Summary Basis of Decision for Tecfidera

Clinical Pharmacology

The mechanism by which Tecfidera (dimethyl fumarate) exerts therapeutic effects in multiple sclerosis (MS) is not known. Dimethyl fumarate and its active metabolite monomethyl fumarate (MMF) have been shown to activate the Nuclear factor (erythroid-derived 2)-like 2 (Nrf2) pathway which is involved in the cellular response to a variety of potentially toxic stimuli, including inflammatory and oxidative stress. Additionally, Tecfidera has demonstrated anti-inflammatory effects which are thought to reduce aberrant immune cell activation, which occurs in auto-immune diseases such as MS.

The clinical pharmacology program included reports on the human pharmacodynamic and pharmacokinetic studies. The clinical pharmacological data support the use of Tecfidera for the specified indication.

For further details on the clinical pharmacology of Tecfidera, please refer to the Tecfidera Product Monograph, approved by Health Canada and available through the Drug Product Database.

Clinical Efficacy

The safety and efficacy of Tecfidera (dimethyl fumarate) for the treatment of patients with relapsing remitting multiple sclerosis (RRMS) were evaluated in two pivotal, Phase III, randomized, placebo-controlled, clinical studies in which patients were treated for up to 24 months. The two studies (Study 1 and Study 2) were similar to each other and to clinical studies that evaluated other recently approved therapies for RRMS, with respect to inclusion/exclusion criteria and the safety and efficacy endpoints evaluated. Both studies compared two doses of Tecfidera, 240 mg twice a day (BID) and 240 mg three times a day (TID) to placebo. Study 2 also included a reference comparator arm in which patients received open label glatiramer acetate (GA). Because GA was administered as a daily subcutaneous injection and Tecfidera was administered orally, patients receiving GA in Study 2 were not blinded to their treatment, but study personnel were blinded to all study treatments.

Patients were randomized 1:1:1 to placebo, Tecfidera 240 mg BID, and Tecfidera 240 mg TID in Study 1; and 1:1:1:1 to placebo, Tecfidera 240 mg BID, Tecfidera 240 mg TID, and GA in Study 2. Study 1 included approximately 400 patients per treatment arm (Tecfidera BID: n = 410; Tecfidera TID: n = 416; placebo: 408) and Study 2 included approximately 350 patients per treatment arm (Tecfidera BID: n = 359; Tecfidera TID: n = 345; placebo: n = 363; GA: n = 350). In total, approximately 760-770 patients were treated with each dose of Tecfidera. Patients randomized to Tecfidera received a reduced dose (120 mg BID or TID) for the first week, and the full dose (240 mg BID or 240 mg TID) from Week 2 onward.

In both studies, patient demographics were representative of RRMS patients, with the majority being female and the median age being 37-39 years. The majority of patients were Caucasian. Patients met the 2005 revised McDonald criteria for RRMS diagnosis, had documentation of at least 1 relapse in the previous year, and had an Expanded Disability Status Scale (EDSS) score of 0 to 5.5 (5.5 corresponds to being ambulatory without aid or rest for approximately 100 metres; the disability precludes full daily activities). Baseline disease characteristics indicated that the median EDSS was 2.0-2.5 and the median disease duration since diagnosis was 3-4 years. Patients who were treatment naïve and patients who had been treated previously with other therapies (mainly interferon-beta therapies) were included, with the majority being treatment naïve. Most types of co-morbid medical conditions were potential reasons for exclusion from these studies, including the presence of abnormal hepatic transaminase levels, low white blood cell counts, and urinalysis (proteinuria, hematuria, glycosuria) at screening. The most common types of medical conditions reported in medical histories were consistent with conditions expected in an MS population (neurological, genitourinary, musculoskeletal conditions). With the exception of smoking and alcohol consumption, potential risk factors for cardiovascular disease were reported in relatively low percentages of patients.

