Introduction and Statement of Symposium Goals Bernard L. Maria, MD, MBA, Executive Director of the Children's Research Institute; Jeffrey Edwin Gilliam Chair and Professor, Medical University of South Carolina, Charleston, SC
The symposium on tuberous sclerosis complex was the third in a series of five symposia titled "Neurobiology of Disease in Children." The conference was supported by the National Institute of Neurological Disorders and Stroke (NINDS), the Tuberous Sclerosis Alliance, and the Child Neurology Society. Dr Maria outlined the following goals for the series:
- Identify research findings that have the potential to enhance clinical practice in child neurology
- Identify future research needed to improve diagnostic accuracy and to develop safe and effective therapies
- Enhance collaboration between basic and clinical scientists
- Introduce students to research on neurologic disease in children
Objectives specific to the tuberous sclerosis complex symposium included description of diagnostic strategies, review of pathogenesis, discussion of therapies and clinical trials, definition of future research directions, and publication of the proceedings.
Current Approaches to Diagnosis of Tuberous Sclerosis Complex
Moderator: E. Steve Roach, MD, Professor of Neurology, Pediatric Neurology, and Pediatrics, Wake Forest University Baptist Medical Center, Winston-Salem, NC
Overview and Approaches to Diagnosis
E. Steve Roach, MD, Wake Forest University Baptist Medical Center, Winston-Salem, NC
Dr Roach reviewed the criteria for clinical and genetic diagnosis of tuberous sclerosis complex. In his presentation, he outlined the associated clinical features of the disease. Tuberous sclerosis complex is an autosomal dominant neurocutaneous disorder with a prevalence of 1 in 6000 to 9000 people. Although a positive family history of tuberous sclerosis complex is frequently observed, spontaneous genetic mutations are common (65-75% of cases). Genetic studies have resulted in the identification of two genes implicated in the disease: TSC1 and TSC2.
Definite tuberous sclerosis complex, as defined by the 1998 consensus conference sponsored by the Tuberous Sclerosis Alliance and the National Institutes of Health (NIH), is diagnosed when at least two major or one major plus two minor features are present. Probable tuberous sclerosis complex includes one major and one minor feature. Possible tuberous sclerosis complex includes one major or two or more minor features. Major features include skin manifestations (ie, facial angiofibromas, ungual fibroma, more than three hypomelanotic macules, and shagreen patch), brain and eye lesions (ie, cortical tuber, subependymal nodules, subependymal giant cell astrocytomas, multiple retinal nodular hamartomas), and tumors in other organs (ie, cardiac rhabdomyoma, lymphangioleiomyomatosis, renal angiomyolipoma). Minor features include multiple randomly distributed pits in dental enamel, rectal polyps, bone cysts, cerebral white-matter migration abnormalities on brain imaging, gingival fibromas, nonrenal hamartomas, retinal achromic patches, confetti skin lesions, and multiple renal cysts. The 1998 criteria do not include symptoms such as seizures or mental retardation to avoid "double counting" (ie, central nervous system lesions cause seizures, and including both in the criteria leads to counting the same symptom twice). Associated neurologic features also include seizures, autism or pervasive developmental disorders, mental retardation, and various learning and behavioral disorders.
The clinical criteria outlined above are useful, despite the availability of a genetic test for tuberous sclerosis complex, because they are quick, accurate, and inexpensive. The genetic test for tuberous sclerosis complex has a false-negative rate of 20%, partly because of complications in the way in which the disease is transmitted. Specifically, in 2% of patients, the mutated TSC gene is not detected in either parent because of germline mosaicism. This might be more frequent when the mother is the carrier and when TSC2 mutations are involved. Despite these difficulties, genetic testing is still indicated when the clinical diagnosis is unclear and/or when parents are making decisions about family planning. Phenotypic variability can be related to various factors: (1) the stronger phenotypic presentation in patients with TSC2 gene mutations, (2) somatic mosaicism, and (3) the specific type of genetic mutation.
Brain Abnormalities in Tuberous Sclerosis Complex
Francis J. DiMario Jr, MD, Professor of Pediatrics and Neurology, University of Connecticut School of Medicine, Farmington, CT; Associate Chair for Academic Affairs and Faculty Development, Department of Pediatrics, Chief, Division of Pediatric Neurology, Connecticut Children's Medical Center, Hartford, CT
Dr DiMario described the central nervous system lesions in tuberous sclerosis complex, the relationship between tuberous sclerosis complex and neuronal migration disorders, and morphometric abnormalities observed in brain imaging studies. Ninety percent of people with tuberous sclerosis complex exhibit at least one supratentorial brain lesion, including cortical tubers, subependymal nodules, subependymal giant cell astrocytomas, white-matter linear migration lines, corpus callosum agenesis or dysplasia, and transmantle cortical dysplasia.
Infratentorial brain lesions are less common (< 2% of patients) and can include linear and gyriform cerebellar folia calcifications, cerebellar nodular white-matter calcifications, agenesis and hypoplasia of the cerebellar hemispheres and vermis, enlargement of the cerebellar hemispheres, and subependymal nodules and tubers in the brain stem and fourth ventricle.
Brain imaging structural abnormalities can occur owing to disturbed neural migration, leading to changes in the number, size, and/or thickness of cortical gyri; heterotopic neuronal aggregates; and variable degrees of cortical cytoarchitectural disorganization (eg, aberrant columnar and laminar arrangement). Specifically, hamartomas are thought to represent focal areas of improperly proportioned cellular components that result from poor cellular organization or differentiation. Tubers, subependymal nodules, and subependymal giant cell astrocytomas result from focal tissue dysplasia, whereas a transmantle cortical dysplasia represents a more diffuse tissue dysplasia.
Dr DiMario and colleagues examined morphometric measures in patients with tuberous sclerosis complex to determine whether abnormalities in neural migration affect whole brain development and function. The authors found that 92% of participants with tuberous sclerosis complex had focal abnormalities (mean 5.7 subependymal nodules, 6.7 tubers). Abnormalities were most common in the periventricular and frontal region, including larger ventricles and a corresponding decrease in gray and white matter. The number of subependymal nodules was associated with increased posterior ventricle size, whereas the number of tubers was associated with increased anterior ventricle size. Subependymal nodules and tubers were also associated with increased frequency of seizures, cognitive impairment, and adaptive functioning. Regional morphometric brain disturbances were observed both in conjunction with and separate from periventricular migration lesions.
