Round 33: CNS Lupus

R. John Looney, MD
Professor of Medicine, University of Rochester

Dr. Looney is an advisor and consultant for Genentech, Roche, Biogen-Idec and Proteolix, but has no financial disclosures related to the material in this article.

Release Date: July 9, 2010
Expiration Date: January 1, 2011

For CME credit,TAKE POST-TEST & EVALUATION

Objectives

After completing this activity, you’ll be able to:

  1. State the Manifestations and pathology of CNS lupus
  2. Explain Vascular cognitive impairment
  3. Discuss Vascular pathogenesis

Central Nervous System Lupus

Central nervous system lupus (CNS lupus) is a very diverse disease:

  • Acute confusional state
  • Psychosis
  • Transverse myelitis
  • Aseptic meningitis
  • Seizures
  • Depression (major)
  • Cognitive dysfunction (moderate or severe)
  • Stroke
  • Demyelinating syndromes
  • Movement disorders

It’s very hard to put this together as one disease.

Occurrence of neuropsychiatric lupus in SLICC Study

In the SLICC inception cohort for neuropsychiatric lupus (NP-SLE) there were nine neurologic manifestations patients experience about 21 months earlier in their disease:

Occurrence of Neuropsychiatric Lupus in SLICC Study

The most common manifestation is seizures and peripheral nerve disease, but cerebrovascular disease and acute confusional states are also prevalent.  The ACR criteria for neuropsychiatric lupus include anxiety, mood disorders and mild cognitive rubs, which leads to 80% of people having neuropsychiatric lupus. The SLICC study excluded these things (mood disorders or anxiety, headaches, mild or moderate depression and mild or moderate cognitive defects) to get down to a more reasonable number of patients––7.5 percent seem to have CNS manifestations early in the disease.

CNS Involvement in SLE at Autopsy

What do autopsy series show about CNS lupus? Ellis seminars in Arthritis and Rheumatism from 1979 show CNS lupus as primarily a vascular problem in the central nervous system. A vasculopathy is seen with changes in the vessels that are generally non-inflammatory, endothelial proliferation, etc. In this particular series, some patients had vasculitis.  However, other series don’t have any vasculitis and so vasculitis is not seen in most of the autopsy series.

Generally, people who die with active CNS lupus show some vasculitis––infarcts and hemorrhages are also present. The primary pathology is not just neuronal dropout; it’s more vascular changes.

Mechanisms of CNS-SLE

The blood vessels and the brain parenchyma are mechanisms of CNS lupus that are important to consider when identifying therapies.

Blood Vessels: The Clinical Problem

Blood Vessels

Above is a patient referred to Rheumatology by Neurology. They originally thought she had MS and they followed her for several years. White matter lesions are apparent on the flair images and they have a characteristic kind of pattern. She had some mild problems with memory and fatigue.  She had a positive ANA, that’s why they sent her to Rheumatology. It turns out she had very low C4 and she has anti-Ro antibodies. Outside of her CNS, not a lot of lupus manifestations: Her history included smoking and elevated blood pressure that she was treated for. She did not have antiphospholipid antibodies.

Leukoaraiosis

How should a patient like this be treated? She is not a lupus patient. This is a patient who has small vessel disease of the CNS. On MRI that looks like what is called leukoaraiosis, which means rarefied white matter.

Leukoaraiosis gives a pattern of confluent areas of white matter disease and also some lacunae, infarcts.  The kinds of changes seen in MRIs with lupus are also seen in MRIs of normal individuals with aging and diabetes and hypertension. Leukoaraiosis is really thought to be a potential cause for non-infarct dementia: white matter lesions and some deterioration of the tissue around the lesions but without real infarcts.

Clinical Correlates of Leukoaraiosis

This has been followed in the general population, people who have leukoaraiosis on their MRIs.

Clinical correlates (cross-sectional)

  • Cognitive defects
  • Decreased ADLs
  • Depression
  • Impaired balance & gait
  • Urinary incontinence

Prognostic significance (longitudinal)

  • Dementia
  • Disability
  • Small vessel strokes
  • Vascular mortality
  • Bleeding with anticoagulation
  • Bleeding with thrombolysis
  • Complications with carotid surgery

Selective Reduction of Blood Flow to White Matter during Hypercapnia

Blood flow imaging of white matter while people were breathing CO2 shows dilatation of the cerebral vessels making it easy to identify the vulnerable vasculature in the brain.

