Summary Basis of Decision for Tresiba

 

Clinical Pharmacology

The primary activity of insulin degludec (IDeg), the medicinal ingredient of Tresiba, is regulation of glucose metabolism. Insulin and its analogues lower blood glucose by stimulating peripheral glucose uptake, especially by skeletal muscle and fat, and by inhibiting hepatic glucose production. Insulin also inhibits lipolysis and proteolysis, and enhances protein synthesis.

The clinical pharmacology program of insulin degludec was comprised of 25 clinical studies conducted in patients with Type 1 diabetes mellitus (T1DM), Type 2 diabetes mellitus (T2DM), or healthy subjects and 7 in vitro studies conducted using human biomaterials (human cell lines, plasma, or tissues).

Two dosage forms were proposed: 100 U/mL and 200 U/mL. Results from studies (i.e., Study 3579 and 3672) comparing total glucose-lowering effect at steady-state demonstrated that there is no clinically relevant difference between 100 U/mL and 200 U/mL dosage forms when the same U/kg dose was administered.

The pharmacodynamic properties of IDeg in patients with T1DM or T2DM following subcutaneous administration were evaluated by the glucose-lowering effect measured by glucose infusion rate (GIR) using the glucose clamp technique. Time to maximum GIR was approximately 12 hours (range 11 to 14 hours) following subcutaneous dosing. The glucose-lowering effect of IDeg extends beyond 42 hours (the entire clamp period) following subcutaneous administration in patients with T1DM. At steady-state, the day-to-day variability of IDeg for glucose-lowering effect was about 20%. At dose levels of 0.4, 0.6, and 0.8 U/kg, the glucose-lowering effects of insulin degludec increased with increasing doses.

In single subcutaneous dose studies conducted in special populations, the systemic exposure (measured by the area under the curve from zero to 120 hours [AUC _0-120h]) decreased with increasing degree of hepatic impairment. The total exposure (AUC _0-120h) of IDeg slightly increased with severity of impaired renal function. As with all insulin products, IDeg should be initiated with caution in patients with hepatic or renal impairment with frequent glucose monitoring, and dosage should be adjusted on an individual basis.

The sponsor did not conduct clinical trials to investigate drug-drug interactions. In vitro drug-drug interactions studies showed that common protein-bound drugs (ibuprofen, warfarin, acetylsalicylate, salicylate), frequently used antidiabetic agents (glimepiride, metformin, sitagliptin, liraglutide) or fatty acids (palmitate, oleate, linoleate) did not affect IDeg binding to human serum albumin at therapeutically relevant drug concentrations. The potential insulin degludec drug-drug interaction (if any) is expected to be similar to that of regular human insulin products.

Overall, the clinical pharmacological data support the use of Tresiba for the specified indication.

For further details, please refer to the Tresiba Product Monograph, approved by Health Canada and available through the Drug Product Database.

Clinical Efficacy

Evaluation of the clinical efficacy of Tresiba was primarily based on three Phase III confirmatory clinical trials (3583, 3585, and 3770) conducted in patients with type 1 diabetes mellitus (T1DM), and five Phase III confirmatory clinical trials (3579, 3672, 3668, 3586, and 3582) conducted in patients with type 2 diabetes mellitus (T2DM). All eight clinical trials were randomised controlled open-label trials with a 'treat-to-target' approach using a predefined target fasting plasma glucose of <5 mmol/L. The primary endpoint in these trials was the change from baseline HbA_1c at end-of-treatment. All eight trials were designed as non-inferiority trials; the non-inferiority criterion for the primary efficacy analysis (estimated mean treatment difference in HbA_1c reduction from baseline to end-of-study) was an upper limit 95% confidence interval (CI) <0.4%.

In addition to these eight clinical trials, one additional Phase III confirmatory trial (3580) provided a superiority comparison of Tresiba with sitagliptin (an oral antidiabetic agent [OAD]).

