Methods of using and compositions containing dulaglutide

Abstract

The present invention relates to methods of using new doses of dulaglutide and compositions containing such higher doses of dulaglutide.

Claims

1. A method of improving glycemic control in a subject in need thereof with type 2 diabetes comprising: a) administering to said subject an initial dose of 0.75 mg of dulaglutide once weekly; b) increasing the dose to 1.5 mg of dulaglutide once weekly; c) increasing the dose to 3.0 mg of dulaglutide once weekly after at least 4 weeks on the 1.5 mg dose; and d) increasing the dose to 4.5 mg of dulaglutide once weekly after at least 4 weeks on the 3.0 mg dose.

2. The method of claim 1, wherein the dose increases of each of steps b) through d) are performed when there is a need to reduce the subject's HbA1c level.

3. The method of claim 2, wherein administration of 4.5 mg of dulaglutide once weekly results in a greater body weight reduction than administration of 1.5 mg of dulaglutide once weekly.

Description

EXAMPLES

(1) Phase 2 Clinical Study.

(2) A Phase 2 clinical trial is designed to assess the safety and efficacy of once weekly dulaglutide 3.0 mg and 4.5 mg, administered following one of two dose titration algorithms, in comparison to placebo in patients with type 2 diabetes mellitus (T2D) treated with metformin monotherapy. The trial is also designed to include exploratory comparisons of 3.0 and 4.5 mg doses to dulaglutide 1.5 mg (the highest dose approved by regulatory agencies). The trial is intended to predict whether the increased doses will provide improved clinical benefits, including greater reduction in HbA1c and greater body weight reduction, with an acceptable safety and tolerability profile.

(3) The study is designed as a multicenter, randomized, double-blind, parallel-arm, placebo-controlled trial with 3 study periods (lead-in, treatment, and safety follow-up) in patients who have T2D with inadequate glycemic control on metformin only.

(4) After screening and a lead-in period, patients are randomized in a 1:1:1:1 ratio to weekly injections of dulaglutide 4.5 mg, 3.0 mg, or 1.5 mg, or placebo, in combination with stable doses of metformin. Within each of the investigational dulaglutide dose (3.0 mg and 4.5 mg) arms, patients are randomly assigned in a 1:1 ratio to one of two dulaglutide dose titration algorithms and participants are treated for 18 weeks after randomization in a double-blind manner, comprising 6 weeks of a titration phase and a 12 week maintenance treatment phase at the final dose. During the titration phase in the 3.0 and 4.5 mg arms, the dose of dulaglutide is escalated over 6 weeks according to one of two algorithms: (1) patients receive dulaglutide 1.5 mg once weekly for the first 4 weeks (Weeks 1-4) followed by dulaglutide 3.0 mg once weekly for the next 2 weeks (Weeks 5-6) (hereafter, “Algorithm 1” or “A1”); or (2) patients received dulaglutide 0.75 mg once weekly for the first 2 weeks (Weeks 1-2) followed by dulaglutide 1.5 mg once weekly for the next 4 weeks (Weeks 3-6) (hereafter, “Algorithm 2” or “A2”). The 2 titration algorithms are chosen based on modeling and simulations including data from several Phase 2 and Phase 3 trials to assess the potential mitigating effect of doubling increases in titrated doses (i.e., Algorithm 1) versus longer exposure at lower titration doses (i.e., Algorithm 2).

(5) A total of 505 patients are screened and 318 patients were randomized to treatment: placebo, 82; dulaglutide 1.5 mg, 81; dulaglutide 3.0 mg, 79; dulaglutide 4.5 mg, 76. One patient randomized to placebo withdrew consent to participate in the study at Visit 4 and did not receive any doses of study drug, so 317 patients received at least 1 dose of study drug and comprised the intent-to-treat (ITT) Population: placebo, 81; dulaglutide 1.5 mg, 81; dulaglutide 3.0 mg, 79; dulaglutide 4.5 mg, 76.

