July 15, 2010

Testimony on Phentermine and Topiramate (Qnexa)

Testimony before FDA’s Endocrinologic and Metabolic Drugs Advisory Committee
Sidney Wolfe, M.D.
Charles Preston, M.D., M.P.H.
Public Citizen’s Health Research Group 

Introduction

Your decision today will greatly influence whether the FDA approves a combination of two old drugs, each of which has significant risks:  an amphetamine and an anti-seizure medication – phentermine and topiramate – for the treatment of obesity. Given the literally insatiable appetite of doctors and patients for new drugs as a quick fix for obesity, there is every reason to believe that, if approved, a combination like this, will be used by millions, over long periods of time far beyond its labeling indications. Because of a long list of safety reasons, this drug should not be approved.

Pharmacology

Few drugs act on as many different sites as topiramate, including GABA receptors, on voltage gated ion channels, on aquaporins (proteins regulating the flow of water), and on carbonic anhydrases.[1] That explains the therapeutic effects for seizures and migraines as well as some of the adverse effects. A few examples include: acute myopia secondary to angle closure glaucoma, cognitive impairment, kidney stones, metabolic acidosis, oligohydrosis and hyperthermia, suicidal ideation, and teratogenicity.

Phentermine acts on adrenergic receptors, with corresponding effects on many different organ systems, including the neurologic, cardiovascular, and gastrointestinal. It increases heart rate and blood pressure, heightens wakefulness, and decreases appetite.

With such a combination of highly active agents, unintended effects are guaranteed. Is thus acceptable?

Off Label Use

Vivus is seeking drug approval for the treatment of obesity, with specific recommendations that Qnexa be used for patients with a BMI greater than 30 or a BMI greater than 27 with related co-morbidities. But does anyone think for a second that this drug won’t make its way into the hands of the high school athlete who wants to trim down before game time or into the hands of a young woman, not really obese, who is worried about her weight? For the athlete, there’s a risk of oligohydrosis (decreased sweating) and hyperthermia,[2] for the young woman, there’s a risk of depression and suicide. We need to know how these patients would fare under Qnexa. But as the phase 3 studies show, the majority of participants were obese, middle age, Caucasian women.[3] We need data that’s more generalizable to the likely target population.

Lack of Long-Term Safety Data

Vivus only conducted trials lasting one year. This is unacceptable. More than other drugs, diet drugs, with their potential for long term use, should be accompanied by long-term safety data of at least 2-5 years. We know that some patients may  take diet drugs for a long time, because when they stop, they usually suffer recurrent weight gain.

Safety Risks

As would be expected of an amphetamine and anti-seizure combination like Qnexa, there is a long list of serious side effects. And because of these side effects, adverse events were the major reason patients failed to complete the trials:  18% of high dose subjects withdrew due to an adverse event, compared to 9% for placebo.[4]

Metabolic Acidosis

One of the most concerning side-effects of the numerous concerning side-effects is metabolic acidosis. We know why this happens – topiramate is a carbonic anhydrase inhibitor. It causes the kidney to lose bicarbonate. Consequently, patients treated with Qnexa experienced a higher percentage of events where serum levels dropped below the normal value of 21, either on two consecutive occasions or as an endpoint.[5] High dose patients experienced this 12.8% of the time versus 2.1% for placebo.[6] And what if the patient also had diarrhea, abused laxatives, or suffered chronic kidney disease? The acidosis would get much worse.

The long-term sequelae of acidosis are very dangerous. There is increased risk of arrhythmia, osteoporosis, and nephrolithiasis (kidney stones), to name a few serious adverse effects. In fact, the Qnexa data show 22 cases of nephrolithiasis in the treatment group compared to 5 in placebo.[7] By pooling the phase 3 treated subjects, we know that this amounts to nearly 1% of treated patients (22/2201).[8] If this drug reaches one million people (it could reach millions more), 10,000 people would get kidney stones. Is that acceptable?

Cognitive Effects

It is clear from previous topiramate studies that cognition is impaired in people who take the drug. This occurs in a number of domains, including attention, memory, and language. The same is seen with Qnexa. It was hypothesized that by adding an amphetamine, the negative cognitive effects of topiramate might be mitigated. But we see just the opposite – that the combination is worse than the topiramate alone. In a summary of treatment emergent adverse events, 36% of patients on high dose topiramate suffered nervous system disorders compared to 49% for high dose combination therapy.[9] The same is true for mid dose topiramate versus mid dose combination therapy, 39% to 41%, respectively.[10]

But what these cognitive deficits actually mean is unclear. And so the briefing document includes a breakdown of symptoms using a test called the Repeatability Battery for the Assessment of Neuropsychological Status (RBANS), which quantifies the various aspects of cognition. The test was performed at baseline, at weeks 4 and 28, or at early termination. Testing showed a highly concerning decrement in attention between placebo and mid and high dose combination at 4 and 28 weeks.[11] That decrement was considered clinically significant.[12] In addition, there were 8 patients in the high dose combination group compared to 2 in placebo whose scores on this test varied by more than 1.5 standard deviations from baseline.[13] This is important because millions of people will be using this drug. They drive. They go to school. They work. The cognitive effects, although acceptable for people who have seizure disorders, are too powerful for mainstream use as a treatment for obesity.

