Switching to zidovudine plus lamivudine plus abacavir maintains viral suppression in patients with high viral load before antiretroviral therapy: a retrospective clinical cohort analysis.

Abstract

Triple nucleoside therapy with combination tablets of zidovudine plus lamivudine plus abacavir (Trizivir) provides simple food-independent twice daily therapy. However, studies in initial therapy suggest that the efficacy of this regimen may be sub-optimal when initiated in individuals with baseline viral loads greater than 100 000 copies/ml [1]. Several switch studies [2,3] have reported data indicating that switching from initial successful therapy to the zidovudine, lamivudine and abacavir combination maintains virological efficacy, and may improve metabolic parameters [4]. Data on whether this triple nucleoside analogue combination is effective in individuals who initiated their original highly active antiretroviral therapy with viral loads greater than 100 000 copies/ml and then switched have not been reported. Using the Chelsea and Westminster Hospital, London database, we retrospectively evaluated the efficacy of switching therapy when the viral load was less than 50 copies/ml from an initially successful regimen to zidovudine, lamivudine and abacavir (with no additional antiretroviral agents). Analyses were performed using the pre-antiretroviral therapy (ART) viral load and CD4 cell count. We identified 32 patients with a pre-ART and at least one post-switch viral load and CD4 cell count. Only three patients had previously received a dual therapy regimen (all didanosine based, without lamivudine). All other patients had commenced on triple or quadruple therapy with either a single protease inhibitor (PI) (n = 16), dual PI (n = 3), non-nucleoside reverse transcriptase inhibitor (n = 8) or non-nucleoside reverse transcriptase inhibitor plus PI (n = 2) plus two nucleoside reverse transcriptase inhibitors. Stavudine was the initial thymidine analogue in 20 and didanosine the initial non-thymidine analogue in nine patients. Nine patients had made regimen modifications before switching to zidovudine, lamivudine and abacavir. The median pre-ART viral load was 193 093 copies/ml [interquartile range (IQR) 42 957–426 084] with 20 patients commencing at more than 100 000 copies/ml. The CD4 cell count was a median of 155 cells/mm3 (IQR 81–304) at baseline and less than 100 CD4 cells/mm3 in 12 patients. Patients received initial ART for a median of 28 months (IQR 15–38) before switching. On switching to zidovudine, lamivudine and abacavir, the median CD4 cell count was 447 cells/mm3 (IQR 325–581). Patients were followed on zidovudine, lamivudine and abacavir for a median of 6.5 months (mean 9.3 months). Over that time the median CD4 cell count rose by a further 29 cells/mm3 to 476 cells/mm3 (IQR 373–618). No patients have experienced virological rebound regardless of their pre-ART viral load or CD4 cell count. Data from our cohort data are consistent with virological control data from clinical trials, demonstrating the maintenance of viral efficacy after switching to the zidovudine, lamivudine and abacavir combination [2,3]. The viral efficacy of the zidovudine, lamivudine and abacavir combination as an alternative or ‘maintenance’ therapy after an initial ‘induction’ ART regimen is maintained regardless of high (> 100 000 cells/ml) and low (< 100 cells/mm3) CD4 cell count before the initiation of ART. Graeme J. Moyle Brian G. Gazzard