HIV treatment response and prognosis in Europe and North America in the first decade of highly active antiretroviral therapy: a collaborative analysis

Summary

Background

Highly active antiretroviral therapy (HAART) for the treatment of HIV infection was introduced a decade ago. We aimed to examine trends in the characteristics of patients starting HAART in Europe and North America, and their treatment response and short-term prognosis.

Methods

We analysed data from 22 217 treatment-naive HIV-1-infected adults who had started HAART and were followed up in one of 12 cohort studies. The probability of reaching 500 or less HIV-1 RNA copies per mL by 6 months, and the change in CD4 cell counts, were analysed for patients starting HAART in 1995–96, 1997, 1998, 1999, 2000, 2001, and 2002–03. The primary endpoints were the hazard ratios for AIDS and for death from all causes in the first year of HAART, which were estimated using Cox regression.

Results

The proportion of heterosexually infected patients increased from 20% in 1995–96 to 47% in 2002–03, and the proportion of women from 16% to 32%. The median CD4 cell count when starting HAART increased from 170 cells per μL in 1995–96 to 269 cells per μL in 1998 but then decreased to around 200 cells per μL. In 1995–96, 58% achieved HIV-1 RNA of 500 copies per mL or less by 6 months compared with 83% in 2002–03. Compared with 1998, adjusted hazard ratios for AIDS were 1·07 (95% CI 0·84–1·36) in 1995–96 and 1·35 (1·06–1·71) in 2002–03. Corresponding figures for death were 0·87 (0·56–1·36) and 0·96 (0·61–1·51).

Interpretation

Virological response after starting HAART improved over calendar years, but such improvement has not translated into a decrease in mortality.

Introduction

Accurate prognostic information on HIV-1 disease progression after starting highly active antiretroviral therapy (HAART) is important for patients, physicians, and health care providers. In 2002, the Antiretroviral Treatment (ART) Cohort Collaboration published estimates of the probability of disease progression up to 3 years after starting HAART, according to baseline age, transmission risk group, CD4 cell count, viral load, and clinical disease stage before HAART based on over 12 000 patients starting treatment between 1995 and 2000 in Europe, USA, and Canada.¹ Prognosis might improve with time given greater physician experience with HAART, earlier diagnosis, appropriate management of associated toxicities, and the availability of more potent, and less toxic, drugs.2, 3 The increasing availability of combined preparations has reduced the pill burden, which might facilitate patient adherence to regimens.4, 5 Conversely, the emergence of drug-resistant strains of HIV circulating in the infected population and changes in the characteristics of the patients starting HAART could be associated with poorer outcomes.6, 7

We analysed the updated database of the ART Cohort Collaboration to examine whether patient characteristics at the time of starting HAART, response to therapy, and disease progression have changed over time, using data combined from 12 cohort studies that followed up antiretroviral-naive patients from when they started therapy.