Diagnosis and management of multiple sclerosis: MRI in clinical practice

Evaluation of machine learning-based classification of clinical impairment and prediction of clinical worsening in multiple sclerosis

BACKGROUND

Robust predictive models of clinical impairment and worsening in multiple sclerosis (MS) are needed to identify patients at risk and optimize treatment strategies.

OBJECTIVE

To evaluate whether machine learning (ML) methods can classify clinical impairment and predict worsening in people with MS (pwMS) and, if so, which combination of clinical and magnetic resonance imaging (MRI) features and ML algorithm is optimal.

METHODS

We used baseline clinical and structural MRI data from two MS cohorts (Berlin: n = 125, Amsterdam: n = 330) to evaluate the capability of five ML models in classifying clinical impairment at baseline and predicting future clinical worsening over a follow-up of 2 and 5 years. Clinical worsening was defined by increases in the Expanded Disability Status Scale (EDSS), Timed 25-Foot Walk Test (T25FW), 9-Hole Peg Test (9HPT), or Symbol Digit Modalities Test (SDMT). Different combinations of clinical and volumetric MRI measures were systematically assessed in predicting clinical outcomes. ML models were evaluated using Monte Carlo cross-validation, area under the curve (AUC), and permutation testing to assess significance.

RESULTS

The ML models significantly determined clinical impairment at baseline for the Amsterdam cohort, but did not reach significance for predicting clinical worsening over a follow-up of 2 and 5 years. High disability (EDSS ≥ 4) was best determined by a support vector machine (SVM) classifier using clinical and global MRI volumes (AUC = 0.83 ± 0.07, p = 0.015). Impaired cognition (SDMT Z-score ≤ -1.5) was best determined by a SVM using regional MRI volumes (thalamus, ventricles, lesions, and hippocampus), reaching an AUC of 0.73 ± 0.04 (p = 0.008).

CONCLUSION

ML models could aid in classifying pwMS with clinical impairment and identify relevant biomarkers, but prediction of clinical worsening is an unmet need.

Advanced MRI Techniques: Diagnosis and Follow-Up of Multiple Sclerosis

Brain and spinal cord imaging plays a pivotal role in aiding clinicians with the diagnosis and monitoring of multiple sclerosis. Nevertheless, the significance of magnetic resonance imaging in MS extends beyond its clinical utility. Advanced imaging modalities have facilitated the in vivo detection of various components of MS pathogenesis, and, in recent years, MRI biomarkers have been utilized to assess the response of patients with relapsing-remitting MS to the available treatments. Similarly, MRI indicators of neurodegeneration demonstrate potential as primary and secondary endpoints in clinical trials targeting progressive phenotypes. This review aims to provide an overview of the latest advancements in brain and spinal cord neuroimaging in MS.

Temporal Relationship Between Serum Neurofilament Light Chain and Radiologic Disease Activity in Patients With Multiple Sclerosis

BACKGROUND AND OBJECTIVES

Serum neurofilament light chain (sNfL) levels correlate with multiple sclerosis (MS) disease activity, but the dynamics of this correlation are unknown. We evaluated the relationship between sNfL levels and radiologic MS disease activity through monthly assessments during the 24-week natalizumab treatment interruption period in RESTORE (NCT01071083).

METHODS

In the RESTORE trial, participants with relapsing forms of MS who had received natalizumab for ≥12 months were randomized to either continue or stop natalizumab and followed with MRI and blood draws every 4 weeks to week 28 and again at week 52 The sNfL was measured, and its dynamics were correlated with the development of gadolinium-enhancing (Gd+) lesions. Log-linear trend in sNfL levels were modeled longitudinally using generalized estimating equations with robust variance estimator from baseline to week 28.