(6) A summary of the changes from baseline in HbA1c (%) and body weight (kg) at Week 18 in the ITT population, excluding data collected post-rescue (patients that had high blood glucose requiring rescue with another therapy) and post study drug discontinuation is provided below in Table 1. The HbA1c data also excludes results for patients demonstrating large, unexplained swings in HbA1c considered to be physiologically implausible and not consistent with the other clinical information available.

(7) TABLE-US-00002 TABLE 1 Change in HbA1c (%) and body weight (kg) at Week 18, ITT population excluding post rescue or study drug discontinuation data. LS Mean Change from Baseline LS Mean Differences (SE) [n] (95% CI) Dula Dula Dula Dula 1.5- Dula 3.0- Dula 4.5- Parameter PBO 1.5 3.0 4.5 PBO Dula 1.5 Dula 1.5 Change in −0.42 −1.24** −1.47** −1.50** −0.82** −0.22 −0.26 HbA1c (%) (0.097) (0.094) (0.097) (0.101) (−1.08, (−0.48, (−0.52, [66] [72] [67] [58] −0.56) 0.03) 0.00) Change in −1.6 −2.9* −4.2**.sup.† −4.4**.sup.† −1.3* −1.3.sup.† −1.5.sup.† body (0.41) (0.40) (0.41) (0.43) (−2.4, (−2.4, (−2.6, weight (kg) [66] [71] [68] [59] −0.2) −0.2) −0.3) Abbreviations: CI = confidence interval; LS mean = least-squares mean; PBO = placebo. *P-value for comparison of dulaglutide vs. placebo <0.05. **P-value for comparison of dulaglutide vs. placebo <0.001. .sup.†P-value for comparison of dulaglutide vs. dulaglutide 1.5 mg <0.05. Notes: Dula X.X represents X.X milligrams of dulaglutide administered once weekly.

(8) At week 18, all three doses of dulaglutide reduced HbA1c significantly from baseline compared to placebo and significantly reduced body weight from baseline compared to placebo. HbA1c declined by a mean of −1.24% in the dulaglutide 1.5 mg group, compared with −1.47% in the dulaglutide 3.0 mg group (LS mean treatment difference, −0.22%; 95% CI−0.48%, 0.03%) and −1.50% in the dulaglutide 4.5 mg group (LS mean treatment difference, −0.26%; 95% CI−0.52%, 0.00%). Body weight declined by a mean of −2.9 kg in the dulaglutide 1.5 mg group, compared with −4.2 kg in the dulaglutide 3.0 mg group (LS mean treatment difference, −1.3 kg; 95% CI−2.4, 0.2) and −4.4 kg in the dulaglutide 4.5 mg group (LS mean treatment difference, −1.5 kg; 95% CI−2.6, −0.3). The incremental efficacy on HbA1c reduction of both higher doses relative to dulaglutide 1.5 mg was greater in subgroups of patients with higher baseline HbA1c levels.

(9) Table 2 summarizes frequency of nausea, vomiting, and diarrhea by treatment, as measured by prevalence (any patient who had a new or ongoing event during the interval), incidence (any patient who had an event begin during the interval), and first onset (any patient who had the first event of that type during the interval) of the events over 3 time periods (Weeks 0-6 [titration period], Weeks 6-10 [first 4 weeks after completion of titration for patients in the dulaglutide 3.0 mg and dulaglutide 4.5 mg groups], and Weeks 0-18 [treatment period]).

