Results of important studies often become available when they are presented at national and international conferences, this being well before publication in a medical journal. Sometimes, however, the study can be terminated with results of safety issues unveiled even earlier. In other words, the study design can be changed or terminated when there is a significant disparity in treatment for a particular study arm; those study subjects may be taking a significantly inferior treatment or one in which a significant toxicity is more common. In this situation a letter is generally sent to all the investigators of a study.
This is the case in a study comparing Trizivir and Sustiva. In this instance a communication was mailed to all HIV practitioners; investigators studying Trizivir vs. Sustiva (efavirenz) felt there to be a significant difference in effectiveness of one treatment arm and one inferior arm was terminated. Let's dissect the information to investigate the possibility of any existing drama in ACTG A5095 (AIDS Clinical Trials Group).
All study participants of ACTG A5095 were naive to antiviral medications (having no experience on any antiviral meds previously); their antiviral treatment arm was chosen at random in a double blinded manner (neither physician nor patient would know which regimen the patient was taking). There was not a CD4 count criteria for entrance into this trial. The study participants were to receive either:
Trizivir consists of Retrovir (AZT), Epivir (3TC) and Ziagen (abacavir). Combivir consists of Retrovir and Epivir.
1,147 patients were followed for an average of 32 weeks. The mean (average) CD4 T-cell count at baseline (start of study) was 238 cells and HIV RNA was 78,825. However, 43% of study patients had viral loads greater than 100,000 copies vs. 57% had HIV RNA below 100,000 at start of study.
Since the National Institutes of Health sponsors and funds all ACTG studies, they sent a letter to physicians informing them of the interim results. They reported that 21% of the patients on Trizivir only had been considered virologic failures vs. 10% in the other two groups (both Sustiva arms) combined. It is worthy to note that this study considers and deems virologic failures for viral loads greater than 200 copies at least four months after starting treatment.
With these results in hand, an independent panel, the Data Safety Monitoring Board (DSMB), recommended stopping the Trizivir arm and is providing the volunteers with the option of an alternative regimen, whether they were failing (viral loads above 200 copies) or still at undetectable levels. It is common to have an independent panel of experts review studies during the ongoing trial, they would intercede for the benefit of the patients. Additionally, the remaining patients of the two other treatment arms who were administered Sustiva were told they were taking Sustiva but still blinded as to which nucleosides were their background treatment.
According to sources familiar with this study, 74% of patients randomized to Trizivir arm and 89% of the pooled Sustiva arms still maintained viral loads undetectable at an average at week 48 (Intent-to-Treat analysis). Approximately 50% of patients had CD4 counts less than 200 and the median CD4 count was 214 cells in this study. In fact, 17% of patients randomized to the Trizivir arm and 20% in the pooled Sustiva arms had CD4 counts between 0-50 cells.
One should understand that all patients in this study had an equal chance of being randomized to any of the three regimens, even when knowing that patients who had either very low CD4 count or very high viral loads before treatment, or at baseline, have an increased risk of virologic failure. Moreover, there have been several studies demonstrating that patients with viral loads greater than 100,000 would be better off with a quadruple regimen (two protease inhibitors, for example) or one that contains Sustiva.
Particularly, another study, DMP-006, was completed several years ago and ran for more than three years. It was historically important because of several significant findings and it led to the quick approval of Sustiva by the FDA. Here patients were randomized to either Sustiva or Crixivan, each with Combivir (vs. Crixivan + Sustiva alone). Sustiva was shown to be superior over Crixivan, and a sub-analysis in this trial demonstrated patients with CD4 count below 100 or high viral RNA at baseline were also successfully treated with Sustiva.
Moreover, other studies showed the benefit of administering dual protease inhibitor regimens in these immune-suppressed patients as being more advantageous for durability (longer time to treatment failure), than a triple therapy regimen. With this in mind, many physicians do not offer a simplified regimen of Trizivir for the patient considered very advanced or in a late stage of disease.
One wonders the rationale for designing a study that potentially places the immune-devastated patient on Trizivir alone. Alternatively, the results of a Sustiva-based treatment (not to mention a quadruple therapy arm containing Sustiva) in this advanced disease population beating out Trizivir was not a revelation nor a shock to us.
The results of this trial should not be overlooked. One does not criticize this ACTG study with its strict study design as being double-blinded and placebo controlled. However, one notes that placing patients in a blinded study, in which Trizivir randomized patients are instructed to take their Sustiva placebo on an empty stomach unfairly abolishes the real life advantage of one Trizivir pill twice daily, with or without food. Also, uncertainty does exist regarding the use of certain three- and four-drug combinations utilizing specific drug classes, which may be one of the questions that the ACTG intended to answer with this trial. However, from a clinician's experience, some things remain as clear as before this study: Trizivir does have its place in the HIV treatment armamentarium. Sustiva-based regimens are very powerful. Also, for the advanced disease patient, more potent regimens need to be considered so that those individuals have a better chance of success with their first regimen.
Daniel S. Berger, M.D., is Medical Director of Chicago's largest private HIV treatment and research center, NorthStar Healthcare, and Clinical Assistant Professor of Medicine at the University of Illinois at Chicago and editor of AIDSInfosource (www.aidsinfosource.com). He also serves as medical consultant and columnist for Positively Aware. Dr. Berger can be reached at DSBergerMD@aol.com or (773) 296-2400.