Report from the American Academy of Neurology annual meeting, Philadelphia, PA, May 4-10, 2019
Clinical tip of the day
Oral DMT to fingolimod: An analysis of the PANGAEA database reported a relapse rate reduction in patients on an oral DMT (teriflunomide, dimethyl fumarate [DMF]) for a mean 1.5 years who were switched to fingolimod (Ziemssen et al. AAN 2019; abstract P3.2.075). Mean reduction in annualized relapse rate (ARR) at 24 months was 73.5%.
Fingolimod to alemtuzumab: ARR was reduced from 1.35 to 0.16 after switching (generally for lack of efficacy) from fingolimod to alemtuzumab (Gonzalez et al. AAN 2019; abstract P3.2.050). The proportion with Gd+ lesions decreased from 59% to 3%. A total of 6% of patients developed thyroid disease in year 1, 21% in year 2.
Natalizumab to ocrelizumab: The OCTAVE study is examining ocrelizumab patients previously treated with natalizumab (Smoot et al. AAN 2019; abstract P3.2.056). Ocrelizumab was started 4-6 weeks after the last dose of natalizumab. To date, 16 patients were relapse-free at 6 months. However, there were four serious adverse events, including one case each of breast cancer and acute cystitis.
Rituximab/ocrelizumab to natalizumab: A small case series reported on three patients who switched from natalizumab to rituximab or ocrelizumab, then switched back to natalizumab (Vollmer et al. AAN 2019; abstract P3.2.103). Patients restarted natalizumab after a single course of anti-CD20 therapy. The reasons for switching back to natalizumab were feeling worse and a return of old symptoms. Mean washout prior to natalizumab restart was 5.8 months.
Escalation vs. de-escalation: A real-world study looked at 786 patients on DMF or fingolimod (Hersh et al. AAN 2019; abstract P3.2.065). The two groups had similar disease control (relapses, Gd+ lesions, new T2 lesions) on treatment. One-third of patients were then switched to a monoclonal antibody (natalizumab, rituximab, ocrelizumab, alemtuzumab), while 39-48% were switched to a first-line injectable or oral DMT. After switching to a MAb, 6% from the DMF group and 11.9% of the fingolimod group relapsed; after switching to an injectable/oral, relapses were more common (14.2% and 18.4%, respectively). There were fewer Gd+ lesions after escalation to a MAb.
Lymphopenia: Four studies reported on the occurrence of lymphopenia during treatment with DMF. The incidence of grade 2 and grade 3 lymphopenia was 24.4% and 6.4%, respectively (Bloch et al. AAN 2019; abstract P4.2.021). Risk factors for developing lymphopenia were age >55 years and low or low-normal baseline lymphocyte count. An integrated safety analysis of phase II/III studies found that periods of prolonged lymphopenia during DMF treatment were not associated with an increased risk of serious infection, including serious herpes zoster infections (Chan et al. AAN 2019; abstract P4.2.023). In patients with lymphopenia who discontinued DMF, the median time to ALC >0.9 x 109/L (grade 0) was 7 weeks. A separate study of older patients (mean age 55 years) reported lymphocyte reconstitution after a median of 22 weeks (Wong et al. AAN 2019; abstract P42.017). The same group reported that lymphopenia improved with a dose reduction to 240 mg/day (Wong et al. AAN 2019; abstract P4.2.013). The average lymphocyte count increased to 0.82 x 109/L with reduced dosing. Over a mean 31-month reduced-dosing period, measures of disability (EDSS, T25FW) were unchanged and no patient experienced a relapse.
TB testing: A German study found that TB testing (IFN-gamma release assay) may be compromised in MS patients currently taking a DMT (Graf et al. AAN 2019; abstract P4.2.014). The chart review found that 5.3% of TB tests were not evaluable, of which one-half remained unevaluable after a second test. In 7 of 8 unevaluable tests, patients were currently receiving a DMT (fingolimod, PEG-IFNb-1a, DMF, alemtuzumab). Antibiotic therapy for latent TB infection remained effective when administered concomitantly with a DMT.
Biotin in progressive MS: The MS-SPI trial of biotin (MD1003) 100 mg TID reported three-month confirmed disability improvement in 12.6% of patients compared to 0% receiving placebo at 12 months (Tourbah et al. Mult Scler 2016;22:1719-1731). In the 12-month open-label extension, 51% discontinued treatment, primarily due to a perceived lack of efficacy (De Seze et al. AAN 2019; abstract P3.2.043). Time to first change in EDSS score (improvement or worsening) and the probability of improvement/worsening at 48 months were not significantly different with biotin versus placebo (De Seze et al. AAN 2019; abstract P3.2.063). The mean EDSS change was numerically better with biotin versus placebo. A real-world study of progressive MS patients receiving biotin for a mean of 12.7 years reported EDSS improvement in 12.4%, stable EDSS in 68.6% and EDSS worsening in 18.9% (Moreau et al. AAN 2019; abstract P3.2.083).
