Tuberous sclerosis complex

Meet the patient

A 7-month-old is referred for clusters of sudden flexion jerks — arms fling out, knees tuck up, thirty at a time on waking. The EEG is chaotic (hypsarrhythmia). You dim the lights, switch on the Wood's lamp, and sweep the trunk: four pale, lance-shaped macules glow back at you. You have the diagnosis before the MRI is booked.^[12][13]

The two questions that decide this child's trajectory are the two that decide every TSC case: which organs are growing hamartomas? (skin, brain, heart, kidney, lung — image them all), and can I switch the mTOR brake back on? (a rapalog, when the lesions threaten). Hold those two questions and the whole topic slots into place.^[1]

One brake, two genes, hamartomas everywhere

TSC is what happens when the cell loses its brake on growth. In a normal cell the proteins hamartin (TSC1) and tuberin (TSC2) hold hands as a heterodimer that acts as a GTPase-activating protein (GAP) for the small GTPase Rheb. Rheb-GTP switches mTORC1 on; the hamartin-tuberin GAP forces Rheb back to GDP and keeps mTORC1 quiet until growth factors arrive.^[1][18]

Lose one allele of TSC1 or TSC2 (germline) and the cell is one hit from disaster; lose the second (a somatic second hit, Knudson-style) and the last working brake disappears. Rheb-GTP accumulates, mTORC1 fires constitutively, S6K1 and 4EBP1 drive translation, and the affected clone expands into a hamartoma — a disorganised overgrowth of cells native to the tissue: cortical neurons, renal tubular epithelium, vascular smooth muscle, adipocytes, fibroblasts.^[1][11][20]

Everyone asks the follow-up: if mTORC1 is an oncogenic pathway, why hamartomas and not carcinomas? Because mTORC1 is a proliferative and survival signal, not a full oncogenic driver. A second mutation beyond TSC1/TSC2 loss is needed to cross into malignancy, and those hits are uncommon — so the lesions stay benign, with only rare malignant conversion (chiefly renal cell carcinoma in a long-standing angiomyolipoma).^[1][11]

Etymology for viva gold: Bourneville disease honours Désiré-Magloire Bourneville, who in 1880 described the potato-hard sclerotic gyri at post-mortem — tuberous = tuber-like, sclerosis = hardening. Epiloia, the older portmanteau, fuses epilepsy, low intelligence and adenoma sebacea — the triad every 1920s textbook printed and every Indian answer key still remembers.^[1]

The cutaneous tetrad — four lesions, four ages

The skin makes the diagnosis in most patients, and each lesion arrives on a clock. Four cutaneous features are the diagnostic backbone; learn them with their age of onset, because the age is as examinable as the lesion.^[1][12]

^{[1] [2] [3]}

Two further cutaneous clues are worth their own sentences. The forehead fibrous plaque — a yellow-brown, hair-bearing or alopecic plaque on the forehead or scalp — can be the only sign in infancy and is easily dismissed as a congenital naevus; in 2021 it was promoted to its own major criterion (fibrous cephalic plaque).^[12] And the confetti-like hypomelanotic macules — scattered 1–3 mm polygonal pits of pigment on the limbs — plus dental enamel pitting are the two minor cutaneous and oral features examiners love to slip into a viva.^[2][3]

The classic trap: the lesion called adenoma sebaceum is neither an adenoma nor sebaceous. Histology shows a fibrovascular hamartoma — blood vessels and fibroblasts, no gland. The 19th-century name stuck; the viva answer and the biopsy report say angiofibroma. Reach for "sebaceous adenoma" in an answer and you have just lost the mark.^[1]

Everyone forgets: a single hypomelanotic macule is not diagnostic — you need three or more, at least 5 mm across, to claim the major criterion. Two macules and a seizure is "possible TSC", not "definite".^[2][3]

The single most useful bedside test for the macules is Wood's lamp examination at 365 nm: darken the room, hold the lamp 10–15 cm off the skin, and sweep trunk and limbs. The macule appears as a sharply demarcated patch of reduced fluorescence against the bluish normal skin — and this is the test that finds the ash-leaf macule on dark skin where it is otherwise invisible.^[1][12]

