REVIEWER: Mark S. Freedman, HBSc, MSc, MD, CSPQ, FANA, FAAN, FRCPC, Professor of Medicine (Neurology), University of Ottawa, Canada
Efficacy
NNT analysis
Safety
Natalizumab discontinuation studies
NeuroSens Survey on first-line orals
The 2015 annual meeting of the American Academy of Neurology (AAN) provided new data on the initiation and optimization of disease-modifying therapy (DMT) for multiple sclerosis. Of particular interest was research on the most recent DMTs, teriflunomide and dimethyl fumarate (DMF), the two oral therapies indicated for first-line treatment. (In Canada, fingolimod is generally recommended as a second-line agent.) Phase III extension data have now been supplemented by real-world observational studies on the effectiveness of therapy in practice, including drug safety and tolerability.
The following summarizes key studies presented at AAN 2015.
Efficacy
Teriflunomide: The efficacy of teriflunomide was demonstrated in TEMSO and TOWER, two phase III trials in relapsing-remitting MS (RRMS) (O’Connor et al. N Engl J Med 2011;365:1293-1303; Confavreux et al. Lancet Neurol 2014;13:247-256). Other phase III studies were TENERE, which compared teriflunomide to subcutaneous interferon beta-1a, and TOPIC, in patients with clinically isolated syndrome (CIS) (Vermersch et al. Mult Scler 2014;20:705-716; Miller et al. Lancet Neurol 2014;13:977-986).
An analysis of data from the TEMSO, TOWER and TOPIC trials examined the time to first relapse in patients receiving one of two daily doses of teriflunomide (7 mg or 14 mg/day) (Honeycutt et al. AAN 2015; abstract P7.279). In all studies there was a consistent improvement in the time to first relapse (25th quartile): 218 days with teriflunomide 14 mg versus 146 days with placebo in TEMSO (hazard ratio 0.719); 369 days versus 188 days in TOWER (HR 0.631); and 609 days versus 317 days in TOPIC (HR 0.660). About 57% of teriflunomide-treated patients were relapse-free in TEMSO and TOWER compared to about 46% of patients in the placebo group.
A separate post-hoc analysis of TEMSO and TOWER reported that teriflunomide 14 mg was significantly more effective then placebo in reducing severe relapses defined according to different criteria (Macdonell et al. AAN 2015; abstract P7.212). In TEMSO, there was a 59% reduction in relapses leading to hospitalization, a 53% reduction in relapses with clinical sequelae, and a 36% reduction in relapses associated with disability (change in EDSS or FS score). Similar reductions (34-54%) were seen for the different definitions of severe relapse in the TOWER analysis.
In the extension to the TENERE trial, ARR remained low in the group switching from subcutaneous interferon-beta-1a to teriflunomide (ARR 0.248 to 0.239) (de Seze et al. AAN 2015; abstract P7.275). There was no washout period. No unexpected adverse effects were associated with switching or ongoing treatment with teriflunomide.
In CIS patients, a reanalysis of TOPIC data using McDonald 2010 diagnostic criteria determined that teriflunomide reduced the risk of conversion to MS by 39.1% versus placebo, similar to the 43% previously reported (Nelson et al. AAN 2015; abstract P7.274). A separate MRI analysis reported that teriflunomide reduced the number of Gd-enhancing lesions per scan by 58.5% versus placebo (Miller et al. AAN 2015; abstract P7.253). Benefits were seen across all subgroups.
Dimethyl fumarate: The efficacy of DMF on reducing relapses in RRMS was determined in the phase III DEFINE and CONFIRM trials (Gold et al. N Engl J Med 2012;367:1098-1107; Fox et al. N Engl J Med 2012;367:1087-1097). In the subgroup of newly-diagnosed patients in DEFINE/CONFIRM with highly active disease, the annualized relapse rate (ARR) was 0.439 (ARR was 0.17 and 0.22 for the full study cohorts) (Phillips et al. AAN 2015; abstract P7.228). The analysis was based on 161 of 1540 patients randomized to the BID and placebo groups (10.4%).
In the subgroup of patients previously treated with interferon beta (n=341), ARR at two years was similar to what was seen in the full intent-to-treat population (ARR 0.19) (Fernandez et al. AAN 2015; abstract P7.231). In the subgroup of patients with baseline EDSS score >3 (n=451), ARR was 0.269 with DMF 240 mg BID and 0.418 with placebo (relative reduction 36%) (Hutchinson et al. AAN 2015; abstract P7.232). The proportion of patients with 12-week confirmed disability progression was lower with DMF BID (15.5% vs. 21.8%), but differences were not statistically significant.
At the conclusion of the two-year pivotal studies, subjects could enroll in the ENDORSE extension and receive DMF 240 mg BID or TID. At five years (three years in ENDORSE), ARR was 0.137 for the BID/BID group and 0.169 for the placebo/BID group (Gold et al. AAN 2015; abstract P7.227; previously presented as Gold et al. ECTRIMS/ACTRIMS 2014, abstract P064). A subgroup analysis based on prior therapy reported that at five years, ARR in the continuous BID group was 0.123 in previously untreated patients, 0.171 in those on one prior injectable, and 0.195 for those on one or more prior injectables (Bar-Or et al. AAN 2015; abstract P7.229). The authors noted that results should be interpreted with caution due to the small sample sizes. In the MRI analysis, 63% of BID/BID patients completing five years were free of new T2 or Gd+ T1 lesions, however, MRIs were obtained in a minority of subjects and the analysis was based on less than 20% of the original BID cohort (Yousry et al. AAN 2015; abstract P7.262).