Two additional therapeutic confirmatory trials evaluating Tresiba dosed three-times weekly were performed in T2DM patients (Trials 3718 and 3724), but failed to achieve non-inferiority of Tresiba versus insulin glargine (IGlar) with respect to hemoglobin A_1c (HbA_1c) reduction, and were not used in support of any aspect of the proposed dosing regimen.

Detailed descriptions of each of the nine Phase III confirmatory trials are as follows:

Type 1 Diabetes Mellitus Trials

In all three clinical trials (3583, 3585, and 3770) conducted in type 1 diabetes mellitus (T1DM) patients, study basal insulin was evaluated in combination with insulin aspart (IAsp) as mealtime insulin. All patients enrolled had previously been treated with a basal-bolus insulin regimen and had HbA_1c ≤10%. The 100 U/mL formulation of Tresiba was evaluated in these trials

Trial 3583

Clinical trial 3583 was a 52-week randomized, controlled, multinational, open-label trial comparing the efficacy and safety of Tresiba and IGlar in patients with T1DM.

The trial design included a screening visit to assess eligibility, followed approximately one week later by a randomization visit (Week 0) to assign treatment groups. A total of 629 patients were randomized in a 3:1 ratio to a treat-to-target basal/bolus insulin regimen with either Tresiba or IGlar, both in combination with IAsp as mealtime insulin. Tresiba was administered once daily with the evening meal. IGlar was administered once daily at any consistent time of day that was in accordance with local labelling.

Beginning at Week 1, the patient's basal insulin dose was titrated weekly, based on pre-breakfast self-measured plasma glucose values. After 8 weeks, bolus insulin doses could be optimized based on pre-meal self-measured plasma glucose values. After 26 weeks, a less intensive bi-weekly visit schedule was introduced. The total treatment period was 52 weeks. At the end of treatment with Tresiba or IGlar, the basal insulin was switched to twice daily neutral protamine Hagedorn (NPH) insulin. A follow-up visit was scheduled one week after the end of treatment to assess patients for any safety issues and to measure insulin antibodies.

Trial 3585

Clinical trial 3585 was a 26-week randomized, controlled, multinational open-label trial comparing the efficacy and safety of Tresiba and insulin detemir (IDet) in patients with T1DM.

The trial design included a screening visit followed approximately one week later by a randomization visit (Week 0) to assign treatment groups. A total of 456 patients were randomized in a 2:1 ratio to a treat-to-target basal/bolus insulin regimen with either Tresiba or IDet, both in combination with IAsp as mealtime insulin. Tresiba was administered once daily with the evening meal. IDet was administered once daily at any consistent time of day that was in accordance with local labelling.

Beginning at Week 1, the patient's basal insulin dose was titrated weekly, based on pre-breakfast self-measured plasma glucose values. After 8 weeks, bolus insulin doses could be optimized based on pre-meal self-measured plasma glucose values. If glycemic control was inadequate after at least 8 weeks of treatment, IDet could be given twice daily. The total treatment period was 26 weeks. At the end of treatment with Tresiba or IDet, the basal insulin was switched to twice daily NPH insulin. A follow-up visit was scheduled one week after the end of treatment to assess patients for any safety issues and to measure insulin antibodies.

Trial 3770

Clinical trial 3770 was a 26 week randomized, controlled, multinational open-label trial comparing the safety and efficacy of Tresiba and IGlar in patients with T1DM. This study provided an evaluation of "flexible" Tresiba dosing.

The trial design included a screening visit followed approximately one week later by a randomization visit (Week 0) to assign treatment groups. A total of 493 patients were randomized in a 1:1:1 ratio to a treat-to-target basal/bolus insulin regimen with either Tresiba administered once daily with the evening meal (Tresiba OD), Tresiba administered once daily at alternating narrow (8 to 12 hours) and wide (36 to 40 hours) dosing intervals (Tresiba Flex) or IGlar administered once daily at any consistent time of day that was in accordance with local labelling.