RESULTS

Of 175 patients enrolled in RESTORE, 166 had serum samples for analysis. Participants with Gd+ lesions were younger (37.7 vs 43.1, p = 0.001) and had lower Expanded Disability Status Scale scores at baseline (2.7 vs 3.4, p = 0.017) than participants without Gd+ lesions. sNfL levels increased in participants with Gd+ lesions (n = 65) compared with those without (n = 101, mean change from baseline to maximum sNfL value, 12.1 vs 3.2 pg/mL, respectively; p = 0.003). As the number of Gd+ lesions increased, peak median sNfL change also increased by 1.4, 3.0, 4.3, and 19.6 pg/mL in the Gd+ lesion groups of 1 (n = 12), 2-3 (n = 18), 4-9 (n = 21), and ≥10 (n = 14) lesions, respectively. However, 46 of 65 (71%) participants with Gd+ lesions did not increase above the 95th percentile threshold of the group without Gd+ lesions. The initial increase of sNfL typically trailed the first observation of Gd+ lesions, and the peak increase in sNfL was a median [interquartile range] of 8 [0, 12] weeks after the first appearance of the Gd+ lesion.

DISCUSSION

Although sNfL correlated with the presence of Gd+ lesions, most participants with Gd+ lesions did not have elevations in sNfL levels. These observations have implications for the use and interpretation of sNfL as a biomarker for monitoring MS disease activity in controlled trials and clinical practice.

Review of multiple sclerosis: Epidemiology, etiology, pathophysiology, and treatment

Multiple sclerosis (MS) is a chronic autoimmune disease with demyelination, inflammation, neuronal loss, and gliosis (scarring). Our object to review MS pathophysiology causes and treatment. A Narrative Review article was conducted by searching on Google scholar, PubMed, Research Gate about relevant keywords we exclude any unique cases and case reports. The destruction of myelinated axons in the central nervous system reserves this brunt. This destruction is generated by immunogenic T cells that produce cytokines, copying a proinflammatory T helper cells1-mediated response. Autoreactive cluster of differentiation 4 + cells, particularly the T helper cells1 subtype, are activated outside the system after viral infections. T-helper cells (cluster of differentiation 4+) are the leading initiators of MS myelin destruction. The treatment plan for individuals with MS includes managing acute episodes, using disease-modifying agents to decrease MS biological function of MS, and providing symptom relief. Management of spasticity requires physiotherapy, prescription of initial drugs such as baclofen or gabapentin, secondary drug options such as tizanidine or dantrolene, and third-line treatment such as benzodiazepines. To treat urinary incontinence some options include anticholinergic medications such as oxybutynin hydrochloride, tricyclic antidepressants (such as amitriptyline), and intermittent self-catheterization. When it comes to bowel problems, one can try to implement stool softeners and consume a high roughage diet. The review takes about MS causes Pathophysiology and examines current treatment strategies, emphasizing the advancements in disease-modifying therapies and symptomatic treatments. This comprehensive analysis enhances the understanding of MS and underscores the ongoing need for research to develop more effective treatments.

Investigating the feasibility of differentiating MS active lesions from inactive ones using texture analysis and machine learning methods in DWI images

BACKGROUND

Magnetic resonance imaging (MRI) is commonly used in conjunction with a gadolinium-based contrast agent (GBCA) to distinguish active multiple sclerosis (MS) lesions. However, recent studies have raised concerns regarding the long-term effects of the accumulation of GBCA in the body. Thus, the purpose of this study is to investigate the possibility of using texture analysis in diffusion-weighted imaging (DWI) and machine learning algorithms to discriminate active from inactive MS lesions without the use of GBCA.

METHODS

To achieve this purpose, we introduce an image processing pipeline. In the proposed pipeline, following registration and alignment of slices, MS lesions from DWI images are segmented and quantized. Next, different texture analysis methods are employed to extract texture features from the lesions. Then, a two-stage feature reduction method is applied, in which the first stage involves a statistical t-test and the second stage relies on principal component analysis (PCA), sequential forward selection (SFS), sequential backward selection (SBS), and ReliefF algorithms. Finally, we use five classifiers logistic regression (LR), support vector machine (SVM), decision tree (DT), K nearest neighbor (KNN), and linear discriminant analysis (LDA) in a 5-fold cross-validation procedure to determine active and inactive MS lesions.

