AMIFAMPRIDINE PHOSPHATE (FIRDAPSEVR ) IS EFFECTIVE AND SAFE IN A PHASE 3 CLINICAL TRIAL IN LEMS SHIN J OH, MD,1 NATALYA SHCHERBAKOVA, MD,2 ANNA KOSTERA-PRUSZCZYK, MD, PhD,3 MOHAMMAD ALSHARABATI, MD,1 MAZEN DIMACHKIE, MD,4 JOSE MUNOZ BLANCO, MD,5 THOMAS BRANNAGAN, MD,6 DRAGANA LAVRNIC´ , MD, PhD,7 PERRY B SHIEH, MD, PhD,8 CHRISTOPHE VIAL, MD,9 ANDREAS MEISEL, MD,10 SAMUEL KOMOLY, MD, PhD, DSc,11 BENEDIKT SCHOSER, MD,12 KUMARASWAMY SIVAKUMAR, MD,13 YUEN SO, MD, PhD,14 and LEMS STUDY GROUP 1 Department of Neurology, University of Alabama at Birmingham, Birmingham, Alabama, USA 2 Russian Academy of Medical Sciences, Scientific Center of Neurology, Moscow, Russia 3 Department of Neurology, Medical University of Warsaw, Poland 4 University of Kansas Medical Center, Kansas City, Kansas, USA 5 Gregorio Maranon Hospital, Madrid, Spain 6 Columbia University Medical Center, New York, New York, USA 7 Clinical Center of Serbia, Clinic of Neurology, Belgrade, Serbia 8 Department of Neurology, University of California, Los Angeles, California, USA 9 Hospital of Lyon, ENMG Service and Neuromuscular Pathology Hospital, Lyon, France 10 Charite Universitatsmedizin Berlin-NeuroCure Clinical Research Center, Berlin, Germany 11 University of Pecs, Department of Neurology, Pecs, Hungary 12 Ludwig-Maximilians-University Munich Friedrich-Baur-Institute, Munich, Germany 13 Neuromuscular Research Center, Phoenix, Arizona, USA [email protected] 14 Stanford University, Stanford, California, USA Accepted 4 February 2016

ABSTRACT: Objective: We evaluated the efficacy and safety ated; the most common adverse events were oral and digital of amifampridine phosphate (FirdapseVR ) for symptomatic treat- paresthesias, nausea, and headache. Conclusions: This study ment in Lambert-Eaton myasthenic syndrome (LEMS). Methods: provides Class I evidence of efficacy of amifampridine phos- Phase 3, randomized, double-blind, study. Patients were treated phate as a symptomatic treatment for LEMS. initially with amifampridine phosphate for 7–91 days, followed by Muscle Nerve 000:000–000, 2016 randomization to continue amifampridine phosphate for 14 days or placebo (7-day taper, 7-day placebo). The primary efficacy endpoints were changes from baseline at day 14 in Quantitative Myasthenia Gravis and Subject Global Impression scores. Results: The coprimary efficacy end points and 1 of the second- INTRODUCTION ary efficacy end points were met, showing a significant benefit Lambert-Eaton myasthenic syndrome (LEMS) is a of aminfampridine phosphate over placebo at Day 14. All 5 primary, secondary, and tertiary endpoints achieved statistical rare, frequently disabling autoimmune neuromus- significance at Day 8. Amifampridine phosphate was well toler- cular disorder that produces serious muscle

Abbreviations: 3,4-DAP, 3,4-diaminopyridine; ACh, acetylcholine; AEs, received honoraria from serving on the Steering Committee for the LMS- adverse events; Ca21, calcium ions; CGI-I, Clinical Global Impression of 002 study (BioMarin). Dr. Meisel served on the LEMS registry publication Improvement; CMAP, compound muscle action potential; CK, creatine advisory board for BioMarin. Dr. Meisel is funded by German Research kinase; ECG, electrocardiogram; EU, European Union; FDA, Food and Foundation (Exc257, SFB-TRR84, SFB-TRR43), German Federal Ministry Drug Administration; FVC, forced vital capacity; GMP, Good manufacture of Education and Research (01EO0801), European Community FP7 practice; IND, Investigational new drug; LEMS, Lambert-Eaton myasthenic (201024), and obtained research support from ThermoFisher Scientific and syndrome; MeDRA, Medical Dictionary for Regulatory Activities; mg, milli- Novartis. Dr. Meisel also has a speaker’s engagement with Novartis. Dr. gram; MMRM, mixed-effect model repeated measure; NS, not significant; Meisel is a co-inventor and co-owner of a patent on anti-infective agents QMG, Quantitative Myasthenia Gravis score; SGI, Subject Global Impres- and immunomodulators used for preventative therapy after an acute cere- sion of improvement; T25FW, Timed 25-foot Walk test; US, United States; brovascular accident that has been filed to the European Patent Office VGCC, voltage-gated calcium channel (PCT/EP03/02246). Dr. Komoly has received honoraria for talks and pay- Key words: 1. Amifampridine phosphate; 2. 3,4-diaminopyridine; 3. 3,4- ment for occasional consultancy or research funding from Teva, Bayer- DAP; 4. Lambert-Eaton myasthenic syndrome; 5. LEMS; 6. Potassium Schering, Serono, and Biogen. Dr. Schoser serves on the scientific advi- channel blockers; 7. Small cell lung cancer (SCLC) sory boards for Genzyme, a Sanofi company; BioMarin Pharmaceutical This study was supported by Catalyst Pharmaceuticals Inc. and BioMarin Inc.; and Audentes Therapeutics. Dr. Sivakumar reports no disclosures. Pharmaceutical Inc. Dr. So has received an honorarium from Catalyst Pharmaceuticals Inc. LEMS Study Group: 2 Disclosure of conflict of interest: Dr. Oh reports he has been a Vitaly Rudnichenko, MD , Oksana Galkina, MD, PhD2 Beata Szyluk, MD3, Richard J. Barohn, MD4, April L McVey, MD4, consultant to Catalyst Pharmaceuticals Inc. Dr. Shcherbakova reports no 4 5 disclosures. Dr. Kostera-Pruszczyk reports no disclosures. Dr. Alsharabati Mamatha Pasnoor, MD , Irene Catalina-Alvarez, MD , Alfredo Traba- Lopez, MD5, Siegfried Kohler, MD10, Sarah Hoffmann, MD10, Gabriella reports no disclosures. Dr. Dimachkie has received research support from 11 11 6 BioMarin Pharmaceutical Inc. and Catalyst Pharmaceuticals Inc. Dr. Deli, MD , Zoltan Pfund MD PhD , Jacqueline Scoon, MD , Louis H Weimer, MD6, Khosro Farhad, MD6, Katherine M Ruzhansky, MD6, Sujata Munoz Blanco has received research support from BioMarin Pharmaceuti- 6 6 cal Inc. and Catalyst Pharmaceuticals Inc. Dr. Brannagan has received P Thawani, MD, MPH research support from Alnylam, Catalyst Pharmaceuticals Inc., ISIS Phar- Corresponding author’s address: Department of Neurology, University maceuticals, and Novartis. He has been a consultant for Baxter, Catalyst of Alabama at Birmingham, University Station, Birmingham, Alabama, Pharmaceuticals Inc., Daishi Sanko, and Grifols. Dr. Lavinic has received 35294. Email address: [email protected]. Tel number: 205-934-2121. Fax speaker honoraria and travel grants from Pfizer, Berlin Chemie, Worwag number: 205-956-6758. Pharma, and Gideon Richter; and research grant support from the Ministry of Education, Science and Technological Development of the Republic of VC 2016 Wiley Periodicals, Inc. Serbia (projects #175031 and #175083). Dr. Shieh has served as an ad- Published online 00 Month 2016 in Wiley Online Library (wileyonlinelibrary. hoc consultant for Catalyst Pharmaceuticals Inc., Sarepta Therapeutics, com). DOI 10.1002/mus.25070 PTC Therapeutics, Cytokinetics, Biogen, and Novartis. He also serves on the Speakers Bureau for Grifols for their immunoglobulin product. Dr. Vial

