Precision medicine in multiple sclerosis

Precision medicine in multiple sclerosis: biomarkers for diagnosis, prognosis, and treatment response

Manuel Comabella, Jaume Sastre-Garriga, and Xavier Montalban

Current Opinion in Neurology

---

 

Purpose of review

This article highlights recent studies on MRI and body fluid molecular biomarkers with potential implications in diagnosis, prognosis, and response to treatment in patients with multiple sclerosis (MS).

 

Recent findings

In the MS diagnostic process, a reappraisal of well known MRI findings, including symptomatic and spinal cord lesions, and incorporation of newer, more specific, features of MS lesions, such as detection of central veins, are in a testing phase aiming to increase diagnostic accuracy. Lesion counts on T2-weighted scans and gadolinium uptake on postcontrast T1-weighted scans are well established prognostic brain MRI biomarkers. Evidences supporting the implementation of brain volume measures for disease and treatment monitoring are increasing, maybe within a ‘zero tolerance’ scheme in the overarching no evidence of disease activity concept. Cerebrospinal fluid (CSF) chitinase 3-like 1 and the light subunit of neurofilaments are consolidating their prognostic role in patients with clinically isolated syndrome. Blood CD62L and CSF IgM oligoclonal bands may become clinically useful biomarkers for progressive multifocal leukoencephalopathy risk stratification in MS patients treated with natalizumab.

 

Summary

Although more studies are needed, current and emerging imaging and molecular biomarkers in MS may contribute to outline the concept of precision medicine in MS.

---

 

INTRODUCTION

 

A better understanding of the individual multiple sclerosis (MS) patient disease profile, prognosis, treatment response, and adverse events is important to develop concepts of personalized medicine or precision medicine in MS, and to look at individual patients instead of patient populations. Biomarkers are of paramount importance in these concepts, insomuch as they have the capacity to capture the different aspects of MS heterogeneity and help, for instance, in MS diagnosis and disease stratification; in the prediction of disease course; in the identification of new therapies beneficial for the disease; and in the development of a personalized therapy based on the prediction of treatment response and identification of patients at high risk for side-effects [1]. The present review will focus on recent imaging and molecular biomarkers that have been either proposed for the first time in MS, or have consolidated their prognostic role in the disease.

 

MRI BIOMARKERS

 

MRI has become the most relevant paraclinical tool for the diagnosis of MS. From a very practical approach, the Magnetic Resonance Imaging in MS (MAGNIMS) network has issued a set of guidelines for the implementation of brain and spinal cord MRI in the diagnostic process [2& ]. A series of statements and recommendations include performance of spinal cord MRI in all patients with spinal cord symptoms at presentation, and in patients with inconclusive or equivocal brain MRI findings, especially in patients with so-called radiologically isolated syndromes. Brain MRI examinations can be performed, according to these recommendations, using a field strength of 1.5 T, but 3.0 T scans are deemed preferable. Pushing in favor of the use of 3 T scanners, recent evidence suggests that detection of central veins in brain lesions of patients with MS, very conspicuous at 7 T, can also be depicted using T2 -weighted sequences at 3 T, helping in the differentiation from microangiopathic lesions [3&&]. Whether this will be of use in 3 T scans performed in clinical practice settings is still to be shown in prospective studies, as early evidence seems to suggest low agreement between observers in counts of lesions with central veins [4& ]. Further improvements to diagnostic criteria are also envisaged and preliminary evidence in favor of the incorporation of symptomatic lesions to the present criteria has been already provided [5& ]. In spite of all this drive toward simplification of MS diagnosis, differentiating patients with MS from cases of neuromyelitis optica spectrum disorders (NMOSD) using MRI may be turning more complex than previously thought as some MRI features closely relate to those found in MS patients, such as the presence of short spinal cord lesions at the initial manifestations of NMOSD [6&&,7&&]. An update on the value of MRI in the diagnosis of MS and NMOSD with present challenges and future directions can be also found in a recent guidelines study by the MAGNIMS consortium [8& ].

 

The clinicoradiological paradox has been long time applied to depict the difficulties in finding close correlations between MRI findings and the natural history of MS. Again, recent guidelines issued by the MAGNIMS network [9& ] have been devoted to the topic of establishing disease prognosis (both in untreated as well as treated patients with MS). In the monitoring of disease progression, these guidelines strongly recommended the use of T2-weighted and contrast-enhanced T1-weighted brain MRI (spinal cord MRI is not generally recommended). The future implementation of brain volume monitoring is suggested, but still limited by the technical, biological, and other confounding factors potentially affecting reliability of measurements. In spite of methodological difficulties, analyses of brain volume datasets from clinical trials have further highlighted this potential, underscoring the fact that the correlation between brain volume changes and disability worsening strengthens when longer time intervals of observation are considered [10& ]. Using well established MRI parameters together with clinical, demographical, and laboratory tests to predict disease evolution after the first manifestations of the disease, two large studies have reached similar conclusions with regard to the highest relevance of MRI measures [11& ,12& ]. Tintore´ et al. [11& ], investigated the impact of baseline demographic, clinical, radiological, and biological factors on prediction of disease outcomes, including disability accumulation in more than 1000 patients with clinically isolated syndromes (CIS) with a follow-up mean of 8 years but up to 18 years; this study concluded that the number of lesions on brain MRI is the highest impact prognostic factor available. Kuhle et al. [12& ] presented a similarly large study in the same group of patients in the earliest phases of disease evolution with a shorter follow-up time but a multicenter approach; this study reached parallel conclusions on the value of brain MRI lesion number to help delineate disease prognosis, together with other biomarkers to be discussed in other sections of this study. The more established value of lesion counts or even brain volume measurements, may be accompanied by other pathological features of brain MRI scans in patients with MS, such as leptomeningeal inflammation, for which the real prevalence is still an open issue [13& ,14& ]. However, the exact value of these findings and their practical application in routine settings is yet to be fully elucidated.

