Colesevelam Hydrochloride (colesevelam hydrochloride 625 mg) Dailymed

14 Clinical Studies

14.1 Primary Hyperlipidemia

Colesevelam hydrochloride reduces total cholesterol (TC), LDL-C, apolipoprotein B (Apo B), and non-high-density lipoprotein cholesterol (non-HDL-C) when administered alone or in combination with a statin in patients with primary hyperlipidemia.

Approximately 1,600 patients were studied in 9 clinical trials with treatment durations ranging from 4 to 50 weeks. With the exception of one open-label, uncontrolled, long-term extension study, all studies were multicenter, randomized, double-blind, and placebo-controlled. A maximum therapeutic response to colesevelam hydrochloride was achieved within 2 weeks and was maintained during long-term therapy.

Monotherapy

In a study in patients with LDL-C between 130 mg/dL and 220 mg/dL (mean 158 mg/dL), colesevelam hydrochloride was given for 24 weeks in divided doses with the morning and evening meals.

As shown in Table 7, the mean LDL-C reductions were 15% and 18% at the 3.8 g and 4.5 g doses. The respective mean TC reductions were 7% and 10%. The mean Apo B reductions were 12% in both treatment groups. Colesevelam hydrochloride at both doses increased HDL-C by 3%. Increases in TG of 9% to 10% were observed at both colesevelam hydrochloride doses, but the changes were not statistically different from placebo.

Table 7: Response to Colesevelam Hydrochloride Monotherapy in a 24-Week Trial - Percent Change in Lipid Parameters from Baseline

Grams/Day	N	TC	LDL-C	Apo B	HDL-C *	Non-HDL-C	TG *
Placebo	88	+1	0	0	–1	+1	+5
3.8 g (6 tablets)	95	–7†	–15†	–12†	+3†	–10†	+10
4.5 g (7 tablets)	94	–10†	–18†	–12†	+3	–13†	+9
*  Median % change from baseline. † p <0.05 for lipid parameters compared to placebo, for Apo B compared to baseline.

In a study in 98 patients with LDL-C between 145 mg/dL and 250 mg/dL (mean 169 mg/dL), colesevelam hydrochloride 3.8 g was given for 6 weeks as a single-dose with breakfast, as a single-dose with dinner, or as divided doses with breakfast and dinner. The mean LDL-C reductions were 18%, 15%, and 18% for the 3 dosing regimens, respectively. The reductions with these 3 regimens were not statistically different from one another.

Combination Therapy

Co-administration of colesevelam hydrochloride and a statin (atorvastatin, lovastatin, or simvastatin) in 3 clinical studies demonstrated an additive reduction of LDL-C. The mean baseline LDL-C was 184 mg/dL in the atorvastatin study (range 156 mg/dL to 236 mg/dL), 171 mg/dL in the lovastatin study (range 115 mg/dL to 247 mg/dL), and 188 mg/dL in the simvastatin study (range 148 mg/dL to 352 mg/dL). As demonstrated in Table 8, colesevelam hydrochloride doses of 2.3 g to 3.8 g resulted in an additional 8% to 16% reduction in LDL-C above that seen with the statin alone.

Table 8: Response to Colesevelam Hydrochloride in Combination with Atorvastatin, Simvastatin, or Lovastatin - Percent Change in Lipid Parameters

Dose/Day	N	TC	LDL-C	Apo B	HDL-C *	Non-HDL-C	TG *
Atorvastatin Trial (4-week)
Placebo	19	+4	+3	–3	+4	+4	+10
Atorvastatin 10 mg	18	–27†	–38†	–32†	+8	–35†	–24†
Colesevelam hydrochloride 3.8 g/Atorvastatin 10 mg	18	–31†	–48†	–38†	+11	–40†	–1
Atorvastatin 80 mg	20	–39†	–53†	–46†	+6	–50†	–33†
Simvastatin Trial (6-week)
Placebo	33	–2	–4	–4†	–3	–2	+6†
Simvastatin 10 mg	35	–19†	–26†	–20†	+3†	–24†	–17†
Colesevelam hydrochloride 3.8 g/Simvastatin 10 mg	34	–28†	–42†	–33†	+10†	–37†	–12†
Simvastatin 20 mg	39	–23†	–34†	–26†	+7†	–30†	–12†
Colesevelam hydrochloride 2.3 g/Simvastatin 20 mg	37	–29†	–42†	–32†	+4†	–37†	–12†
Lovastatin Trial (4-week)
Placebo	26	+1	0	0	+1	+1	+1
Lovastatin 10 mg	26	–14†	–22†	–16†	+5	–19†	0
Colesevelam hydrochloride 2.3 g/Lovastatin 10 mg Together	27	–21†	–34†	–24†	+4	–27†	–1
Colesevelam hydrochloride 2.3 g/Lovastatin 10 mg Apart	23	–21†	–32†	–24†	+2	–28†	–2
*  Median % change from baseline. † p <0.05 for lipid parameters compared to placebo, for Apo B compared to baseline.

