On December 1, 2004, Johnson & Johnson Pharmaceutical Research & Development announced discontinuation of the current clinical development program for its controlled-release formulation of topiramate in obesity and type 2 diabetes.
Department of Human Nutrition Centre for Advanced Food Studies, The Royal Veterinary and Agricultural University, Rolighedsvej 30, DK-1958 Frederiksberg C, Denmark. E-mail: firstname.lastname@example.org
Obesity is a growing worldwide epidemic that is associated with serious medical complications. Although diet and exercise are often effective treatment in the short run, the weight lost is usually regained unless intensive intervention is maintained. Pharmacological adjunctive measures are clearly needed, and topiramate is currently being investigated as a possible new therapy for obesity. This paper reviews three clinical trials of topiramate for weight reduction in obese subjects: a 6-month dose-ranging study, a 2-year study of weight loss, and a 44-week study in subjects who had previously lost weight on a low-calorie diet. All three studies found topiramate to be significantly more efficacious than placebo. Notably, weight loss continued for 1 year and, perhaps, could have continued for a longer period. Topiramate was generally well tolerated, with adverse events being mild to moderate and mostly related to the central nervous system. Paresthesia was a frequent occurrence, but it did not lead to withdrawal of >5% of the subjects in any study.
The prevalence of obesity has risen rapidly in recent decades, not only in the United States (1), but also globally (2). Both overweight and obesity are associated with medical complications, such as type 2 diabetes (3), hypertension (4, 5), and osteoarthritis (6). Mortality from obesity-related causes such as diabetes, cardiovascular disease, and cancer increases with severity of the obesity (7). Dietary restriction in combination with behavioral modification and an exercise program is generally the preferred initial treatment for obesity and can produce a weight loss of 5% to 10% (8). Nevertheless, unless very intensive treatment is maintained, as occurred in the Diabetes Prevention Study (9), patients who lose weight through these means almost always regain much of the weight within 1 year and almost all of the weight by the end of 5 years (10). The search for pharmacological treatment for obesity has been ongoing, but currently only two medications for long-term weight loss are approved by the Food and Drug Administration (11, 12).
Among the medications being investigated for treatment of obesity is topiramate. Topiramate, a sulfamate-substituted monosaccharide, is currently approved for adjunctive treatment of partial onset seizures or generalized tonic-clonic seizures and for seizures associated with Lennox-Gastaut syndrome (13). As discussed by Shank et al. (14), topiramate's mechanism of action as an anticonvulsant may include blockade of voltage-activated Na+ channels, enhancement of γ-aminobutyric acid-evoked currents, inhibition of kainate-evoked currents, inhibition of high-voltage—activated Ca2+ channels, and inhibition of carbonic anhydrase. All of these effects may result from a postulated ability of topiramate to bind to dephosphorylated regulatory subunits of the various receptors and channels, thus, both exerting direct allosteric effects and blocking subunit rephosphorylation.
Data from studies of topiramate conducted in several animal models have suggested its efficacy as a weight loss agent (15). These observations led to clinical trials of topiramate for treatment of obesity. To date, three randomized, controlled studies of weight loss in obese subjects have been published: a 6-month dose-ranging study (16), a 2-year weight loss trial (17), and a 1-year trial weight maintenance study in obese subjects who lost weight by consuming a low-calorie diet for 8 weeks (18). In each study, all subjects participated in a program of behavioral modification called Pathways to Change, which was designed to encourage weight loss through education in the areas of diet, nutrition, physical activity, psychosocial structuring, and support (16). Subjects were prescribed an individualized diet that was ∼600 kcal/d less than their anticipated caloric requirements for weight maintenance. Study design and inclusion and exclusion criteria were similar between studies, except for differences reflecting their specific goals. However, before completion of the two longer term studies (17, 18), the topiramate immediate-release clinical research program in obesity and diabetes was discontinued to develop a new controlled-release formulation that could enhance tolerability and simplify dosing. Accordingly, efficacy results are reported for modified intent-to-treat populations, consisting of subjects who enrolled early enough to have the opportunity to complete at least 60 (17) and 44 (18) weeks of medication therapy.
