Progressive Reduction in Body Weight after Treatment with the Amylin

0021-972X/07/$15.00/0 Printed in U.S.A.

The Journal of Clinical Endocrinology & Metabolism 92(8):2977–2983 Copyright © 2007 by The Endocrine Society doi: 10.1210/jc.2006-2003

Progressive Reduction in Body Weight after Treatment with the Amylin Analog Pramlintide in Obese Subjects: A Phase 2, Randomized, Placebo-Controlled, Dose-Escalation Study Louis Aronne, Ken Fujioka, Vanita Aroda, Kim Chen, Amy Halseth, Nicole C. Kesty, Colleen Burns, Cameron W. Lush, and Christian Weyer Weill-Cornell Medical College (L.A.), New York, New York 10021; Scripps Clinic (K.F.), San Diego, California 92007; University of California, San Diego (V.A.), San Diego, California 92093; and Amylin Pharmaceuticals, Incorporated (K.C., A.H., N.C.K., C.B., C.W.L., C.W.), San Diego, California 92121 Context: In previous 1-yr trials, treatment with pramlintide (120 ␮g), an analog of the ␤-cell hormone amylin, induced sustained reductions in A1C and body weight in insulin-using subjects with type 2 diabetes. Objective: To assess the potential of pramlintide as an antiobesity agent, we assessed the weight effect, safety, and tolerability of pramlintide in non-insulin-treated obese subjects with and without type 2 diabetes at doses greater than previously studied. Design/Setting: We performed a randomized, double-blind, placebocontrolled, multicenter study. Patients: A total of 204 obese subjects [80/20% female/male, age 48 ⫾ 10 yr, and body mass index 37.8 ⫾ 5.6 kg/m2 (mean ⫾ SD)] participated in the study. Intervention: For 16 wk, without concomitant lifestyle intervention, subjects self-administered pramlintide (nonforced dose escalation ⱕ 240 ␮g) or placebo via sc injection three times a day before meals.

N

EUROHORMONAL SIGNALS secreted from adipose tissue, the intestine, and pancreatic islets are known to play a pivotal role in the physiological regulation of glucose and energy homeostasis (1– 6). In diabetes, these advances in endocrine science have translated into several hormone-based therapies that have become an integral part of the therapeutic arsenal, including insulin, glucagon-like peptide-1, and amylin agonists. In contrast, despite major advances in our understanding of the neurohormonal control of food intake and body weight, the development of hormone-based therapeutics for obesity has proved unsuccessful to date. Amylin is a neuroendocrine peptide hormone that is cosecreted with insulin by pancreatic ␤-cells in response to meals. It has centrally mediated glucoregulatory and anFirst Published Online May 15, 2007 Abbreviations: BMI, Body mass index; ITT, intent-to-treat; LOCF, last observation carried forward; TID, three times a day. JCEM is published monthly by The Endocrine Society (http://www. endo-society.org), the foremost professional society serving the endocrine community.

Main Outcome Measures: Weight, waist circumference, tolerability, and safety were the main outcome measures. Results: Pramlintide was generally well tolerated, with 88% of subjects able to escalate to the maximum dose of 240 ␮g. Withdrawal rates were similar between placebo (25%) and pramlintide-treated subjects (29%). Subjects completing 16 wk of pramlintide treatment experienced placebo-corrected reductions in body weight of 3.7 ⫾ 0.5% (3.6 ⫾ 0.6 kg; P ⬍ 0.001) and waist circumference (3.6 ⫾ 1.1 cm; P ⬍ 0.01). Approximately 31% of pramlintide-treated subjects achieved ⱖ5% weight loss (vs. 2% placebo; P ⬍ 0.001). More pramlintide than placebo-treated subjects reported improvements in appetite control (72% vs. 31%), weight control (63% vs. 24%), and overall well-being (52% vs. 17%). No unexpected safety signals were observed. The most common adverse event reported was mild, transient nausea. Pramlintide-treated subjects not reporting nausea experienced weight loss similar to those who did (3.6 ⫾ 0.5% and 3.9 ⫾ 0.5%, respectively). Conclusion: These results support continued evaluation of pramlintide as a potential treatment for obesity. (J Clin Endocrinol Metab 92: 2977–2983, 2007)

orexigenic actions in rodents, and exerts these effects via binding to specific amylin receptors in the area postrema (7), a region of the hindbrain that serves an important role in the reception and integration of peripheral (humoral and vagal afferent) satiety signals in humans (1). The glucoregulatory actions of amylin complement those of insulin in postprandial glucose control by suppressing inappropriate glucagon secretion and regulating gastric emptying (6). Furthermore, in diet-induced obese rodents, amylin administration resulted in sustained reductions in food intake and meal size, as well as in a fat-specific reduction in body weight (8). Pramlintide is a synthetic, soluble analog of human amylin that retains a broad range of the pharmacological actions of the native hormone, including receptor binding, and differs from amylin by only three amino acids (6, 9). Pramlintide is currently approved in the United States as an antihyperglycemic agent for patients with diabetes as an adjunctive therapy to mealtime insulin. In previous trials, insulin-using subjects treated with pramlintide before meals experienced desirable glucoregulatory effects, such as reductions in A1C, postprandial glucose excursions, and total