Amifampridine phosphate in LEMS MUSCLE & NERVE Month 2016 1 FIGURE 1. Study Design weakness and symptoms of autonomic dysfunc- Societies9 and was approved for use in this indica- tion.1 Sanders estimated the prevalence of LEMS tion by the European Medicines Agency in Decem- in the United States (US) to be 1/100,000 on the ber 2009. Amifampridine phosphate has received basis of prevalence of small cell lung cancer.2 orphan drug designation by the FDA and has also In LEMS, voltage-gated calcium channel (VGCC) been granted Breakthrough Therapy designation. 6 auto-antibodies block influx of calcium ions (Ca21) The multicenter, randomized, double-blind, pla- into the nerve terminal, inducing insufficient release cebo-controlled withdrawal study we report was of acetylcholine (ACh) that produces muscle weak- designed to evaluate the efficacy and long-term ness.3,4 About 50% of patients with LEMS, particu- safety of the phosphate salt form of amifampridine larly male smokers, present with an underlying in patients with LEMS. malignancy, usually small cell lung cancer.5 MATERIALS AND METHODS For symptomatic treatment, pyridostigmine, This multicenter, randomized, phase 3 guanidine, 4-aminopyridine, and 3,4-diaminopyri- “withdrawal trial” was conducted at 18 sites in the dine (3,4-DAP) have been tried. Among these, 3,4- US, European Union (EU), and Russian Federa- DAP (amifampridine) in its base form has been tion. This study was approved by the local ethics found to be most effective and safe.6 committees or institutional review boards for com- 3,4-DAP base has not received Food and Drug pliance with the international and US regulations Administration (FDA) approval for the sympto- and good practice guidelines for pharmaceuticals matic management of LEMS in the United States for human use. All patients gave written informed (US), and at present has limited availability consent. This study was registered as NCT01377922 through Treatment Investigational New Drugs at www.ClinicalTrials.gov. This study was designed (IND) studies, expanded-access programs, clinical and sponsored by Biomarin Pharmaceutical Inc and trials, and compounding pharmacies.6 The poten- Catalyst Pharmaceuticals Inc. tial exists for batch-to-batch variability and issues of lack of reliability in drug quality, potency, and risk Patients. Patients enrolled in the study were at of under- and overdose due to compounding least age 18 years and had a confirmed LEMS errors.7 A review and analysis of compounded 3,4- diagnosis, either previously treated with 3,4-DAP DAP products concluded that they are subject to base (treatment-experienced) or na€ıve (treatment- substantial variability (22.2% - 125.2%) in active naive) with documented acquired (typical) proxi- drug substance content and are noncompliant with mal muscle weakness and at least 1 of the follow- the good manufacturing practice (GMP) standard ing: compound muscle action potential that of 95% to 105% range limit of declared label con- increased 2-fold after maximum voluntary con- tent.7 To overcome the limitations currently associ- traction of the tested muscle or a positive anti-P/ ated with the use of 3,4-DAP base, an effort was Q-type VGCC antibody test. 10,11 For treatment- made to develop a formulation that meets FDA na€ıve patients, the Quantitative Myasthenia Gravis regulatory requirements. Amifampridine phos- (QMG) score was required to be 5. 12 A forced phate, the salt form of 3,4-DAP, has been shown to vital capacity (FVC) of 80% of predicted and have superior stability compared with the base, 8 60% of predicted was required among patients and an oral formulation containing amifampridine who were and were not receiving 3,4-DAP base at phosphate, equivalent to 10 mg base, has been screening, respectively. Patients were required to developed under current GMPs. have normal swallowing function, as defined by Amifampridine phosphate has been recom- the ability to swallow 4 ounces of water without mended as a first-line symptomatic treatment for coughing or throat clearing, and to have com- LEMS by the European Federation of Neurologic pleted anti-cancer treatment 3 months prior to