During the 26 week treatment period, both basal and bolus insulin doses were self-titrated by patients, according to a titration guideline. Further guidance for insulin self-titration was provided at weekly visits or phone contacts with the trial site. Basal insulin dose was to be adjusted on Mondays, Wednesdays and Fridays, based on pre-breakfast self-measured plasma glucose values. Bolus insulin doses were adjusted based on pre-meal values. At the end of treatment with Tresiba or IGlar, the basal insulin was switched to twice daily NPH insulin. A follow-up visit was scheduled one week after the end of treatment to assess patients for any safety issues and to measure insulin antibodies.

Efficacy Results

With respect to change in HbA_1c from baseline, and with a non-inferiority margin of 0.4%, treatment with Tresiba was non-inferior to treatment with control basal insulin. Estimated treatment differences (Tresiba - control insulin) were:

• Trial 3583          -0.01 % (95% CI -0.14; 0.11)
• Trial 3585          -0.09 % (95% CI -0.23; 0.05)
• Trial 3770          +0.17% (95% CI +0.04; 0.30)

Analysis of change from baseline fasting plasma glucose was a confirmatory efficacy endpoint for Trials 3583 and 3585; however, statistical significance could not be inferred due to failure of the preceding endpoint in the hierarchical testing strategy that was used to control type 1 error in the confirmatory statistical analyses.

Type 2 Diabetes Mellitus Trials

The T2DM review was based primarily on six Phase III trials. Five of the six trials (3579, 3672, 3668, 3586, and 3582) compared Tresiba with IGlar and were designed as non-inferiority trials. The sixth trial (3580) was a superiority trial comparing Tresiba with sitagliptin. All patients enrolled had been diagnosed with T2DM for at least six months and had been treated with antidiabetic therapy for at least 3 months. A baseline body mass index (BMI) ≤ 40 kg/m² was required. The 100 U/mL formulation of Tresiba was evaluated in these trials, except for study 3672 which allowed for a higher BMI (<45 kg/m²) and evaluated the 200 U/mL formulation of Tresiba. Trial sites were located in Asia, Europe, North America, South Africa and South America.

Trial 3579

Clinical trial 3579 was a 52-week randomized, controlled, multinational open-label trial comparing the efficacy and safety of Tresiba and IGlar in insulin-naive patients with T2DM treated with metformin, with or without additional OAD treatment, and with HbA_1c 7.0 to 10.0%.

The trial design included a screening visit to assess eligibility followed one week later by a randomization visit (Week 0). At the randomization visit, the patient's current antidiabetic treatment was discontinued, except for metformin and (if applicable) dipeptidyl peptidase 4 (DPP-4) inhibitor. A total of 1,030 patients were randomized in a 3:1 ratio to treatment with either 10 units of Tresiba or 10 units of IGlar. Tresiba was administered once daily with the evening meal. IGlar was administered once daily at any consistent time of day that was in accordance with local labelling.

Beginning at Week 1, the patient's basal insulin dose was titrated weekly, based on pre-breakfast self-measured plasma glucose values. The total treatment period was 52 weeks. At the end of treatment with Tresiba or IGlar, the basal insulin was switched to twice daily NPH insulin. A follow-up visit was scheduled one week after the end of treatment to assess patients for any safety issues and to measure insulin antibodies.

Trial 3672

Clinical trial 3672 was a 26-week randomized, controlled, multinational open-label trial comparing the efficacy and safety of Tresiba 200 U/mL and IGlar in insulin-naïve patients T2DM treated with metformin, with or without additional OAD treatment, and with HbA_1c 7.0 to 10.0%.

The trial design included a screening visit to assess eligibility followed one week later by a randomization visit (Week 0). At the randomization visit, the patient's current antidiabetic treatment was discontinued, except for metformin and (if applicable) DPP-4 inhibitor. A total of 460 patients were randomized in a 1:1 ratio to treatment with either 10 units of Tresiba or 10 units of IGlar. Tresiba was administered once daily with the evening meal. IGlar was administered once daily at any consistent time of day that was in accordance with local labelling.

