Skip secondary menu

Research Plan For Tuberous Sclerosis


Research Plan For Tuberous Sclerosis

July 2003


Executive Summary

In a December 12, 2001, Concurrent Resolution (S.Con.Res.69, H.Con.Res.25), Congress stated its support for the improved detection and treatment of tuberous sclerosis and expressed the sense of Congress that the National Institutes of Health (NIH) should take a leadership role and provide to the Congress a five-year research plan for tuberous sclerosis. This report transmits the research plan for tuberous sclerosis in accordance with the Congressional Resolution.

To begin the planning process, the National Institute of Neurological Disorders and Stroke (NINDS), the Tuberous Sclerosis Alliance, and the NIH Office of Rare Diseases (ORD) organized a major international symposium on tuberous sclerosis complex (TSC). Held in Chantilly, Virginia, in September 2002, the conference brought together researchers, clinicians, and patient advocates to assess the state of tuberous sclerosis research, define the key challenges, and recommend strategies to address them. The NINDS organized the recommendations into a strategic plan, with additional input from the conference participants and representatives from the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK); National Institute of Child Health and Human Development (NICHD); National Heart, Lung, and Blood Institute (NHLBI); National Institute of Mental Health (NIMH); National Institute of Arthritis and Musculoskeletal and Skin Diseases (NIAMS); National Cancer Institute (NCI); and ORD.

Although the Congressional Resolution was for a five-year research plan, the needs of the TSC patient community and the challenges inherent in the science, which were reinforced by the proceedings of the symposium, mandate a broader and longer-range, forward-thinking vision. The plan is thus organized around long-term (that is, greater than 10-year) broad goals, with the expectation that significant progress toward the more specific objectives can be achieved within a five-year time frame.

The broad goals are:

  1. To determine the molecular and cellular basis of TSC;
  2. To understand and treat the symptoms of TSC;
  3. To understand the expression of TSC symptoms across the life span, and to identify factors that affect this expression;
  4. To develop resources that accelerate TSC research; and
  5. To create new research opportunities.

The NIH and the broader TSC research community will use these goals as a guide for developing focused research projects, programs, and infrastructure. The identified goals and objectives will also provide a yardstick for measuring progress in TSC research. At the end of Fiscal Year 2004, and annually thereafter, the NIH will evaluate scientific advances in relation to the goals of the plan and identify needed modifications or activities, such as workshops or solicitations, to ensure continued progress against tuberous sclerosis.


The Congressional Concurrent Resolution passed on December 12, 2001 (S.Con.Res.69, H.Con.Res.25), states, in part:

"That it is the sense of Congress that -

  1. all Americans should take an active role in the fight against tuberous sclerosis by all means available to them, including early and complete clinical testing and investigating family histories;
  2. the role played by national and community organizations and health care providers in promoting awareness of the importance of early diagnosis, testing, and ongoing screening should be recognized and applauded;
  3. the Federal Government has a responsibility to-
    1. endeavor to raise awareness about the importance of the early detection of, and proper treatment for, tuberous sclerosis;
    2. increase funding for research so that the causes of, and improved treatment for, tuberous sclerosis may be discovered; and
    3. continue to consider ways to improve access to, and the quality of, health care services for detecting and treating tuberous sclerosis; and
  4. the Director of the National Institutes of Health should take a leadership role in the fight against tuberous sclerosis by acting with appropriate offices within the National Institutes of Health to provide to the Congress a five-year research plan for tuberous sclerosis."



Planning Process

In December 2001, Congress passed a Joint Resolution (S.Con.Res.69, H.Con.Res.25) mandating that NIH develop a five-year research plan for tuberous sclerosis complex (TSC). To begin the planning process, the National Institute of Neurological Disorders and Stroke (NINDS), the Tuberous Sclerosis Alliance, and the NIH Office of Rare Diseases (ORD) organized a major international TSC symposium. Held in Chantilly, Virginia, in September 2002, the conference brought together researchers, clinicians, and patient advocates to review the state of the field, define the key challenges in tuberous sclerosis research, and develop strategies to address these challenges. The summary of the meeting is provided in the Appendix. The NINDS organized the recommendations from the meeting into a strategic plan, with input from the conference participants and representatives from the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK); the National Institute of Child Health and Human Development (NICHD); the National Heart, Lung, and Blood Institute (NHLBI); the National Institute of Mental Health (NIMH); the National Institute of Arthritis and Musculoskeletal and Skin Diseases (NIAMS); the National Cancer Institute (NCI); and the ORD.

Description of Tuberous Sclerosis

Tuberous sclerosis complex (TSC) is a genetic disorder associated with mutations in two different genes-TSC1 and TSC2-which cause benign tumors called hamartomas. These tumors can form in many different organs, primarily the brain, eyes, heart, kidney, skin, and lungs. In most individuals affected by TSC only a subset of these organs is affected. Epileptic seizures (which often begin as infantile spasms) and learning and behavioral problems (including autism) are also common in individuals with TSC. There is no cure for TSC; treatment is symptomatic and may include anticonvulsant therapy for seizures, drug therapy for neurobehavioral problems, treatment of high blood pressure caused by kidney dysfunction, and surgery to remove growing tumors. The prognosis for individuals afflicted with TSC varies in accordance with the severity of the specific symptoms.