2 Amifampridine phosphate in LEMS MUSCLE & NERVE Month 2016 the screening visit. If the patient was receiving a base. During Part 1 (open-label), patients received peripherally acting cholinesterase inhibitor, a stable a total daily dose of amifampridine phosphate of dose was required for >7 days prior to screening. If 15 to 80 mg/day, given in 3 to 4 divided doses, the patient had received permitted oral immuno- with a maximum single dose of 20 mg. Treatment- suppressants, a stable dose was required for >90 experienced patients were converted to amifampri- days prior to screening. Patients with known seiz- dine phosphate at either an equivalent or lower ure(s), brain metastasis, or serious electrocardio- dose of 3,4-DAP base. Amifampridine phosphate gram (ECG) abnormalities were excluded. was initiated in treatment na€ıve patients and esca- lated in 5- to 10-mg increments every 4 or 5 days Study Design. The study was conducted in 4 parts (Figure 1). Subjects who successfully met all inclu- to a maximum of 80 mg/day, based on optimal sion criteria at screening were entered into the benefit. Patients had to receive at least 30 mg/day open-label run-in (Part 1). During Part 1, amifam- to enter Part 2 of the study. Patients were required pridine phosphate was titrated for each individual to be on stable background therapy during the patient to a dose that produced optimal neuromus- study. Placebo tablets were indistinguishable from cular benefit and tolerability in the opinion of the amifampridine phosphate tablets and were admin- investigator. Treatment-na€ıve patients were required istered consistent with the dose and dose regimen to have an improvement of at least 3 points in the of amifampridine phosphate. QMG score from the initial evaluation. An improve- Efficacy Assessments. Efficacy endpoints were ment in the QMG score was not required at the identified based upon discussions with the US time of study entry for treatment-experienced FDA. The coprimary efficacy endpoints were patients. Those who continued to the double-blind changes from baseline to Day 14 (the end of Part treatment discontinuation part were required to 3) in QMG score and in Subject Global Impression have received 91 consecutive days of treatment (SGI) score. 12,13 Baseline was defined as the meas- (base or phosphate), including at least 7 consecu- urements obtained at the beginning (Day 1) of tive days of open-label amifampridine phosphate, Part 2 double-blind Treatment. Efficacy assess- and to be on a stable background therapy regimen. ments were also performed at Day 8 in Part 2. In Part 2 (double-blind treatment discontinua- The QMG is a physician-rated evaluation consist- tion), patients were randomly allocated to either ing of 13 assessments mainly designed for clinical continuation of treatment of amifampridine phos- trials in patients with myasthenia gravis. Each assess- phate at the dose established during the open- ment is rated 0 to 3, where 0 indicates “no weakness” label run-in, or to downward titration of the dose and 3 indicates “severe weakness.” The 13 individual to zero and discontinuation of treatment. This was assessment scores were totaled to obtain a QMG accomplished by substituting an increasing propor- score that provided a quantitative assessment of tion of matching placebo tablets for amifampridine muscle strength. For QMG assessment, lower scores phosphate tablets starting on Day 2 and ending on reflected better muscle strength. 12,14 Day 7, at which point all tablets were placebo. The SGI is a 7-point scale on which patients rate Blood samples were collected for pharmacoki- their global impression of the effects of a study netic analysis on Day 1 of Part 2 prior to dosing. treatment (1 5 terrible to 7 5 delighted) on their Patients in Part 3 (double-blind treatment), LEMS symptoms. For SGI assessment, higher ratings who were randomized to amifampridine phosphate reflected a higher level of patient satisfaction.13 in part 2 continued to receive the same dose regi- Secondary endpoints included the changes men for 7 additional days. Patients for whom the from baseline in Clinical Global Impression of dose was titrated downward to placebo in Part 2 Improvement (CGI-I) and Timed 25-foot Walk test remained on placebo for 7 days. Efficacy and safety (T25FW) speed at Day 14.15,16 assessments were performed on Days 8 and 14. If The CGI-I is a 7-point scale that captures the required, rescue treatment was permitted during investigator’s global impression of improvement or the treatment discontinuation and double-blind worsening (1 5 very much improved; 7 5 very parts of the study. much worse) baseline status. An investigator scored In Part 4 (open-label safety assessment), the CGI-I scale based on changes in symptoms, patients were offered the option to participate in a 15 behavior, and functional abilities. On this assess- long-term safety assessment for up to 2 years (not ment, lower ratings reflected a higher level of reported here). improvement as perceived by the investigator. The Drug Regimen, Placebo, and treatment. Active treat- T25FW is a quantitative mobility and leg function ment was administered as tablets of amifampridine performance test based on a timed 25-foot walk. phosphate (FirdapseVR , Catalyst Pharmaceuticals) Following a rest period of at least 5 minutes, the equivalent to 10 mg of amifampridine (3,4-DAP) T25FW was repeated.16 The measurement for the

Amifampridine phosphate in LEMS MUSCLE & NERVE Month 2016 3 FIGURE 2. Disposition of Study Patients