RESULTS

In this study, we collected and prepared 255 active/inactive MS lesions from MRI scans of 34 patients diagnosed with MS, with a mean age of 35.56±10.89. Among 89 texture features extracted, 63 features showed statistically significant differences between the means of active and inactive lesions (P<0.05). The SVM classifier with the PCA feature reduction algorithm demonstrated the best performance with an average accuracy of 0.960 (±0.024), specificity and precision of 1.0, sensitivity of 0.913 (±0.053), and AUC of 0.957 (±0.027).

CONCLUSION

Our study indicates that DWI changes detected using texture analysis-based machine learning models can precisely differentiate active from inactive MS lesions. This finding provides valuable clinical information for the early diagnosis and effective monitoring of MS disease.

The Role of MRI in Differentiating Demyelinating and Inflammatory (not Infectious) Myelopathies

Demyelinating and inflammatory myelopathies represent a group of diseases with characteristic patterns in neuroimaging and several differential diagnoses. The main imaging patterns of demyelinating myelopathies (multiple sclerosis, neuromyelitis optica spectrum disorder, acute disseminated encephalomyelitis, and myelin oligodendrocyte glycoprotein antibody-related disorder) and inflammatory myelopathies (systemic lupus erythematosus-myelitis, sarcoidosis-myelitis, Sjögren-myelitis, and Behçet's-myelitis) will be discussed in this article, highlighting key points to the differential diagnosis.

Previous disease-modifying treatments influence T lymphocyte kinetics in people with multiple sclerosis switching to ocrelizumab

BACKGROUND

Recently, concern has been raised about the influence of the previous disease-modifying treatments (DMTs) on the clinical efficacy of ocrelizumab (OCR). We aimed to evaluate whether the previous DMT affects the lymphocyte subset kinetics in people with Multiple Sclerosis (MS) switching to OCR.

METHODS

This is a multicenter, retrospective, real-world study on consecutive MS patients who started or switched to OCR. We grouped them by prior DMT in: (i) naïve-to-treatment (NTT), (ii) switching from fingolimod (SF) and (iii) switching from natalizumab (SN). Differences in absolute lymphocyte count and lymphocyte subset count changes, considering the period from baseline to 6 months, over all the three groups were assessed with an inverse-probability-weighted regression adjustment model.

RESULTS

Mean T CD4+ cell count reduction from baseline to the six-month follow-up was more pronounced in the SN group compared to the NTT (p = 0,026). Further, patients in the SF group experienced a less pronounced CD4 T cell number decrease than both NTT and SN groups (p = 0,04 and p < 0,001, respectively). Patients in the SF group experienced an increase in CD8 T cell absolute number, whereas those in the NTT and SN groups experienced a significant decrease (p = 0,015 and p < 0,001, respectively). Patients experiencing early inflammatory activity showed a lower CD8+ cell count at baseline than stable patients (p = 0,02).

CONCLUSIONS

Previous DMTs influence the lymphocyte kinetics in people with MS switching to OCR. Reassessment of these findings over a larger population may help optimize the switch.

Association of vitamin D serum levels and vitamin D supplementation with B cell kinetics and disease activity in Multiple Sclerosis patients treated with ocrelizumab: an Italian multi-center study

BACKGROUND

Several observational studies have shown an association between low circulating levels of 25-hydroxy- vitamin D (25(OH)D) and an increase in inflammatory activity in Multiple Sclerosis (MS). Among its immunomodulatory functions, 25(OH)D suppresses proliferation and immunoglobulin production of B cells. 25(OH)D supplementation has been associated with better radiological outcomes in MS patients treated with interferon (IFN)-B, glatiramer acetate, fingolimod, natalizumab and rituximab. Our study is aimed at analyzing the association of 25(OH)D serum levels and supplementation with B cell kinetics and clinical-radiological outcomes of people with MS treated with ocrelizumab.

