Advances in Neuroimaging for Amyotrophic Lateral Sclerosis

Michael O'Leary

May 22, 2015

Neurology Advisor

Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disorder characterized by progressive muscular paralysis reflecting degeneration of motor neurons in the primary motor cortex, brain stem, and spinal cord. 

Diagnosing ALS in the early stage of the disease can be difficult because of variable presentation, and common mimic disorders, according to Erik Pioro, MD, PhD, director of the section of amyotrophic lateral sclerosis and related disorders at Cleveland Clinic.

Published site ...

“While ways to identify lower motor neuron dysfunction in ALS (like electromyography) are easily accessible, we have limited means to assess upper motor neuron abnormalities in patients, and quickly rule out ALS mimics,” Pioro said.

With no established objective markers of upper motor neuron and extramotor involvement in ALS, diagnostic workup is very complex and currently based on "clinical presentation, progression of symptoms, and exclusion of other diseases, supported by neurophysiological and neuroimaging examinations."

Much of the complexity stems from the phenotypic expression of ALS, which is highly heterogeneous and determined by four elements: body region of onset, upper and lower motor neuron involvement, rate of progression, and cognitive impairment.4

A definitive diagnosis depends on both lower and upper neuron involvement, which is often not clinically evident.4 About 10% of cases are classified as familial, and now more than half of these have an identified mutation. However, most patients have sporadic ALS with no obvious family history.3

A comprehensive diagnostic workup includes the following procedures:4

• Thorough neurological examination
• Blood and urine studies
• Electrodiagnostic tests
• X-rays
• Spinal tap
• Muscle and/or nerve biopsy
• Myelogram of cervical spine

All of these factors contribute to an average delay of one year from initial symptom onset to diagnosis, which hampers biomarker research as trial populations inevitably enroll patients with “atypical,” slowly progressive disease, leaving out those with more aggressive disease.5 More importantly, more than 40% of ALS patients undergo incorrect medical treatment, including surgical intervention, emphasizing the need for reliable diagnostic and prognostic biomarkers for ALS.6

The Role of Neuroimaging

Although ALS is a complex condition, neuroimaging could enable objective phenotypic classification of patients, provide information on the brain regions affected, and identify which brain alterations correspond to various symptom progression rates. Neuroimaging could also identify disease stage and monitor progressive degenerative changes.7

Currently, conventional MRI is used to rule out upper cervical cord lesions and other conditions that may mimic ALS.9 Efforts to develop accurate diagnostic biomarkers for ALS have concentrated on multimodal applications of advanced neuroimaging methods.

Michael Knopp, MD, PhD, vice chair of research at Ohio State University's Wexner Medical Center said that the complexity of ALS means it is unlikely a single biomarker will give the reader confidence to make the right diagnosis. Instead, he sees a consensus emerging on a complex series of imaging studies.

“Part of what is currently driving the community is putting consensus guidelines or best practices together that can then serve as an integrated biomarker by tapping into morphometric imaging, functional imaging using diffusion tensor, and also the neural functioning imaging, and then combining the biochemical readout coming from spectroscopy,” Knopp said. “I think what the community has recognized is that this has to be done in a structured way, and that if we use it in a structured way, the different components allow us to achieve a more precise diagnosis and more precise assessment of what is happening.”

In a review of the role of neuroimaging in ALS, researchers outlined three main areas of neuroimaging research:8

• Anatomical and functional changes in ALS have been identified on structural (MRI), functional MRI (fMRI), PET, and SPECT neuroimaging. This includes the spread of cortical and subcortical lesions.
• MRI and radiotracers can identify central nervous system alterations that could improve the accuracy of diagnosing ALS with sensitivity and specificity that could be used in a clinical setting.
• These techniques are being used to assess promising biomarkers of how motor and non-motor lesions progress, which will be used as both clinical markers of prognosis and as biological markers in research evaluating how well experimental treatments are working.

In a 2014 proof-of-concept study, Bradley Foerster, MD, of the University of Michigan, and colleagues tested diffusion tensor imaging (DTI) and magnetic resonance spectroscopy (MRS) to accurately identify diagnostic biomarkers for ALS. Subjects included 29 ALS patients and 30 matched healthy controls. Both groups underwent MRI with MRS to measure g-aminobutyric acid (GABA) and DTI to measure fractional anisotropy (FA) of the corticospinal tract.

Researchers found that DTI FA measures combined with MRS resulted in a significant improvement in positive and negative likelihood ratios, and they concluded that the combination leads to significantly increased diagnostic accuracy in distinguishing ALS patients with well-established disease from healthy controls.5

In a review of advanced neuroimaging techniques, researchers examined potential diagnostic value of multimodal neuroimaging techniques of examining brain structure. One technique combined high resolution MRI to measure focal loss of gray or white matter, with voxel-based and surface-based morphometry (VBM and SBM). Using automated software, the acquired data was compared to normal controls to study regional differences.6

The researchers found that reports vary, with some showing reduced gray matter in motor and extramotor regions, while others only show reduced gray matter in extramotor regions. The differences are thought to be due to sample sizes, image processing, and statistical analysis, but also because of the clinical, cognitive, and genetic characteristics of patients.