T25FW was the average speed, expressed in feet/ study drug and had any post-treatment safety minute, of the 2 completed walks. The compound information collected.18 Efficacy endpoints were muscle action potential (CMAP), a tertiary end- summarized using descriptive statistics at each mea- point, was obtained in the abductor digiti minimi surement time point. Treatment differences were muscle at rest with a single maximal stimulation of assessed at a significance level of 0.05 for all end- the ulnar nerve at the wrist.10 Because amifampri- points. A mixed-effect model repeated measures dine phosphate acts rapidly (Tmax 5 0.5 hour) and (MMRM) model was used to analyze each of the has a short half-life (T1=2 5 2.5 6 0.73 hours), effi- coprimary endpoints. For each coprimary endpoint, cacy assessments were required to be performed 45 the model included treatment, time (Day 8, 17 minutes after taking a dose of medication. Day 14), treatment-by-time interaction, and double- Safety Assessments. Safety assessments included blind baseline scores as fixed effects and patient as a monitoring for adverse events (AEs) using Medical random effect. The P-values corresponding to the Dictionary for Regulatory Activities (MeDRA) treatment comparisons were determined using a codes, clinical laboratory assessments, vital-signs permutation test. P-values determined by permuta- assessments, physical examinations, and ECGs. tion remove all assumptions of normality of distribu- Pregnancy testing was also required for women of tion. Both the results for the QMG and SGI were childbearing potential. required to be statistically significant in order to deem the study positive. A ‘per-protocol’ analysis Statistical Analyses. A sample size of at least 36 was also conducted and included patients in whom patients was required to detect a 2.44-unit differ- assessments were conducted in accordance to the ence in the mean change of the QMG scores protocol. All patients who received at least 1 dose between the 2 treatment groups, assuming a type I of amifampridine phosphate and had any post- error of 0.05 and a common standard deviation of treatment safety information collected were 2.5. 18–21 included in the safety analysis. A final safety analysis Two prespecified populations were analyzed: 1) will be performed after completion of the Part 4 the full analysis set (intent to treat) consisted of all open-label extension and will be reported randomized patients who received at least 1 dose separately. of amifampridine phosphate or placebo in Part 2 and who had at least 1 post-baseline efficacy assess- RESULTS ment, and 2) the safety population included all Patients Demographics. Seventy-four patients were enrolled patients who received at least 1 dose of screened for eligibility, and 54 entered Part 1 of the

4 Amifampridine phosphate in LEMS MUSCLE & NERVE Month 2016 Table 1. Demographic Data Parts 2 and 3

Part 1 Amifampridine phosphate Placebo Overall N 53162238 Caucasian, N (%) 48 (90.6) 13 (81.3) 21 (95.5) 34 (89.5) Women, N (%) 35 (66.0) 9 (56.3) 14 (63.6) 23 (60.5) Mean age (range), years 52.1 (20–88) 51.6 (25–67) 51.5 (21–88) 51.5 (21–88) Treatment-experienced, N (%) 11 (20.8) 3 (18.8) 7 (31.8) 10 (26.3) PEF: 100%, N (%) 32 (60.4) 9 (56.3) 12 (54.5) 21 (55.3) VGCC antibody-positive, N (%) 48 (90.6) 15 (93.8) 20 (90.9) 35 (92.1) History of cancer, N (%) 9 (17.0) 3 (18.6) 3 (13.6) 6 (15.6)

PEF5post-exercise facilitation; VGCC5voltage-gated calcium channel. study (Figure 2); 20 patients did not meet all eligi- exposure to treatment was 630 days in the placebo bility criteria. The safety population included all 53 group and 365 days in the amifampridine phos- patients who received 1 dose of amifampridine phate group. There were no remarkable differen- phosphate and who had any post-treatment safety ces between groups in other parameters. During information collected (1 patient was excluded from Part 2 of the study, 1 placebo patient was rescued the safety population because no post-baseline because of an increase in the QMG score >5 and data were collected). Among 54 patients in Part 1, was advanced to Part 4. Thus, 37 patients com- 14 did not complete the Part 1 open-label study for pleted Part 2 of the study. During Part 3 of the various reasons, and 2 were advanced to Part 4 of study, 1 placebo patient was rescued because of the study. increase in the QMG score >5 and was advanced Thirty-eight patients (placebo, n 5 22; amifam- to Part 4. Thus, 36 patients completed Part 3 of pridine phosphate, n 5 16) completed the 3- the study. month, open-label run-in period (Part 1) and were Forty patients have entered the 2-year open- enrolled in the subsequent 2-week, double-blind, label safety extension part (38 patients from the randomization period. Thus, the full analysis set randomized part and 2 from the run-in part). The included all 38 patients who received 1 dose of result of Part 4 will be reported separately. amifampridine phosphate or placebo in Part 2 and who had 1 post-baseline efficacy assessment. At baseline, the majority of patients (73.7%) Primary Efficacy Endpoints at Day 14 in Part 3. The were treatment-na€ıve, and there were more women coprimary efficacy end points of the change in the (66%) enrolled than men (Table 1). Prior median QMG and SGI scores were met and showed a

Table 2. Full Analysis Scores on Days 8 and 14 in the Primary, Secondary, and Tertiary Endpoints Baseline Day 8* Day 14† (N538) (N537) (N536) Primary endpoints QMG Amifampridine phosphate 6.4 6 3.22 6.4 6 3.08 6.7 6 4.09 Placebo 5.6 6 3.99 9.5 6 3.58 7.9 6 2.85 P value (95% CI) <0.0001 (24.9, 21.9) 0.0452 (23.4, 0.0) SGI Amifampridine phosphate 5.6 6 1.26 5.4 6 1.20 4.9 6 1.57 Placebo 5.9 6 1.22 3.5 6 1.99 3.2 6 1.70 P value (95% CI) 0.0010 (0.9, 3.1) 0.0028 (0.7, 3.0) Secondary endpoints CGI-I Amifampridine phosphate 2.6 6 0.63 3.5 6 0.97 3.6 6 1.50 Placebo 2.5 6 0.98 4.6 6 1.53 4.8 6 1.45 P value (95% CI) 0.0170 (22.0, 20.2) 0.0252 (22.1, 20.1) TFW25 (ft/min) Amifampridine phosphate 254 6 126 263 6 141 253 6 126 Placebo 255 6 111 205 6 109 244 6 116 P value (95% CI) 0.0302 (4.8, 89.9) NS (226.8, 43.8) Tertiary endpoint CMAP (mV) Amifampridine phosphate 5.7 6 3.72 5.7 6 4.12 5.7 6 3.72 Placebo 6.5 6 3.26 4.9 6 3.35 5.0 6 3.26 P value (95% CI) 0.0065 (0.5, 2.6) NS (21.0, 1.6)

QMG: Quantitative Myasthenia Gravis score; SGI: Subject Global Impression of improvement; CGI-I: Clinical Global Impression of Improvement; TFW25: Timed 25-foot Walk test; NS: not significant, CMAP: compound muscle action potential. *1 placebo patient was rescued. †1 placebo patient was rescued.

