14 CLINICAL STUDIES

14.1 Overview of Clinical Studies

RYBELSUS has been studied as monotherapy and in combination with metformin, sulfonylureas, sodium-glucose co-transporter-2 (SGLT-2) inhibitors, insulins, and thiazolidinediones in patients with type 2 diabetes mellitus. The efficacy of RYBELSUS was compared with placebo, empagliflozin, sitagliptin and liraglutide. RYBELSUS has also been studied in patients with type 2 diabetes mellitus with mild and moderate renal impairment.

In patients with type 2 diabetes mellitus, RYBELSUS produced clinically significant reduction from baseline in HbA1c compared with placebo.

The effectiveness of RYBELSUS (formulation R2 - 1.5 mg, 4 mg and 9 mg strengths) [see Dosage and Administration (2.2)] and RYBELSUS (formulation R1 - 3 mg, 7, mg and 14 mg strengths) [see Dosage and Administration (2.3)] has been established as an adjunct to diet and exercise to improve glycemic control in adults with type 2 diabetes mellitus based on adequate and well-controlled studies of RYBELSUS (formulation R1) in adult patients with type 2 diabetes mellitus [see Clinical Pharmacology (12.3)]. Below is a display of the efficacy results of the adequate and well-controlled studies of RYBELSUS (formulation R1) in adult patients with type 2 diabetes mellitus.

The efficacy of RYBELSUS was not impacted by baseline age, sex, race, ethnicity, BMI, body weight, duration of diabetes and degree of renal impairment.

14.2 Monotherapy Use of RYBELSUS in Patients with Type 2 Diabetes Mellitus

In a 26-week double-blind trial (NCT02906930), 703 adult patients with type 2 diabetes mellitus inadequately controlled with diet and exercise were randomized to RYBELSUS 3 mg, RYBELSUS 7 mg or RYBELSUS 14 mg once daily or placebo. Patients had a mean age of 55 years and 51% were men. The mean duration of type 2 diabetes mellitus was 3.5 years and the mean BMI was 32 kg/m2. Overall, 75% were White, 5% were Black or African American and 17% were Asian; 26% identified as Hispanic or Latino ethnicity.

Monotherapy with RYBELSUS 7 mg and RYBELSUS 14 mg once daily for 26 weeks resulted in a statistically significant reduction in HbA1c compared with placebo (see Table 4).

Table 4. Results at Week 26 in a Trial of RYBELSUS as Monotherapy in Adult Patients with Type 2 Diabetes Mellitus Inadequately Controlled with Diet and Exercise

	Placebo	RYBELSUS 7 mg	RYBELSUS 14 mg
Intent-to-Treat (ITT) Population (N)a	178	175	175
HbA1c (%)
Baseline (mean)	7.9	8	8
Change at week 26b	-0.3	-1.2	-1.4
Difference from placebob [95% CI]		-0.9 [-1.1; -0.6]c	-1.1 [-1.3; -0.9]c
Patients (%) achieving HbA1c <7%	31	69	77
FPG (mg/dL)
Baseline (mean)	160	162	158
Change at week 26b	-3	-28	-33

a The intent-to-treat population includes all randomized patients. At week 26, the primary HbA1c endpoint was missing for 5.6%, 8.6% and 8.6% of patients randomized to placebo, RYBELSUS 7 mg and RYBELSUS 14 mg, respectively. Missing data were imputed by a pattern mixture model using multiple imputation (MI). Pattern was defined by randomized treatment and treatment status at week 26. During the trial, additional anti-diabetic medication was initiated as an add on to randomized treatment by 15%, 2% and 1% of patients randomized to placebo, RYBELSUS 7 mg and RYBELSUS 14 mg, respectively.

b Estimated using an ANCOVA model based on data irrespectively of discontinuation of trial product or initiation of rescue medication adjusted for baseline value and region.

c p<0.001 (unadjusted 2-sided) for superiority, controlled for multiplicity.

The mean baseline body weight was 88.6 kg, 89.0 kg and 88.1 kg in the placebo, RYBELSUS 7 mg, and RYBELSUS 14 mg arms, respectively. The mean changes from baseline to week 26 were -1.4 kg, -2.3 kg and

-3.7 kg in the placebo, RYBELSUS 7 mg and RYBELSUS 14 mg arms, respectively. The difference from placebo (95% CI) for RYBELSUS 7 mg was -0.9 kg (-1.9, 0.1) and for RYBELSUS 14 mg was -2.3 kg (-3.1, -1.5).

