Some of the pros and cons of protease inhibitor (PI)-sparing regimens were discussed by Dr. Christine Katlama from the Hopital de la Pitie-Salpetriere, Paris, France. These points include:
55 percent of subjects in the Ziagen arm and 45 percent in the Crixivan arm completed week 48. Adverse advents were the main reason for stopping therapy, with slightly more leaving from the Crixivan arm. Only 5 percent of subjects in each group discontinued treatment because of virologic failure. Seven percent of the 280 patients treated with Ziagen developed hypersensitivity, whereas four of the patients (not on Ziagen) in the Crixivan arm were judged to have Ziagen hypersensitivity.
In the intent-to-treat analysis, both arms had similar results using a standard viral load test (400 copies). Treatment was considered a failure for those participants who changed therapy, left the study for any reason, had missing data, or whose viral load did not go below 400 copies. Fifty-one percent of both groups had viral loads below 400 copies at week 48. In the as-treated analysis, 86 percent of the Ziagen arm and 94 percent of the Crixivan arm had viral loads below 400 copies. Looking at ultrasentive viral load testing (50 copies), more people in the Crixivan arm were below the limits of the test. This was especially true for participants who entered the study with viral loads above 100,000 copies, though there was no difference in the time to viral rebound. CD4 counts were similar in the two treatment groups.
These findings may mean that for people with viral loads between 10,000 to 100,000 copies, the triple nucleoside regimen of Ziagen and Combivir is equivalent to the protease inhibitor-containing regimen of Crixivan and Combivir. However, in patients with higher viral loads, the protease-containing regimen may be better at suppressing viral load.
The study included 298 individuals from France, the U.S., Poland, Spain, Germany, Belgium, Italy, Portugal, Hungary, Canada and the Netherlands. Participants in this trial were randomized to one of three regimens:
About 80 percent of the 298 participants have completed 48 weeks. At study entry, mean viral load was 25,119 copies and CD4 cells were 447. Rates of viral suppression in the intent-to-treat analysis, using standard viral load testing (below 500 copies), was 59 percent for the Crixivan arm, 55 percent for the Viramune arm, and 57 percent for the three nucleoside arm. Rates in the as-treated analyses were 95 percent, 91 percent, and 90 percent, respectively. Viral suppression below 50 copies (ultrasensitive assay) was not as good in the triple nucleoside arm.
Similar results were found for participants entering the study with the highest viral loads, with the three nucleoside arm being less effective at suppressing viral load below 50 copies. CD4 counts were similar among the three treatment groups, with increases up to 167 CD4 cells at 48 weeks.
This study may show more support for using PI-sparing regimens in individuals without prior treatment experience who have low-to-moderate baseline viral loads. In people with higher baseline viral loads who have been on antiretrovirals, PI-containing regimens may be more effective.
Results of CNA 3005 and the Atlantic study show the same trends as the 72-week data of DuPont's 006 study, reported in this issue of Positive Living. That study showed that the triple nonnucleoside combination of Sustiva/Retrovir/Epivir was superior to Crixivan/Retrovir/Epivir, even in people with viral loads over 100,000 copies. While favoring PI-containing regimens for people with advanced disease is common among physicians, Katlama stressed the fact that clinical progression rates remain very low in most large studies of PI-sparing regimens. Lack of data in patients with advanced disease is the major issue with PI-sparing regimens, Katlama believes.