2977

2978

J Clin Endocrinol Metab, August 2007, 92(8):2977–2983

daily insulin use (10 –14). In addition, acute and long-term adjunctive pramlintide treatment (120 ␮g) reduced food intake and body weight, respectively, in insulin-using subjects with type 2 diabetes (10, 11, 15). Although these previous studies were not designed as obesity pharmacotherapy trials, a consistent finding was a significant reduction in body weight. This weight loss is noteworthy because it occurred in subjects with type 2 diabetes, upon concomitant insulin therapy, and in the face of a significant A1C reduction, factors that all favor weight gain (16 –20). To examine whether pramlintide may also have potential as an antiobesity agent, we conducted a randomized, double-blind, placebo-controlled study in noninsulin-using obese subjects assessing the weight effect, safety, and tolerability of pramlintide, at doses greater than previously studied. Subjects and Methods Subjects Subjects [intent-to-treat (ITT), n ⫽ 204] were obese male and female individuals, between the ages of 30 and 70 yr, with or without type 2 diabetes. Other entry criteria included a body mass index (BMI) of ⱖ30 to ⱕ50 kg/m2, baseline clinical laboratory tests judged by the investigators to be not clinically significant, prestudy weight fluctuations of less than 3 kg 2 months before screening, typical consumption of three meals per day, and euthyroid. Women were surgically sterile, postmenopausal, or practicing appropriate contraception. Obese subjects with type 2 diabetes were eligible if they had an A1C ⱕ 8% and were not treated with antidiabetic agents, except for stable doses of metformin. Exclusion criteria included clinically active significant cardiac disease, untreated or poorly controlled hypertension (sitting blood pressure ⬎ 160/95 mm Hg), clinically significant history of hepatic disease, a malignant disease requiring chemotherapy, psychiatric illness, eating disorders (including anorexia, bulimia, and/or binge eating), or gastrointestinal disorders. Subjects were also ineligible if they currently were enrolled in or planned to enroll in a weight-loss program. Excluded concomitant medications included weight loss agents, antidiabetic agents (with the exception of metformin), steroids, or drugs that affect gastrointestinal motility. The study protocol was approved by the institutional review board of each study site or by a centralized institutional review board, and all patients provided written informed consent. This study was conducted in accordance with the principles described in the Declaration of Helsinki (1964), including all amendments up to and including the 1996 South African revision.

Study design This multicenter (22 centers in the United States), randomized, double-blind, placebo-controlled study consisted of four study periods: a 1-wk placebo lead-in, 4-wk nonforced dose escalation, 12-wk maintenance, and 8-wk off-treatment follow-up. Study medication (pramlintide or matching volumes of placebo) was self-administered via sc injection three times a day (TID), 15 min before the morning, midday, and evening meals. After the placebo lead-in, subjects were randomized (2:1) to 16 wk of treatment with either pramlintide or placebo. Pramlintide treatment was initiated at 60 ␮g TID and increased in 30-␮g increments every 3 d, as tolerated, during the 4-wk nonforced dose-escalation period up to a maximum allowed dose of 240 ␮g. Based on their individually achieved dose at the end of the escalation period, subjects were treated for another 12 wk at one of the following maintenance doses: 120, 180, or 240 ␮g TID. At wk 16, study medication was discontinued, and subjects had a follow-up visit 8 wk later. Placebotreated subjects underwent the same nonforced dose-escalation regimen, using matching injection volumes. The study did not use a lifestyle intervention program. All subjects were instructed to maintain their normal eating and activity patterns, and not to start any new diet or exercise programs during the course of the study.

Aronne et al. • Pramlintide Treatment for Obesity

Measurements Key endpoints included changes in body weight and waist circumference, as well as safety and tolerability. Other evaluations included responses to a nonvalidated end-of-study questionnaire. Weight and waist circumference were recorded at baseline and each subsequent visit. Safety assessments included incidence and severity of treatmentemergent adverse events occurring over the 16-wk treatment period. Other safety signals were assessed by evaluation of concomitant medications, physical examination findings, vital signs, electrocardiograms, and clinical laboratory measures. Tolerability was assessed by the proportion of subjects able to escalate to 240 ␮g TID. Other laboratory pharmacodynamic parameters and vital signs included glycosylated hemoglobin (A1C), fasting plasma glucose, fasting insulin, triglycerides, total cholesterol, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, systolic blood pressure, and diastolic blood pressure. At the end of active treatment (wk 16), patients responded to a blinded, nonvalidated questionnaire assessing subjective perceptions of using the study medication with a five-point Likert scale (strongly disagree, disagree, neutral, agree, and strongly agree). The four key statements to which subjects responded were: 1) I feel that using the study medication improved how I feel overall, 2) I feel that the benefits of using the study medication are worth the inconvenience of taking injections, 3) I feel that using the study medication made it easier for me to control my appetite, and 4) I feel that using the study medication made it easier for me to control my weight.