14.3 Combination Therapy Use of RYBELSUS in Patients with Type 2 Diabetes Mellitus

Combination with Metformin

In a 26-week trial (NCT02863328), 822 adult patients with type 2 mellitus diabetes were randomized to RYBELSUS 14 mg once daily or empagliflozin 25 mg once daily, all in combination with metformin. Patients had a mean age of 58 years and 50% were men. The mean duration of type 2 diabetes mellitus was 7.4 years and the mean BMI was 33 kg/m2. Overall, 86% were White, 7% were Black or African American and 6% were Asian; 24% identified as Hispanic or Latino ethnicity.

Treatment with RYBELSUS 14 mg once daily for 26 weeks resulted in a statistically significant reduction in HbA1c compared to empagliflozin 25 mg once daily (see Table 5).

Table 5. Results at Week 26 in a Trial of RYBELSUS Compared to Empagliflozin in Adult Patients with Type 2 Diabetes Mellitus in Combination with Metformin

	RYBELSUS 14 mg	Empagliflozin 25 mg
Intent-to-Treat (ITT) Population (N)a	411	410
HbA1c (%)
Baseline (mean)	8.1	8.1
Change at week 26b	-1.3	-0.9
Difference from empagliflozinb [95% CI]	-0.4 [-0.6, -0.3]c
Patients (%) achieving HbA1c <7%	67	40
FPG (mg/dL)
Baseline (mean)	172	174
Change at week 26b	-36	-36

a The intent-to-treat population includes all randomized patients. At week 26, the primary HbA1c endpoint was missing for 4.6% and 3.7% of patients randomized to RYBELSUS 14 mg and empagliflozin 25 mg, respectively. Missing data were imputed by a pattern mixture model using multiple imputation (MI). Pattern was defined by randomized treatment and treatment status at week 26. During the trial, additional anti-diabetic medication was initiated as an add on to randomized treatment by 1.9% and 1.2% of patients randomized to RYBELSUS 14 mg and empagliflozin 25 mg, respectively.

b Estimated using an ANCOVA based on data irrespectively of discontinuation of trial product or initiation of rescue medication adjusted for baseline value and region.

c p<0.001 (unadjusted 2-sided) for superiority, controlled for multiplicity.

The mean baseline body weight was 91.9 kg and 91.3 kg in the RYBELSUS 14 mg and empagliflozin 25 mg arms, respectively. The mean changes from baseline to week 26 were -3.8 kg and -3.7 kg in the RYBELSUS 14 mg and empagliflozin 25 mg arms, respectively. The difference from empagliflozin (95% CI) for RYBELSUS 14 mg was -0.1 kg (-0.7, 0.5).

Combination with Metformin or Metformin with Sulfonylurea

In a 26-week, double-blind trial (NCT02607865), 1864 adult patients with type 2 mellitus diabetes on metformin alone or metformin with sulfonylurea were randomized to RYBELSUS 3 mg, RYBELSUS 7 mg, RYBELSUS 14 mg or sitagliptin 100 mg once daily. Patients had a mean age of 58 years and 53% were men. The mean duration of type 2 diabetes mellitus was 8.6 years, and the mean BMI was 32 kg/m2. Overall, 71% were White, 9% were Black or African American and 13% were Asian; 17% identified as Hispanic or Latino ethnicity.

Treatment with RYBELSUS 7 mg and RYBELSUS 14 mg once daily for 26 weeks resulted in a statistically significant reduction in HbA1c compared to sitagliptin 100 mg once daily (see Table 6).

Table 6. Results at Week 26 in a Trial of RYBELSUS Compared to Sitagliptin 100 mg Once Daily in Adult Patients with Type 2 Diabetes Mellitus in Combination with Metformin or Metformin with Sulfonylurea

