Contributions
Abstract: P546
Type: Poster Sessions
Abstract Category: Pathology and pathogenesis of MS - Biomarkers
Background: Sensory and motor evoked potentials (SEP; MEP) quantitatively measure signal conduction in multiple sclerosis (MS). They may serve as outcomes in clinical trials. Here we estimate sample sizes to detect differences in longitudinal group changes.
Methods: 21 patients with MS (mean age: 47.3 years, median EDSS: 3.0, 74% relapsing, 26% progressive) were evaluated at 2 time points (tp) within a 30 day window with median and tibial SEP and upper (UL) and lower limb (LL) MEP in 3 centers1-3. Five neurophysiologists independently marked all curves using a custom server-based software, EPMark. N20, P40, cortico-muscular-latencies (CxM) and two quantitative EP-scores (qEPS=sum of z-transformed results/number of tests; unit=mean standard deviation) were analyzed. Modified qEPS (mqEPS) comprised all 8 tests, qMEP all 4 CxM.
From mixed effects linear models (outcome: tp 1 and 2 combined) total variance (V) and its components were estimated assuming 2 central readers. Subject related V (Vsub), intra-individual test-retest V (VTRT), standard deviation of assumed longitudinal change (SDlong) and effect size e=SDlong were used to estimate sample sizes necessary to detect differences in mean group change (a=5%, b=90%) using: n=(1.96+1.28)^2*(2*VTRT+SDlong^2)/e^2. SDlong was calculated from the assumed change of Vsub over time (proportion of Vsub not explained by its inter-correlation over time: SDlong=sqrt[4*Vsub*(1-r)] with r=0.85 for correlation between true underlying baseline and follow-up values). High effect size is justified as it relates to SDlong and not to the larger observed longitudinal variance SDlong+VTRT.
Results: For single EP tests, estimated sample sizes range between 50 and 60 subjects per arm to detect group differences in longitudinal change d ranging from 0.7 to 4.2ms (MEP-UL: n=59, d=2.4ms; MEP-LL: n=52, d=4.2ms; N20: n=59, d=0.7ms; P40: n=55, d=3.7ms). For qEPS, sample sizes and d are smaller (mqEPS: n=54, d=1.2; qMEP: n=45, d=1.7).
Conclusion: Our results corroborate the approach of using multimodal EP to assess therapeutic effects of remyelinating substances: the estimated sample sizes allow conducting efficient trials, the differences in mean group change are reasonable provided that an efficient drug shortens latencies. Multimodal EP optimizes sensitivity to change and has high construct validity.
Disclosure: The study was supported by an unconditional research grant (investigator-initiated trial grant) from Biogen Inc., Cambridge, Massachusetts, USA.
Martin Hardmeiers institution has received fees from Roche for consultancy services; his research is or was supported by the Swiss Multiple Sclerosis Society and the Swiss National Science Foundation SPUM 33CM30_124115 and 33CM30_140338. Dr F Jacques has received honorariums and unrestricted grants from Biogen, Sanofi-Genzyme, Merck-Serono for participating in ad boards, doing presentations and for conducting investigator initiated trials. Philipp Albrecht reports grants, personal fees and non-financial support from Allergan, Biogen, Ipsen, Merz Pharmaceuticals, Novartis, and Roche, personal fees and non-financial support from Bayer Healthcare, and Merck, and non-financial support from Sanofi-Aventis/Genzyme. Habib Bousleiman has no conflict of interest to declare. Christian Schindler has no conflicts of interest to disclose. L. Leocani received honoraria for consulting services from Merck, Roche, Biogen and for speaking activities from Teva; research support from Merck, Biogen, Novartis; travel support from Merck, Roche, Biogen, Almirall. Peter Fuhr´s research is or was supported by the Swiss National Science Foundation SPUM 33CM30_124115 and 33CM30_140338 (PI), Swiss Multiple Sclerosis Society, Synapsis Foundation, Parkinson Schweiz, Novartis Research Foundation, Gossweiler Foundation, Freiwillige Akademische Gesellschaft Basel, Mach-Gaensslen-Stiftung, Botnar Foundation, Bangerter Foundation, and by unconditional research grants from industry (Roche, AbbVie, Biogen, General Electrics).
