European Journal of Clinical Pharmacology

, 63:123

The effect of topiramate on energy balance in obese men: a 6-month double-blind randomized placebo-controlled study with a 6-month open-label extension

Authors

    • Division of Kinesiology (PEPS), Department of Social and Preventive MedicineLaval University
  • Jean-Philippe Chaput
    • Division of Kinesiology (PEPS), Department of Social and Preventive MedicineLaval University
  • Sonia Bérubé-Parent
    • Division of Kinesiology (PEPS), Department of Social and Preventive MedicineLaval University
  • Denis Prud’homme
    • School of Human KineticsUniversity of Ottawa
  • Claude Leblanc
    • Division of Kinesiology (PEPS), Department of Social and Preventive MedicineLaval University
  • Natalie Alméras
    • Department of Food Sciences and NutritionLaval University
  • Jean-Pierre Després
    • Lipid Research CenterCHUL Research Center
    • Quebec Heart Institute
Clinical Trials

DOI: 10.1007/s00228-006-0220-1

Cite this article as:
Tremblay, A., Chaput, J., Bérubé-Parent, S. et al. Eur J Clin Pharmacol (2007) 63: 123. doi:10.1007/s00228-006-0220-1
  • 119 Views

Abstract

Objective

Topiramate (TPM) has been reported to reduce body weight beyond a placebo in the treatment of obese participants, but the effect of this agent on components of energy balance has not yet been established in humans. Thus, the aim of this study was to study the impact of TPM on food preferences, measures of satiety, food intake, resting metabolic rate (RMR), and 24-h energy expenditure.

Methods

The study design consisted of a 6-month, single-center, randomized, double-blind, parallel group, placebo-controlled trial with a 6-month open-label extension. The study included 68 sedentary men with abdominal obesity (waist circumference ≥100 cm), of between 25 and 55 years of age, with a dyslipidemic profile and a body mass index (BMI) ≥27 and ≤40 kg/m2.

Results

Treatment with TPM produced significant changes in anthropometric variables and body composition compared with placebo. However, at the end of the 1-year study, the placebo/TPM group showed similar weight loss and reduction in body fatness compared with the TPM/TPM group. For instance, at the end of the 12-month intervention, mean percentage of body weight loss from baseline was about −5% in both groups (−4 kg fat loss). Topiramate treatment reduced energy intake, be it in the context of an ad libitum buffet-type meal or under free living conditions. The 24-h daily energy expenditure (DEE) assessed by whole-body indirect calorimetry adjusted for body weight and age was not altered by TPM treatment.

Conclusion

Topiramate treatment produced significantly greater weight loss than placebo and the majority of this loss was explained by a decrease in body fat stores. Most of the weight loss effect produced by TPM therapy was observed within a period of 6 months. Finally, TPM treatment had an impact on energy balance through a reduction in food intake that appears to have created an energy deficit of about 30,000–40,000 kcal compared with treatment with the placebo over 6 months.

Keywords

AppetiteBody compositionEnergy balanceFood preferencesTopiramate

Introduction

Currently, sibutramine [1] and orlistat [2] are approved for long-term use in the treatment of obesity. However, as obesity is a heterogenous condition, there is a need to develop novel agents with new mechanisms of action and with an acceptable side-effect profile. In this regard, topiramate (TPM), an agent currently approved for the treatment of epilepsy, has been reported to induce significant weight loss as a “side effect”.

Although the effect of TPM on body weight is documented, the mechanism through which this drug influences energy balance is not clear. In a variety of animal models [35], TPM has been shown to reduce appetite and to interfere with the efficiency of energy utilization. This effect on efficiency of energy utilization may reflect TPM’s ability to stimulate lipoprotein lipase activity in brown adipose tissue and skeletal muscle [5, 6], thus potentially increasing thermogenesis and substrate oxidation. TPM also increases expression of uncoupling proteins 2 and 3 in adipose tissue and skeletal muscle, thus directly decreasing efficiency of energy utilization [4].

To date, three randomized, controlled studies of weight loss by TPM in obese individuals have been published: a 6-month dose-ranging study [7], a 2-year weight loss trial [8], and a 1-year trial weight maintenance study in obese participants who lost weight after having been treated with a low-calorie diet for 8 weeks [9]. These three studies all demonstrated the efficacy of TPM in inducing weight loss as an adjunct to diet and behavioral modification therapy. Notably, weight loss appeared to continue for 1 year and, perhaps, could have continued for a longer period. TPM was generally well tolerated, with adverse events being mild to moderate and mostly related to the central nervous system. Paresthesia has been reported, but this side effect did not lead to withdrawal of >5% of the participants in any study.

However, the impact of this agent on components of energy balance (energy expenditure and food intake) has not yet been established in humans. The aim of the present study was therefore to determine the impact of TPM on food preferences, measures of satiety, food intake, resting metabolic rate (RMR), and 24-h energy expenditure in order to evaluate whether variations in body fat indices produced by TPM treatment were due to a decrease in food intake/appetite and/or an increase in energy expenditure. The tolerability of the compound and adverse experiences were also extensively investigated in the present study, but only a safety overview will be given here.