Structural Magnetic Resonance Imaging of Tuberous Sclerosis Complex
Edward Bullmore, PhD, Professor of Psychiatry; Khanum Ridler, PhD, Research Associate, University of Cambridge, Cambridge, UK
Dr Bullmore reviewed the work of Dr Khanum Ridler describing morphometric magnetic resonance imaging (MRI) studies in individuals affected with tuberous sclerosis complex. Two traditional approaches are used in examining morphometry. The first is the "region of interest" approach, in which the volume of the area most likely to be affected by tuberous sclerosis complex is estimated. The benefits of the approach include accessibility to expert information, definition of regions as native space, and support of regional-level factorial modeling. The disadvantages include the need for a prior hypothesis, a lack of comprehensiveness, the intensity of human labor required, and less-than-perfect interrater reliability. Recent advances have led to computational approaches in which an axial image of the brain is obtained, the brain image is lifted from the rest of the image, and estimates of brain tissues are made. This approach requires no prior knowledge of the brain abnormality; is comprehensive, robust, and reliable; and supports voxel and cluster level computational modeling.
Drawbacks include a lack of accessibility to expert information and the use of standard space modeling. Automated anatomic modeling is a new approach that combines aspects of region of interest and computational methods. In this method, brain maps are put into the space, but a template divides the gray matter into 45 regions of interest in each hemisphere.
Dr Bullmore described the results of two case-controlled studies that used computational morphometry to study patients with tuberous sclerosis complex. These studies collectively suggested that bilateral symmetric deficits in subcortical gray matter and intrahemispheric white matter could be considered markers of tuberous sclerosis complex. Tubers were distributed symmetrically and were most often located in the frontal lobe. The density of tubers was greatest in the parietal lobes, and tubers were more frequent in the cingulate gyrus than expected. Tuber volume was positively correlated with the number of subependymal nodules and negatively correlated with gray-matter volume. However, the location of tubers and subependymal nodules was unrelated to the location of gray- or white-matter deficits. Furthermore, when brain volume was corrected for the volume of tubers and subependymal nodules, gray-matter deficits were still apparent. In a group of patients of average intelligence, cognitive deficits in working memory and immediate and delayed recall were noted. In contrast, recognition memory, inhibition, and processing speed were intact. Memory scores were correlated with subcortical gray-matter volume.
However, tuber and subependymal nodule volumes were unrelated to cognitive processing skills. The authors concluded: "Computational morphometry can complement and add value to anatomical characterization of neurodevelopmental syndromes."
Behavioral and Cognitive Aspects of Tuberous Sclerosis Complex
Penny Prather, PhD, Department of Neurology and Psychiatry, Massachusetts General Hospital, Harvard Medical School, Boston, MA
Dr Prather reviewed the literature that suggests that tuberous sclerosis complex can be considered a risk factor for cognitive and developmental problems. Early descriptions of tuberous sclerosis complex emphasized the presence or absence of mental retardation or "delay." Although few studies examine cognition closely, as more patients are identified with tuberous sclerosis complex, it becomes apparent that cognitive outcomes vary more than was previously reported. Studies largely using parent surveys and reports of functional abilities estimate that 37 to 65% of children with tuberous sclerosis complex exhibit mental retardation. In one of only two studies to use structured tests, intellect was normally distributed, with the exception of a large cluster of children (30%) functioning in the severely deficient range. Neuropsychologic studies ( N = 5) suggest additional deficits in executive functioning and attention in children with tuberous sclerosis complex. In addition, children with tuberous sclerosis complex are at risk of symptoms of autism (≥ 40% of patients) and exhibit an increased risk of disruptive behavior disorders (eg, impulsivity and aggression) and social problems (eg, poor judgment and social awkwardness).
Early onset of seizures (< 3 years of age; infantile spasm) and/or intractable seizures appear to be associated with an increased risk of neurodevelopmental and cognitive problems. Moreover, the number of tubers present on brain MRI is positively correlated with age at seizure onset and negatively correlated with global cognitive functioning. In this respect, it is unclear whether tubers, seizures, or both are specifically related to the learning deficits seen in children with tuberous sclerosis complex. Similar factors can be associated with autism in tuberous sclerosis complex. Specifically, Bolton and colleagues (2002) found that the concurrent finding of tubers in the temporal lobe and the early onset of temporal lobe seizures (< 12 months of age) were associated with the diagnosis of autism, implicating disruption of subcortical-orbitofrontal circuits.
Dr Prather summarized her research on neuropsychologic profiles in a sample of individuals affected with tuberous sclerosis complex ranging in age from 3 months to 50 years. Fifty-eight percent of the individuals in her sample were described as having cognitive function within normal limits, whereas 42% were described as delayed. Among children aged 6 to 18 years who completed a full testing battery (21%), deficits in regulation or executive functioning (62%), visual-motor organization and planning (54%), memory or information retrieval (38%), language (24%), and spatial (19%) abilities were observed. She concluded with a summary of consensus recommendations for neuropsychologic evaluation of children with tuberous sclerosis complex. Guidelines emphasize the importance of repeated evaluation at times associated with developmental shifts (ie, infancy, toddlerhood, preschool, grade-school entry, entry to upper primary grades, and adolescence).
Autism and Tuberous Sclerosis Complex
Max Wiznitzer, MD, Associate Professor of Pediatrics, Division of Pediatric Neurology, Case Western Reserve University, Rainbow Babies and Children's Hospital, Cleveland, OH Dr Wiznitzer reviewed the core symptoms of autism, including impaired social interaction and communication and repetitive and restricted interests present before the age of 3 years. Much evidence for comorbidity between autism spectrum disorders and tuberous sclerosis complex comes from an extensive series of case studies. However, the methodology in many of these case studies is limited (eg, incomplete evaluation of tuberous sclerosis complex or autism, waxing and waning of the symptoms of autism). In group studies of autism, 1 to 4% of children exhibit tuberous sclerosis complex. Conversely, in group studies of tuberous sclerosis complex, 25% or more of children exhibit autistic features. However, several factors make it difficult to interpret this research.
These include unrepresentative samples, failure to use "gold standard" assessment tools, and the presence of other disorders associated with symptoms of autism (eg, mental retardation and seizures). In tuberous sclerosis complex, an equal number of boys and girls are diagnosed with autism. This is in contrast to studies of classic autism, in which boys outnumber girls.
Several possible reasons for the increased risk of autism in children with tuberous sclerosis complex were discussed. First, the genes associated with tuberous sclerosis complex can directly affect the development of brain regions associated with autism. Research indicates that the TSC gene products are expressed in regions of the brain implicated in autism (ie, frontal and temporal lobes). Second, autism spectrum disorders can occur in association with tuberous sclerosis complex-related brain disorganization, including cognitive impairment, seizures, and tuber location.