Statistical analysis A total of 150 subjects (100 pramlintide and 50 placebo) completing the study was considered sufficient to detect a significant difference of 1.7 kg (sd of ⬃3 kg) in mean body weight change from baseline to wk 16 between the pramlintide and placebo groups, with approximately 90% power at the 0.05 significance level. The ITT population included all randomized subjects who received at least one injection of double-blind study medication. The evaluable population included all ITT subjects who remained in the study through wk 16 with no major protocol deviations. Summaries of safety and tolerability were conducted using the ITT population. Changes in body weight and waist circumference were analyzed using both the ITT and evaluable populations. Missing data for the ITT population were imputed using the last observation carried forward (LOCF) method. Laboratory pharmacodynamic parameters and vital signs (A1C, fasting plasma glucose, fasting insulin, triglycerides, total cholesterol, highdensity lipoprotein cholesterol, low-density lipoprotein cholesterol, and systolic and diastolic blood pressure) were analyzed using the evaluable population. Subgroup analyses [by initial weight loss response (ⱖ1% and ⬍1% weight loss by wk 4), diabetes status (subjects with and without type 2 diabetes), gender, and the occurrence of treatment-emergent nausea] were conducted to characterize further the weight effect of pramlintide. Correlation analysis was used to study the interrelationship between change in body weight and change in waist circumference. The Pearson correlation coefficients (with P values) are presented. Changes from baseline in body weight, waist circumference, and other laboratory pharmacodynamic parameters and vital signs were analyzed using a general linear model, including factors for treatment, baseline body weight stratum, diabetes status, and study site. P values were based on the least squares mean differences between pramlintide and placebo in the change from baseline to each visit. A sensitivity analysis was conducted to explore the impact of missing data on the change in body weight using data from the ITT population. A mixed effects model was used with factors for treatment, time, treatment-bytime interaction, baseline body weight stratum, and study site. The percentage of subjects achieving ⱖ5% weight loss from baseline to wk 16 and the incidence of adverse events were analyzed using Fisher’s exact test. For all analyses, a P value of ⬍0.05 was considered statistically significant. End-of-study questionnaire responses were summarized for each treatment with the number and percentage of subjects responding in the following pooled categories: strongly agree/agree, neutral, and disagree/strongly disagree. All parameters were reported as mean ⫾ se unless otherwise noted. Demographics data were mean ⫾ sd.



Results Baseline demographics and disposition

The baseline demographics and characteristics of the placebo and pramlintide treatment groups were well matched (Table 1). The majority of subjects were female (80%), Caucasian (78%), and nonsmokers (92%), with an average BMI of 37.8 kg/m2. The population included 44 subjects with type 2 diabetes (placebo n ⫽ 15; pramlintide n ⫽ 29). Overall withdrawal rates were similar between the placebo (25%) and pramlintide (29%) treatment groups. Pramlintide treatment was not associated with increased withdrawal due to adverse events, compared with placebo treatment. During the escalation period, one pramlintide and one placebo-treated subject withdrew due to adverse events (injection site rash and nausea, respectively). With the nonforced dose-escalation approach used, 88% of subjects randomized to pramlintide were able to reach the highest possible dose of 240 ␮g TID, and 8% reached a maintenance dose of 180 ␮g TID. Due to the small number of subjects treated with a maintenance dose less than 240 ␮g, all pramlintide subjects were pooled into a single group. During the maintenance period, the main reason for withdrawal was withdrawal of consent (15% in both placebo- and pramlintide-treatment groups). Body weight and waist circumference

Placebo-treated subjects had minimal changes in body weight over the 16-wk treatment period (Fig. 1A). In contrast, TABLE 1. Baseline demographic characteristics and disposition ITT (n ⫽ 204)

Male/female (%) Caucasian/other (%) Without/with type 2 (%) Nonsmoker/smoker (%) Age (yr) Height (cm) Body weight (kg) BMI (kg/m2) Waist circumference (cm)

Placebo (n ⫽ 67)

Pramlintide (n ⫽ 137)

19/81 79/21 78/22 93/7 49 ⫾ 10 167 ⫾ 8 104.9 ⫾ 19.1 37.6 ⫾ 5.5 109.1 ⫾ 15.1

20/80 77/23 79/21 92/8 48 ⫾ 10 167 ⫾ 8 105.5 ⫾ 18.4 37.9 ⫾ 5.7 109.5 ⫾ 14.8

Disposition

Escalation period Withdrew Withdrawal of consent Adverse event Othera Subjects escalating to 240 ␮g TIDb Maintenance period Withdrew Withdrawal of consent Adverse event Othera Completed maintenance Evaluable