Materials and methods

Subject eligibility

Non-smoking, sedentary men with abdominal obesity (waist circumference ≥100 cm), between 25 and 55 years of age, with dyslipidemic profile of fasting triglyceride levels ≥1.90 and <4.50 mmol/l with corresponding HDL-cholesterol concentration <1.20 mmol/l or fasting triglyceride levels ≥1.50 mmol/l and <1.90 mmol/l with corresponding HDL-cholesterol ≤0.90 mmol/l, and with a body mass index (BMI) ≥27 and ≤40 kg/m2 were selected to participate in this study through advertisements in a local newspaper. In addition, all participants were weight stable (varying no more than 4 kg) for at least 2 months prior to enrollment. Exclusion criteria included endocrine diseases, abnormal hepatic liver function tests or renal disease, history of schizophrenia, psychotic or major affective disorder, epilepsy, history of eating disorders, significant cardiovascular disease, diabetes, hypertension, medication that could interfere with the mechanism of action or metabolism of TPM within 3 months prior to enrollment, medication known to alter blood pressure, plasma lipids, plasma glucose and/or insulin within 28 days before enrollment, severe systemic illness, heparin allergies, problem of alcohol abuse, and history of nephrolithiasis. All participants gave their written consent to participate in this study, which received approval of the Laval University Ethics Committee.

Study design

The study was a 6-month, single-center, randomized, double-blind, parallel group, placebo-controlled trial with a 6 month open-label extension (Fig. 1). A subgroup of participants was randomized to participate in a 24-h energy expenditure substudy. A placebo control was used to establish the frequency and magnitude of changes in clinical endpoints that may occur in the absence of active treatment. Participants were instructed not to start a diet and/or exercise program intended to induce weight loss during the course of the study.
https://static-content.springer.com/image/art%3A10.1007%2Fs00228-006-0220-1/MediaObjects/228_2006_220_Fig1_HTML.gif
Fig. 1

Study design and dosage measurements

Upon completion of all screening procedures, participants who qualified for enrollment were randomized at a 1:1 ratio to receive TPM or placebo b.i.d. for 6 months. The randomization was stratified by participation in the 24-h energy expenditure substudy, i.e., a separate randomization schedule was provided for those who participated in the 24-h energy expenditure substudy and for those who did not. This ensured balance between the two treatment groups in both the overall study sample and in the subgroup of patients who participated in the 24-h energy expenditure substudy.

Dosage and administration

The study drug was titrated over a 12-week period until a dose of 200 mg b.i.d. (or the maximal tolerated dose without reported adverse events) was reached. This dose was maintained during the stabilization period until the completion of the double-blind phase. Upon completion of the 6th month, all participants began an open-label TPM treatment. For the group that had received placebo over the first 6 months, the TPM dose was titrated over a 12-week period until a dose of 200 mg b.i.d. (or the maximal tolerated dose) was reached. Similarly, this dose was maintained during the stabilization period until the completion of the open-label phase. For the TPM group, 200 mg b.i.d. (or the maximal tolerated dose) was maintained until the completion of the open-label phase. During the second titration period, the double-blind study was tapered at the same rate as the open-label drug was introduced in order to avoid breaking the blind nature of the first study phase.

Patients were evaluated at six study visits: a screening visit, a baseline visit within a month of the initial screening visit, and at months 3, 6, 9, and 12. Screening procedures included physical examination, 12-lead electrocardiogram, medical history, and clinical laboratory tests.

Anthropometric measurements

Body weight and height were measured according to standardized procedures recommended at the Airlie Conference [10]. BMI was calculated as body weight divided by height squared (kg/m2). Skinfold thicknesses were measured with a Harpenden caliper at six sites (biceps, triceps, medial calf, subscapular, abdominal, and suprailiac) following standardized procedures [11].

Body density was obtained from the mean of six measurements with the hydrostatic weighing technique [12]. Before immersion in the hydrostatic tank, the helium dilution method of Meneely and Kaltreider [13] was used to determine the pulmonary residual volume. The percentage of total body fat was determined from body density with the equation of Siri [14]. Body fat mass and fat-free mass were estimated from body weight and the derived percentage of body fat.

Satiety, food preferences, and energy intake

Standardized breakfast and visual analogue scales measurements

The standardized breakfast given to participants after a 12-h overnight fast consisted of white bread (100 g), butter (12 g), peanut butter (16 g), cheddar cheese (40 g), and orange juice (250 g). The meal was designed to have a food quotient of 0.85 and an energy content of 3,067 kJ (733 kcal). Participants were instructed to eat the meal within a period of 30 min or less. Desire to eat, hunger, fullness, and prospective food consumption (PFC) were rated in the fasting state and at 0, 10, 20, 30, 40, 50, and 60 min after the standardized test meal on 150 mm VAS, which were adapted from Hill and Blundell [15]. More details on the VAS measurements have been previously described [16]. It is important to mention that the VAS measurements in our laboratory have been shown to be highly reliable both before and in response to a meal [17].