Since testing methods that permit the identification of less severe manifestations of TSC have been developed, estimates of the frequency of this disorder have risen sharply. Population-based studies indicate a prevalence of 8-9 per 100,000 individuals (O'Callaghan et al., 1998). The Tuberous Sclerosis Alliance estimates that 50,000 Americans and 1 million individuals worldwide have TSC and that the incidence of TSC is approximately 1 in 6,000 live births.

NIH Overview of Tuberous Sclerosis Research

The NIH supports a broad range of studies on TSC, from basic science to clinical research. Basic research studies are under way to investigate the molecular and cellular mechanisms by which mutations in the TSC genes lead to neurological, kidney, and lung abnormalities, as well as the relationship between tuber formation and seizure development. NIH-supported scientists have identified a crucial step in the regulation of cell growth by the TSC2 gene. Investigators are also developing various animal models of tuberous sclerosis to investigate its genetic basis and to evaluate gene delivery as a potential therapeutic intervention. The NIH supports research to develop assays, or tests, to characterize mutations occurring in either of the two genes responsible for TSC; such tests would enable a comprehensive analysis of the molecular basis of TSC and would aid physicians and family members in confirming diagnosis and providing information on prognosis. Investigators are conducting clinical studies to test the hypothesis that abnormalities in tryptophan metabolism contribute to the development of both epilepsy and autism in children with tuberous sclerosis; to examine what causes skin tumors to develop in patients with TSC; and to understand the molecular and cellular basis for the development of lymphangioleiomyomatosis-LAM-a severe destructive lung disease, in patients with tuberous sclerosis complex.

The Plan for Future Research

The goals of the plan were developed to address the needs of the TSC patient community and the state of the science. The goals listed in headings 'A' through 'E' represent the broad, long-range vision for TSC research, which will require more than five years, indeed likely more than 10 years, to achieve in full. However, we expect that significant progress toward the more specific (bulleted) objectives can be made within a five-year time frame.

A. Determine the Molecular and Cellular Basis of TSC

TSC is inherited in an autosomal dominant pattern in that the loss of a single copy of a gene leads to the disease. TSC has been associated with mutations in two different genes-TSC1 and TSC2. The products of these genes have been identified: the TSC1 gene encodes a novel protein called hamartin and the TSC-2 gene product is a protein called tuberin. Hamartin and tuberin bind to each other and are likely to function as tumor suppressors; that is, they play critical roles in regulating cell growth and proliferation. Recent evidence suggests that hamartin and tuberin act in the growth factor (insulin) signaling pathway.

Understanding how genetic mutations ultimately lead to cellular abnormalities will be essential for developing effective therapies for TSC. Drug discovery requires a molecular target whose function can be modified; this target can be the mutated protein itself, an interacting protein, or a protein that acts downstream. To identify an appropriate potential drug target for TSC, it will be necessary to identify the complex molecular pathways underlying the disease, characterize the pathway components, and elucidate how the pathways regulate cellular behavior.

Specific objectives:

  • Determine whether tubers and other TSC-associated overgrowths develop through a "two-hit" mutational model in which spontaneous loss of the second copy of the gene in certain cells causes them to proliferate into tumors.
  • Identify hamartin/tuberin-binding proteins and additional downstream components of the TSC1/TSC2 signaling pathway.
  • Determine how the TSC1/TSC2 pathway interacts with other signaling cascades (e.g., the Wnt, mTOR, and P13K-Akt pathways).
  • Determine the role of TSC1/TSC2-mediated signaling in cell cycle regulation, cell proliferation, cell adhesion, and cell motility.


B. Understand and Treat the Symptoms of TSC

1. Dermatological Lesions

Dermatologic hamartomas in individuals with TSC can lead to bleeding, chronic irritation, and cosmetic disfigurement. Additional functional problems can arise, including tooth maldevelopment, airway obstruction, or visual occlusion. Current treatments for dermatological hamartomas are surgical, generally involving laser treatment and abrasion. Insight into the mechanisms by which these tumors form may aid in the development of nonsurgical, drug-based interventions.

Specific objectives:

  • Develop and characterize cell lines from skin tumors in TSC patients.
  • Develop and characterize animal models of TSC that exhibit skin lesions.
  • Determine the cell of origin of skin tumors and the role of cell-autonomous and non-autonomous factors in their progression.
  • Investigate how loss of TSC1/TSC2 function causes angiogenesis, alterations in extracellular matrix, and changes in skin pigmentation.
  • Identify growth factors that stimulate angiogenesis and fibrogenesis in TSC skin tumors.
  • Investigate nonsurgical treatments for skin lesions, including topical medications.