Symptoms of autism are prevalent in children with mental retardation, regardless of etiology. Given that mental retardation is prevalent in this population, autism spectrum disorders could be related to intellect. Alternatively, seizures could disrupt central nervous system organization, resulting in symptoms of autism spectrum disorders that could wax and wane along with seizure activity. The presence of tubers in specific brain regions also could be explanatory, although these findings were not conclusive. Finally, there could be linkage between the TSC genes and genes related to autism spectrum disorders. A genome scan implicated chromosome 16 in autism spectrum disorders. In that the TSC2 gene is located on chromosome 16, these genetic studies suggest that linkage between autism spectrum disorder and tuberous sclerosis complex might be possible. However, this seems unlikely because of differences in gender ratio, the spectrum of cognitive impairment, and the frequency of seizure disorders between the tuberous sclerosis complex and autism spectrum disorder populations. Dr Wiznitzer suggested that physicians should consider the possibility of tuberous sclerosis complex in individuals with autism, particularly if seizures and/or mental retardation are present.
Conversely, monitoring individuals with tuberous sclerosis complex for autism spectrum disorders is also indicated, with particular attention to the possibility of emergence or worsening of symptoms associated with seizure onset.
Question and Answer Session 1
Dr Roach: Can you elaborate on how the morphometric analysis was performed on the scans?
Dr DiMario: The measures were conducted primarily on axial sections at the level of the caudate.
Audience Member: Why does there seem to be frontal predominance of lesions in tuberous sclerosis complex?
Dr DiMario: There is no good answer for that question, but there must be a biologic explanation relating to the ventricular zone in that region.
Audience Member: Do you know if changes occur over time in tuberous sclerosis? In attention-deficit hyperactivity disorder (ADHD) and schizophrenia, changes can be shown in longitudinal studies.
Dr DiMario: We haven't looked at it ourselves, but patients with frequent seizures taking antiepileptic drugs could well have progressive cortical changes.
Dr Roach: The classic dogma has been that lesions develop before 22 weeks' gestation, and everything' is then "set in stone." On the other hand, a normal brain is continuing to evolve after birth, and some patients seem to acquire lesions. There is often a question about whether the imaging technology has become more sensitive or whether a lesion is truly progressive.
Audience Member: There appears to be no anatomic correlation between gray-matter deficits and the distribution of the tubers, but is it possible that reciprocal connections between the two areas could account for the gray-matter deficits?
Dr Bullmore: I think that's a very interesting point. Generally, these structures are part of circuits, and presumably that is why damage to a region is associated with impairment of so many different aspects of working memory. One of the things we'd like to do is take the data analysis forward a little bit and see if we can understand whether there are differences between the cases, not just on a region-by-region basis but in terms of the circuits. I think the idea that there might be some kind of remote connections or correlations between anatomically remote areas of regional deficit is absolutely correct.
Dr Maria: Why do you think serial recognition memory seems to be much more robust? You showed us very impressive thalamic involvement. One would have expected serial recognition memory to be as impaired. Any thoughts on why that might be?
Dr Bullmore: No, it's very difficult to explain that, and I think I should make a number of points. One is that this general approach to taking structural data and psychological test scores and trying to understand one in terms of the other is a rich approach and could be used to look at various psychopathologies and to understand abnormal variation. There isn't a great hypothesis about what is expected. It would be very nice in a few years' time to be able to come back to the question.
Dr Roach: That's a pretty remarkable group of tuberous sclerosis patients with an average IQ of 114. Have you studied more typical individuals with IQs of 50 to 60?
Dr Bullmore: I think the choice of higher IQ participants was pragmatic because we wanted good quality imaging data to examine the relationship between structure and function. But it would be certainly very interesting and important to know more about the anatomic profile and also structure and function, with perhaps a more representative sample of patients.
Audience Member: Do the data show that the presence of infantile spasms is a risk factor for cognitive dysfunction in tuberous sclerosis? The numbers seem small to draw that conclusion.
Dr Prather: I agree that the data do not yet show a clear association between the presence of infantile spasms and cognitive dysfunction in tuberous sclerosis.
Dr Roach: That's a good point. There was an earlier paper by Charles Shepherd and Manny Gomez that looked at number and variety of spasms. Quite clearly, if there are many lesions, spasms are more likely, but cognitive dysfunction could be related to the number of lesions in the first place.
Dr Maria: Are there any longitudinal data in the way of neuropsychological assessment that might give us some sense of natural history? That's my first question. And my second question is what would you recommend as an approach to neuropsychological investigation? Are we saying now that every child who would be entering kindergarten would be required to have a full neuropsychological evaluation? What would you generally recommend?
Dr Prather: There are some consensus guidelines coming out, and I think this is in the "forewarned is forearmed" requirement, but I think it's very helpful for a number of reasons to have re-evaluations at regular intervals in tuberous sclerosis. I believe that children with tuberous sclerosis should be evaluated when entering school, in the fourth grade, and again in the ninth grade.
Audience Member: Dr Wiznitzer, I wanted to address your comments about linkage studies on chromosome 16. You were a little bit less enthusiastic with reference to the TSC2 gene because of the difference in the gender ratio.
Dr Wiznitzer: No, what I meant is that if we're looking for an independent linkage between an autism gene and autism just goes along for the ride, we probably would expect to see the male:female ratio that we see in autism.
Audience Member: Dr Wiznitzer, you provided a very elegant and extensive survey of this tuberous sclerosis association. Recognizing that your population of pediatric neurologic patients comprises several thousand children with autistic spectrum disorders, how common is tuberous sclerosis in your population of patients?
Dr Wiznitzer: In my patient population, I have yet to identify a patient with autism who did not have other clinical features of tuberous sclerosis, including epilepsy.
Audience Member: Can tuberous sclerosis present with strictly behavioral problems? Dr Wiznitzer: It would be unusual for behavioral problems to be a presenting complaint in the absence of developmental delay. According to the American Academy of Neurology guidelines, children with unexplained global developmental delay should have MRI. But even before considering an MRI, it is important toconduct a dermatologic examination. In my opinion, the physical examination is paramount to making the diagnosis of tuberous sclerosis. I usually do not order neuroimaging for individuals with autism.
Audience Member: Do brain lesions in tuberous sclerosis shrink or disappear, as has been reported for cardiac lesions?
Dr Roach: The heart lesions shrink within 2 to 3 years, but lesions in a 6 year old are not likely to vanish. However, postmortem exams can show residual cardiac lesions that were too small for imaging resolution. There is no information about central nervous system lesions disappearing.
Audience Member: And you don't know why the heart lesions can be smaller?
Dr Roach: The cardiac lesions are rich in glycogen, and there's speculation that the glycogen is used up.
Dr Maria: I would normally think about the functional disability as related to the cortical lesions, but Dr Bullmore's data suggest a relation with deep gray-matter structures such as the thalamus. Please clarify.