(10) TABLE-US-00003 TABLE 2 Summary of Prevalence, Incidence, and First Onset of Nausea, Vomiting, and Diarrhea by Treatment and Time Period, Intent-to-Treat Population. Event/ Placebo Dula 1.5 Dula 3.0 Dula 4.5 Parameter/ (N = 81) (N = 81) (N = 79) (N = 76) Time Period m/M (%) m/M (%) m/M (%) m/M (%) Nausea Prevalence Wks 0-6 3/81 (3.7)  16/81 (19.8)  16/79 (20.3) 17/76 (22.4) Wks 6-10 1/81 (1.2)  9/81 (11.1)  8/78 (10.3) 14/75 (18.7) Wks 0-18 4/81 (4.9)  18/81 (22.2)  19/79 (24.1) 23/76 (30.3) Incidence Wks 0-6 3/81 (3.7)  16/81 (19.8)  16/79 (20.3) 17/76 (22.4) Wks 6-10 1/81 (1.2)  6/81 (7.4)  5/78 (6.4) 10/75 (13.3) Wks 0-18 4/81 (4.9)  18/81 (22.2)  19/79 (24.1) 23/76 (30.3) Onset Wks 0-6 3/81 (3.7)  16/81 (19.8)  16/79 (20.3) 17/76 (22.4) Wks 6-10 1/81 (1.2)  2/81 (2.5)  3/78 (3.8) 5/75 (6.7) Wks 0-18 4/81 (4.9)  18/81 (22.2)  19/79 (24.1) 23/76 (30.3) Vomiting Prevalence Wks 0-6 2/81 (2.5)  7/81 (8.6)  7/79 (8.9) 5/76 (6.6) Wks 6-10 3/81 (3.7)  2/81 (2.5)  2/78 (2.6) 4/75 (5.3) Wks 0-18 4/81 (4.9)  9/81 (11.1)  8/79 (10.1) 10/76 (13.2) Incidence Wks 0-6 2/81 (2.5)  7/81 (8.6)  7/79 (8.9) 5/76 (6.6) Wks 6-10 2/81 (2.5)  2/81 (2.5)  2/78 (2.6) 3/75 (4.0) Wks 0-18 4/81 (4.9)  9/81 (11.1)  8/79 (10.1) 10/76 (13.2) Onset Wks 0-6 2/81 (2.5)  7/81 (8.6)  7/79 (8.9) 5/76 (6.6) Wks 6-10 2/81 (2.5)  2/81 (2.5)  1/78 (1.3) 3/75 (4.0) Wks 0-18 4/81 (4.9)  9/81 (11.1)  8/79 (10.1) 10/76 (13.2) Diarrhea Prevalence Wks 0-6 5/81 (6.2)  6/81 (7.4)  11/79 (13.9)  9/76 (11.8) Wks 6-10 2/81 (2.5)  4/81 (4.9)  10/78 (12.8)  9/75 (12.0) Wks 0-18 9/81 (11.1) 9/81 (11.1) 19/79 (24.1) 15/76 (19.7) Incidence Wks 0-6 5/81 (6.2)  6/81 (7.4)  11/79 (13.9)  9/76 (11.8) Wks 6-10 1/81 (1.2)  3/81 (3.7)  5/78 (6.4) 7/75 (9.3) Wks 0-18 9/81 (11.1) 9/81 (11.1) 19/79 (24.1) 15/76 (19.7) Onset Wks 0-6 5/81 (6.2)  6/81 (7.4)  11/79 (13.9)  9/76 (11.8) Wks 6-10 1/81 (1.2)  2/81 (2.5)  5/78 (6.4) 5/75 (6.7) Wks 0-18 9/81 (11.1) 9/81 (11.1) 19/79 (24.1) 15/76 (19.7) Abbreviations: m = number of patients with event during the interval; M = number of patients with data during the interval; N = number of patients randomized and treated; Wks = weeks.

(11) Frequency of nausea and vomiting was higher in all three dulaglutide groups compared to placebo during all 3 time intervals. There was a dose-response in the frequency of nausea overall (Weeks 0-18) across the dulaglutide groups, with the greatest frequency in the dulaglutide 4.5 mg group (30.3%). In each dulaglutide treatment group, nausea was most frequent during the titration period (Week 0-6) and decreased from Weeks 6-10 and beyond. Prevalence of vomiting ranged from 6.6% to 8.9% across the dulaglutide groups during the titration period, and decreased during the Week 6-10 period. During all 3 time intervals, frequency of diarrhea was generally similar in the placebo and dulaglutide 1.5 mg groups and higher in the higher-dose groups. More frequent reporting of diarrhea in the dulaglutide 3.0 mg and 4.5 mg groups during the initial 6 weeks continued during the Week 6-10 period, and decreased thereafter. Incidence of diarrhea was greater in the higher-dose groups compared to the 1.5 mg group during the titration period before patients received the 3.0 mg or 4.5 mg doses; therefore, it appears that the differences were not entirely dose-related. In terms of incidence of severe nausea, diarrhea, and vomiting, few patients had any severe events, and incidence was generally similar in all four treatment groups.