The blue areas (under A) show a decrease in blood flow during Hypercapnia.  B shows a sort of map of leukoaraiosis in a group of people.  A and B match up reasonably well indicating that the areas of leukoaraiosis are where the vulnerable vessels in the brain are appearing.

White Matter Disease and Vascular or Mixed Dementia

If white matter disease is present on MRI there is a chance of developing dementia over time. People who have a more severe disease have a very high risk of developing dementia over time.

People who have leukoaraiosis get vascular cognitive impairment, which is distinct from the neuronal cognitive impairment. White matter and subcortical gray matter lesions are associated with impairments in attention, executive function, and processing speed. A lot of neuropsychiatric testing will miss some of these defects; tests need to be chosen carefully to find this defect.

Below is a patient with a T2 and a flair image. The flair images are much better at breaking out the white matter lesions.

t2

SPECT scans: a radioactively tagged chemical that accumulates in areas based on blood flow and some of the characteristics of the tissue.

Below is a heat map looking at Z score. The red areas are the areas that are the coldest in terms of blood flow––decreased blood flow.

Heat Map Looking at Z Score

Cognitive Dysfunction in SLE

The cognitive dysfunction of NPSLE consists of deficits in information processing speed, attention, and executive function, combined with preservation of language, a profile more reminiscent of the frontal-subcortical dementias (Huntington’s disease and Parkinson’s disease) and of white matter dementias such as MS, than of Alzheimer’s disease.

Causes of Vascular Disease in SLE

  1. Conventional risk factors
  2. Vascular inflammation
    No immune complex deposit – activation of endothelial cells and leukocytes leading to vasculopathy and leukothrombosis
    Immune complexes deposit – vasculitis and capillaritis
  1. APL antibodies (and TTP)
  2. Type I interferon and endothelial cell precursors

Systemic lupus erythematosus:  brain (gross specimen and photomicrograph)

Figure Below: Left, Gross specimen reveals multiple cerebral infarctions. Right, Photomicrograph demonstrates occlusion of small vessels by both leukoaggregation (A) and a fibrin thrombus (B) as the causes of the infarctions. Vasculitis was not found.

gross specimen and photomicrograph

Intramural platelet deposition in cerebral vasculopathy of SLE is reminiscent of what is seen with TTP or chronic DIC––deposition of platelets and fibrin in the wall of the vessels.

Activated CD8+ T lymphocytes correlate with disease activity in patients with SLE may contribute to some of the vasculopathy and vascular disease in CNS lupus.

Neuropsychiatric antiphospholipid antibodies (APLS): thrombotic and ischemic effects

  • Stroke
  • Transient ischemic attack
  • Cerebral venous sinus thrombosis
  • Ocular ischemia
  • Acute ischemic encephalopathy
  • Seizure
  • Cognitive impairment
  • Dementia
  • Optic atrophy
  • Transverse myelitis
  • MS-like disease
  • Chorea
  • Migraine
  • Psychiatric disturbance

Non-thromboembolic Manifestations of APLS

  • Livedo reticularis*
    • Activation of endothelial cells
    • Vasculopathy
  • Heart valve lesions
  • Chorea
  • Thrombocytopenia

None of these things respond to anticoagulation.

Interferon-promotes abnormal vasculogenesis in lupus

A recent “hot topic” has been the observation of alpha interferon in patients with lupus.  Alpha interferon can cause inflammation in tissues––photosensitive skin rash. However, there may be some affects of alpha interferon on the vasculature as well. Looking at endothelial cell precursor cells per ml of blood in normal individuals versus lupus individuals, there is a decrease of about 50% overall. Looking at these endothelial cells precursors in blood as a function of SLEDAI, there is a correlation. People with the more active SLEDAI have very few of the endothelial cell precursors in blood.

