Leonardo Pantoni Compares Features of CADASIL and NOTCH3 Negative Patients

February 4, 2010


Leonardo Pantoni

Leonardo Pantoni, MD, PhD, from the Department of Neurological and Psychiatric Sciences of the University of Florence, compares the clinical, familial, and MRI features of CADASIL and NOTCH3 negative patients, in his paper on the topic recently published in Neurology® (2010;74:57-63). He spoke with José Merino, MD, AAN.com Science editor.

AAN.com: Please describe the methodology and results of your study.

Pantoni: We performed NOTCH3 gene analysis (exons 2-23) in 81 patients, because Cerebral Dominant Arteriopathy with Subcortical Infarcts and Leukoencephalopathy (CADASIL) was clinically suspected. The disease suspicion was confirmed in 16/81 (20 percent) probands by finding a mutation leading to a cysteine substitution within the epidermal growth factor (EGF)-like repeats of the NOTCH3 receptor, while in 65 patients no pathogenic mutation was found. We then compared the clinical, familial, and neuroimaging features of patients with and without pathogenic mutations.

The following features were significantly more frequent in CADASIL than in NOTCH3 negative patients: history of migraine (73 vs. 39 percent), stroke before the age of 60 among relatives (71 vs. 32 percent), severe leukoencephalopathy (94 vs. 62 percent), white matter changes extended to the anterior temporal lobes (93 vs. 45 percent), external capsule involvement (100 vs. 50 percent), and presence of lacunar infarcts (100 vs. 65 percent). The frequency of vascular risk factors was balanced between the two groups. No feature was peculiar to either group.

AAN.com: What are the phenotypic characteristics of CADASIL?

Pantoni: The typical clinical features of CADASIL are migraine, frequently with aura, which is usually the initial manifestation and can be present as early as in the second decade of life; ischemic cerebrovascular events; mood disorders; and cognitive decline. The course is usually progressive leading to dementia and disability. Less commonly, some patients may have epilepsy, visual deficits, and reversible disturbances of consciousness. From the radiological point of view, CADASIL is characterized on MRI by diffuse and severe white matter changes, frequently extending to the anterior pole of the temporal lobe and multiple lacunar infarcts. Also, depending on the patient's age, these radiological features can be milder or more severe. In its full clinical-radiological expression, CADASIL is quite easily recognizable, but the phenotype is highly variable and many patients present with an incomplete picture, advanced age, or with uncommon symptoms such as brain hemorrhages (instead of the typical ischemic manifestations) or transient disturbances of consciousness. Unfortunately, there are no pathognomonic features of the disease, although the presence of white matter hyperintensities in the anterior part of the temporal lobes and in the external capsule is typical of the disease.

AAN.com: How is CADASIL diagnosed? Is there a role for skin testing now that the genetic test is widely available?

Pantoni: CADASIL is usually initially suspected because of the clinical features, and neuroimaging (MRI) provides additional support for the suspicion of the disease. CADASIL is caused by mutations of the NOTCH3 gene located on chromosome 19, and mutations can occur on each of the 23 exons codifying for the Epidermal Growth Factor (EGF)-like repeats. The diagnosis is confirmed when genetic testing shows a mutation of the gene. A large number of mutations that lead to a cysteine substitution have been reported. Clinicians should keep in mind, however, that some patients may have atypical mutations that may not be detected with our current tests.

Until a few years ago, not all the laboratories were able to screen the whole gene, and the diagnosis of CADASIL required the demonstration, by electron microscopy, of deposits of granular eosinophilic material on the basal membrane in a skin biopsy. Now that the gene test is widely available, skin testing is rarely done.

AAN.com: Until now, CADASIL has been synonymous with a NOTCH3 mutation, yet in a sample with clinical features of CADASIL, you found mutations with clear pathogenic significance in only 20 percent of your sample. Does this mean that the term CADASIL should be a phenotypic descriptor, with several subtypes based on genetic etiology?

Pantoni: This could indeed be the case. In our sample of NOTCH3 negative patients, a subgroup was phenotypically indistinguishable from CADASIL in both clinical and neuroimaging terms. At present, researchers are exploring two issues:

  • Whether these "CADASIL-like" patients bear alterations in other genes, as is seen in diseases where mutations of different genes are responsible for the same disease
  • If confirmation can be found for the pathogenic value of the alterations not leading to a cysteine substitution

  AAN.com: What do you think is the significance of NOTCH3 mutations that do not lead to a cysteine substitution?