 

MS therapy research has recently become a story of success, with newer drugs being incorporated to treatment algorithms almost yearly. This has intensified the search for biomarkers of disease evolution under therapy, to rapidly gauge what has been called in several ways, but is most frequently termed ‘treatment response.’ The previously alluded to MAGNIMS guidelines on treatment monitoring [9& ] have emphasized the role of brain MRI studies, indicating the need for baseline MRI examinations, possibly to be acquired once treatment effects have taken off (usually 6 months), and the appropriateness of follow-up scans performed 12 months after therapy onset to measure new T2 lesion formation and active inflammation on postcontrast T1-weighted scans. The exact threshold for lesion activity beyond which a patient should be considered as a suboptimal responder to a given drug is a matter of intense debate at the moment. The ‘zero tolerance’ threshold has been popularized under the acronym of no evidence of disease activity (NEDA), but evidence suggests that this may be unsuited to application in clinical practice on the grounds of a very low negative predictive value for disability progression over reasonably mid to long-term follow-up periods [15&&]. The incorporation of brain volume assessments to the definition of response to therapy is also a subject of intense research; several lines of evidence [16& ,17&&] are suggesting that there is an added value of brain volume loss estimations to complement the previously mentioned measures of tissue inflammation with information on the neurodegenerative process ongoing in MS. The incorporation of brain volume loss as a further parameter in the NEDA construct to give raise to NEDA-4 has already been suggested and first evidences of their usefulness are being communicated [18&&].

 

Table 1 summarizes the main findings in relation to potential new MRI markers in diagnosis, prognosis, and treatment monitoring that have been the topic of recent publications in the field and have been highlighted in the present review.

 

BODY FLUID BIOMARKERS

 

Biomarkers in the early phases of multiple sclerosis

Cerebrospinal fluid (CSF) biomarkers in patients presenting with a CIS suggestive of central nervous system demyelination are important to identify patients at high risk for conversion to MS and, more importantly, for the development of neurological disability. In this respect, it is worth highlighting two large multicenter studies recently conducted in CIS patients [12& ,19& ]. The first study determined a panel of biochemical variables related with known disease environmental factors in serum samples from 1047 CIS patients. Low vitamin D levels were associated with increased risk for MS in a univariate analysis, hence reinforcing the role of vitamin D in the early phases of MS [12& ]. In the second study, the prognostic role of the astrocyte-derived chitinase 3-like 1 (CHI3L1), which was first suggested in a proteomic study conducted in CIS patients classified according to extremes of outcome in terms of conversion to MS [20], was later validated in a large cohort study including CSF samples from 813 CIS patients [19& ]. High CSF CHI3L1 levels resulted in a risk factor for conversion to MS independent of strong predictors of conversion to MS such as brain MRI abnormalities and IgG oligoclonal bands. More importantly, CSF increased CHI3L1 levels were the only significant independent risk factor associated with the development of disability in multivariate Cox regression models [19& ]. It is important to remark that independent studies, published soon thereafter, further confirmed the prognostic role of CHI3L1 in the early phases of MS. In this regard, by using a proteomic approach with later verification of results by ELISA, CHI3L1 was identified as one of the best predictors of conversion to MS in CIS patients [21&&]. In another study conducted in patients with optic neuritis with more than 10 years of follow-up, CSF CHI3L1 in combination with MRI and age were the best predictors of MS risk in a multivariable analysis. In addition, CHI3L1 predicted long-term cognitive impairment evaluated by the Paced Auditory Serial Addition Test [22&&]. Finally, in a recent study including CIS and also relapsing-remitting MS (RRMS) patients, high CHI3L1 levels were associated with earlier progression to Expanded Disability Status Scale (EDSS) 3 and EDSS 6, thus confirming the role of CHI3L1 in the development of neurological disability [23& ].

 

Several studies also support a prognostic role for the light chain subunit of neurofilaments (NF-L) in the early phases of MS. In a longitudinal study, CSF NF-L levels were predictive of conversion to MS in CIS patients [23▪]. In patients with optic neuritis, NF-L predicted visual outcome [24] and the long-term disability evaluated by the MS severity scale and the nine hole peg test [22▪▪]. In cognitively preserved CIS patients, CSF levels of NF-L were associated with functional MRI correlates of attention performance [25]. It is important to remark that NF-L have also been measured in peripheral blood and, for instance, serum NF-L levels were increased in CIS patients compared to healthy controls [26▪]. In this regard, the finding of a high correlation between serum and CSF levels may support the use of serum as a more accessible body fluid to determine the levels of NF-L in future biomarker studies [26▪,27].