(12) Although the incidence of nausea, vomiting, and diarrhea was increased in patients treated with dulaglutide in a dose-dependent manner, and the proportion of patients discontinuing study treatment due to adverse events was higher in comparison to the dulaglutide 1.5 mg group, the difference in incidence between the high dose and 1.5 mg groups were modest, and the frequency of relevant GI events and treatment discontinuations was lower than those observed in completed dulaglutide trials of shorter duration that included patients who received nontitrated doses ≥3.0 mg, indicating the titration algorithms A1 and A2 had a beneficial effect.

(13) The number of patients randomized by titration algorithm in the dulaglutide 3.0 mg and 4.5 mg groups are as follows: dulaglutide 3.0 mg A1, 41; dulaglutide 3.0 mg A2, 38; dulaglutide 4.5 mg A1, 39; dulaglutide 4.5 mg A2, 37. Table 3 summarizes frequency of nausea, vomiting, and diarrhea by dose and titration algorithm for the 2 higher dose groups, as measured by prevalence, incidence, and first onset of the events over 3 key time periods (Weeks 0-6 [titration period], Weeks 6-10, and Weeks 0-18 [treatment period]).

(14) TABLE-US-00004 Summary of Prevalence, Incidence, and First Onset of Nausea, Vomiting, and Diarrhea by Dose and Titration Algorithm (Dulaglutide 3.0 mg and Dulaglutide 4.5 mg) and Time Period, Intent-to-Treat Population. Abbreviations: Al = algorithm 1; A2 = algorithm 2; m = number of patients with event during the interval; M = number of patients with data during the interval; N = number of patients randomized and treated; Wks = weeks. Prevalence counts any patient who had a new or ongoing event during the interval. Incidence counts any patient who had an event begin during the interval. Onset counts any patient who had the first event of that type during the interval. Event/ Treatment/Algorithm Groups Parameter/ 3.0 A1 3.0 A2 4.5 A1 4.5 A2 Algorithm Groups Time (N = 41) (N = 38) (N = 39) (N = 37) A1 (N = 80) A2 (N = 75) Period m/M (%) m/M (%) m/M (%) m/M (°%) m/M (%) m/M (%) Nausea Prevalence Wks 0-6 11/41 (26.8) 5/38(13.2) 7/39(17.9) 10/37 (27.0) 18/80 (22.5) 15/75 (20.0) Wks 6-10 5/40(12.5) 3/38 (7.9) 6/38(15.8) 8/37(21.6) 11/78 (14.1) 11/75(14.7) Wks 0-18 12/41 (29.3) 7/38(18.4) 11/39 (28.2) 12/37 (32.4) 23/80 (28.8) 19/75 (25.3) Incidence Wks 0-6 11/41 (26.8) 5/38(13.2) 7/39(17.9) 10/37(27.0) 18/80 (22.5) 15/75 (20 0) Wks 6-10 2/40 (5.0) 3/38 (7.9) 6/38(15.8) 4/37(10.8) 8/78 (10.3) 7/75 (9.3) Wks 0-18 12/41 (29.3) 7/38(18.4) 11/39 (28.2) 12/37(32.4) 23/80 (28.8) 19/75 (25.3) Onset Wks 0-6 11/41 (26.8) 5/38(13.2) 7/39(17.9) 10/37(27.0) 18/80 (22.5) 15/75 (20.0) Wks 6-10 1/40 (2.5) 2/38 (5.3) 3/38 (7.9) 