Buffet-type meal

In order to measure food intake in an experimental context where energy intake and macronutrient preferences can be assessed, a buffet-type meal was served in the laboratory to participants as described by Arvaniti et al. [17]. Briefly, at noon on the same day of the administration of the standardized breakfast, a cold buffet-type meal including a variety of foods was offered and the participants were instructed to eat ad libitum. Participants had 30 min to eat their meal and portions of each food were larger than the participant’s expected intake. There was a large diversity in the protein, lipid, and carbohydrate content of foods in order to detect macronutrient preferences. All foods were weighed before and at the end of the buffet to the nearest gram to quantify the exact intake of each type of food. Energy, protein, lipid, and carbohydrate intake were calculated using the Canadian Nutrient File [18] and/or information on food labels. It is important to mention that this assessment was not performed at month 9 for all participants.

3-day dietary record

All participants were invited to complete a 3-day dietary record including 2 weekdays and 1 weekend day [19] in order to evaluate energy intake and diet quality. A dietician explained to the participants how to complete their records and to measure quantities of ingested foods. Mean daily energy intake as well as macronutrient intake were estimated by a dietician using a computerized version of the Canadian Nutrient File [18].

Energy expenditure, fat oxidation, and body energy loss

Resting metabolic rate measurement

Resting metabolic rate (RMR) was measured by indirect calorimetry in the morning, after a 12-h overnight fast. After a 15-min resting period, expired gas collection was performed through a mouthpiece for 15 min while the nose was clipped during the whole measurement. Oxygen and carbon dioxide concentrations were determined by non-dispersive infrared analysis (Uras 10 E; Hartmann & Braun, Mannheim, Germany), whereas pulmonary ventilation was assessed using an S-430A measurement system (KL Engineering, Ventura, CA, USA). The Weir formula [20] was used to determine the energy equivalent of oxygen volume. As previously reported [21], RMR measurements as performed in our facilities provide a reliability coefficient of 0.9 and a coefficient of variation of less than 6%. The determination of fat oxidation was assessed through the calculations previously described by Frayn [22], while assuming that protein oxidation contributes to 10% of total energy expenditure measured under these conditions.

24-hour energy expenditure

Twenty-four hour total energy expenditure was measured in a subgroup of participants with a whole-body indirect calorimeter that has been shown to ensure highly reproducible data in our laboratory, with a within-subject coefficient of variation of 2.8% [23]. This assessment was performed at baseline, month 6 and month 12 in the subgroup of participants. Participants entered the calorimeter at approximately 7:30 am after an overnight fast (12 h). During their stay in the calorimeter, participants were maintained as much as possible in energy balance by using the resting energy expenditure performed at the initial visit, by extrapolating this value over a 24-h period, and by then multiplying this value by an activity factor of 1.32, as previously described [24]. This established individual energy intake was maintained for the three measurements of 24-h energy expenditure to ensure that a similar thermic effect of food is maintained under all measures. Moreover, before each visit to the metabolic chamber, participants were asked to refrain from exercise and eliminate consumption of foods or beverages containing caffeine for 24 h.

Body energy loss

Mean body energy loss was calculated from hydrostatic weighing measurements by assuming that the energy equivalent of fat and lean tissues corresponded to 9,300 and 1,020 kcal/kg respectively [25].

Statistical analyses

The sample size determination was based on 30 participants per treatment group and estimates for the power calculation were based on a 3-month double-blind study of fenfluramine and placebo in a similar population of viscerally obese men [26]. Two-sided Student’s t test was used to compare means of baseline characteristics of study participants in the safety population. The safety population consisted of all randomized participants who received at least one dose of the study drug and provided at least one safety assessment after the beginning of the double-blind treatment (35 placebo and 33 TPM). In addition, a paired Student’s t test was used to assess changes from baseline at months 3, 6, 9, and 12 in both the modified intent-to-treat (MITT) population and the completers’ population. The double-blind MITT population consisted of all randomized participants who received at least one dose of the study drug and provided at least one evaluation of a primary or secondary efficacy variable during the double-blind phase ( 35 placebo and 32 TPM) and the open-label MITT population consisted of all double-blind completers who received at least one dose of open-label study drug and provided at least one evaluation of a primary or secondary efficacy variable during the open-label phase (29 placebo/TPM and 19 TPM/TPM). The completers’ population consisted of all randomized participants who had at least 161 days of exposure to the double-blind study drug (n = 29 placebo and 20 TPM) and the open-label completer’s population consisted of all double-blind completers who had at least 161 days of exposure to the open-label study drug (n = 21 placebo/TPM and 13 TPM/TPM). For comparisons of changes between the two groups (placebo/TPM and TPM/TPM), we used the unpaired Student’s t test. Multiple regression analysis was performed for the two groups in the completers’ population with 24-h daily energy expenditure (DEE) as the dependent variable and body weight and age as the predicting variables. The following equation was retained to predict daily energy expenditure: Predicted DEE (kcal/day) = 742.875−5.274 age + 19.176 weight (kg) (R2 = 0.76) [27]. This regression equation was established by using baseline values and was used to predict the theoretical DEE values in both groups at months 6 and 12. The predicted DEE values were then compared with measured DEE values using a paired Student’s t test. Data are given as mean ± standard deviation (SD) unless otherwise noted. Statistical significance was set at p < 0.05. All statistical analyses were performed using the JMP program version 3.2.2 (SAS Institute, Cary, NC, USA).