 

Finally, other recently proposed biomarkers in CIS patients were the albumin quotient as a measure of the blood-brain barrier breakdown, and the B-cell chemokine C-X-C motif chemokine ligand 13 (CXCL13). An increase in the albumin quotient at the time of the CIS event was associated with higher brain atrophy and greater disability [28]. CSF levels of CXCL13 were significantly higher in CIS patients who converted to MS, and levels correlated with CSF parameters, such as cell count, total protein, IgG index, and presence of IgG and IgM oligoclonal bands [29].

 

Biomarkers associated with disease course and activity

Regarding biomarkers that distinguish between MS and other pathological conditions or between different clinical forms of MS, it is worth mentioning the transcriptional regulator high-mobility group box protein 1, that discriminated between patients with relapse-onset MS from patients with primary progressive MS, based on the increased mRNA expression levels observed in peripheral blood cells from RRMS and secondary progressive MS patients [30▪]. MicroRNAs are also an area of expanding research, and microRNA-based signatures identified by next-generation sequencing in whole blood discriminated between patients with CIS/RRMS and patients with NMOSD [31]. Of note, the kynurenine pathway, which is involved in the breakdown of the essential amino acid tryptophan and has been found dysregulated in several psychiatric and neurodegenerative conditions, was upregulated in the CSF of primary progressive MS patients in a pattern more similar to other inflammatory neurological diseases than to patients with relapse-onset MS [32▪].

 

In several studies, proteomics-based technologies were applied to identify MS disease activity biomarkers. In this regard, a proteomic approach (isobaric tagging for relative and absolute protein quantification) was used to identify serum biomarkers associated with benign or aggressive MS progression phenotypes and led to the identification of 11 candidates related with processes, such as inflammation, opsonization, and complement activation that discriminated between both phenotypes with high sensitivity and specificity [33]. In another proteomic study in which two-dimensional electrophoresis was applied to CSF samples, two isoforms of the vitamin D-binding protein, and the apolipoprotein E allowed to discriminate between MS patients with aggressive and benign disease courses [34]. Finally, in another proteomic-based study combining isobaric tagging for relative and absolute protein quantification and mass spectrometry, the calcium-binding protein secretogranin-1 was found to be decreased in the CSF during the disease course compared with the early phases of MS [35].

 

Concerning other disease activity biomarkers, it is worth mentioning the surprising findings with B-cell-activating factor (BAFF), a key survival factor for B cells that was found significantly increased in plasma samples from stable MS patients compared with relapsing patients [36▪]. These observations prompt to reconsider the reported association between high BAFF levels and more severe disease activity, and may relate with the unexpected failure of therapeutic strategies targeting BAFF [37]. Finally, in a 15-year follow-up study high CSF levels of the heavy chain subunit of neurofilaments were associated with more severe disability progression, particularly in patients with RRMS disease course [38▪].

 

Treatment response biomarkers

 

IFNβ

At the genetic level, an intronic variant of solute carrier family 9, subfamily A (SLC9A9), a gene encoding a Na+–H+ exchanger found in endosomes, was associated with the nonresponse to IFNβ. Initial results were validated in a combined analysis including three independent cohorts, and SLC9A9 expression was downregulated in peripheral blood cells from MS patients with more activated lymphocyte profile and at increased risk for relapses [39▪▪].

 

At the transcriptomic level, one study proposed the NLR family, pyrin domain containing 3 (NLRP3) inflammasome as response biomarker to IFNβ. Interestingly, mRNA expression levels for NLRP3 and its target IL-1β, were upregulated in peripheral blood cells from nonresponders to IFNβ regardless of the response criteria used to classify patients [40▪]. Another recent transcriptomic study [41] was built on previous findings by the authors, in which combinations of gene triplets mainly composed of immune and apoptosis-related molecules were found to be highly predictive of the response to IFNβ at baseline [42]. In the more recent study, the authors found acceptable predictive accuracies for some of these previous gene triplets in an independent cohort of IFNβ-treated patients, and proposed additional predictive combinations, being the CASP2/IL-10/IL12Rb1 gene triplet the top-scoring classifier [41].

 

Glatiramer acetate

Studies of response biomarkers to glatiramer acetate remain scarce. In one recent study [43], response gene to complement 32 (RGC-32), FasL, and IL-21 were proposed as biomarkers of glatiramer acetate response based on the increased mRNA expression levels of RGC-32 and FasL and decreased expression of IL-21 observed in peripheral blood cells from responders compared to nonresponders. Among these three genes, RCG-32, a gene product induced by complement, had associated the best estimates of detecting a good responseto glatiramer acetate [43].

 

Natalizumab

Studies proposing natalizumab-related biomarkers are fast increasing and can be classified into two categories, biomarkers of response to natalizumab and biomarkers of risk for progressive multifocal leukoencephalopathy (PML). In relation to the first category, an increase in the number of circulating hematopoietic stem and progenitor cells (CD45lowCD34+) was associated with clinical remission in patients treated with natalizumab and proposed as responsebiomarker to this treatment [44▪▪]. Other candidates proposed as potential biomarkers of response to natalizumab were the lymphocytosis induced by natalizumab [45] and the oxysterols [46]. To expand further, the mean percentage of lymphocytes over a period of 10 natalizumab administrations was significantly higher in responders than in partial responders to treatment [45]. Oxysterols are oxidized derivatives of cholesterol that were found to be reduced in the CSF of MS patients by the effect of natalizumab and levels correlated with performance in the symbol digit modalities test [46].

