Open-label treatment period Pre-randomisation period Screening period Baseline period Fingolimod 0.5 mg iDMT 4 weeks ±7 days Day 0 Days 1–336/12 months (visits at months 1, 3, 6, 9 and 12) Randomisation (1:1) Fingolimod 0.5 mg GA IFNβ 0 0.5 1.0 1.5 2.0 2.5 0 0.5 1.0 1.5 2.0 2.5 n=382 n=370 n=194 n=127 p=0.2610 p=0.8151 n=167 n=119 p=0.0001 p=0.0879 a) b) BVL from Baseline (%) BVL from Baseline (%) Fingolimod 0.5 mg GA IFNβ 0 0.5 1.0 1.5 2.0 0 0.5 1.0 1.5 2.0 n=348 n=327 a) b) cGMVL from Baseline (%) cGMVL from Baseline (%) n=160 n=53 p=0.4877 p=0.4391 n=137 n=56 p=0.0001 p=0.0015 Fingolimod 0.5 mg GA IFNβ –4 2 4 6 8 –4 2 0 0 –2 –2 4 6 8 n=73 n=73 a) b) Score change from Baseline Score change from Baseline n=37 n=35 n=30 n=30 p=0.1324 p=0.0446 p=0.7674 p=0.7852 Introduction • Various injectable disease-modifying therapies (iDMTs) are available for relapsing forms of multiple sclerosis (RMS), including glatiramer acetate (GA), interferon (IFN) β-1a (for intramuscular [IM] or subcutaneous [SC] injection) and IFNβ-1b (for SC injection) • Adherence to these iDMTs is poor among patients with RMS, which can reduce their effectiveness 1,2 • Fingolimod 0.5 mg is a once-daily oral therapy approved for RMS 3 • PREFERMS was a 12-month, Phase 4, open-label, prospective study designed specifically to compare retention rates in patients with RMS treated with either fingolimod or an iDMT (GA or an IFNβ) 4 • During the study, one treatment switch was allowed. Effects of treatment could thus be analysed across two different periods – Baseline to end of randomised treatment (EoRT): the period during which patients received only their original randomised treatment – Baseline to end of study (EoS): includes time on original randomised treatment plus any treatment switch Objective • Compare radiological and cognitive outcomes in patients randomised to fingolimod, GA or IFNβ before (at EoRT) and after (at EoS) any treatment switches Methods Study design • PREFERMS was a 12-month, Phase 4, open-label, active-controlled, randomised, multicentre study conducted at 117 sites in the USA • Primary endpoint was patient retention on randomised treatment over 12 months • Enrolled patients were either treatment-naïve or had previously received only one class of iDMT (GA or IFNβ) • Patients were randomised 1:1 to either fingolimod or a preselected iDMT (GA, IFNβ-1a IM or IFNβ-1a/b SC) and observed quarterly for 12 months • One on-study treatment switch was allowed after 3 months, or earlier for efficacy or safety reasons (Figure 1) Analyses • Post hoc analyses were conducted in the full analysis set (FAS) at EoRT and EoS to evaluate: – exposure-adjusted percentage brain volume loss (BVL) from Baseline – exposure-adjusted percentage cortical grey matter volume loss (cGMVL) from Baseline – change from Baseline in oral Symbol Digit Modalities Test (SDMT) scores • Analyses were performed according to patients’ original randomised treatment groups and were for hypothesis generation only (the study was not powered to detect treatment effects related to switching) • No adjustment was made for multiple comparisons Radiological and Cognitive Outcomes Among Patients Randomised to Interferon Beta, Glatiramer Acetate or Fingolimod in PREFERMS Samuel F. Hunter 1 , Bruce A.C. Cree 2 , Xiangyi Meng 3 , Lesley Schofield 3 , Scott Kolodny 3 , Nadia Tenenbaum 3 , Florian P. Thomas 4,5 on behalf of the PREFERMS investigators 1 Advanced Neurosciences Institute, Franklin, TN, USA; 2 University of California San Francisco, San Francisco, CA, USA; 3 Novartis Pharmaceuticals Corporation, East Hanover, NJ, USA; 4 Seton Hall-Hackensack Meridian School of Medicine, South Orange, NJ, USA; 5 Hackensack University Medical Center, Hackensack, NJ, USA EP1663 References 1. Devonshire V, et al. Eur J Neurol 2011;18:69–77 2. Patti F. Patient Prefer Adherence 2010;4:1–9 3. Novartis Pharmaceuticals Corporation. Prescribing information. Gilenya ® . 