Evid Based Med: first published as 10.1136/ebm.1996.1.216 on 1 December 1996. Downloaded from Lamivudine plus zidovudine was an effective Initial therapy for HIV-1 infection

Katlama C, Ingrand D, Loveday C, et globin level, platelet count, creati- treated with combination therapy al.yfor the Lamivudine European HIV nine level, or serum amylase level; (rather than zidovudine monother- Working Group. Safety and efficacy previous anti-HIV therapy other apy) for 24 weeks to prevent 1 addi- of lamivudine-zidovudine combi- than zidovudine; history of periph- tional patient from having a CD4+ nation therapy in antiretroviral- eral neuropathy; or intolerance to cell count below baseline, 95% CI 2 naive patients. A randomized zidovudine. Follow-up was 88%. to 4; the relative risk reduction was controlled comparison with zidovu- 77%, CI 55% to 89%.}* Patients dine monotherapy. JAMA. 1996 Intervention who received combination therapy Jul 10; 276:118-25. Patients were allocated to zidovudine, had a greater mean reduction in 600 mg/d, and lamivudine, 300 mg viral load at 24 weeks than did pa- twice daily (n = 65), or zidovudine tients who received zidovudine Objective plus lamivudine placebo (n = 64). Af- monotherapy (P = 0.008 by the im- To determine the effectiveness of ter 24 weeks, all patients could receive mune capture method and P < 0.001 lamivudine and zidovudine combi- lamivudine. by the Roche method). The groups nation therapy compared with zido- Main outcome measures did not differ for clinical events or vudine monotherapy in patients + toxic effects, which were mild. with HIV-1. Change in CD4 cell count, HTV-1 RNA viral load, and adverse and Conclusion Design toxic events. Lamivudine plus zidovudine combi- 24~week randomized, double-blind, Main results nation therapy was well tolerated placebo-controlled trial. Analysis was by intention to treat. and more effective than zidovudine + Setting At 24 weeks, patients who received monotherapy in increasing CD4 14 hospitals in Europe. combination therapy had an in- cell count and reducing viral load in crease in CD4+ cells of 80 cells/^L patients with HIV-l. Patients compared with a decrease of 10 Sources of funding: Glaxo Wellcome Research 129 patients who were > 18 years of cells/fiL in patients on zidovudine and Development and Agence Nationak de age (mean age 35 y, 74% men) and mono-therapy (P< 0.001). 13% of Recherches sur le SIDA, France. were infected with HIV-1, had re- patients who received combination For article reprint: Dr. C. Katlama, Service + ceived minimal previous zidovudine therapy had a CD4 cell count below des Maladies Infectieuses, Parasitaires et therapy (< 4 weeks), had a CD4+ cell baseline compared with 57% of pa- Tropicales, Hospitaller Pitie Salpetriere, http://ebm.bmj.com/ count between 100 and 400/pL, tients who received zidovudine 47Bdde I'Hopital, 7SO 13 Paris, France. and had a Karnofsky score > 70. Ex- monotherapy {P < 0.001}*. {This ab- FAX 33-142-160-126, clusion criteria were abnormal liver solute risk reduction of 44% means "Numbers calculated from data in article. function, neutrophil count, hemo- that 2 patients would need to be Commentary on September 26, 2021 by guest. Protected copyright. These 2 clinical trials confirm the results of weeks, patients were allowed to receive drugs and avoiding die selective pressure an earlier report (1) and show that the com- open-label combination therapy; the results that favors the appearance of drag-resistant bined activity of zidovudine and lamivudine suggest that effects are maintained at 48 HIV mutants. More than 80% of the pa- is more effective than monotherapy in re- weeks, but the study design and number of tients treated with combination zidovudine ducing plasma HIV RNA and increasing patients do not allow for valid conclusions. and lamivudine developed the 184V muta- CD4+ cell counts in patients with HIV in- Limited effectiveness of zidovudine in the tion associated with resistance to lamivudine, fection. The authors also extend their find- early treatment of HIV infection became but this may delay the appearance of zido- ings to patients with a mean of > 2 years of evident about 4 years ago (3) and led to sev- vudine-resistant mutants and preserve previous zidovudme therapy. In both the eral trials of combination therapies. Not zidovudine susceptibility. Although neither naive and experienced groups, the superior only zidovudine and lamivudine but also study was able to show clinical effectiveness suppression of HIV with the combined regi- zidovudine and didanosine; zidovudine and or what is more important, a meaningful men was well documented by 3 methods zalcitabine; and more recently 3 drug survival benefit, this is hardly surprising and further supported by increased CD4+ combinations, including a protease-inhibi- given the relentless but very slow course of cell counts, decreased immune activation tor (4), have been reported to be superior HTV infection during clinical latency. Also, markers (p-2 microglobulin and neopterin), to zidovudine monotherapy in controlled a much longer observation period is needed and a trend toward fewer Centers for Dis- clinical trials. Such treatments are also gen- to determine whether the antiviral effect di- ease Control and Prevention category C erally well-tolerated (5). The rationale for minishes with time. The decrease in HIV events. A meta-analysis of 4 similar trials using combinations of 2 or more drugs in- RNA that was achieved, however, is an im has confirmed these results (2). After 24 cludes maximizing the synergistic effects of (Continued on page 217) 216 Therapeutics Evidente-Based Medicine November/December 19%