Beginning at Week 1, the patient's basal insulin dose was titrated weekly, based on pre-breakfast self-measured plasma glucose values. The total treatment period was 26 weeks. At the end of treatment with Tresiba or IGlar, the basal insulin was switched to twice daily NPH insulin. A follow-up visit was scheduled one week after the end of treatment to assess patients for any safety issues and to measure insulin antibodies.

Trial 3668

Clinical trial 3668 was a 26-week randomized, controlled, multinational open-label trial comparing the efficacy and safety of Tresiba and IGlar in patients with T2DM treated with either OADs alone, OADs in combination with basal insulin or with basal insulin alone, and with HbA_1c 7.0 to 11.0%. Permitted OADs were any combination of metformin, sulfonylureas, glinides and/or pioglitazone. This study assessed a flexible Tresiba dosing regimen (alternating narrow [8 - 12 hours] and wide [36 - 40 hours] dosing intervals) in T2DM patients.

The trial design included a screening visit to assess eligibility, followed one week later by a randomization visit (Week 0). At the randomization visit, the patient's current antidiabetic treatment was discontinued, except for up to three previous OADs (metformin, sulfonylureas, glinides and pioglitazone, as applicable). A total of 687 patients were randomized in a 1:1:1 ratio to either 10 units of Tresiba administered once daily with the evening meal (Tresiba OD), 10 units of Tresiba administered once daily with a flexible dosing regimen (Tresiba Flex) or 10 units of IGlar administered once daily at any consistent time of day that was in accordance with local labelling.

Beginning at Week 1, the patient's basal insulin dose was titrated weekly, based on pre-breakfast self-measured plasma glucose values. The total treatment period was 26 weeks. At the end of treatment with Tresiba or IGlar, the basal insulin was switched to twice daily NPH insulin. A follow-up visit was scheduled one week after the end of treatment to assess patients for any safety issues and to measure insulin antibodies.

Trial 3586

Clinical trial 3586 was a 26-week randomized, controlled, multinational open-label trial comparing the efficacy and safety of Tresiba and IGlar, in insulin-naïve patients with T2DM treated with metformin and/or a sulfonylurea or glinide, with or without an alpha-glucosidase inhibitor or DPP-4 inhibitor, and with HbA_1c 7.0 to 10.0%. This study was conducted in Asia only.

The trial design included a screening visit to assess eligibility, followed one week later by a randomization visit (week 0). At the randomization visit, the patient's current OAD treatment was continued, except for DPP-4 inhibitors. A total of 435 patients were randomized in a 2:1 ratio to treatment with either 10 units of Tresiba or 10 units of IGlar. Tresiba was administered once daily with the evening meal. IGlar was administered once daily at any consistent time of day that was in accordance with local labelling.

Beginning at Week 1, the patient's basal insulin dose was titrated weekly, based on pre-breakfast self-measured plasma glucose. The total treatment period was 26 weeks. At the end of treatment with Tresiba or IGlar, the basal insulin was switched to twice daily NPH insulin. A follow-up visit was scheduled one week after the end of treatment to assess patients for any safety issues and to measure insulin antibodies.

Trial 3582

Clinical trial 3582 was a 52-week randomized, controlled, multinational open-label trial comparing the efficacy and safety of Tresiba and IGlar in patients with T2DM treated with any insulin regimen with or without OADs, and with HbA_1c 7.0 to 10.0%.

The trial design included a screening visit to assess eligibility, followed one week later by a randomization visit (week 0). At the randomization visit, the patient's current antidiabetic treatment was discontinued, except for metformin and pioglitazone (if applicable). A total of 1,006 patients were randomized in a 2:1 ratio to treatment with either Tresiba or IGlar, both in combination with IAsp as mealtime insulin. Tresiba was administered once daily with the evening meal. IGlar was administered once daily at any consistent time of day that was in accordance with local labelling.