NNT analysis
An indirect comparison of treatment outcomes with first-line orals was performed by determining the number needed to treat (NNT) to prevent one relapse, one relapse requiring hospitalization, and one patient from experiencing disability progression (Leist et al. AAN 2015; abstract P3.245). The post-hoc analysis used data from phase III trials of teriflunomide (TEMSO, TOWER) and DMF (DEFINE, CONFIRM).
NNT results for preventing one relapse were comparable for teriflunomide 14 mg (5.9 and 5.6 in the two trials) and DMF 240 mg BID (5.3, 5.6). In contrast, the risk of relapse leading to hospitalization was significantly reduced in both teriflunomide trials but not in the DMF trials. In consequence, NNT was substantially lower for teriflunomide (12.5 and 20) versus DMF (50 and 50). The risk of disability progression was significantly reduced in TEMSO, TOWER and DEFINE, but not in CONFIRM. The NNT to prevent one patient experiencing disability progression was similar in the two teriflunomide studies (13.8 and 17.4), and inconsistent for the two DMF trials (10.8 and 30.2).
Safety
Teriflunomide: Two studies have investigated the long-term safety of teriflunomide. Following the completion of the phase II study, 147 of 160 subjects entered the extension and have now received teriflunomide for up to 12 years (Kremenchutzky et al. AAN 2015; abstract P7.223). Cumulative drug exposure was 990 patient-years. ARR during the extension was 0.19 with teriflunomide 14 mg. A total of 18.2% discontinued due to adverse effects. The most common adverse effects were nasopharyngitis, hypoesthesia and fatigue. No new adverse effects emerged during long-term treatment.
A pooled analysis of data from the phase II study as well as TEMSO, TOWER and TOPIC found no increased infection risk with teriflunomide (Leist et al. AAN 2015; abstract P7.268). White-blood cell (WBC) counts declined in the first six months of treatment, then remained stable thereafter. Most patients remained within the normal range (3.8-10.7 x 109/L). The frequency of low lymphocyte or neutrophil counts was less than 2.6%. Cytopenias were not associated with infections. The overall rate of discontinuation due to adverse effects was 1.2% in the teriflunomide 14 mg group. The most common reason for discontinuation was elevated liver enzymes; discontinuation was specified in the study protocol. There were no hematological or lymphoproliferative malignancies. No new or unexpected safety concerns were identified.
DMF: A retrospective single-centre study examined the safety of DMF in real-world practice (Chaves et al. AAN 2015; abstract P3.241). Data were available for 145 patients followed for a median of 8.7 months. Flushing was reported in 57.3% of patients in the first month of starting DMF, increasing to 71.4% in month 9. GI side effects were also common in the first month (53.0%), but declined to 10.7% by month 9. The rate of lymphopenia and/or leucopenia was 26.5% at month 3, increasing to 50.7% at month 6, and 62.0% at month 9. The rates of Grade 2 leucopenia and/or Grade 3 lymphopenia were 2.0%, 4.5% and 6.9%, respectively, for the three time points. Transaminitis occurred in 19.4% but was generally mild. The rate of treatment discontinuation was 15.9%, primarily due to GI adverse effects.
A separate retrospective study examined the prevalence of lymphopenia during treatment with DMF to determine if the rate in clinical practice was higher than that reported in clinical trials (4% in DEFINE) (Longbrake et al. AAN 2015; abstract P3.240). Among 144 patients treated for up to 14 months, the prevalence of Grade 2 or 3 lymphopenia was 10% at months 4-6, and 28% after >12 months. Lymphopenia was persistent and appeared to be more common in patients older than age 55 years.
Natalizumab discontinuation studies
It is often advisable to discontinue treatment with natalizumab after 24 infusions due to the increasing risk of progressive multifocal leukoencephalopathy (PML), and it is generally assumed that a high-potency agent would then be required. However, three studies presented at AAN 2015 examined the feasibility of starting oral therapy post-natalizumab.
A retrospective study examined records for 78 patients who stopped natalizumab (Vu et al. AAN 2015; P3.290). In a group of 22 patients who switched to DMF, 40% experienced a relapse after 4.3 months (9% relapsed on natalizumab); 1 of 9 relapses was associated with MRI changes. Mean EDSS score improved (from 2.7 to 2.4), but mean Timed 25 Foot Walk (T25FW) worsened slightly (from 7.5 to 7.8 seconds). For the 14 patients who switched to teriflunomide, four patients (28%) experienced a relapse after 8.5 months; one relapse was associated with MRI changes. Mean EDSS scores improved (from 3.0 to 2.6), but T25FW worsened somewhat (from 7.2 to 7.7 seconds).
A survey and chart review involving 66 patients who had switched from natalizumab to DMF reported that two-thirds remained stable (Amjad et al. AAN 2015; abstract P3.289). Stable disease was less likely in females, patients younger than age 60 years, and those on natalizumab for less than two years, although differences were not significant. Patients on DMF for >6 months were more likely to be unstable compared to those on DMF for a shorter duration.
A pilot study performed monthly 3T MRI on 15 patients switching from natalizumab to teriflunomide (Edwards et al. AAN 2015; abstract P3.295). Mean age was 48 years; mean number of natalizumab infusions was 52. Subjects began teriflunomide within four weeks of natalizumab discontinuation. Overall, 13 of 15 patients were stable on all MRI parameters after switching to teriflunomide, and no patients had new/enlarging T2 lesions. Two patients had new Gd+ lesions. One patient had a 5-mm pericallosal lesion at month 3 with no symptoms which resolved at month 4. One patient had two 3-mm lesions in the pons at month 6; the patient reported leg weakness two weeks later, which resolved after a course of IV steroids. All patients were able to continue treatment with teriflunomide.
NeuroSens Survey on first-line orals
This poll is currently disabled.