I've never seen so many doctors so reluctant to accept the results of a clinical trial. Everyone wanted so badly for Trizivir to show stellar results.
As three medications in one tablet taken twice a day, Trizivir's easy-to-take, easy-to-tolerate profile makes it an attractive option (if you're not allergic to it). This is in spite of the fact that the three medications in Trizivir (made by GlaxoSmithKline) -- Retrovir (AZT), Epivir (3TC) and Ziagen (abacavir) -- are all in the same drug class, nucleoside analogue, or nuke for short. Usually, combination therapy consists of at least two of the four drug classes on the market.
Here, when Trizivir was compared to Trizivir plus Sustiva or Combivir plus Sustiva, people on Trizivir alone had more virologic failure. As a result, the Data and Safety Monitoring Board (DSMB) overseeing this clinical trial, ACTG 5095, stopped the Trizivir arm of the study early. (Combivir is made up of two of the three drugs in Trizivir -- Retrovir and Epivir.) But this clinical trial was a little more complicated than it would seem, as Dr. Daniel Berger explained in the May/June issue of Positively Aware (see "The Buzz").
Dr. Roy M. Gulick of Cornell University faced an onslaught of questions when he presented the results of this study in Paris during the International AIDS Society conference in July. ACTG 5095 was a large trial of 1,200 people, 19% of them women, more than half of all participants people of color (60%). (ACTG stands for AIDS Clinical Trials Group.) All participants were treatment naive -- that is, they were taking anti-HIV therapy for the first time. The treatment naive have the best results in clinical trials, as do most people taking therapy for the first time.
Trizivir did worse at lowering viral load whether or not people started with more or less than 100,000. Twice as many people had virologic failure, defined as going above 200 viral load twice at 16 weeks or later, as compared to the Sustiva arms of the study (21% vs. 11%).
Moreover, the time to virologic failure was shorter with Trizivir. For most of the Trizivir failures, that was within four weeks. Looking at whether people started out with more than 200 T-cells or less, the results were the same. The simple conclusion of the study, not surprisingly, was that, "In treatment-naive patients, [Trizivir] was inferior to [Sustiva]-containing treatment in terms of rates and time to virologic failure."
One audience member pointed out Trizivir's advantages in simplicity, plus lack of drug interactions and co-pays for other drugs. Another asked about the clinical disadvantages of Trizivir for the 20% of failures who didn't have enough viral load to run a resistance test (1,000). In other words, they still have relatively low viral load and wouldn't be expected to get sick.
Gulick said that, "A strict definition of virologic failure is needed in this study to power the results [make them statistically relevant]. But in the clinic you can use your judgment and look at what patients want to do with the information."
Another audience member raised the question of drug resistance. If most people had the Epivir signature mutation (M184V) at the time of first failure (before the routine confirmation viral load test), doesn't this raise the issue of inadherence or of inaccurate reporting of good adherence? Gulick said that 22% of failures had wild type virus -- indicating that those people took little or no medicine. (Nevertheless, this would point to another glitch in the Trizivir glamour -- it's supposed to lead to better adherence, not worse, because it's more convenient to take.)
Another audience member asked, "Are we to give up Trizivir forever?" Gulick replied, "Trizivir did have efficacy. If you look at other studies, that can help you."
Yet another member of the audience pointed out that because Sustiva was in two of the three arms of the study, that the Sustiva arms actually had half of the failures. "That's worse, because you fail two classes [of drugs]. Half the failure with double the resistance is not so great." Another doctor made the same point after the presentation: when you have resistance to Sustiva, you have resistance to the entire non-nucleoside analog class of drugs. He called that "an expensive failure."
A Florida researcher said that when clinical trial participants don't have early strong results in viral load drop, the clinic brings in those participants and focuses on adherence, and "nine out of 10 times, that's the problem. The pill count is off."
One doctor pointed out that because the study was placebo-controlled -- so that everyone took the same amount of pills in the same way -- Trizivir's simplicity of only two pills a day was compromised. He said Trizivir was studied "with one arm tied behind its back."
Doctors also lamented the loss of another advantage if they have to cut back on Trizivir use -- one co-pay for three drugs.
Perhaps the biggest question raised: Is a three-nuke regimen really inferior? Gulick pointed out that the remaining Sustiva arms of the trial have not been unblinded. (Participants are still taking medicine in a blinded fashion -- neither they nor the clinic staff know which regimen they're on.) In other words, maybe Trizivir with Sustiva will have better results than Combivir with Sustiva. Nevertheless, Trizivir didn't make it on its own.
There's no question that people are doing well on Trizivir. They can continue their regimen with the same careful monitoring anyone else should be getting while on HIV therapy. (They can also choose to switch or intensify with another drug if they want.)
The question is, should anyone be started on Trizivir as their first regimen? The answer remains to be seen. So far, it may be that only people with low viral loads should be started on it. In this trial, it was recommended that even the people who were undetectable with Trizivir add Sustiva, or that they switch out Trizivir for Combivir and add Sustiva.