Low rate of toxicity-related treatment change suggests HIV treatment tolerable in the long-term

Michael Carter
Published: 29 May 2008

HIV-positive patients with an undetectable viral load infrequently change their anti-HIV treatment because of side-effects nowadays, according to the results of a British study published in the May 31st edition of AIDS.

The study found that some anti-HIV treatment combinations, for example those including tenofovir (Viread), efavirenz (Sustiva) or atazanavir (Reyataz) were less likely to be changed because of toxicities than others. The researchers think that in many cases, patients doing well on anti-HIV treatment will be able to stay on their combination of drugs for a long period of time.

Anti-HIV treatment significantly reduces the risk of illness and death in people with HIV. But anti-HIV drugs cannot cure HIV and it is likely that many HIV-positive individuals will need to take antiretroviral therapy for decades.

Side-effects are a major drawback of anti-HIV drugs. Studies have suggested that as many as 50% of patients starting anti-HIV treatment need to change at least one of their drugs within a year because of toxicities. Many of these side-effects occur soon after treatment has been started, and there is little information on the need to change treatment because of side-effects in patients who have achieved and maintained an undetectable viral load in the longer term.

Researchers at the Royal Free Hospital in London therefore performed a study to see the rate of treatment changes in patients starting anti-HIV treatment for the first time. All the patients in the study achieved an undetectable viral load (below 50 copies/ml) within six months of starting anti-HIV treatment and who never had an increase in their viral load to detectable levels. All the patients were taking either 3TC (lamivudine, Epivir) or FTC (emtricitabine, Emtriva) In particular, the researchers wanted to see what proportion of changes were due to side-effects.

A total of 508 patients were included in the study who contributed a total of 912 years of follow-up between 2000 and 2005. Overall, there were 357 treatment changes. The reason for changing treatment was recorded in 279 instances, and half of these changes were due to side-effects. The next most common reason for changing treatment was patient choice (18%), followed by poor adherence (4%).

The side-effects most frequently associated with a change in treatment were those of the central nervous system (33 instances, 23%), followed by lipodystrophy (28 instances, 19%). Other toxicities associated with a significant number of treatment changes included diarrhoea, anaemia, increased blood fats, and nausea and vomiting.

Statistical analysis showed that the overall rate of treatment change was 39 per 100 person years. The rate of treatment change due to side-effects was 15.4 per 100 person years.

Factors associated with an increased risk of treatment change due to side-effects were being on d4T rather than AZT (p = 0.0027), taking Kaletra rather than efavirenz (p = 0.03), being a heterosexual woman (p = 0.013) or gay man (p = 0.034) rather than a heterosexual man, and being older (every ten year increase in age, p = 0.034).

The investigators found that patients were significantly less likely to change treatment if they were taking tenofovir (p = 0.001) as their second NRTI drug, and atazanavir (p = 0.036) as their third drug.

However, the investigators think that the true rate of treatment changes due to side-effects is likely to be higher than that reported in their study. They comment, “treatment changes recorded as being due to reasons other than specific toxicities, such as patient or physician choice, may in fact be driven by toxicities.”

They conclude, “in patients who have never experienced virological failure, the rate of treatment change due to toxicities is low with certain regimens” [i.e. those containing efavirenz, tenofovir, and atazanavir] “associated with an even lower rate of change.” They add, “this suggests that so long as virological failure is avoided, some regimens are so far proving to be sufficiently stable to suggest that very long-term use is potentially feasible.”


Lodwick RK et al. Stability of antiretroviral regimens in patients with viral suppression. AIDS 22: 1039 – 1046, 2008.