Beginning at Week 1, the patient's basal insulin dose was titrated weekly, based on pre-breakfast self-measured plasma glucose values. After 8 weeks, bolus insulin doses could be optimized based on pre-meal self-measured plasma glucose values. The total treatment period was 52 weeks. At the end of treatment with Tresiba or IGlar, the basal insulin was switched to twice daily NPH insulin. A follow-up visit was scheduled one week after the end of treatment to assess patients for any safety issues.

Efficacy Results

With respect to change in HbA_1c from baseline, and with a non-inferiority margin of 0.4%, treatment with Tresiba was non-inferior to treatment with IGlar in all five non-inferiority trials (3579, 3672, 3586, 3582, and 3668) conducted in T2DM patients. Estimated treatment differences (IDeg - IGlar) were:

• Trial 3579:          +0.09 % (95% CI -0.04; 0.22)
• Trial 3672:          +0.04 % (95% CI -0.11; 0.19)
• Trial 3668:          +0.04% (95% CI -0.12; 0.20)
• Trial 3586:          +0.11 % (95% CI -0.03; 0.24)
• Trial 3582:          +0.08 % (95% CI -0.05; 0.21)

Some studies (3579, 3572, 3586 and 3582) included change from baseline in centrally measured fasting plasma glucose as a confirmatory efficacy endpoint; however, statistical significance could not be inferred for these analyses due to failure of a preceding endpoint in the hierarchical testing strategy that was used to control type 1 error in the confirmatory statistical analyses.

Trial 3580

Clinical trial 3580 was a 26-week randomized, controlled, multinational open-label trial comparing the efficacy and safety of Tresiba and sitagliptin in insulin-naïve patients with T2DM treated with 1 to 2 OADs (metformin, sulphonylurea [SU], glinides and/or pioglitazone) and with HbA_1c 7.0 to 10.0%.

The trial design included a screening visit to assess eligibility, followed one week later by a randomization visit (Week 0). A total of 458 patients were randomized in a 1:1 ratio to treatment with either 10 units of Tresiba or 100 mg of sitagliptin, in combination with the previous OAD(s). Tresiba was administered once daily when preferred by the subject, during waking hours and within a dosing window of a minimum of 8 hours and a maximum of 40 hours between injections. Sitagliptin was administered once daily. A follow-up visit was scheduled one week after the end of treatment to assess patients for any safety issues.

Beginning at Week 1, for patients randomized to Tresiba, the basal insulin dose was titrated weekly based on pre breakfast self-measured plasma glucose values. The dose of sitaglitpin was fixed. The total treatment period was 26 weeks.

Efficacy Results

Trial 3580 conducted in T2DM patients provided a superiority comparison of Tresiba versus sitagliptin. With respect to change in HbA_1c from baseline, treatment with Tresiba was superior to treatment with sitaglitpin. The estimated treatment difference (IDeg - sitagliptin) was -0.43% (95% CI -0.61; -0.24).

 

 

Indication

The sponsor proposed the following indication for Tresiba: "Tresiba (insulin degludec) is an ultra-long-acting basal insulin analogue indicated for once daily subcutaneous administration for the treatment of adults with diabetes mellitus to improve glycemic control."

Following review of the NDS, Health Canada recommended the following indication: "Tresiba is indicated for once-daily treatment of adults with diabetes mellitus to improve glycemic control."

For more information, refer to the Tresiba Product Monograph, approved by Health Canada and available through Drug Product Database.

Clinical Safety

The safety of Tresiba was based primarily on the critical analysis of data derived from 11 therapeutic confirmatory trials. These 11 confirmatory trials consisted of nine Phase III clinical trials previously described in the Clinical Efficacy section and two additional therapeutic confirmatory trials (trials 3718 and 3724) which evaluated Tresiba dosed three times weekly in T2DM patients. Although the clinical trials 3718 and 3724 failed to achieve non-inferiority and were not used in support of any aspect of the proposed dosing regimen, these two trials where included as part of the safety data analysis.

