February 24, 2005
GS 99-9903, the registrational study for tenofovir (TDF, Viread), which was published last year in the Journal of the American Medical Association, clearly showed the relationship between the onset of several metabolic complications and the use of stavudine as part of a nucleoside backbone in antiretroviral-naive patients. In that study, stavudine was associated with a significantly higher incidence of hypertriglyceridemia, hypercholesterolemia, lipoatrophy and peripheral neuropathy when it was compared with tenofovir.
Many other studies have shown the benefits of using a stavudine nucleoside-sparing regimen for the treatment of HIV infection in naive and experienced patients. Other studies have also shown similar gains by sparing thymidine analogs all together for patients initiating antiretroviral therapy. For the most part, the efficacy observed with thymidine analog nucleoside-sparing regimens has been comparable to the response observed with either zidovudine (AZT, Retrovir)- or stavudine-containing regimens. But the rate of adverse events seen with thymidine analogs has flipped the scale toward favoring the use of other reverse transcriptase inhibitors such as tenofovir or abacavir (ABC, Ziagen).
This study, which was conducted by Daniel Podzamczer and colleagues from Spain, aimed to compare the effectiveness of abacavir versus stavudine in patients who had never been on HIV treatment. The researchers also carefully looked at the effect of these 2 therapies on lipodystrophy and lipid abnormalities.
Two hundred thirty-seven patients were equally randomized to receive either abacavir or stavudine in combination with lamivudine + efavirenz. The researchers then assessed metabolic parameters such as lipodystrophy and mitochondrial toxicity by using several objective and subjective parameters. They measured venous lactate, high-density lipoprotein cholesterol (HDLc), low-density lipoprotein cholesterol (LDLc), apoprotein A1 and apoprotein B. They also collected peripheral blood mononuclear cells for mitochondrial DNA and obtained dual energy X-ray absorptiometry (DEXA) scans for objective anthropometric measures.
One hundred twenty-two patients received a regimen of stavudine + lamivudine + efavirenz, while 115 patients received abacavir + lamivudine + efavirenz. Baseline characteristics were quite similar between both groups, with a median CD4+ cell count of 213 CD4+ cells and an HIV RNA of 5.2 logs.
After 96 weeks, abacavir was found to be superior to stavudine in an intent-to-treat (ITT) analysis (47.5% in the stavudine group versus 60.9% in the abacavir group achieved an HIV RNA of less than 50 copies). The response in the as-treated population was 85.3% versus 87.5% in the stavudine and abacavir groups, respectively. CD4+ cell counts increased by 294 cells for the stavudine group versus 263 cells for the abacavir group.
When both the physicians and patients were asked about any observed changes in body fat, both reported body fat loss over face, buttock and/or extremities in a statistically greater amount in the stavudine group than the abacavir group. These reports were then correlated with objective anthropometric measures by DEXA scans. The changes perceived by patients and physicians did not statistically differ from the ones measured by DEXA scans. Despite the higher peripheral fat loss in the stavudine arm, total body weight did not change significantly between both groups.
Other significant changes in metabolic parameters between both groups were with regards to the HDLc and the apoprotein A1 level. There were no changes in the lactate, total cholesterol (TC), LDLc, LDL/HDL ratio, triglycerides and apoprotein B levels.
Fifty-nine patients discontinued therapy in the stavudine group, while 40 patients discontinued therapy in the abacavir group. The incidence of abacavir hypersensitivity reaction was not clearly reported, but there were 4 deaths in the study (2 in each group).
The authors concluded that patients treated with stavudine showed a higher rate of lipoatrophy than patients treated with abacavir. In addition, abacavir had a better lipid profile, lower triglycerides, a higher increase in HDL and apoprotein A1, and a decreased TC:HDL ratio.
This study started to enroll in 2001, at which time we did not know what we know today about the use of stavudine in naive patients. It is possible that some of the discontinuations in the stavudine arm may have been caused by the perceived body changes that had already been attributed to stavudine. Those discontinuations potentially affected the ITT analysis and probably underscored the rate of other metabolic complications in the patients who discontinued the study earlier due to stavudine toxicity.
Overall, this study was quite well conducted and confirmed once again some of the metabolic alterations that can occur with the use of stavudine in this population.
There are 2 main lessons from this study. The first is the validation of previous findings observed on the Gilead 903 Study with regard to the metabolic abnormalities seen with the use of stavudine in antiretroviral-naive patients. This study also served to remind us that abacavir is another option as a thymidine-sparing nucleoside backbone regimen, and is not associated with the metabolic changes seen in patients treated with stavudine.