Epilepsy drugs: getting it right the first time

  • Published on

  • View

  • Download


Re ection and Reaction476 http://neurology.thelancet.com Vol 6 June 2007Epilepsy drugs: getting it right the rst timefor many years to reduce motor uctuations,2 and the concept of continuous dopaminergic stimulation has lately gained wide acceptance in the treatment of PD.3In addition to its symptomatic bene ts, there are data that suggest treatment with dopamine agonists can delay or prevent the emergence of dyskinesias if started early in the course of the disease,4 and perhaps also can delay the progressive motor deterioration.3 If these arguments are correct, it might be advantageous to give these drugs in a formulation that will provide stable blood concentrations for as long as possible. This has led to the development of transdermal drugs, such as rotigotine.5,6 The article by Poewe and co-workers in the current issue adds important support to the concept that dopamine agonists can reduce the duration of o -time in advanced PD. 7 Disappointingly, rotigotine was slightly less e cacious than oral pramipexole (although technically it was non-inferior), which might be due to particular pharmacodynamic e ects of rotigotine. Although pramipexole is a speci c dopamine D2 receptor agonist, rotigotine also acts on D1 receptors. The interaction between the activation of these two receptor subtypes in PD is complex and not well understood.The present article provides important information on the e cacy of transdermal rotigotine as symptomatic therapy in advanced PD. Other agents that have been suggested for the same indication include amantadine, selegiline, rasagiline, transdermal lisuride, and entacapone, and although each of these has its own mechanism of action,8,9 their relative advantages remain to be seen.Furthermore, each agent might have a di erent adverse-events pro le; in the case of dopamine agonists, these include dopamine dysregulation syndrome and impulse control disorder.10 Although long-acting drugs such as rotigotine are theoretically less likely to cause these potentially serious side e ects, there are no published data on this issue.Amos D KorczynSieratzki Chair of Neurology, Tel-Aviv University Medical School, Ramat-Aviv 69978, Israelneuro13@post.tau.ac.ilI have acted as a consultant to Schwarzpharma.1 Korczyn AD. Pathophysiology of drug-induced dyskinesias. Neuropharmacology 1972; 11: 6017.2 Rabey JM, Nissipeanu P, Inzelberg R, Korczyn AD. Bene cial e ect of cabergoline, a new long-lasting D2 agonist in the treatment of Parkinsons disease. Clin Neuropharmacol 1994; 17: 28693.3 Olanow CW, Obeso JA, Stocchi F. Drug insight: continuous dopaminergic stimulation in the treatment of Parkinsons disease. Nat Clin Prac Neurol 2006; 2: 38292.4 Rascol O, Brooks DJ, Korczyn AD, De Deyn PP, Clarke CF, Lang AE. A ve year study of the incidence of dyskinesia in patients with early Parkinsons disease who were treated with ropinirole or levodopa. N Engl J Med 2000; 342: 148491.5 The Parkinson Study Group. A controlled trial of rotigotine monotherapy in early Parkinsons disease. Arch Neurol 2003; 60: 172128.6 Watts RL, Jankovic J, Waters C, Rajput A, Boroojerdi B, Rao J. Randomized, blind, controlled trial of transdermal rotigotine in early Parkinsons disease. Neurology 2007; 68: 27276.7 Poewe WH, Rascol O, Quinn N, et al, on behalf of the Sp515 investigators. E cacy of pramipexole and transdermal rotigotine in advanced Parkinsons disease: a double-blind, double-dummy, randomised controlled trial. Lancet Neurology 2007; 6: 51320.8 Waters CH, Kapil D, Hauser RA, Molho E, Bertoni JM. Zydis selegiline reduces o -time in Parkinsons disease patients with motor uctuations: a 3 month, randomized, placebo-controlled study. Mov Disord 2004; 19: 42632.9. The Parkinson Study Group. A randomized placebo-controlled trial of rasagiline in levodopa-treated patients with Parkinsons disease and motor uctuations. Arch Neurol 2005; 62: 24148. 10 Tippmann-Peikert M, Park JG, Boeve BF, Shepard JW, Silber MH. Pathologic gambling in patients with restless legs syndrome treated with dopminergic agonists. Neurology 2007; 68: 3013.A new diagnosis of epilepsy is a disturbing event for patients and their families, who understandably want the seizures to stop immediately. Initially choosing the drug with the best chance of success is important because drug changes are expensive, frustrating for patients, and laborious for physicians. However, there has been resistance to making the new antiepileptic drugs (AEDs) the rst choice for new-onset epilepsy because it is not clear that they are more e cacious than cheaper standard drugs.1 The Standard and New Antiepileptic Drugs (SANAD) study by Marson and colleagues, recently published in The Lancet, evaluated ve drugs for partial epilepsy and three for generalised epilepsy, on the parameters of e cacy, tolerability and cost e