Older anti-epileptic drugs associated with failure of HIV treatment

An interaction means that some older anti-epileptic drugs and antiretroviral therapy should not be used together, a US study published in the open-access journal AIDS Research and Therapy shows.

Both types of therapy are metabolised by the body using the P450 pathway in the liver, reducing levels of anti-HIV drugs and leading to sub-optimal control of the virus.

Although the theoretical risk of a drug interaction is known from pharmacokinetic studies in healthy volunteers, this study provides strong clinical evidence that older anti-epileptic drugs should not be used in patients taking antiretroviral therapy.

Most of the anti-epileptics used in resource-limited settings use the P450 pathway, and the investigators believe the ramifications of the interaction with HIV therapy “may be substantial.”

HIV-positive individuals use anti-epileptic drugs widely. In addition to controlling seizures, they are also a therapy for neuropathy, depression and bipolar mood disorders.

Many first-generation anti-epileptic drugs such as phenytoin, carbamazepine and phenobarbital are metabolised by the P450 enzyme. The same pathway is also used by antiretroviral drugs in the protease inhibitor and non-nucleoside reverse transcriptase (NNRTI) classes as well as the CCR5 inhibitor maraviroc.

Because these medications are metabolised in a similar way there is a potential for interactions. These could lead to reduced blood levels of antiretroviral drugs and therefore inadequate control of HIV.

Investigators therefore retrospectively compared virologic outcomes in patients taking antiretroviral therapy with P450-inducing anti-epileptics with two control populations.

The first consisted of patients who were taking HIV treatment and anti-epileptics that did not use the P450 pathway; the second comprised patients who were only taking HIV therapy.

Virologic failure was defined as lack of viral suppression to below 400 copies/ml six months after the start of therapy, or a sustained rebound to above 400 copies/ml after a period of undetectable viral load.

A total of 19 patients were treated with concurrent HIV therapy and anti-epileptic drugs that used the P450 pathway.

These patients were significantly more likely than the 85 individuals taking anti-epileptics that did not use this pathway to experience virologic failure (63% vs. 27%; p = 0.009).

After both six and twelve months of therapy, patients taking the older anti-epilepsy drugs were less likely to have an undetectable viral load than the patients taking non-P450-inducing epilepsy therapy (33% vs. 71%, p = 0.016; 36% vs. 75%, p = 0.018).

The risk of treatment failure was increased four-fold for patients taking P450-using anti-epileptics compared to patients taking epilepsy drugs that did not use this pathway (OR = 4.19; 95% CI, 1.54-11.44, p = 0.005).

Rates of virologic failure were also higher among the patients taking older epilepsy therapy when compared to individuals not receiving anti-epileptics (63% vs. 43%). Differences in outcomes between these two groups were significant when the investigators took into account baseline viral load (p = 0.046).

“This study is the first demonstrating clinically meaningful outcomes in participants receiving overlapping treatment with [P450-inducing anti-epileptic drugs] and HAART,” comment the investigators.

They emphasise, “the impact is so robust, that we were able to demonstrate this despite the small numbers of patients receiving [P450-inducing anti-epileptic drugs.]”

The researchers recommend that wherever possible HIV-positive patients should receive anti-epileptics that do not use the P450 pathway.

However, in many resource-limited settings the only available epilepsy drugs are older therapies with a risk of interactions.

In these circumstances the investigators recommend closer clinical monitoring, and if possible the use of HIV therapy that does not involve a risk of interactions.

They note that one antiretroviral option would be raltegravir. But they acknowledge that its price means that it is not readily available in poorer countries. Another possibly offered by the researchers is triple NRTI therapy, but the poor virological outcomes associated with such regimens are also noted.

Okulicz JF et al. Virologic outcomes of HAART with concurrent use of cytochrome P450-inducing antiepileptics: a retrospective case control study. AIDS Research and Therapy, 8: 18, 2011 (click here for the open access paper).