METHODS

We have retrospectively collected clinical and radiological data from 136 MS patients who have been treated with ocrelizumab, have undergone at least two treatment cycles and for whom data on serum 25(OH)D levels and intake were available. The patients were divided into three groups according to baseline 25(OH)D serum levels: deficient (≤19,9 ng/ml), insufficient (20-29,9 ng/ml), and normal range 25(OH)D (>30 ng/ml). According to 25(OH)D intake, we divided our population into users and non-users. To explore B cell kinetics at six- and twelve-month follow-ups, the patients were divided into two groups: with fast repopulation (FR) and slow repopulation rate (SR), based on the reappearance or non- appearance of CD19 at each time point.

RESULTS

When considering the entire population, the mean 25(OH)D serum level (sd) was 26.27 ng/ml (14.15). 43 (31,62%) patients were classified as deficient, 52 (38,24%) were classified as insufficient, and 41 (30,14%) showed 25(OH)D serum levels within the normal range. 60.29% (82/136) of the patients were classified as users, and 39.70% (54/136) as non-users. Over the eighteen-month treatment period, we observed a significant difference between the 25(OH)D users and the non-users as concerns the number of scans with at least one new/enlarging T2 lesion (2% vs 15.38%, respectively; p= 0.025). In the multinomial regression model, 25(OH)D deficiency (serum levels ≤19,9 ng/ml) was significantly associated with a higher likelihood of disease activity during a follow-up of eighteen months (p = 0.029, RRR = 4.84, Confidence Interval (CI) 1.17 - 20.01). After six months, there were 30/136 FR patients (22,05%), whereas only 22/136 (16,17%) showed early B cell reappearance at twelve month follow up. 86.66% of the patients in the FR group showed 25(OH)D levels lower than 30 ng/ml (25(OH)D deficiency or insufficiency), whereas only 65.09% of the SR patients presented vitamin D levels lower than 30 ng/ml (p= 0.024). In the logistic regression model, 25(OH)D serum levels below 30 ng/ml were associated with a higher likelihood of early B cell reappearance at six month follow up (p= 0.042).

CONCLUSIONS

25(OH)D supplementation and serum levels might be associated with B cell kinetics and radiological activity of patients with MS treated with ocrelizumab.

Practical Aspects of novel MRI Techniques in Neuroradiology: Part 2 - Acceleration Methods and Implications for Individual Regions

BACKGROUND

 Recently introduced MRI techniques facilitate accelerated examinations or increased resolution with the same duration. Further techniques offer homogeneous image quality in regions with anatomical transitions. The question arises whether and how these techniques can be adopted for routine diagnostic imaging.

METHODS

 Narrative review with an educational focus based on current literature research and practical experiences of different professions involved (physicians, MRI technologists/radiographers, physics/biomedical engineering). Different hardware manufacturers are considered.

RESULTS AND CONCLUSIONS

 Compressed sensing and simultaneous multi-slice imaging are novel acceleration techniques with different yet complimentary applications. They do not suffer from classical signal-to-noise-ratio penalties. Combining 3 D and acceleration techniques facilitates new broader examination protocols, particularly for clinical brain imaging. In further regions of the nervous systems mainly specific applications appear to benefit from recent technological improvements.

KEY POINTS

  · New acceleration techniques allow for faster or higher resolution examinations.. · New brain imaging approaches have evolved, including more universal examination protocols.. · Other regions of the nervous system are dominated by targeted applications of recently introduced MRI techniques..

CITATION FORMAT

· Sundermann B, Billebaut B, Bauer J et al. Practical Aspects of novel MRI Techniques in Neuroradiology: Part 2 - Acceleration Methods and Implications for Individual Regions. Fortschr Röntgenstr 2022; DOI: 10.1055/a-1800-8789.