ectiveness.2,3 The authors concluded that lamotrigine (a new AED) is usually the best choice for partial epilepsies, and valproate (a standard AED) is the best choice for generalised epilepsies. It is refreshing to read some declarative statements on this topic after many years of ambiguity; however, the conclusions of this study must be quali ed.Most previous active-control studies were brief, of insufficient power to detect small differences Re ection and Reactionhttp://neurology.thelancet.com Vol 6 June 2007 477in efficacy, and used fixed dosing schedules. Furthermore, the majority were commercially sponsored, two-way comparisons between a new drug and carbamazepine: pharmaceutical companies have been uneager to enter their drugs in a multihorse race; therefore, in practice, these essential multidrug comparisons must be sponsored by government funding. The SANAD investigators overcame many of these obstacles. The study was sponsored by the UK National Health Service and comprised two arms: arm A aimed to enroll patients with partial-onset seizures, using carbamazepine as the standard compared with lamotrigine, oxcarbazepine, or topiramate; and arm B aimed to enroll patients with generalised-onset seizures, using valproate compared with lamotrigine or topiramate. Remarkably, 2437 patients aged 4 years and older were enrolled and most were followed for >2 years, with some followed for 6 years, making this the largest reported randomised epilepsy trial. Investigators were not blinded, and flexible titration and dosing schedules were permitted; the disadvantages of this approach are offset by the applicability of the results to the real-world setting of AED use. The maintenance doses selected were reasonable, although somewhat conservative by North American standards, particularly for children. Lamotrigine was better than carbamazepine on a primary outcome measuretime to treatment failure; however, this was because it was better tolerated, not because it was more e cacious. Several smaller, shorter studies have led to the same conclusion,4 as did a large study of new-onset seizures in the elderly.5 Therefore, this result is not surprising. There have been hints from small, retrospective studies that valproate is more e ective than lamotrigine or topiramate for generalised epilepsies;6 however, SANAD was the rst large, well-controlled study to demonstrate this convincingly. The real cost of a drug includes not only the acquisition cost but also the cost of the resources used, such as laboratory tests, and clinic and hospital visits, and the savings per seizure avoided. When these data from SANAD were analysed, the existing drugs were not necessarily cheaper: lamotrigine and oxcarbazepine were as cost-e ective as carbamazepine.The authors were careful to note several limitations of the study but two in particular deserve emphasis. First, lamotrigine and valproate are not the best rst choices in all situations and for large sections of the population: oxcarbazepine is the best choice, compared with lamotrigine or carbamazepine, when it is desirable to achieve quickly the therapeutic dose; valproate is not the best choice for women who might become pregnant;7 and topiramate is a better choice for the obese than valproate. Second, the study was not designed to address important considerations in drug choicesuch as drug interactionsother than e cacy, tolerability, and cost. Enzyme-inducing drugs, including carbamazepine, interact strongly with statins and other commonly used medications, with both biological and economic consequences. Long-term side e ects are also an issue; there is evidence that some older drugs, including valproate, are bad for bone health.8 Finally, lamotrigine and valproate are not always benign, and rare but serious adverse e ects can occur.The importance of SANAD is that we now have a new standard for comparison with future drugs for partial epilepsy: lamotrigine should replace carbamazepine for this purpose in clinical trials, whereas valproate remains the standard for generalised-onset seizures in appropriate patients. However, the quest for the ideal drug is not nished because individual patient characteristics must be taken into account and newer drugs are available. It would be regrettable if the SANAD recommendations led to more restrictions imposed by health authorities and insurance companies, rather than fewer. Ultimately, the factors in uencing drug choice must be weighed in the balance, and careful selection Alfred Pasieka/Science Photo LibraryThe printed journal includes an image merely for illustrationRe ection and Reaction478 http://neurology.thelancet.com Vol 6 June 2007must be made by the physician, in consultation with each patient.Edward FaughtUniversity of Alabama at Birmingham Birmingham, Alabama, USA faught@uab.eduI have received research grants, speakers fees, and consultancy fees from the manufacturers of