2. Renal, Pulmonary, Lymphatic, and Cardiac Pathology

TSC can cause lesions in the kidneys and lungs, as well as cardiac problems. In the kidneys, these lesions are called renal angiomyolipomas; in the lungs, they are referred to as cystic lesions and nodules. Lymphatic abnormalities are also common. Abnormal growths and cystic destruction of lung tissue are characteristic of lymphangioleiomyomatosis (LAM), a rare lung disease that occurs as an isolated disorder (sporadic LAM) and in women with tuberous sclerosis. Abnormalities in the genes and proteins that cause TSC also cause LAM.

Specific objectives:

  • Determine the cell type of origin of renal angiomyolipomas and pulmonary cystic lesions/nodules.
  • Establish cell culture models for renal and pulmonary TSC, including LAM.
  • Understand how TSC leads to the abnormal cellular growth, lymphatic abnormalities, and cystic destruction of lung tissue characteristic of LAM.
  • Determine the origin of LAM cells and the role of cell motility in the development of LAM.
  • Determine the role of vascular endothelial growth factor (VEGF) receptors and other angiogenic factors in LAM.
  • Initiate clinical trials to test the potential of rapamycin for treating renal angiomyolipomas and pulmonary disease.
  • Investigate causes and possible treatments for TSC-associated cardiac arrhythmias and congenital cardiac abnormalities.

3. Central Nervous System Pathology and Epilepsy

Epilepsy is the most common neurological symptom among individuals with TSC, and the management of epilepsy accounts for a large percentage of the care that they require. Currently available antiepileptic drugs are usually ineffective, particularly in the case of infantile spasms.

The molecular and cellular bases for TSC-associated epilepsy and seizure intractability are not well understood. Hamartomas in the brain resemble tubers (hence the name "tuberous sclerosis") and are cortical abnormalities characterized by disorganized cells, cells with aberrant structure (morphology), and an increased number of astrocytes. It has been hypothesized that tubers cause seizures. Seizure intractability may be partially explained by the expression of multidrug resistance genes, which has been documented in TSC cells. When expressed in brain capillary endothelial cells, multidrug resistance gene products may lower the concentration of antiepileptic drugs so that they are below therapeutic levels in the brain even if systemic drug levels are high.

Specific objectives:

  • Understand the functions of wild-type tuberin and hamartin in specific cell and tissue types in the central nervous system (CNS) (e.g., neurons, astrocytes).
  • Determine how the loss of tuberin/hamartin function alters cell proliferation and affects the development of specific brain regions (e.g., hippocampus).
  • Analyze the cellular composition, development, and electrophysiological properties of tubers; develop better methods for tuber collection.
  • Determine by neuroimaging whether tubers, or specific tuber subtypes, are associated with seizures in TSC patients.
  • Identify TSC patients at risk for epilepsy and predict whether the epilepsy will be intractable or treatable.
  • Investigate the effects of antiepileptic drugs in model systems.
  • Investigate the role of Multidrug Resistance type 1 (MDRI1) gene expression in CNS pathology.
  • Develop better treatment strategies for TSC-associated epilepsy, particularly for infantile spasms.
  • Evaluate the effectiveness of surgery for seizure prevention.

4. Cognitive and Behavioral Problems

Cognitive and behavioral dysfunction affect more than half of all individuals with TSC, at all ages. Significant developmental delays, some associated with autism spectrum disorders, are observed in about half of all children with TSC. The causes of these cognitive and behavioral problems are not known. One hypothesis is that tubers are responsible for the cognitive deficits. TSC individuals have been found to have deficits in attention, retrieval, and integration, which are attributed to the regions of the brain that most commonly harbor tubers. It has also been proposed that epileptic seizures in the temporal lobes, perhaps related to the presence of tubers, disrupt the development of key social representations during an early critical period.

Specific objectives:

  • Determine whether tubers, specific tuber subtypes, or other TSC brain abnormalities cause behavioral and cognitive problems.
  • Develop improved neuroimaging techniques for detecting TSC-associated brain malformations.
  • Determine whether the presence of seizures or seizure frequency is predictive of specific TSC cognitive and behavioral symptoms.
  • Develop and test models of the pathogenesis of TSC-associated autism.
  • Formulate improved clinical guidelines for neuropsychological evaluation; develop detailed strategies for timing assessments and choosing appropriate interventions.
  • Develop neuropsychological interventions for cognitive and behavioral problems.


C. Understand the Expression of TSC Symptoms (phenotype) Across the Life Span and Identify Factors that Affect this Expression

Understanding the extent and basis of variability in the TSC phenotype, over the lifetime of an individual and among TSC patients, will be critical for developing therapies and new diagnostic approaches. In conducting clinical trials, it is necessary to know whether an observed change in condition is due to the effects of therapy or to the natural progression of the disease. Identification of correlations between symptoms and particular mutations or modifier genes may aid in diagnosing TSC and choosing the most appropriate treatment regimen.

Specific objectives:

  • Conduct multi-center, large-scale natural history studies of the progression of various TSC symptoms.
  • Perform focused natural history studies to analyze symptoms that may serve as potential outcome measures in clinical trials.
  • Conduct genotype-phenotype studies to determine whether there is a correlation between particular mutations and symptoms.
  • Identify modifier genes that influence the variable expression of TSC.