In all 3 studies, the LDL-C reduction achieved with the combination of colesevelam hydrochloride and any given dose of statin therapy was statistically superior to that achieved with colesevelam hydrochloride or that dose of the statin alone. The LDL-C reduction with atorvastatin 80 mg was not statistically significantly different from the combination of colesevelam hydrochloride 3.8 g and atorvastatin 10 mg.

Pediatric Therapy

The safety and efficacy of colesevelam hydrochloride in pediatric patients were evaluated in an 8-week, multicenter, randomized, double-blind, placebo-controlled, parallel-group study followed by an open-label phase, in 194 boys and postmenarchal girls 10 to 17 years of age (mean age 14.1 years) with HeFH, taking a stable dose of an FDA-approved statin (with LDL-C >130 mg/dL) or naïve to lipid-lowering therapy (with LDL-C >160 mg/dL). This study had 3 periods: a single-blind, placebo stabilization period; an 8-week, randomized, double-blind, parallel-group, placebo-controlled treatment period; and an 18-week, open-label treatment period. Forty-seven (24%) patients were taking statins and 147 (76%) patients were statin-naïve at screening. The mean baseline LDL-C at Day 1 was approximately 199 mg/dL.

During the double-blind treatment period, patients were assigned randomly to treatment: Colesevelam hydrochloride 3.8 g/day (n=64), colesevelam hydrochloride 1.9 g/day (n=65), or placebo (n=65). In total, 186 patients completed the double-blind treatment period. After 8 weeks of treatment, colesevelam hydrochloride 3.8 g/day significantly decreased plasma levels of LDL-C, non-HDL-C, TC, and Apo B and significantly increased HDL-C. A moderate, non-statistically significant increase in TG was observed versus placebo (Table 9).

Table 9: Response to C olesevelam Hydrochloride 3.8 g Compared to Placebo in Pediatric Patients 10 to 17 Years of Age – Mean Percent Change in Lipid Parameters from Baseline to Week 8

Treatment Difference	TC	LDL-C	Apo B	HDL-C	Non-HDL-C	TG*
	(N=128)	(N=128)	(N=124)	(N=128)	(N=128)	(N=128)
Colesevelam hydrochloride 3.8 g vs. Placebo	-7†	-13†	-8†	+6†	–11†	+5
* For triglycerides, median % change from baseline. † p≤0.05 for lipid parameters compared to placebo Values represent LS mean. Only patients with values at both study baseline and endpoint are included in this table. Study baseline was defined as the last value measured before or on Day 1 prior to the first dose of randomized study medication. Results were based on the ITT population with LOCF.

During the open-label treatment period patients were treated with colesevelam hydrochloride 3.8 g/day. In total, 173 (89%) patients completed 26 weeks of treatment. Results at Week 26 were consistent with those at Week 8.

14.2 Type 2 Diabetes Mellitus

Colesevelam hydrochloride has been studied as monotherapy and in combination with metformin, pioglitazone, sulfonylureas, and insulin. In these studies, colesevelam hydrochloride and placebo were administered either as 3 tablets twice daily with lunch and dinner or as 6 tablets with dinner alone.

Monotherapy

The efficacy of colesevelam hydrochloride 3.8 g/day as anti-diabetes monotherapy was evaluated in a randomized double-blind, placebo-controlled trial involving 357 patients (176 colesevelam hydrochloride and 181 placebo) with type 2 diabetes mellitus who were treatment-naïve or had not received antihyperglycemic medication within 3 months prior to the start of the study. Statin use at baseline was reported in 13% of the colesevelam hydrochloride-treated patients and 16% of the placebo-treated patients.

Colesevelam hydrochloride resulted in a statistically significant reduction in HbA1c of 0.27% compared to placebo (Table 10).