Levodopa powder: The SPAN-PD extension is examining the longer-term efficacy of levodopa inhalation powder (CVT-301) in PD (Farbman et al. AAN 2019; abstract P2.8.048). During the initial double-blind phase, CVT-301 was associated with improvements in UPDRS motor scores at 30 minutes post-dosing (84 mg -9.83 vs. placebo -5.91) (LeWitt et al. Lancet Neurol 2019;18:145-154). At 12 months, 73.0% of patients receiving CVT-301 84 mg achieved an ON state within 60 minutes; results were not significantly different from 1-month values, suggesting that treatment efficacy was maintained.
A survey of 62 patients (average age 71 years) in a medical cannabis program found that 77% experienced improvements in motor symptoms and over one-half had improvements in non-motor symptoms (Myers et al. AAN 2019; abstract P2.8.016). Motor symptom improvement was most common for tremor and spasticity, although improvements in gait instability, dyskinesia and rigidity were also found. Non-motor symptoms showing improvement included sleep disturbance, anxiety, depression and nausea. Adverse effects included somnolence, disorientation and dizziness, with 6.4% discontinuing due to adverse effects.
Clinical tip: Elevated von Willebrand Factor antigen and Factor VIII activity levels during an MS relapse may be predictive of clinical resistance to methylprednisolone, according to a clinical case series (Roy et al. AAN 2019; abstract P3.2.089).
Clinical tip of the day
Progressive MS: EXPAND was the pivotal study of siponimod in SPMS and reported a significant reduction in 3- and 6-month confirmed disability progression versus placebo (Kappos et al. Lancet 2018;391:1263-1273). A new subgroup analysis has examined cognitive outcomes as assessed by the Symbol Digit Modalities Test (SDMT) (Benedict et al. AAN 2019; abstract P3.2.051). Sustained change was defined as a difference from baseline >4 points on all subsequent assessments. Siponimod was associated with a reduction in the proportion of patients with sustained SDMT worsening (hazard ratios 0.72 and 0.76 for patients with and without baseline cognitive impairment). The proportion of patients with sustained SDMT improvement was significantly higher in patients without baseline impairment (HR 1.49).
Neurofilament-light: NfL in CSF and blood (serum or plasma) is emerging as a potentially useful biomarker of neuroaxonal injury. A UCSF study has reported that NfL levels in CSF and peripheral blood are strongly correlated in RRMS and SPMS but not in PPMS (Harp et al. AAN 2019; abstract P2.2.082). CSF and blood NfL results were also correlated in a study of radiologically isolated syndrome (RIS) and were predictive of disease activity (Thouvenot et al. AAN 2019; abstract S37.003). In the EXPAND study of siponimod in SPMS, high baseline sNfL levels were correlated with more rapid SDMT worsening (HR 1.41) (Kuhle et al. AAN 2019; abstract S12.009). A separate study reported that patients with cognitive impairment have higher baseline sNfL levels and greater increases over 5 years compared to MS patients without cognitive impairment (Jakimovski et al. AAN 2019; abstract S37.002). The domains primarily affected were visuospatial memory and processing speed. Uher and colleagues have proposed sNfL cut-off values for predicting MRI activity: NfL >90th percentile was predictive of MRI disease activity (odds ratio 2.6-3.4); and NfL < 30th percentile was predictive of little or no activity with only 10.9% developing >1 active lesions) (Uher et al. AAN 2019; abstract S37.005). The authors suggested that sNfL may reduce the need for annual MRIs.
NfL and DMTs:
Fingolimod: An analysis of the LONGTERMS study of RRMS reported that baseline NfL was reduced 40% with fingolimod, an effect that was sustained for 9.5 years (Cohen et al. AAN 2019; abstract P3.2.032). A FREEDOMS database study found that sNfL levels were correlated with brain-volume loss at month 24 (Sormani et al. AAN 2019; abstract S37.006). Part of the effect that sNfL had on brain-volume loss occurred independently of inflammation.
Alemtuzumab: In the CARE-MS I comparison of alemtuzumab and IFNb-1a, median sNfL levels declined 46% vs. 32%, respectively, from baseline (Kuhle et al. AAN 2019; abstract P3.2.045). At month 24, a higher proportion of alemtuzumab-treated patients had sNfL reductions (81% vs. 72%) and >50% reductions (69% vs. 57%).