D. Develop Resources that Facilitate and Accelerate TSC Research

1. Basic Research

Model systems are essential for understanding the mechanisms that underlie human disorders. Rodent models for TSC currently exist, but they do not reproduce the symptoms of the human disorder.

Specific objectives:

  • Develop mouse models that permit the spatial and temporal regulation of TSC1 and TSC2 expression.
  • Develop mouse models that recapitulate the CNS abnormalities in TSC patients (e.g., epilepsy, tubers, and behavioral/cognitive deficits).
  • Facilitate the storage and widespread dissemination of existing cell and animal models.

2. Translational Research

Before therapeutics can be tested in humans, they must be extensively tested in models. To accelerate the process of preclinical therapy development, it will be important to develop specific resources and make them available to the TSC research community.

Specific objectives:

  • Develop cell culture, invertebrate, and vertebrate animal models suitable for testing potential therapeutics.
  • Test identified candidate therapeutics (e.g., rapamycin, anti-PDGFD, interferon gamma, anti-VEGF) in model systems.
  • Perform high-throughput screening of compound libraries in biochemical and cell culture systems.
  • Develop biomarkers and imaging techniques that permit longitudinal animal studies and reduce reliance on tissue biopsies and autopsies.

3. Clinical Research

Studies involving patients with TSC or tissues from these individuals also require specific tools and research resources. Currently, these resources are either not available or are available only in individual laboratories. In addition to developing such resources, additional steps must be taken to facilitate effective natural history studies and clinical trials.

Specific objectives:

  • Establish a repository of tissue from TSC patients that includes brain, lung, skin, and kidney samples (both primary tissue and immortalized cell lines).
  • Explore the utilization of existing research centers to facilitate TSC research (e.g., the NICHD Mental Retardation and Developmental Disabilities Research Centers).
  • Develop a general clinical database of TSC patient information (including genotype information).
  • Develop focused subcomponents within the clinical database (e.g., establish a database dedicated to epilepsy that includes data on genotype, seizure subtypes, tuber subtypes, history and efficacy of antiepileptic drug use).
  • Develop a database component to track the effects of drugs on TSC symptoms in TSC patients enrolled in trials for other disorders.
  • Develop/identify a multicenter clinical network to be used in natural history studies and clinical trials.
  • Develop outcome measures for clinical trials.
  • Identify the most promising therapeutic agents (e.g., rapamycin, gleevec) and develop infrastructure for mechanism-based clinical trials.
  • Perform pilot clinical trials that lay the groundwork for large-scale Phase III efficacy trials.

E. Create New Research Opportunities

Specific objectives:

  • Attract new investigators to TSC research and provide new training opportunities for junior investigators.
  • Develop workshops and satellite conferences associated with larger meetings to increase awareness of TSC in the scientific community.
  • Establish mechanism for regular therapy development conferences.
  • Develop partnerships between industry, academia, and voluntary organizations to develop therapies for TSC.
  • Encourage international collaborations in TSC research.



The identification of the genes that cause TSC provides a remarkable opportunity for progress in TSC research, but the multi-system complexity of the disease and the wide variations in disease manifestation (phenotype) presents enormous challenges. The NIH, working through the NINDS, NIDDK, NICHD, NHLBI, NIMH, NIAMS, NCI, and ORD, believes that a multi-pronged approach, incorporating basic, translational, and clinical research, will permit the TSC community to seize this opportunity and address these challenges.

Challenged by the complexities of tuberous sclerosis and the scope of this long-term vision for TSC research, we anticipate that significant progress will be made towards achieving elements of this plan within the next five years. At the end of fiscal year 2004, and annually thereafter, we will evaluate scientific advances in relation to the goals of the plan and identify modifications or activities, such as workshops or solicitations, to ensure continued progress against TSC. As our understanding of the causes of TSC increases, we believe that new therapeutic interventions will be developed to reduce the burden of this devastating disorder.


Conference Summary: New Perspectives in Tuberous Sclerosis Research
September 19-22, 2002
Chantilly, Virginia


Tuberous sclerosis complex (TSC) is a multi-system disease characterized by abnormal growths (hamartomas). These growths are distributed at multiple sites throughout the body. The severity and symptomatic expression of TSC is highly variable, making the disease a challenging one to diagnose. While severe manifestations may be seen in individuals diagnosed in childhood, mild forms of the disease may be observed in women and men diagnosed in adulthood. TSC is a genetic disorder but more than half the individuals with TSC have a spontaneous rather than an inherited mutation. The major organ systems involved include brain, kidneys, lungs, heart, eyes, and skin. Central nervous system involvement is the most common feature. The majority of patients first come to the clinic for treatment of seizures, and TSC is one of the most common neurogenetic syndromes associated with epilepsy and autism.