Amifampridine phosphate in LEMS MUSCLE & NERVE Month 2016 5 Table 3. Per-protocol Analysis Scores on Day 14 in the Primary and Secondary Endpoints Baseline Day 14 (N526) (N526) Primary endpoints QMG Amifampridine phosphate 6.5 6 3.34 5.8 6 3.46 Placebo 5.9 6 4.12 8.2 6 2.99 P value (95% CI) 0.0048 (24.7, 20.9) SGI Amifampridine phosphate 4.9 6 0.99 5.1 6 1.10 Placebo 5.6 6 1.26 3.0 6 1.59 P value (95% CI) 0.0019 (0.9, 3.4) Secondary endpoints CGI-I Amifampridine phosphate 2.7 6 0.67 2.8 6 0.92 Placebo 2.6 6 1.03 4.7 6 1.58 P value (95% CI) 0.0024 (23.0, 20.7)

QMG: Quantitative Myasthenia Gravis score; SGI: Subject Global Impression of improvement; CGI-I: Clinical Global Impression of Improvement. significant benefit of amifampridine phosphate versus 93.8% on amifampridine phosphate, and at over placebo at Day 14, 42.9% versus 75% had scores of 4 to 7 on placebo compared to amifampridine phosphate, Day 14 in the full analysis (Table 2). Over this time respectively. frame, the mean increase in QMG score was 0.3 There was also a strong correlation between the among patients randomized to amifampridine change from baseline in QMG score and the phosphate and 2.2 among those randomized to change from baseline in SGI score (the R2 placebo [mean difference, 21.7 in favor of ami- statistic50.56), using the full analysis set. fampridine phosphate; 95% confidence interval (CI), 23.4, 20.0]. This difference was statistically Secondary and Tertiary Efficacy Endpoints at Day 14 significant (P50.0452). in Part 3. At Day 14, the CGI-I was significantly Among the 13 domains in the QMG, the arm improved among patients who received amifampri- domain showed the most change (95% CI, 21.8, dine phosphate as compared with those who 20.6; P < 0.0001). Considering that the arm and received placebo in the full analysis set. Mean leg domains of the QMG represent a measure of scores were 3.6 and 4.7 for amifampridine phos- proximal muscle weakness characteristic of LEMS, phate and placebo, respectively, with a statistically an analysis of the arm plus leg domains was per- significant (P50.0267) mean difference of 21.1 formed. At Day 14 the results were statistically sig- (95% CI: 22.1, 20.1) in favor of amifampridine nificant in favor of amifampridine phosphate phosphate. There was no statistically significant dif- (P50.0195). ference in changes in scores in T25FW and CMAP A ‘responder analysis’ evaluation of whether amplitude at Day 14. patients experienced clinically significant disease Amifampridine phosphate efficacy at Day 8 in progression (defined as an increase of at least 3 Part 2. All 5 primary, secondary, and tertiary end- points) in their QMG scores from baseline to Day points achieved statistical significance at Day 8 8 and Day 14 showed that amifampridine phos- (Table 2), suggesting that placebo patients experi- phate was superior to placebo. At Day 8, 59.1% of enced a clinically and statistically significant wor- placebo patients had clinically significant disease sening of symptoms during the amifampridine progression versus 6.3% on amifampridine phos- phosphate withdrawal period and indicating that phate; and at Day 14, 38.1% versus 12.5% had pro- amifampridine phosphate was effective. The differ- gression on placebo compared to amifampridine ence between amifampridine phosphate and pla- phosphate, respectively. cebo in QMG score at Day 8 was 3.4 (P < 0.0001), The mean change in SGI at Day 14 was 20.7 satisfying standard criteria for “clinical signifi- and 22.7 in the amifampridine phosphate and pla- cance”, as defined as a QMG score change of 2.6 cebo groups, respectively [mean difference, 1.7 in points.12 The efficacy of amifampridine phosphate favor of amifampridine phosphate (95% CI, 0.7, is further supported by objective CMAP results at 3.0; P50.0028)]. The comparison at Day 8 was also Day 8. significant in favor of amifampridine phosphate (Table 2). Per-protocol analysis at Day 14. To more thor- An evaluation of patient satisfaction with the oughly evaluate the effects of treatment and the treatment (SGI score of 4 to 7, partially satisfied to observed differences between Day 8 and Day 14, a delighted) at Day 8 and Day 14 showed amifampri- per-protocol analysis was performed in 26 patients dine phosphate to be superior to placebo. At Day (Table 3). This analysis was performed because the 8, 45.5% of placebo patients had scores of 4 to 7 assessment protocol at Day 14 was not followed in

