Clinical Trial Result Information

Title of Study:
(XENDOS study) Weight Reducing and Type 2 Diabetes Preventing Effects of Xenical in Obese Patients

Fast Facts:

Protocol number:	BM15421
Sponsor:	Hoffman-La Roche Inc
Company division:	Pharmaceutical
Product name:	Xenical
Generic name:	orlistat
Phase of development:	III
Therapeutic area, approved indication:	Obesity
Date of report:	8/21/2002

Clinical study summary:
This multicenter, double-blind, placebo-controlled, randomized parallel design study was conducted to determine the effects of orlistat (Xenical) on weight control and type 2 diabetes prevention in obese patients with either normal or impaired glucose tolerance.

Study center(s) :
22 centers in Sweden.

Objectives:
Primary: To determine 1) if 120mg Xenical, as compared with placebo,(both tid) could prevent or delay the development of type 2 diabetes in obese patients, and 2) the long term weight control effects of 120mg Xenical and placebo for up to 4 yrars. Secondary: (1)To determine the effects of Xenical on other aspects of obesity related risk factors; and (2) to determine the safety and tolerability in obese patients of 120 mg Xenical tid administered orally for up to 4 years.

Methodology:
Eligible patients were entered into the study and randomized 1:1 to either the placebo or Xenical treatment groups. During the baseline visit and for the duration of the study, all patients received lifestyle intervention counseling consisting of a nutritionally balanced, hypocaloric diet and encouragement to increase physical activity. The study was designed to follow patients for up to 4 years of treatment, with the earliest possible analysis date after 2 years of treatment if a sufficient number of patients had converted to type 2 diabetes. Because the number of type 2 diabetes cases at the end of 2 years of treatment was not sufficient, the study was conducted for an additional 2 years.

Number of patients (planned/analyzed):
3304 randomized

Diagnosis and main criteria for inclusion:
Patients between the ages of 30 and 60 years with body mass index (BMI) >=30 kg/m² and an oral glucose tolerance test classification of normal or impaired at baseline.

Test product, dose and mode of administration or test procedure:
Xenical/120 mg/po/tid

Duration of treatment:
4 years

Reference therapy, dose and mode of administration or reference procedure:
Placebo/po/tid

Criteria for evaluation (efficacy, safety):
Primary efficacy parameters: (1) Time to onset of diabetes mellitus and (2) change in body weight from baseline to the end of treatment. Secondary efficacy parameters: Change from baseline in BMI, serum lipids, fasting insulin, fasting glucose, blood pressure, pulse rate, cardio thrombotic markers, anthropometric measurements, and time to onset of impaired glucose tolerance. Safety parameters: adverse events, clinical laboratory tests (hematology, fasting chemistry, and fat-soluble vitamins), 12-lead electrocardiography, and mammography.

Statistical methods:
Efficacy was analyzed for all patients who had baseline efficacy assessment and >=1 post baseline efficacy measurement (intent-to-treat [ITT] population). Survival analysis techniques were used to analyze primary efficacy (time to onset of diabetes). Quantitative changes in primary and secondary efficacy parameters were analyzed at yearly time points using an analysis of covariance model Although ITT last observation carried forward (LOCF) data was the primary data set for assessing efficacy, observed data and completers were also provided.

Summary (efficacy, safety, other results):
Efficacy – Xenical significantly (P<0.01) delayed the onset of type 2 diabetes such that at the end of 4 years of treatment the cumulative rate of diabetes was 9.04% for the placebo group compared with 6.15% for the Xenical group (see Table). The hazard ratio (an estimate of relative risk) indicates that Xenical treatment significantly decreased the hazard of diabetes mellitus by 37.3% relative to placebo. In patients with impaired glucose tolerance at baseline, the difference in the cumulative rate was highly significant (P<0.01). Xenical treatment significantly decreased the hazard of diabetes mellitus in this population by 44.9% relative to placebo. The least squares mean (LSM) change for body weight from baseline to the end of the first year of treatment for the ITT (LOCF) population was –6.19 kg for the placebo group and –10.56 kg for the Xenical group. The difference between treatment groups was statistically significant (P<0.001) and was maintained for each of the yearly time points throughout the entire 4-year treatment period. After 4 years of treatment, 44.8 of Xenical treated and 28.0% of placebo treated patients lost >=5% of baseline body weight (P<0.001) and 20.6% of Xenical treated and 10.3% of placebo treated patients lost >=10% of baseline body weight (P<0.001). Throughout the entire 4-year treatment period, Xenical-treated compared with placebo-treated patients had statistically significant (P<0.001) reductions in several metabolic risk factors. These included BMI, total cholesterol, low-density lipoprotein (LDL) cholesterol, LDL:HDL (high-density lipoprotein) ratio, fasting insulin, fasting glucose, systolic blood pressure, diastolic blood pressure, pulse rate, plasminogen activator inhibitor, and waist circumference.