 

Regarding biomarkers predicting adverse events to natalizumab, it should be noted the association between the presence of lipid-specific IgM oligoclonal bands and the decreased risk for PML in patients treated with natalizumab, a finding that may indicate an excessive and PML-protective immunosuppression in patients positive for IgM oligoclonal bands [47▪▪]. In another study [48▪▪], the role of the L-selectin (CD62L) expressed by CD4+ T cells as a potential biomarker for PML risk [49] was recently confirmed in a large multicenter study, and CD62L in combination with the anti-JC virus antibody index may improve the identification of natalizumab-treated patients at high risk for PML [48▪▪]. However, a recent study using cryopreserved peripheral blood mononuclear cells collected as part of natalizumab clinical trials questioned the utility of CD62L as PML risk biomarker [50].

 

Fingolimod

In a recent study, fingolimod was found to modify the CSF NF-L levels [51▪]. Of note, reductions in the NF-L levels after 1 year of treatment with fingolimod were associated with improvements in relapse and MRI outcomes [51▪], suggesting the use of NF-L as a biomarker to monitor the response to fingolimod as well as its potential neuroprotective effects.

 

Figure 1 summarizes the main findings in relation to consolidated and emerging body fluid biomarkers that have been the topic of recent publications in the field and have been highlighted in the present review.

 

CONCLUSION

Clear recommendations for routine implementation of MRI in the diagnosis and monitoring of both treated and untreated patients with MS have been recently produced. In parallel, newer, more specific features of MRI lesions have been described, namely the presence of central veins in typical MS lesions, and the evidence in favor of the incorporation of brain volume measurements as a disease-monitoring tool is fast growing. Composite clinical and MRI measures such as NEDA and NEDA-4 are still to prove their usefulness in routine clinical practice. Regarding body fluid biomarkers, CHI3L1 and NF-L have consolidated their prognostic role in CIS patients not only in terms of predicting conversion to MS but also in relation with the development of disability, and it is probably time we consider their incorporation into routine clinical practice. Other biomarkerssuch as SLC9A9 or the NLRP3 inflammasome need further confirmation in independent studies. Finally, biomarkers in natalizumab-treated patients like the lipid-specific IgM oligoclonal bands in the CSF and CD62L in peripheral blood may help to improve current algorithms evaluating the risk of PML. Although more studies are needed, a great deal of progress has already been made in the field of biomarkers in MS, and recent findings suggest that precision medicine, a utopia in the past, is slowly but steadily becoming reality in MS.

 

---

 

ACKNOWLEDGEMENTS

The authors thank the ‘Red Española de Esclerosis Múltiple (REEM)’ sponsored by the ‘Fondo de Investigación Sanitaria’ (FIS), Ministry of Science and Innovation, Spain, and the ‘Ajuts per donar Suport als Grups de Recerca de Catalunya’, sponsored by the ‘Agència de Gestió d’Ajuts Universitaris i de Recerca’ (AGAUR), Generalitat de Catalunya, Spain.

 

Financial support and sponsorship

None.

 

Conflicts of interest

M.C. has received compensation for consulting services and speaking honoraria from Bayer Schering Pharma, Merk Serono, Biogen-Idec, Teva Pharmaceuticals, Sanofi-Aventis, Genzyme, and Novartis. J.S.-G. has received compensation for consulting services and speaking honoraria from Merck-Serono, Biogen-Idec, Teva, Genzyme, Almirall, and Novartis. X.M. has received speaking honoraria and travel expenses for scientific meetings, has been a steering committee member of clinical trials or participated in advisory boards of clinical trials in the past years with Bayer Schering Pharma, Biogen Idec, EMD Merck Serono, Genentech, Genzyme, Novartis, Sanofi-Aventis, Teva Phramaceuticals, Almirall, and Roche.

 

---

 

REFERENCES AND RECOMMENDED READING

Papers of particular interest, published within the annual period of review, have been highlighted as:

▪ of special interest

▪▪ of outstanding interest

---

 

REFERENCES

1. Comabella M, Montalban X. Body fluid biomarkers in multiple sclerosis. Lancet Neurol 2014; 13:113–126. 2. & Rovira A` , Wattjes MP, Tintore´ M, et al., MAGNIMS study group. Evidencebased guidelines: MAGNIMS consensus guidelines on the use of MRI in multiple sclerosis-clinical implementation in the diagnostic process. Nat Rev Neurol 2015; 11:471–48

2. A practically oriented set of guidelines designed to be readily applied in the diagnostic process in the day-to-day clinical practice of physicians managing patients with suspected MS.

3. && Mistry N, Abdel-Fahim R, Samaraweera A, et al. Imaging central veins in brain lesions with 3-T T2-weighted magnetic resonance imaging differentiates multiple sclerosis from microangiopathic brain lesions. Mult Scler 2015; pii: 1352458515616700. [Epub ahead of print] PubMed PMID: 26658816. Increasing the pathological specificity toward inflammatory demyelination of MRI findings would be a major step forward in the field of MS diagnosis; the presence of central veins in hyperintense lesions in optimized T2 -weighted scans performed in clinically available 3 T magnets is bound to be very helpful here.