6 Amifampridine phosphate in LEMS MUSCLE & NERVE Month 2016 reported 3 TEAEs (muscle weakness, myalgia, Table 4. Treatment-emergent Adverse Events during Open-label Run-in (part 1) chronic sinusitis). None of these TEAEs were seri- ous or resulted in study drug discontinuation. Only Previously Treatment-na€ıve Treated Overall 1 event was deemed drug related by the investiga- Symptoms* (N542) (N511) (N553) tor, and this was in the placebo group. In Part 3, double-blind treatment, 18.8% of Oral paresthesia 20 (47.6)* 1 (9.1) 21 (39.6) Digital paresthesia 18 (42.9) 0 18 (34.0) patients randomized to amifampridine phosphate Headache 5 (11.9) 0 5 (9.4) reported 3 TEAEs (nasopharyngitis, otitis externa, Nausea 5 (11.9) 0 5 (9.4) urinary tract infection) compared to 27.3% of Diarrhea 4 (9.5) 1 (9.1) 5 (9.4) patients randomized to placebo who reported 11 Constipation 3 (7.1) 0 3 (5.7) Dizziness 3 (7.1) 0 3 (5.7) TEAEs [asthenia, sensation of heaviness, periodon- Sore throat 3 (7.1) 0 3 (5.7) titis, dental pulpitis, fatigue, muscle weakness, Upper respiratory 3 (7.1) 0 3 (5.7) blood creatine kinase (CK) increase, depression, tract infection dyspnea]. None of these TEAEs were serious or Limb pain 3 (7.1) 0 3 (5.7) resulted in study drug discontinuation. Two events Fall 3 (7.1) 0 3 (5.7) in the placebo group were deemed drug related by Each symptom listed here was reported in at least 3 patients. the investigator. In 1 patient, a liver panel was *N (%) abnormal; mildly elevated CK was noted in 1 pla- cebo patient. No seizures, serious ECG abnormal- 10 patients; more specifically, amifampridine phos- ities (including prolonged QT interval), or serious phate was taken 1 night prior to assessment instead laboratory abnormalities were reported in the clini- of 45 minutes prior to the assessment. In this anal- cal trial. Side effects were generally mild and ysis, there was a 0.7-point decrease in QMG score waned over time with continued use of amifampri- among patients who received amifampridine phos- dine phosphate. Fewer side effects were reported phate, as compared with a 2.3-point increase in treatment-experienced patients compared to among patients who received placebo. This differ- treatment-na€ıve patients. ence was 3.0, which was statistically significant DISCUSSION (P50.0048) in favor of amifampridine phosphate. Amifampridine is the non-proprietary active Thus, the statistical value of significance was ingredient name for 3,4-DAP. In 2009, the Euro- increased from 0.0453 in the full analysis to 0.0048 pean Commission granted marketing approval to in the per-protocol analysis. Further, the 3.0-point amifampridine phosphate as an orphan drug for difference between study drug and placebo satisfies symptomatic treatment of LEMS. Amifampridine the criterion of “clinical significance”, as defined phosphate has been shown to have superior stabil- by a difference in QMG score of 3.0. ity compared with the 3,4-DAP base and can be A similar pattern of increased difference stored at room temperature; in the EU, amifampi- among the full analysis set and per-protocol popu- dine phosphate has been shown to have a 3 year lation was seen for the SGI and CGI-I scores. No shelf life.8,22 statistical significance was observed for either the Amifampridine is a voltage-gated potassium T25FW or CMAP in this analysis. channel blocker.23 The blocking action prolongs Safety. Amifampridine phosphate was well toler- depolarization of nerve action potentials. This ated, and there were no unexpected or notable increases the open time of VGCCs, thereby increas- adverse events. No serious adverse events attribut- ing presynaptic calcium levels. The increased cal- able to amifampridine phosphate were reported cium influx enhances ACh release, which then during the study. In Part 1 (N 5 53), serious binds to muscle receptors, resulting in improved adverse events (none attributed to drug treatment) muscle function. Thus, amifampridine is ideally were reported in 3 subjects, only 1 of whom did suited for symptomatic treatment for LEMS. not proceed into Part 2. Overall, the most common Since the first use of amifampridine in LEMS adverse events were oral (40%) and digital (34%) in 1983, its efficacy and safety have been docu- paresthesia, nausea, and headache (Table 4). mented consistently in multiple case reports and 1 During the double-blind treatment discontinua- open trial in more than 70 patients.24,25 In these tion (Part 2), 37.5% of patients randomized to ami- reports, moderate-to-marked functional improve- fampridine phosphate reported a total of 11 TEAEs ment was seen in patients who received amifampri- (headache, erythema, subcutaneous abscess, upper dine at doses ranging from 24 to 80 mg per day. respiratory infection, dizziness, diarrhea, chest pain, The maximum daily dose should not exceed 80 seasonal allergy, oropharyngeal pain) compared to mg, because seizure has been reported in 3 13.6% of patients randomized to placebo who patients on a daily regimen of 100-mg.20,26