2979

pramlintide-treated subjects experienced significant weight loss from baseline as early as wk 2, which was progressive up to wk 16, with no evidence of a plateau. At wk 16, the placebo-corrected reduction in weight after pramlintide treatment was statistically significant compared with placebo in the evaluable (3.7 ⫾ 0.5%, P ⬍ 0.001; 3.6 ⫾ 0.6 kg, P ⬍ 0.001) and ITT population using LOCF (ITT-LOCF 2.8 ⫾ 0.5%, P ⬍ 0.001; 2.7 ⫾ 0.5 kg, P ⬍ 0.001) and the mixed effects model (3.2 ⫾ 0.5%, P ⬍ 0.001; 3.1 ⫾ 0.5 kg, P ⬍ 0.001). Therefore, our primary study results were robust, not only in the ITT-LOCF analysis, but also when analyzed via mixed effect models. At the end of the follow-up period, 8 wk after treatment cessation, the placebo-corrected reduction in body weight was still statistically significant (evaluable 2.5 ⫾ 0.7%; P ⬍ 0.01), even though pramlintide-treated subjects had on average regained one third of the overall weight loss observed by wk 16. At wk 16, 30 evaluable pramlintide-treated subjects achieved ⱖ5% weight loss from baseline, compared with only one of the evaluable placebo-treated subjects (31% vs. 2%, P ⬍ 0.001; ITT-LOCF 23% vs. 3%, P ⬍ 0.001) (Fig. 1B). Evaluable pramlintide-treated subjects achieving ⱖ5% weight loss by wk 16 lost an average of 7.6 ⫾ 0.5 kg, compared with 1.8 ⫾ 0.3 kg in pramlintide-treated subjects achieving less than 5% weight loss. The reduction in weight in pramlintide-treated subjects was accompanied by a significant reduction in waist circumference compared with placebo-treated subjects after 16 wk of treatment (evaluable 4.3 ⫾ 0.6 vs. 0.7 ⫾ 0.9 cm, P ⬍ 0.01; ITT-LOCF 3.8 ⫾ 0.5 vs. 0.8 ⫾ 0.7 cm, P ⬍ 0.01) (see supplemental Table 1, published as supplemental data on The Endocrine Society’s Journals Online web site at http://jcem. endojournals.org). This reduction in abdominal obesity was proportionate to the overall weight loss in that correlation analysis revealed a significant, positive relationship between the change in weight and change in waist circumference (pramlintide r ⫽ 0.55, P ⬍ 0.001; placebo r ⫽ 0.51, P ⬍ 0.001). Changes in glycemic control parameters, lipoprotein profiles, and blood pressure at wk 16 are provided in supplemental Table 1. Subgroup analyses

2 (3%) 0 (0%) 1 (1%) 1 (1%) 63 (94%)

7 (5%) 0 (0%) 1 (1%) 6 (4%) 121 (88%)

15 (22%) 10 (15%) 1 (1%) 4 (6%) 50 (75%) 48c (72%)

33 (24%) 20 (15%) 4 (3%) 9 (7%) 97 (71%) 97 (71%)

All data are mean ⫾ SD unless otherwise indicated. Numbers may not add up to 100% due to rounding. a Other includes administrative, lost to follow-up, protocol violation, and investigator decision. b Or equivalent volume of placebo. c Before unblinding, two placebo-treated subjects deemed not evaluable due to missed medication and visits.

Initial weight loss response. A larger proportion of pramlintidethan placebo-treated subjects achieved ⱖ1% weight loss by wk 4 (76% vs. 25%). In the subgroup of subjects achieving ⱖ1% weight loss by wk 4, pramlintide-treated subjects (evaluable n ⫽ 74) experienced greater weight loss than placebo-treated subjects (evaluable n ⫽ 12) by wk 16 (4.6 ⫾ 0.4% vs.1.3 ⫾ 0.6%, respectively). Of subjects achieving ⱖ1% weight loss by wk 4, by wk 16, pramlintide-treated subjects (evaluable n ⫽ 23) experienced a reduction in body weight of 0.7 ⫾ 0.4%, while placebo-treated subjects (evaluable n ⫽ 36) gained 0.5 ⫾ 0.5%. Diabetes status. Pramlintide-mediated weight loss at wk 16 was statistically different from placebo both in subjects without (placebo-corrected: evaluable n ⫽ 113, 3.7 ⫾ 0.6%, P ⬍ 0.001; ITT-LOCF n ⫽ 160, 2.8 ⫾ 0.6%, P ⬍ 0.001; and mixed effects model n ⫽ 160, 3.3 ⫾ 0.6%, P ⬍ 0.001) and with type

2980



FIG. 1. Change in body weight. A, Mean (⫾SE) change from baseline in body weight (kg) for pramlintide (filled squares) and placebo (open circles) treated subjects during the escalation, maintenance, and follow-up study periods for evaluable (n ⫽ 145; placebo n ⫽ 48; pramlintide n ⫽ 97) and mixed effects model (wk 16 only; n ⫽ 204; placebo n ⫽ 67; pramlintide n ⫽ 137). B, Proportion of evaluable subjects achieving ⱖ5% reduction in weight by wk 16. **, P ⬍ 0.01. ***, P ⬍ 0.001.