Dr Bullmore: The functional consequences resulting from given lesions may be attributed to disruption of neuronal circuits rather than direct involvement of tissue by lesions. Although we can document functional problems from focal deficits, the function is probably normally served not just by regions but by circuits of which they're a constituent part. And I think the other thing to say, as Dr Roach was pointing out, is that particular sample was not a very impaired sample, thus perhaps not representative of usual tuberous sclerosis.
Dr Prather: Just to add to that, the statistics tell what's most frequent. It may just be that the frontotemporal is most common; that's what most people have, but they have other things too.
Dr DiMario: I think also the circuitry is at issue. Even areas that seem normal on MRI may be microscopically affected.
Dr Roach: I think that's part of the issue; you have to distinguish between what you find normal or above normal and then consider the whole question: Is there a correlation in someone that's poorly functioning and someone that's better functioning?
Dr Roach: Dr Wiznitzer, if you're having trouble sorting out the relative contribution to autism of lesion location, lesion type, and so on and cognitive function, do you see enough people with tuberous sclerosis who have autism who are not functioning to do a big study and just not include anyone who's retarded?
Dr Wiznitzer: They have to be out there. I have not seen any, no, but then we have to assume that if you're doing a background or if you're dealing with a tuberous sclerosis association... I would guess that few people have a high number of individuals in their patient population who have autistic features and tuberous sclerosis complex but have normal intelligence and have epilepsy.
Treatment and Management of Complications
Moderator: E. Steve Roach, MD
Managing Epilepsy in Tuberous Sclerosis Complex
Elizabeth A. Thiele, MD, PhD, Director, Tuberous Sclerosis Complex Comprehensive Clinical Program; Director, Pediatric Epilepsy Program, Massachusetts General Hospital, Harvard Medical School, Boston, MA
Dr Thiele described epilepsy as "the most common disorder in tuberous sclerosis complex." Seizures occur in up to 92% of people with tuberous sclerosis complex and are most common during the first year of life. All seizure types, except typical absence seizures, can be observed in children with tuberous sclerosis complex. Although increased numbers of tubers are associated with seizures, 10% of people with tuberous sclerosis complex who have intractable tuberous sclerosis complex exhibit normal MRI findings. Abnormal electroencephalograms (EEGs), such as those with epileptiform features (ie, focal spike or sharp-wave discharges, multifocal epileptiform abnormalities, hypsarrhythmia, and generalized spike-and-wave discharges) and diffuse slowing, are commonly observed. The presence of seizures is associated with a poorer cognitive outcome. Treatments are similar to those offered to patients without tuberous sclerosis complex who have epilepsy, including medication, vagus nerve stimulation, the ketogenic diet, and surgery. However, some children with tuberous sclerosis complex and epilepsy respond poorly to conventional antiseizure therapies.
Tuberous sclerosis complex is the leading cause of infantile spasms, accounting for 25% of all cases. At least one third of all patients with tuberous sclerosis complex have infantile spasms, typically between 4 and 6 months of age. For one third of children with tuberous sclerosis complex, spasms are preceded by partial seizures. These spasms can be difficult to detect and can manifest as subtle minor changes in infant behavior (eg, indifference to the social environment, irritability). EEG abnormalities do not necessarily include hypsarrhythmia and often demonstrate one or more foci. Research has suggested that vigabatrin is the preferred treatment for infantile spasm, and in many studies, vigabatrin eliminated spasms in most children with tuberous sclerosis complex. Vigabatrin is an irreversible γ-aminobutyric acid (GABA) transaminase inhibitor typically prescribed in doses of 100 to 200 mg/day titrated up over several days. Treatment usually lasts for 1 year, but it is unclear whether shorter treatment periods (eg, 4-6 months) are also effective. Unfortunately, vigabatrin can result in ophthalmologic toxicity in 5 to 20% of patients and include the reduction of 30 Hz flicker cone b-wave amplitude in full-field electroretinography, concentric visual field deficits, and reversible effects on the outer retina. Little is known about the effects of dose and length of treatment on toxicity and whether all of the ophthalmologic effects are reversible. Accordingly, electroretinographic examinations are obtained periodically and used to plan treatment. If other seizure types are present, additional medications might be necessary. Early treatment of infantile spasm might preserve cognitive functioning in some children. However, even without treatment, one third of children with tuberous sclerosis complex and infantile spasms show generally good overall cognitive outcomes. Dr Thiele noted that vigabatrin is not approved by the US Food and Drug Administration (FDA), limiting its availability for treatment in the United States. The session ended with a review of the remaining questions on etiology and the nature of seizures in tuberous sclerosis complex.
Epilepsy Surgery for Children with Tuberous Sclerosis Complex
Howard L. Weiner, MD, Daniel Miles, MD, Orrin Devinsky, MD, New York University Comprehensive Epilepsy Center, New York University School of Medicine, New York, NY
Surgical treatment of epilepsy is challenging in people with tuberous sclerosis complex because multiple bilateral lesions located in eloquent cortex are common. Thus, many children with tuberous sclerosis complex are rejected for surgical treatment. However, independent studies (~ 50) suggest that surgery is generally effective. These studies are limited by their inclusion of predominantly older children or adolescents, resections of single lesions, and the failure to use invasive monitoring strategies to locate the epileptogenic area.
Dr Weiner advocates the use of innovative approaches in children with tuberous sclerosis complex when the surgery is safe, effective for both seizure control and promotion of good child development, and conducted within a comprehensive epilepsy center. He described multistage surgery for epilepsy. This approach has been used on a selected group of children with tuberous sclerosis complex and can be used with children who exhibit one lesion, multiple lesions, or no lesions. Monitoring strategies include grid, strip, and depth electrode placement. For bilateral lesions, electrodes are placed to locate the epileptogenic area, and the lesion is resected.
Electrodes are then replaced to determine whether the epileptogenic region was completely resected, and the lesion is resected again if necessary. This technique helps identify adjacent or distal epileptogenesis and is especially useful when presurgical data suggest eloquent cortex or bihemispheric involvement.
This procedure has been evaluated in a small series of 16 children (5 boys), in which 13 received the three-stage surgery. Ten of these children are seizure free at follow-up, three have rare seizures (two have seizures associated with high fever), and two exhibited a 50% reduction in seizure frequency. Three of these children had unilateral seizure onset that could not be detected without invasive monitoring. All but one child required the removal of two or more tubers, and two children underwent bilateral resections. In this sample, one child suffered an infection related to the operation and three experienced transient hemiparesis. Dr Weiner ended by emphasizing that MRI often misses lesions, that lesionectomy is rarely sufficient alone, and that invasive monitoring is critical to the success of surgical treatment for seizure disorders in this population.