Amifampridine phosphate in LEMS MUSCLE & NERVE Month 2016 7 Three randomized, placebo-controlled trials sis of those patients in whom protocol instructions with oral 3,4-DAP base and 1 with intravenous 3,4- were followed demonstrated that those assigned to DAP base have demonstrated its efficacy in patients amifampridine phosphate had sustained improve- with LEMS. 19,20,23,27 These 4 trials showed a ment in QMG scores through the end of the significant improvement in muscle strength or double-blind treatment period compared with QMG score and CMAP amplitude in patients patients who received placebo (P50.0048). The with LEMS.19,20,26,27 In a crossover study, patients difference is explained by the finding that 10 treated with 3,4-DAP base showed significantly bet- patients took their last dose of study drug the eve- ter CMAP, LEMS classification, Medical Research ning prior to final assessments instead of the Council muscle strength, QMG, and subjective morning of. Based upon the pharmacokinetics of symptom scores (P 5 0.0017–0.0246).19 Based on amifampridine phosphate, the medication dosing analyses of the results of these 4 randomized, con- time would have implications on efficacy evalua- trolled studies in a total of 54 patients with LEMS, tions and bias the results against the treatment a Cochrane review in 2011 concluded that there group. This was confirmed by the per-protocol was “limited but moderate to high-quality evidence analysis. Nevertheless, the QMG score was still stat- showing that over days 3,4-DAP improved muscle istically significant in favor of the treatment group strength and CMAP in LEMS.21 for the full analysis data set. One hypothesis is that This is the phase 3, double-blind, randomized some pharmacodynamic effect may extend beyond study with the largest number of patients that has those predicted solely on the pharmacokinetics of lasted longer than 8 days. We randomized 38 the drug; however, additional research is required patients at multiple academic centers and eval- to evaluate this hypothesis. Notably, the SGI score uated the efficacy of amifampiridine phosphate at appeared less effected by the medication dosing Day 8 (Part 2) and Day 14 (Part 3). All previous time. This is due to the fact that SGI is based randomized trials were conducted at a single aca- upon the patient’s impression of the effects of the demic center, involved 7-26 patients, and analyzed study medication during the previous week. the results at days 3 to 8 in phase 2.19,20,26 In con- According to Barohn et al. the QMG score trast, our study demonstrated significant efficacy of change should be >2.6 to be “of clinical signifi- amifampiridine phosphate over placebo in all 5 cance” based on 5 MG patients and 4 controls.14 If endpoints at Day 8 in Part 2 and a significant this criterion is applied, both the Oh et.al. study steady efficacy of amfampiridine phosphate com- and the present study at Day 8 in Part 2 and at pared with placebo in both primary endpoints and Day 14 in the per-protocol analysis showed a 1 of the 2 secondary endpoints at Day 14 in Part 3 “clinically significant improvement”.19 Considering in the full analysis. that the QMG score was primarily designed for If the primary endpoints had been set at Day 8 assessing clinical trials in myasthenia gravis, this in Part 2 similar to what was done in the previous finding is remarkable. This was also supported by studies, this study would have shown that amifam- the improvement in the QMG score in the com- pridine phosphate was significantly more effica- bined arm and leg QMG domains. This indicates cious than placebo based on all primary, that the QMG score is still valid for clinical evalua- secondary, and tertiary endpoints. 19,20,26 tion in LEMS. In 2 previous phase 2 studies, 3,4-DAP base This study also used the subjective evaluation showed a significant improvement in QMG over pla- by the patient as measured using SGI-score as a cebo: a 2.76 score difference in a prior study by Oh coprimary endpoint. There was a significant differ- et al. and 2.44 in a study by Sanders et al.19,20 In ential of 1.7 in favor of amifampridine phosphate these 2 studies, the evaluation was performed at over placebo for this end point. A subjective symp- days 3 to 6. In the present study, at Day 8 in Part 2, tom score was first used in LEMS in the Oh et.al. the highest differential of 3.1 occured, indicating study; consistent with the findings of this study, clinically significant improvement associated with there was a statistically significant improvement of amifampiridine phosphate over placebo. This study 3,4-DAP base treated group.19 showed the lowest differential of 1.8 at Day 14 in The CGI-I score and T25FW test were secondary Part 3, that was, however, statistically significant. endpoints. The T25FW test was used in the 4- Thus, there was an unexpected decrement in aminopiridine study in multiple sclerosis.16 The response to treatment in the QMG between days 8 CGI-I score showed a significant improvement with and 14. Amifampridine phosphate has a rapid amifampridine phosphate over placebo. Thus, one absorption (Tmax50.5 hours), and relatively short can conclude that subjective evaluation by patients half-life (2.5 hours). For this reason, assessments and by clinicians confirmed the objective findings of were specified to be performed 45 minutes after the QMG score. Our study also showed a significant taking a dose of medication. A per-protocol analy- improvement in CGI-I score with amifampridine