In patients with T1DM, 1,102 patients were exposed to Tresiba and 467 patients were exposed to comparators. Patients enrolled in these studies were most frequently white (80.4%), males (56.3%), with a mean age of 42.7 years. Patients had a mean body mass index (BMI) of 25.8 kg/m², a mean duration of diabetes of 17.3 years, a baseline mean HbA_1c of 7.8%, and a baseline mean fasting plasma glucose level of 9.6 mmol/L.

In patients with T2DM, 3,173 were exposed to Tresiba and 1,802 were exposed to comparators. Patients were most frequently white (73.1%), males (56.5%), with a mean age of 57.9 years. Patients had a mean BMI of 30.9 kg/m², a mean duration of diabetes of 10.2 years, a baseline mean HbA_1c of 8.3%, and a baseline mean fasting plasma glucose level of 9.4 mmol/L.

The most commonly observed serious and severe adverse event in patients using insulin preparations, including Tresiba, was that of hypoglycemia. The actual frequency of hypoglycemic adverse events with use of Tresiba can be difficult to assess given identification of this adverse event depends on several factors. These factors may include: the applied definition of hypoglycemia, diabetes type, insulin dose, intensity of glucose control, background therapies, and other intrinsic and extrinsic patient factors. Therefore, caution is warranted if comparing rates of hypoglycemia between products, as this may be misleading and not representative of rates to be expected in clinical practice.

Based on therapeutic confirmatory trials conducted in patients with T1DM and T2DM, hypoglycemic episodes reported as adverse events were restricted to events of severe hypoglycemia. Severe hypoglycemia was defined as an episode requiring assistance of another person to actively administer carbohydrate, glucagon, or other resuscitative actions. The therapeutic confirmatory trials also assessed the occurrence of confirmed hypoglycemia, which comprised episodes of severe hypoglycemia as well as minor hypoglycemic episodes with a plasma glucose <3.1 mmol/L. The percentage of patients with T1DM randomized to Tresiba who experienced at least one episode of severe hypoglycemia and confirmed hypoglycemia, respectively, was as follows:

• Trial 3583 + IAsp 52 weeks, 12.3% and 95.6%.
• Trial 3585 + IAsp 26 weeks, 10.6% and 93.0%.
• Trial 3770 + IAsp 26 weeks; Tresiba being administered at the same time each day, 12.7% and 99.4%.
• Trial 3770 + IAsp 26 weeks; Tresiba being administered at alternating times, 10.4% and 93.9%.

The percentage of patients with T2DM randomized to Tresiba who experienced at least one episode of severe hypoglycemia and confirmed hypoglycemia, respectively, was as follows:

• Trial 3579 + 1 to 2 OADs insulin naïve 52 weeks, 0.3% and 46.5%.
• Trial 3672 + 1 to 2 OADs insulin naïve 26 weeks, 0% and 28.5%.
• Trial 3668 T2DM ± 0 to 3 OADs 26 weeks; Tresiba being administered at the same time each day, 0.9% and 43.8%.
• Trial 3668 T2DM ± 0 to 3 OADs 26 weeks; Tresiba being administered at alternating times, 0.4% and 50.9%.
• Trial 3582 T2DM ± 0-2 OADs + IAsp 26 weeks, 4.5% and 80.9%.

Other frequently reported adverse events with use of Tresiba included viral upper respiratory tract infections, upper respiratory tract infections, headache, and diarrhea. Viral upper respiratory infections in T1DM patients were observed in 24.0% of patients treated with Tresiba versus 22.5% in patients treated with a comparator. For T2DM patients, viral upper respiratory infections were reported in 12.0% Tresiba-treated patients versus 9.7% in comparator-treated patients. Upper respiratory tract infections reported by T1DM patients was 11.9% in patients treated with Tresiba versus 10.1% in patients treated with a comparator. Headaches in T1DM patients were reported in 11.8% Tresiba-treated patients versus 10.5% in comparator-treated patients. Type 2 diabetes mellitus patients noted headaches with 8.8% in Tresiba-treatment group versus 6.7% in the comparator group. Diarrhea was reported in T1DM patients in 4.1% Tresiba-treated patients versus 4.9% in comparator-treated patients. Type 2 diabetes mellitus patients reported diarrhea in 6.2% Tresiba-treated patients versus 7.2% in comparator-treated patients.

