Background:  The percentage of patients who initiated HAART during primary HIV-1 infection has decreased in France from 95% in 1996 to < 50% in 2003. The objectives of this study were to describe the prognostic factors for clinical and immunological progression in these untreated patients at time of primary infection.

Methods:  Patients are included in the prospective multicenter PRIMO Cohort if they are diagnosed with an acute or recent HIV-1 infection. No specific therapeutic guidelines are given. Patients who did not initiate HAART at enrolment were included in this analysis. Progression was defined as a clinical AIDS event, or a decrease in CD4 cell count < 350/mm³, or a decrease in CD4 < 500/mm³ associated with a fall in the CD4 % ≥ 20% in a ≥ 6-month period. Data were censored at the time of treatment in patients who initiated it during follow-up. Risk factors for progression were studied by using a Cox proportional hazard model.

Results:  We followed 75 untreated patients for a median time of 18.5 months:  70% had presented with a symptomatic posts infection. Characteristics at baseline were (median value): HIV RNA 4.4 log copies/mL (range 1 to 7; 2); HIV DNA in PBMC 2.9 log copies/10⁶ cells (1.8 to 4.2); CD4 cell count 648/mm³ (207 to 1508). Median time between infection and enrolment was 55 days. In 27 patients (36%), clinical or immunological progression developed during follow-up:  1 patient had a recurrent bacterial pneumonia, the others had a sustained low CD4 count or a pronounced CD4 decrease. The rate of occurrence of an end-point was 25%, 44%, and 49% at 12, 24, and 36 months, respectively. A low CD4 cell count, a low CD4 % and a high HIV DNA at baseline were associated with disease progression in univariate analysis, but a high HIV DNA level remained the only significant prognostic factor in the multivariate analysis with an OR of 3.13 [1.2 to 8.0] (p = 0.02). Other baseline variables (age, gender, year of infection, number of signs of primary infection or length of the symptoms, hemoglobin, plasma HIV RNA) had no prognostic value. HAART was initiated during follow-up in 20 patients, mainly based on CD4 changes.

Conclusions:  There are many concerns about whether to treat at time of primary HIV infection. This study shows that untreated patients must continue to benefit from a close clinical and laboratory monitoring, especially patients with a high DNA level early at the time of primary infection, since immunodeficiency may worsen.

Keywords: primary infection; prognostic factors; HIV DNA