Summary of Efficacy Parameters– LOCF Data, ITT Population
	Year 1	Year 2	Year 3	Year 4
Diabetic Status – All ITT Patients
Cumulative rate of diabetes: placebo vs Xenical*	2.00% vs 0.99%	4.29% vs 2.05%	6.98% vs 4.45%	9.04% vs 6.15%
4-year cumulative reduction in relative risk (hazard) of diabetes relative to placebo	37.3%*
Diabetic Status – Patients with IGT at Baseline
Cumulative rate of diabetes: placebo vs Xenical*	8.27% vs 3.71%	15.70% vs 6.80%	23.53% vs 16.32%	28.77% vs 18.79%
4-year cumulative reduction in relative risk (hazard) of diabetes relative to placebo	44.9%*
Body Weight (kg)
LSM difference from placebo	-4.37*	-4.00*	-3.24*	-2.72*
Categorical loss >=5% placebo (%) vs Xenical (%)	45.1 vs 72.8*	35.2 vs 60.7*	30.2 vs 50.1*	28.0 vs 44.8*
>=10% placebo (%) vs Xenical (%)	20.8 vs 41.0*	15.8 vs 32.4*	11.7 vs 23.5*	10.3 vs 20.6*
Body Mass Index (kg/m 2)
LSM difference from placebo	-1.48*	-1.35*	-1.09*	-0.91*
Total cholesterol (% change)
LSM difference from placebo	-7.37*	-5.47*	-5.13*	-4.99*
LDL cholesterol (% change)
LSM difference from placebo	-9.96*	-8.16*	-7.94*	-7.80*
HDL cholesterol (% change)
LSM difference from placebo	-4.06*	-1.80*	-1.24*	-1.09
LDL:HDL ratio
LSM difference from placebo	-0.20*	-0.21*	-0.20*	-0.20*
Fasting insulin (pmol/L)
LSM difference from placebo	-8.44*	-8.86*	-10.09*	-9.22*
Fasting glucose (mmol/L)
LSM difference from placebo	-0.11*	-0.14*	-0.10*	-0.11*
Systolic blood pressure (mmHg)
LSM difference from placebo	-2.46*	-1.68*	-1.58*	-1.52*
Diastolic blood pressure (mmHg)
LSM difference from placebo	-1.32*	-1.10*	-0.76*	-1.05*
Pulse Rate (pulse/min)
LSM difference from placebo	-1.79*	-1.64*	-1.13*	-1.15*
Fibrinogen ( m mol/L)
LSM difference from placebo	0.06	-0.09	-0.19*	-0.16*
Plasminogen activator inhibitor (mcgmol/L)
LSM difference from placebo	-4.08*	-4.10*	-3.30*	-2.60*
Waist circumference (cm)
LSM difference from placebo	-2.88*	-2.81*	-2.07*	-1.79*
*P<.05.