2016. Available from: https://www.pharma.us.novartis.com/product/pi/pdf/gilenya.pdf (Accessed 14 August 2017) 4. Cree B, et al. Neurology 2016;86(Suppl 16):P3.115 Acknowledgements Editorial support was provided by Oxford PharmaGenesis, Oxford, UK, funded by Novartis Pharmaceuticals Corporation Disclosures Samuel F. Hunter has received consulting fees and/or research support from and/or served on speakers’ bureaux for Acorda Therapeutics, Avanir, Bayer HealthCare Pharmaceuticals, Novartis, Osmotica, Roche/Genentech, Sanofi Genzyme and Teva Neuroscience. Bruce A.C. Cree has received personal compensation for consulting from AbbVie, Biogen, EMD Serono, GeNeuro and Novartis. Xiangyi Meng, Lesley Schofield, Scott Kolodny and Nadia Tenenbaum are employees of Novartis Pharmaceuticals Corporation. Florian P. Thomas has served as a speaker and/or consultant for Acorda Therapeutics, Biogen, Genzyme, Novartis and Teva Neuroscience, and has received research support from Biogen and Teva Neuroscience © 2017 Novartis Pharmaceuticals Corporation Poster presented at the 7 th Joint Congress of the European Committee for Treatment and Research in Multiple Sclerosis–Americas Committee for Treatment and Research in Multiple Sclerosis, October 25–28, 2017, Paris, France This study was sponsored by Novartis Pharmaceuticals Corporation, United States Table 1. PREFERMS patient demographics and Baseline characteristics in the FAS Characteristic Fingolimod 0.5 mg (n=433) GA (n=231) p-value IFNβ (n=197) p-value Age, years 41.6 (10.86) 42.2 (10.31) 0.4492 41.1 (10.48) 0.6124 Sex, n (%) Male Female 125 (28.9) 308 (71.1) 61 (26.4) 170 (73.6) 0.5011 46 (23.4) 151 (76.6) 0.1488 Race, n (%) Caucasian Black Asian Native American Pacific Islander Other 354 (81.8) 67 (15.5) 1 (0.2) 1 (0.2) 0 10 (2.3) 188 (81.4) 36 (15.6) 1 (0.4) 1 (0.4) 0 5 (2.2) ND 159 (80.7) 33 (16.8) 0 0 2 (1.0) 3 (1.5) ND Height, cm 168.5 (9.02) 167.4 (10.50) 0.1918 167.7 (9.07) 0.3578 Weight, kg 82.87 (20.70) 84.74 (23.06) 0.2780 82.28 (21.75) 0.7388 BMI, kg/m 2 29.17 (6.68) 30.24 (7.89) 0.0657 29.17 (7.15) 0.9998 Duration of MS since diagnosis, years 4.43 (6.67) 4.70 (6.17) 0.6110 3.67 (5.68) 0.1661 Duration of MS since first symptom, years 7.30 (8.21) 7.51 (7.86) 0.7490 6.81 (7.43) 0.4727 Number of relapses in the past year 0.6 (0.95) 0.5 (0.85) 0.4972 0.6 (1.06) 0.9111 Number of relapses in the past 2 years 0.9 (1.51) 0.9 (1.34) 0.9335 0.9 (1.50) 0.7691 EDSS score 2.36 (1.56) 2.37 (1.53) ND 2.52 (1.48) ND T2 lesion volume, cm 3 7.65 (11.60) 7.35 (8.88) ND 7.55 (11.53) ND Normalised brain volume, cm 3 1521.42 (83.91) 1499.49 (97.82) ND 1524.85 (79.17) ND Number of Gd+ lesions 1.08 (3.75) 0.63 (2.38) ND 1.11 (3.64) ND Data are mean (SD) unless stated otherwise Between-group comparisons were made using the χ 2 test (excluding missing values) for categorical variables and a two-sample t-test for continuous variables comparing either GA or IFNβ with fingolimod 0.5 mg BMI, body mass index; EDSS, Expanded Disability Status Scale; FAS, full analysis set; GA, glatiramer acetate; Gd+, gadolinium enhancing; MS, multiple sclerosis; ND, not determined Figure 1. PREFERMS study design Figure 2. Exposure-adjusted BVL from Baseline to a) EoRT and b) EoS Figure 3. Exposure-adjusted cGMVL from Baseline to a) EoRT and b) EoS Figure 4. Change from Baseline in oral SDMT scores at a) EoRT and b) EoS Patients were allowed one switch from randomised treatment Reason for switch within 3 months: safety or efficacy Reason for switch at 3–12 months: safety, efficacy, tolerability or convenience iDMT, injectable disease-modifying therapy Data are mean (SD) Rank analysis of covariance adjusted for treatment, treatment-naïvety, corresponding Baseline values and age BVL, brain volume loss; EoRT, end of randomised treatment; EoS, end of study; GA, glatiramer acetate; IFN, interferon; SD, standard deviation Data are mean (SD) Rank analysis of covariance adjusted for treatment, treatment-naïvety, corresponding