The most frequently reported adverse events leading to treatment withdrawal included hypoglycemia, weight increase, and major adverse cardiovascular events.

Based on the 11 therapeutic confirmatory trials, it was demonstrated that there was a greater frequency and rate of medication errors in patients exposed to Tresiba compared to the comparator. Serious adverse events due to medication errors were relatively infrequent; however, they were exclusively reported in patients exposed to Tresiba. All reported serious adverse events occurred when bolus insulin was injected instead of basal insulin. Similarly, medication errors of moderate to severe intensity were reported at a markedly higher frequency in patients exposed to Tresiba (Number of patients [n]) = 25; 2.3%; 3.6 events per 100 patient years of exposure [PYE]) compared to comparator (n = 2; 0.4%; 0.7 events per 100 PYE). Overall, medication errors associated with Tresiba were similarly reported in patients with T1DM or T2DM. To minimize potential accidental medication errors, patients must visually check the Tresiba label and verify the correct dialed dosage units on the delivery device before each injection. These instructions have also been included in the Tresiba Product Monograph. As for patients who are blind or visually impaired, it is recommended these patients obtain assistance from another person who has good vision and is trained in using the delivery device.

As part of the safety analysis for Tresiba, the sponsor conducted a Major Adverse Cardiovascular Event (MACE) meta-analysis based on the pooled Phase III clinical trial data. During review of this data, a signal of increased cardiovascular risk was identified. However, adequate characterization of the cardiovascular safety profile of Tresiba was precluded by deficiencies in both the cardiovascular safety database and the sponsor's MACE meta-analysis. Therefore, Health Canada issued a Notice of Deficiency (NOD) to the sponsor on October 3, 2012 requesting several new MACE analyses based on an updated cardiovascular safety database. The sponsor submitted additional post-hoc MACE analyses in a NOD Response, but these did not exclude a potential increase in the risk of major cardiovascular events associated with Tresiba. As a result, a NOD-Withdrawal (NOD-W) was issued by Health Canada to the sponsor on July 12, 2013. In the NOD-W letter issued to the sponsor, Health Canada requested new data based on a long-term cardiovascular outcome trial (CVOT) to be conducted in T2DM patients at high cardiovascular risk, in order to rule out the identified cardiovascular safety signal.

To address the identified cardiovascular safety risk, the sponsor subsequently developed the DEVOTE (4080) trial. The DEVOTE trial was a prospective, long-term, multicentre, multi-national, randomized, double-blinded, parallel-group, active-controlled trial designed to assess cardiovascular safety of Tresiba compared to IGlar in patients with T2DM at high risk of cardiovascular events.

Following analysis of the interim results of the DEVOTE trial, Health Canada concluded that the DEVOTE interim results were statistically valid and demonstrated no evidence of increased cardiovascular risk associated with Tresiba. Definitive conclusions regarding the cardiovascular risk of Tresiba (including in select subpopulations) will be deferred until the final report analysis of the DEVOTE trial has been comprehensively evaluated. The sponsor has committed to filing the final report for DEVOTE to Health Canada as a supplemental New Drug Submission in early 2018.

As seen with other insulin products, another risk factor associated with the administration of Tresiba is the possible formation of anti-insulin antibodies. The detection of antibody formation is highly dependent on the sensitivity and specificity of the assay and may be influenced by several factors such as: assay methodology, sample handling, timing of sample collection, concomitant medication, and underlying disease. For these reasons, comparison of the incidence of antibodies to Tresiba with the incidence of antibodies in other studies or to other products may be misleading. As such, caution should be used when interpreting these results.