4. & George IC, Sati P, Absinta M, et al. Clinical 3-tesla FLAIR MRI improves diagnostic accuracy in multiple sclerosis. Mult Scler 2016; pii: 1352458515624975. [Epub ahead of print] PubMed PMID: 26769065. A study focusing on the clinical implementation of the use of the ‘central vein sign’ suggesting that further steps include validation in multicenter studies together with creation of consensus criteria of easy application in clinical practice settings.

5. & Caucheteux N, Maarouf A, Genevray M, et al. Criteria improving multiple sclerosis diagnosis at the first MRI. J Neurol 2015; 262:979–987. A new proposal of diagnostic criteria that includes the visualization of the symptomatic lesion as a primary criterion. Further testing and validation in prospective cohorts are needed.

6. && Kim HJ, Paul F, Lana-Peixoto MA, et al. MRI characteristics of neuromyelitis optica spectrum disorder: an international update. Neurology 2015; 84:1165–1173. Definitively disregarding the older doctrine that brain MRI scans should be normal in patients with NMOSD, this consensus study comparatively assesses MRI findings in MS and NMOSD patients.

7. && Flanagan EP, Weinshenker BG, Krecke KN, et al. Short myelitis lesions in aquaporin-4-IgG-positive neuromyelitis optica spectrum disorders. JAMA Neurol 2015; 72:81–87. Blurring further the borders between NMOSD and MS, the present study demonstrates that nonextensive (i.e., short) transverse myelitis are not uncommon in NMOSD, but may still have some distinguishing features such as a central location and being hypointense on T1-weighted scans.

8. & Filippi M, Rocca MA, Ciccarelli O, et al., MAGNIMS Study Group. MRI criteria for the diagnosis of multiple sclerosis: MAGNIMS consensus guidelines. Lancet Neurol 2016; 15:292–303. A thorough review of recent evidence with proposed modifications to the current diagnostic criteria for MS.

9. & Wattjes MP, Rovira A` , Miller D, et al., MAGNIMS study group. Evidencebased guidelines: MAGNIMS consensus guidelines on the use of MRI in multiple sclerosis: establishing disease prognosis and monitoring patients. Nat Rev Neurol 2015; 11:597–606. A practically oriented set of guidelines designed to be readily applied in the monitoring of treated and untreated patients with MS in the clinical practice setting.

10. & Radue EW, Barkhof F, Kappos L, et al. Correlation between brain volume loss and clinical and MRI outcomes in multiple sclerosis. Neurology 2015; 84:784–793. Very large datasets coming from high-quality clinical trials will be instrumental to assess the clinical value of MRI parameters such as brain volume measures. In this study data from 3635 patients, total 8616 patient-years serve well to demonstrate that brain volume loss is associated with other measures of disease severity, adding predictive power to models explaining disability.

11. & TintoreM, Rovira A` , Rı´o J, et al.Defining high, medium and low impact prognostic factors for developing multiple sclerosis. Brain 2015; 138:1863–1874. The study presents a multivariate analysis of predictive factors of disease evolution after a first episode suggestive of MS. Such approach allows weighing the relative impact of such factors in one of the largest and longest followed inception cohorts in MS research.

12. & Kuhle J, Disanto G, Dobson R, et al. Conversion from clinically isolated syndrome to multiple sclerosis: a large multicentre study. Mult Scler 2015; 21:1013–1024. A very large multicenter study investigating prognostic factors after a first attack suggestive of MS over a relatively short follow-up time again suggesting the value of T2-weighted hyperintense lesions independently of other body fluid biomarkers such as the presence of oligoclonal bands in the CSF. Environmental factors known to play roles in MS, such as tobacco, measured by cotinine levels, humoral responses against Epstein–Barr virus, and vitamin D were also investigated, although only a role for vitamin D was partially confirmed, with low serum vitamin D increasing the risk for clinically definite MS.

13. & Absinta M, Vuolo L, Rao A, et al. Gadolinium-based MRI characterization of leptomeningeal inflammation in multiple sclerosis. Neurology 2015; 85:18– 28. In this study, focal leptomeningeal enhancement was observed in 25% of patients with MS but in only one out of 37 healthy controls (2.7%), being twice as frequent in primary progressive MS patients.

14. & Eisele P, Griebe M, Szabo K, et al. Investigation of leptomeningeal enhancement in MS: a postcontrast FLAIR MRI study. Neurology 2015; 84:770–775. Contrary to the previous study, focal leptomeningeal enhancement was found only in one out of 112 patients with MS. Differences may lie in technical issues, such as sequence specifications or dose and delay of gadolinium administration. In any case, diffuse leptomeningeal enhancement should still raise a red flag for the diagnosis of MS.

15. && Rotstein DL, Healy BC, Malik MT, et al. Evaluation of no evidence of disease activity in a 7-year longitudinal multiple sclerosis cohort. JAMA Neurol 2015; 72:152–158. No evidence of disease activity (NEDA – absence of relapses, disability worsening, and new T2 lesions) is found to be difficult to maintain over time (only 7.9% of treated patients after 7 years) and even though positive predictive values of NEDA at 2 years for stability are high (78.3%), the negative predictive values are characteristically low, around 40%.

16. & Pe´rez-Miralles FC, Sastre-Garriga J, Vidal-Jordana A, et al. Predictive value of early brain atrophy on response in patients treated with interferon b. Neurol Neuroimmunol Neuroinflamm 2015; 2:e132. Brain volume changes after 1 year of therapy with IFNb are independently predictive of EDSS status after 4 years. Best cutoff to predict worse outcomes on therapy is calculated at 0.86% of brain volume loss.

