Evoked potentials in clinical trials for multiple sclerosis

DELMEP: a deep learning algorithm for automated annotation of motor evoked potential latencies

The analysis of motor evoked potentials (MEPs) generated by transcranial magnetic stimulation (TMS) is crucial in research and clinical medical practice. MEPs are characterized by their latency and the treatment of a single patient may require the characterization of thousands of MEPs. Given the difficulty of developing reliable and accurate algorithms, currently the assessment of MEPs is performed with visual inspection and manual annotation by a medical expert; making it a time-consuming, inaccurate, and error-prone process. In this study, we developed DELMEP, a deep learning-based algorithm to automate the estimation of MEP latency. Our algorithm resulted in a mean absolute error of about 0.5 ms and an accuracy that was practically independent of the MEP amplitude. The low computational cost of the DELMEP algorithm allows employing it in on-the-fly characterization of MEPs for brain-state-dependent and closed-loop brain stimulation protocols. Moreover, its learning ability makes it a particularly promising option for artificial-intelligence-based personalized clinical applications.

Motor evoked potentials for multiple sclerosis, a multiyear follow-up dataset

Multiple sclerosis (MS) is a chronic disease affecting millions of people worldwide. Through the demyelinating and axonal pathology of MS, the signal conduction in the central nervous system is affected. Evoked potential measurements allow clinicians to monitor this process and can be used for decision support. We share a dataset that contains motor evoked potential (MEP) measurements, in which the brain is stimulated and the resulting signal is measured in the hands and feet. This results in time series of 100 milliseconds long. Typically, both hands and feet are measured in one hospital visit. The dataset contains 5586 visits of 963 patients, performed in day-to-day clinical care over a period of 6 years. The dataset consists of approximately 100,000 MEP. Clinical metadata such as the expanded disability status scale, sex, and age is also available. This dataset can be used to explore the role of evoked potentials in MS research and patient care. It may also be used as a benchmark for time series analysis and predictive modelling.

Measurement(s)	Abnormal upper-limb motor evoked potentials • Abnormal lower-limb motor evoked potentials • Kurtzke Expanded Disability Status Scale Clinical Classification
Technology Type(s)	Transcranial Magnetic Stimulation • performing a clinical assessment
Factor Type(s)	Date of birth • Sex • Measurement date
Sample Characteristic - Organism	Homo sapiens
Sample Characteristic - Location	Belgium

Deciphering the Morphology of Motor Evoked Potentials

Motor Evoked Potentials (MEPs) are used to monitor disability progression in multiple sclerosis (MS). Their morphology plays an important role in this process. Currently, however, there is no clear definition of what constitutes a normal or abnormal morphology. To address this, five experts independently labeled the morphology (normal or abnormal) of the same set of 1,000 MEPs. The intra- and inter-rater agreement between the experts indicates they agree on the concept of morphology, but differ in their choice of threshold between normal and abnormal morphology. We subsequently performed an automated extraction of 5,943 time series features from the MEPs to identify a valid proxy for morphology, based on the provided labels. To do this, we compared the cross-validation performances of one-dimensional logistic regression models fitted to each of the features individually. We find that the approximate entropy (ApEn) feature can accurately reproduce the majority-vote labels. The performance of this feature is evaluated on an independent test set by comparing to the majority vote of the neurologists, obtaining an AUC score of 0.92. The model slightly outperforms the average neurologist at reproducing the neurologists consensus-vote labels. We can conclude that MEP morphology can be consistently defined by pooling the interpretations from multiple neurologists and that ApEn is a valid continuous score for this. Having an objective and reproducible MEP morphological abnormality score will allow researchers to include this feature in their models, without manual annotation becoming a bottleneck. This is crucial for large-scale, multi-center datasets. An exploratory analysis on a large single-center dataset shows that ApEn is potentially clinically useful. Introducing an automated, objective, and reproducible definition of morphology could help overcome some of the barriers that are currently obstructing broad adoption of evoked potentials in daily care and patient follow-up, such as standardization of measurements between different centers, and formulating guidelines for clinical use.

Electrophysiological markers and predictors of the disease course in primary progressive multiple sclerosis

Background:

Currently no valid surrogate marker exists for primary progressive multiple sclerosis (PPMS).

Objective:

Our aim was to prospectively investigate multimodal evoked potentials (EPs) as markers and predictors of the disease course in PPMS.

