TuberOus SClerosis registry to increase disease Awareness (TOSCA) – baseline data on 2093 patients

Background Tuberous sclerosis complex (TSC) is a rare autosomal dominant genetic disorder. Many gaps remain in the understanding of TSC because of the complexity in clinical presentation. The TuberOus SClerosis registry to increase disease Awareness (TOSCA) is an international disease registry designed to address knowledge gaps in the natural history and management of TSC. Here, we present the baseline data of TOSCA cohort. Methods Patients of any age diagnosed with TSC, having a documented visit for TSC within the preceding 12 months, or newly diagnosed individuals were included. The registry includes a “core” section designed to record detailed background information on each patient including disease manifestations, interventions, and outcomes collected at baseline and updated annually. “Subsections” of the registry recorded additional data related to specific features of TSC. Results Baseline “core” data from 2093 patients enrolled from 170 sites across 31 countries were available at the cut-off date September 30, 2014. Median age of patients at enrollment was 13 years (range, 0–71) and at diagnosis of TSC was 1 year (range, 0–69). The occurrence rates of major manifestations of TSC included – cortical tubers (82.2%), subependymal nodules (78.2%), subependymal giant cell astrocytomas (24.4%), renal angiomyolipomas (47.2%), lymphangioleiomyomatosis (6.9%), cardiac rhabdomyomas (34.3%), facial angiofibromas (57.3%), forehead plaque (14.1%), ≥ 3 hypomelanotic macules (66.8%), and shagreen patches (27.4%). Epilepsy was reported in 1748 (83.5%) patients, of which 1372 were diagnosed at ≤ 2 years (78%). Intellectual disability was identified in 451 (54.9%) patients of those assessed. TSC-associated neuropsychiatric disorders (TAND) were diagnosed late, and not evaluated in 30–50% of patients. Conclusion TOSCA is the largest clinical case series of TSC to date. It provided a detailed description of the disease trajectory with increased awareness of various TSC manifestations. The rates of different features of TSC reported here reflect the age range and referral patterns of clinics contributing patients to the cohort. Documentation of TAND and LAM was poor. A widespread adoption of the international TSC assessment and treatment guidelines, including use of the TAND Checklist, could improve surveillance. The registry provides valuable insights into the necessity for monitoring, timing, and indications for the treatment of TSC. Electronic supplementary material The online version of this article (doi:10.1186/s13023-016-0553-5) contains supplementary material, which is available to authorized users.


Background
Tuberous sclerosis complex (TSC) is a rare genetic disorder characterized by the development of benign tumors in several organs of the body [1]. The birth incidence of the disorder is approximately 1 in 5800 individuals [2]. TSC is caused by genetic mutations in either of the TSC1 or TSC2 genes [3]. Based on routine diagnostic techniques, a pathogenic mutation is detected in up to 85-90% of individuals with a clinical diagnosis of TSC [1]. In the remaining 10-15% patients with 'no mutation identified' , next generation DNA sequencing (NGS), a high-throughput sequencing, identified mosaic or intronic mutations in TSC1 or TSC2 genes in a vast majority suggesting that it is unlikely that a third TSC gene exists [4]. Mutations of TSC1 or TSC2 gene result in overactivation of the mammalian target of rapamycin (mTOR) complex 1, a key intracellular regulator of cell growth and proliferation, resulting in the hamartomatous lesions found in multiple organs [5,6]. Recent research has helped us understand the pathophysiology of TSC, which has led to the use of mTOR inhibitors for the treatment of certain manifestations of TSC including subependymal giant cell astrocytomas (SEGAs) and renal angiomyolipomas [7][8][9][10]. The recently revised guidelines for the surveillance and management of TSC provided updated recommendations for standard, optimal care for patients [10].
There is, however, still a lack of clarity with respect to the natural history of many of the TSC manifestations, their variability, the age-related expression pattern, and their prognostic roles. Gaps also exist in understanding the rare symptoms and comorbidities of TSC, the relationship between genotype and phenotype, and the various interventions, treatments, and their outcomes. An improved understanding of the natural history of TSC is essential in order to evaluate the benefitrisk ratio of any intervention accurately. TuberOus SClerosis registry to increase disease Awareness (TOSCA) was established as a multicenter, international disease registry with the specific aim to gather clinical data on this rare disease in a systematic and longitudinal way. TOSCA consists of a "core" dataset representing the diagnostic characteristics and core associated clinical features, and "petal projects" which represent detailed sub-projects focusing on specific TSC manifestations. The results from the baseline core data of the TOSCA cohort are reported here.