2/37 (5.4) 4/78(5.1) 4/75 (5.3) Wks 0-18 12/41 (29.3) 7/38(18.4) 11/39 (28.2) 12/37(32.4) 23/80 (28.8) 19/75 (25.3) Vomiting Prevalence Wks 0-6 6/41 (14.6) 1/38 (2.6) 3/39 (7.7) 2/37 (5.4) 9/80(11.3) 3/75 (4.0) Wks 6-10 1/40 (2.5) 1/38(2.6) 0/38 4/37(10.8) 1/78(1.3) 5/75 (6.7) Wks 0-18 6/41 (14.6) 2/38(5.3) 4/39(10.3) 6/37(16.2) 10/80(12.5) 8/75(10.7) Incidence Wks 0-6 6/41 (14.6) 1/38(2.6) 3/39 (7.7) 2/37 (5.4) 9/80(11.3) 3/75 (4.0) Wks 6-10 1/40 (2.5) 1/38(2.6) 0/38 3/37(8.1) 1/78(1.3) 4/75 (5.3) Wks 0-18 6/41 (14.6) 2/38 (5.3) 4/39(10.3) 6/37(16.2) 10/80(12.5) 8/75(10.7) Onset Wks 0-6 6/41 (14.6) 1/38(2.6) 3/39 (7.7) 2/37 (5.4) 9/80(11.3) 3/75 (4.0) Wks 6-10 0/40 1/38(2.6) 0/38 3/37 (8.1) 0/78 4/75 (5.3) Wks 0-18 6/41 (14.6) 2/38 (5.3) 4/39(10.3) 6/37(16.2) 10/80(12.5) 8/75(10.7) Diarrhea Prevalence Wks 0-6 7/41 (17.1) 4/38(10.5) 6/39(15.4) 3/37(8.1) 13/80 (16.3) 7/75 (9.3) Wks 6-10 4/40(10.0) 6/38(15.8) 5/38(13.2) 4/37(10.8) 9/78(11.5) 10/75(13.3) Wks 0-18 8 41 (19 5) 11/38 (28.9) 9/39(23.1) 6/37(16.2) 17/80 (21.3) 17/75 (22.7) Incidence Wks 0-6 7/41 (17.1) 4/38(10.5) 6/39(15.4) 3/37(8.1) 13/80(16.3) 7/75(9.3) Wks 6-10 0/40 5/38 (13.2) 5/38(13.2) 2/37 (5.4) 5/78 (6.4) 7/75 (9.3) Wks 0-18 8/41 (19.5) 11/38 (28.9) 9/39(23.1) 6/37(16.2) 17/80 (21.3) 17/75 (22.7) Onset Wks 0-6 7/41 (17.1) 4/38 (10.5) 6/39 (15 4) 3/37(8.1) 13/80 (16.3) 7/75(9.3)

(15) The frequency of nausea is similar with A1 and A2 overall; there is a difference within the 3.0 mg group, with lower frequency within the A2 subgroup compared to the A1 subgroup, but this difference is considered unlikely to be related to the titration algorithms, as suggested by similar frequencies within the 4.5 mg group A1 and A2 subgroups. Most of the events of nausea occur during the titration period, with few new events reported upon uptitration to final doses (Weeks 6-10).

(16) The frequency of vomiting is similar with A1 and A2 overall; more patients in the A1 subgroup than the A2 subgroup had events within the dulaglutide 3.0 mg group and more patients had events in the A2 subgroup than in the A1 subgroup within the dulaglutide 4.5 mg group, but, similar to nausea as described above, these differences are considered unrelated to the titration algorithms, since differences between the A1 and A2 subgroups in the 3.0 mg and 4.5 mg groups were in the opposite direction. Most of the events of vomiting with A1 occur during the titration period, with few new events reported upon uptitration to final doses (Weeks 6-10). With A2, the frequency of vomiting was lower during the titration period, remained at similar levels during the Week 6-10 period, and started decreasing after Week 10 and remained lower up to Week 18.

