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


Zalcitabine was as effective as didanosine for zidovudine nonresponders

ACP J Club. 1994 July-Aug;121:3. doi:10.7326/ACPJC-1994-121-1-003

Source Citation

Abrams DI, Goldman AI, Launer C, et al. A comparative trial of didanosine or zalcitabine after treatment with zidovudine in patients with human immunodeficiency virus infection. N Engl J Med. 1994 Mar 10;330:657-62.



To evaluate the effectiveness and safety of didanosine and zalcitabine in patients with human immunodeficiency virus (HIV) infection in whom zidovudine treatment had failed.


Randomized controlled trial with a median follow-up of 16 months.


133 clinical sites in 16 community clinical trial units.


467 patients (mean age, 38 y; 90% men) who had an acquired immunodeficiency syndrome (AIDS)-defining condition or 2 CD4+ lymphocyte cell counts ≤ 300/mL and had treatment with zidovudine that led to drug intolerance or progression of the disease (the development of an opportunistic infection or malignancy, AIDS dementia complex, or failure to thrive). Exclusion criteria were previous treatment with or contraindications to the study drugs; a history of pancreatitis or peripheral neuropathy; an uncontrolled seizure disorder; AIDS dementia complex stage 2 or worse; or current treatment for an active AIDS-defining opportunistic infection or neoplasm.


230 patients were allocated to receive didanosine buffered sachets, 500 mg/d, and 237 patients were allocated to receive zalcitabine tablets, 2.25 mg/d.

Main Outcome Measures

Death, disease progression, CD4+ cell count changes, and adverse effects.

Main Results

157 patients (68%) taking didanosine died or had disease progression compared with 152 patients (64%) taking zalcitabine {95% CI for the 4% difference, -4% to 13%}*. 100 patients (43%) taking didanosine died compared with 88 patients (37%) taking zalcitabine {CI for the 6% difference -2% to 15%}*. CD4+ cell counts declined by about 3 cells/mo after the first 2 months in both treatment groups. Treatment was discontinued because of toxicity in 94 patients (41%) and 88 patients (37%) assigned to didanosine and zalcitabine, respectively. Diarrhea and abdominal pain occurred more often with didanosine and peripheral neuropathy and stomatitis occurred more frequently with zalcitabine.


Zalcitabine was as effective as didanosine in delaying disease progression and death in patients with human immunodeficiency virus infection who had not responded or were intolerant to treatment with zidovudine.

Source of funding: National Institute of Allergy and Infectious Diseases.

For article reprint: Dr. L. Deyton, National Institute of Allergy and Infectious Diseases, National Institutes of Health, 6003 Executive Boulevard, Bethesda, MD 20892. FAX 301-496-0701.

* Numbers calculated from data in article.


Didanosine benefited HIV-infected patients failing on zidovudine

Previous studies have shown that patients who are clinically stable on zidovudine for at least 16 weeks have fewer AIDS-defining events or deaths if they are switched to didanosine (1). Spruance and colleagues studied patients taking zidovudine for at least 6 months with clinical deterioration in the final 12 weeks. Switching to didanosine decreased disease progression, with the greatest benefit seen in those with less advanced disease (CD4+ ≥ 100/mL) and a longer duration of zidovudine therapy (> 12 mo). This is consistent with the understanding that antiretroviral agents work less well in patients who are severely immunodeficient and that zidovudine-resistant HIV strains are detected within months of initiating zidovudine.

Zalcitabine has been shown to be beneficial in combination with zidovudine for patients with CD4+ counts < 300/mL (2). The study by Abrams and colleagues also lends support to zalcitabine use as monotherapy. In patients who are intolerant to or show disease progession while on zidovudine, zalcitabine was at least as effective as didanosine in preventing disease progression or death. The possibly lower death rate in the zalcitabine groups remains to be confirmed.

These studies assist us in many ways. First, they make us more comfortable switching from zidovudine to other antiretroviral agents in patients with adverse reactions or clinical progression. Second, the toxicity profiles shown by Abrams and colleagues help us choose between didanosine and zalcitabine as a second-line agent. Patients with preexisting neuropathies or poor oral hygiene may do better on didanosine. Zalcitabine may be a better choice in persons who have preexisting diarrhea. Previous episodes of pancreatitis, heavy alcohol ingestion, or concomitant intravenous pentamidine should also favor zalcitabine use because of the absence of cases of pancreatitis in the group receiving this drug.

Lastly, it is important to temper our optimism by noting that although the patients in both studies benefited from didanosine or zalcitabine, the rates of disease progression or death were 42% at a median of 47 weeks and 66% at a median of 64 weeks.

The finite benefits to reverse transcriptase monotherapy are becoming clearer. Although more studies will clarify the roles of these agents, giant steps in the control of HIV disease progression will probably be made through the use of agents that interfere at other stages in the HIV replicative cycle, such as protease inhibitors. In addition, the combined or alternating use of 2 or 3 reverse transcriptase inhibitors may prove more beneficial than monotherapy alone. Meanwhile, a cautious and organized use of reverse transcriptase inhibitors remains our only Food and Drug Administration-approved approach to the patient infected with HIV. Recently published guidelines that take into account the patient's clinical status and CD4 count are consistent with the 2 studies presented here (3). For now, the reverse transcriptase inhibitors provide a temporizing effect on disease progression and future research will define how best to improve our approach.

Fred A. Zar, MD
St. Francis Hospital Evanston, Illinois, USA