Results

Participants

A total of 68 patients were randomized (n = 35 placebo and 33 TPM). The baseline characteristics of the safety population were balanced between the two groups, as shown in Table 1. Of these, 48 patients completed the double-blind phase (29 placebo and 19 TPM) and 34 completed the open-label phase (n = 21 placebo/TPM and 13 TPM/TPM). With regard to the subgroup of participants included in the 24-h energy expenditure substudy, 22 patients were randomized (13 placebo and 9 TPM). Of these, 18 patients completed the double-blind phase (11 placebo and 7 TPM) and 16 completed the open-label phase (10 placebo/TPM and 6 TPM/TPM). It is also to be noted that they did not differ in any respect from the study group as a whole. The participants’ disposition from screening to study completion is shown in Fig. 2.
Table 1

Baseline characteristics of patients (safety population)

Variable

Placebo (n = 35)

TPM (n = 33)

Age (years)

42.6 ± 7.8

43.1 ± 7.6

Body weight (kg)

97.6 ± 11.7

97.6 ± 8.8

BMI (kg/m2)

31.7 ± 2.6

31.8 ± 2.6

Systolic blood pressure (mmHg)

113 ± 11

115 ± 11

Diastolic blood pressure (mmHg)

80 ± 8

79 ± 8

Fasting total cholesterol (mmol/l)

4.90 ± 0.68

4.90 ± 0.66

Fasting triglycerides (mmol/l)

2.61 ± 1.03

2.54 ± 0.89

Fasting plasma glucose (mmol/l)

5.82 ± 0.49

5.90 ± 0.52

Values are mean ± SD.

Values are not significantly different between the two groups.

https://static-content.springer.com/image/art%3A10.1007%2Fs00228-006-0220-1/MediaObjects/228_2006_220_Fig2_HTML.gif
Fig. 2

Disposition of participants

Changes in anthropometric variables, body composition, and RMR

Tables 2 and 3 present values over time and changes from baseline during the double-blind and open-label phases for the MITT and completers’ population respectively. Topiramate produced significant changes in these variables compared with placebo. However, at the end of the study, the placebo/TPM group essentially showed similar results compared with the TPM/TPM group. Indeed, for the placebo/TPM and TPM/TPM groups respectively, weight loss was −4.9% and −4.6%, reduction in BMI was −5.0% and −3.8%, loss of body fat mass was −14.2% and −12.5%, change in fat-free mass was −2.2% and −2.8%, reduction in percentage of body fat was −9.4% and −8.7%, the decrease in the sum of 6 skinfold thicknesses was −12.7% and −11.8%, and the change in RMR was −5.2% and −9.8% in the MITT population at month 12. Comparable results were observed for changes in adiposity in the completers’ population. However, the decrease in body weight in this population was significantly greater in the Placebo/TPM than in the TPM/TPM participants, as shown in Table 3. In addition, this table indicates that this difference was mostly explained by the greater decrease in fat-free mass that was found in the Placebo/TPM subgroup.
Table 2

Values over time and changes from baseline during the double-blind and open-label phases for anthropometric, body composition, and RMR measurements in the MITT population

Variable

Placebo/TPM

TPM/TPM

n

Mean ± SD

Change

n

Mean ± SD

Change

Weight (kg)

  Baseline

35

97.6 ± 11.7

 

32

97.0 ± 8.4

 

  Month 3

33

98.4 ± 11.6

0.8

29

94.7 ± 10.1

−2.3ac

  Month 6

33

98.7 ± 11.9

1.1

30

93.8 ± 11.4

−3.2ac

  Month 9

26

94.2 ± 12.4

−3.4a

17

91.8 ± 14.1

−5.2a

  Month 12

27

92.8 ± 12.3

−4.8a

17

92.5 ± 14.5

−4.5a

BMI (kg/m2)

  Baseline

35

31.7 ± 2.6

 

32

31.8 ± 2.6

 

  Month 3

33

31.7 ± 2.8

0.0

28

31.2 ± 3.0

−0.6

  Month 6

33

31.8 ± 2.8

0.1

29

30.8 ± 3.4

−1.0bd

  Month 9

26

30.6 ± 2.7

−1.1a

17

30.3 ± 4.0

−1.5b

  Month 12

27

30.1 ± 2.6

−1.6a

17

30.6 ± 4.1

−1.2b

Skinfold thicknesses (mm)

