We found that giving atazanavir in a well established cohort of patients receiving HAART was safe and effective. A total of 5873 HIV1 infected patients have been followed up in the HAART era and of these, 241 patients were given atazanavir, comprising 4.1% of the total cohort. From the 241 atazanavir exposed patients, 231 patients (95.8%) received atazanvir (300 mg daily) boosted with ritonavir (100 mg daily) and the remaining 10 received atazanavir alone (400 mg daily).
A total of 113 patients (46.8%) switched to atazanavir from other antiretrovirals (table 1
). Of the 241 patients, 76 (31.5%) switched to atazanavir secondary to an adverse drug reaction with a previous antiretroviral, nine (3.7%) switched because of adherence issues, and a further 28 after the premature end of a local efavirenz based trial.
 | Table 1 Drugs switched to atazanavir |
Figure 1
shows the changes in viral load (intent to treat and on treatment) and CD4 count in both the PI naive (fig 1A
) and experienced (fig 1B
) patients. For PI naive patients in an intent to treat analysis, viral suppression for <500 copies/ml and <50 copies/ml, occurred in 72% and 60% of patients at one year respectively. For PI experienced patients in an intent to treat analysis, viral suppression for <500 copies/ml and <50 copies/ml, occurred in 90% and 57% of patients at one year respectively. The mean CD4 count rise between PI naive and experienced patients showed no significant increases.
 | Figure 1(A) Viral suppression (on treatment and intent to treat) and change in CD4 count in PI naive patients. (B) Viral suppression (on treatment and intent to treat) and change in CD4 count in PI experienced patients. |
Figure 2
shows the decline in viral load in patients who were and were not PI experienced over one year. While PI naive experienced patients experienced a greater fall in viral load, no significant differences were seen between these two groups. Switching to atazanavir when the viral load was <50 copies or <500 copies/ml led to continued viral suppression for a year, in both an intent to treat and on treatment analysis (fig 3
).
 | Figure 2Change in viral load in PI experienced and naive patients over 12 months. |
 | Figure 3Viral suppression over one year in patients with previously suppressed viraemia. |
Table 2
shows the changes in cholesterol over 12 months of follow up and shows that the proportion of patients with a total cholesterol greater than 6.5 mmol/l decreased over time while the average cholesterol showed a sustained decrease.
| Table 2Changes in cholesterol over 12 months of follow up. The proportion of patients with a raised total cholesterol (TC>6.5 mmol/l) is shown. |
Figure 4
shows the main toxicity previously seen in other studies, unconjugated hyperbilirubinaemia. A small increase in bilirubin was seen over the 12 month period, supporting retrospective data that has evaluated atazanavir related bilirubin increase as an adherence marker.
28 | Figure 4Incidence of hyperbilirubinaemia. |
There were 17 virological failures and we found that the mutations 10F/I, 84V, and 90M are prevalent in resistant patients, as shown by other data.
29 A total of 47 patients stopped atazanavir because adherence or lost to follow up issues (20 patients), virological failure (9 patients), toxicity (9 patients), a structured treatment interruption (4 patients), acute hepatitis C (1 patient), patient request (1), and never taking the drug (1). Of the nine patients who stopped taking atazanavir because toxicity, four stopped because of jaundice, two because of diarrhoea, one because of depression, one because of a flare of hepatitis C, and one because of an abacavir hypersensitivity reaction in their HAART regimen. A total of four patients died, one because of to bacterial pneumonia, one because of Hodgkin's disease, and one because of progressive multifocal leucencephalopathy.
While these data are not randomised and do not show long term durability or potency, in a prospective cohort of 241 patients who took atazanavir with one year of follow up data, we show that (1) atazanavir was well tolerated with few cases of virological failure, (2) its use was successful in patients who were PI naive and experienced, (3) switching to atazanavir when the viral load was <50 copies/ml led to continued viral suppression, (4) cholesterol concentrations decreased in patientss receiving atazanavir, (5) the major adverse event was mild hyperbilirubinaemia, which rarely led to treatment cessation, and (6) failure with ritonavir/atazanavir was not associated with the development of resistance mutations. Although some may consider that the intent to treat data were disappointing, it is important to note that a number of patients who stopped atazanavir were lost to follow up, a problem with prospective cohort studies in general.
Drug side effects and patient reported symptoms are the foremost variables predictive of non‐adherence to HAART.
30 Future HIV standard of care will emphasise the use of once daily therapies, especially as HAART penetrates the under‐developed world.
31