Natalizumab: In the ASCEND trial in SPMS, baseline sNfL levels were associated with Gd+ lesion number, T2 lesion volume, brain atrophy and worsening disability at 96 weeks (Kapoor et al. AAN 2019; abstract S12.008). Natalizumab reduced sNfL in SPMS patients with and without inflammatory disease activity.
HSCT: In an HSCT study, baseline sNfL and glial fibrillary acid protein (GFAP), a marker of astrocytosis, were elevated and sNfL was correlated with inflammatory activity (relapses, lesions) (Thebault et al. AAN 2019; abstract S37.004). After autologous HSCT, sNfL levels remained elevated and Tau/GFAP levels increased; increases were correlated with early pseudoatrophy. At 6 months post-transplant, sNfL levels declined; GFAP declined but remained elevated compared to controls.
McDonald misapplied: A survey of 160 MS specialists and residents found that McDonald diagnostic criteria (Thompson et al. Lancet Neurol 2018;17:162-173) are frequently misapplied (Solomon et al. AAN 2019; abstract S6.001). Atypical presentations that were considered typical MS included complete transverse myelopathy (residents 35%, MS specialists 15%), bilateral optic neuritis/unilateral ON with poor visual recovery (residents 17%, MS specialists 10%), and intractable vomiting/nausea/hiccups (residents 20%, MS specialists 5%). Optic nerve involvement was often misidentified as fulfilling the dissemination in space criterion (residents 31%, MS specialists 26%), and radiologically-isolated syndrome (RIS) was incorrectly believed to meet MS diagnostic criteria (residents 48%, MS specialists 12%). Most believed that non-specific historical sensory and coordination symptoms met McDonald criteria for dissemination in time (residents 88%, MS specialists 65%). Many did not identify the criteria of periventricular lesions (residents 62%, MS specialists 39%) or juxtacortical lesions (residents 81%, MS specialists 46%).
Stroke: A survey of 238 U.S. stroke neurologists found no consensus on the timing of anticoagulants for secondary prevention in patients with acute ischemic stroke (AIS) due to paroxysmal atrial fibrillation (Rybinnik et al. AAN 2019; abstract S35.001). For small AIS without hemorrhagic transformation, 24% would start anticoagulation within 48 hours of stroke onset; 80% would anticoagulate by seven days. For greater stroke severity, 29% would anticoagulate within seven days, 48% in 7-14 days and 23% after 14 days. Direct oral anticoagulants were preferred. ASA was preferred in patients ineligible for direct anticoagulants.
Clinical tip: Oral cladribine demonstrates predictable lymphocyte kinetics when sequenced after fingolimod or natalizumab (Hodgkinson et al. AAN 2019; abstract P4.2.034). The effect of cladribine on lymphocyte counts was somewhat less after fingolimod, which may be due to a short washout period and late lymphocyte egress from secondary lymphoid organs after cladribine initiation. Cladribine effects on lymphocyte counts after natalizumab were increased, which may be due to the higher number of circulating lymphocytes during treatment with natalizumab. Lymphocyte counts should be within normal range when initiating cladribine, and PML should be excluded prior to initiating a DMT after fingolimod or natalizumab.
Pregnancy: A small study found that during pregnancy there was a general improvement in all symptom domains, most notably for cognition and sensory symptoms (Nayak et al. AAN 2019; abstract P4.2.102). Only bowel/bladder symptoms worsened in 22% of patients, with symptoms persisting into the postpartum period; it was not determined if symptom worsening was due to MS or was pregnancy-related. A separate study looked at serum neurofilament-light (NfL) during pregnancy (Cuello et al. AAN 2019; abstract P4.2.092). Serum NfL values were higher in the first trimester compared to those of healthy pregnant controls. In the third trimester, sNfL was elevated only in patients experiencing a relapse. Non-relapsing pregnant MS patients had lower sNfL levels than non-pregnant MS controls in the first and third trimesters and in the 6 weeks after childbirth.
Ocrelizumab in PPMS: ORATORIO demonstrated a significant 24% reduction in three-month confirmed disability progression (CDP) with ocrelizumab vs. placebo (Montalban et al. N Engl J Med 2017;376:209-20). Reduction in 24-week CDP was 25%. For the extended controlled period and open-label extension, analysis of CDP was changed to confirmation at 48 weeks (Wolinsky et al. AAN 2019; abstract P3.2.031). The proportion of patients on continuous ocrelizumab with 48-week confirmed disability progression was 30.5% at week 168, 34.8% at week 192 and 43.7% at week 264 (5 years). A separate survey of 146 U.S. neurologists reported that ocrelizumab was the most common DMT used to treat PPMS (Naismith et al. AAN 2019; abstract P1.2.102). However, 51% said they used agents that have not been shown to be effective in PPMS, such as oral DMTs (19%), other monoclonal antibodies (15%), glatiramer acetate (10%) and interferons (7%).