  Baseline

34

172.0 ± 31.4

 

31

172.5 ± 29.7

 

  Month 3

29

171.8 ± 26.2

−0.2

22

163.4 ± 30.9

−9.1bd

  Month 6

31

173.3 ± 30.1

1.3

25

164.2 ± 39.2

−8.3bd

  Month 9

23

160.4 ± 32.9

−11.6a

16

153.8 ± 41.7

−18.7a

  Month 12

24

150.2 ± 38.1

−21.8a

16

152.1 ± 45.2

−20.4a

Body fat mass (kg)

  Baseline

35

30.2 ± 5.9

 

32

30.4 ± 5.9

 

  Month 3

31

29.8 ± 6.0

−0.4

24

29.0 ± 6.5

−1.4bd

  Month 6

31

30.0 ± 6.1

−0.2

25

28.1 ± 7.3

−2.3bc

  Month 9

22

26.5 ± 5.8

−3.7a

16

26.5 ± 8.6

−3.9a

  Month 12

23

25.9 ± 6.5

−4.3a

16

26.6 ± 8.9

−3.8a

Fat-free mass (kg)

  Baseline

35

67.3 ± 8.4

 

32

66.8 ± 5.4

 

  Month 3

31

68.2 ± 7.5

0.9

24

65.2 ± 5.7

−1.6d

  Month 6

31

68.3 ± 7.7

1.0

25

65.1 ± 6.4

−1.7d

  Month 9

22

66.8 ± 8.3

−0.5

16

64.3 ± 7.0

−2.5b

  Month 12

23

65.8 ± 8.1

−1.5

16

64.9 ± 7.1

−1.9

Percentage of body fat (%)

  Baseline

35

30.9 ± 4.2

 

32

31.1 ± 4.4

 

  Month 3

31

30.3 ± 3.8

−0.6

24

30.5 ± 4.5

−0.6

  Month 6

31

30.4 ± 3.9

−0.5

25

29.8 ± 5.1

−1.3

  Month 9

22

28.3 ± 3.6

−2.6a

16

28.5 ± 6.5

−2.6b

  Month 12

23

28.0 ± 4.7

−2.9a

16

28.4 ± 6.6

−2.7b

RMR (kcal/day)

  Baseline

35

1,834 ± 249

 

32

1,851 ± 257

 

  Month 3

28

1,903 ± 358

69

23

1,736 ± 264

−115d

  Month 6

31

1,763 ± 222

−71

24

1,680 ± 216

−171b

  Month 9

21

1,703 ± 160

−131b

15

1,655 ± 270

−196b

  Month 12

24

1,739 ± 226

−95

16

1,669 ± 202

−182b

BMI  body mass index, RMR  resting metabolic rate.

The skinfold thicknesses were the sum of biceps, triceps, medial calf, abdomen, subscapular, and suprailiac skinfold measurements.

aSignificantly different from baseline, p < 0.01; bp < 0.05.

cSignificantly different from placebo/TPM group, p < 0.01; dp < 0.05.

Table 3

Values over time and changes from baseline during the double-blind and open-label phases for anthropometric, body composition, and RMR measurements in the completers’ population

Variable

Placebo/TPM

TPM/TPM

n

Mean ± SD

Change

n

Mean ± SD

Change

Weight (kg)

  Baseline

29

98.3 ± 11.0

 

20

98.5 ± 9.5

 

  Month 3

29

98.2 ± 11.2

−0.1

19

94.9 ± 11.8

−3.6ac

  Month 6

28

97.2 ± 9.3

−1.1

18

92.9 ± 14.0

−5.6ac

  Month 9

20

92.6 ± 12.2

−5.7a

13

93.7 ± 13.1

−4.8a

  Month 12

20

90.6 ± 12.1

−7.7a

12

94.2 ± 14.1

−4.3ad

BMI (kg/m2)

  Baseline

29

31.8 ± 2.6

 

20

32.6 ± 2.7

 

  Month 3

29

31.8 ± 2.7

0.0

19

31.4 ± 3.2

−1.2bd

  Month 6

28

31.7 ± 2.6

−0.1

18

30.7 ± 3.9

−1.9bd

  Month 9

20

30.3 ± 2.6

−1.5a

13

30.3 ± 4.0

−2.3a

  Month 12

20

29.7 ± 2.4

−2.1a

12

30.5 ± 4.2

−2.1b

Skinfold thicknesses (mm)

  Baseline

28

170.5 ± 29.4

 

19

179.5 ± 29.4

 

  Month 3

27

172.7 ± 25.9

2.2

17

164.3 ± 32.5

−15.2bd

  Month 6

28

172.7 ± 29.5

2.2

18

158.4 ± 38.5

−21.1ad

  Month 9

20

158.5 ± 34.4

−12.0a

13

163.0 ± 39.6

−16.5b

  Month 12

20

148.6 ± 39.6

−21.9a

11

152.7 ± 43.7

−26.8a

Body fat mass (kg)

  Baseline

29

30.3 ± 5.9

 