Six-Month Dose-Ranging Study
Bray et al. (16) enrolled 385 subjects at 17 centers in the United States. Subjects had a BMI ≥ 30 and <50 kg/m2 (or ≥27 to 50 kg/m2 in subjects with controlled hypertension and/or dyslipidemia). Subjects were excluded if they had uncontrolled hypertension, diabetes, liver disease, renal dysfunction, and/or cardiovascular, endocrine, neurological, or psychiatric disease.
Subjects were randomized to receive topiramate 64, 96, 192, or 384 mg/d, or placebo, administered in divided twice-daily doses (Figure 1). Dosing was titrated upward, beginning with 16 mg/d for the 1st week, increasing to 32 mg/d for the 2nd week, and then rising weekly by 32 mg/d until the target dose was reached; the titration phase was considered to last for 12 weeks regardless of the actual time required to reach the target dose. The titration phase was followed by a 12-week maintenance phase and a 2-week taper and evaluation phase.
In the intent-to-treat group with last observation carried forward analysis, mean weight loss at 24 weeks was 2.6% of initial body weight in the placebo group and 5.0%, 4.8%, 6.3%, and 6.3% in the four topiramate groups (64, 96, 192, and 384 mg/d, respectively). Weight loss for all topiramate groups was significantly greater than for placebo (p < 0.05) at all time-points after week 4 of the titration period. Weight loss with the two higher doses of topiramate was similar and greater than weight loss with the two lower doses. Compared with the placebo group, more subjects in each topiramate treatment group lost at least 5% of their initial weight (Figure 2). Weight loss continued throughout the study and had not yet reached a plateau at week 24. Among the completers, notable weight loss was also seen by week 4, and this continued to decline throughout the 24-week observation period (Figure 3).
Although all subjects were required to have normal blood pressure at baseline, all topiramate groups experienced significant decreases in systolic blood pressure compared with placebo (1.2 mm Hg for the placebo group and 5.8, 7.5, 7.1, and 4.5 mm Hg for topiramate 64, 96, 192, and 384 mg/d, respectively). Significant decreases in diastolic blood pressure were also seen for all topiramate groups except at the 384 mg/d dose (1.6 mm Hg for placebo and 3.2, 4.6, 4.3, and 2.0 mm Hg, respectively, for the topiramate groups). There were no significant differences between groups in glucose, insulin, glycosylated hemoglobin, or fasting lipids in this basically healthy population.
Long-Term Weight Loss Study
The design of this study is shown in Figure 4. Subjects completed a 6-week single-blind placebo run-in phase, with the Pathways to Change behavioral modification program as active therapy. Subjects who lost at least 6% of their original body weight during the placebo run-in phase were excluded from the study. Subjects who remained eligible were randomized to placebo or to 96, 192, or 256 mg/d topiramate. Premature termination of the study led to efficacy results being reported for a modified intent-to-treat population who enrolled early enough to have potentially completed at least 60 weeks of treatment (8 weeks of titration and 52 weeks of maintenance). Of the 1289 subjects randomized, 854 met the criteria for the modified intent-to-treat population.
Subjects randomized to placebo therapy lost 1.7% of their initial body weight, whereas losses for the three topiramate groups were 7%, 9.1%, and 9.7%, respectively (Figure 5). In the placebo and three topiramate-treated groups, the percentages of subjects losing at least 5% of initial body weight were 18%, 54%, 61%, and 67%, respectively (difference between each topiramate-treated group and placebo, p < 0.05). The percentages of subjects who lost at least 10% of initial body weight were 6%, 29%, 40%, and 44%, respectively (difference between each topiramate-treated group and placebo, p < 0.05). Subjects treated with topiramate continued losing weight throughout the study period, and weight loss continued after 1 year in the small number of subjects for whom data were available (Figure 6). In contrast, weight loss in subjects receiving placebo occurred up to week 32, then reached a plateau, followed by an upward trend after 1 year.
In placebo-treated subjects, systolic blood pressure increased by 0.41 mm Hg and diastolic blood pressure by 1.04 mm Hg. In the three topiramate-treated groups, systolic blood pressure decreased by 3.1, 5.7, and 4.6 mm Hg, respectively, and diastolic blood pressure by 1.3, 3.4, and 2.4 mm Hg, respectively. Differences between each topiramate group and the placebo group were statistically significant (p < 0.001). There were no significant changes in blood lipids in any group.
