Clinical manifestations

Prognosis and complications

The authors of these 3 studies joined their efforts and founded the “Radiologically Isolated Syndrome Consortium” (RISC) in 2009 to study this state of subclinical disease, understand its evolution in time, and clarify treatment strategies. The RISC has started many research projects and continues to conduct multicenter multinational studies related to radiologically isolated syndrome (52). The initial studies on radiologically isolated syndrome focused on individual factors predicting conversion to clinical disease such as spinal cord lesions, pregnancy, and MRI and CSF findings (50; 69; 51). In a major multinational study lead by the consortium, a cohort of 451 individuals with radiologically isolated syndrome were evaluated for demographic features and risk factors for conversion to clinical disease at 5 years and then in 277 of the initial 451 at 10 years (70; 53). The mean age at diagnosis of radiologically isolated syndrome was 37.2 (range:11-74) years with a mean clinical follow-up time of 4.4 (range: 0.01–21.1) years and 6.7 (range:1-26.6) years in the 2 studies, respectively. The cumulative probability of a first clinical event was 34.7% at 5 years, and 51.2% at 10 years. In the multivariate model (younger) age of less than 37, positive CSF for oligoclonal bands, and the spinal cord and posterior fossa lesions on MRI were identified as baseline factors associated with a subsequent first clinical event by 10 years. The probability of a first clinical event at 10 years follow-up was 29% for individuals with at least 1 risk factor, 54% for 2 risk factors, 68% with 3 risk factors, and 87% for all 4 risk factors. Although the presence of contrast enhancement at baseline was not associated with an increased risk for a clinical conversion, there was the observation of continuing radiological activity during follow-up. A positive family history for multiple sclerosis, observed in 10.0% of the study group, was not predictive for a clinical event (53). These findings may assist the physician in treatment decisions with radiologically isolated syndrome.

Some individuals with radiologically isolated syndrome later present with progressive disease suggestive of primary progressive onset rather than a relapse (40). Eleven point seven percent of these patients fulfilled criteria for primary progressive multiple sclerosis. This rate of primary progressive multiple sclerosis at onset within the radiologically isolated syndrome population is similar to that seen in the general multiple sclerosis populations in an age-dependent manner, further suggesting that radiologically isolated syndrome is biologically part of the multiple sclerosis spectrum. This finding necessitates redefining primary progressive disease and a review of the concept of progression! Because primary progressive multiple sclerosis may develop from radiologically isolated syndrome, it suggests that these individuals could go through the relapsing form of the disease subclinically. Then, once they enter in the so-called secondary progressive stage, they start showing the first clinical signs of the disease, hence diagnosed as primary progressive multiple sclerosis. It may be hypothesized that they accumulate central nervous system damage and have asymptomatic radiological relapses that remain clinically silent possibly due to efficient repair plasticity. Once a certain threshold is reached, the disease manifests itself clinically and appears to be primary progressive (92; 39). This pattern is consistent with a number of studies showing that all progressive forms of multiple sclerosis whether primary, secondary, or single-attack progressive are similar in terms of the age at onset of progression (15; 43; 94; 39). It may be hypothesized that the progressive forms of multiple sclerosis have similar biology, differentiated by the plasticity of the central nervous system and severity of inflammatory degeneration. This view has considerable treatment implications.

Radiologically isolated syndrome in the pediatric population. In a multinational/multicenter study, 38 children (£ 18 years) fulfilling the radiologically isolated syndrome criteria were followed for a mean of 4.8±5.3 years. A first clinical event consistent with central nervous system demyelination occurred in 42% of the study population at a median of 2.0 years, a higher rate in a shorter time compared with the adult population, whereas radiologic evolution developed in 61% (60). Oligoclonal bands in the CSF and spinal cord lesions on MRI were associated with an increased risk of a first clinical event. In an expanded cohort of 61 children with radiologically isolated syndrome, the same group confirmed their earlier results and showed that oligoclonal bands increased the specificity of magnetic resonance imaging criteria for predicting clinical conversion (59). Incidental lesions suggestive of multiple sclerosis found on MRI in the absence of clinical symptoms have a significant impact in children regarding counselling and management issues.