Methods
The study methods have been described in detail previously [11]. All patients gave informed consent. TOSCA is a multicenter, international disease registry that was designed to collect data on patients with TSC from many countries worldwide. Patients of any age with a diagnosis of TSC (definite, probable, or possible) and with a documented clinical visit for TSC within the past 12 months or newly diagnosed with TSC were eligible. The certainty of diagnosis was based on the 1998 revised Gomez criteria. [12] The registry consists of a "core" section and subsections or "petals". The "core" section collected general information on patients' background which included demographic data, family history, prenatal history, and disease features such as neurological and neuropsychiatric, renal, cardiovascular, pulmonary, dermatological, and others. This information was collected at baseline and is being updated annually. Subsections ("petals") are being developed as research projects to record in-depth data related to specific disease manifestations. Pediatric and adult sites with specialists in managing one or more aspects of TSC were included in the registry. Patients will be followed up for up to five years and an interim analysis will be conducted every year.

Overall findings
TOSCA recruited a total of 2223 patients from 170 centers in 31 countries (Fig. 1), over half of them (57%) from neuropediatric/pediatric clinics (Fig. 2). At the cutoff date September 30, 2014, complete baseline data from 2093 patients (1009 male and 1084 female) were available. Median age at inclusion in TOSCA was 13 years (range, 0-71). Patient distribution by age at inclusion in TOSCA is shown in Fig. 3

Neurological manifestations
Cortical tubers and subependymal nodules were the most commonly reported neurological manifestations (reported in 82.2 and 78.2% of patients, respectively). A total of 510 (24.4%) patients had ever been diagnosed with SEGA (Table 2). For TOSCA, a broad definition of SEGA was adopted, based on the presence of a lesion near the foramen of Monro without specific criteria for size or growth. Median age at SEGA diagnosis was 8 years (range 0-51) (Additional file 1). SEGA was diagnosed before age 2 years in 132 (26.4%), before age 18 years in 278 (55.6%), and after age 18 years in 90 (18%) patients. SEGA was present at the time of recruitment in TOSCA in 422 (82.7%) patients. Of these, 195 (46.2%) were bilateral and 155 (36.7%) showed growth since the previous scan. Growth was reported in 10 out

Cardiovascular manifestations
Cardiac rhabdomyomas, which were the most frequent cardiovascular manifestations found in 717 (34.3%) patients (Table 2). These were diagnosed at a mean age of 3.1 years. Cardiac rhabdomyomas reported earlier and were still present in 483 (67.4%) patients at the time of Collectively, manifestations previously thought to be rare were reported in 316 (15.1%) patients; bone sclerotic foci in 87 patients, scoliosis in 46 patients, thyroid adenoma in 15 patients, spleen angiomyolipoma in 5 patients, pancreatic neuroendocrine tumor in 5 patients, and hemihypertrophy (abnormal growth on one side of the body), calvarial sclerosis and thickening (each in 2 patients).
Co-morbidities were reported in 347 (16.6%) patients; cardiovascular co-morbidities were the most frequent (44 patients). Other less common comorbidities included malignancies (15 patients), dyslipidemia (17 patients) and diabetes (5 patients). Other features of TSC reported were bone cysts (65 patients), non-renal hamartoma (excluding liver, 34 patients), and hamartomatous rectal polyps (8 patients). Figure 4 depicts a distinctive pattern to age-related emergence and prevalence of TSC manifestations. Hypomelanotic macules, subependymal nodules, cortical tubers and cardiac rhabdomyomas were reported from age ≤ 2, (presumably as soon as they were looked for), and their prevalence did not change. In contrast, the prevalence of SEGAs and retinal hamartomas peaked in childhood, and the prevalence of renal angiomyolipomas, facial angiofibromas, forehead plaques and shagreen patches went on increasing into adulthood. However all these lesions were reported in a small number of patients from age ≤ 2 years. In contrast, pulmonary LAM and ungual fibromas presented later but also became more prevalent in adults.