2 diabetes (placebo-corrected: evaluable n ⫽ 33, 3.7 ⫾ 1.0%, P ⬍ 0.001; ITT-LOCF n ⫽ 44; 2.4 ⫾ 0.9%; P ⬍ 0.01; and mixed effects model n ⫽ 44, 3.3 ⫾ 0.9%, P ⬍ 0.001) (Fig. 2).

52% vs. 17% reported that the study medication improved how they felt overall (Fig. 3D).

Gender. Weight-loss effects for pramlintide vs. placebotreated subjects were similar between males (evaluable n ⫽ 31; 3.2 ⫾ 0.9 vs. 0.2 ⫾ 1.2 kg) and females (evaluable n ⫽ 114; 3.8 ⫾ 0.4 vs. 0.1 ⫾ 0.4 kg).

Safety

End-of-study questionnaire

In the end-of-study questionnaire administered at wk 16, the majority of pramlintide-treated subjects reported that they had a positive response to study medication: 72% of pramlintide vs. 31% of placebo-treated subjects reported that the study medication made it easier to control their appetite (Fig. 3A); 63% vs. 24% reported that the study medication made it easier to control their weight (Fig. 3B); 61% vs. 43% reported that the perceived benefits of the study medication outweighed the inconvenience of injections (Fig. 3C); and

FIG. 2. Change in body weight by diabetes status. Mean (⫾SE) change from baseline in body weight (kg) for pramlintide (filled squares) and placebo (open circles) treated subjects (A) without diabetes (evaluable n ⫽ 113; placebo n ⫽ 35; pramlintide n ⫽ 78; mixed effects model wk 16 only; n ⫽ 160; placebo n ⫽ 52; pramlintide n ⫽ 108) and (B) with type 2 diabetes (evaluable n ⫽ 32; placebo n ⫽ 13; pramlintide n ⫽ 19; mixed effects model wk 16 only; n ⫽ 44; placebo n ⫽ 15; pramlintide n ⫽ 29). *, P ⬍ 0.05. **, P ⬍ 0.01. ***, P ⬍ 0.001.

Even at a pramlintide dose (240 ␮g) that was 2-fold higher than the highest dose tested in previous long-term studies in insulin-using subjects with type 2 diabetes, pramlintide appeared to be generally well tolerated. No unexpected safety signals were observed as assessed by evaluation of concomitant medications, physical examination findings, vital signs, electrocardiograms, and clinical laboratory measures. There was no evidence of central nervous system-related adverse events or adverse effects on vital signs. The most frequent treatment-emergent adverse events were injection site-related events and nausea (Table 2). Consistent with observations from previous studies, the most common adverse event among pramlintide-treated subjects that occurred more frequently than in placebo-



2981

There were no reports of moderate or severe hypoglycemia. Mild hypoglycemia adverse events (defined as any reports from subjects or investigators of symptoms consistent with hypoglycemia that may or may not have been documented by glucose monitoring at the time of symptoms and resolved spontaneously without medical intervention) were reported by 11 pramlintide-treated subjects (three with type 2 diabetes). Discussion

FIG. 3. Responses to an end-of-study questionnaire (evaluable n ⫽ 145; placebo n ⫽ 48; pramlintide n ⫽ 97). Subjects responded to the following statements: I feel that using the study medication made it easier for me to control my appetite (A); I feel that using the study medication made it easier for me to control my weight (B); I feel that the benefits of using the study medication are worth the inconvenience of taking injections (C); and I feel that using the study medication improved how I feel overall (D). Data presented as proportions of placebo- and pramlintide-treated subjects (%) who responded strongly disagree/disagree (pooled in dark gray), neutral (light gray), and strongly agree/agree (pooled in white).

treated subjects was mild, transient nausea, which was most prevalent during the first few weeks of treatment and gradually subsided (Fig. 4A). Of the 137 pramlintide-treated subjects, there was only one episode of severe nausea. One pramlintide and one placebo-treated subject withdrew due to nausea. Of note, reductions in body weight for pramlintide-treated subjects were independent of nausea, with similar mean reductions in body weight observed among pramlintidetreated subjects who reported nausea as an adverse event during the study and those who did not (evaluable 3.9 ⫾ 0.5% vs. 3.6 ⫾ 0.5%; ITT-LOCF 3.2 ⫾ 0.4% vs. 2.8 ⫾ 0.4%; and mixed effects model 3.7 ⫾ 0.5% vs. 3.2 ⫾ 0.4%) (Fig. 4B). Injection-related adverse events, which had a similar overall incidence in pramlintide- and placebo-treated subjects, included injection-site erythema and injection-site bruising, were predominantly of mild intensity, and resolved without sequelae.