The mean baseline LDL-C was 121 mg/dL in the monotherapy trial. Colesevelam hydrochloride treatment resulted in a placebo-corrected 11% reduction in LDL-C. Colesevelam hydrochloride treatment also reduced serum TC, ApoB, and non-HDL-C (Table 11). The mean change in body weight was -0.6 kg for colesevelam hydrochloride and -0.7 kg for placebo treatment groups.

Table 10: Glycemic Parameters in a 24-Week Placebo-Controlled Study of Colesevelam Hydrochloride Monotherapy in Patients with Type 2 Diabetes

	Colesevelam hydrochloride 3.8 g/day	Placebo
HbA1c (%), Mean
N	175	169
Baseline	8.25	8.17
Change from baseline*	-0.26	0.01
Treatment difference (p-value)	-0.27 (p = 0.013)
FPG (mg/dL), Mean
N	172	166
Baseline	172	168
Change from baseline*	-4.6	5.7
Treatment difference (p-value)	-10.3 (p = 0.037†)
*Least-squares mean change calculated from an Analysis of Covariance model. †Nominal p = value, not controlled for multiplicity testing. FPG = fasting plasma glucose.

Table 11: Percent Change in Lipid Parameters in a 24-Week Placebo-Controlled Study of Colesevelam Hydrochloride Monotherapy in Patients with Type 2 Diabetes

Dose/Day	N*	TC	LDL-C	Apo B	HDL-C	Non-HDL-C	TG†
Colesevelam hydrochloride 3.8 g	162	-3.3‡	-10.0‡	-5.6‡	1.7	-4.4‡	-15.5
Placebo	160	1.8	1.2	0.9	-0.1	3.0	5.8
* The number of patients with analyzable data, i.e., a baseline and post-treatment value (last observation carried forward), varied slightly among different parameters. The N given represents the smallest number of patients included in the analysis for any parameter. † Median % change from baseline. ‡ p<0.001 for lipid parameters compared to placebo (This more stringent criterion for statistical significance accounts for multiplicity testing of the lipid parameters, which were secondary endpoints in the diabetes trials.).

Add-on Combination Therapy

The efficacy of colesevelam hydrochloride 3.8 g/day in patients with type 2 diabetes mellitus was evaluated in 5 double-blind, placebo-controlled add-on therapy trials involving a total of 1,691 patients with baseline HbA1c 7.5 9.5%. Patients were enrolled and maintained on their pre-existing, stable, background anti-diabetic regimen. Statin use at baseline was reported in 41% of the colesevelam hydrochloride-treated patients and 48% of the placebo-treated patients.

In 3 add-on combination therapy trials (metformin, sulfonylurea and insulin), treatment with colesevelam hydrochloride resulted in a statistically significant reduction in HbA1c of 0.5% compared to placebo. Similar placebo-corrected reductions in HbA1c occurred in patients who received colesevelam hydrochloride in combination with metformin, sulfonylurea, or insulin monotherapy or combinations of these therapies with other anti-diabetic agents. In the pioglitazone trial, treatment with colesevelam hydrochloride resulted in a statistically significant reduction in HbA1c of 0.32% compared to placebo. In the metformin, pioglitazone, and sulfonylurea trials, treatment with colesevelam hydrochloride also resulted in statistically significant reductions in FPG of at least 14 mg/dL compared to placebo.

Colesevelam hydrochloride had consistent effects on HbA1c across subgroups of age, gender, race, body mass index, and baseline HbA1c. Colesevelam hydrochloride’s effects on HbA1c were also similar for the two dosing regimens (3 tablets with lunch and with dinner or 6 tablets with dinner alone).

The mean baseline LDL-C was 104 mg/dL in the metformin study (range 32 to 214 mg/dL), 107 mg/dL in the pioglitazone study (range 48 to 263 mg/dL), 106 mg/dL in the sulfonylurea study (range 41 to 264 mg/dL), 102 mg/dL in the insulin study (range 35 to 204 mg/dL). In these trials, colesevelam hydrochloride treatment was associated with a 12% to 16% reduction in LDL-C levels. The percentage decreases in LDL-C were of similar magnitude to those observed in patients with primary hyperlipidemia. Colesevelam hydrochloride treatment was associated with statistically significant increases in TG levels in the studies of patients on insulin, patients on a sulfonylurea, and patients on pioglitazone but not in the study of patients on metformin. The clinical significance of these increases is unknown. Colesevelam hydrochloride is contraindicated in patients with TG levels >500 mg/dL [see Contraindications (4)], and periodic monitoring of lipid parameters including TG is recommended [see Warnings and Precautions (5.1) and Adverse Reactions (6.1)].