Tuberous sclerosis shows an autosomal dominant pattern of inheritance. Two genes have been identified within the last 11 years (TSC1 in 1997 and TSC2 in 1992). The TSC1 gene is located on chromosome 9 and encodes a 130 kD protein called hamartin. The TSC2 gene, located on chromosome 16, is contiguous with the gene for polycystic kidney disease. The TSC-2 gene product, a protein called tuberin, is a GTPase activating leucine zipper protein. Hamartin and tuberin bind to each other and are likely to function as tumor suppressors.

Research in tuberous sclerosis is as broad as the disease itself, addressing the minutiae of molecular genetics to the behavioral studies of cognition and from clinical observations in the operating room to drug discovery and development of new therapies. This meeting was unusual in that it bridged many disciplines and brought together a diverse group of experts. Critical issues in all areas of tuberous sclerosis were presented and discussed. Participants were asked to reflect on future directions and to develop a template for further research initiatives.

Highlights of Recent Research


Epilepsy is the most common neurological symptom among individuals with TSC, and the management of epilepsy accounts for a large percentage of the care that they require. Approximately 90 percent of patients have seizures, and an astounding 85 percent of seizures begin in the first year of life. Infantile spasms occur at a very high rate and are associated with especially poor prognosis.

Hamartomas in the brain resemble tubers, hence the name tuberous sclerosis. Cortical lamination is disrupted within cortical tubers. Abnormal cells found in tubers include dysplastic neurons, large astrocytes, and giant cells displaying markers of neuronal, glial, and immature phenotypes. The number of tubers and their location varies, although there is a preferential accumulation within frontal and parietal cortex. The abnormal placement of neurons and the disorganized lamination that are typical of tubers suggest that the underlying developmental defect may be in neuronal migration. Neuronal maturation may also be incomplete as about 70 percent of neurons in tubers have either fewer or thinner spines than normal.

Epilepsy in the majority of individuals with TSC is intractable. Currently available antiepileptic drugs are usually ineffective, particularly in the case of infantile spasms. Surgery thus becomes an important option. Michael Duchowny (Miami Children's Hospital) discussed the fact that not all tubers are the same, and some are more epileptogenic than others. Using depth electrodes, Duchowny and collaborators have shown that the core of a tuber can be epileptogenic. Propagation of electrical activity spreads to surrounding tissue and, due to aberrant connections, to other parts of the brain. If a particular tuber is shown to be an epileptic focus, surgical removal of that tuber can provide relief from seizures even while other tubers remain. The reasons underlying epileptogenesis in tubers in particular, and in epilepsy in general, are unclear. Intrinsic hyperexcitability of the neurons is likely to be a contributing factor. PET scans have shown that uptake of alphamethyltryptophan is higher in epileptogenic tubers than in nonepileptogenic tubers. This suggests that alterations in serotonin receptor activity may be involved. Seizure intractability may be partially explained by expression of multidrug resistance genes, which have been documented in TSC cells. When expressed in brain capillary endothelial cells, multidrug resistance gene products may lower the concentration of antiepileptic drugs so that they are below therapeutic levels in the brain even if systemic drug levels are high.

Cognition and Behavior

Behavioral and cognitive dysfunction affects more than half of all individuals with TSC. Significant developmental delays, with or without autism spectrum disorders, are seen in about half of all children with TSC. In fact, autism was first diagnosed in a child with TSC.

Given the high prevalence of autism spectrum disorders in individuals with TSC (up to 60 percent in children with TSC compared to 0.15 percent in the general population worldwide), tuberous sclerosis may be a useful model system with which to investigate autistic spectrum disorders. One theory, proposed by Patrick Bolton (Cambridge, United Kingdom) is that epileptic seizures in the temporal lobes, perhaps related to the presence of tubers, disrupt the development of key social representations during an early critical period. A simple, very early sign of these disruptions is the inability to tell where someone else's eyes are directed. These early changes are thought to lead to later impairments in the ability to infer the intent and emotions of other people.

There is significant risk for specific learning disabilities, even in the 50 percent of TSC patients whose development is within normal limits. Neuropsychological testing of patients 3 months to 45 years old at the TSC Comprehensive Clinic at the MGH (Boston, Massachusetts) has shown that consistent deficits in attention, retrieval, and integration can be identified in individuals who otherwise tested in the overall normal cognitive range. This is consistent with the distribution of tubers in TSC, which are most commonly found in frontal, parietal, and subcortical regions of the brain. The development of children with TSC should be followed closely. Specifically, Dr. Prather used her recent findings to recommend that all children should receive a cognitive assessment at ages 5-6 (before entering school) and at age 9 (at the end of the primary elementary grades). As with all children with attentional difficulties, children with TSC are likely to benefit from more structure and repetition in the early elementary grades to reinforce basic skills, followed with help on organization and planning in the upper primary grades and accommodations and extra support through high school.

Renal Development

TSC is also associated with lesions in the kidneys and lungs. Hamartomas in these organs consist of smooth muscle cells, blood vessels, and, in kidneys, fat tissue. Renal disease is a significant source of mortality in TSC, with the most life-threatening risk due to the rupture of vessels within the hamartomas.