If endothelial cell precursors are put into tissue culture, a nice lawn of cells grows. Insert alpha interferon––they don’t grow. In addition to affect on endothelial cell precursors there is association with vascular endothelial dysfunction in alpha interferon.

Vascular Disease (In Summary)

There are multiple plausible pathologic processes that might be important for vascular disease in lupus beyond the conventional risk factors:

  • Leukoagglutination;
  • Thromboembolic disease;
  • Vasculopathy; and
  • Vasculitis (in some people).

These are some of the potential immune targets to go after when trying to treat this:

  • Type III – immune complexes
    • Complement and Fc receptor
    • Alpha interferon
  • Anti-phospholipid antibodies
  • Plasma cells
  • General immune activation

There are also vascular targets:

  • Risk factors for vascular disease
  • Interventions for clotting
  • Interventions for vasculopathy

Brain Parenchyma

Mechanisms of CNS-SLE

The other major mechanism with CNS-SLE is an effect on the brain parenchyma directly, primarily through autoantibodies.  Other mechanisms could include the effect of cytokines or choroid dysfunction. Immune complexes are seen in the choroid plexus where CSF is synthesized but it is unknown if there is an abnormality in the synthesis of CSF.

Antibodies Associated with CNS-SLE

What are some of the autoantibodies associated with CNS lupus?

  • Anti-phospholipid antibodies
  • Anti-neuronal antibodies
    • Anti-NSPA / anti-ribosomal P
    • Anti-NMDA resceptor / anti-dsDNA
  • Other antibodies
    • Anti-aquaphorin 4 (Neuromyelitis optica)
    • Antibodies associated with paraneoplastic neurological disease, e.g. anti-Hu
    • ANCA

Antibodies reactive to the surface antigens of the Neuroblastoma cells

In a fairly recent paper (Kang, Lupus 2008) took a neuroblastoma cell line, threw in serum from different types of individuals and then did flow cytometry to look for IgG staining.  They found it is very easy to find anti-neuronal antibodies that are binding to the neuroblastoma as sort of a surrogate for that in patients with CNS lupus.  Non-CNS lupus and all these other autoimmune disease or healthy controls were markedly lower. The same thing was true in the CSF––there were a lot of the anti-neuronal antibodies in the CSF.

Anti-DNA antibodies cross-react with the NMDA NR2 receptor

Betty Diamond had a long series of research where she was characterizing anti-DNA antibody in mice with lupus. She made a hybridoma, then characterized the peptide and then found that that antibody actually cross-reacted with a peptide in the NMDA NR2 receptor in the brain. She showed that CSF from patients with lupus actually combine this peptide whereas normals don’t , and that CSF and lupus patients also have anti-DNA. What she went on to show that the antibody could induce the CSF from lupus patients to induce apoptosis of neuronal cells in tissue culture; whereas the CSF from normal individuals does not.

In a mouse model, Diamond could immunize with this peptide or immunize with a control. If immunized with the peptide just by itself, nothing happened. They developed antibodies that would cross-react with neurons but they didn’t have any disease unless the blood brain barrier was disrupted.  When the blood brain barrier was disrupted, there was apoptosis in the mice that have been immunized with the peptide; the mice immunized with the control immunigen didn’t have this.

Affinity-purified anti-ribosomal P recognize a cell-surface protein

The other antibody that has been long known to be associated with CNS problems is anti-ribosomal P.  Anti-ribosomal P recognizes a protein in a cytoplasm of a lot of cells, but will also recognize protein on the surface of neurons.

This antibody also has physiological effects.  If the antibody is introduced, there’s a dose dependent increase in death of cells just like with the anti-DNA antibodies.  However, there are also calcium fluxes causing effects on cell function and viability.

CSF Antibody to Neuronal Cells

In 1981 Bluestein (Am J Med) did a study using neuroblastoma cells with a radioimmunoassay to show that lupus patients with CNS disease had an increase in antibodies that bind the neuroblastoma cells (higher than the people with no neurological disease).

He also looked at specific binding––how much binding is a function of how much IgG is in the sample.  When compared to serum, the CSF in these individuals is much higher. It looks like these antibodies are either being synthesized in, or somehow by, the central nervous system.