17. && Sormani MP, De Stefano N, Francis G, et al. Fingolimod effect on brain volume loss independently contributes to its effect on disability. Mult Scler 2015; 21:916–924. Brain volume loss over 2 years and presence of relapses during the first year of therapy with fingolimod are able to explain 73% of the effect of this drug on disability worsening after 2 years in a post hoc analysis of the Fingolimod Research Evaluating Effects of Daily Oral Therapy in Multiple Sclerosis trial involving 992 patients.

18. && Kappos L, De Stefano N, Freedman MS, et al. Inclusion of brain volume loss in a revised measure of ’no evidence of disease activity’ (NEDA-4) in relapsingremitting multiple sclerosis. Mult Scler 2015; pii: 1352458515616701. [Epub ahead of print] PubMed PMID: 26585439. The study, on a pooled dataset of both phase III placebo-controlled trials of fingolimod, suggests that NEDA-4, resulting from the incorporation of brain volume measurements to NEDA-3, may result in an improved prediction of disability worsening in treated patients.

19. & Canto´ E, Tintore´ M, Villar LM, et al. Chitinase 3-like 1: prognostic biomarker in clinically isolated syndromes. Brain 2015; 138:918–931. The prognostic role of CHI3L1, first suggested in a previous proteomic study conducted by the same group [18&&], was confirmed in the largest validation study of a CSF biomarker conducted in CIS patients. Interestingly, a CHI3L1 value of 170 ng/ml resulted in the best cutoff point to classify CSF protein levels into high and low. This cutoff had clear prognostic implications since CIS patients with CHI3L1 levels above the cutoff value had shorter time to MS and earlier disability progression, with a 5-year difference as median time to reach and EDSS 3.0 compared with patients with CSF CHI3L1 values below the 170 ng/ml cutoff point. CHI3L1 immunostaining performed in human brain tissue identified reactive astrocytes as the main central nervous system cell population involved in CHI3L1 expression.

20. Comabella M, Ferna´ndez M, Martin R, et al. Cerebrospinal fluid chitinase 3-like 1 levels are associated with conversion to multiple sclerosis. Brain 2010; 133:1082–1093.

21. && Hinsinger G, Gale´otti N, Nabholz N, et al. Chitinase 3-like proteins as diagnostic and prognostic biomarkers of multiple sclerosis. Mult Scler 2015; 21:1251–1261. In a CSF proteome comparison, CHI3L1 and CHI3L2 were found to be more abundant in RRMS patients than in controls. Of note, the authors identified a cutoff value for CSF CHI3L1 of 189 ng/ml, which was close to the cutoff of 170 ng/ml identified in the study conducted by Canto´ et al. [19& ], and also had prognostic implications insomuch as CIS patients with high CSF CHI3L1 converted more rapidly to MS. It is worth mentioning that serum CHI3L1 levels also had discriminative power to identify MS converters. Finally, CSF levels of CHI3L1 and CHI3L2 accurately distinguished controls from RRMS and progressive MS patients.

22. && Modvig S, Degn M, Roed H, et al. Cerebrospinal fluid levels of chitinase 3-like 1 and neurofilament light chain predict multiple sclerosis development and disability after optic neuritis. Mult Scler 2015; 21:1761–1770. In this study, the authors investigated the role of eight selected CSF molecular biomarkers (CHI3L1, osteopontin, neurofilament light chain, myelin basic protein, C-C motif chemokine ligand 2, CXCL10, CXCL13, and matrix metallopeptidase 9) in 86 CIS patients presenting with optic neuritis in terms of conversion to clinically definite MS and long-term disability. Of the abovementioned CSF biomarkers, in multivariable Cox regression models, only CHI3L1 predicted conversion to clinically definite MS. In addition, CHI3L1 correlated with the Paced Auditory Serial Addition Test. When disability was evaluated, none of the molecular biomarkers predicted the MS severity scale score in multivariable analyses, although the light subunit of neurofilaments was a significant predictor of MS severity scale score in the univariable analyses.

23. & Martı´nez MA, Olsson B, Bau L, et al. Glial and neuronal markers in cerebrospinal fluid predict progression in multiple sclerosis. Mult Scler 2015; 21:550–561. A large panel of neuronal and glial markers, including the NF-L and CHI3L1 were investigated in the CSF of 109 CIS patients and 192 RRMS patients. NF-L and CHI3L1 levels in the CSF were increased in fast converters to clinically definite MS (<1 year). Cox hazard regression model revealed that high CSF CHI3L1 levels were an independent risk factor to reach EDSS 3 and 6.

24. Modvig S, Degn M, Sander B. Cerebrospinal fluid neurofilament light chain levels predict visual outcome after optic neuritis. Mult Scler 2015; pii: 1352458515599074. [Epub ahead of print] PubMed PMID: 26283696.

25. Tortorella C, Direnzo V, Taurisano P, et al. Cerebrospinal fluid neurofilament tracks fMRI correlates of attention at the first attack of multiple sclerosis. Mult Scler 2015; 21:396–401.