Biological basis

Etiology and pathogenesis

Pathology of radiologically isolated syndrome and what we have learned from the imaging studies. There is only a single report based on the biopsies of 3 patients with radiologically isolated syndrome who had MRI for non-multiple sclerosis causes. Brain MRI showed a tumefactive enhancing lesion with additional nonenhancing white matter lesion/s in each. Pathology confirmed inflammatory demyelinating disease compatible with multiple sclerosis pathology in all (41).

Access to central nervous system tissue in individuals diagnosed with radiologically isolated syndrome is not expected. However, it is still possible to obtain some information through more advanced MRI technology and software programs.

In order to assess brain damage in asymptomatic first-degree relatives of patients with sporadic and familial multiple sclerosis patients in the study of De Stefano and colleagues above, study subjects and healthy volunteers underwent brain MRI and magnetization transfer imaging on a mobile MR scan (19). In asymptomatic first-degree relatives of patients with established multiple sclerosis, magnetization transfer ratios of focal brain lesions were lower than in control white matter whereas magnetization transfer ratios values from the normal-appearing white matter and brain volumes from asymptomatic individuals were similar to healthy controls. Thus, despite morphological (imaging) evidence of subclinical disease, the central nervous system damage remained very limited and likely is under the influence of yet unknown protective mechanisms and therefore, remains subclinical in these individuals.

In subjects with radiologically isolated syndrome who are studied with magnetic transfer imaging and volumetric measurements, there are some lesions with similarities to lesions in relapsing-remitting multiple sclerosis (22). However, it was much milder than in relapsing-remitting multiple sclerosis patients. This suggests that the lack of clinical manifestations in radiologically isolated syndrome subjects may be due to more limited and less-severe tissue involvement. In another imaging study of 15 individuals with radiologically isolated syndrome, the same group found cortical lesions in 40% of the subjects, another finding commonly seen in patients with clinical multiple sclerosis (32). These subjects had higher white matter lesion volumes than those without cortical lesions but no more cortical atrophy or cognitive decline. These subjects also had oligoclonal bands in their CSF and most had cervical lesions on spinal MRI, suggesting that they are more likely to convert to clinical multiple sclerosis than individuals with radiologically isolated syndrome who are free from such affection. In another study from Spain, regional cortical thinning was primarily distributed in frontal and temporal lobes (47).

Decreased brain NAA/Cr levels in subjects with radiologically isolated syndrome suggest that axonal damage can be significant even at this early stage (86). However, not all individuals with radiologically isolated syndrome show these abnormalities.

Thalamic atrophy, a metric associated with neurodegeneration in early stages of multiple sclerosis, was also shown in radiologically isolated syndrome (05). The normalized cerebellar white matter volume and the anterior cerebellar gray matter volume were significantly decreased in individuals with radiologically isolated syndrome compared to healthy controls, suggesting that regional brain atrophy develops prior to an initial clinical episode in these people (29).

In radiologically isolated syndrome with visible spinal lesions, there was no evidence of spinal cord atrophy but the magnetization transfer ratio was lower. This suggests that inflammation and demyelination may reflect microstructural changes detectable in the very earliest stage of multiple sclerosis, a finding consistent with the known pathologic mechanisms of multiple sclerosis (02).

The central vein sign in the diagnosis and differential diagnosis of multiple sclerosis has become a reliable marker in the differential diagnosis between multiple sclerosis and multiple sclerosis-like lesions (76; 64; 57). Most individuals with radiologically isolated syndrome have a central vein sign in their lesions, (78). In this study, the proportion of central vein sign lesions correlated with the presence of spinal cord lesions. Spinal cord lesions are a major risk factor for clinical conversion in individuals with radiologically isolated syndrome, so the central vein sign in this population is also likely to have prognostic value.

Susceptibility-based imaging in multiple sclerosis demonstrates a paramagnetic rim sign indicative of chronic active demyelination and associated with the presence of iron inside phagocytes (01). This sign was also detected in individuals with radiologically isolated syndrome. The lesions with paramagnetic rims correlated with white matter lesions showing the central vein sign (77). These findings collectively suggest that many of the lesions observed in radiologically isolated syndrome develop from perivenous inflammation and demyelination and that most lead to chronic, active inflammation.