(17) Prevalence and incidence of diarrhea during the titration period (Weeks 0-6) were higher in the dulaglutide 3.0 mg A1 and dulaglutide 4.5 mg A1 subgroups and lower in the dulaglutide 3.0 mg A2 and 4.5 mg A2 subgroups. During Weeks 6-10, prevalence and incidence were lower in the dulaglutide 3.0 mg A1 and dulaglutide 4.5 mg A2 subgroups and higher in the dulaglutide 3.0 mg A2 and dulaglutide 4.5 mg A1 subgroups.

(18) The observed pattern of incidence of these events in A1 versus A2 suggest that starting with a lower dulaglutide dose (dulaglutide 0.75 mg [used in A2] vs. dulaglutide 1.5 mg [used in A1]) may reduce GI tolerability events during the titration period. Slower reduction in frequency of these events in the A2 group versus the A1 group during the final uptitration phase (Weeks 6-10) suggests a longer titration period may be needed to allow for development of tachyphylaxis to GI side effects of dulaglutide.

(19) As noted above, earlier studies on doses of 3.0 mg and greater resulted in increases in pulse rate which contributed, in part, to discontinuation of the 3.0 mg arm in an earlier study. Thus, changes in pulse rate were also measured in the present study, and the data collected are provided in Table 4 below:

(20) TABLE-US-00005 TABLE 4 Change from baseline in heart rate by treatment in ITT population. Time LS mean change in PR (bpm) (95% CI) [n] (wks) Placebo Dula 1.5 Dula 3.0 Dula 4.5 2  0.22 (−1.50, 1.95)  1.77 (0.03, 3.50)  1.87 (0.12, 3.62) 1.91 (0.12, 3.70) [80] [80] [79] [76] 4 −0.98 (−2.62, 0.66)  1.09 (−0.57, 2.75)  1.38 (−0.28, 3.05) 1.97 (0.25, 3.69) [79] [78] [79] [74] 6 −1.97 (−3.71, −0.23)  0.52 (−1.22, 2.27)  1.72 (−0.04, 3.47) 2.07 (0.27, 3.87) [76] [77] [77] [74] 7 −0.85 (−2.45, 0.75)  2.13 (0.50, 3.75)  2.37 (0.75, 4.00) 3.94 (2.27, 5.61) [77] [76] [77] [74] 8 −0.06 (−1.90, 1.79)  1.83 (−0.03, 3.68)  1.86 (−0.00, 3.73) 4.70 (2.78, 6.62) [76] [76] [76] [72] 9  0.04 (−1.86, 1.77)  0.79 (−1.04, 2.62)  2.28 (0.44, 4.11) 3.85 (1.96, 5.75) [76] [75] [76] [71] 10 −1.80 (−3.53, −0.08)  0.78 (−0.97, 2.52)  2.18 (0.43, 3.93) 2.65 (0.84, 4.46) [77] [76] [77] [71] 14  0.28 (−1.44, 2.00) −0.22 (−1.97, 1.53)  0.58 (−1.15, 2.31) 2.11 (0.29, 3.92) [76] [73] [78] [69] 18 −1.40 (−3.20, 0.41) −0.71 (−2.55, 1.12) −0.04 (−1.86, 1.78) 0.70 (−1.18, 2.58) [75] [72] [75] [70] Abbreviations: LS = least squares; PR = pulse rate; wks = weeks; CI = confidence interval; n = sample size.

(21) As indicated in Table 4 above, mean changes in pulse rate across dulaglutide groups were similar during the entire treatment period and at the end of the study were similar.