Non-Neurologic Manifestations of Tuberous Sclerosis Complex
David N. Franz, MD, Associate Professor of Pediatrics and Neurology, Director, Tuberous Sclerosis Clinic, University of Cincinnati College of Medicine, Cincinnati Children's Hospital Medical Center, Cincinnati, OH
Dr Franz described the non-neurologic aspects of tuberous sclerosis complex, including dental, cardiac, renal, and pulmonary manifestations. Dr Franz noted the presence of dental pits and craters in permanent teeth in 100% of people with tuberous sclerosis complex and the frequent presence of gingival fibromas. He stated that dental examinations are useful in screening for tuberous sclerosis complex because multiple dental pits or craters and gingival fibromas are diagnostic of tuberous sclerosis complex. People with tuberous sclerosis complex also can exhibit fibromas in the toes.
Potential cardiac symptoms include ventricular tachycardia and fibrillation, arrhythmia, bradycardia, and heart block. In the case of cardiac rhabdomyoma, Dr Franz emphasized the importance of managing patients medically and allowing time for improvement in those whose tumors are not obstructive, rather than pursuing surgical approaches, because many of these fibromas tend to regress over time.
Kidney complications are the second most common cause of morbidity in people with tuberous sclerosis complex (after central nervous system complications). Angiomyolipomas are also observed, and polycystic kidney disease occurs in 5 to 8% of patients. Although not typically malignant, angiomyolipomas can grow and destroy kidney tissue, resulting in complications such as hemorrhage, hematuria, hypertension, and renal failure. Angiomyolipomas vary in composition, consisting of some combination of fat, smooth muscle, and blood vessels. Because abnormal vessels are present in angiomyolipomas, the risk of hemorrhage increases as lesion size increases (3.5-6 cm). Treatment options include observation, embolization, and total or partial nephrectomy. Dr
Franz recommended embolization because it preserves kidney tissue, prevents hemorrhage, and reduces kidney size and tumor volume. The risks include embolization of unwanted targets, infection, and postembolization syndrome, which can be reduced by presurgical corticosteroid treatment.
Lung manifestations include lymphangioleiomyomatosis, multifocal cysts, and multifocal micronodular pneumocyte hyperplasia (a benign condition sometimes mistaken for malignancy). Lymphangioleiomyomatosis in tuberous sclerosis complex can be distinguished from sporadic lymphangioleiomyomatosis, an uncommon, uninherited condition (incidence 0.0001%). Sporadic lymphangioleiomyomatosis is more likely to be progressive than lymphangioleiomyomatosis seen in individuals with tuberous sclerosis complex. Both types of lymphangioleiomyomatosis occur almost exclusively in women (up to 40% of women with tuberous sclerosis complex) and are associated with angiolipomas. Symptoms include cystic lung disease, dyspnea, cough, hemoptysis, recurrent pneumothoraces, lymphatic obstruction, and retroperitoneal and hilar adenopathy. Dr Franz concluded by speculating whether treatment with rapamycin inhibitors will be effective in treating some of the clinical manifestations of tuberous sclerosis complex.
Question and Answer Session 2
Audience Member: I have a practical question. What do you do about the availability of vigabatrin? We've been grappling with this in the practice community.
Dr Thiele: The question is a good one. I think this is a major question for the child neurology community. Vigabatrin is one of the few effective drugs I use that has not been approved by the Food and Drug Administration. However, the use of the drug is associated with a financial burden on families and a liability burden on practitioners. We do have some colleagues who don't feel comfortable prescribing vigabatrin, which I think is fine and readily defendable. And then, oftentimes, patients are referred to physicians outside this country. That's also fine. My concern with that is there may be a month or 2-month wait before the child is started on vigabatrin. I know a lot of our patients get vigabatrin from various patients, and I know in Mexico you can buy it at Sam's over the counter. There are different ways of getting it. I think there needs to be strong advocacy for having the drug made available in the United States.
Dr Roach: Written documentation of the benefits of vigabatrin could help address liability issues. Audience Member: Are there any data regarding age of exposure to vigabatrin and risk of retinal toxicity? And perhaps a related question, have any of your patients developed electroretinogram abnormalities on the drug? If so, how has that been managed?
Dr Thiele: It is not known what the risk of retinal toxicity is in relation to age of exposure to vigabatrin. I think when the reports first came out, they were in older individuals, and those people could be readily tested. I think now that we're relying on electroretinograms, it is not clear what the significance of those changes are, especially in younger children. At this point, we deal with patients on a case-by-case basis.
Audience Member: Is there any information about the natural history of disease in children with tuberous sclerosis, epilepsy, and behavioral problems?
Dr Thiele: There is no published natural history study in tuberous sclerosis. I can tell you from my experience with my patients, plenty of these kids do outgrow seizures and continue to go on infinitum without recurrence of the seizures. That clearly plays a role when you're looking at your way of treating it; should surgery be a strong option or not? As with other causes of epilepsy, I think seizure control is really of paramount importance in tuberous sclerosis.
Audience Member: Please discuss the methodology used in obtaining electroretinograms.
Dr Thiele: We refer to the ophthalmologist, who will first try to do an unsedated electroretinogram. Regrettably, many children require sedation to obtain good data.
Audience Member: How long should one treat with vigabatrin?
Dr Thiele: I typically keep kids on it for a year. I know there is the hope that a shorter duration of treatment will at least be effective in some kids. Audience Member: Dr Weiner, what was the age of your youngest surgical candidate with tuberous sclerosis? Did you use any positron emission tomography (PET) studies to correlate studies?
Dr Weiner: The youngest patient is the one I have in the hospital now, a 10-month-old boy. Most patients referred come with PET imaging data, but I have not found the information critically important in planning surgery.
Audience Member: I was unclear in terms of your technique; is this dependent upon finding interictal sites? How are you doing that with the very young child, who, in my experience, rarely tolerates being placed in the monitoring unit for several days to record the seizures?
Dr Weiner: All this work is really done with ictal data. I don't think any of the children in this series were done primarily on interictal spikes. And it's a challenge for the child, it's a challenge for the parents. I can tell you that, with this recent fresh experience with the 10 month old, he tolerated it quite well. His mother tolerated it less well. Basically, they come through it okay. But most of the patients come in with families with fears that with behavioral abnormalities, they're not going to be able to tolerate lying in bed for 2 weeks at a time.
Audience Member: Do you use sedation?
Dr Weiner: We don't use sedation at all. We use some pain management, light pain management with the pediatric intensive care unit staff.
Dr Roach: In younger patients, seizure frequency is high and is driving the surgical option. Thus, monitoring often does not have to be for a prolonged period.