26. & Disanto G, Adiutori R, Dobson R, et al. Serum neurofilament light chain levels are increased in patients with a clinically isolated syndrome. J Neurol Neurosurg Psychiatry 2016; 87:126–129. Levels of the NF-L were measured in serum samples of 100 CIS and 92 healthy controls using a previously validated highly sensitive electrochemiluminescencebased immunoassay. Serum NF-L levels were increased in CIS patients; however, they could not discriminate between fast converters and CIS patients who remained as a CIS after long follow-up. Although more studies of NF-L in serum are needed, these findings open the possibility of replacing CSF by serum as more accessible body fluid to measure NF-L levels.

27. Kuhle J, Barro C, Disanto G, et al. Serum neurofilament light chain in early relapsing remitting MS is increased and correlates with CSF levels and with MRI measures of disease severity. Mult Scler 2016; pii: 1352458515623365. [Epub ahead of print] PubMed PMID: 26754800.

28. Uher T, Horakova D, Tyblova M, et al. Increased albumin quotient (QAlb) in patients after first clinical event suggestive of multiple sclerosis is associated with development of brain atrophy and greater disability 48 months later. Mult Scler 2015; pii: 1352458515601903. [Epub ahead of print] PubMed PMID: 26362893.

29. Ferraro D, Galli V, Vitetta F, et al. Cerebrospinal fluid CXCL13 in clinically isolated syndrome patients: association with oligoclonal IgM bands and prediction of multiple sclerosis diagnosis. J Neuroimmunol 2015; 283:64– 69.

30. & Malhotra S, Fissolo N, Tintore´ M, et al. Role of high mobility group box protein 1 (HMGB1) in peripheral blood from patients with multiple sclerosis. J Neuroinflammation 2015; 12:48. The high-mobility group box protein 1 has been shown to mediate functions of pathogenic relevance for disorders like MS, such as proinflammatory cytokine production, T-cell proliferation, and cell migration. mRNA expression levels of highmobility group box protein 1 were increased in peripheral blood mononuclear cells of RRMS and secondary progressive MS patients, suggesting their potential use as biomarker to distinguish patients with relapse-onset MS from patients with primary progressive MS.

31. Keller A, Leidinger P, Meese E, et al. Next-generation sequencing identifies altered whole blood microRNAs in neuromyelitis optica spectrum disorder which may permit discrimination from multiple sclerosis. J Neuroinflammation 2015; 12:196.

32. & Aeinehband S, Brenner P, Sta˚hl S, et al. Cerebrospinal fluid kynurenines in multiple sclerosis; relation to disease course and neurocognitive symptoms. Brain Behav Immun 2016; 51:47–55. The kynurenine pathway is involved in the breakdown of tryptophan and includes the kynurenic acid, which is mainly produced by astrocytes and blocks the N-methyl-D-aspartate glutamatergic receptors, and the quinolinic acid, which is mainly produced by microglia and acts as an agonist of N-methyl-D-aspartate receptors. A balance in favor of the kynurenic acid may result in neuroprotection by counteracting the excitotoxic effect of the quinolinic acid. Primary progressive MS patients were characterized by an overall upregulation of kynurenine pathway metabolites, which differed from the pattern observed in RRMS and secondary progressive MS patients and suggest differences in the underlying disease mechanisms of these clinical forms.

33. Tremlett H, Dai DL, Hollander Z, et al. Serum proteomics in multiple sclerosis disease progression. J Proteomics 2015; 118:2–11.

34. Perga S, Giuliano Albo A, Lis K, et al. Vitamin D binding protein isoforms and apolipoprotein e in cerebrospinal fluid as prognostic biomarkers of multiple sclerosis. PLoS One 2015; 10:e0129291.

35. Kroksveen AC, Jaffe JD, Aasebø E, et al. Quantitative proteomics suggests decrease in the secretogranin-1 cerebrospinal fluid levels during the disease course of multiple sclerosis. Proteomics 2015; 15:3361–3369.

36. & Kannel K, Alnek K, Vahter L, et al. Changes in blood B cell-activating factor (BAFF) levels in multiple sclerosis: a sign of treatment outcome. PLoS One 2015; 10:e0143393. In this prospective longitudinal study, the levels of the B cell-activating factor (BAFF) were determined in plasma samples from 170 RRMS patients and 87 controls. Overall, plasma BAFF levels were higher in MS patients compared with controls. However, within the RRMS group, BAFF levels were significantly increased in stable MS patients without relapses compared with relapsing patients. These observations challenge the hypothesis that increased BAFF levels in peripheral blood are always related with more severe forms of the disease. On a different note, BAFF levels were increased by the effect of IFNb, a finding reported in previous publications proposing BAFF as a response biomarker to this treatment.

37. Kappos L, Hartung HP, Freedman MS, et al. Atacicept in multiple sclerosis (ATAMS): a randomised, placebo-controlled, double-blind, phase 2 trial. Lancet Neurol 2014; 13:353–363.

38. & Petzold A. The prognostic value of CSF neurofilaments in multiple sclerosis at 15-year follow-up. J Neurol Neurosurg Psychiatry 2015; 86:1388–1390. The study supports the long-term (15 years) prognostic value of the heavy subunit of neurofilaments in MS patients with different clinical forms, and high CSF heavy subunit of neurofilaments levels were associated with more severe disability progression, particularly in RRMS patients.