Safety: The most common adverse events were gastrointestinal (GI) disorders, including fatty/oily stool, fecal urgency, flatus with discharge, oily evacuation, and oily spotting. The majority were mild or moderate in intensity and were considered unrelated or remotely related to study medication. The incidence of all GI adverse events decreased progressively with each additional year of treatment. Seven placebo-treated and 2 Xenical-treated patients died either during or after treatment ; all these deaths were considered unrelated to study medication. Over the 4-year period, 13% of patients in the placebo group and 15% of patients in the Xenical group had >=1 serious adverse event. GI disorders (2% in both treatment groups) were the most frequent serious adverse events. The majority of serious adverse events were considered unrelated or remotely related to study medication, and their incidence did not show a significant dependency on treatment duration. Overall, 4% of placebo-treated patients and 8% of Xenical-treated patients withdrew prematurely for adverse events or laboratory abnormalities. The higher incidence of withdrawals in the Xenical group was accounted for by a higher incidence of premature withdrawals due to GI events. The annual incidence of premature withdrawals for safety reasons decreased from Year 1 to Year 4. No clinically significant changes in the mean laboratory values from baseline were seen in the 2 treatment groups. Markedly low lymphocyte and neutrophil counts and markedly high ALT, GGT, TSH, and PTH concentrations were the most frequent marked laboratory abnormalities. The marked laboratory abnormalities were largely isolated events, and the annual incidence of the majority of the marked laboratory abnormalities was low (<1%) in both treatment groups. Vitamin A, 1,25-dihydroxy vitamin D, 25-hydroxy vitamin D, vitamin E, and vitamin K1 were assessed during the study. Treatment with Xenical decreased the mean 25-hydroxy vitamin D, vitamin E, and vitamin K1 levels. However, the mean levels of all vitamins assessed at any time during the 4 year treatment period in both groups remained well within the reference ranges. Furthermore, most patients who received vitamin supplementation therapy because of a low baseline vitamin level or a repeated low vitamin level during treatment had their respective vitamin levels normalized to the reference range at the last determination. Female patients were followed up with mammograms for 4 years after starting study treatment. Breast abnormalities from 58 patients were diagnosed during the study. Of these, 4 placebo treated patients and 1 Xenical treated patient had breast cancer, and 1 additional Xenical -treated patient had breast cancer in situ. All other breast abnormalities were benign changes, with the types and occurrence being generally the same between the 2 groups. Bone function was investigated by measures including urine N-telopeptide, serum under-carboxylated osteocalcin, and total serum osteocalcin. There were virtually no differences between the Xenical treated patients and the placebo treated patients for any of these evaluations. The lack of effect was confirmed by bone mass and bone density measurements done over the 4 year study using computerized tomography and/or dual-energy X-ray absorptiometry, which showed no loss over time and virtually no difference between treatment groups. Additional body composition analyses revealed that most of the long term weight loss came from fat stores and the fat loss, as expected, was greater in the Xenical treated patients due to the greater overall weight loss seen in this group. The relative composition of the lost weight was similar between treatment groups. Compared with placebo treatment, no clinically significant findings of physical examination and electrocardiogram associated with Xenical treatment were seen during the study.

Conclusions:
The results of this study show that Xenical and lifestyle changes significantly prevented or decreased the development of type 2 diabetes mellitus compared with placebo and lifestyle changes. Over a 4 year treatment period, there was a 37.3% relative reduction in the incidence of diabetes mellitus in the Xenical and lifestyle group compared with the placebo and lifestyle group. In the subgroup of patients with impaired glucose tolerance at baseline, there was a 44.9% relative reduction in the incidence of diabetes in the Xenical group compared with the placebo group. Xenical maintained a statistically significant and clinically meaningful weight loss effect relative to placebo over 4 years of treatment. Xenical had beneficial effects on metabolic risk factors associated with obesity over 4 years. In addition, the safety and tolerability profile of Xenical over 4 years of treatment showed no evidence of worsening tolerability over time.

Publications (references, if available):
Torgerson J, Boldrin M, Hauptman J, et al. Xenical in the Prevention of Diabetes in Obese Subjects (XENDOS) study. Diabetes Care 2004; 27: 155-161
Torgerson JS. Preventing diabetes in the obese; the XENDOS study and its context. British Journal of Diabetes and Vascular Disease 2004; 4: 22-27
Torgerson JS, Arlinger K, Kappi M et al, Principles for enhanced recruitment of subjects in a large clinical trial. Control Clin Trials 2001; 22: 515-525