due either to incorrect diagnosis or to randomisation error.

Sixty per cent of participants were male, with a median age of 32; 56% were gay or bisexual men.  However, all South African participants were women, so in effect, 96% of participants outside South Africa were male and 90% were gay and bisexual men.

Participants had been infected for a median of 12 weeks at the time of randomisation (range 9 to 15 weeks), and had a median CD4 count of 556 cells/mm³ and viral load of 4.53 log ₁₀ copies (around 30,000 copies/ml).

Participants were randomised equally to the three study arms and followed for a median of 4.2 years. Ninety-one per cent of participants received a regimen of AZT/3TC (Combivir) and lopinavir/ritonavir (Kaletra). Approximately 6% of participants already had at least one drug resistance mutation at baseline, suggesting that they had been infected with drug-resistant virus. Nucleoside analogue and non-nucleoside reverse transcriptase inhibitor mutations were equally common (13 and 11 patients respectively).

After 4.5 years of planned follow-up, participants in the 48-week treatment arm had a significantly reduced risk of reaching a CD4 count below 350 or needing to start long-term treatment (hazard ration 0.63, 95% confidence interval 0.45 – 0.90, p=0.01) when compared to the standard-of-care arm.

Participants in the 12-week treatment arm did not show a significant reduction in their risk of CD4 decline or in their need to start long-term treatment.

However, when the time taken to reach the primary endpoint was compared, the average delay in the 48-week treatment arm when compared to the standard of care was 65 weeks.

In other words, 48 weeks of treatment soon after infection only delayed disease progression to the extent that a person treated soon after infection would need four months less treatment during their life than someone who didn’t take treatment immediately after infection.

But a secondary post hoc analysis showed that among people who started treatment in the 48-week arm less than 12 weeks after becoming infected, there was a somewhat greater reduction in the risk of disease progression (hazard ratio 0.48, 95% CI 0.30 – 0.78, p=0.003). Calculation of the subsequent  delay in starting treatment among this group of patients was not presented.

Among those who started treatment more than 12 weeks after infection, there appeared to be no significant difference in the risk of disease progression between the standard-of-care group and the 48-week treatment group.

This finding suggests that identifying very recent infection will be critical if any benefit of an early course of treatment is to be realised.

Short courses of treatment may carry risks however. While early treatment could limit the size of the viral reservoir, removal of treatment would allow the viral reservoir to swiftly expand. Stopping treatment might also cause inflammatory changes that would increase the risk of other serious, non-AIDS defining illnesses.

However, patients in the 48-week treatment arm of the SPARTAC study continued to have lower levels of viral load than untreated patients in the control arm for at least 60 weeks after stopping treatment, and their average CD4 cell count remained 135 cells/mm³ higher than the control arm after 4.5 years follow-up. But the rate of decline of CD4 count did not differ between the treatment arms and the standard-of-care arm.

Levels of the inflammatory markers IL-6 and d-dimer did not go up after treatment interruption, unlike in the SMART study of treatment interruption.

There was no difference in the rate of serious adverse events between patients in the treatment arms and patients in the standard of care arm.