       Document 2859
 DOCN  M94A2859
 TI    Effect of rate of CD4 decline on the efficacy of early ZDV therapy in
       asymptomatic HIV infection.
 DT    9412
 AU    Easterbrook PJ; Yu LM; Babiker A; Gazzard BG; Boag F; McLean K; Chelsea
       and Westminster Hospital, London, U.K.
 SO    Int Conf AIDS. 1994 Aug 7-12;10(1):209 (abstract no. PB0265). Unique
       Identifier : AIDSLINE ICA10/94369716
 AB    OBJECTIVE: To determine whether asymptomatic HIV infected patients with
       rapid CD4 decline are a subgroup who might benefit most from early ZDV
       therapy. METHODS: Analysis of data on 197 patients from 3 centres
       participating in the Concorde trial: a randomized double-blind
       comparative trial of ZDV (1g/day) vs. placebo (P) in asymptomatic HIV
       infected individuals. Rate of CD4 decline was estimated using linear
       regression analysis in 156 patients (P = 81; ZDV = 75) who had > or = 3
       CD4 measurements before and after randomization. Median CD4 cell count
       at entry was 437/mm3 (P) and 462/mm3 (ZDV), respectively. Patients were
       stratified into quartiles (groups 1-4) according to rate of CD4 decline
       prior to randomization, where gp. 1 had the fastest (median = -180
       cells/6 mth) and gp. 4 the slowest rate (median = +10 cells/6 mth) of
       CD4 decline. Outcome measures were (i) magnitude of change in rate of
       CD4 decline after therapy and (ii) clinical endpoints: CDC IV events,
       AIDS or death. Multivariate models were used to examine the interaction
       between ZDV and prior rate of CD4 decline on outcomes (i) and (ii).
       RESULTS: The greatest reduction in rate of CD4 decline with early ZDV
       vs. placebo was in gp. 1 patients (difference (ZDV-P) in change in rate
       of CD4 decline = 93 cells, p = 0.09) compared to gp2 (10 cell diff, p =
       0.26); gp 3 (4 cell diff, p = 0.79) and gp4 (45 cell diff, p = 0.64).
       However, when the analysis was restricted to those patients with > or =
       5 CD4 counts before and after randomization, this trend was no longer
       apparent. The impact of early ZDV on clinical events (21 in ZDV; 30 in
       P) did not differ according to rate of CD4 decline in either analysis.
       Similarly, in multivariate analyses, rate of CD4 decline had no
       influence on the efficacy of ZDV for outcomes (i) or (ii). CONCLUSION:
       These preliminary findings suggest that rate of CD4 cell decline may not
       be important determinant of the response to ZDV therapy. If ongoing
       analyses in a larger cohort of patients confirm these findings, then
       asymptomatic patients with more rapid CD4 decline do not appear to be a
       subgroup likely to benefit from early ZDV therapy. We also propose to
       examine the role of SI/NSI phenotype on ZDV response in these patients.
 DE    Comparative Study  Double-Blind Method  HIV Infections/*DRUG THERAPY
       Leukocyte Count  Treatment Outcome  *T4 Lymphocytes
       Zidovudine/*THERAPEUTIC USE  CLINICAL TRIAL  MEETING ABSTRACT
       RANDOMIZED CONTROLLED TRIAL

       SOURCE: National Library of Medicine.  NOTICE: This material may be
       protected by Copyright Law (Title 17, U.S.Code).