Psychiatric Effects

Anti-epileptic drugs have long been known to increase the risk of suicidal ideation relative to placebo. In a meta-analysis of placebo controlled trials conducted by the FDA, topiramate was shown to have a statistically significant 2.53 times greater odds of suicidal ideation than placebo (in fact, this was the highest point estimate of the statistically significant estimates of the 10 other AEDs FDA analyzed).[14]

This risk of suicidality is especially concerning because Qnexa is being proposed for use in the obese – a population known to have higher rates of depression. And in fact, the rate of depression in the high dose treatment was more than double placebo, 7.7% vs 3.4%, respectively.[15] The high dose combination group discontinued the study because of anxiety, sleep, and depression, 4 to 7 times more commonly than those in placebo group.[16]

Teratogenicity

The study populations of OB-302 and OB-303 heavily favored women, 83% and 70%, respectively. Combined with the age distribution, this indicates a large proportion of people who will use this drug will be women of child bearing age.[17] Therefore, we strongly agree with FDA’s conclusion that there is a high likelihood of exposed pregnancies to this drug.[18] This is of particular concern because of a “repeated pattern of craniofacial congential malformations in animal studies, the UK pregnancy registry, the North American AED pregnancy registry, and the FDA AERS database”.[19]

At doses just twice as much as the high dose Qnexa combination therapy (200 mg/day), the UK Epilepsy and Pregnancy registry suggested a 4.8% malformation rate.[20] The North American Antiepileptic Pregnancy registry showed topiramate exposed pregnancies to have a higher prevalence of malformations than controls, 4.1% to 1.6%.[21] In the FDA’s adverse reporting system, 64 analyzable reports of monotherapy exposure were found, and of these, nearly 33% had craniofacial defects and 30% had skeletal abnormalities (when information on timing and dose were known, the majority reported use of 200 mg/day or less (29/45)).[22] Although there were no birth defects reported in the Qnexa trials, 13 pregnancies in the treatment group (out of 34 reported in total) were carried to term despite an aggressive pregnancy prevention protocol.[23]

Phentermine Effects

Phentermine also has some serious adverse effects. Because of its mechanism of action as an adrenergic agonist, it is contraindicated in patients with advanced arteriosclerosis and cardiovascular disease.

The adrenergic effects are concerning because in populations with cardiovascular disease, we don’t want to stress the heart unnecessarily. But because relatively few people with a history of MI were enrolled, we don’t know how this medication will affect them.[24] Although the rates of ischemic cardiovascular events were similar in treatment and placebo groups, there were 4 MIs in the treatment group and 0 in the placebo group.[25] It is not surprising then that Vivus has volunteered to conduct a large cardiovascular outcomes trial if Qnexa is approved.[26]

The cardiovascular effects that were seen were alarming. In high dose Qnexa therapy, for example, heart rate increased by 15 beats per minute above placebo.[27] Arrhythmia occurred at an incidence of 4.7% compared to 1.8% for placebo.[28]

Conclusion

In sum, this is not a novel therapy. This is simply a repackaging of two old drugs, each of which has substantial dangers. For many reasons, including the risks of off-label use, the lack of long-term safety data, the concerns related to metabolic acidosis, cognition, mood and suicidal ideation, teratogenicity, and cardiovascular effects, this drug should not be approved.

Perhaps one of the most important prior studies relevant to today’s decision was a randomized, placebo controlled trial of topiramate controlled release (CR) for overweight and obese patients with type 2 diabetes. The treatment group underwent 16 weeks of therapy, up to a dose of 175 mg/day. The investigators found that the cognitive and psychiatric effects were similar to those seen with Qnexa namely higher rates of anxiety, memory difficulties, and insomnia compared to placebo – and they concluded that the “CNS and psychiatric adverse event profile of topiramate CR makes it unsuitable for the treatment of obesity and diabetes.”[29] We couldn’t agree more strongly.

---

[1] Shank RP, Maryanoff BE. Molecular pharmacodynamics, clinical therapeutics, and pharmacokinetics of topiramate. CNS Neurosci Ther. 2008 Summer;14(2):120-42. Review.