20

31.8 ± 5.7

 

  Month 3

29

30.1 ± 6.1

−0.2

19

29.5 ± 7.1

−2.3bc

  Month 6

28

29.9 ± 5.8

−0.4

18

27.5 ± 7.9

−4.3ac

  Month 9

20

26.4 ± 6.1

−3.9a

13

28.2 ± 7.6

−3.6a

  Month 12

20

25.8 ± 6.9

−4.5a

12

27.9 ± 8.2

−3.9a

Fat-free mass (kg)

  Baseline

29

67.9 ± 7.4

 

20

66.7 ± 6.0

 

  Month 3

29

68.1 ± 7.4

0.2

19

65.5 ± 6.3

−1.2

  Month 6

28

67.3 ± 6.2

−0.6

18

65.4 ± 7.4

−1.3

  Month 9

20

66.2 ± 8.0

−1.7

13

65.5 ± 7.2

−1.2

  Month 12

20

64.8 ± 7.8

−3.1b

12

66.3 ± 7.5

−0.4d

Percentage of body fat (%)

  Baseline

29

30.7 ± 3.8

 

20

32.1 ± 4.0

 

  Month 3

29

30.5 ± 3.8

−0.2

19

30.7 ± 4.9

−1.4d

  Month 6

28

30.6 ± 3.9

−0.1

18

29.1 ± 5.4

−3.0bc

  Month 9

20

28.4 ± 3.8

−2.3a

13

29.6 ± 5.6

−2.5b

  Month 12

20

28.2 ± 5.0

−2.5a

12

29.1 ± 5.8

−3.0b

RMR (kcal/day)

  Baseline

29

1,817 ± 243

 

32

1,862 ± 238

 

  Month 3

26

1,922 ± 365

105

23

1,742 ± 283

−120d

  Month 6

26

1,746 ± 223

−71

24

1,693 ± 238

−169b

  Month 9

18

1,678 ± 137

−139b

12

1,688 ± 290

−174b

  Month 12

19

1,735 ± 234

−82

12

1,696 ± 211

−166b

The skinfold thicknesses were the sum of biceps, triceps, medial calf, abdomen, subscapular, and suprailiac skinfold measurements.

aSignificantly different from baseline, p < 0.01; bp < 0.05.

cSignificantly different from placebo/TPM group, p < 0.01; dp < 0.05.

Changes in satiety, food preferences and energy intake

No obvious difference could be noted in the changes observed from baseline and between the two groups (placebo/TPM and TPM/TPM) with regard to the VAS measurements following the standardized breakfast (data not shown). Thus, no significant change was observed between the two groups and the satiating effect of breakfast was observed both in the placebo group during the double-blind phase as well as in the TPM group during the open-label phase.

However, results obtained following the administration of the buffet-type meal served in the laboratory revealed that TPM reduced the increase in caloric intake to be expected following weight loss (Table 4). Particularly, all three macronutrient intakes were reduced at month 3 in the TPM/TPM group and these changes were significantly different from the placebo/TPM group (p < 0.05). Moreover, the change in carbohydrate intake was significantly different from baseline at month 12 in the TPM/TPM group of completers.
Table 4

Values over time and changes from baseline during the double-blind and open-label phases for food preferences by the use of a buffet-type meal in the completers’ population

Characteristic

Placebo/TPM

TPM/TPM

 

n

Mean ± SD

Change

n

Mean ± SD

Change

Total calories (kcal)

Baseline

29

1,153 ± 367

 

20

1,243 ± 339

 

Month 3

27

1,200 ± 410

47

19

1,088 ± 380

−155ab

Month 6

28

1,140 ± 292

−13

17

1,164 ± 365

−79

Month 12

20

1,076 ± 509

−77

12

1,152 ± 305

−91

Lipid (kcal)

Baseline

29

478 ± 192

 

20

504 ± 149

 

Month 3

27

513 ± 225

35

19

463 ± 197

−41b

Month 6

28

455 ± 168

−23

17

494 ± 181

−10

Month 12

20

443 ± 256

−35

12

487 ± 173

−17

Carbohydrate (kcal)

Baseline

29

460 ± 140

 

20

515 ± 163

 

Month 3

27

472 ± 174

12

19

436 ± 164

−79ab

Month 6

28

487 ± 129

27

17

473 ± 152

−42

Month 12

20

445 ± 206

−15

12

463 ± 112

−52a

Protein (kcal)

Baseline

29

215 ± 82

 

20

224 ± 89

 

Month 3

27

215 ± 68

0

19

190 ± 78

−34b

Month 6

28

198 ± 57

−17

17

196 ± 71

−28

Month 12

20

188 ± 95

−27

12

202 ± 82

−22

aSignificantly different from baseline, p < 0.05.

bSignificantly different from placebo/TPM group, p < 0.05.