Discussion
The TOSCA natural history study represents the largest clinical collection of TSC data to date. It is not possible to derive an accurate absolute prevalence of TSC or its individual manifestations in the general population from this dataset because it was ascertained from a specialist clinic population. However, the data do record the relative prevalence of different manifestations within this cohort and their natural history. Reassuringly, our data shows a number of similarities to what has been reported in previous studies of TSC [1,14]. For instance, TSC2 mutations were more common than TSC1 mutations [4,15] and the prevalence of certain disease features such as cortical tubers, subependymal nodules and epilepsy were similar to previous reports [1,14]. However, the frequencies of some of the other features such as SEGA, TAND, and renal angiomyolipomas differed from the published data [1,14]. Potential reasons for these similarities and differences are discussed in more details below.
SEGAs were reported in 24.4% of the patients, which is much higher than the previously reported rate of 10-15% [1,11]. The higher frequency of SEGAs seen in this cohort could be because the majority of centers included in TOSCA were specialist neurology centers. There also is ongoing discussion with respect to the most accurate definition of SEGA, which may have been of influence on the number of SEGA reported in TOSCA. In 2012, a European Consensus Group defined SEGA as a tumor in a TSC patient that is usually characterized by a location near the foramen of Monro, > 0.5 cm in diameter, with any documented growth, and gadolinium enhancement on neuroimaging [16]. Later that year, an international panel of experts defined SEGA as a lesion at the caudothalamic groove with either a size of more than 1 cm in any direction or a subependymal lesion at any location that has shown serial growth on consecutive imaging regardless of size [17]. Most SEGAs show avid enhancement after contrast administration; however, a growing subependymal lesion even in the absence of enhancement should be considered a SEGA [17]. Median age at SEGA diagnosis was 8 years but more than a quarter of patients were diagnosed with SEGA already before age two years, highlighting the young age at onset and early need for follow-up. When compared to the last scan, 36.7% of ongoing SEGAs were reported to have grown in size. Since SEGAs are known to grow over time, there are existing recommendations for their regular follow-up and timely management [10,16]. Median time between scans was 1 year and median time between SEGA diagnosis and start of treatment was less than 1 year, reflecting good clinical practice with respect to SEGA follow-up and management in the TOSCA cohort. Although SEGA growth was most common between ages 5-18 years, growth after age 18 years remains possible as was shown in this cohort. This highlights the necessity to stay vigilant to possible symptoms related to SEGA-growth also at adult age. In this analysis, surgery was the most common mode of treatment for SEGAs followed by mTOR inhibitors. Until the recent approval of everolimus for the treatment of SEGAs associated with TSC [9,18], surgery was the only treatment option. For acutely symptomatic SEGAs, surgery and cerebrospinal fluid diversion remain the treatment of choice [10]. However, for asymptomatic growing SEGAs, mTOR inhibitors may be considered especially in multisystem disease [10], since mTOR inhibitors have also been found to benefit other manifestations of TSC [19][20][21][22][23]. Due to high rate of regrowth of residual tumors [24], mTOR inhibitors should be recommended for those patients with SEGA, in whom total surgical removal of SEGA is not possible. Complete SEGA resection might be more difficult to achieve in the presence of bilateral SEGA [17], which were present in over one third of patients in this cohort.
Similar to previous reports [3], this analysis of TOSCA data showed that epilepsy (83.5%) was the most commonly reported clinical presentation of TSC. Focal seizures were the most common type of seizures followed by infantile spasms. GABAergics (vigabatrin) were most frequently used, both for focal seizures and for infantile spasms. This finding is in line with European recommendations made by TSC Consensus meeting for SEGA and epilepsy management in 2012, which recommended vigabatrin both for infantile spasms and focal seizures in infants in the first year of life [25]. Also the guidelines from the 2012 International TSC consensus Conference recommend vigabatrin as first line and adrenocorticotropic hormone as the second line treatment for infantile spasms in individuals with TSC [10]. The use of other anti-epileptic drugs; e.g., sodium channel blockers and fructose derivatives will be described in more detail in a subsequent paper. Despite a high rate of refractory epilepsy, alternative treatment options such as the ketogenic diet and epilepsy surgery were not commonly used in this population.
Recently, mTOR inhibitors have been thought to be useful for the treatment of refractory seizures [26][27][28]. Results from a phase III, randomized, double-blind, placebo-controlled study (EXIST-3; NCT01713946) could tell us the efficacy and safety of 2 trough-ranges of everolimus (an mTOR inhibitor) as adjunctive therapy in patients with TSC who have refractory seizures.
The TAND domain showed lower rates of behavioral and psychiatric disorders than previously reported [29][30][31]. The rates of intellectual disability were similar to previous reports [29,30] but there was no clear evidence of a bimodal distribution of IQ/DQ. Very high rates of academic difficulties and of neuropsychological deficits were reported, and represent the first report of the potential magnitude of academic and neuropsychological deficits in TSC. Strikingly though, diagnoses of ASD, ADHD, anxiety and depression were made very late, and the TAND domain was characterized by very high rates of missing data. These findings suggest that, even in the TOSCA cohort, TAND is underdiagnosed and therefore undertreated. A major challenge in maintaining a database like TOSCA is to ensure that data are complete, especially since these are collected from many centers over a long time period. Missing data for TAND suggests that a considerable number of patients were never assessed for TAND. It is apparent that there is a specific need to educate clinicians to assess all the patients with TSC for TAND. In order to address the need, to increase the awareness of TAND and the importance of screening for these difficulties, the Neuropsychiatry Panel at the 2012 Tuberous Sclerosis Complex International Consensus Conference developed a simple tool called the TAND checklist [31,32]. The neuropsychiatry panel recommended at least an annual screening for TAND and comprehensive formal evaluation for TAND at key developmental time points: infancy (0-3 years), preschool (3-6 years), pre-middle school (6-9 years), adolescence (12-16 years), early adulthood (18-25 years), and as needed thereafter. Management strategies should be based on the TAND profile of each patient and should be based on evidence-based good practice guidelines/practice parameters for individual disorders (e.g., autism spectrum disorder, attention deficit hyperactivity disorder, anxiety disorder). The TAND checklist can serve as an ideal guide to facilitate the discussion between the healthcare professionals and patients [31].