Consistent with numerous previous studies in insulinusing subjects with types 1 and 2 diabetes (10 –14), pramlintide treatment in obese, noninsulin-using subjects elicited a statistically significant reduction in body weight. Since insulin therapy is often accompanied by weight gain and insulin-using subjects with type 2 diabetes have previously been reported to be poorly responsive to nonpharmacological or pharmacological weight loss interventions (16 –20), we had hypothesized that greater weight loss might be achievable with pramlintide in obese, noninsulin-treated subjects. We studied this hypothesis by examining higher than previously tested doses, using a nonforced dose-escalation regimen. The first important finding of the study was that higher doses of pramlintide were generally well tolerated in obese subjects. Although subjects had the option to stay on lower maintenance doses of 120 and 180 ␮g, approximately 90% were able to escalate to the highest dose of 240 ␮g TID. Second, at this higher dose, the mean reduction in body weight with pramlintide treatment over 16 wk was approximately twice that previously observed over a similar time frame in insulin-treated subjects with type 2 diabetes treated with lower pramlintide doses (120 ␮g) (10 –12). Although it remains to be determined whether the greater magnitude of weight loss in the present study is due to higher pramlintide doses, the lack of concomitant insulin therapy, or both, the results constitute a proof of concept that pramlintide may have therapeutic use as an antiobesity treatment. Finally, the study revealed no evidence of novel safety signals for pramlintide doses of up to 240 ␮g TID. Instead, the safety profile was generally consistent with that observed in previous studies in insulin-treated subjects with diabetes, with the exception that, as expected, no cases of moderate or severe hypoglycemia were observed in obese, noninsulintreated subjects. Although there were a few reports of mild hypoglycemia in this study, no cases of hypoglycemia were reported in a subsequent study in obese subjects (21). Consistent with previous studies in insulin-using subjects with diabetes, nausea was the most commonly reported adverse event associated with pramlintide compared with placebo treatment. However, nausea was typically mild in intensity, short lived, and not a major reason for withdrawal. This finding, combined with the fact that approximately 90% of subjects were able to escalate to the highest maintenance dose (240 ␮g), suggests that this pramlintide dose was generally well tolerated by obese, noninsulin-using subjects. Initial weight loss, obtained during the first 2– 4 wk of treatment, is reported as predictive of subsequent weight loss response (22). In this study pramlintide-treated subjects ex-

2982



TABLE 2. Treatment-emergent adverse events occurring at a frequency ⱖ 5% and with a higher incidence in pramlintide-treated than placebo-treated subjects (ITT subjects, n ⫽ 204) Adverse event

Injection site-related eventsb Nausea Hypoglycemia Diarrhea Dizziness

Placebo (n ⫽ 67)

Pramlintide (n ⫽ 137)

No. (%)

Intensitya (x/y/z)

No. (%)

Intensitya (x/y/z)

28 (42) 15 (22) 1 (1) 5 (7) 4 (6)

27/1/0 15/0/0 1/0/0 5/0/0 4/0/0

59 (43) 52 (38)c 11 (8) 12 (9) 9 (7)

55/4/0 44/7/1 11/0/0 9/3/0 8/1/0

a

Where x/y/z refers to the number of subjects experiencing adverse events by most severe intensity (mild/moderate/severe). Injection site-related events included bruising, burning, discomfort, erythema, hemorrhage, nodule, pain, pruritus, rash, scar, stinging, swelling, urticaria, and vesicles. c P ⬍ 0.05 compared with placebo. b

perienced significant reductions in body weight as early as wk 2. Furthermore, pramlintide-treated subjects experiencing ⱖ1% weight loss within the first 4 wk (76% of all pramlintide-treated subjects) achieved an average weight loss of 4.6% at wk 16. Because even moderate weight loss (5%) can reportedly attenuate the risk of obesity related comorbidities such as diabetes, cardiovascular disease, and dyslipidemia (23, 24), one standard efficacy criterion for pharmacological weight loss trials is the proportion of subjects achieving a reduction of ⱖ5% of initial body weight (25). Although the present study was only of 16-wk duration and did not use concomitant lifestyle intervention, more than 30% of pramlintidetreated subjects lost ⱖ5% body weight, a significantly greater percentage than with placebo treatment (2%). Within this cohort, the average weight loss was 7.6 kg, without evidence of a weight loss plateau at wk 16. These findings suggest that a pramlintide treatment may induce considerable weight loss in a subset of obese subjects, warranting further studies in which pramlintide is tested over longer periods of times and in conjunction with lifestyle intervention. The mechanism(s) by which amylin agonism reduces body weight has been extensively studied in animals and humans (6, 8, 26, 27). In a review of the nonclinical literature, Lutz (5) noted that amylin fulfills all main criteria for a peripheral satiety signal, including dose-dependent reductions in food intake and meal size that occur at near-physiological concentrations and are independent of malaise. Amylin’s anFIG. 4. Occurrence of nausea over time and the relative change in body weight by occurrence of treatment-emergent nausea. A, Cumulative percentage of pramlintide (filled) and placebo (open) treated subjects reporting nausea each week during the placebo lead-in, escalation, and maintenance study periods (ITT n ⫽ 204; placebo n ⫽ 67; pramlintide n ⫽ 137). B, Mean (⫾SE) relative change from baseline in body weight (kg) for pramlintide-treated subjects who reported any episodes of nausea (gray squares) and who never reported nausea (black squares) during the escalation, maintenance, and follow-up study periods for evaluable (n ⫽ 97; pramlintide with nausea n ⫽ 37; pramlintide no nausea n ⫽ 60) and mixed effects model (wk 16 only; n ⫽ 137; pramlintide with nausea n ⫽ 52; pramlintide no nausea n ⫽ 85).