(22) In terms of determining the optimal dose titration strategy, patients who started treatment with dulaglutide 0.75 mg (for 2 weeks, A1) had fewer GI issues initially than patients who started on dulaglutide 1.5 mg (for 4 weeks, A2), but these differences were not maintained upon further uptitration to 1.5 mg with A1. Differences were also noted between A1 and A2 in the dulaglutide 4.5 mg group once the patients were uptitrated to their final doses (in the A1 arm from 3.0 mg to 4.5 mg; in the A2 arm from 1.5 mg to 4.5 mg), with tolerability modestly poorer in the patients in the A2 arm. These results suggest that further adjustments in the algorithms may be capable of optimizing dulaglutide titration with higher doses with respect to duration of titration and dose levels needed to further minimize the risk of tolerability issues.

(23) Phase 3 Study.

(24) A Phase 3, multicenter, randomized, double-blind, parallel-arm study is designed with 3 study periods (Lead-In, Treatment, and Safety Follow-up) in patients with T2D with inadequate glycemic control on metformin only; the study has a 52-week treatment duration, with primary endpoint at 36 weeks. The study is designed to assess the efficacy and safety of once weekly dulaglutide 3.0 mg and 4.5 mg in comparison to once weekly dulaglutide 1.5 mg.

(25) A minimum sample size of approximately 1800 patients assuming 15% dropout rate are enrolled (randomized) in order to obtain approximately 510 completers per arm at 36 weeks. At Visit 3, patients are randomized in a 1:1:1 ratio to weekly injections of dulaglutide 4.5 mg, 3.0 mg, or 1.5 mg, in combination with stable doses of metformin. The investigational doses of 4.5 and 3.0 mg are administered following a titration algorithm designed based on PK/PD modeling of nausea and vomiting events from the Phase 2 study described above, with the goal of further alleviating gastrointestinal adverse events. PK/PD exposure-response modeling of the Phase 2 study data also predicts that the daily incidence of nausea and vomiting at higher dulaglutide doses would be lowest in algorithms starting at a 0.75 mg dose with a slower dose escalation over an 8-week period, allowing adequate time for tolerance to develop. Based on these findings, patients in the present study are titrated through sequential 4-week treatment segments beginning with 0.75 mg once weekly followed by 1.5 mg once weekly. At week 8, patients randomized to the dulaglutide 1.5 mg group continue on this dose for the remainder of the Treatment Period. Patients randomized to the dulaglutide 3.0 mg group are escalated to 3.0 mg once weekly at Week 8, and will continue on this dose for the remainder of the Treatment Period. Patients assigned to the dulaglutide 4.5 mg group are escalated to 3.0 mg once weekly at Week 8 for 4 weeks, followed by escalation to their final dose of 4.5 mg once weekly at Week 12. Study participants are treated for 52 weeks, with the primary objectives assessed at Week 36. This gradual, step-wise dose titration strategy is intended to further improve GI tolerability at the 3.0 mg and 4.5 mg doses relative to that observed in the Phase 2 study described above.

(26) Formulation Stability Studies

(27) A demonstration batch is prepared containing various concentrations of dulaglutide in the formulation used for the currently approved dulaglutide doses, and samples are filled into the commercial primary packaging for dulaglutide, as set forth in Table 5 below:

(28) TABLE-US-00006 TABLE 5 Demonstration batch composition. Dulaglutide Concentration 3.0, 6.0, 9.0, and 12.0 mg/mL Sodium Citrate 2.74 mg/mL Citric Acid Anhydrous 0.13 mg/mL Mannitol 46.4 mg/mL Polsyorbate-80 0.02% Primary Packaging BD Gen 2 1 mL Long, West Uncoated Plunger

(29) Samples are placed on stability for up to 24 months. After 3 months, however, the study was discontinued due to observed decreases in the level of PS80. The data obtained on PS80 concentrations are analyzed in JMP 12.1.0 software for predicting PS80 content in 24-month long-term storage condition. Activation energy (E.sub.a) of PS80 degradation is calculated and used for predicting PS80 content over long-term storage. Based on a second-degree polynomial model, the point estimate for activation energy, E.sub.a, is 15.24 kcal/mol and the lower 95% confidence limit was 12.85 kcal/mol. Based on this analysis, with starting PS80 level at 0.02% (w/v), the 24-month prediction of PS80 for the 12 mg/mL dulaglutide formulation is predicted to be below the specification limit for current commercial dulaglutide. The 24-month prediction of PS80 for the 9 mg/mL dulaglutide formulation would also be close to the specification limit.