	RYBELSUS 7 mg	RYBELSUS 14 mg	Sitagliptin 100 mg
Intent-to-Treat (ITT) Population (N)a	465	465	467
HbA1c (%)
Baseline (mean)	8.4	8.3	8.3
Change at week 26b	-1	-1.3	-0.8
Difference from sitagliptinb [95% CI]	-0.3 [-0.4; -0.1]c	-0.5 [-0.6; -0.4]c
Patients (%) achieving HbA1c <7%	44	56	32
FPG (mg/dL)
Baseline (mean)	170	168	172
Change at week 26b	-21	-31	-15

a The intent-to-treat population includes all randomized patients. At week 26, the primary HbA1c endpoint was missing for 5.8%, 6.2% and 4.5% of patients randomized to RYBELSUS 7 mg, RYBELSUS 14 mg and sitagliptin 100 mg, respectively. Missing values were imputed by a pattern mixture model using multiple imputation (MI). Pattern was defined by randomized treatment and treatment status at week 26. During the trial, additional anti-diabetic medication was initiated as an add on to randomized treatment by 2.4%, 1.1% and 2.8% of patients randomized to RYBELSUS 7 mg, RYBELSUS 14 mg and sitagliptin 100 mg, respectively.

b Estimated using an ANCOVA based on data irrespectively of discontinuation of trial product or initiation of rescue medication adjusted for baseline value, background medication and region.

c p<0.001 (unadjusted 2-sided) for superiority, controlled for multiplicity.

The mean baseline body weight was 91.3 kg, 91.2 kg and 90.9 kg in the RYBELSUS 7 mg, RYBELSUS 14 mg and sitagliptin 100 mg arms, respectively. The mean changes from baseline to week 26 were -2.2 kg, -3.1 kg and -0.6 kg in the RYBELSUS 7 mg, RYBELSUS 14 mg and sitagliptin 100 mg arms, respectively. The difference from sitagliptin (95% CI) for RYBELSUS 7 mg was -1.6 kg (-2.0, -1.1) and RYBELSUS 14 mg was -2.5 kg (-3.0, -2.0).

Combination with Metformin or Metformin with SGLT-2 Inhibitors

In a 26-week, double-blind, double-dummy trial (NCT02863419), 711 adult patients with type 2 diabetes mellitus on metformin alone or metformin with SGLT-2 inhibitors were randomized to RYBELSUS 14 mg once daily, liraglutide 1.8 mg subcutaneous injection once daily or placebo. Patients had a mean age of 56 years and 52% were men. The mean duration of type 2 diabetes mellitus was 7.6 years and the mean BMI was 33 kg/m2. Overall, 73% were White, 4% were Black or African American and 13% were Asian; 6% identified as Hispanic or Latino ethnicity.

Treatment with RYBELSUS 14 mg once daily for 26 weeks resulted in statistically significant reductions in HbA1c compared to placebo. Treatment with RYBELSUS 14 mg once daily for 26 weeks resulted in non-inferior reductions in HbA1c compared to liraglutide 1.8 mg (see Table 7).

Table 7. Results at Week 26 in a Trial of RYBELSUS Compared to Liraglutide and Placebo in Adult Patients with Type 2 Diabetes Mellitus in Combination with Metformin or Metformin with SGLT-2i

	Placebo	RYBELSUS 14 mg	Liraglutide 1.8 mg
Intent-to-Treat (ITT) Population (N)a	142	285	284
HbA1c (%)
Baseline (mean)	7.9	8	8
Change at week 26b	-0.2	-1.2	-1.1
Difference from placebob [95% CI]		-1.1 [-1.2; -0.9]c
Difference from liraglutideb [95% CI]		-0.1 [-0.3; 0]
Patients (%) achieving HbA1c <7%	14	68	62
FPG (mg/dL)
Baseline (mean)	167	167	168
Change at week 26b	-7	-36	-34

a The intent-to-treat population includes all randomized patients. At week 26, the primary HbA1c endpoint was missing for 5.6%, 4.2% and 2.5% of patients randomized to placebo, liraglutide 1.8 mg and RYBELSUS 14 mg, respectively. Missing values were imputed by a pattern mixture model using multiple imputation (MI). Pattern was defined by randomized treatment and treatment status at week 26. During the trial, additional anti-diabetic medication was initiated as an add on to randomized treatment by 7.7%, 3.2% and 3.5% of patients randomized to placebo, liraglutide 1.8 mg and RYBELSUS 14 mg respectively.

b Estimated using an ANCOVA based on data irrespectively of discontinuation of trial product or initiation of rescue medication adjusted for baseline value, background medication and region.

c p<0.001 (unadjusted 2-sided) for superiority, controlled for multiplicity.

The mean baseline body weight was 93.2 kg, 95.5 kg and 92.9 kg in the placebo, liraglutide 1.8 mg and RYBELSUS 14 mg arms, respectively. The mean changes from baseline to week 26 were -0.5 kg, -3.1 kg and

-4.4 kg in the placebo, liraglutide 1.8 mg and RYBELSUS 14 mg arms, respectively. The difference from placebo (95% CI) for RYBELSUS 14 mg was -3.8 kg (-4.7, -3.0). The difference from liraglutide 1.8 mg for RYBELSUS 14 mg was -1.2 (-1.9, -0.6).