Body weight did not significantly increase from baseline with colesevelam hydrochloride therapy, compared with placebo, in any of the add-on combination diabetes studies.

Add-on Combination Therapy with Metformin

Colesevelam hydrochloride 3.8 g/day or placebo was added to background anti-diabetic therapy in a 26-week trial of 316 patients already receiving treatment with metformin alone (N=159) or metformin in combination with other oral agents (N=157). A total of 60% of these patients were receiving ≥1,500 mg/day of metformin. In combination with metformin, colesevelam hydrochloride resulted in statistically significant placebo-corrected reductions in HbA1c and FPG (Table 12). Colesevelam hydrochloride also reduced TC, LDL-C, Apo B, and non-HDL-C (Table 13). The mean percent change in serum LDL-C levels with colesevelam hydrochloride compared to placebo was -16% among statin users and statin non-users; the median percent change in serum TG levels with colesevelam hydrochloride compared to placebo was -2% among statin users and 10% among statin non-users. The mean change in body weight was -0.5 kg for colesevelam hydrochloride and -0.3 kg for placebo.

Table 12: Glycemic Parameters in a 26-Week Placebo-Controlled Study of Colesevelam Hydrochloride in Combination with Metformin in Patients with Type 2 Diabetes

	Total Patient Population		Metformin Alone		Metformin in Combination with Other Oral Anti-diabetic Agents
	Colesevelam hydrochloride 3.8 g/day	Placebo	Colesevelam hydrochloride 3.8 g/day	Placebo	Colesevelam hydrochloride 3.8 g/day	Placebo
HbA1c (%), Mean
N	148	152	79	76	69	76
Baseline	8.1	8.1	8.2	8.2	8.1	8.
Change from baseline*	-0.4	0.2	-0.4	0.0	-0.4	0.3
Treatment difference (p-value)	-0.5 (p < 0.001)		-0.5 (p = 0.002)		-0.6 (p < 0.001)
FPG (mg/dL), Mean
N	149	152	79	76	70	76
Baseline	178	174	184	180	171	168
Change from baseline*	-3	11	-7	8	0	13
Treatment difference (p- value)	-14 (p = 0.01)		-14 (p = 0.07)		-14 (p = 0.10)
* Least-squares mean change calculated from an Analysis of Covariance model.

Table 13: Percent Change in Lipid Parameters in a 26-Week Placebo-Controlled Study of Colesevelam Hydrochloride in Combination with Metformin in Patients with Type 2 Diabetes

Dose/Day	N*	TC	LDL-C	Apo B	HDL-C	Non-HDL-C	TG†
Total Patient Population
Colesevelam hydrochloride 3.8 g	125	-4‡	-12‡	-4‡	1	-6‡	12
Placebo	126	3	4	4	0	5	7
Metformin Alone
Colesevelam hydrochloride 3.8 g	66	-3	-9	-2	1	-4	15
Placebo	61	2	0	1	-2	4	8
Metformin in Combination with Other Oral Anti-diabetic Agents
Colesevelam hydrochloride 3.8 g	59	-6‡	-15‡	-6‡	1	-7‡	8
Placebo	65	4	7	7	2	6	5
* The number of patients with analyzable data, i.e., a baseline and post-treatment value (last observation carried forward), varied slightly among different parameters. The N given represents the smallest number of patients included in the analysis for any parameter. † Median % change from baseline. ‡ p<0.001 for lipid parameters compared to placebo (This more stringent criterion for statistical significance accounts for multiplicity testing of the lipid parameters, which were secondary endpoints in the diabetes trials.).

Add-on Combination Therapy with Pioglitazone

Colesevelam hydrochloride 3.8 g/day or placebo was added to background anti-diabetic therapy in a 24-week trial of 562 patients already receiving treatment with pioglitazone alone (N=51) or pioglitazone in combination with other oral agents (N=511). Of these, most were on dual therapy with metformin (N=298) or triple therapy with metformin and a sulfonylurea (N=139). In combination with pioglitazone-based therapy, colesevelam hydrochloride resulted in statistically significant reductions in HbA1c and FPG compared to placebo (Table 14). Colesevelam hydrochloride also reduced TC, LDL-C, Apo B, and non-HDL-C but increased serum TG (Table 15). The mean change in body weight was 0.8 kg for colesevelam hydrochloride and 0.4 kg for placebo.