The above-mentioned MRI studies in radiologically isolated syndrome have a number of major implications. The central vein sign and the paramagnetic rim sign are additional findings to consider in the differential diagnosis of this syndrome, especially when the Okuda criteria are not fulfilled. Many of these imaging studies reveal that central nervous system changes already exist in this preclinical stage. Microstructural and metabolic changes in the central nervous system tissue, years before the onset of the clinical disease, as well as the demonstration that cortical, thalamic, and cerebellar atrophy is similar to that of clinically isolated syndrome and relapsing-remitting multiple sclerosis shows that radiologically isolated syndrome is the earliest stage of clinical multiple sclerosis. However, changes are not uniform in all these individuals and not all of them convert to clinical disease. Some protective mechanisms may be effective in some of these individuals but not in others!

Nonimaging biomarkers in radiologically isolated syndrome.

Cerebrospinal fluid studies. Asymptomatic twins or first-degree relatives of multiple sclerosis patients can have oligoclonal bands and other CSF findings supportive of multiple sclerosis (Xu and McFarlin 1984; 24). In Sweden, 19% of the healthy siblings of multiple sclerosis patients have oligoclonal bands, in contrast to 4% in unrelated healthy controls, as well as increased antiviral reactivity, named as a multiple sclerosis immunopathic trait (35). In the same population asymptomatic siblings had CSF neurofilament light chain protein and glial fibrillary acidic protein levels similar to controls (36). However, their siblings with clinical multiple sclerosis had higher levels of these proteins, which are sensitive markers of axonal damage and astrogliosis. The lack of an increase in neurofilaments, despite the presence of oligoclonal bands and increased antivirus immune-reactivity suggests that despite biological (immunopathic) evidence of subclinical disease, these individuals are free from central nervous system damage and remain asymptomatic. Because there was no MRI data in this study, we can’t argue whether these asymptomatic siblings also had asymptomatic brain and spinal lesions.

The presence of CSF oligoclonal bands is a predictive factor for clinical conversion both in adult and pediatric subjects with radiologically isolated syndrome (59; 53). Oligoclonal bands and CSF neurofilament light chain protein levels are also independent predictors for clinical conversion in patients with radiologically isolated syndrome and are associated with a shorter time to onset of clinical multiple sclerosis (61). On the other hand, chitinase-3-like protein, a biomarker that predicts conversion from clinically isolated syndrome to relapsing-remitting multiple sclerosis, did not influence conversion to clinical disease in radiologically isolated syndrome (61; 89).

Progranulin is a fundamental neurotrophic factor involved in inflammatory wound repair. It is elevated in the CSF of individuals with radiologically isolated syndrome and is suggested to be an early marker of inflammatory repair before axonal disintegration occurs (72).

Multiple sclerosis cases were identified among active duty military personnel with at least 1 presymptomatic serum sample stored in the US Department of Defense Serum Repository. Serum neurofilament light protein levels were higher in these presymptomatic people compared to matched controls, both at a median of 6 years and at 1 year before the clinical onset of multiple sclerosis (08). This study confirms a prodromal phase for multiple sclerosis lasting several years and that neuroaxonal damage may already occur during this phase. This study was not conducted in a cohort of people with radiologically isolated syndrome but this asymptomatic period is comparable to the study above in which neurofilament light protein levels in the CSF predicted clinical conversion in patients with radiologically isolated syndrome (61). Serum neurofilament light protein was confirmed to be evidence of disease activity defined as clinical conversion and/or new MRI activity in individuals within radiologically isolated syndrome (89).

Preclinical phase – is there a multiple sclerosis prodrome?

The radiologically isolated syndrome is imaging evidence that there is an asymptomatic period of multiple sclerosis. Some work done in this preclinical phase raises concerns whether this prodromal phase is truly asymptomatic.