orexigenic effect in rodents is dependent upon amylin receptor signaling within the hindbrain (area postrema) and is accompanied by changes in neuronal activity in the lateral hypothalamus, reminiscent of those after feeding (28). Pairfeeding studies in diet-induced obese rodents indicate that amylin-mediated weight loss is largely attributable to reduced caloric intake and is fat specific, with relative preservation of lean mass (29). These preclinical findings are consistent with results of previous clinical studies that suggest that pramlintide may induce weight loss by modulating satiation and reducing food intake. A single injection of pramlintide (120-␮g dose) has been reported to increase satiety and reduced food intake at a solitary ad libitum buffet in obese subjects (15). This satiogenic mechanism of action was further confirmed in a recent study, which demonstrated that pramlintide-mediated weight loss in obese subjects is accompanied by sustained reductions in 24-h food intake (21). Although food intake was not examined in the present study, in the endof-study questionnaire, a majority of pramlintide-treated subjects agreed that pramlintide assisted in appetite and weight control. In conclusion, this study is the first to demonstrate progressive weight loss with pramlintide in obese, noninsulinusing subjects. Although several gastrointestinal peptide hormones have reduced food intake acutely in obese human subjects (1–5), to our knowledge, the present study, showing 3.7% placebo-corrected weight loss over 16 wk, represents


the most robust clinical proof of concept reported to date for the antiobesity potential of a satiogenic peptide hormone. Further studies are warranted to examine the weight-lowering effect of pramlintide at different doses and dose frequencies, in conjunction with lifestyle intervention, and over prolonged periods of time.


13.

14.

Acknowledgments 15.

Received September 12, 2006. Accepted May 3, 2007. Address all correspondence and requests for reprints to: Christian Weyer, M.D., M.A.S., Executive Director, Clinical Research, Amylin Pharmaceuticals, Inc., 9360 Towne Centre Drive, San Diego, California 92121. E-mail: [email protected]. Disclosure Summary: L.A. and K.F. consult for and have received grant support (2004 –present) from Amylin Pharmaceuticals, Inc. V.A. has received grant support (2004 –present) from Amylin Pharmaceuticals, Inc. K.C., A.H., N.C.K., C.B., C.W.L., and C.W. are employed by and have equity interest in Amylin Pharmaceuticals, Inc.

References 1. Badman MK, Flier JS 2005 The gut and energy balance: visceral allies in the obesity wars. Science 307:1909 –1914 2. Wynne K, Stanley S, McGowan B, Bloom S 2005 Appetite control. J Endocrinol 184:291–318 3. Woods SC, Lutz TA, Geary N, Langhans W 2006 Pancreatic signals controlling food intake; insulin, glucagon and amylin. Philos Trans R Soc Lond B Biol Sci 361:1219 –1235 4. Huda MS, Wilding JP, Pinkney JH 2006 Gut peptides and the regulation of appetite. Obes Rev 7:163–182 5. Lutz TA 2005 Pancreatic amylin as a centrally acting satiating hormone. Curr Drug Targets 6:181–189 6. Young AA 1997 Amylin’s physiology and its role in diabetes. Curr Opin Endocrinol Diab 4:282–290 7. Beaumont K, Kenney MA, Young AA, Rink TJ 1993 High affinity amylin binding sites in rat brain. Mol Pharmacol 44:493– 497 8. Mack C, Hoyt J, Moore C, Jodka C, Sams-Dodd F 2003 Sustained reduction in food intake and body weight in high fat-fed rats during 28-day amylin infusion. Diabetes 52(Suppl 1):A389 (Abstract 1690-P) 9. Young AA, Vine W, Gedulin BR, Pittner R, Janes S, Gaeta LSL, Percy A, Moore CX, Koda JE, Rink TJ, Beaumont K 1996 Preclinical pharmacology of pramlintide in the rat: comparisons with human and rat amylin. Drug Dev Res 37:231–248 10. Hollander P, Maggs DG, Ruggles JA, Fineman M, Shen L, Kolterman OG, Weyer C 2004 Effect of pramlintide on weight in overweight and obese insulintreated type 2 diabetes patients. Obes Res 12:661– 668 11. Hollander PA, Levy P, Fineman MS, Maggs DG, Shen LZ, Strobel SA, Weyer C, Kolterman OG 2003 Pramlintide as an adjunct to insulin therapy improves long-term glycemic and weight control in patients with type 2 diabetes: a 1-year randomized controlled trial. Diabetes Care 26:784 –790 12. Ratner RE, Want LL, Fineman MS, Velte MJ, Ruggles JA, Gottlieb A, Weyer C, Kolterman OG 2002 Adjunctive therapy with the amylin analogue pram-