Table 14: Glycemic Parameters in a 24-Week Placebo-Controlled Study of Colesevelam Hydrochloride in Combination with Pioglitazone-Based Therapy in Patients with Type 2 Diabetes

	Colesevelam Hydrochloride 3.8 g/day	Placebo
HbA1c (%), Mean
N	271	276
Baseline	8.2	8.1
Change from baseline*	-0.34	-0.02
Treatment difference (p-value)	-0.32 (0.0001)
FPG (mg/dL), Mean
N	268	270
Baseline	155	157
Change from baseline*	-4.8	+9.9
Treatment difference (p- value)	-14.7 (< 0.0001)
* Least-squares mean change calculated from an Analysis of Covariance model.

Table 15: Percent Change in Lipid Parameters in a 24-Week Placebo-Controlled Study of Colesevelam Hydrochloride in Combination with Pioglitazone-Based Therapy in Patients with Type 2 Diabetes

Dose/Day	N*	TC	LDL-C	Apo B	HDL-C	Non-HDL-C	TG†
Total Patient Cohort
Colesevelam hydrochloride 3.8 g	262	-3‡	-9‡	-5‡	+3	-5‡	+14‡
Placebo	262	+3	+7	+4	+1	+5	+2
* The N given represents the smallest number of patients included in the analysis for any parameter. † Median % change from baseline. ‡ p<0.001 for lipid parameters compared to placebo.

Add-on Combination Therapy with Sulfonylurea

Colesevelam hydrochloride 3.8 g/day or placebo was added to background anti-diabetic therapy in a 26-week trial of 460 patients already treated with sulfonylurea alone (N=156) or sulfonylurea in combination with other oral agents (N=304). A total of 72% of these patients were receiving at least half-maximal doses of sulfonylurea therapy. In combination with a sulfonylurea, colesevelam hydrochloride resulted in statistically significant placebo-corrected reductions in HbA1c and FPG (Table 16). Colesevelam hydrochloride also reduced TC, LDL-C, Apo B, and non-HDL-C, but increased serum TG (Table 17). The mean percent change in serum LDL-C levels with colesevelam hydrochloride compared to placebo was -18% among statin users and -15% among statin non-users; the median percent increase in serum TG with colesevelam hydrochloride compared to placebo was 29% among statin users and 9% among statin non-users. The mean change in body weight was 0.0 kg for colesevelam hydrochloride and 0.4 kg for placebo.

Table 16: Glycemic Parameters in a 26-Week Placebo-Controlled Study of Colesevelam Hydrochloride in Combination with Sulfonylurea in Patients with Type 2 Diabetes

 

	Total Patient Population		Sulfonylurea Alone		Sulfonylurea in Combination with Other Oral Anti-diabetic Agents
	Colesevelam hydrochloride 3.8 g/day	Placebo	Colesevelam hydrochloride 3.8 g/day	Placebo	Colesevelam hydrochloride 3.8 g/day	Placebo
HbA1c (%), Mean
n	218	218	69	80	149	138
Baseline	8.2	8.3	8.2	8.4	8.2	8.3
Change from baseline*	-0.3	0.2	-0.3	0.5	-0.4	0.0
Treatment difference (p-value)	-0.5 (p < 0.001)		-0.8 (p < 0.001)		-0.4 (p < 0.001)
FPG (mg/dL), Mean
n	218	217	70	80	148	137
Baseline	177	181	181	186	175	178
Change from baseline*	-4	10	3	15	-11	4
Treatment difference (p- value)	-14 (p = 0.009)		-12 (p = 0.18)		-14 (p = 0.03)
* Least-squares mean change calculated from an Analysis of Covariance model.