Cognitive function was investigated in individuals with radiologically isolated syndrome in 2 separate studies. Depending on the type of the neuropsychological testing and the number of tests failed, about one fourth or more of the radiologically isolated syndrome cohort showed mild cognitive impairment (50; 04).When quantitative MRI metrics were included, comparable levels of lesion load and brain atrophy were detected in patients with well-established relapsing-remitting multiple sclerosis (04). Based on these studies, mild cognitive impairment is present in some individuals with radiologically isolated syndrome and they have a similar profile to the cognitive impairment seen in clinically isolated syndrome and relapsing-remitting multiple sclerosis (50; 04; 46). It is noteworthy that neither the individuals with radiologically isolated syndrome who failed on tests nor their family members noticed any cognitive deficits or changes (04). In an earlier study, 4 individuals with (at that time defined as) “subclinical multiple sclerosis” had mild cognitive impairment on neurocognitive testing done at different times after the initial imaging study (37). This finding was independent of clinical conversion to multiple sclerosis and didn’t progress over the years!

Because at least some individuals with radiologically isolated syndrome have a similar cognitive profile to multiple sclerosis patients, this raises the question of whether these individuals are likely to be multiple sclerosis patients with an undiagnosed “clinically isolated syndrome” who present with cognitive dysfunction (50). However, neither these individuals with radiologically isolated syndrome nor their family members were aware of any sign of cognitive deficits or functional deficits. There may be some signs that remain subclinical unless explored with special techniques and tools, similar to the demonstration of subclinical disease by imaging. Such “sophisticated findings” may not predict conversion to clinical disease and may correlate with a “subclinical disease state” that may persist for long periods or even a lifetime. In an Israeli cohort of radiologically isolated syndrome in which cognitive performance was below average for age- and education-matched norms, cognitive performance did not differ between individuals with radiologically isolated syndrome who converted to multiple sclerosis and those who remained asymptomatic (63). These findings suggest that cognitive deficits may develop independently of focal and inflammatory disease activity (46).

When finger motor performance was quantitated with engineered gloves, worse finger motor performance was found in individuals with radiologically isolated syndrome compared to healthy controls (09). Subclinical motor impairment was related to diffuse microstructural white matter damage on DTI. The impaired hand motor performance could be detected only by sophisticated technology in these individuals with radiologically isolated syndrome who didn’t have any functional/clinical impairment in their daily life, suggesting that some motor signs may remain subclinical.

Therefore, such “sophisticated findings” may not always predict conversion to clinical disease and may correlate with a “subclinical disease state” that may remain as such for long periods or even for a lifetime (81).

Is preclinical cognitive performance affected long before the clinical onset of multiple sclerosis? This question was addressed in a number of studies. In the Norwegian multiple sclerosis registry, men born between 1950 and 1995 and who underwent the conscription examination when they were 18 to 19 years of age, some later developed multiple sclerosis. Their cognitive performance was correlated with a cohort of randomly selected controls from the Norwegian Conscript Service database (16). Men developing their first clinical multiple sclerosis symptoms up to 2 years after the examination scored significantly lower than controls, corresponding to a 6-intelligence quotient point difference, and those who developed primary progressive multiple sclerosis scored 4.6 to 6.9 points lower than controls up to 20 years prior to first progressive symptoms. These findings suggest that relapsing/remitting multiple sclerosis may start years prior to the clinical presentation and that primary progressive multiple sclerosis could begin even decades prior to the first apparent clinically detectable progressive symptoms (16). An Argentinean study looking at school performance as a marker of cognitive decline prior to diagnosis of multiple sclerosis showed similar results (80). On the other hand, in a Swedish study in which the time of multiple sclerosis was not considered, a significant difference wasn’t observed (34). From these studies it may be said that cognitive problems may be present in all multiple sclerosis phenotypes before apparent clinical symptoms develop.

Multiple sclerosis patients reported higher rates of fatigue or depression in the years prior to the initial neurologic episode of their diagnosis compared with non-multiple sclerosis controls (12; 07). More frequent use of health care such as more hospital admissions, physician claims, and prescriptions in patients with multiple sclerosis than in controls in the 5 years before a first demyelinating event, according to health administrative data, suggested the existence of a measurable multiple sclerosis prodrome (97). The same Canadian group then studied the major causes for more physician and hospital encounters and found that these were related to nervous, sensory, musculoskeletal, and genito-urinary systems, and in another study reported that fatigue, sleep disorders, anemia, and pain were more common within this period (98; 99). Consistent with these symptoms, cases had more psychiatrist and urologist encounters and higher proportions of musculoskeletal, genito-urinary, or hormonal prescriptions. In a similar study conducted in the UK, multiple sclerosis patients had a higher risk of presenting up to 10 years prior to index date with gastric, intestinal, urinary, and anorectal disturbances, anxiety, depression, insomnia, fatigue, headache, and various types of pain (23). A systematic review assessing studies of morbidities before or at disease onset confirmed that anxiety, depression, migraine, and lower cognitive performance are more common in multiple sclerosis patients than in control populations and appear during a multiple sclerosis prodrome (100).