Direct imaging of white matter ultrashort T2∗ components at 7 Tesla

PURPOSE

To demonstrate direct imaging of the white matter ultrashort T₂^∗ components at 7 Tesla using inversion recovery (IR)-enhanced ultrashort echo time (UTE) MRI. To investigate its characteristics, potentials and limitations, and to establish a clinical protocol.

MATERIAL AND METHODS

The IR UTE technique suppresses long T₂^∗ signals within white matter by using adiabatic inversion in combination with dual-echo difference imaging. Artifacts arising at 7 T from long T₂^∗ scalp fat components were reduced by frequency shifting the IR pulse such that those frequencies were inverted likewise. For 8 healthy volunteers, the T₂^∗ relaxation times of white matter were then quantified. In 20 healthy volunteers, the UTE difference and fraction contrast were evaluated. Finally, in 6 patients with multiple sclerosis (MS), the performance of the technique was assessed.

RESULTS

A frequency shift of -1.2 ppm of the IR pulse (i.e. towards the fat frequency) provided a good suppression of artifacts. With this, an ultrashort compartment of (68 ± 6) % with a T₂^∗ time of (147 ± 58) μs was quantified with a chemical shift of (-3.6 ± 0.5) ppm from water. Within healthy volunteers' white matter, a stable ultrashort T₂^∗ fraction contrast was calculated. For the MS patients, a significant fraction reduction in the identified lesions as well as in the normal-appearing white matter was observed.

CONCLUSIONS

The quantification results indicate that the observed ultrashort components arise primarily from myelin tissue. Direct IR UTE imaging of the white matter ultrashort T₂^∗ components is thus feasible at 7 T with high quantitative inter-subject repeatability and good detection of signal loss in MS.

Measuring treatment response to advance precision medicine for multiple sclerosis

Objective

To assess the independent contributions of clinical measures (relapses, Expanded Disability Status Scale [EDSS] scores, and neuroperformance measures) and nonclinical measures (new brain magnetic resonance imaging [MRI] activity and serum neurofilament light chain [sNfL] levels) for distinguishing natalizumab‐treated from placebo‐treated patients.

Methods

We conducted post hoc analyses using data from the AFFIRM trial of natalizumab for multiple sclerosis. We used multivariable regression analyses with predictors (EDSS progression, no relapse, new or enlarging MRI activity, brain atrophy, sNfL levels, and neuroperformance worsening) to identify measures that independently discriminated between treatment groups.

Results

The multivariable model that best distinguished natalizumab from placebo was no new or enlarging T2 or gadolinium‐enhancing activity on MRI (odds ratio; 95% confidence interval: 7.2; 4.7–10.9), year 2 sNfL levels <97.5th percentile (4.1; 2.6–6.2), and no relapses in years 0–2 (2.1; 1.5–3.0). The next best‐fitting model was a two‐component model that included no MRI activity and sNfL levels <97.5th percentile at year 2. There was little difference between the three‐ and two‐component models.

Interpretation

Nonclinical measures (new MRI activity and sNfL levels) discriminate between treatment and placebo groups similarly to or better than clinical outcomes composites and have implications for patient monitoring.

The contemporary role of MRI in the monitoring and management of people with multiple sclerosis in the UK

BACKGROUND

Compare the contemporary use of magnetic resonance imaging (MRI) in the monitoring and management of people with MS in the UK to current consensus guidelines.

METHODS

This retrospective multicentre audit of clinical practice gathered data on 2567 patients with MS from 25 MS centres across the UK.

RESULTS

Routine monitoring (44.7%), and recent clinical relapse (20.3%) were the most common scan indications. In routine monitoring, the addition of spinal imaging to brain showed no significant difference in disease modifying treatment (DMT) decision at subsequent clinical review. Approximately 1 in 5 gadolinium administered scans showed enhancement, and in 1 in 20 patients, gadolinium enhancement was the only evidence of radiological disease activity. Mean inter-scan intervals in relapsing-remitting MS for routine monitoring was 19.2 months (SD 20.7) with wide variation between centres. Only 53.8% of patients under progressive multifocal leukoencephalopathy (PML) surveillance met the recommended scanning frequency. MRI protocols demonstrated heterogeneity in the sequences used for diagnostic, monitoring and PML surveillance scans.