All of the efficacy endpoints were standard, validated endpoints for assessing the efficacy of MS therapies in clinical trials, including clinical endpoints that assessed effects of treatment on relapses [annualized relapse rate (ARR), proportion of patients relapsed at 2 years], confirmed progression of disability, and magnetic resonance imaging (MRI) endpoints that assessed changes in inflammatory lesion numbers and volume, which are biomarkers for central nervous system (CNS) tissue damage and loss. Flushing is a very common and visible adverse event (AE) with Tecfidera, which potentially could have unblinded treatment assignments. Although procedures were in place in the studies to minimize the potential for study personnel to become unblinded to treatment assignments in patients who experienced flushing AEs, there was no way to avoid patients potentially being unblinded to their treatment if they experienced flushing AEs. Analyses of clinical efficacy endpoint results in patients with and without flushing AEs did not reveal any meaningful impact of these AEs on efficacy outcomes. Overall, there were no meaningful differences between the treatment effects observed with 240 mg BID and 240 mg TID for any of the endpoints in either study, suggesting that there was no additional benefit from the 240 mg TID dose. Therefore, only results for the 240 mg BID dose are described.

In Study 1, the primary endpoint was the proportion of patients relapsed at 2 years. The proportion of patients relapsed was 27% with Tecfidera compared to 46% with placebo, and the corresponding hazard ratio was 0.51 [95% confidence interval (CI) 0.40, 0.66], representing a reduction in the risk of relapse of 49% (p <0.0001). There were four secondary endpoints; two MRI endpoints and two clinical endpoints. For the MRI endpoints, patients treated with Tecfidera had fewer new/newly enlarging T2 hyperintense lesions and fewer gadolinium (Gd)-enhancing lesions at 2 years than patients that received placebo. The mean number of T2 hyperintense lesions at 2 years was 16.5 in the placebo group versus (vs.) 3.2 in the Tecfidera group, corresponding to an 85% decrease relative to placebo. The mean number of Gd-enhancing lesions at 2 years was 1.8 in the placebo group vs. 0.1 in the Tecfidera group, corresponding to a 90% decrease relative to placebo. The two clinical secondary endpoints were the adjusted ARR at 2 years and 12-week confirmed progression of disability, as measured by protocol-defined worsening from baseline EDDS that was subsequently confirmed at 12 weeks after the onset of the progression. The ARR was reduced by 53% and the proportion of patients with confirmed progression was reduced by 38% in patients treated with Tecfidera compared to placebo.

In Study 2, the primary endpoint was the adjusted ARR at 2 years. The adjusted ARR at 2 years was 0.401 (95% CI, 0.329, 0.488) for patients on placebo compared with 0.224 (95% CI, 0.179, 0.282) for patients on Tecfidera, which corresponded to a relative reduction in ARR of 44% (p <0.0001). Four secondary endpoints were tested; two MRI endpoints and two clinical endpoints. For the MRI endpoints, patients treated with Tecfidera had fewer new/newly enlarging T2 hyperintense lesions and fewer T1 hypointense lesions at 2 years than patients that received placebo. The mean number of T2 hyperintense lesions at 2 years was 19.9 in the placebo group vs. 5.7 in the Tecfidera group, corresponding to a 71% decrease relative to placebo. The mean number of T1 hypointense lesions at 2 years was 8.1 in the placebo group vs. 3.8 in the Tecfidera group, corresponding to a 57% decrease relative to placebo. Clinical secondary endpoints were the proportion of patients relapsed and 12-week confirmed disability progression at 2 years. The proportion of patients relapsed at 2 years was reduced by 34% and the proportion of patients with confirmed disability progression was reduced by 21% with Tecfidera compared to placebo. Overall, the treatment effects in Study 2 were smaller in magnitude than those in Study 1, particularly the treatment effect for the 12-week confirmed disability progression endpoint. Study 2 also included GA as a reference comparator. Results from GA vs. placebo comparisons confirmed the sensitivity of the study.