These observations indicate that some people with multiple sclerosis are likely to have mild cognitive changes and non-multiple sclerosis specific health-related problems in the years before they develop clinical multiple sclerosis. What we may discover as radiologically isolated syndrome may not be isolated! However, the concept of “multiple sclerosis prodrome” (93) doesn’t seem to have practical consequences unless combined with further and more solid evidence.

Differential diagnosis

The diagnosis of radiologically isolated syndrome depends on what we may name “1 positive criterion,” which is a pattern consistent with multiple sclerosis on an MRI study which was performed for non-multiple sclerosis causes, and “1 negative criterion” which is the absence of any clinical symptom or sign suggestive of multiple sclerosis, ever.

The MRI criteria for a diagnosis of radiologically isolated syndrome are outlined in Table 1. Some physicians, rather than considering these detailed criteria, would prefer McDonald’s dissemination in space MRI criteria (Thompson et al 2017). It should be kept in mind that when the periventricular and juxtacortical regions are considered, those lesions should touch the ventricles or the cortex, respectively, and that subcortical lesions should not be mistaken for either of them. In many non-multiple sclerosis cases, periventricular and juxtacortical lesions may coexist and may be confused with demyelinating lesions. Neurologically normal patients with small vessel disease, migraine, and many of the so-called “nonspecific white matter abnormalities” can have white matter abnormalities that are easily confused with multiple sclerosis lesions (44; 85; 82).

The practicing neurologist should be extremely careful before making a diagnosis of radiologically isolated syndrome, which also has important psychological consequences on the individual. When in doubt, the lack of the central vein within these lesions may add in differentiating these ischemic or nonspecific lesions from multiple sclerosis-related white matter lesions (85; 76; 64; 82).

Once the MRI is deemed to be consistent with “radiologically isolated syndrome,” the other major diagnostic criterion is to be sure that the individual has had no symptom related to multiple sclerosis, ever. The next task of the neurologist should be to requestion that he or she is truly asymptomatic for multiple sclerosis and is indeed a “definite” case of radiologically isolated syndrome. There are a number of non-multiple sclerosis and nonspecific neuropsychiatric symptoms that may mimic a demyelinating event. Somatoform symptoms, dizziness, visual blurring related to the eye, and numbness due to nerve entrapments are symptoms that require clinical expertise in differential diagnosis and disease mechanisms.

Diagnostic workup

Once a diagnosis of radiologically isolated syndrome is made by a brain MRI combined with a negative clinical history, there are a number of approaches that may be taken. One is to decide not to go with further workup and to just follow the individual by an MRI after 3 to 6 months. Further workup should rule out any other disorder that may mimic these MRI lesions such as systemic disorders that cause central nervous system lesions (82).

Further workup should also estimate the risk of clinical conversion and prognosis. These include evaluating risk factors seen on images of the spinal cord, CSF oligoclonal bands, IgG, and antiviral profiles, visual evoked potentials, and optical coherence tomography to estimate the extent of subclinical disease (48; 28; 42). Neurofilament light chain protein may help predict clinical conversion and can be studied serum with highly-sensitive single-molecule arrays technology.

Management

Currently, there is no evidence to support early treatment in individuals with radiologically isolated syndrome and neither the Food Drug Administration nor the European Medicines Agency have approved any disease-modifying treatments for radiologically isolated syndrome. However, in common practice, mainly in the U.S. and partly in Europe, a significant number of physicians start off-label treatment to these individuals. Two separate surveys, 1 in the U.S. and the other in Europe had confirmed this attitude (91; 26)! The protreatment decisions were mostly influenced by observing 1 or more spinal cord lesions on MRI, gadolinium-enhancing lesions, and more than 1 new or enlarging T2 brain lesion at a follow-up scan. In Argentina, neurologists are more conservative and less likely to start treatment in radiologically isolated syndrome (13). However, there was a general consensus in all surveys to perform a lumbar puncture to evaluate the presence of oligoclonal bands and to perform a spinal cord MRI.