Table 17: Percent Change in Lipid Parameters in a 26-Week Placebo-Controlled Study of Colesevelam Hydrochloride in Combination with Sulfonylurea in Patients with Type 2 Diabetes

Dose/Day	N*	TC	LDL-C	Apo B	HDL-C	Non-HDL-C	TG†
Total Patient Population
Colesevelam hydrochloride 3.8 g	186	-5‡	-16‡	-6‡	1	-6‡	20‡
Placebo	193	0	1	1	0	1	1
Sulfonylurea Alone
Colesevelam hydrochloride 3.8 g	57	-5	-14‡	-5	-1	-6	17
Placebo	68	0	1	1	1	0	-1
Sulfonylurea in Combination with Other Oral Anti-diabetic Agents
Colesevelam hydrochloride 3.8 g	129	-5	-18‡	-7‡	1	-6	21‡
Placebo	125	0	0	1	0	1	2
* The number of patients with analyzable data, i.e., a baseline and post-treatment value (last observation carried forward), varied slightly among different parameters. The N given represents the smallest number of patients included in the analysis for any parameter. † Median % change from baseline. ‡ p<0.001 for lipid parameters compared to placebo (This more stringent criterion for statistical significance accounts for multiplicity testing of the lipid parameters, which were secondary endpoints in the diabetes trials.).

Add-on Combination Therapy with Insulin

Colesevelam hydrochloride 3.8 g/day or placebo was added to background anti-diabetic therapy in a 16-week trial of 287 patients already treated with insulin alone (N=116) or insulin in combination with oral agents (N=171). At baseline, the median daily insulin dose was 70 units in the colesevelam hydrochloride group and 65 units in the placebo group. In combination with insulin, colesevelam hydrochloride resulted in a statistically significant placebo-corrected reduction in HbA1c (Table 18). Colesevelam hydrochloride also reduced LDL-C and Apo B, but increased serum TG (Table 19). The mean percent change in serum LDL-C levels with colesevelam hydrochloride compared to placebo was 13% among statin users and statin non-users; the median percent increase in serum TG levels with colesevelam hydrochloride compared to placebo was 24% among statin users and 17% among statin non-users. The mean change in body weight was 0.6 kg for colesevelam hydrochloride and 0.2 kg for placebo.

Table 18: Glycemic Parameters in a 16-Week Placebo-Controlled Study of Colesevelam Hydrochloride in Combination with Insulin in Patients with Type 2 Diabetes

	Total Patient Population		Insulin Alone		Insulin in Combination with Oral Anti-diabetic Agents
	Colesevelam hydrochloride 3.8 g/day	Placebo	Colesevelam hydrochloride 3.8 g/day	Placebo	Colesevelam hydrochloride 3.8 g/day	Placebo
HbA1c (%), Mean
n	144	136	54	55	90	81
Baseline	8.3	8.2	8.2	8.3	8.3	8.2
Change from baseline*	-0.4	0.1	-0.4	0.2	-0.4	0.0
Treatment difference (p-value)	-0.5 (p < 0.001)		-0.6 (p < 0.001)		-0.4 (p < 0.001)
FPG (mg/dL), Mean
n	144	136	54	55	90	81
Baseline	165	151	165	163	165	143
Change from baseline*	2	16	8	17	-4	14
Treatment difference (p- value)	-15 (p = 0.08)		-9 (p = 0.51)		-18 (p = 0.09)
* Least-squares mean change calculated from an Analysis of Covariance model.

Table 19: Percent Change in Lipid Parameters in a 16-Week Placebo-Controlled Study of Colesevelam Hydrochloride in Combination with Insulin in Patients with Type 2 Diabetes

Dose/Day	N*	TC	LDL-C	Apo B	HDL-C	Non-HDL-C	TG†
Total Patient Cohort
Colesevelam hydrochloride 3.8 g	129	-3	-12‡	-4	-1	-3	23‡
Placebo	121	1	1	1	0	1	0
Insulin Alone
Colesevelam hydrochloride 3.8 g	46	-3	-12	-5	0	-3	19
Placebo	48	2	4	2	3	2	-2
Insulin in Combination with Oral Anti-diabetic Agents
Colesevelam hydrochloride 3.8 g	83	-4	-13	-4	-1	-3	25‡
Placebo	73	-1	-3	0	-1	-1	2
* The number of patients with analyzable data, i.e., a baseline and post-treatment value (last observation carried forward), varied slightly among different parameters. The N given represents the smallest number of patients included in the analysis for any parameter. † Median % change from baseline. ‡ p<0.001 for lipid parameters compared to placebo (This more stringent criterion for statistical significance accounts for multiplicity testing of the lipid parameters, which were secondary endpoints in the diabetes trials.).