When to Start ART: Two Major Cohort Studies Published

[Guest guest]

One study suggests that treatment should be started before the CD4 count falls

below 350 cells/mm3; the other suggests before 500 cells/mm3.*

Because no prospective clinical trial has directly addressed the issue of

immediate versus deferred antiretroviral therapy (ART) for asymptomatic

HIV-infected patients, clinical guideline panels have relied mainly on cohort

study data to inform their recommendations.

This month, findings were published from two major collaborative cohort studies,

both of which were presented at scientific meetings in the past year (JW AIDS

Clin Care Nov 10

2008 <http://aids-clinical-care.jwatch.org/cgi/content/full/2008/1110/1>

and Mar 9 2009

<http://aids-clinical-care.jwatch.org/cgi/content/full/2009/309/3>

In the NA-ACCORD study, investigators conducted two parallel analyses using data

pooled from 22 cohorts in North America. Patients were eligible for the first

analysis if, at any point during the study period (1996–2005), they had a CD4

count of 351 to 500 cells/mm3 and no history of AIDS-defining illness or ART

use. Of the 8362 patients who met these criteria, 2084 started ART immediately

(within 6 months after entering the CD4-cell range of interest), and 6278

deferred.

Deferral of therapy was associated with a 69% increased risk for death during

follow-up. The second analysis was identical to the first, except that patients

had CD4 counts >500 cells/mm3.

Again, only about 25% of patients (2220 of 9155) started ART immediately, and

deferral of therapy was associated with a significant increase in mortality risk

(94%).

In both analyses, patients who started ART earlier generally had more-favorable

baseline characteristics (e.g., lower rates of injection-drug use [iDU] and

hepatitis C virus coinfection), and they achieved higher rates of virologic

suppression with treatment, suggesting a

better rate of medication adherence.

However, even after adjustment for some of these factors, the survival advantage

of earlier therapy remained.

The merits of early therapy were also assessed by the When to Start

Consortium, which evaluated pooled data from 15 cohorts in Europe and North

America. Patients were eligible for analysis if they started ART in 1998 or

later with a CD4 count <550 cells/mm3, no history of AIDS-defining illness, and

a presumed acquisition mode other than IDU.

In order to estimate the lead time before starting therapy and the clinical

events that took place during this time, the investigators applied cohort data

from the pre-ART era (1989–1995) to the current era. More than 24,000 patients

were included in the analysis.

Deferral of ART until the CD4 count fell to between 251 and

350 cells/mm3 (compared with 351–450 cells/mm3) was associated with a 28%

increased risk for AIDS or death; no significant increase was seen for death

alone.

Comment

These two studies represent a fascinating example of how advanced

statistical techniques can be applied to observational data to try and

answer a clinical question — in this case, when to start ART in asymptomatic

patients.

Although the observational nature of these studies prevents either

research group from claiming to " prove " that therapy should be started at higher

CD4-cell counts — a point that both editorials hasten to mention — the data do

add to a substantial body of evidence showing that untreated HIV infection has

negative consequences, even at higher CD4-cell counts.

(Neither study can claim that those who start ART earlier are the same as those

who defer — they may have been more " health seeking " and hence did better for

reasons unrelated to ART.) The NA-ACCORD data are particularly striking and are

even stronger than those presented at the 2009 Retrovirus Conference.

Although the two research groups reach different conclusions

about the precise threshold for when to start therapy, they are consistent in

the most important message — that treatment should be started before the CD4

count falls below 350 cells/mm3, and perhaps earlier. *— E. Sax, MD*

*http://www.medscape.com/viewarticle/703308?src=mp & spon=1 & uac=67363AY*

_____________________

Dr RK Sood

drrksood@...

+91 9418064077, +91 9445157327