Of the 145 subjects (with available information at baseline) who had impaired glucose tolerance or type 2 diabetes at baseline and received at least one dose of topiramate, glucose tolerance was normal by 60 weeks in 72 (50%). Among subjects who received at least one dose of placebo, 19 of 48 (40%) had normal glucose tolerance by 60 weeks.
Long-Term Weight Loss Maintenance
The design of this study is shown in Figure 7. Subjects who had at least an 8% weight loss during the 8-week low-calorie (800 to 1000 kcal/d) diet run-in period were randomized to therapy with placebo, 96 mg/d topiramate, or 192 mg/d topiramate. Minimum BMI was 34 kg/m2 at study entry and 30 kg/m2 at randomization after dietary weight loss (30 and 27 kg/m2 in the presence of controlled hypertension and/or dyslipidemia). The Pathways to Change behavioral modification program was initiated at randomization, along with a diet 600 kcal/d below estimated weight maintenance needs. The study was originally scheduled to include an 8-week titration phase and a 52-week maintenance phase, but due to premature termination, results were reported for a modified intent-to-treat population consisting of subjects who enrolled early enough to have the opportunity to complete at least 36 weeks of the maintenance phase. Of the 701 subjects initially enrolled, 561 met the dietary weight loss criterion and were randomized; of these, 293 were randomized early enough to be included in the modified intent-to-treat population.
During the 8-week dietary run-in period, all groups lost between 10% and 11% of their initial body weight. During the 44 weeks after randomization and initiation of study medication, modified intent-to-treat subjects who received placebo regained 1.8% of their initial body weight, whereas those given 96 mg/d or 192 mg/d topiramate lost an additional 5.2% and 6.4%, respectively, of their initial body weight (Figure 8). More topiramate-treated than placebo-treated subjects maintained 50% to 74%, 75%, or 100% of their initial dietary weight loss through week 44 (Figure 9).
All groups experienced improvements in preenrollment cardiovascular and metabolic risk factors, although the largest improvement occurred during the low-calorie weight loss phase from enrollment through week 44. The decreases in systolic blood pressure were 6.4 mm Hg in the placebo group, 7.4 mm Hg in the 96 mg/d topiramate group (p = 0.014 vs. placebo), and 10.2 mm Hg in the 192 mg/d topiramate group (p < 0.001 vs. placebo). The decreases in diastolic blood pressure were 2.8, 3.5 (both NS vs. placebo), and 4.5 (p < 0.05 vs. placebo) mm Hg, respectively. Improvements in lipid profile were modest, but for triglycerides and total cholesterol, improvements were significantly greater in each topiramate group than in the placebo group.
Safety and Tolerability Results
In all studies, topiramate was generally well tolerated, especially at doses of 192 mg/d or less. Most of the adverse events were mild to moderate in severity and related to the central nervous system (CNS).1 Most prominent of these was paresthesia, which was experienced by 42% of the topiramate-treated subjects in the short-term dose-ranging study (16), 57% of those in the long-term weight loss study (17), and 59% of those in the long-term weight loss maintenance study (18). Subjects generally found the paresthesia to be tolerable, and only 4% (16), 5% (17), and 3% (18) of the subjects withdrew from the studies because of paresthesias.
Other common adverse events were also CNS-related: fatigue, difficulty with concentration or attention, depression, and difficulty with memory. In the long-term weight loss study of Wilding et al. (17), which was typical of the other studies, these adverse events were seen in 23%, 11%, 11%, and 10%, respectively, of topiramate-treated subjects compared with 20%, 5%, 3%, 8%, and 7% of placebo-treated subjects.
In the longer term studies (17, 18), most subjects withdrew due to the trials’ premature termination. Only six subjects completed the weight loss study, all because they had achieved the maximum permitted weight loss (17). However, there were 484 subjects who completed 1 year of the planned 2 years (103 placebo, 133 96 mg/d topiramate, 123 192 mg/d topiramate, and 125 256 mg/d topiramate). In addition, 89 subjects (28 placebo, 31 96 mg/d topiramate, 30 192 mg/d topiramate) completed the scheduled 60 weeks of the weight loss maintenance study (18).