Work in the Herzlinger laboratory (New York University, New York, New York) has shown that renal angiolipomas are not the result of dedifferentiation of mature renal epithelia to an embryonic pluripotent cell. This is supported by the fact that hamartin and tuberin, the protein products of the TSC1 and TSC2 genes, are thought to be involved more in regulating cell growth than cell differentiation. Instead, it may be stromal cells that are the culprit cell type responsible for generating renal angiolipomas. Stromal cells differentiate into fibroblasts and vascular smooth muscle. They express TSC1 and are extremely sensitive to deregulated growth upon loss of TSC function. In addition, they secrete paracrine factors (notably FGF-7) that serve to pattern the size and number of renal tubules. Overgrowth of stromal cells when TSC1 is lost could lead directly to angiolipomas due to excess fibroblasts and smooth muscle cells, as well as oversecretion of paracrine factors that could result in the formation of renal cysts.

Pulmonary Function

Pulmonary involvement in TSC consists primarily of cystic lesions and nodules in the lungs. These occur independently of each other; cysts are almost exclusively found in women but nodules occur in both men and women. Abnormal growths and cystic destruction of lung tissue is characteristic of lymphangioleiomyomatosis (LAM ), a rare lung disease that occurs as an isolated disorder (sporadic LAM) and in women with TSC.

In a large scale screen of women with TSC, Joel Moss (NIH, Bethesda, Maryland) found that lung function was preserved even if cystic lesions were present, suggesting that a large population of individuals with TSC are asymptomatic or have very mild lung disease. The prevalence of LAM in women with TSC is estimated to be about 34 percent, although severe respiratory disease occurs in less than one percent of women with TSC. In screening women with LAM for TSC, brain scans revealed meningiomas in 10 women out of 250 who were screened; 3 of these women met the diagnostic criteria for TSC. Eventual respiratory failure due to destruction of lung tissue may require lung transplantation, which remains the most viable option for end-stage disease. As predictors of time to transplant or death, histological assessment of disease (LAM histology score), DLCO (in the case of mild disease only), the cardiopulmonary exercise test, and high resolution CT scans are all useful.

LAM nodules in the lungs and AML (angiomyolipomas) in the kidney both contain abnormal smooth muscle cells that express smooth muscle and melanoma antigens. While similar to each other, cells in LAM and AML nodules are not like other smooth muscle or melanoma cells. Nodules have a characteristic composition, although quite a bit of variability is observed. The center of the LAM nodule contains proliferating cells, these are surrounded by epithelia-like cells and, these in turn may be surrounded by cells that are hyperplastic pneumocytes.

LAM nodules, like other growths in TSC, are not malignant. However, they do possess some metastatic potential. Work from the Henske laboratory (Fox Chase Cancer Center, Philadelphia, Pennsylvania) demonstrates that recurrence of LAM in a healthy transplanted lung is due to metastasis of cells from the host (n=2).

Dermatologic Involvement

Dermatologic hamartomas in individuals with TSC are harmless but nonetheless disfiguring, particularly if they are on the face. Skin tumors are frequently numerous and include facial angiofibromas, collagenomas (Shagreen patches), and periungal fibromas. They are extremely accessible to study, in contrast to the parallel internal tumors. In touch preparations of TSC skin lesions, Tom Darling (Uniformed Services University of the Health Sciences, Bethesda , Maryland) reported that allelic deletion of the TSC2 gene was observed in nearly every sample. The tumors were a heterogenous mixture of cells with 18 percent to 50 percent showing allelic deletion of TSC2. These results are consistent with the two-hit theory, where individuals with a germ line defect in one allele lose the second allele in a tumor. Deletion of the TSC1 gene was not observed in any sample. Cells from TSC skin lesions can be cultured and have been propagated for up to 5 passages. These cells express higher levels of cytokines that may be one way that the neoplastic cells influence the growth of surrounding normal cells.

Most skin lesions on the face are angiofibromas that are either flat and red or raised and unpigmented or red. The presence of facial angiofibromas can lead to bleeding, chronic irritation, and cosmetic disfigurement. Additional functional problems can arise including tooth maldevelopment, airway obstruction, or visual occlusion. Mark Mausner (Mausner Plastic Surgery Center, Rockville, Maryland) described the results of treatment with a variety of specialty lasers that greatly improved appearance and alleviated functional problems. Many of the treated facial angiofibromas did not recur, and when they did, regrowth was extremely slow. Thus, lesions should be treated early since it was noted in all individuals with TSC that facial angiofibromas increase rapidly in size and number during puberty.

Animal Models of TSC

The Eker rat, first described in the 1960s by Eker and Mossige, spontaneously develops tumors in the kidney, uterus, and spleen. It is now known that these tumors are due to loss of function of the tsc2 gene. The Eker rat, and cell lines established from its tumors, are useful tools in understanding tuberin function and the renal manifestations of tuberous sclerosis. Interestingly, the Eker rat (bred on a Long-Evans background) rarely develops cortical tubers. To develop an Eker rat model for TSC seizures, various "second-hit" approaches have been tried. So far, work in Scott Baraban's laboratory (University of California, San Francisco, California) exposing Eker rats prenatally to hydroquinone has not resulted in tuber formation. A different "second-hit" was used by Philip Schwartzkroin (University of California, Davis, California) who gave young postnatal Eker pups a dose of radiation. The brains of these rats had a significantly lower seizure threshold and contained both dysmorphic neurons and large cells that were GFAP positive.