The lower rate of renal angiomyolipomas reported in this population was likely attributable to the fact that the cohort had a younger median age. This is clear from Fig. 4 that the prevalence in adults (e.g., 80% in age > 40) is the same as in other studies. Also, the lower than expected prevalence of complications of angiomyolipoma such as hypertension, microscopic hematuria, and impaired renal function (reported in the literature as 27, 25-50, and 40% respectively) [33][34][35] reported here is a reflection of the young age of this cohort, who have not had time to develop the most common adult renal complications; and to good practice in specialist clinics of active surveillance and pre-emptive treatment of enlarging angiomyolipomas. The finding that 27.8% of those with angiomyolipomas had received treatment for them (presumably mostly pre-emptive) supports this suggestion. An important finding is that a significant number of children (55 patients aged <18 years) needed intervention for their angiomyolipomas. It is also of note that the occurrence of angiomyolipomas was not statistically different in women compared to men. A previous report [36] found angiomyolipomas complications were more common in women than men, and two thirds of the recruits for EXIST-2 [8] were women; both findings implying that if there is no sex difference in angiomyolipomas prevalence, women are more vulnerable to developing complications. Renal angiomyolipomas can cause considerable morbidity including complications like aneurysm and hemorrhage [37]. Moreover, renal complications have been associated with the most common cause for death in adult patients with TSC [38]. Timely diagnosis and treatment are therefore of utmost importance. The main objective in treating renal angiomyolipomas is to prevent hemorrhage and to preserve renal function. TSC Consensus Conference guidelines recommend embolization followed by corticosteroids as the firstline of treatment for angiomyolipomas presenting with acute hemorrhage [10]. An mTOR inhibitor is the recommended first-line therapy for asymptomatic, growing angiomyolipoma > 3 cm in diameter [10]. In this cohort, renal angiomyolipomas were most commonly treated with embolization followed by mTOR inhibitors.
LAM is the main pulmonary manifestation of TSC, which is seen in about 40% of women of reproductive age [39,40]. A recent study has reported a higher prevalence (up to 80%), especially in women over 40 years of age [39]. In this dataset, LAM was reported much less frequently compared to what has been published in literature. This could probably be explained partly by the young average age of the cohort and probably also because screening may have been based on clinical symptoms rather than high-resolution chest computed tomography (HRCT). However 40% of these patients were symptomatic from their LAM. Cudzilo et al. reported that most women with TSC develop cystic changes consistent with LAM [39], highlighting the importance of routine surveillance using CT scan. As per the guidelines, baseline pulmonary function testing and HRCT must be performed in females 18 years or older, even if asymptomatic [10]. Screening for LAM in female patients and adult males (symptomatic) with TSC as per the recommendations would be helpful. The high occurrence of symptomatic retinal hamartomas (12.6%) is a new finding and has implications for surveillance. In this analysis of TOSCA, it was also noted that most of the patients were asymptomatic. It is therefore crucial to highlight the importance of regular surveillance in all patients with TSC even in the absence of symptoms to help prevent complications.
With regard to the age at occurrence, most of the manifestations in majority of the patients were diagnosed at the median age of 1 year. The mean age at diagnosis for cardiac rhabdomyoma was 3.1 years. The highest incidence of cardiac tumors is in children below 2 years of age reaching up to 65-80% [40]. This late age at diagnosis of cardiac rhabdomyomas could be related to age at diagnosis of TSC. Apart from those diagnosed with rhabdomyomas on fetal ultrasonography, other patients may have had an echocardiogram organized after their diagnosis of TSC was made.
Early, sometimes prenatal, diagnosis of TSC enables the beginning of surveillance and thus prevention or amelioration of complications such as epilepsy, intellectual disability, autistic behaviors and tumors development [17,41,42]. As TSC is a genetic disease, the family members must also be assessed. Family counselling must be done and genetic testing carried out [10]. Current molecular techniques enable the TSC1/TSC2 mutation detection in more than 95% of patients [4]. In TOSCA participants, genetic testing was not reported in about 40% of the patients, which might be due to ethical or financial reasons. Five patients reported both TSC1 and TSC2 mutations, this unexpected finding is being investigated and will be reported in more detail in future publications.
Though TOSCA is a large clinical case series, it must be noted that patients were recruited through clinical centers with expertise in TSC and milder cases may not always be seen at these centers. The study design therefore includes potential limitations inherent to clinical case series, albeit large-scale, multinational ones. Nevertheless, participation of a large number of centers with complementary expertise has helped inclusion of a huge number of patients with TSC, which reveals occurrence rates of complications that are likely to be representative of hospital clinical practice. This baseline paper of TOSCA provides a detailed description of the disease trajectory of TSC. The registry can provide valuable insights into the necessity for monitoring, the timing, and the indications for treatment of this disease. Further follow-up studies of TOSCA including research projects will provide more detail in understanding the treatment interventions and outcomes.