Methods:

Twenty-two PPMS patients were prospectively examined with visual, somatosensory and motor EPs and Expanded Disability Status Scale (EDSS) assessments at baseline (T0) and at six-month intervals over three years. Spearman rank correlation was used to determine the relationship between EP measures and EDSS. The relationship between disease evolution and a numerical score derived from z-transformed EP-latencies ( s-EP-Q) and baseline characteristics was further assessed using multivariable linear regression analysis.

Results:

s-EP-Q correlated with EDSS score at all points in time in cross-sectional comparison (0.53≤rs ≤0.68; 0.0007≤p≤0.0232) and also longitudinally by trend ( rs=0.46, p=0.0740). The s-EP-QT0 correlated with the EDSS score at year 3 (T6) ( rs=0.77, p<0.0001). The s-EP-Q changes became statistically significant six months before corresponding changes were seen in the EDSS score. EDSST6 as predicted by EDSST6= −1.027+0.037* age+0.217* s-EP-QT0 + 0.695* EDSST0 correlated with the observed values ( rs=0.92, p<0.0001).

Conclusions:

Multimodal EPs correlate well with disability in PPMS, and allow some prediction of the disease course over three years. These findings support a role of EPs as surrogate markers in clinical trials in PPMS.

Prediction of long-term disability in multiple sclerosis

BACKGROUND

Little is known about the predictive value of neurophysiological measures for the long-term course of multiple sclerosis (MS).

OBJECTIVE

To prospectively investigate whether combined visual (VEP) and motor evoked potentials (MEP) allow prediction of disability over 14 years.

METHODS

A total of 30 patients with relapsing-remitting and secondary progressive MS were prospectively investigated with VEPs, MEPs and the Expanded Disability Status Scale (EDSS) at entry (T0) and after 6, 12 and 24 months, and with cranial MRI scans at entry (T2-weighted and gadolinium-enhanced T1-weighted images). EDSS was again assessed at year 14 (T4). The association between evoked potential (EP), magnetic resonance (MR) data and EDSS was measured using Spearman's rank correlation. Multivariable linear regression was performed to predict EDSS(T4) as a function of z-transformed EP-latencies(T0). The model was validated using a jack-knife procedure and the potential for improving it by inclusion of additional baseline variables was examined.

RESULTS

EDSS values(T4) correlated with the sum of z-transformed EP-latencies(T0) (rho = 0.68, p < 0.0001), but not with MR-parameters(T0). EDSS(T4) as predicted by the formula EDSS(T4) = 4.194 + 0.088 * z-score P100(T0) + 0.071 * z-score CMCT(UE, T0) correlated with the observed values (rho = 0.69, p < 0.0001).

CONCLUSION

Combined EPs allow prediction of long-term disability in small groups of patients with MS. This may have implications for the choice of monitoring methods in clinical trials and for daily practice decisions.

Prospective clinical and electrophysiological follow-up on a multiple sclerosis population treated with interferon beta-1 a

Objective To analyse transcranial magnetic stimulation (TMS) variables in a prospective six-month follow-up pilot study on patients suffering from relapsing-remitting multiple sclerosis (RRMS), satisfying inclusion criteria for interferon (IFN) beta-1a treatment.

Background So far, no predictive factors are available as to the course of RRMS treated with IFN beta-1 a.

Design/methods Fifteen RRMS patients were studied before (month 0 (M0)) and after IFN beta-1a onset (M3, M6). The parameters analysed were motor functional score (mFS), Expanded Disability Status Scale (EDSS), and TMS variables - central motor conduction time (CMCT) and amplitude ratio (AR).

Results Four of the six patients with no motor signs at inclusion, subsequently showed signs of pyramidal dysfunction. All had abnormal M0_TMS variables. The number of M0_TMS abnormalities per patient was greatest in the group that showed mFS worsening, and was significantly correlated with M6_EDSS. The M0_CMCT was significantly correlated with M6_EDSS. During follow-up, the number of patients with abnormal TMS variables decreased from 12/15 to 4/15, and the total number of abnormalities decreased from 33.3 to 16.7%.