Several lines of transgenic mice have been developed. Jack Arbiser's laboratory (Emory University, Atlanta, Georgia) generated a transgenic mouse line that expresses a dominant negative allele of tuberin behind a constitutive cytomegalovirus promoter. The mice are viable and fertile. They develop neoplastic nodules in the kidney, as well as skin lesions that are reminiscent of collagenomas (Shagreen patch).

A tsc1 conditional knockout, under the control of the glial specific GFAP promoter, results in mice with abnormally high numbers of astrocytes (up to 6 times more than normal). David Gutmann (Washington University, St. Louis, Missouri) described how the astrocytes grow abnormally in vitro, having lost contact inhibition of growth. Interestingly, disorganized neuronal layers are evident in the hippocampus. Most interesting, the tsc1 conditional knockout mice develop seizures after about 1 month of age and start to die after 3-4 months. The seizures occur with a frequency of 0.6/hour. Kevin Ess (Washington University, St. Louis, Missouri) described the seizures in more detail, noting that most seizures are generalized, although focal onset in some seizures could be detected by hippocampal depth electrodes. Interrictal EEG and behavior were severely abnormal. Electrophysiology performed in the hippocampus of mutant mice showed that neither paired pulse facilitation or presynaptic facilitation were detectable, indicating that short- term plasticity in these mice is impaired.

David Kwiatkowski's laboratory (Brigham and Women's Hospital, Boston, Massachusetts) has generated 2 other conditional knockouts of the tsc1 gene controlled either by the synapsin promoter (expressed in all neurons) or by the nestin promoter (expressed in neuroepithelial cells). The tsc1/synapsin mice die by 6 weeks of age. Their brains have mildly enlarged cells, no astrogliosis is present, and some disorganization in the hippocampus is observed. Seizures can be induced by physical manipulation (tail spinning). The tsc1/nestin mice develop spontaneous seizures and die by 4 months of age. Their brains show enlarged neuron-like cells and disorganized laminae in the cortex and hippocampus. Cells have also been cultured from these mouse embryos and are proving to be useful in dissecting tsc1 and tsc2 intracellular signaling pathways.

Molecular Biology and Biochemical Pathways

Tuberin (TSC2) is a GTPase activating protein, with homology to Rap1GAP. The kinase AKT, which is stimulated by growth factors, phosphorylates tuberin on 2 sites. Hamartin (TSC1) is not phosphorylated by AKT. Hamartin, thought to interact with the cytoskeletal proteins ezrin-radixin-moesin, binds to the light chain of neurofilament and activates the small GTPase Rho. Tuberin and hamartin form a complex, and the presence of both proteins seems to be required in most of the pathways described below.

Hamartin and tuberin have been shown to be part of the growth factor (insulin) signaling pathway. This pathway is well conserved from flies to humans, and many of its molecular details are well understood. A pivotal component of the cascade, mTOR (mammalian Target Of Rapamycin), was identified because the immunosuppressant drug rapamycin inhibits its activity. The placement of tuberin and hamartin in this pathway upstream of mTOR is of intense interest due to the therapeutic possibilities. While many of the details concerning the role of hamartin and tuberin in this pathway have been elucidated, particularly for the Drosophila homologs of tsc1 and tsc2, many questions remain, particularly concerning exactly how directly hamartin and tuberin interact with mTOR.

Aside from the growth factor/mTOR pathway, hamartin and tuberin affect cell cycle control. Cyclin D levels increase when tuberin levels decrease. Cyclin D is regulated by Wnt/_-catenin, and Baldwin Mak (University of Washington, Seattle, Washington) presented evidence that hamartin/tuberin act at the level of the _-catenin degradation complex. A third pathway was described by Cheryl Walker (M.D. Anderson Cancer Center, Smithville, Texas), whose laboratory has demonstrated that tuberin leads to stabilization of the HIF2a subunit of HIF (hypoxia-induced factor). This results in elevated HIF activity, which increases the expression of VEGF (vascular endothelial growth factor), and ultimately leads to tumor angiogenesis.

Recommendations for Future Research

Recommendations for future TSC research were made with the broad acknowledgment that collaborative efforts were going to be needed and that resources and reagents should be shared and made widely available to all interested investigators. Several common areas for future work were cited by investigators in each of the systems affected by TSC.

Tissue Bank

Conference participants expressed overwhelming and unanimous support for the establishment of a bank of TSC tissue from brain, lungs, skin, and kidneys. Both primary and immortalized cells would be useful. Skin samples in particular are readily obtainable but so far have not been widely available to researchers. Examples of the many ways that the tissue bank could be used are to characterize the different cell types in a tumor, to examine molecular pathways involving hamartin and tuberin, or to identify growth factors that stimulate angiogenesis and fibromagenesis. A collaborative group effort to perform electrophysiology on fresh tuber samples from human brains was also proposed.