Pantoni: At present it is unclear what they mean. We need to confirm the role of these mutations in producing the disease. At present, this is done by searching for the same gene alterations in population samples and exploring whether these alterations just represent polymorphisms or not. The role of skin biopsy in this sense is somehow limited because, if GOM are not found, the presence of the disease cannot be ruled out, given the limited sensitivity of the skin exam. We are also studying the RNA product of these alterations to see whether it is altered.

AAN.com: Could some of the NOTCH3 mutation negative patients have sporadic small vessel disease due to a clustering of risk factors, despite a positive family history of stroke, migraine, or mood and cognitive changes?

Pantoni: Yes but we must keep in mind that all the disturbances that make up the picture of CADASIL (migraine, mood disorders, and stroke) are influenced by genetics. Therefore, some of these NOTCH3 negative patients could in effect be sporadic cases (i.e., without single gene alteration) in whom not only a clustering of risk factors occurs but also a familial susceptibility for each single disturbance could lead to a CADASIL-like picture. Based on our experience, we avoid excluding from genetic screening patients with vascular risk factors, provided they have the clinical-neuroimaging picture of CADASIL.

AAN.com: In which patients do you test for NOTCH3 mutations? Since some cases with de novo mutations have been described, should the diagnosis be considered in someone with the phenotypic characteristics of the disease in the absence of a family history?

Pantoni: Very few de novo cases have been reported in the literature. We suspect CADASIL in all patients in whom the clinical picture is highly suggestive, when:

  • There are at least three typical clinical features of the disease and extensive white matter changes on MRI
  • White matter hyperintensities extend to the anterior part of the temporal lobe
  • A subcortical stroke occurs in the absence of vascular risk factors
  • A complete work-up has excluded other rare causes of stroke

However, we also usually require a suggestive family history. Sometimes a reliable family history is not obtainable. We have genetically screened a few patients without family history when they have a highly suggestive clinical picture and temporal lobe involvement on MRI.

Considering the differential diagnosis of CADASIL, we perform laboratory tests to exclude other causes of white matter disease such as multiple sclerosis and metabolic or infective leukoencephalopathies. If the patient has a history of stroke and the MRI shows lacunar infarcts, we explore the rarer causes of stroke, such as Fabry disease and vasculitis.

AAN.com: What data would be needed to establish an algorithm to identify patients who should be screened for NOTCH3 and other—still unidentified—genetic mutations?

Pantoni: As also evidenced by our research, we lack a pathognomonic feature of the disease. We are currently working to develop an algorithm for the pre-genetic screening of CADASIL patients that could provide a predictive value for the diagnosis and also limit the requirement of genetic testing. At present, it seems that the variables needed for developing such an algorithm will be derived from a combination of family history, clinical features, and neuroimaging findings. For the possible unidentified mutations, large studies are needed with a strong genetic substrate and capability.

AAN.com: A diagnosis of CADASIL has significant personal and familial implications, and there is no treatment to prevent or slow the progression of the disease. What do you tell patients about the implications of the genetic test? Do you test family members of probands?

Pantoni: Before performing the genetic test and obtaining a consensus for it, we always have a discussion with the patient and his or her family. We illustrate the autosomal dominant nature and the natural history of the disease, including the high variability of it. We clearly state that a specific therapy is not yet available, but we also tell them that we will do our best to control the other classical vascular risk factors if they are present and that we will follow up with them regularly. Moreover, we have ongoing clinical trials to offer these patients.

We are aware that talking about the possibility of a genetic disease may result in psychological consequences to the patient. One additional problem with CADASIL in this regard is that mood disorders are part of the phenotype of the disease and therefore some patients may enter a depressive phase after this communication. In our experience, the patients and their families are usually reassured by the presence of a team specifically dedicated to this problem.

Considering the family members of probands, at the time of communicating the diagnosis to the patient, we declare our availability to perform the test in other family members even if they are asymptomatic. We believe this is their right, but we never force this decision. There is also another relevant issue, related to the above, with clear ethical implications: today it is possible to perform a prenatal diagnosis of the disease and some young patients have already started to ask us for this. This is of course a delicate issue, considering that the disease may have a variable natural history and some patients show a more benign course than previously thought.

Author Disclosure

Dr. Merino performed a one-time consultation with staff from Bell, Falla and Associates.