Conclusion
This international registry provides a better understanding of the TSC manifestations, and facilitates development of better management and surveillance strategies for patients with TSC. Following patients over the years will help in understanding any changes in the treatment and outcome of the different manifestations. Baseline analysis has highlighted the serious import of epilepsy and SEGA in children, and renal angiomyolipomas in children as well as adults. We found inadequate surveillance for TAND; hopefully this will be improved in future by widespread adoption of the international guidelines including use of the TAND checklist. Clinicians need to be alert for rare complications but especially changes in retinal hamartomas. We believe comprehensive surveillance will lead to more pro-active pre-emptive treatment and better outcomes in future. Subsequent analyses are planned yearly to allow the clinical course of the disease over time to be evaluated.
Competing interests RC, GG, and MR are employees of Novartis. All authors are members of TOSCA Scientific Advisory Board and have received honorarium from Novartis. JCK and BZ report receiving grants from Novartis. ACJ, SJ, and VC received lecture fees from Novartis. MS has received personal fees for consultancy activities and lectures and financial support for investigator driven research from Novartis. PJdV has been a co-principal investigator on research studies funded by Novartis.

Consent for publication
All patients were informed that the reason for enrolling in the study was to collect natural history data for scientific analysis and publication (an obligation mandated by the Good Clinical Practice guidelines) and they signed consent forms with that understanding.
Ethics approval and consent to participate All patients (or their legal representatives) provided written informed consent before enrolling in the TOSCA disease registry. The study was designed, implemented, and reported in accordance with the principles of Good Clinical Practice, Good Pharmacoepidemiology Practices of the International Society for Pharmacoepidemiology (ISPE 2008), the STROBE (Strengthening and Reporting of Observational Studies in Epidemiology) guidelines, and the ethical principles laid down in the Declaration of Helsinki, and all local regulations. The study protocol and all amendments were reviewed and approved by independent ethics committee/institutional review board for each center (