About The Study

Avanti III was the third in a series of small international studies designed to investigate multi-drug combinations in antiretroviral-naive volunteers, and identify the most promising regimens for testing in large clinical trials. This trial tested AZT plus 3TC versus AZT plus 3TC plus nelfinavir.

Study Approach

This was an international, double-blind (neither doctors nor volunteers knew which therapy volunteers received), comparative trial where volunteers, with CD4 count between 150 and 500 cells/mm3, were randomized to receive either AZT(200mg td)/3TC(150mg bd) or AZT/3TC/nelfinavir(750mg td) for 52 weeks. Volunteers were assessed regularly for viral load, CD4 count and safety. Due to changes in treatment guidelines, participants were permitted to switch to open-label AZT/3TC/NFV on completion of 28 weeks of blinded therapy.

Study population

One hundred and two volunteers from Europe, Canada and Australia were randomized. The median baseline viral load (log (10) copies/ml) was 4.8 for the AZT/3TC group and 5.0 for the AZT/3TC/nelfinavir group, while the median baseline CD4 count was 287 and 279 respectively.


The median decrease in viral load over 28 weeks was -0.98 for the AZT/3TC group and -1.85 for the AZT/3TC/nelfinavir group. The percentage of volunteers with a viral load below 500 copies at 28 weeks was 18% for the two-drug arm and 83% for the three-drug arm. The data at week 52 showed that more participants in the AZT/3TC group reported nausea, while more participants in the AZT/3TC/NFV group reported diarrhea and throat & tonsil discomfort.


Both treatment groups demonstrated improvement in viral load over the 28 week period. However, volunteers in the three-drug arm reached a lower viral load compared to volunteers in the two-drug arm. In addition, more volunteers had a viral load below 500 copies/ml after 28 weeks of treatment in the AZT/3TC/nelfinavir group.

Note: These results were taken from abstract 8 presented at the 5th Conference on Retroviruses and Opportunistic Infections in February 1998. The 52-week data were taken from a Glaxo poster presented in Glasgow in November 1998.