Abstract: P546
Type: Poster Sessions
Abstract Category: Pathology and pathogenesis of MS - Biomarkers
Background: Sensory and motor evoked potentials (SEP; MEP) quantitatively measure signal conduction in multiple sclerosis (MS). They may serve as outcomes in clinical trials. Here we estimate sample sizes to detect differences in longitudinal group changes.
Methods: 21 patients with MS (mean age: 47.3 years, median EDSS: 3.0, 74% relapsing, 26% progressive) were evaluated at 2 time points (tp) within a 30 day window with median and tibial SEP and upper (UL) and lower limb (LL) MEP in 3 centers1-3. Five neurophysiologists independently marked all curves using a custom server-based software, EPMark. N20, P40, cortico-muscular-latencies (CxM) and two quantitative EP-scores (qEPS=sum of z-transformed results/number of tests; unit=mean standard deviation) were analyzed. Modified qEPS (mqEPS) comprised all 8 tests, qMEP all 4 CxM.
From mixed effects linear models (outcome: tp 1 and 2 combined) total variance (V) and its components were estimated assuming 2 central readers. Subject related V (Vsub), intra-individual test-retest V (VTRT), standard deviation of assumed longitudinal change (SDlong) and effect size e=SDlong were used to estimate sample sizes necessary to detect differences in mean group change (a=5%, b=90%) using: n=(1.96+1.28)^2*(2*VTRT+SDlong^2)/e^2. SDlong was calculated from the assumed change of Vsub over time (proportion of Vsub not explained by its inter-correlation over time: SDlong=sqrt[4*Vsub*(1-r)] with r=0.85 for correlation between true underlying baseline and follow-up values). High effect size is justified as it relates to SDlong and not to the larger observed longitudinal variance SDlong+VTRT.
Results: For single EP tests, estimated sample sizes range between 50 and 60 subjects per arm to detect group differences in longitudinal change d ranging from 0.7 to 4.2ms (MEP-UL: n=59, d=2.4ms; MEP-LL: n=52, d=4.2ms; N20: n=59, d=0.7ms; P40: n=55, d=3.7ms). For qEPS, sample sizes and d are smaller (mqEPS: n=54, d=1.2; qMEP: n=45, d=1.7).
Conclusion: Our results corroborate the approach of using multimodal EP to assess therapeutic effects of remyelinating substances: the estimated sample sizes allow conducting efficient trials, the differences in mean group change are reasonable provided that an efficient drug shortens latencies. Multimodal EP optimizes sensitivity to change and has high construct validity.
Disclosure: The study was supported by an unconditional research grant (investigator-initiated trial grant) from Biogen Inc., Cambridge, Massachusetts, USA.
Martin Hardmeiers institution has received fees from Roche for consultancy services; his research is or was supported by the Swiss Multiple Sclerosis Society and the Swiss National Science Foundation SPUM 33CM30_124115 and 33CM30_140338. Dr F Jacques has received honorariums and unrestricted grants from Biogen, Sanofi-Genzyme, Merck-Serono for participating in ad boards, doing presentations and for conducting investigator initiated trials. Philipp Albrecht reports grants, personal fees and non-financial support from Allergan, Biogen, Ipsen, Merz Pharmaceuticals, Novartis, and Roche, personal fees and non-financial support from Bayer Healthcare, and Merck, and non-financial support from Sanofi-Aventis/Genzyme. Habib Bousleiman has no conflict of interest to declare. Christian Schindler has no conflicts of interest to disclose. L. Leocani received honoraria for consulting services from Merck, Roche, Biogen and for speaking activities from Teva; research support from Merck, Biogen, Novartis; travel support from Merck, Roche, Biogen, Almirall. Peter Fuhr´s research is or was supported by the Swiss National Science Foundation SPUM 33CM30_124115 and 33CM30_140338 (PI), Swiss Multiple Sclerosis Society, Synapsis Foundation, Parkinson Schweiz, Novartis Research Foundation, Gossweiler Foundation, Freiwillige Akademische Gesellschaft Basel, Mach-Gaensslen-Stiftung, Botnar Foundation, Bangerter Foundation, and by unconditional research grants from industry (Roche, AbbVie, Biogen, General Electrics).