16. 17.

18.

19.

20. 21.

22. 23. 24. 25. 26. 27.

28. 29.

2983

lintide leads to a combined improvement in glycemic and weight control in insulin-treated patients with type 2 diabetes. Diabetes Technol Ther 4:51– 61 Whitehouse F, Kruger DF, Fineman M, Shen L, Ruggles JA, Maggs DG, Weyer C, Kolterman OG 2002 A randomized study and open label extension evaluating the long-term efficacy of pramlintide as an adjunct to insulin therapy in type 1 diabetes. Diabetes Care 25:724 –730 Ratner RE, Dickey R, Fineman M, Maggs DG, Shen L, Strobel SA, Weyer C, Kolterman OG 2004 Amylin replacement with pramlintide as an adjunct to insulin therapy improves long-term glycaemic and weight control in type 1 diabetes mellitus: a 1-year, randomized controlled trial. Diabet Med 21:1204 – 1212 Chapman I, Parker B, Doran S, Feinle-Bisset C, Wishart J, Strobel S, Wang Y, Burns C, Lush C, Weyer C, Horowitz M 2005 Effect of pramlintide on satiety and food intake in obese subjects and subjects with type 2 diabetes. Diabetologia 48:838 – 848 Klein S 2004 Long-term pharmacotherapy for obesity. Obes Res 12(Suppl): 163S–166S Kelley DE, Bray GA, Pi-Sunyer FX, Klein S, Hill J, Miles J, Hollander P 2002 Clinical efficacy of orlistat therapy in overweight and obese patients with insulin-treated type 2 diabetes: a 1-year randomized controlled trial. Diabetes Care 25:1033–1041 Finer N, Bloom SR, Frost GS, Banks LM, Griffiths J 2000 Sibutramine is effective for weight loss and diabetic control in obesity with type 2 diabetes: a randomised, double-blind, placebo-controlled study. Diabetes Obes Metab 2:105–112 UK Prospective Diabetes Study (UKPDS) Group 1998 Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet [Erratum (1999) 354:602] 352:837– 853 Purnell JQ, Weyer C 2003 Weight effect of current and experimental drugs for diabetes mellitus: from promotion to alleviation of obesity. Treat Endocrinol 2:33– 47 Smith SR, Blundell JE, Burns C, Ellero C, Schroeder B, Kesty N, Chen K, Halseth A, Lush C, Weyer C 2007 Pramlintide-treatment reduces 24-hour caloric intake and meal sizes, and improves control of eating in obese subjects: a 6-week translational research study. Am J Physiol, in press Astrup A, Rossner S 2000 Lessons from obesity management programmes: greater initial weight loss improves long-term maintenance. Obes Rev 1:17–19 Vidal J 2002 Updated review on the benefits of weight loss. Int J Obes Relat Metab Disord 26(Suppl 4):S25–S28 Goldstein DJ, Potvin JH 1994 Long-term weight loss: the effect of pharmacologic agents. Am J Clin Nutr 60:647– 657 Food and Drug Administration 1996 Guidance for the clinical evaluation of weight-control drugs. Rockville, MD: Division of Metabolic and Endocrine Drug Products, Food and Drug Administration Kruger DA, Gloster, MA 2004 Pramlintide for the treatment of insulin-requiring diabetes mellitus: rationale and review of clinical data. Drugs 64: 1419 –1432 Chapman I, Parker B, Doran S, Wishart J, Lush C, Chen K, LaCerte C, Burns C, McKay R, Weyer C, Horowitz M 2007 Low-dose pramlintide reduced food intake and meal duration in healthy, normal weight subjects. Obesity 15:1179 – 1186 Rowland NE, Crews EC, Gentry RM 1997 Comparison of Fos induced in rat brain by GLP-1 and amylin. Regul Pept 71:171–174 Roth JD, Anderson, C 2006 Antiobesity effects of the ␤-cell hormone amylin in diet-induced obese rats: effects on food intake, body weight, composition, energy expenditure, and gene expression. Endocrinology 147:5855–5864

JCEM is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.