The Zodiac study (otherwise known as CNA30021) was a 48-week, randomized clinical trial comparing efavirenz (EFV, Sustiva, Stocrin) given with abacavir (ABC, Ziagen) 600 mg and lamivudine (3TC, Epivir) 300 mg once daily or abacavir 300 mg and lamivudine 150 mg twice daily in 770 antiretroviral-naive persons. The reasons for doing this trial were two-fold: to determine whether abacavir and lamivudine can really be given once daily as suggested by their intracellular half life, and to prepare the way for the new co-formulation of abacavir and lamivudine that GlaxoSmithKline is bringing to market.

Brian Gazzard presented the primary results of this trial in September 2003 at ICAAC. This presentation presented new data on the resistance profile among those patients who experienced viral failure. It used the U.S. Food and Drug Administration (FDA)-mandated TLOVR (time to loss of virologic response) analysis, which gives fairly similar results to the more familiar intent-to-treat, missing-equals-failure analysis. As we have come to expect from a regimen that uses efavirenz, lamivudine and a third drug, the response was excellent, with 68% achieving a viral load of less than 50 copies/mL.

The various virologic and CD4 outcomes were essentially identical for the once-daily and twice-daily regimens. Virologic failure occurred in only 8% of those on the twice-daily regimen and 10% of those on the once-daily regimen. Many of those classified as virologic failures maintained viral loads of less than 400 copies/mL, although follow up was limited. Thus, only 31 patients out of the 770 had a sample that was suitable for determining resistance.

Phenotypic testing at baseline (using the Virologic phenosense assay) showed reduced susceptibility to at least one drug in 1/15 (7%) of the patients in the twice-daily group, compared to 5/16 (31%) in the once-daily group, but the details were not given. At treatment failure, efavirenz resistance was seen in about 60% of the twice-daily group and about 80% of those in the once-daily group. Lamivudine resistance was detected at failure in 30% of the twice-daily group and 60% of the once-daily group. Resistance to abacavir was unusual at about 15%. Zidovudine (ZDV, Retrovir)/stavudine (d4T, Zerit) resistance was not seen.

Were there any surprises in the genotypic studies? In a word, no. Somewhat more patients in the once-daily group had baseline mutations (5/16 compared to 2/15). At failure, the majority of patients had efavirenz-resistance mutations (14/15 in the twice-daily group and 12/16 in the once-daily group). Resistance mutations to lamivudine at M184 were also common (5/16 and 2/15, respectively). Only one subject developed the K65R mutation.

So what did we learn from this analysis? Virologic failure is uncommon in efavirenz-containing regimens, but, when it occurs, it usually involves efavirenz resistance with or without lamivudine resistance. That is not new, but confirms several other studies, including the original efavirenz studies (DMP 006) and the more recent GS 903 study. There did not appear to be any significant difference in resistance between once-daily and twice-daily abacavir and lamivudine, but that conclusion has to be tempered by the small number of patients and the difference at baseline. Eliminating zidovudine from the regimen did not lead to new or surprising patterns of abacavir resistance, but as with the tenofovir (TDF, Viread), 3TC and efavirenz regimen used in GS 903, the overall potency and success of the regimen might have masked subtle differences in patterns that may come out with widespread use.

Are we ready to use abacavir and lamivudine once daily? So far, the data are consistent and encouraging, but the FDA will have to review the complete data set before they can decide on whether to approve the lamivudine/abacavir combo as a once-daily regimen.

Reference

Abstract: Analysis of Virologic Failure in a Clinical Trial of Abacavir Once Daily Versus Twice Daily With Lamivudine and Efavirenz (Poster 551)
Authored by: C. Craig, C. Stone, T. Bonny, H. Zhao, D. Gordon, S. Castillo, D. Paes
Affiliations: GlaxoSmithKline, Stevenage, UK; GlaxoSmithKline, Research Triangle Park, NC; GlaxoSmithKline, Greenford, UK