Audience Member: What are the preinvasive criteria to go on to the invasive monitoring?
Dr Miles: Most of these children probably had two or three extensive monitoring sessions. Audience Member: Can you comment on the patients for whom surgery was not an option?
Dr Weiner: The patients I am seeing have been carefully screened for surgical eligibility. I can tell you that three of these patients went in for bilateral strip studies. We generally placed four holes and a small craniotomy at the top to just place survey strips. In those cases, I was very guarded about the possibility of going forward and conveyed that to the parents, and I was very surprised that we were able to lateralize the seizures. So, in general, I'm not seeing the patients, obviously, who are being rejected for surgery. I think they don't really even come to bilateral study.
Audience Member: I have a question about routine surveillance of sonograms; how are they influenced by the patient's age? And the last one is, as they cut down the frequency, is there any evidence of whether computed tomography (CT) or MRI is better? Is there a certain frequency?
Dr Roach: The closest thing in the literature that specifically addresses that is from the consensus conference, a series of recommendations on routine follow-ups once the diagnosis has been made. And that was in the Journal of Child Neurology of 1999, I think a year after the diagnostic criteria. And that came from basically the same group in Annapolis that did the diagnostic criteria division, and the approach is trying to do routine surveillance targeted toward features where early identification offers some advantage. And also where you have relatively noninvasive means of monitoring and also where you have some treatment. So, using that kind of scheme, you more or less took away extra eye exams, we took away chest CTs, and so on. This was not so much evidence, but at least the consensus was, at least among children, some routine imaging is merited. I don't think there's too much evidence to favor CT versus MRI; both are going to pick up enlargement, so I don't think there's too much indication to favor one versus the other. I think if it's available, MRI certainly shows it more eloquently. And the recommendation was to scan every 1 or 3 years, realizing that we're not in a position evidence-wise to come down real strong on the recommendation it had to be done every year, which, for a lot of these kids, clearly is too much. Even then, if you do it yearly and a child gets a giant cell tuber, it can enlarge and become symptomatic during that time. That same paper basically has the same recommendation: ultrasonography every 1 to 3 years. And most of us tend to shape that a little bit, depending on what the last finding is.
Dr Franz: If you do an initial sonogram, and it looks absolutely clean, you may be at lower risk of having an angiomyolipoma early on, but we still screen those kids. When you get to a lesion above 4 cm, we like to do a magnetic resonance angiogram to assess the presence of aneurysms. There is also a recommendation for the screening for lymphangioleiomyomatosis in women with tuberous sclerosis when they reach adulthood. Some people think, "What's the point; there's no treatment for lymphangioleiomyomatosis, and there's nothing you can do about it." I think that's a very worthwhile thing to do for a couple of reasons. Number one, the care for adults with tuberous sclerosis complex is rather abysmal. Lymphangioleiomyomatosis is misdiagnosed; they're told they have emphysema or they're told they have lung disease. The other thing is there's a number of lines of evidence that say if you have lymphangioleiomyomatosis, estrogen exposure may accelerate it or make it worse. I think it's worth suggesting to a person they may not want to take estrogen contained in contraceptives and things like that.
Dr Roach: One additional point: the 1998 paper that we published in the Journal of Child Neurology was a longitudinal study of ultrasonography of kidneys in 60 some patients with tuberous sclerosis complex. One thing that stood out is the idea that this old notion that if your initial sonogram of the kidney doesn't show a tumor, then you're home free is clearly wrong. The people who had normal renal sonograms at the time of diagnosis, or at least at the time of a perfect ultrasound study, fully half of those later developed a recognizable tumor. My guess is there would be a time when you could say, "Well, if you haven't developed one of these by the time you're 20 or something, your chances of getting one go down."
Audience Member: My question is about vigabatrin and infantile spasms. How often do you use vigabatrin by itself without other anticonvulsants in tuberous sclerosis complex patients?
Dr Thiele: If it's infantile spasms, I typically start kids on it as initial therapy. If they respond, then it's vigabatrin and only vigabatrin. If they don't respond and have spasms, then other treatments for spasm, such as adrenocorticotropic hormone (ACTH)? If they develop partial seizures, like I said, I haven't found vigabatrin particularly effective. I guess with all the cases I have with spasms and tuberous sclerosis, I would say two thirds of them are on vigabatrin alone. Audience Member: When you withdraw the vigabatrin, do they stay off all anticonvulsants?
Dr Thiele: That would be the goal, that they continue to be seizure free. If they do develop other seizure types, sometimes they go back on the anticonvulsants. If they were already on anticonvulsants, once they've been seizure free for a year, I'd get them off vigabatrin. Sometimes they are off the other medication before or after that time mentioned.
Audience Member: What about after staying off all medicine, after you withdraw?
Dr Thiele: That's a small number of kids. I would think I probably have started maybe 50 or 60 kids on vigabatrin for tuberous sclerosis, and I would guess that maybe 40% of those kids might be off medications and seizure free.
Dr Maria: As you know, one of the things we're trying to do today is to define some priorities in the way of future directions. My question is for Dr Franz. From a neurosurgical perspective, what are the important questions remaining, and how would you prioritize those?
Dr Franz: I didn't really have time to touch on that. What we're involved in is coordinating a national study looking at all the surgical cases that have been done in the United States up until now as a basis for proposing a prospective study. And one thing we're interested in looking at obviously is outcome, not only seizure outcome but also functional outcome. And then trying to correlate some of these issues such as imaging findings and how that relates to surgery and how that relates to outcome. So, in other words, I think the trend in this small series is that we're pushing the envelope somewhat, and the question is "Are we having any real impact on the disease?" That's the real question. So those are some of the things we're focusing on with this retrospective study. We're looking at radiology, we're looking at pathology, we're trying to correlate with a number of operations as to whether that has any overall impact, trying to correlate the sort of risk factors with outcome. But I think that noninvasive modalities for trying to understand the epileptic tubers are going to be critical and obviously molecular studies as well.
Audience Member: We had a patient in yesterday, and I wonder if anybody else had had the experience of managing a patient with abnormal signal in the pancreas or thyroid gland?
Dr Roach: Yes, I think each of those has been described. As I recall in Dr Gomez's book, one of the editions had mentioned at least both pancreatic and thyroid. Liver was very common. We didn't talk about the liver today; we looked at that in Dallas. About 25% of the children overall have benign liver lesions. As David mentioned, the lung is probably the most common one that we didn't talk about.
Audience Member: What are the statistics for getting rid of anticonvulsants in patients with tuberous sclerosis?
Dr Roach: That paper basically came out a couple of weeks ago. I think 15 patients, or 40%, managed to get off medicine.
