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


Zidovudine, nevirapine, and didanosine in combination led to a more sustained reduction in plasma viral load than double-drug therapy

ACP J Club. 1998 Sep-Oct;129:42. doi:10.7326/ACPJC-1998-129-2-042

Source Citation

Montaner JS, Reiss P, Cooper D, et al., for the INCAS Study Group. A randomized, double-blind trial comparing combinations of nevirapine, didanosine, and zidovudine for HIV-infected patients. The INCAS trial. JAMA. 1998 Mar 25;279:930-7.



In patients who have HIV-1 infection without AIDS, what are the virologic and immunologic effects of various combinations of zidovudine, nevirapine, and didanosine.


Randomized, double-blind, controlled trial with 1-year follow-up.


Italy, the Netherlands, Canada, and Australia.


151 patients (mean age 37 y, 93% men) who were ≥ 18 years of age and had HIV-1 infection, CD4+ cell counts of 0.20 to 0.60 × 109/L, no previous AIDS-defining illnesses, and no previous antiretroviral therapy. Exclusion criteria included absolute neutrophil count < 1.0 × 109/L, platelet count < 0.8 × 1012/L, hemoglobin level < 95 g/L for men and < 90 g/L for women, or Karnofsky score < 80 points.


Patients were allocated to zidovudine plus nevirapine (n = 47), zidovudine plus didanosine (n = 53), or zidovudine plus didanosine and nevirapine (triple-drug therapy, n = 51). Doses were zidovudine, 200 mg 3 times/d; nevirapine, 200 mg once/d for 2 weeks and 200 mg twice/d thereafter; and didanosine, 125 or 200 mg twice/d based on body weight.

Main outcome measures

Changes in plasma HIV-1 RNA level and CD4+ cell count.

Main results

Analysis was by intention to treat. Patients who received triple-drug therapy had greater decreases in HIV-1 RNA levels than did those who received zidovudine plus nevirapine or zidovudine plus didanosine {median logarithmic change at 1 year -2.31 vs -0.24 and -1.04}*. The effect in the triple-drug therapy group was maximal at week 16 when 68% of patients had HIV-1 RNA levels < 20 copies/mL. More patients who received triple-drug therapy had HIV-1 RNA levels < 20 copies/mL at all visits between 40 and 52 weeks compared with those who received zidovudine plus nevirapine or zidovudine plus didanosine (45% vs 0% and 6%, P < 0.001 for both comparisons). Triple-drug therapy led to a greater increase in mean CD4+ cell counts than did zidovudine plus didanosine or zidovudine plus nevirapine (mean increase 0.139 × 109/L vs 0.087 × 109/L and -0.006 × 109/L, P = 0.03 for triple-drug therapy vs zidovudine plus didanosine). Rates of disease progression or death were 23%, 25%, and 12% for zidovudine plus nevirapine, zidovudine plus didanosine, and triple-drug therapy groups, respectively (P = 0.08 for triple-drug therapy vs zidovudine plus didanosine).


In patients who had HIV-1 infection without AIDS, triple-drug therapy with zidovudine, nevirapine, and didanosine led to greater and more sustained decreases in HIV-1 RNA levels and greater increases in CD4+ cell counts than did zidovudine plus nevirapine or zidovudine plus didanosine.

Sources of funding: Boehringer Ingelheim Pharmaceuticals; Instituto Superiore di la Sanita; National AIDS Therapy Evaluation Centre; Canadian HIV Trials Network; British Columbia Centre for Excellence in HIV/AIDS; Australian National Council on AIDS and Related Diseases.

For correspondence: Dr. J.G. Montaner, St. Paul's Hospital-University of British Columbia, 667-1081 Burrard Street, Vancouver, British Columbia V6Z 1Y6, Canada. FAX 604-631-5210.

*Numbers supplied by author.


The study by Montaner and colleagues is another in a series of well-designed anti-retroviral treatment trials that adds a substantial piece to the puzzle. Although the picture is still incomplete, clinically useful information can be gleaned. The authors used an ultrasensitive viral load detection assay (20 copies/mL threshold) but showed that 50% of the patients who received 3 drugs (2 nucleosides plus a nonnucleoside reverse transcriptase inhibitor) maintained viral suppression below detectable levels. Studies of triple-drug regimens that included a protease inhibitor showed similar findings.

This study confirms that triple-drug regimens are superior to double-drug combinations. This finding was most apparent at the lower levels of viral detection (20 vs 400 copies/mL). Patients with the lowest viral nadirs predictably had more pronounced long-term benefit; however, a surprising finding was that patients receiving the nucleoside and nonnucleoside combination actually fared worse than those receiving the double-nucleoside treatment. Used alone, nonnucleosides are not very potent and resistance quickly develops; judicious use is critical.

This study used patients with no previous antiretroviral exposure. Can these results be generalized to previously treated patients? Probably not. Nearly all studies show that treatment-naive patients do better.

The bottom line for the practicing clinician remains fuzzy. Recommendations for patients with either relatively high viral loads, relatively low CD4+ cell counts, or both should include ≥ 3 medications: a protease inhibitor and 2 nucleosides that the patient has not previously received. For patients with low viral loads, high CD4+ cell counts, or both, such regimens are also effective, but alternative options may exist. Based on this study, 1 such option may be 2 nucleosides and a nonnucleoside reverse transcriptase inhibitor.

Aaron E. Glatt, MD
Catholic Medical Center of Brooklyn and Queens, Inc.Jamaica, New York, USA