Logistical considerations for the establishment and maintenance of a tissue bank were discussed. Coordinating surgery on patients with sending tissue samples to researchers seemed complicated, particularly if the patients and researchers are not at the same institution. There are also regulations about the sharing of tissue and data, especially across international borders. The tissue bank should probably be established with the idea that eventually the Federal Government will assume responsibility for running it. Finally, donors like to know what happened with the tissue that was donated, so a mechanism needs to be developed where the researcher can easily provide results.

Molecular and Cellular Mechanisms

Identification of the TSC1 and TSC2 genes has spurred great advances in understanding the disease mechanisms of tuberous sclerosis, particularly in the lung and kidney. More research is unquestionably needed on the multiple putative pathways that have been identified thus far. How these signaling pathways affect cell adhesion, migration, proliferation, growth, and differentiation also need to be explored.

How hamartin and tuberin affect brain development and whether the growth factor/mTOR pathway is involved in brain pathology are questions that need to be answered. The origin of tubers needs to be understood. Understanding the electrophysiology of tubers is also a priority since this may lead to insights into epileptogenesis.

In addition to the basic science implications, understanding biological mechanisms is critical to developing new therapies. Drug discovery requires a molecular target, an enzyme or receptor whose function can be modified. The more clearly a disease process is understood at a molecular and cellular level, the more likely that key proteins are targeted for drug intervention.

Clinical Trials/Databases

Several ideas for clinical trials and clinical databases were proposed.

Clinical Trial to Test Rapamycin in TSC Patients
This proposal was for a Phase I/II pilot study to investigate the feasibility and efficacy of treatment with rapamycin in individuals with TSC. Individuals with TSC who have renal angiomyolipomas would be enrolled, inclusion criteria would include angiomyolipomas (AML) 2-5 cm in size and good performance status. LAM patients with AML 2-5 cm in size would also be eligible. The trial would aim to start with about 20 individuals in the United States and 20 patients in the United Kingdom. The study participants would receive a clinical exam and MRI upon entry into the study. Rapamycin would be administered for 8 weeks, starting with 1 mg/day (renal transplant patients currently start with 2-5 mg/day). As the study progresses, the dose could be increased if the safety profile is favorable. The primary endpoint would be AML size, and secondary endpoints would include the status of other organs that are involved, toxicity, and other biomarkers. A call was made to establish a working group to draw up a final protocol, secure funding, and develop a database in preparation for conducting the trial.

Longitudinal Study of Cognitive Outcome in Individuals with TSC
This proposal was for a multicenter, multidisciplinary, international, prospective, longitudinal study to identify the principal determinants of cognitive and behavioral outcome in TSC. Serial neurocognitive assessments would be made, along with neuroradiological and EEG assessments and behavioral assessments. Treatments and medications would be recorded. Centers in both the United States and the United Kingdom would participate using a common core set of instruments but with different and complementary emphases on attention, autism spectrum disorder, and language. A call was also made to establish a working group to determine details of the study with respect to timing of visits, choice of assessment tools, and appropriate sources of funding.

Epilepsy Database
A multicenter effort is needed to collect data on epilepsy in individuals with TSC from as many clinicians as possible. Data on genotype; seizure subtypes; the use of antiepileptic drugs and their efficacy and side effects; and tuber location as determined by MRI, SPECT, PET, surgical outcome and tissue pathology all need to be collected and systematized. Thus a picture of the natural history of the disease may emerge, along with treatments that are likely to be effective. A particular emphasis was placed on the need to focus on patients with infantile spasms, as these individuals are the youngest and their seizures among the most intractable.

Clinical TSC Database
A database to collect information on multiple aspects of TSC was also proposed. The database would help to determine the natural variance in the disease and to establish parameters for appropriate clinical endpoints that are easy to assay. The goal would be to enroll about 2,000 individuals with TSC, with a focus on adults with TSC. For proliferative and metastatic symptoms, understanding the progression and variance will help to quantify the expected placebo response. The visits would take place every 6 months to 1 year, depending on the individual's age (more frequent up to age 3). A Web-based electronic data capture system would be implemented. Questions that the database should address include the following: (1) How does tuber burden affect seizure refractoriness? (2) How does tuber burden correlate with cognitive abilities? (3) How does seizure refractoriness correlate with cognition? (4) What is the rate of progression of AML lesions? (5) What is the rate of progression of LAM lesions after the first signs of subclinical evidence on CT? And (6) which serum-based biomarkers correlate with disease status?

Attracting New Investigators

Research opportunities and training fellowships are needed to attract additional researchers and clinicians. TSC is a rare genetic disorder but should be positioned more prominently because of the possible insights that could be gained from understanding the disease mechanism(s). Informational sessions could be held to increase awareness of TSC, and workshops at professional meetings should be organized. A great deal of enthusiasm was also expressed for organizing a Gordon Conference devoted to tuberous sclerosis.


Reviewed October 1, 2003

Last updated April 14, 2010