In another study, Adriano Chiò, of the ALS Centre of the Department of Neuroscience of the University of Turin and Molinette Hospital, wrote "findings from large studies have shown that frontotemporal and parietal loss or thinning, which is seen in patients with ALS with normal cognitive functions, is more severe in patients with cognitive impairment and ALS-FTD than in those with only motor ALS.”1

At the same time, SBM studies of patients with ALS show that cortical thinning is a consistent alteration specific to upper motor neuron degeneration. Mimic conditions and patients with PMA do not show these cortical changes.1

While functional MRI using tasks to investigate cognitive and behavioral performance have shown increased activity in cortical regions associated with motor processing, Foerster and Verstraete concluded that that this type of measure is unlikely to result in diagnostic markers because of challenges involving quantification and standardization.

“It is expected that techniques with much higher temporal resolutions than fMRI, such as magnetoencephalography, will provide future insights into functional connectivity changes in the context of ALS,” they wrote.6

Studying neuronal receptors and protein expression in ALS is also uniquely possible with positron emission tomography (PET). Foerster and Verstraete also reviewed PET studies aimed at examining brain receptors and metabolites, including the γ- aminobutyric acid A (GABAA) receptor and the serotonergic 5-hydroxytryptamine1A receptor.

The studies reviewed showed patients with sporadic ALS had reductions in [11C]flumazenil receptor binding in the motor cortex and motor association areas. There also may be different neuronal vulnerabilities for ALS patients who have a SOD1 mutation compared with sporadic ALS, leading to differences in their cortical imaging signature.2 While promising, Foerster and Verstraete concluded that new PET agents will provide more potential biomarkers for ALS.5

In a review of PET imaging in ALS, researchers found that the largest study to date involved 195 ALS patients and 40 controls. FDG-PET using generalized linear discriminant model resulted in 95% sensitivity and 83% specificity in distinguishing ALS patients from healthy controls.9

Although advances in ALS neuroimaging have been substantial, further progress is needed in developing biomarkers for the presymptomatic period, which is currently only feasible for familial ALS.

“Ongoing research into patients with preclinical familial ALS to identify biomarkers of disease onset is very important,” Pioro said, “but may become even more useful if such indicators of pathology can reveal disease at the earliest stages of symptom onset in nonfamilial patients.”

In an opinion piece, Martin Turner of Oxford University's John Radcliffe Hospital and Michael Benatar, of Miller School of Medicine, University of Miami, said such presymptomatic biomarkers are essential for studying both disease onset and symptom onset as well as overcoming biases inherent in studies that focus exclusively on people already affected.2

The article listed several research needs, among them the need to standardize techniques and data sharing. Such standards for sample collection, storage, and analysis are required not only for those affected, but also for relevant healthy controls in order to be able replicate findings that allow the validation essential to advancing the understanding and treatment of ALS. The authors recommended that a biomarker development arm should be a mandatory component in all future therapeutic trials. Multivariate modeling is also required because of the complex biology of ALS.2

“Hopefully we will [soon] have a palette of biomarkers, [including] blood, CSF, electrodiagnostic, and neuroimaging-based, to achieve early and accurate identification of ALS in patients who will be able to derive the greatest benefits from present and emerging therapies,” said Pioro.

Michael O'Leary is a freelance medical writer based in the greater Seattle Area. This article was medically reviewed by Pat F. Bass III, MD, MS, MPH.

References

1. Chiò A, Pagani M, Agosta F, Calvo A, Cistaro A, Filippi M. Neuroimaging in amyotrophic lateral sclerosis: insights into structural and functional changes. Lancet Neurol. 2014;13(12):1228-40.
2. Turner MR, Benatar M. Ensuring continued progress in biomarkers for amyotrophic lateral sclerosis. Muscle Nerve. 2015;51(1):14-8.
3. ALS Association. “Familial Amyotrophic Lateral Sclerosis (FALS) and Genetic Testing”http://www.alsa.org/about-als/genetic-testing-for-als.html
4. ALS Association. “Diagnosing ALS” http://www.alsa.org/about-als/diagnosing-als.html
5. Foerster BR, Carlos RC, Dwamena BA, et al. Multimodal MRI as a diagnostic biomarker for amyotrophic lateral sclerosis. Ann Clin Transl Neurol. 2014;1(2):107-14.
6. Verstraete E, Foerster BR. Neuroimaging as a new diagnostic modality in amyotrophic lateral sclerosis. Neurotherapeutics. 2015;12(2):403-16.
7. Rajagopalan V, Pioro EP. Comparing brain structural MRI and metabolic FDG-PET changes in patients with ALS-FTD: 'the chicken or the egg?' question. J Neurol Neurosurg Psychiatr. 2014;
8. Pradat PF, El mendili MM. Neuroimaging to investigate multisystem involvement and provide biomarkers in amyotrophic lateral sclerosis. Biomed Res Int. 2014;2014:467560.
9. Quartuccio N, Van weehaeghe D, Cistaro A, Jonsson C, Van laere K, Pagani M. Positron emission tomography neuroimaging in amyotrophic lateral sclerosis: what is new?. Q J Nucl Med Mol Imaging. 2014;58(4):344-54.