In the dose-ranging study (16), 21% of subjects on any topiramate dose, compared with 11% of those on placebo, withdrew due to an adverse event. In the long-term weight loss study (17), the rate of withdrawal due to adverse events was 11% for placebo, 18% for 96 mg/d topiramate, 24% for 192 mg/d topiramate, and 22% for 256 mg/d topiramate. In the weight loss maintenance study (18), 8% of placebo subjects, 13% of those on 96 mg/d topiramate, and 20% of those on 192 mg/d topiramate withdrew due to an adverse event.
In addition to paresthesia, the most common adverse events leading to withdrawal were difficulty with memory, difficulty with concentration or attention, mood problems, fatigue, depression, and dizziness (16, 17, 18). These events resolved after withdrawal of treatment.
These three studies (16, 17, 18) demonstrate the weight loss efficacy of topiramate as an adjunct to diet and behavioral modification therapy. The weight loss effect is observed quickly; Bray et al. (16) found statistically significant differences between placebo and every topiramate dose starting at week 4. The weight loss effect persists; Astrup et al. (18) found weight loss continued through week 44, whereas Wilding et al. (17) saw weight loss continue for at least 1 year with indications that it may continue beyond 1 year. In addition, a consistently greater proportion of topiramate-treated subjects than placebo-treated subjects lost at least 5%, and 10% of their initial body weight.
These studies also demonstrate efficacy of the Pathways to Change behavioral modification program. In the Bray et al. (16) and Wilding et al. (17) studies, subjects lost weight even when they were randomized to placebo, whereas in the weight loss maintenance study of Astrup et al. (18), placebo subjects regained only a small fraction of the weight they had lost on a strict low-calorie diet. The efficacy of topiramate appeared to peak at a dose of ∼192 mg/d. Bray et al. (16) found identical weight loss at 192 and 384 mg/d, with a smaller proportion of subjects with 5% and 10% weight loss at the higher dose. Wilding et al. (17) reported more responders and slightly greater weight loss with 256 mg/d topiramate than with the 192 mg/d dose, but the difference in weight loss was not statistically significant. In the weight loss maintenance study of Astrup et al. (18), 192 mg/d topiramate showed fairly modest superiority to the 96 mg/d dose. The frequency of adverse events was generally similar at all doses from 96 to 384 mg/d. Some trials reported dose-related increases in frequency of paresthesia, memory difficulties, and somnolence, but these relationships were not consistent across the trials (16, 17, 18).
Although head-to-head comparisons with orlistat and sibutramine are not available, comparison with generally similar studies of these two medications are possible. Finer et al. (19) and Wirth and Krause (20) have carried out 1-year weight loss studies comparable with that of Wilding et al. (17), whereas Hill et al. (21) and Apfelbaum et al. (12) have carried out 1-year studies of weight loss maintenance comparable with that of Astrup et al. (18).
Finer et al. (19) randomized subjects to either placebo or 120 mg orlistat three times a day, together with a diet estimated to be 600 kcal/d below maintenance needs. At the end of 1 year, subjects on orlistat had lost 8.5% of their initial body weight compared with 5.4% for subjects on placebo. A 5% response occurred in 35% and 21% of subjects, respectively, whereas a 10% response occurred in 28% and 17%, respectively; differences were statistically significant for both responses. There were also significant decreases in total cholesterol, low-density lipoprotein (LDL)-cholesterol, and the LDL-to-high-density lipoprotein-cholesterol ratio for both responses.
The study by Wirth and Krause (20) used a somewhat different design. Sibutramine (15 mg/d) was administered open-label for 4 weeks; only subjects who lost at least 2 kg or 2% of their initial body weight during this period were eligible to continue. They were then randomized either to placebo, continuous treatment with 15 mg/d sibutramine, or alternating treatment with 15 mg/d sibutramine for 1 to 12, 19 to 30, and 37 to 48 weeks and placebo for the intervening weeks, followed by placebo for 6 weeks. There was no formal dietary intervention.
