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


Combination nucleoside therapy was no better than zidovudine alone for advanced HIV infection

ACP J Club. 1997 May-Jun;126:65. doi:10.7326/ACPJC-1997-126-3-065

Related Content in this Issue
• Companion Abstract and Commentary: Combination nucleoside therapy or ddI alone delayed the progression of HIV infection compared with zidovudine alone

Source Citation

Saravolatz LD, Winslow DL, Collins G, et al., and Investigators for the Terry Beirn Community Programs for Clinical Research on AIDS. Zidovudine alone or in combination with didanosine or zalcitabine in HIV-infected patients with the acquired immunodeficiency syndrome or fewer than 200 CD4 cells per cubic millimeter. N Engl J Med. 1996 Oct 10;335:1099-106.



To compare zidovudine monotherapy or zidovudine combined with didanosine (ddI) or zalcitabine (ddC) for disease progression and survival in patients with advanced HIV infection.


Randomized, double-blind, placebo-controlled trial with median follow-up of 35 months.


21 community clinics in 17 U.S. cities.


1102 patients who were ≥ 13 years old (mean age 37.5 y, 92% men, 55% white, 68% homosexual men) and had an AIDS-defining condition or CD4+ cell count of < 200 cells/mm3 (mean value 119/cells mm3). Exclusion criteria included abnormal laboratory values, a history or symptoms of pancreatitis or peripheral neuropathy, intolerance to study drugs, or advanced AIDS-related dementia. 77% of patients had previously used zidovudine (mean duration of use 12 mo). Follow-up was 97%.


Patients were assigned to zidovudine alone (n = 372), zidovudine plus ddI (n = 363), or zidovudine plus ddC (n = 367). Drug dosages were 200 mg 3 times/d for zidovudine, 200 mg 2 times/d for ddI, and 0.75 mg 3 times/d for ddC.

Main outcome measures

Disease progression (AIDS-defining condition) or death.

Main results

Analysis was by intention-to-treat. The study had sufficient power to show a reduction of 33% in disease progression or death with combination therapy. Overall, the groups did not differ for rates of disease progression or death (66% for zidovudine, 63% for zidovudine plus ddC, and 62% for zidovudine plus ddI; P = 0.24). The risk for esophageal or pulmonary candidiasis, cytomegalovirus disease, malignant conditions, other diseases, and all events (first and subsequent) and the rate of decline in CD4+ cell counts did not differ among groups. Patients who received zidovudine plus ddI had a lower risk for Pneumocystis carinii pneumonia (relative risk 0.65, 95% CI 0.44 to 0.96) and systemic Mycobacterium avium complex infection (relative risk 0.66, CI 0.44 to 0.98) than did patients who received zidovudine monotherapy. The groups did not differ for most adverse events; however, patients who received zidovudine plus ddI had a higher rate of adverse events that caused permanent discontinuation of study drugs (P < 0.05).


For patients who have advanced HIV infection, disease progression did not differ among patients who were receiving zidovudine monotherapy and those who received zidovudine combined with didanosine or zalcitabine.

Sources of funding: National Institute of Allergy and Infectious Diseases; Bristol-Myers Squibb; Glaxo Wellcome; Hoffmann-LaRoche.

For article reprint: Dr. L. Deyton, National Institutes of Health, 6003 Executive Boulevard, Bethesda, MD 20892, USA. FAX 301-402-0369.


The results of the trials by Hammer and Saravolatz and their colleagues should be considered with those of the recently published trial by the Delta Coordinating Committee (1). The overall results found by ACTG 175 are in general agreement with those found by the Delta Coordinating Committee and show that treatment with a combination of 2 nucleoside-analogue reverse-transcriptase inhibitors is more beneficial in decreasing disease progression and death than is treatment with zidovudine alone. These results have had a major effect on clinical practice; it is now a standard practice to treat patients with a combination of at least 2 nucleoside analogues.

The results of the study by Saravolatz and colleagues would seem to be at odds with those of ACTG 175. Major differences, however, are found in the patient population; this may have affected the overall results. As Corey and Holmes (2) state in an accompanying commentary, the study population in the trial by Saravolatz and colleagues included patients who had more advanced disease, received study treatment for a shorter period, had greater intolerance of study drugs, and were more likely to have taken zidovudine previously. When these differences are taken into account, the results of the 2 trials are comparable: In the trial by Saravolatz and colleagues, treatment with zidovudine and ddI consistently delayed disease progression compared with treatment with zidovudine alone throughout the study period, and the incidence of certain opportunistic infections was significantly lower. In a subgroup analysis of patients who previously had limited anti-retroviral experience or were inexperienced at entry, a benefit from combination therapy compared with zidovudine monotherapy was shown.

Both trials make it clear that the addition of ddC to zidovudine in patients with extensive previous use of zidovudine is not an appropriate treatment strategy. The addition of ddI to current zidovudine therapy was, however, beneficial in zidovudine-experienced patients. The results of recent trials suggest that a more appropriate clinical strategy might include the more potent protease inhibitors, especially in patients with advanced disease (3). Preliminary evidence shows that the addition of (or switching to) a potent protease inhibitor combined with a new nucleoside analogue may be effective (4).

One outcome of ACTG 175 that clinicians will find difficult to explain and that differs from the results of a previous trial (5) is the similar benefits seen with ddI monotherapy compared with zidovudine plus ddI. The relatively high incidence of zidovudine mutations present at baseline in the antiretroviral-inexperienced patients reported in a subset of patients in this study (6) may help to explain this result and account for the lower-than-expected rise in CD4+ cell count in patients in the zidovudine monotherapy group. Because of the greater effect that combination therapies have on surrogate markers and the general agreement from the results of other trials of the clinical benefit of combination therapies, most clinicians would not accept using ddI monotherapy as standard therapy when starting treatment for HIV infection.

It was not the intention of these trials to ascertain the optimum time to start anti-retroviral treatment. This remains uncertain with the treatments that are currently available. Limited efficacy of therapies (although clearly improved), drug intolerance, and poor compliance remain important issues in deciding when to start treatment. Little doubt exists, however, that the results of trials of combination therapies have given both the patient and physician more confidence about initiating treatment. I hope that the results from clinical trials of more potent combination regimens will further improve clinical effectiveness.

Ian G. Williams, MB, ChB
University College London Medical SchoolLondon, England, UK

Ian G. Williams, MB, ChB
University College London Medical School
London, England, UK


1. The Delta Coordinating Committee. Lancet. 1996;348:283-91.

2. Corey L, Holmes K. N Engl J Med. 1996; 335:142-4.

3. Cameron DW, Heath-Chiozzi M, Kravcik S, et al. Abstracts of the 11th International Conference on AIDS, Vancouver, BC, July 7-12, 1996:24.

4. Gulick R, Mellors J, Havlir D, et al. Abstracts of the 11th International Conference on AIDS, program supplement, Vancouver, BC, July 7-12, 1996:19.

5. Dolin R, Amatod DA, Fischl MA, et al. Arch Intern Med. 1995;155:961-74.

6. Rey D, Pi T, Diehl LJ, et al. In 5th International Workshop on HIV Drug Resistance, Whistler, BC, July 3-6, 1996:38.