Conclusions TMS variables might be predictive of disease progression. The improvement observed here in the TMS variables may reflect an improvement in MS patients undergoing IFN beta treatment. Multiple Sclerosis 2007; 13: 348-356. http://msj.sagepub.com

Pattern reversal visual evoked potentials as a measure of visual pathway pathology in multiple sclerosis

BACKGROUND

Pattern reversal visual evoked potentials (PRVEPs) have a well-documented role in diagnosis of multiple sclerosis (MS), but their value as a visual function surrogate remains controversial.

METHODS

We evaluated PRVEP in 37 patients with MS who were participating in a long-term follow-up study following a phase III trial of interferon beta-1a (Avonex). Patients were examined to determine the Kurtzke Extended Disability Status Score (EDSS), multiple sclerosis functional composite (MSFC), contrast letter acuity (CLA), and had cranial MRI scans to determine whole brain atrophy (BPF). PRVEP was evaluated for P100 latency, amplitude, and waveform morphology. Two summary scores were created: for Score A, abnormal latencies, morphologies, and amplitudes of each individual eye were added; for Score B, abnormal latencies, morphologies, and amplitude ratio between eyes was determined. Sixteen patients in this group also had PRVEP at the time they enrolled in the clinical trial, eight years previously.

RESULTS

At the follow-up exam, over 75% of patients had abnormal PVEP parameters while visual acuity (VA) was abnormal only in 59%. Increased PRVEP latency over the eight-year period correlated with deterioration assessed by EDSS (P = 0.006), BPF (P = 0.0001), and MSFC (P = 0.0041). Score A was significantly correlated with EDSS, BPF, CLA, cognitive function, and quality of life assessed with the Sickness Impact profile. No correlation was seen with the MSFC.

CONCLUSIONS

The results indicate that PRVEP measures MS-related pathology, and can provide not only diagnostic but also prognostic information during evaluation of MS patients.

Evoked potentials for evaluation of multiple sclerosis

The role of evoked potentials (EP) in the assessment of multiple sclerosis (MS) has changed over the last decade. This is largely due to progress in imaging techniques. But while MRI has a greater diagnostic sensitivity, EP remain a useful diagnostic tool in many clinical situations. Moreover, recent studies demonstrate the utility of EP for monitoring and predicting the course of the disease in patient groups, although not yet in individuals. For these purposes, EP show better results than conventional MRI. In the near future, new developments in electrophysiology, immunology and imaging may allow to differentiate between different subtypes of MS early in the course, and consequently to tailor therapeutic measures more precisely to the individual patients.

Neurophysiological investigations in multiple sclerosis

The advent of magnetic resonance imaging techniques has greatly reduced the diagnostic value of neurophysiological tests, particularly evoked potentials, in multiple sclerosis patients, because of the higher sensitivity in revealing subclinical involvement of the central nervous system. Technical progress and new methods of investigating afferent and efferent nervous pathways would seem to increase the sensitivity in detecting neural dysfunction, but the 'clinical gain' is modest at best. More promising is the utilization of neurophysiological tests to quantify the severity of white matter involvement. Transversal and longitudinal studies have demonstrated good correlations between neurophysiological parameters and disability measures, indicating that a battery of neurophysiological tests could be useful in monitoring the disease evolution in single patients and as surrogate endpoints in clinical trials. Further studies are needed for a better definition of the applications of evoked potentials and other neurophysiological techniques. Finally, event-related potentials and advanced electroencephalogram techniques, such as coherence analysis, could provide useful information on the pathophysiology of cognitive dysfunction, so common in multiple sclerosis patients, and with a strong impact on the quality of life.

Serial evoked potential studies and MRI imaging in chronic progressive multiple sclerosis

Measurements of serial evoked potential latencies and plaque burden on MRI scans are often obtained during clinical studies of multiple sclerosis patients to provide additional information to the disability-based primary endpoints. The ideal laboratory-based marker of progression would be expected to significantly change over the time period of study. Serial visual (VEP) and brainstem auditory evoked potentials (BAEP) and MRI scans of 11 chronic progressive MS patients were obtained over a 1.5 year period in a clinical study. Over this period, there was no significant change in disability as measured by the Kurtzke EDSS, Ambulation Index or Neurological Rating Score. The VEP P100 significantly progressed over the period of study. However, the MRI T(2) plaque burden and BAEP I-V intrapeak latency did not significantly progress over the 1.5 years. We conclude that, in chronic progressive MS, serial visual evoked potential tests may complement standard disability-based endpoints to assess disease progression.