Comorbidities: A new study has suggested a link between food allergies and MS disease activity (Fakih et al. AAN 2019; abstract P4.2.068). The analysis used data from the CLIMB (Comprehensive Longitudinal Investigation of MS at the Brigham and Women’s Hospital) study, which included questionnaires on food, drug and environmental allergies. Patients with self-reported food allergy had a higher relapse rate (adjusted relapse rate ratio 1.27) and more Gd+ lesions (odds ratio 2.53). Drug and environmental allergies did not appear to have an effect. The findings suggest an association between the microbiome/gut immunoglobulins and CNS inflammation.
An analysis of the US Department of Defense database examined comorbidities before and after an MS diagnosis (MS, n=8695, non-MS, 86,934) (Persson et al. AAN 2019; abstract P4.2.058). Prior to MS diagnosis, there was an increased prevalence of depression (22.7% vs. 16.4%), other psychiatric diagnoses (9.8% vs. 6.7%), asthma (4.5% vs. 3.7%), suicidal behaviours (1.3% vs. 0.9%), and epilepsy (0.8% vs. 0.4%) compared to non-MS controls. Infections in the year preceding diagnosis were also more common (46.5% vs. 38.3%). Comorbidities were more common in the median 7 years after an MS diagnosis, including epilepsy (incidence rate ratio 4.90), spasticity (IRR 4.86), neuropathy (IRR 3.78), treated depression (IRR 3.12), venous thromboembolism (IRR 2.54), peripheral vascular disease (IRR 2.49), hospitalized infection (IRR 2.43), myocardial infarction (IRR 2.11) and bowel dysfunction (IRR 2.05). The cancer rate was similar in MS patients and controls.
Acute treatment: Ubrogepant, an oral calcitonin gene-related peptide (CGRP) receptor antagonist, is in development for acute migraine. A phase II trial reported superiority of ubrogepant 100 mg versus placebo in pain-free rate at two hours (25.5% vs. 8.9%), although not for two-hour headache response (Voss et al. Cephalalgia 2016;36:887-898). Two phase III trials (ACHIEVE I and II) are unpublished. [ACHIEVE II results will be presented on Wednesday at Symposium 38, 1:00-3:00 pm.] An analysis from ACHIEVE I/II found that 21% of patients treated with ubrogepant 100 mg were pain-free at 2 hours post-dose, and 61% reported pain relief (Viswanathan et al. AAN 2019; abstract P2.10.012). A 1-year safety analysis reported asymptomatic elevations of liver enzymes that resolved with continued dosing; bilirubin was not elevated (Ailani et al. AAN 2019; abstract P2.10.009). A separate study reported no apparent QTc prolongation at doses up to 400 mg (Jakate et al. AAN 2019; abstract P2.10.003).
Prophylaxis: A 52-week study of the CGRP-targeted monoclonal antibody fremanezumab reported similar efficacy with the monthly (starting dose 675 mg, 225 mg/month thereafter) and quarterly (675 mg q3monthly) dosing regimens in chronic migraine (Yeung et al. AAN 2019; abstract P1.10.008). The mean number of headache days of at least moderate severity decreased from baseline (mean 14.2-14.7 days) to month 12 (5.9-6.1 days). The proportion of patients with ≥50% reduction in migraine days was 52.5% for quarterly dosing and 47.7% for monthly dosing at month 12. [Additional long-term data will be presented on Wednesday at Symposium 38, 1:00-3:00 pm.] A safety study of fremanezumab reported a discontinuation rate of 21% at 12 months (Ning et al. AAN 2019; abstract P1.10.015). The most common adverse effects were injection-site reactions (26–33%). No safety signals with respect to hepatotoxicity or anaphylaxis were observed.
A meta-analysis (three trials, one meta-analysis) examined the usefulness of antihypertensive agents in migraine prophylaxis (Dorosh et al. AAN 2019; abstract P4.10.014). The angiotensin receptor blocker candesartan significantly reduced migraine days by >50% in 43% of patients compared to 23% for placebo; telmisartan also appeared effective in a post-hoc analysis. The ACE inhibitor enalapril was associated with a 50% reduction in headache severity in 48% of patients vs. 11% with placebo.
Clinical tip: Previous studies have reported a reduction in MS relapse activity during pregnancy and an increased relapse risk after childbirth, although most patients (72%) in the PRIMS study were relapse-free in the three months postpartum (Vukusic et al. Brain 2004;;127(Pt 6):1353-1360). A new analysis of health records found no increased relapse risk postpartum (Langer-Gould et al. AAN 2019; abstract S6.007). Exclusive breastfeeding was associated with a reduction in postpartum relapses (hazard ratio 0.58); re-starting a modestly effective DMT had no effect on relapse risk.