Combination in patients with Type 2 Diabetes Mellitus and Moderate Renal Impairment with Metformin alone, Sulfonylurea alone, Basal Insulin alone or Metformin in Combination with either Sulfonylurea or Basal Insulin

In a 26-week, double-blind trial (NCT02827708), 324 adult patients with moderate renal impairment (eGFRCKD-EPI 30-59 mL/min/1.73 m2) were randomized to RYBELSUS 14 mg or placebo once daily. RYBELSUS was added to the patient’s stable pre-trial antidiabetic regimen. The insulin dose was reduced by 20% at randomization for patients on basal insulin. Dose reduction of insulin and sulfonylurea was allowed in case of hypoglycemia; up titration of insulin was allowed but not beyond the pre-trial dose.

Patients had a mean age of 70 years and 48% were men. The mean duration of type 2 diabetes mellitus was 14 years and the mean BMI was 32 kg/m2. Overall, 96% were White, 4% were Black or African American and 0.3% were Asian; 6.5% identified as Hispanic or Latino ethnicity. 39.5% of patients had an eGFR value of 30 to 44 mL/min/1.73 m2.

Treatment with RYBELSUS 14 mg once daily for 26 weeks resulted in a statistically significant reduction in HbA1c from baseline compared to placebo (see Table 8).

Table 8. Results at Week 26 in a Trial of RYBELSUS Compared to Placebo in Patients with Moderate Renal Impairment

	Placebo	RYBELSUS 14 mg
Intent-to-Treat (ITT) Population (N)a	161	163
HbA1c (%)
Baseline (mean)	7.9	8
Change at week 26b	-0.2	-1
Difference from placebob [95% CI]		-0.8 [-1.0; -0.6]c
Patients (%) achieving HbA1c <7%	23	58
FPG (mg/dL)
Baseline (mean)	164	164
Change at week 26b	-7	-28

a The intent-to-treat population includes all randomized patients including patients on rescue medication. At week 26, the primary HbA1c endpoint was missing for 3.7% and 5.5% of patients randomized to placebo and RYBELSUS 14 mg, respectively. Missing values were imputed by a pattern mixture model using multiple imputation (MI). Pattern was defined by randomized treatment and treatment status at week 26. During the trial, additional anti-diabetic medication was initiated as an add on to randomized treatment by 10% and 4.3% of patients randomized to placebo and RYBELSUS 14 mg, respectively.

b Estimated using an ANCOVA based on data irrespectively of discontinuation of trial product or initiation of rescue medication adjusted for baseline value, background medication, renal status and region.

c p<0.001 (unadjusted 2-sided) for superiority, controlled for multiplicity.

The mean baseline body weight was 90.4 kg and 91.3 kg in the placebo and RYBELSUS 14 mg arms, respectively. The mean changes from baseline to week 26 were -0.9 kg and -3.4 kg in the placebo and RYBELSUS 14 mg arms, respectively. The difference from placebo (95% CI) for RYBELSUS 14 mg was -2.5 kg (-3.2, -1.8).

Combination with Insulin with or without Metformin

In a 26-week double blind trial (NCT03021187), 731 adult patients with type 2 diabetes mellitus inadequately controlled on insulin (basal, basal/bolus or premixed) with or without metformin, were randomized to RYBELSUS 3 mg, 7 mg and 14 mg once daily or placebo once daily. All patients reduced their insulin dose by 20% at randomization to reduce the risk of hypoglycemia. Patients were allowed to increase the insulin dose only up to the starting insulin dose prior to randomization.

Patients had a mean age of 61 years and 54% were men. The mean duration of type 2 diabetes mellitus was 15 years and the mean BMI was 31 kg/m2. Overall, 51% were White, 7% were Black or African American and 36% were Asian; 13% identified as Hispanic or Latino ethnicity.

Treatment with RYBELSUS 7 mg and 14 mg once daily for 26 weeks resulted in a statistically significant reduction in HbA1c from baseline compared to placebo once daily (see Table 9).