Genetic Strategies in Tuberous Sclerosis Complex
Moderator: David H. Gutmann, MD, PhD, Donald O. Schnuck Family Professor of Neurology, Washington University School of Medicine, St. Louis, MO
Clinical Genetics of Tuberous Sclerosis Complex
Hope Northrup, MD, Director, Division of Medical Genetics, Professor, Department of Pediatrics, The University of Texas Medical School at Houston, Houston, TX
Dr. Northrup noted that tuberous sclerosis complex was first identified as a genetic disorder over 100 years ago. In 1935, Gunther and Penrose identified tuberous sclerosis complex as an autosomal dominant disorder but thought that 50% of cases were sporadic. However, today we think that 70% of tuberous sclerosis complex cases are sporadic, whereas 30% are familial. In the 1980s, it became apparent that the disease is heterogeneous, showing linkage to more than one gene on two different chromosomes. Linkage of the TSC1 gene on chromosome 9 was reported in 1987. This discovery was followed by the discovery of linkage for the TSC2 gene (chromosome 16) in 1992.
Currently, we know that the TSC2 gene is located on the short arm of chromosome 16 and is about 41 exons in size. Two of the exons are alternatively spliced, but mutations have never been discovered in these exons. The TSC1 gene is located on the long arm of chromosome 9 and is 23 exons in length. The first two exons are alternatively spliced. The gene products of the TSC1 (hamartin) and TSC2 (tuberin) regulate key intracellular growth control pathways, suppress the actions of ribosomal S6 kinase, and inhibit the mammalian target of rapamycin (mTOR). Hamartin and tuberin form a single signaling complex, which is critical for regulating a number of important cellular processes related to cell growth and size.
Dr Northrup described her work using data on mutations in more than 400 people with tuberous sclerosis complex to provide prognostic information. More than 1000 mutations are known to exist. Most mutations in the TSC1 gene constitute inactivating or nonsense mutations. In contrast, missense mutations constitute close to 25% of mutations. In families, close to 50% have mutations involving the TSC1 gene; however, TSC2 gene mutations are overrepresented in sporadic patients. Mosaicism occurs on an infrequent basis (1-2%) in tuberous sclerosis complex. Dr Northrup described her research indicating that phenotypes with neurologic deficits were more closely associated with mutations in the TSC2 gene than in the TSC1 gene. Similarly, more severe additional features of tuberous sclerosis complex were related to TSC2 gene mutations.
These findings were consistent with prior literature in suggesting a more severe phenotype for people with TSC2 gene mutations.
Molecular Genetics of Tuberous Sclerosis Complex: What Is New in 2003?
Elizabeth Petri Henske, MD, Member, Medical Science Division, Fox Chase Cancer Center, Philadelphia, PA
Dr Henske described findings suggesting that cells in the brain that lack TSC2 can have an unusual ability to migrate, as seen in lymphangioleiomyomatosis, and an unusual ability to differentiate, as seen in renal tuberous sclerosis complex. The authors investigated the possibility of a mutation in the TSC2 gene in patients with sporadic lymphangioleiomyomatosis, suggesting that sporadic lymphangioleiomyomatosis was either a mild form of tuberous sclerosis complex or reflected genetic mosaicism. Preliminary evidence suggested mutations in the TSC2 gene in angiomyolipomas of women with sporadic lymphangioleiomyomatosis. In a study of five patients with no mutations in their angiomyolipomas, researchers found no mutations in the peripheral blood, ruling out the mild tuberous sclerosis complex hypothesis. Furthermore, the only abnormal genes were found in abnormal lung and kidney tissue, arguing against genetic mosaicism. A third model suggested that cells were actually metastasizing from the kidney to the lungs, causing lymphangioleiomyomatosis. This was hard to understand because angiomyolipomas are benign tumors. The researchers examined a single case in which lymphangioleiomyomatosis recurred after transplant. A mutation was found in the lymphangioleiomyomatosis before transplant, and the mutation was not present in normal lung tissue. After transplant, the same mutation was found in a recurring lymphangioleiomyomatosis, suggesting the spread of lymphangioleiomyomatosis cells. More research is necessary to determine how these findings relate to the central nervous system, but researchers now wonder if tubers might result from abnormal migration in the central nervous system.
In addition, researchers wished to know whether cells deficient in TSC2 gene expression also had an unusual ability todifferentiate. They sought to determine whether the three types of cells found in an angiomyolipoma (ie, fat, muscle, and blood vessel) resulted from the same precursor cell. First, researchers examined whether blood vessels in angiomyolipomas were neoplastic or reactive. Five types of blood vessels were identified, and vessels were both reactive (those without loss of heterozygosity) and neoplastic (those with loss of heterozygosity). Thus, it seems that kidney cells lacking TSC2 gene expression are unusually plastic and are therefore able to differentiate into these three cell types. Whether central nervous system lesions result from abnormal differentiation during cortical development is not yet known. Dr Henske closed by reviewing research suggesting that rapamycin might be effective in treating tuberous sclerosis complex.
TSC Genes and the Brain
Peter B. Crino, MD, PhD, PENN Epilepsy Center, Department of Neurology, University of Pennsylvania Health System, Philadelphia, PA
Dr Crino began by pointing out that the link between loss of function of hamartin or tuberin and the change from a normal progenitor cell to a giant cell is unknown. This is important because tubers are characterized by disorganized cortical lamination, aberrant dendritic arbors and axonal projections, astrocytic proliferation, and abnormal cell morphology (ie, dysplastic and heterotopic neurons and giant cells).
Dr Crino described the mechanism for tuber formation during brain development. Gene mutations in either of the two TSC genes influence neural precursors between weeks 7 and 20 of gestation to result in disrupted cell division, abnormal cell differentiation, dysregulated cell size control, and abnormal cellular migration. However, it is not known whether this effect develops via haploinsufficiency (single germline mutation) or through a second "hit" caused by a stochastic or random DNA event. Dr Crino's group hypothesized that giant cells occur as a result of a second hit, whereas adjacent dysplastic cells are haploinsufficient and exhibit only the germline TSC gene mutation. To test this hypothesis, tubers were obtained intraoperatively from patients undergoing surgery for seizures. Examination of the tubers indicated that two tuber sections exhibited evidence of S6 kinase hyperactivation, with the giant cells being the only cells with this abnormal S6 kinase activation. In addition, researchers examined perituberal tissue and did not find this abnormal S6 kinase activation. They then dissected out individual giant cells and sequenced the messenger ribonucleic acid. In these experiments, Dr Crino and colleagues detected only the one mutation in dysplastic neurons. In the cerebellum, dysplastic cells express tuberin, but giant cells do not.