8 Amifampridine phosphate in LEMS MUSCLE & NERVE Month 2016 phosphate at Day 8. A significant improvement with 11. Lennon VA, Kryzer TJ, Griesmann GE, O’Suilleabhain PE, Windebank AJ, Woppmann A, et al. Calcium-channel antibodies in 3,4-DAP base over placebo has been reported in the Lambert–Eaton syndrome and other paraneoplastic syndromes. abstract form in a phase 2 study using the endpoint N Engl J Med 1995;332:1467–1474. 28 12. Jaretzki A III, Barohn RN, Ernsthoff RM, Kaminski HJ, Keesy JC, of 3 TUG study at 3.5 days. Our study also showed Penn AS, et al. Myasthenia gravis. Recommendations for clinical significant improvement at Day 8 in the CMAP as research standards. Neurology 2000;55:16–23. 19,20,26 13. Farrar JT, Young JP Jr, LaMoreaux L, Werth JL, Poole RM. Clinical previously reported. importance of changes in chronic pain intensity measured on an 11- Amifampridine phosphate therapy was well tol- point numerical pain rating scale. Pain 2001;94(2):149–158. 14. Barohn RJ, McIntire D, Herbelin L, Wolfe GI, Nations S, Bryan WW. erated with an excellent safety profile and without Reliability testing of the quantitative myasthenia gravis score. Ann N any notable adverse events. The most common Y Acad Sci 1998;41:769–772. 15. Goodman AD, Brown TR, Krupp LB, et al. Sustained-release oral side effects reported by patients were tongue and fampridine in multiple sclerosis: a randomised, double-blind, con- finger paresthesia, mainly in treatment-na€ıve trolled trial. Lancet 2009;373(9665):732–738. 16. Cutter GR, Baier ML, Rudick RA, Cookfair DL, Fischer JS, Petkau J, patients. These are expected drug effects of ami- et al. Development of a multiple sclerosis functional composite as a fampridine phosphate rather than adverse events, clinical trial outcome measure. Brain 1999;122(Pt 5):871–882. 17. Catalyst Pharmaceutical Partners. A Phase 3, Multicenter, Double- as noted in previous studies. No serious cardiac blind, Placebo-controlled Randomized Discontinuation Study Fol- abnormalities were detected in this study, confirm- lowed by an Open-label Extension Period to Evaluate the Efficacy ing the previous BioMarin study of amifampridine and Safety of Amifampridine phosphate (3,4-Diaminopyridine Phos- 29 phate) in Patients with Lambert-Eaton Myasthenic Syndrome phosphate in 59 healthy volunteers. There has (LEMS). Clinical Study Protocol, page 74. Catalyst Pharmaceutical been 1 case report of prolonged QTc interval with Partners, Inc. Coral Gables, Florida 30 18. Sanders DB, Howard JF, Massey JM. 3,4-diaminopyridine in the treat- 60-mg 3,4-DAP base in a LEMS patient. ment of Lambert-Eaton myasthenic syndrome and myasthenia gravis. A limitation of our trial was that the protocol Ann N Y Acad Sci 1993;681:588–590. 19. Oh SJ, Claussen GG, Hatanaka Y, Morgan MB. 3,4-diaminopyridine is was not followed in 10 patients as discussed above. more effective than placebo in a randomized, double-blind cross-over From these findings, we conclude that this drug study in LEMS. Muscle Nerve 2009;40(5):795–800. 20. Sanders DB, Massey JM, Sander LL, Edwards LJ. A randomized trial study of amifampridine phosphate (FirdapseVR ) of 3,4-diaminopyridine in Lambert-Eaton syndrome. Neurology 2005; provides Class I evidence of efficacy as a sympto- 54:603–607. 21. Keogh M, Sedehizaden S, Maddison P. Treatment for Lambert-Eaton matic treatment for LEMS with an acceptable toler- myasthenic syndrome. Cochrane Database Syst Rev 2011;(2): ability and safety profile. CD003279. 22. Quartel A, Turbeville S Lounsbury D. Current therapy for Lambert– Eaton myasthenic syndrome: development of 3,4-diaminopyridine The Study Group wishes to acknowledge and thank all the site phosphate salt as first-line symptomatic treatment. Current Medical coordinators and personnel, without whom this study would not Research & Opinion 2010;26:1363–1375. have been possible. 23. Thomsen RH. Wilson DF. Effects of 4-aminopyridine and 3,4-diami- nopyridine on transmitter release at the neuromuscular junction. REFERENCES J Pharmacol Exp Ther 1983;227:260–265. 1. O’Neill JH, Murray NM, Newsom-Davis J. The Lambert-Eaton myas- 24. Lundh H, Nilsson O, Rosen I. Novel drug of choice in Eaton-Lambert thenic syndrome. A review of 50 cases. Brain 1988;11 (Pt 3):577–596. syndrome. J Neurol Neurosurgery Psychiatry 1983;46:684–685. 2. Sanders DB. Lambert-Eaton myasthenic syndrome: diagnosis and 25. Lundh H, Nisson O, Rosen I, Johansson S.. Practical aspects of 3,4- treatment. Ann NY Acad Sci 2003;998:500–508 diaminopyridine treatment in the Lambert-Eaton myasthenic syn- 3. Newsom-Davis J. The emerging diversity of neuromuscular junction drome. Acta Neurol Scand 1993;88:136–140. disorders. Acta Myol 2007;26(1):5–10. 26. McEvoy KM, Windebank AJ, Daube JR, Low PA. 3,4-diaminopyridine 4. Lambert EH, Elmqvist D. Quantal components of end-plate poten- in the treatment of Lambert-Eaton myasthenic syndrome. N Engl J tials in the myasthenic syndrome. Ann NY Acad Sci 1971;183: Med 1989:321:1567–1571. 183–199. 27. Witz PW, Verschuuren JJ, van Dijk JG, de Kam ML, Schomaker RC, 5. Gutmann L, Phillips LH, Gutmann L. Trends in the association of van Hasselt JG, et al. Efficacy of 3,4 diaminopyridine and pyridostig- Lambert-Eaton myasthenic syndrome with carcinoma. Neurology mine in the treatment of Lambert-Eaton myasthenic syndrome: a 1992;42:848–850. randomized, double-blind, placebo-controlled, cross-over study. Cline 6. Oh SJ, Sieb JP. Update on amifampridine phosphate as a drug of Pharmacol Ther 2009;86:44–48. choice in Lambert-Eaton myasthenic syndrome. US Neurology 2014; 28. Sanders D, Harati Y, Smith AG, Peltler A, Marburger T, Lou JS, et al. 10(2):83–89. The DAPPER study team. Results from the dapper study: inpatient 7. Green DM, Jones AC, Brain KR. Content variability of active drug double blind, placebo-controlled withdrawal study of 3,4 diaminopyri- substance in compounded oral 3,4-diaminopyridine products. J Clin dine base (3,4-DAP) in subjects with Lambert-Eaton myasthenic syn- Pharm Ther. 2012;37:53–57. drome. 2015 abstract. Muscle Nerve 2015; 52: AANEM annual 8. Raust JA, Goulay-Dufay S, Le Hoang MD, Pradeau D, Guyon F, Do B. meeting abstracts. S13 Stability studies of ionised and non-ionised 3,4-diaminopyridine: 29. Catalyst Pharmaceutical Partners Reports Positive Cardiac Safety hypothesis of degradation pathways and chemical structure of degra- Results of Its Phase 3 Product, Firdapse Tablets. In: GlobeNewsWise. dation products. J Pharm Biomed Anal. 2007;43:83–88. January 8, 2014. Available at http://globenewswire.com/news- 9. Skeie GO, Apostolskib S, Evolic A, Gilhus LE, Illa I, Harms L., et al. release/2014/01/08/601054/10063346/en/Catalyst-Pharmaceutical- Guidelines for treatment of autoimmune neuromuscular transmis- Partners-Reports-Positive-Cardiac-Safety-Results-of-Its-Phase-3-Product- sion disorders. Gilhus LE, Eur J Neurol 2010;17:893–902. Firdapse-Tablets.html. Accessed 2015 May 22. 10. Oh SJ, Kurokawa K, Claussen GC, Ryan HF Jr. Electrophysiological 30. Kanzato N, Motomura M, Suehara M, Arimua K. Lambert-Eaton diagnostic criteria of Lambert-Eaton myasthenic syndrome. Muscle myasthenic syndrome with ophthalmoparesis and pseudoblepharo- Nerve 2005;32:515–520. spasm. Muscle Nerve 1999:22:1727–1730.

Amifampridine phosphate in LEMS MUSCLE & NERVE Month 2016 9