Table 9. Results at Week 26 in a Trial of RYBELSUS Compared to Placebo in Adult Patients with Type 2 Diabetes Mellitus in Combination with Insulin alone or with Metformin

	Placebo	RYBELSUS 7 mg	RYBELSUS 14 mg
Intent-to-Treat (ITT) Population (N)a	184	182	181
HbA1c (%)
Baseline (mean)	8.2	8.2	8.2
Change at week 26b	-0.1	-0.9	-1.3
Difference from placebob [95% CI]		-0.9 [-1.1; -0.7]c	-1.2 [-1.4; -1.0]c
Patients (%) achieving HbA1c <7%	7	43	58
FPG (mg/dL)
Baseline (mean)	150	153	150
Change at week 26b	5	-20	-24

a The intent-to-treat population includes all randomized patients. At week 26, the primary HbA1c endpoint was missing for 4.3%, 4.4% and 4.4% of patients randomized to placebo, RYBELSUS 7 mg and RYBELSUS 14 mg, respectively. Missing values were imputed by a pattern mixture model using multiple imputation (MI). Pattern was defined by randomized treatment and treatment status at week 26. During the trial, additional anti-diabetic medication was initiated as an add on to randomized treatment by 4.9%, 1.1 % and 2.2% of patients randomized to placebo, RYBELSUS 7 mg and RYBELSUS 14 mg, respectively.

b Estimated using an ANCOVA based on data irrespectively of discontinuation of trial product or initiation of rescue medication adjusted for baseline value, background medication and region.

c p<0.001 (unadjusted 2-sided) for superiority, controlled for multiplicity.

The mean baseline body weight was 86.0 kg, 87.1 kg and 84.6 kg in the placebo, RYBELSUS 7 mg and RYBELSUS 14 mg arms, respectively. The mean changes from baseline to week 26 were -0.4 kg, -2.4 kg and

-3.7 kg in the placebo, RYBELSUS 7 mg and RYBELSUS 14 mg arms, respectively. The difference from placebo (95% CI) for RYBELSUS 7 mg was -2.0 kg (-3.0, -1.0) and for RYBELSUS 14 mg was -3.3 kg (-4.2, -2.3).

14.4 Cardiovascular Outcomes Trial in Patients with Type 2 Diabetes Mellitus and Cardiovascular Disease

PIONEER 6 (NCT02692716) was a multi-center, multi-national, placebo-controlled, double-blind trial. In this trial, 3,183 adult patients with inadequately controlled type 2 diabetes mellitus and atherosclerotic cardiovascular disease were randomized to RYBELSUS 14 mg once daily or placebo for a median observation time of 16 months. The trial compared the risk of a Major Adverse Cardiovascular Event (MACE) between RYBELSUS 14 mg and placebo when these were added to and used concomitantly with standard of care treatments for diabetes and cardiovascular disease. The primary endpoint, MACE, was the time to first occurrence of a three-part composite outcome which included cardiovascular death, non-fatal myocardial infarction and non-fatal stroke.

Patients eligible to enter the trial were 50 years of age or older and had established, stable, cardiovascular, cerebrovascular, peripheral artery disease, chronic kidney disease or NYHA class II and III heart failure or were 60 years of age or older and had other specified risk factors for cardiovascular disease. In total, 1,797 patients (56.5%) had established cardiovascular disease without chronic kidney disease, 354 patients (11.1%) had chronic kidney disease only, and 544 patients (17.1%) had both cardiovascular disease and kidney disease; 488 patients (15.3%) had cardiovascular risk factors without established cardiovascular disease or chronic kidney disease. The mean age at baseline was 66 years and 68% were men. The mean duration of diabetes was 14.9 years and mean BMI was 32 kg/m2. Overall, 72% were White, 6% were Black or African American and 20% were Asian; 16% identified as Hispanic or Latino ethnicity. Concomitant diseases of patients in this trial included, but were not limited to, heart failure (12%), history of ischemic stroke (8%) and history of a myocardial infarction (36%). In total, 99.7% of the patients completed the trial and the vital status was known at the end of the trial for 100%.

For the primary analysis, a Cox proportional hazards model was used to test for non-inferiority of RYBELSUS 14 mg to placebo for time to first MACE using a risk margin of 1.3. Type-1 error was controlled across multiple tests using a hierarchical testing strategy. Non‑inferiority to placebo was established, with a hazard ratio equal to 0.79 (95% CI: 0.57, 1.11) over the median observation time of 16-months. The proportion of patients who experienced at least one MACE was 3.8% (61/1591) for RYBELSUS 14 mg and 4.8% (76/1592) for placebo.