Baseline values and age cGMVL, cortical grey matter volume loss; EoRT, end of randomised treatment; EoS, end of study; GA, glatiramer acetate; IFN, interferon; SD, standard deviation Data are least-squares mean (95% CI) Analysis of covariance adjusted for treatment, treatment-naïvety, corresponding Baseline values and age. Higher scores indicate better performance CI, confidence interval; EoRT, end of randomised treatment; EoS, end of study; GA, glatiramer acetate; IFN, interferon; SDMT, Symbol Digit Modalities Test Results • 875 patients were randomised and 861 (98.4%) were included in the FAS (fingolimod, n=433; GA, n=231; IFNβ, n=197) • Demographics and Baseline characteristics were similar across the three treatment groups (Table 1) • The majority of patients (58.1%) randomised to either GA or IFNβ switched to fingolimod during the study, whereas only 6.2% of the fingolimod group switched to an iDMT Percentage BVL from Baseline • EoRT: mean percentage BVL was lowest in the fingolimod group, followed by the GA, then IFNβ group (0.48%, 0.67% and 0.85%, respectively). BVL in IFNβ was significantly greater than the fingolimod group (Figure 2a) – Median (interquartile range [IQR]) percentage BVL values in the three groups were 0.37% (–0.94, 0.11), 0.46% (–1.19, 0.17) and 0.85% (–1.49, –0.23), respectively • EoS: mean percentage BVL was similar in the fingolimod and GA groups (0.50% and 0.52%, respectively), and had reduced to 0.56% in the IFNβ group. There were no significant differences between groups (Figure 2b) – Median (IQR) percentage BVL values in the three groups were 0.40% (–0.94, 0.09), 0.40% (–0.89, 0.05) and 0.59% (–1.00, 0.02), respectively Percentage cGMVL from Baseline • EoRT: mean cGMVL was numerically lowest in the GA group, followed by the fingolimod and then IFNβ groups (0.08%, 0.09% and 0.52%, respectively; Figure 3a) – Median (IQR) cGMVL was 0.02% (–0.45, 0.40), 0.12% (–0.45, 0.24) and 0.48% (–1.02, 0.00), for fingolimod, GA and IFNβ groups, respectively • EoS: mean cGMVL followed the same pattern among the three groups as at EoRT (0.12%, 0.13% and 0.34%, respectively; Figure 3b) – Median (IQR) cGMVL values were 0.03% (–0.49, 0.35), 0.12% (–0.55, 0.44) and 0.28% (–0.74, 0.19) in the fingolimod, GA and IFNβ groups, respectively Change in oral SDMT scores from Baseline • EoRT: the greatest improvements in oral SDMT scores were observed in the fingolimod group (Figure 4a) – Least-squares mean (95% confidence interval [CI]) score change was 3.4 (1.4, 5.4) with fingolimod and 1.0 (–1.6, 3.6) with GA – Scores worsened with IFNβ during the same time frame (–0.7 [–4.0, 2.6]) • EoS: oral SDMT scores had improved to a similar extent in the fingolimod and GA groups, and an improvement was also observed in the IFNβ group (3.2 [1.1, 5.3], 2.7 [0.0, 5.4] and 2.5 [–1.0, 6.1], respectively; Figure 4b) Conclusions • During the randomised phase of the PREFERMS study, patients receiving fingolimod tended to have less BVL, less cGMVL and better cognitive outcomes than those receiving IFNβ or GA • At EoS, improvements in BVL and cGMVL in the IFNβ group and in cognition in the IFNβ and GA groups may have been due to the high proportion of patients switching to fingolimod Scan QR code to download this poster and view videos of PREFERMS primary data and of PREFERMS switch data. Also available at: http://novartis.medicalcongressposters.com/Default.aspx?doc=cfb72 And via Text Message (SMS) Text: Qcfb72 To: 8NOVA (86682) US only +18324604729 North, Central and South Americas; Caribbean; China +447860024038 UK, Europe and Russia +46737494608 Sweden and Europe Copies of this poster obtained through the QR (Quick Response) code are for personal use only and may not be reproduced without written permission of the authors Presenter email address: sfhunter@neurosci.us DOWNLOAD THIS POSTER AND VIEW VIDEOS OF PREFERMS PRIMARY DATA AND OF PREFERMS TREATMENT SWITCH DATA Scan to view a video presentation