39. && Esposito F, Sorosina M, Ottoboni L, et al. A pharmacogenetic study implicates SLC9A9 in multiple sclerosis disease activity. Ann Neurol 2015; 78:115– 127. In this genome-wide association study conducted in MS patients treated with IFNb, G allele carriers of rs9828519, a polymorphism located in the SLC9A9 gene, were associated with a lack of response to IFNb. A meta-analysis across three independent replication datasets confirmed the initial findings. The study corresponds to the first genome-wide association study of IFNb response in MS that identifies a polymorphism exceeding the threshold of genome-wide signifi- cance in the discovery cohort.

40. & Malhotra S, Rı´o J, Urcelay E, et al. NLRP3 inflammasome is associated with the response to IFN-b in patients with multiple sclerosis. Brain 2015; 138:644–652. There is increasing evidence of a role for inflammasomes in experimental autoimmune encephalomyelitis and MS. In this study, the authors aimed to investigate the role of inflammasomes (NLRP3, NLRP1, NLRC4, and absent in melanoma 2) and related cytokines (IL-1b, IL-10, and IL-18) in the response to IFNb in MS patients. mRNA expression levels for NLRP3 and IL-1b were significantly increased in peripheral blood cells from IFNb nonresponders compared with responders. These findings were irrespective of the response criteria applied to classify patients (clinical criteria after 2 years of treatment or clinical– radiological criteria after 1 year of treatment).

41. Baranzini SE, Madireddy LR, Cromer A, et al. Prognostic biomarkers of IFNb therapy in multiple sclerosis patients. Mult Scler 2015; 21:894–904. 42. Baranzini SE, Mousavi P, Rio J, et al. Transcription-based prediction of response to IFNbeta using supervised computational methods. PLoS Biol 2005; 3:e2.

43. Kruszewski AM, Rao G, Tatomir A, et al. RGC-32 as a potential biomarker of relapse and response to treatment with glatiramer acetate in multiple sclerosis. Exp Mol Pathol 2015; 99:498–505.

44. && Mattoscio M, Nicholas R, Sormani MP, et al. Hematopoietic mobilization: potential biomarker of response to natalizumab in multiple sclerosis. Neurology 2015; 84:1473–1482. Absolute counts of hematopoietic stem and progenitor cells (HSPC) were higher in blood from patients treated with natalizumab compared to untreated patients. Longitudinal determinations of HSPC over 1 year of natalizumab treatment revealed a significant but highly variable increase in the number of CD34þ cells. When this variability was analyzed further, HSPC nonmobilizer patients showed persistent MRI and clinical disease activity compared to mobilizers (evaluated by the number of gadolinium-enhancing lesions and presence of relapses). Based on these findings, the authors proposed the hematopoietic mobilization as response biomarker to natalizumab.

45. Signoriello E, Lanzillo R, Brescia Morra V, et al. Lymphocytosis as a response biomarker of natalizumab therapeutic efficacy in multiple sclerosis. Mult Scler 2015; pii: 1352458515604381. [Epub ahead of print] PubMed PMID: 26453682.

46. Novakova L, Axelsson M, Malmestro¨m C, et al. Reduced cerebrospinal fluid concentrations of oxysterols in response to natalizumab treatment of relapsing remitting multiple sclerosis. J Neurol Sci 2015; 358:201–206.

47. && Villar LM, Costa-Frossard L, Masterman T, et al. Lipid-specific immunoglobulin M bands in cerebrospinal fluid are associated with a reduced risk of developing progressive multifocal leukoencephalopathy during treatment with natalizumab. Ann Neurol 2015; 77:447–457. MS patients with positive lipid-specific IgM oligoclonal bands are characterized by a highly inflammatory disease. This IgM oligoclonal bands-associated inflammatory profile may play a protective role against the development of PML in patients treated with natalizumab. This hypothesis was elegantly supported by the results from this study in which the authors observed that patients with lipid-specific IgM oligoclonal bands in the CSF had lower risk of suffering PML compared to patients without IgM oligoclonal bands.

48. && Schwab N,Schneider-Hohendorf T,PignoletB,et al.PML riskstratification using anti-JCV antibody index and L-selectin. Mult Scler 2015. [Epub ahead of print] The role of CD62L expressed by CD4þ T cells as risk biomarker for PML was confirmed in a large multicenter study and, as previously reported [49], the percentage of CD4þ T cells expressing CD62L was significantly lower in those natalizumab-treated patients who later developed PML.

49. Schwab N, Schneider-Hohendorf T, Posevitz V, et al. L-selectin is a possible biomarker for individual PML risk in natalizumab-treated MS patients. Neurology 2013; 81:865–871.

50. Lieberman LA, Zeng W, Singh C, et al. CD62L is not a reliable biomarker for predicting PML risk in natalizumab-treated R-MS patients. Neurology 2016; 86:375–381.

51. & Kuhle J, Disanto G, Lorscheider J, et al. Fingolimod and CSF neurofilament light chain levels in relapsing-remitting multiple sclerosis. Neurology 2015; 84:1639–1643. The NF-L was measured in the CSF of a subgroup of patients who participated in the Fingolimod Research Evaluating Effects of Daily Oral Therapy in Multiple Sclerosis study at baseline and after 1 year of treatment. Although CSF NF-L levels were overall reduced by the effect of fingolimod, it is important to mention that at the 1-year timepoint, NF-L levels were higher in patients who experienced relapses during the study, and NF-L levels correlated with the new/enlarging T2 lesions count.

Full Article