In clinical trials conducted in T1DM patients, 95.9% of patients who received Tresiba OD were positive for anti-insulin antibodies at least once during the clinical trials, including 89.7% that were positive at baseline. In clinical trials conducted in T2DM patients, 31.5% of patients who received Tresiba OD were positive for anti-insulin antibodies at least once during the studies, including 14.5% that were positive at baseline. The antibody incidence rates for T2DM may be under reported due to potential assay interference by endogenous insulin in samples in these patients. The presence of antibodies that affect clinical efficacy may necessitate dose adjustments to correct for tendencies, toward hyper- or hypoglycemia. As for the incidence of anti-insulin degludec antibodies, this has not yet been established.

Other possible risks associated with the use of Tresiba are as follows:

Hyperglycemia

Inadequate dosing and/or discontinuation of treatment in patients requiring insulin may lead to hyperglycemia and potentially to diabetic ketoacidosis. Furthermore, concomitant illness, especially infections, may lead to hyperglycemia and thereby cause an increased insulin requirement.

Hypokalemia

Hypokalemia is among the potential clinical adverse effects associated with the use of all insulin therapies. This potential clinical adverse effect may be relevant in patients who are on potassium lowering drugs or losing potassium through other means (e.g., diarrhea). Untreated hypokalemia may cause respiratory paralysis, ventricular arrhythmia, and death. As such, it is recommended that potassium levels be monitored in patients at risk for hypokalemia, if indicated (e.g., patients using potassium-lowering medications, patients taking medications sensitive to serum potassium concentrations).

Allergic Reactions

Severe, life-threatening, generalized allergy, including anaphylaxis, generalized skin reactions, angioedema, bronchospasm, hypotension, and shock may occur with any insulin, including Tresiba and may be life threatening. Hypersensitivity (manifested with swelling of tongue and lips, diarrhea, nausea, tiredness, and itching) and urticaria were rare and reported in 0.9% of patients treated with Tresiba.

Lipodystrophy

Long-term use of insulin, including Tresiba, can cause lipodystrophy at the site of repeated insulin injections. Lipodystrophy includes lipohypertrophy (thickening of adipose tissue) and lipoatrophy (thinning of adipose tissue) and may affect insulin absorption. It is therefore recommended to rotate insulin injection sites within the same region to reduce the risk of lipodystrophy. In the clinical program, lipodystrophy, lipohypertrophy, or lipoatrophy was rare and reported in 0.3% of patients treated with Tresiba.

Injection Site Reactions

Patients taking Tresiba may experience injection site reactions, including injection site hematoma, pain, hemorrhage, erythema, nodules, swelling, discoloration, pruritus, warmth, and injection site mass. In the clinical program, injection site reactions occurred rarely, i.e., in 3.8% of patients treated with Tresiba.

Weight Gain

Weight gain can occur with insulin therapy, including Tresiba, and has been attributed to the anabolic effects of insulin. In the clinical program after 52 weeks of treatment, patients with T1DM treated with Tresiba gained an average of 1.8 kg and patients with T2DM treated with Tresiba gained an average of 3.0 kg.

Peripheral Edema

Insulin, including Tresiba, may cause sodium retention and edema. In the clinical program, peripheral edema occurred in 0.9% of patients with T1DM and 3.0% of patients with T2DM treated with Tresiba.

Conclusion:

Overall, based on the critical analysis of the pooled safety data derived from the Phase III clinical trials, there is no pronounced trend suggestive of a clinically relevant safety risk associated with Tresiba in patients with T1DM or T2DM. These results are supported by additional extension studies, interim analysis of the cardiovascular outcome trial (DEVOTE), and international post-market experience, providing supplementary evidence suggestive of an overall favourable safety profile for Tresiba.

For more information, refer to the Tresiba Product Monograph, approved by Health Canada and available through the Drug Product Database.