This raises the possibility that in some lupus patients the blood brain barrier doesn’t have to be broken. The bad antibodies could be there all the time causing a slow deterioration rather than episodes that causes the problem.

Targets

  • Immune targets (antibodies, and how o get rid of them)
    • Anti-neuronal antibodies
    • Anti-aquaporin antibodies
  • Blood brain barrier
    • MMP-9 inhibition
  • Neuroprotection
    • Block receptors bound by antibodies

Problems with pathogenesis

There are too many clinical manifestations to line up nicely with pathogenic mechanisms.  There are too many plausible pathogenic mechanisms and too many potential targets. This makes it very hard to do clinical trials that separate all these different possibilities.

Magnetic Resonance Imaging (MRI)

MRI may be a better outcome measure to use in trials of CNS-SLE than clinical outcomes.

  1. Neuropsychiatric testing is clinically difficult to get done because of cost.
  2. Some of the damage that occurs or some of the problems that occur may have a threshold effect. In other words, neuropsychiatric testing may not show anything until a threshold is crossed. This is certainly true for leukoaraiosis in the general population––MRI shows it long before neuropsychiatric testing.
  3. Early intervention is key.

Conventional MRIs in patients admitted with lupus show white matter lesions as the most common and infarcts not far behind.

Cerebral MRI abnormalities and neuropsychiatric manifestations

Cerebral MRI

The slides above represent a population-based study from Finland that looked at conventional MRIs and CSF volume, which is a measure of atrophy and T1 or T2 volume in patients who are controlled.

There’s an enormous difference in the conventional MRI findings in lupus versus normal and in people with NPSLE.  Interestingly, the people who don’t have neuropsychiatric lupus still have abnormalities.

Atrophy of Hippocampus in SLE3

Atrophy in Hippocampus in SLE

A series of patients with lupus were studied twice a mean of 19 months apart. They were all suspected of having neuropsychiatric lupus. They are people who had two different scans, which showed atrophy of the hippocampus over this 19-month period of time, which is very statistically significant.

Progression of Small Vessel Disease (Lacunae and Leukoaraiosis)

In people with small vessel disease, two parameters can explain 74% of the variance in executive function: pre-morbid IQ and their fractional anisotropy. To reproduce this in some lupus patients and see if it holds true could give a very good parameter for following people.

Functional Anisotropy in SLE

There are abnormalities in functional anisotropy in lupus.  The thalamus, the corpus callosum, parietal and frontal white matter lesions are all very different in normal controls versus neuropsychiatric lupus.

Evidence of reversible axonal dysfunction using NMR spectroscopy

Another measure that can be examined is the NMR spectroscopy––comparing different chemicals in the brain. When used to examine people who’ve gone from active disease to inactive disease or inactive disease to active disease, parameters can be defined for certain chemicals. This is probably not a good outcome measure because there’s too much platform time.  The patient is in the MRI for a long period of time.

Functional MRI: BOLD images

The functional MRI using the blood oxygen level determination shows a larger area that changes for lupus patients.

Treating Vascular Cognitive Impairment?

What should be done for the patient?

  • Treat risk factors –blood pressure, smoking, lipids, DM, etc.
  • Other interventions for vascular disease
    • ASA and other anti-platelet agents
    • Statins for patients without high cholesterol
    • EPA +DHA (Wright ARD 2008)
  • Treat APL antibodies (what about non-thromb?)
  • Treat lupus (need for biomarkers)
    • Hydroxychloroquine (for everyone; check levels)
    • Glucocorticoids and immunosuppression
    • Anti-interferon?
  • Follow-up – imaging; neuropsychiatric testing.

For CME credit,TAKE POST-TEST & EVALUATION

All information contained within the Johns Hopkins Arthritis Center website is intended for educational purposes only. Physicians and other health care professionals are encouraged to consult other sources and confirm the information contained within this site. Consumers should never disregard medical advice or delay in seeking it because of something they may have read on this website.

Updated: August 16, 2012

Arthritis Center

About Arthritis Center

Founded in 1998, the Arthritis Center at Johns Hopkins is dedicated to providing quality education to patients and healthcare providers alike.