Initiating treatment in people with radiologically isolated syndrome, mainly at presentation, unfortunately results in unnecessary treatment in a significant number of people who don’t have any clinical disease and who probably would not develop multiple sclerosis in their lifetime (10; 21; 33; 45; 20; 58). Besides, it’s not so uncommon to see individuals misdiagnosed as having radiologically isolated syndrome and in whom disease-modifying treatments are initiated based on their MRI, which in fact don’t fulfil the Okuda criteria. These nonspecific white matter lesions are a well-known facet in individuals who are misdiagnosed as having clinical multiple sclerosis and put on therapy (84; 82; 38). However, there are some who advocate treatment in selected individuals with radiologically isolated syndrome and other risk factors (Brassat and Lebrun 2012; 67).

It may be wise to do a spinal tap in high-risk individuals with radiologically isolated syndrome who have posterior fossa and spinal cord lesions along with enhancing lesions on their MRI to determine whether they also have CSF oligoclonal bands and elevated serum neurofilament light chain protein levels. This approach may help the caring neurologist to decide how to follow and whether to start treatment in these asymptomatic individuals. Our approach is to follow these individuals by MRI every 6 months in the first year and then yearly for the next 2 years, and then at year 5 or until they develop any clinical symptom suggestive of multiple sclerosis (81). A consensus statement from the magnetic resonance imaging in multiple sclerosis study group (MAGNIMS) similarly suggests that clinical and MRI monitoring should be considered at 6- to 12-month intervals in this asymptomatic population. There have been several other algorithms suggested for the long-term evaluation of people with radiologically isolated syndrome regarding when to consider treatment (79; 03).

There are 2 ongoing clinical trials to determine whether approved disease-modifying therapies for multiple sclerosis will prevent or delay the onset of a first episode consistent with clinical multiple sclerosis in individuals with radiologically isolated syndrome. The first is a 2-year multicenter, randomized, double-blind clinical trial with the oral medication dimethyl fumarate in the United States, the ARISE study (71). The second trial has the same design with oral teriflunomide in Europe, the TERIS study (54). Once completed these clinical trials are expected to provide answers as to whether early treatment should be initiated in individuals with radiologically isolated syndromes and if yes, in whom and at what stage.

So, what should be our treatment decision for people with radiologically isolated syndrome? It would be safer to wait for the results of the ongoing 2 treatment trials, ARISE and TERIS. However, there may be some individuals in whom the neurologist may feel a need to start treatment, such as young individuals with high-risk MRI factors such as spinal cord and posterior fossa lesions and CSF-oligoclonal bands. Each of these 4 risk factors increase the probability to convert to clinical disease at 10 years, from 29% when only 1 exists to 87% when all are present together (53). High-risk individuals can be followed more closely. With 3 to 4 of the outlined risk factors and continued new and enhancing lesions on consecutive MRI studies, treatment may be considered before clinical symptoms and signs develop.

Needless to say, because these people don’t present with clinical symptoms, “an attack”, they shouldn’t be treated with steroids even if they do have gadolinium-enhancing lesions.

Special considerations

Pregnancy

In a single study addressing this issue, the mean number of active lesions on a subsequent brain MRI scan was significantly higher in the pregnant group with radiologically isolated syndrome compared to the nonpregnant control group with radiologically isolated syndrome (51). However, it was not clear whether the increase seen on MRI had increased risk for clinical conversion.

Anesthesia

No data exist regarding anesthesia in individuals with radiologically isolated syndrome. It’s likely that general anesthesia and epidural anesthesia may not have an effect, as in people with multiple sclerosis.

Surgery. No data exist regarding surgical procedures in individuals with radiologically isolated syndrome.

Acknowledgments. We wish to thank Darin Okuda, Christine Lebrun, Daniel Pelletier, and Orhun Kantarci, the core group of our RISC, and also Naila Makhani and Cristina Azavedo; over the years our close collaboration and work in both adult-onset and pediatric-onset radiologically isolated syndrome had allowed us to understand and learn further of this syndrome.