CONCLUSIONS

MS centres across the UK demonstrate varied practice and protocols when using MRI to monitor people with MS. In this cohort, gadolinium use and spinal imaging demonstrates limited impact on subsequent DMT decisions.

Update on brain MRI for the diagnosis and follow-up of MS patients

Over the past decades, MRI has become a major tool in the diagnosis and the follow-up of patients with multiple sclerosis (MS), especially for monitoring the effectiveness of therapy. The recent international recommendations issued for the standardization of neurological and radiological clinical practices converge on many points. In this setting, recommendations made by the "Observatoire français de la sclérose en plaques", the French MS registry, can be distinguished by its interdisciplinary complementarity, its longevity, its size, and its positions in direct connection with the clinic. Hence, after suspicions of gadolinium deposition in the brain, with multiple warning from the American and European health authorities, a national consultation took place and resulted in limitation to useful injections. The precautionary principle prevailing, the patient receives a limited quantity of contrast product even if no clinically harmful manifestation has been detected to date. The result of this round table bringing together neurologists and neuroradiologists from specialized centers was published in the form of a recommendation in early 2020. The interest of this project also lies in the constant improvement of the management of patients with MS and the possibility of developing advanced techniques to assist the clinician. The aim of this review is to explain to the neurologist, the interest of following this imaging protocol both in his/her clinical practice and in the possibilities that this opens up.

Digital Twins for Multiple Sclerosis

An individualized innovative disease management is of great importance for people with multiple sclerosis (pwMS) to cope with the complexity of this chronic, multidimensional disease. However, an individual state of the art strategy, with precise adjustment to the patient's characteristics, is still far from being part of the everyday care of pwMS. The development of digital twins could decisively advance the necessary implementation of an individualized innovative management of MS. Through artificial intelligence-based analysis of several disease parameters - including clinical and para-clinical outcomes, multi-omics, biomarkers, patient-related data, information about the patient's life circumstances and plans, and medical procedures - a digital twin paired to the patient's characteristic can be created, enabling healthcare professionals to handle large amounts of patient data. This can contribute to a more personalized and effective care by integrating data from multiple sources in a standardized manner, implementing individualized clinical pathways, supporting physician-patient communication and facilitating a shared decision-making. With a clear display of pre-analyzed patient data on a dashboard, patient participation and individualized clinical decisions as well as the prediction of disease progression and treatment simulation could become possible. In this review, we focus on the advantages, challenges and practical aspects of digital twins in the management of MS. We discuss the use of digital twins for MS as a revolutionary tool to improve diagnosis, monitoring and therapy refining patients' well-being, saving economic costs, and enabling prevention of disease progression. Digital twins will help make precision medicine and patient-centered care a reality in everyday life.

Recent advances in development of nanomedicines for multiple sclerosis diagnosis

Multiple sclerosis (MS) is a neurodegenerative disease with a high morbidity and disease burden. It is characterized by the loss of the myelin sheath, resulting in the disruption of neuron electrical signal transmissions and sensory and motor ability deficits. The diagnosis of MS is crucial to its management, but the diagnostic sensitivity and specificity are always a challenge. To overcome this challenge, nanomedicines have recently been employed to aid the diagnosis of MS with an improved diagnostic efficacy. Advances in nanomedicine-based contrast agents in magnetic resonance imaging scanning of MS lesions, and nanomedicine-derived sensors for detecting biomarkers in the cerebrospinal fluid biopsy, or analyzing the composition of exhaled breath gas, have demonstrated the potential of using nanomedicines in the accurate diagnosis of MS. This review aims to provide an overview of recent advances in the application of nanomedicines for the diagnosis of MS and concludes with perspectives of using nanomedicines for the development of safe and effective MS diagnostic nanotools.