(30) In order to evaluate the risk of PS80 content out-of-specification (OOS) during 24 months of long-term storage, a Monte Carlo simulation is performed in R 3.4.0 software, leveraging historical dulaglutide data. The OOS risks associated with each high concentration level are summarized in Table 6 below:

(31) TABLE-US-00007 TABLE 6 Monte Carlo Simulation Results for 24-month PS80 OOS Risk. Probability of PS80% (w/v) OOS after 24 months Dulaglutide 2-8° C. Storage Concentration E.sub.a = 15.24 E.sub.a = 12.85  6 mg/mL 5 × 10.sup.−5 0.06692  9 mg/mL 0.10279 0.96191 12 mg/mL 0.89239 1

(32) The results from the Monte Carlo simulation showed excessive degradation using the asymptotic lower confidence limit of the activation energy and thus high-risk levels for shelf-life OOS. PS80 is present in the dulaglutide formulation to provide protection against physical stress, and the PS80 lower specification limit was set to ensure sufficient PS80 is present throughout maximum possible shelf-life and in-use periods to provide that physical protection.

(33) In view of the preliminary data analysis from the development stability study indicating that polysorbate-80 hydrolysis was increased in the higher strength dulaglutide formulations, it was determined to explore modifications to the formulation. Increasing the level of polysorbate-80 in the formulation may result in adequate intact polysorbate-80 at end-of-shelf life, but the hydrolysis of higher amounts of polysorbate-80 may result in appearance of visible particles. Thus, a surfactant range study was initiated in order to assess the impact of varying levels of surfactant on stability.

(34) Samples were prepared containing dulaglutide concentrations of 3, 6, 9 and 12 mg/mL, 10 mM citrate buffer, 46.4 mg/mL mannitol and PS80 concentrations ranging from 0.002% to 0.05%, filled into vials, and placed on stability at 30° C. Samples are pulled at 0, 12, 25, 35, 46 and 74 days and tested for free oleic acid (FOA)—a hydrolysis product of PS80—and particulate matter. Data show an increase in FOA with increasing rates for higher dulaglutide strength formulations. At the same concentration of dulaglutide, the increased rates of FOA are the same for formulations having 0.02% and 0.05% starting level of polysorbate-80. Particulate matter, as measured by micro-flow imaging (MFI) is equivalent for all formulations, regardless of dulaglutide or polysorbate-80 level. The results support the ability to increase starting levels of polysorbate-80—as much as 0.05% without compromising dulaglutide stability.

(35) Based on results of the development stability study and the surfactant study, a detailed statistical analysis is conducted of all available data. The analysis concludes that the level of starting polysorbate-80 in the formulation, for the 6.0 and 9.0 mg/mL strengths, should be increased from 0.02% to 0.025% (w/v). This increased level is selected to provide high probability that even at end-of-shelf life, there will be sufficient polysorbate-80—meet the current approved specifications for commercial product.

(36) A second demonstration batch is prepared containing the same concentrations as used in the first demonstration batch described above, with the exception of an increase in PS80 concentration, as set forth in Table 7 below:

(37) TABLE-US-00008 TABLE 7 Second demonstration batch parameters. Dulaglutide Concentration 3.0, 6.0, 9.0, and 12.0 mg/mL Sodium Citrate 2.74 mg/mL Citric Acid Anhydrous 0.13 mg/mL Mannitol 46.4 mg/mL Polsyorbate-80 0.025% Primary Packaging BD Gen 2 1 mL Long Syringe, West Uncoated Plunger

(38) Samples are placed on stability at 5° C., 25° C. and 30° C. Results from samples held up to six months at 5° C. and 25° C. and one month at 30° C. indicate that the concentrated formulations are chemically and physically stable, and that the PS80 concentration will remain above the lower specification limit throughout shelf-life.