AZT: attacking AIDS early and hard.

The witches in Shakespeare's Macbeth assured the Scottish monarch that "Macbeth shall never vanquish'd be until, Great Birnam wood to high Dunsinane hill, Shall come against him." Those assurances rang hollow when Macbeth's enemies marched to Dunsinane under cover of the trees they had cut down in Birnam wood.

Literature and legend often require protagonists to survive by resolving such a paradox. Now, two conflicting reports in the Aug. 17 New England Journal of Medicine pose a paradox to physicians about when they should treat patients with drugs that inhibit HIV, the virus that causes AIDS.

Although the studies, one from a U.S. and one from a European team, present apparently contradictory results on whether early treatment with the drug AZT (zidovudine) prolongs life, new understanding of HIV infection suggests that treating patients within months of infection offers the best chance of stymieing the virus.

Early treatment of HIV infection has a checkered past. In 1990, a U.S. team announced that treating asymptomatic HIV-positive individuals with AZT maintained their concentrations of CD4 cells - white blood cells that HIV destroys. Researchers hoped that high CD4 counts would slow the disease.

But in April 1994, the Anglo-French Concorde study indicated that early treatment with AZT neither prolongs survival nor slows the onset of opportunistic infections. Subsequently, some doctors decided to wait for their patients' immune systems to falter before prescribing AZT.

In 1994, however, researchers reported that giving HIV-infected mothers AZT shortly before they gave birth decreased the risk of passing HIV to their newborns (SN: 2/26/94, p. 134). This finding bolstered the theory that treating HIV with antiviral drugs early on could hold off disease.

The new results seem to reinforce the Concorde findings. The researchers found that people who take AZT after a long diseasefree period, when their CD4 counts are still high - around 700 - lived no longer than people who initiated therapy when their immune systems began to weaken, as indicated by CD4 counts of less than 500.

The European researchers, on the other hand, treated patients so early in their infection that they had not yet developed antibodies to HIV. Study collaborator Luc Perrin of Geneva University Hospital in Switzerland notes that most participants suffered from an HIV-associated flulike illness known as anti-retroviral syndrome, which may strike when a person is first infected and often heralds a poor prognosis.

Perrin and his colleagues found that starting AZT within 3 months of infection decreased the number of minor HIV-associated infections and kept CD4 cell counts up. The researchers don't know whether the drug will result in longer survival, but they point out that they are hitting the disease at a vulnerable stage.

"We [need] a very strong antiviral effect early in the infection," says Sabine Kinloch-de Loes of Geneva University Hospital. It won't cure these patients, but "it could change the prognosis."

During the earliest period of infection, HIV replicates rapidly. It mutates frequently as it reproduces, which helps it to evade the immune system and become resistant to retroviral drugs like AZT, 3TC, and protease inhibitors.

Any HIV-infected person carries a swarm of different strains, "but a person is initially infected with [only] a few types of viruses," says Perrin. The limited variety of HIV at this point makes it more susceptible to drugs. The researchers advocate early treatment, preferably with combinations of antiviral drugs.

Steven Schnittman of the National Institute of Allergy and Infectious Diseases in Bethesda, Md., supports the European strategy. "The only way to really know is to look at the patients in the long run," says Schnittman. "But it is the most promising thing we have seen" for early infection.