Accordingly, Table 5 shows values over time and changes from baseline for macronutrient food intake in the completers’ population by the use of a 3-day dietary record. There was an obvious reducing effect of TPM on food intake, and such reduction was accompanied by a decrease in carbohydrate and lipid intakes. It is also interesting to note that this reducing effect of TMP on food intake appeared to be less marked at month 12 in the TPM/TPM group.
Table 5

Values over time and changes from baseline during the double-blind and open-label phases for macronutrient food intake in the completers’ population

Characteristic

Placebo/TPM

TPM/TPM

 

n

Mean ± SD

Change

n

Mean ± SD

Change

Total daily calories (kcal)

Baseline

28

2,636 ± 468

 

18

2,913 ± 734

 

Month 3

21

2,631 ± 536

−5

17

2,452 ± 700

−461ab

Month 6

17

2,634 ± 598

−2

10

2,605 ± 407

−308ab

Month 9

15

2,176 ± 563

−460a

8

2,507 ± 868

−406a

Month 12

16

2,176 ± 507

−460a

9

2,802 ± 837

−111b

Lipid (kcal)

Baseline

28

932 ± 232

 

18

1,044 ± 297

 

Month 3

21

964 ± 248

−32

17

847 ± 204

−197ab

Month 6

17

929 ± 258

−3

10

908 ± 162

−136ab

Month 9

15

721 ± 203

−211a

8

866 ± 281

−178a

Month 12

16

741 ± 246

−191a

9

981 ± 364

−63b

Carbohydrate (kcal)

Baseline

28

1,181 ± 300

 

18

1,358 ± 371

 

Month 3

21

1,157 ± 336

−24

17

1,165 ± 450

−193ab

Month 6

17

1,173 ± 297

−8

10

1,195 ± 271

−163ab

Month 9

15

977 ± 360

−204a

8

1,212 ± 542

−146

Month 12

16

974 ± 283

−207a

9

1,289 ± 376

−69b

Protein (kcal)

Baseline

28

448 ± 89

 

18

433 ± 116

 

Month 3

21

433 ± 117

−15

17

388 ± 108

−45

Month 6

17

463 ± 121

15

10

427 ± 90

−6

Month 9

15

431 ± 123

−17

8

410 ± 105

−23

Month 12

16

407 ± 96

−41

9

466 ± 226

33

Data are according to the mean of 3-day dietary records.

aSignificantly different from baseline, p < 0.05.

bSignificantly different from placebo/TPM group, p < 0.05.

Energy expenditure, fat oxidation and body energy loss

In order to better document the variation in 24-h daily energy expenditure (DEE) over time, reference equations established from baseline body weight and age in both groups were used to predict DEE at months 6 and 12 in the completers’ population. The predicted values were then compared with their measured values (Fig. 3). No significant difference was found between the predicted and measured values or between the two groups.
https://static-content.springer.com/image/art%3A10.1007%2Fs00228-006-0220-1/MediaObjects/228_2006_220_Fig3_HTML.gif
Fig. 3

Comparison between measured and predicted daily energy expenditure (DEE) values in the completers’ population. Values are expressed as mean ± SEM. Note: a subgroup of participants was randomized to participate in a 24-h energy expenditure substudy. No significant difference between predicted and measured values

With regard to the respiratory quotient (RQ), values over time and changes from baseline did not reveal any statistically significant effect. In addition, no difference was observed between the two groups at months 6 and 12 (data not shown).

In respect of the mean body energy loss over time in the completers’ population (Fig. 4), results confirmed the efficacy of TPM compared with placebo. Indeed, a mean estimated body energy loss of 41,316 and 4,332 kcal was observed at month 6 for the TPM/TPM and placebo/TPM groups respectively. Figure 4 also illustrates the recovery in the placebo group exposed to the open-label phase, where the body energy loss in the placebo/TPM group even exceeded that observed in the TPM/TPM group at month 12.
https://static-content.springer.com/image/art%3A10.1007%2Fs00228-006-0220-1/MediaObjects/228_2006_220_Fig4_HTML.gif
Fig. 4

Mean body energy loss over time during the double-blind and open-label phases in the completers’ population. *Significantly different from months 3 and 6, p < 0.01; **significantly different from month 9, p < 0.05; † significantly different from month 3, p < 0.01; ‡ significantly different from the placebo group, p < 0.01; § significantly different from the TPM/TPM group, p < 0.05. See Table 3 for the numbers of participants

Safety overview

Adverse events occurred in at least 5% of the TPM-treated participants and were more frequent in TPM-treated individuals than in those receiving placebo. Most of these events were related to the central nervous system (CNS). Paresthesia was the most common, seen in 55% of participants receiving TPM compared with 11% of those on placebo. Although common, paresthesia was mild to moderate in nature and led to the withdrawal of only 3% of TPM-treated participants. Other adverse events occurring more frequently in TPM-treated patients compared with those on placebo included fatigue (24 vs. 17%), somnolence (15 vs. 9%), insomnia (13 vs. 6%), difficulty with concentration/attention (30 vs. 3%), and impotence (9 vs. 0%). For the double-blind phase, adverse events led to withdrawal of 5 participants (14%) in the placebo group and 9 (27%) in the TPM group (p < 0.05 vs. placebo). Adverse events that led to withdrawal and that were significantly more frequent among TPM-treated participants vs. placebo-treated individuals were: paresthesia, difficulty with concentration/attention, impotence, and dizziness.