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

Clinical Safety

Two Phase III clinical studies (described in Clinical Efficacy) provided safety data from 769 patients treated with Tecfidera 240 mg BID, 761 patients treated with 240 mg TID dose, and 771 patients that received placebo. When the safety data from a Phase II study that evaluated the 240 mg TID dose and lower doses are considered, the totality of safety data from the placebo-controlled MS clinical studies come from approximately 1,700 patients exposed to Tecfidera for up to 2 years. In addition, there are safety data from patients who have been enrolled in an ongoing Phase III open-label extension study in which some patients have been exposed for up to 4 years.

In the Phase III placebo-controlled studies, adverse events (AEs) with an incidence of >10% for patients treated with Tecfidera and at a higher rate than for patients that received placebo mainly included flushing (34% Tecfidera vs. 5% placebo) and gastrointestinal AEs, such as diarrhea, nausea, upper abdominal pain, abdominal pain (48% Tecfidera vs. 36% placebo). Infection AEs reported for >10% of patients treated with Tecfidera included urinary tract and upper respiratory tract infections, but these AEs were reported at only slightly higher frequencies with Tecfidera compared to placebo. The high frequencies of flushing and gastrointestinal AEs were consistent in the Phase III studies and in the Phase II study in RRMS patients, and were also common in the clinical pharmacology studies in healthy volunteers after single doses. The majority of these AEs were mild to moderate in severity and occurred most frequently during the first month of treatment. Serious flushing or gastrointestinal AEs were reported for <1% of patients treated with Tecfidera, and flushing or gastrointestinal AEs led to discontinuation of treatment for <1% of patients treated with Tecfidera. To minimize these AEs, treatment is initiated at half the recommended dose (120 mg BID, 240 mg total/day) for the first week of treatment, and then increased to the recommended dose of 240 mg BID (480 mg total/day) in the second week of treatment. As in the clinical studies, temporary dose reductions to half the recommended dose for up to one month or treatment interruptions may be used to manage these AEs. Administration with food may improve gastrointestinal tolerability. A clinical pharmacology study in healthy volunteers in which subjects received 325 mg acetylsalicylic acid 30 minutes before taking Tecfidera, over a 4-day period, showed that acetylsalicylic acid did not affect the pharmacokinetics of Tecfidera and in some subjects may have helped to attenuate flushing AEs. Because longer term co-administration of acetylsalicylic acid has not been evaluated, the use of acetylsalicylic acid to manage persistent flushing AEs is not recommended.

Other safety concerns of special interest included reductions in white blood cell (WBC) and lymphocyte counts; infections (including serious infections, potential opportunistic infections) and the occurrence of infections in patients with low lymphocyte counts; malignancies; hepatic disorders; renal disorders, and cardiac disorders. These safety concerns were of special interest due to: 1) a potential relationship to the pharmacodynamic effects on the immune system [increased risks of infections, potential opportunistic infections and malignancies (via reductions in white blood cell and lymphocyte counts)]; 2) non-clinical findings (renal); and/or 3) observations from approximately 20 years of post-market experience in Germany with psoriasis patients treated with another oral dimethyl fumarate product (immune, renal, hepatic, and cardiovascular effects).

The main findings from these analyses of safety concerns of special interest are the following:

• Tecfidera causes reductions in WBC counts (10%-15%) and lymphocyte counts (30%) from baseline values, which occur over the first year of treatment. Counts remain at these reduced levels throughout the remainder of treatment, and show some recovery by 4 weeks after treatment is discontinued, but the time to recovery to baseline levels was not established in the clinical studies. In the Phase III clinical studies, there were clinically significant lymphocyte counts of <0.5 x 10⁹/L (lower limit of normal is 0.91 x 10⁹/L) for 6% of patients treated with Tecfidera compared to <1% for patients that received placebo. Analyses that compared serious and non-serious infection rates in these patients and patients that did not have clinically significant lymphocyte reductions did not suggest a greater risk of infections (including serious infections) for patients with clinically significant reductions but, the number of patients with lymphocyte counts <0.5 x 10⁹/L in the Phase III clinical studies was too small to conclude that there would be no increased risk of serious infections for these patients. No strong signal has emerged from the analyses of low lymphocyte counts and the occurrence of infections/serious infections in the clinical studies in MS patients, but there have been 4 cases of the opportunistic infection progressive multifocal leukoencephalopathy (PML) reported in psoriasis patients treated for several years with other preparations of dimethyl fumarate. In 2 of these cases there were no apparent risk factors or concomitant medications, but the patients had prolonged severe lymphopenia (reductions in lymphocyte counts) while on treatment. The reductions in WBC and lymphocyte counts that occur with Tecfidera warrant periodic monitoring (complete blood counts), vigilance for the risk of infections and administration of vaccines during treatment is not recommended. All of this has been captured in the Tecfidera Product Monograph. Certain aspects of the effects of Tecfidera on the immune system remain to be characterized, such as: time required for WBC and lymphocyte counts to return to baseline levels; populations of lymphocytes that are affected; the mechanism(s) by which lymphocyte and WBC counts are decreased; and whether the reduction in lymphocytes affects immune responses to new pathogens or vaccines.
• There were asymptomatic elevations in hepatic transaminases (alanine transaminase and aspartate aminotransferase), which were generally between >1 x the upper limit of normal (ULN) and <3 x ULN. These have consistently been reported more frequently in patients treated with Tecfidera compared to placebo in the clinical studies. These elevations were reported mainly during the first 6 months of treatment. There were few serious hepatic AEs or hepatic AEs/transaminase elevations that led to treatment discontinuations. Although no strong signal has emerged for hepatic AEs from the clinical study data in MS patients, patients with elevated hepatic transaminases at baseline were excluded from the clinical studies and Tecfidera has not been studied in individuals with hepatic impairment. In addition, serious hepatic AEs have been reported rarely during post-marketing experience in Germany with another oral dimethyl fumarate product used in patients with psoriasis. The potential risk of hepatic AEs during Tecfidera treatment has been addressed in the Tecfidera Product Monograph by including precautions for hepatic transaminase elevations, patients with hepatic impairment, and a recommendation for periodic monitoring of hepatic function.
• The analyses of renal AEs indicated that there was a slightly higher incidence of proteinuria in patients treated with Tecfidera compared to those that received placebo in the MS clinical studies, but laboratory observations of proteinuria were not progressive. Other findings included a higher incidence of positive urine ketones in patients treated with Tecfidera compared to patients on placebo and an increased frequency of decreased serum 1, 25-dihydroxyvitamin D and elevated parathyroid hormone, with no untoward clinical consequences in the clinical trials. The overall incidences of other renal AEs reported in the clinical studies were low and similar for patients on Tecfidera and patients on placebo, and there were no serious AEs of renal failure. Although no strong signal for renal toxicity emerged from the clinical studies, patients with abnormal urinalysis findings at baseline were excluded from the clinical studies and Tecfidera has not been studied in individuals with renal impairment. In addition, serious renal AEs have been reported rarely during post-marketing experience in Germany with an oral fumarate product used in patients with psoriasis. The potential risk of renal AEs during Tecfidera treatment has been addressed in the Tecfidera Product Monograph by including precautionary information about the proteinuria, positive urine ketones, and effects on serum 1, 25-dihydroxyvitamin D and parathyroid hormone levels in the Product Monograph and a recommendation for periodic urinalysis to monitor renal function.

Overall, the available safety data indicated that flushing and gastrointestinal AEs are the most common AEs associated with Tecfidera treatment and that these AEs are most common during the first month of treatment. Most flushing or gastrointestinal AEs were mild to moderate in severity and there were very few serious AEs or AEs leading to treatment discontinuation. These AEs may be managed in part by administration with food (gastrointestinal symptoms) and/or temporary dose reductions or treatment interruptions.

The available safety data at this time do not indicate an increased risk of infections, hepatic or renal AEs, but do indicate that monitoring of WBC and lymphocyte counts, hepatic transaminases, and renal function by urinalysis is warranted during treatment.

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