Current issues of ACP Journal Club are published in Annals of Internal Medicine


Immediate zidovudine was no more effective than deferred zidovudine in symptom-free HIV infections

ACP J Club. 1994 Nov-Dec;121:73. doi:10.7326/ACPJC-1994-121-3-073

Source Citation

Concorde Coordinating Committee. Concorde: MRC/ANRS randomised double-blind controlled trial of immediate and deferred zidovudine in symptom-free HIV infection. Lancet. 1994 Apr 9;343:871-81.



To compare immediate and deferred zidovudine for mortality, progression to acquired immunodeficiency syndrome (AIDS) or AIDS-related complex (ARC), and safety in patients who had symptom-free human immunodeficiency virus (HIV) infection.


Randomized, double-blind, placebo-controlled trial with a median follow-up of 3.3 years.


74 centers in the United Kingdom, Ireland, and France.


1761 patients > 13 years of age (mean age 32 y, 85% men) with symptom-free HIV infection. Exclusion criteria were pregnancy or breastfeeding; use of immunomodulating or antiviral drugs except acyclovir; weight < 40 kg; Karnofsky performance scores < 90; or abnormal creatinine, hepatic function, hemoglobin, platelet counts, or neutrophil counts. 42% and 41% of participants in the immediate and deferred groups, respectively, had CD4+ counts >500/µL at enrollment. Follow-up was 99%.


877 patients were assigned to zidovudine, 250 mg 4 times/d, and 872 patients were assigned to placebo until they developed ARC or AIDS, when they were started on open-label zidovudine. After 1989, zidovudine could be started when CD4+ counts decreased to 500/µL if judged to be clinically indicated.

Main outcome measures

Mortality, progression to ARC or AIDS, and adverse effects.

Main results

During the study, 418 patients (48%) in the deferred group were given zidovudine because ARC or AIDS developed (174 patients), CD4+ counts decreased (239 patients), or other reasons (5 patients). The groups did not differ for mortality or progression to ARC or AIDS (Table). The 3-year survival was 92% for the immediate group and 94% for the deferred group. The immediate group spent 81% of trial time before progression to ARC or AIDS on zidovudine compared with 16% for the deferred group. At 3 months and for the rest of the study, the immediate group had higher CD4+ cell counts. The rate of adverse effects was lower in the deferred group.


Among initially symptom-free patients with HIV infection, no differences in mortality or morbidity were found between those who started zidovudine immediately and those in whom zidovudine was deferred until development of ARC or AIDS.

Sources of funding: Medical Research Council (UK and Ireland); Agence Nationale de Recherches sur le SIDA; Institut National de la Santé et de la Recherche Médicale; Wellcome Foundation.

For article reprint: Dr. J.H. Darbyshire, MRC HIV Clinical Trials Centre, Department of Clinical Epidemiology, National Heart and Lung Institute, Royal Brompton National Heart and Lung Hospital, Sydney Street, London SW3 6NP, England, United Kingdom. FAX 44-351-8946.

Table. Immediate vs deferred zidovudine treatment for symptom-free HIV infection*

Outcomes at a median follow-up of 3.3 y Immediate zidovudine treatment Deferred zidovudine treatment RRI (95% CI) NNH
Death 11% 9% 26% (-6 to 67) Not significant
Progression to AIDS, ARC, or death 30% 33% 7% (-7 to 19) Not significant

*ARC = AIDS-related complex. Other abbreviations defined in Glossary; RRI, RRR, NNH, NNT, and CI calculated from data in article.


Nothing is more illustrative of the rapid pace of recent developments in therapeutics for HIV infection than the fact that the critically important and highly cited 1994 publication of the Concorde trial now represents little more than a historical artifact. The trial compared 2 strategies of initiating zidovudine monotherapy and purported to show that deferring initiation until CD4+ counts fell or clinical symptoms developed was equally effective as beginning therapy immediately. Interpretation of the original trial was hampered by the fact that the deferral strategy was subjectiveand was neither well described nor easily operationalized.

The loss of relevance stems directly from the fact that monotherapy and dual therapy are both dead issues with respect to currently licensed drugs. The current standard of care involves highly active anti-retroviral therapy (HAART) with ≥ 3 agents whose goal is to maintain plasma viral load levels below the limits of quantitation of the most sensitive available assays for as long as possible. This has been shown to be of clinical benefit in randomized clinical trials as well as in clinic and population-based studies. Unfortunately, virological rebound occurs frequently in clinical practice. Partially suppressive therapy, regardless of the reason (e.g., incomplete adherence, suboptimal prescription, pharmacokinetic issues), promotes the development of viral resistance. This has become a major therapeutic challenge because of the limited number of treatments available and the substantial cross-resistance between the members of each drug class. However, ≥ 2 prospective studies have now shown the efficacy of using resistance testing to tailor therapy (1, 2). To what extent this will translate into improved long-term outcomes in clinical practice remains to be established.

The optimal time for the initiation of therapy remains controversial. A more recent paper by the Concorde authors involving long-term follow-up of this and other zidovudine monotherapy trials (3) sought again to suggest no benefit for early initiation. However, it is not clear that trials involving initiation of monotherapy can inform the debate on when to initiate HAART. Although the rationale for early therapy is powerful, so are the difficulties associated with this strategy, such as adherence, resistance, adverse effects, quality of life, and cost. Uncertainties about the optimal time for treatment initiation are likely to remain until either the treatment strategies are substantially improved in terms of safety, tolerability, effectiveness, and cost; or conclusive evidence on this issue is derived from long-term clinical trials involving HAART. Alas, the latter remains an unlikely prospect when the therapeutic field continues to move at such a rapid pace.

Julio S.G. Montaner, MD
Martin T. Schechter, MD, PhDUniversity of British ColumbiaVancouver, British Columbia, Canada


1. Durant J, Clevenbergh P, Halfon P, et al. Can HIV genotype determination be useful for individualized adaptation of antiretroviral therapy: The VIRADAPT French Study [Abstract]. Abstracts of the Fourth International Congress on Drug Therapy in HIV Infection, Glasgow, Scotland. AIDS. 1998;12(Suppl 4):S16.

2. Baxter JD, Mayers DL, Wentworth DN, et al. A pilot study of the short-term effects of antiretroviral management based on plasma genotypic antiretroviral resistance testing (GART) in patients failing antiretroviral treatment. In: Program and abstracts of the 6th Conference on Retroviruses and Opportunistic Infections, January 31 - February 9, 1999, Chicago IL. Abstract 8.

3. Joint Concorde and Opal Coordinating Committee. Long-term follow-up of randomized trials of immediate versus deferred zidovudine in symptom-free HIV infection. AIDS. 1998;12:1259-65.