Discussion

Energy balance is influenced by numerous biological systems involving substrates, hormones, and neuronal messengers that fix the precision with which energy intake is balanced with expenditure over time. From a clinical standpoint, it is likely that under given environmental conditions, variations in these systems can permit energy balance at a reduced level of body fat. One of the objectives of the pharmacotherapy of obesity is to develop new molecules with the hope that they could reduce body energy stores without compromising the ability to maintain energy balance at a satisfactory satiety level. As indicated above, TPM is one of these molecules that was proposed as an anti-obesity agent when its use for the treatment of epilepsy was also found to promote weight loss. The weight-reducing effect of TPM has since been confirmed in well-controlled clinical trials [79], which confirmed previous animal studies that had demonstrated the ability of TPM to favor a negative energy balance [36]. However, these investigations have not provided clear answers to the following questions:
  1. 1.

    What is the magnitude of body energy loss that can be induced by TPM therapy alone?

     
  2. 2.

    Is this body energy loss accompanied by a decrease in satiety?

     
  3. 3.

    What are the respective effects of TPM treatment on energy intake and expenditure?

     
These questions were documented in the present study, which overall demonstrated the potential of TPM per se to induce a significant body weight loss in obese patients.

The experimental design of this study was very instrumental regarding the ability to answer the first of the above questions, mainly because of the use of two complementary experimental strategies. In the first part of the experimental protocol, the double-blind random assignment of participants to either TPM or placebo confirmed the ability of TPM to change body weight set point without any specific dietary restriction. Following the first 6-month intervention period, a difference of about 4 kg of weight loss was observed between particpants having received TPM in comparison to those who were subjected to the placebo. Specifically, TPM treatment was able to induce an additional body energy loss of about 30,000–40,000 kcal over a 6-month period compared with the placebo treatment. This observation was reinforced by the results obtained in the second part of the protocol in which an open-label experimental strategy was used. Interestingly, the participants who had received TPM in the first phase of the protocol did not display any further weight loss over the next 6 months. On the other hand, the group that was exposed to the placebo treatment in the first part of the protocol exhibited a significant body weight loss during Phase 2. Also of interest is the fact that this weight loss was sufficient to reach a body weight/fat loss that was comparable to that observed in individuals having received TPM throughout the 12-month period of the protocol. Taken together, these results suggest that TPM can induce a substantial body energy loss without any dietary energy restriction. Our results showed that this corresponds to a change in body weight set point of about 4 kg. Finally, our results demonstrate that the continuation of a TPM treatment for a period exceeding 6 months of time is not accompanied by additional changes in body energy.

Another important question underlying the present study pertains to the ability to maintain an acceptable level of satiety in response to the spontaneous weight loss induced by TPM. In this regard, our results clearly showed that eating variables such as perceived hunger, fullness, desire to eat, and prospective food consumption were not significantly modified by TPM treatment. This finding is of great clinical relevance since weight loss has generally been reported to promote an increase in hunger and desire to eat [16]. In addition, the data collected when participants were eating ad libitum in the context of a buffet-type meal revealed that this lack of change in hunger was accompanied by a decrease in spontaneous energy intake. This observation is concordant with what was documented by the 3-day dietary record, which showed that TPM promoted a significant decrease in energy, lipid, and carbohydrate intake. These results thus demonstrate that the weight-reducing effect of TPM could be explained by a significant decrease in spontaneous energy intake. Furthermore, it is noteworthy to emphasize that this effect was observed without any significant increase in hunger level.

As indicated above, this study was also designed to determine the extent to which TPM might also affect body thermogenesis. In this case, we had to control for the confounding effect of the additional weight loss induced by TPM compared with the placebo condition. Indeed, it is well established that body weight loss favors a decrease in energy expenditure that can significantly exceed the values predicted by the loss of fat and fat-free mass [28, 29]. Specifically, we thus calculated predicted values of energy expenditure that would take into account the weight loss produced by TPM. The comparison of post-treatment measured and predicted values of energy expenditure revealed no treatment effect of TPM on energy expenditure, be it measured in resting state or over 24 h. This is not so surprising since available research generally shows that in humans, most of the compounds known to affect energy balance mainly produce their effect via a reduction in energy intake.

In summary, the results of this study show that TPM can induce a significant decrease in body energy without altering the capacity to maintain habitual levels of satiety. The weight-reducing effect of TPM is explained by a decrease in spontaneous energy/macronutrient intake, whereas no quantitative impact on energy expenditure seems to be detectable. These results thus indicate that TPM has a significant effect on systems involved in the regulation of energy balance.

Acknowledgements

This study was funded by Johnson & Johnson Pharmaceutical Research & Development, L.L.C.

Copyright information

© Springer-Verlag 2007