Continuous treatment proved somewhat more efficacious than alternating treatment, although differences were not statistically significant (20). During the period of randomized treatment, subjects on continuous sibutramine lost 4.0% of their initial weight, whereas those on placebo gained 0.2%. A 5% weight loss was seen in 65% of subjects on continuous sibutramine compared with 35% of those on placebo. For 10% responders, the corresponding figures were 32% and 13%. There was a significant decline in triglycerides levels among sibutramine-treated subjects but, despite the weight loss, no change in blood pressure. When these efficacy results were compared with those of Wilding et al. (17), all topiramate doses were found to produce greater weight loss compared with placebo than did maximal doses of orlistat or sibutramine, with weight losses for 192 mg/d topiramate approaching twice the losses observed with either of the other medications. Particularly notable is the superiority of topiramate over sibutramine, even though in the study by Wirth and Krause (20), a bias was introduced in the results by including only subjects who had responded to a brief open-label trial of sibutramine.
The design of the study by Hill et al. (21) was similar to that of Astrup et al. (18), except that the initial weight loss of 8% or greater was achieved through a 6-month diet at an estimated caloric deficit of 4180 kJ/d. Subjects were then assigned to a maintenance diet plus either placebo or orlistat at a dose of 30, 60, or 120 mg three times a day (21). At the end of 1 year, only the 120-mg dose had proven statistically superior to placebo, with subjects in that group regaining 32.8% of the weight initially lost compared with 56% in the placebo group. The authors noted that subjects in all orlistat groups had significantly greater improvements in total and LDL-cholesterol levels than did those given placebo. Changes in metabolic risk factors and blood pressure were not notable or consistent.
In the study by Apfelbaum et al. (12), the randomization criterion was weight loss of at least 6 kg on a 4-week very low-calorie (220 to 800 kcal/d) diet. Subjects were randomized to either sibutramine 10 mg/d or placebo, along with dietary counseling intended to reduce their caloric intake to ∼20% to 30% below preenrollment levels. From randomization to the end of the trial, sibutramine-treated subjects lost 5.2 kg, whereas those assigned to placebo gained 0.5 kg. With a mean weight at screening of ∼104 kg, this corresponds to weight changes of −5.0% and +0.5%, respectively. Differences between sibutramine and placebo were significant at all time-points.
The percentages of subjects who retained at least 100%, 50%, and 25% of their run-in weight loss were 75%, 95%, and 96%, respectively, for sibutramine-treated subjects and 42%, 73%, and 80%, respectively, for subjects given placebo. All differences were statistically significant. Subjects on sibutramine had improvements in triglycerides and high-density lipoprotein-cholesterol that were significantly better than those of subjects on placebo. At 6 months, diastolic blood pressure was significantly higher in sibutramine-treated subjects than placebo-treated subjects (change of +1.5 vs. −1.9 mm Hg), but the difference was no longer significant at 12 months. In addition, pulse rate was significantly higher with sibutramine than with placebo at all time-points.
All three medications were generally well tolerated. Adverse events with topiramate were generally CNS-related, with paresthesia—a known effect of carbonic anhydrase inhibition—being especially prominent. Orlistat acts as a gastric lipase inhibitor; thus, it reduces absorption of dietary fat. As expected, most of the adverse events seen with this medication were gastrointestinal. In the study by Finer et al. (19), gastrointestinal adverse events were 26% more common in orlistat-treated subjects than in those receiving placebo. Hill et al. (21) likewise reported that gastrointestinal adverse events were seen in 27% more subjects receiving 120 mg/d orlistat than those receiving placebo (21). Furthermore, fat malabsorption can lead to loss of fat-soluble vitamins. Both Finer et al. (19) and Hill et al. (21) reported significantly greater decreases in serum vitamin E and β-carotene with orlistat than with placebo.
Adverse effects with sibutramine were generally similar to those with placebo (20). The major unwanted effect with sibutramine was a rise in blood pressure or failure to see the drop expected as a result of weight loss (12, 20). Because obese individuals are often hypertensive—and, indeed, weight reduction is often initiated to decrease blood pressure and its associated cardiovascular risks (22)—this effect can be a significant drawback in many instances.
Administration of topiramate in conjunction with diet and a behavioral modification program is effective for weight reduction in obese persons. Furthermore, weight reduction continues for at least 1 year, which is longer than that reported for other weight loss medications. Topiramate is generally well tolerated. Although paresthesia is common, it is generally mild to moderate and withdrawal due to its occurrence is infrequent.
Nonstandard abbreviations: CNS, central nervous system; LDL, low-density lipoprotein.