all of the drugs included in SANAD: Abbott, GlaxoSmithKline, Novartis, Ortho-McNeil, P zer and Shire.1 Epilepsy (adults)newer drugs. London: National Institute for Clinical Excellence, 2004. Technology appraisal guidance TA76, National Institute for Clinical Excellence (http://www.nice.org.uk/page.aspx?o=109115).2 Marson AG, Al-Kharousi AM, Alwaidh M, et al. The SANAD study of e ectiveness of carbamazepine, gabapentin, lamotrigine, oxcarbazepine, or topiramate for treatment of partial epilepsy: an unblinded randomised controlled trial. Lancet 2007; 369: 100015.Magnesium for neuroprotection after traumatic brain injuryWe thank Maas and Murray for their thoughtful comments1 on our report detailing the outcomes of 499 patients with head injuries, randomised to either placebo or magnesium treatment.2 In the latter group, we initially used a high dose but subsequently changed to a low dose of supplemental magnesium. Maas and Murray have asked several questions in their Re ection and Reaction and, accordingly, we are responding to their inquiries.First, we would like to address their point as to what information was available to the data and safety monitoring board at the time of their decision to restart the trial. By way of background, our original intent was to treat with 100185 mmol/L magnesium but the NIH study section, which reviewed our grant in 1997, recommended that we use a higher dose. After the rst 118 patients were randomised to placebo or high-dose magnesium, the unblinded principal investigator (NRT) and colleagues noted a widening di erence in mortality, and brought this information to the attention of the blinded principal investigator (HRW). A decision was made to notify immediately the data and safety monitoring board of the morbidity and mortality for each group in addition to doing an analysis of the relevant clinical information. The medical records for each death were also available. After an on-site evaluation and additional analyses, the board concurred, as did the appropriate o cials at the NIH, with the recommendation to continue the trial at the lower, originally proposed treatment dose. Second, Maas and Murray raise questions about the original sample size: Was the target sample size 380, so the restarted trial could indeed stand alone, or was the target 500, so that a pooled primary analysis would be more natural? They also note our failure to present a prospective power calculation. Our original goal, as outlined in the initial grant application and on pages 3233 of the paper, was to randomise 400 patients, a sample size based on the pre-trial power analysis. As stated, Simulations showed that this sample size [400] gave at least 95% power to detect an increase of 10 percentage points on the dichotomised Glasgow outcome scale for severely injured participants; and similar improvement on other outcome measures and for cases with moderate severity. Furthermore, the nal sample size of 381 in the low-dose group would have essentially had the same power.Third, they question our decision to analyse the high-dose and low-dose groups separately, noting that: There was substantial overlap in the target magnesium concentrations. The decision to analyse the di erent dose levels separately was made when the study was restarted at the lower dose, and was incorporated into the protocol at that time. The analysis presented was according to protocol. Furthermore, because these two groups were not similar in their treatment, we felt that it was not appropriate to combine them. Fourth, and nally, we would urge caution in accepting their contention that the results of the current study clearly challenge the existing perception that magnesium administration is safe. We believe it would be more accurate to state that prolonged 3 Marson AG, Al-Kharousi AM, Alwaidh M, et al. The SANAD study of e ectiveness of valproate, lamotrigine, or topiramate for generalized and unclassi able epilepsy: an unblinded randomised controlled trial. Lancet 2007; 369: 101626.4 Brodie MJ, Richens A, Yuen AW, for the UK Lamotrigine/Carbamazepine Monotherapy Trial Group. Double-blind comparison of lamotrigine and carbamazepine in newly diagnosed epilepsy. Lancet 1995; 345: 47679.5 Rowan AJ, Ramsay RE, Collins JF, et al, for the VA Cooperative Study 428 Group. New onset geriatric epilepsy: a randomised study of gabapentin, lamotrigine, and topiramate. Neurology 2005; 64: 186873.6 Nicolson A, Appleton RE, Chadwick DW, Smith DF. The relationship between treatment with valproate, lamotrigine, and topiramate and the prognosis of the idiopathic generalised epilepsies. J Neuro Neurosurg Psychiatry 2004; 75: 7579.7 Meador KJ, Baker GA, Finnell RH, et al. In utero antiepileptic drug exposure: fetal death and malformations. Neurology 2006; 67: 40712.8 Sheth R. Bone health in epilepsy. Lancet Neurol 2004; 9: 516.Epilepsy drugs: getting it right the first timeReferences