Tubers are thought to arise during embryogenesis within the ventricular zone, where a second somatic mutation occurs, affecting the neuroprogenitor cell. The mTOR pathway is disrupted, resulting in dysregulated cell size and proliferation. It is hypothesized that these cells then migrate in an abnormal fashion to the cortex to generate in abnormal collections of inappropriately positioned neurons. In conclusion, Dr Crino described recently published data suggesting that cortical tubers express markers of recent cell proliferation in both the Eker rat brain nodules and human subependymal giant cell astrocytomas. Dr Crino suggested that giant cells, tubers, and subependymal nodules derive from the same population of progenitor cells.
Mouse Models
David H. Gutmann, MD, PhD, Department of Neurology, Washington University School of Medicine, St. Louis, MO
Dr Gutmann began by describing the advantages and disadvantages of using animal models for the study of tuberous sclerosis complex. Animal models are quite instructive for understanding the molecular and cellular pathogenesis of disease, identifying targets for therapeutic drug design, preclinical evaluation of potential therapies, and developing more refined methods for following human disease. He described four methods for creating mouse models. First, human tissue can be implanted in mice. Because this method allows for the use of human tissue in a natural genetic or cellular situation, it can more accurately predict human responses to therapies. However, this approach is not useful for studying nontumor tissue. The second option is to use naturally occurring animal models (eg, the Eker rat; Tsc2 gene mutation). The benefit of the naturally occurring rat model is the presence of a natural genetic or cellular situation that is potentially manipulable. However, the condition in a rat might differ from the human condition and might not recapitulate all aspects of human disease, including exhibiting a different response to therapeutic agents. A third option is to ablate the disease gene to generate mice that are genetically similar to humans. However, this approach has similar drawbacks to those of the natural rat model, including the problem that loss of TSC gene expression results in embryonic lethality. Finally, tissue-specific inactivation of the disease gene in question can be employed. The benefits are similar to those of gene ablation, but this method potentially eliminates the problem of embryonic lethality. In this method, only one tissue type is targeted.
Dr Gutmann described research from his laboratory in which an enzyme that mediates DNA recombination and excision was specifically expressed in astrocytes to develop a mouse that lacks Tsc1 expression in the brain. The researchers studied astrocytes from these Tsc1-deficient mice to learn more about the molecular pathogenesis of tuberous sclerosis complex. In their studies, Dr Gutmann and colleagues found evidence of abnormal migration and differentiation of neurons. In addition, loss of Tsc1 expression resulted in increased astrocyte proliferation.
The Tsc1 conditional knockout mice exhibited abnormal behavior, beginning around 1 month of age. By 2 months of age, rhythmic jerking of the front paws was noted, suggesting the development of seizures. Mouse EEG confirmed that this rhythmic jerking represented seizures. Although it was not immediately obvious how Tsc1 loss in astrocytes might result in neuronal hyperexcitability, recent studies by Dr Gutmann and Dr Michael Wong have suggested that Tsc1-deficient astrocytes have reduced glutamate transport. These findings might be useful in the design of future therapies aimed at treating epilepsy in children with tuberous sclerosis complex.
Question and Answer Session 3
Dr Gutmann: Let me start out with an incredibly unfair question. If the model is that the TSC1 and TSC2 genes are interacting to form a functional complex, how would you suppose that patients with TSC1 or TSC2 would present with a severe phenotype, or does it matter?
Dr Northrup: Some of the things that I've been reading suggest they probably have some functions that are more related to one of the proteins versus the other, that are more related to hamartin and more related to tuberin. I would postulate that a tuberin mutation affecting the TSC2 gene may be involved in other functions that are more critical and more likely to cause more severe phenotype.
Dr Bullmore: Does anybody have any idea why there should be a high mutation rate in these two genes? What are the data like on the variation of these genes in the normal population?
Dr Northrup: I'm going to answer your second question first. These genes are extremely variable, and that's been a really tricky thing for us in trying to do mutational analysis. There are polymorphic variants all over the place, even within exons, so it's really difficult. When we see what we may end up classifying as a missense mutation in a child, we then have to trace back to parents, look at our whole data set, has it ever been described. Unless it's a protein mutation, it's up in the air as to whether we feel comfortable calling it a missense mutation or just calling it a variation. In terms of your second question, I don't think we really understand why there is a high mutation rate.
Dr Gutmann: We do not know why that particular region is a hot spot.
Dr Bullmore: Have you found any mutations in both these genes, and do any mutations in the binding regions have any different phenotype that relate to the combination of these two proteins? Do specific mutations relate to the phenotype?
Dr Northrup: No mutations have been concurrently found in both genes. Because of the large number of mutations, phenotype-genotype correlations have not been studied.
Audience Member: Are your cells missing both copies of the PS2 gene?
Dr Henske: The question is whether in the sporadic lymphangioleiomyomatosis, are there patients with both copies of the PS2 gene, and yes, there are, which actually alludes to the one question in the explanation of one hit or two hits.
Audience Member: You mentioned TSC2 but not TSC1 with the angiomyolipoma. Is there anything about that?
Dr Henske: I just used TSC2 as an example. TSC1 also causes angiomyolipoma, although we have found very few TSC1 mutations in the sporadic lymphangioleiomyomatosis in the population.
Audience Member: Why don't men with angiomyolipoma get lung disease?
Dr Henske: The question of my career is why don't men get lymphangioleiomyomatosis. We would love to have the answer. Men get angiomyolipoma, but they virtually never get pulmonary manifestations. That must be telling us something about hormone regulation and the TSC genes.
Audience Member: My question is, you have the giant cells in the tubers that clearly have two hits, and you have dysplastic cells in the tuber that have one hit, and you have the normal cortical neurons that have one hit. What differentiates dysplastic cells and cortical neurons? Why aren't cortical neurons dysplastic?
Dr Crino: That's a fantastic question. What it suggests is at the molecular genetic level, the molecular genetic phenotype of the cell is supposedly the normal cortex, is exactly identical to a cell within an aberrant morphology within a tuber. What that suggests is that potentially the reason dysplastic neurons form does not have to do with the cell itself. It may not be a downstream cause. But it may be a contextual event because you have these large giant cells forming early on in the development. Maybe the lineage of giant cells has to do with neuronal migration. You're quite right that, in fact, the gene type of dysplastic neurons seems to be identical to a completely normal and unaffected cortex.
Audience Member: My question is, do you think the second-hit event relates somehow to the calcifications?
Dr Crino: We have some evidence in the lab to suggest there may be, in keeping with migration of cells into tubers, enhanced cell death within tubers that could theoretically account for calcifications from an exhausted blood supply. However, there is no final answer to the question of why tubers
