519   CD4⁺Lymphocyte Level Is Better than HIV-1 Plasma Viral Load in Determining When To Initiate HAART.

T. R. Sterling*, R. E. Chaisson, J. G. Bartlett, and R. D. Moore.
Johns Hopkins Univ. Sch. of Med., Baltimore, MD.

Background:The optimal timing for initiation of highly active antiretroviral therapy (HAART) is unclear. The validity of current treatment guideline cutoffs (viral load >20,000 copies/ml or CD4⁺lymphocyte level <500/ mm³) has not been assessed in men and women.

Methods:The impact of HAART (>90 days of a regimen including an HIV-1 protease inhibitor, nonnucleoside reverse transcriptase inhibitor, or triple nucleosides with abacavir) on new opportunistic infection (OI) or death (combined endpoint) was assessed according to CD4⁺level and viral load preceding initiation of therapy. Patients were enrolled in care after 7/1/96 and within 90 days of first viral load. Study period was through 6/30/ 00. Viral load was determined by RT-PCR.

Results:There were 1162 patients: 71% male, 78% black, 47% injection drug users, median age 38 years. 675 patients received HAART, with median time on therapy 533 days (range: 91—1460), and 76% of patients achieved viral load <400 copies/ml. In a multivariate proportional hazards model adjusting for CD4⁺level and prior OI, viral load <20,000 copies/ml vs. >20,000 copies/ml prior to starting HAART was not associated with disease progression (Chi-square (C²) = 2; hazard ratio (HR) = 0.8; 95% CI: 0.6, 1.1; p = 0.2). This also held when viral load cutoffs of 100,000 or 500,00 copies/ml were used. In contrast, CD4⁺level <200/mm³prior to starting HAART was strongly associated with disease progression (C²= 32; HR = 4.3; 95% CI: 2.6, 7.2; p < 0.0001), but CD4⁺level 201—350 was not (C²=2.5; HR = 1.6; 95% CI: 0.9, 2.9; p = 0.11) compared to CD4⁺level 350—500/mm³. Compared to patients who did not receive HAART (dual nucleosides only, or no therapy), HAART was associated with a decreased risk of disease progression among patients with CD4⁺level <200 cells/ mm³(Kaplan-Meier [K-M] log rank p = 0.01) but not >200 cells/mm³(K-M log rank p = 0.33). There were no differences in disease progression between men and women within any CD4 or viral load stratum.

Conclusions:Based on levels prior to initiating HAART, CD4⁺lymphocyte level is a better predictor of disease progression than viral load. Current treatment guidelines should place greater emphasis on CD4⁺level than viral load for both men and women.