Is it TSC or NF1? — the neurocutaneous face-off

TSC is not a variant of neurofibromatosis. Both are autosomal dominant neurocutaneous syndromes, and both are exam staples — but they share nothing but the syllabus. Learn the one-line discriminator for each row and you will never confuse them.^[1]

^[1]

The one-line discriminator: ash-leaf and angiofibromas go with TSC; cafe-au-lait and neurofibromas go with NF1. If a stem gives you six cafe-au-lait macules and axillary freckling, that is NF1 — full stop — and reaching for an mTOR inhibitor would be the wrong pathway entirely.^[1]

The 2012/2021 diagnostic criteria — reproduced verbatim

These are criteria examiner will ask you to reproduce, word for word. A definite diagnosis is a pathogenic TSC1/2 variant alone, OR two major features, OR one major plus two or more minor features. A possible diagnosis is one major feature, OR two or more minor features.^[2][3]

A. Genetic criterion — sufficient on its own

Identification of a pathogenic variant in TSC1 or TSC2 by validated genetic testing makes the diagnosis definite, independent of any clinical feature. A pathogenic variant is one that is clearly protein-truncating (nonsense, frameshift, canonical splice, large deletion) or a missense with demonstrated loss of function; a variant of uncertain significance does NOT qualify.^[2][3]

B. Major features (12)

1. Hypomelanotic macules (3 or more), at least 5 mm diameter.^[2][3]
2. Angiofibromas (3 or more) or forehead fibrous plaque.^[2][3]
3. Fibrous cephalic plaque (separated out in the 2021 update).^[2]
4. Ungual fibromas (2 or more) — Koenen tumours.^[2][3]
5. Shagreen patch (connective-tissue naevus).^[2][3]
6. Multiple retinal hamartomas.^[2][3]
7. Cortical dysplasias (cortical tubers and cerebral white-matter radial migration lines).^[2][3]
8. Subependymal nodules (2 or more).^[2][3]
9. Subependymal giant cell astrocytoma (SEGA).^[2][3]
10. Cardiac rhabdomyoma.^[2][3]
11. Lymphangioleiomyomatosis (LAM).^[2][3]
12. Renal angiomyolipoma (2 or more).^[2][3]

C. Minor features (7)

1. Confetti skin lesions. 2. Dental enamel pits (more than 3). 3. Intraoral fibromas (2 or more). 4. Retinal achromic patch. 5. Multiple renal cysts. 6. Non-renal hamartomas. 7. Sclerotic bone lesions.^[2]

(Source: Northrup H et al., Pediatric Neurology 2021, PMID 34399110.)^[2]

^[2]

The consultant confession: the 2021 update did two things candidates still miss — it split fibrous cephalic plaque out as its own major criterion (it used to hide under angiofibromas), and it confirmed that LAM plus renal AML together counts as two major features, which on its own secures a definite diagnosis in an adult woman who never had a skin exam.^[2]

Organ involvement — where the hamartomas grow

TSC touches every organ, and each organ has a signature lesion. The brain carries the morbidity; the kidneys and the lungs carry the mortality; the skin carries the diagnosis.^[1][2]

• Brain — cortical tubers (the epileptic substrate), subependymal nodules lining the ventricles, and SEGA near the foramen of Monro in 5–15 percent; epilepsy in 80–90 percent; autism and intellectual disability in up to half; the TAND cluster (TSC-Associated Neuropsychiatric Disorders) is near-universal and needs structured screening.^[1]
• Heart — cardiac rhabdomyoma in roughly half of infants, often the presenting sign on fetal echocardiography; usually multiple, intramural, and regressing spontaneously through childhood, so intervention is reserved for arrhythmia or outflow obstruction.^[1]
• Kidneys — renal angiomyolipoma in 70–80 percent of adults, with bleeding risk once a lesion passes 3–4 cm or grows an aneurysm of 5 mm or more; plus cysts and rare renal cell carcinoma within a long-standing AML.^[1][11][7]
• Lungs — LAM, almost exclusively in adult women, with diffuse thin-walled cysts, recurrent pneumothorax, chylous effusions, progressive airflow obstruction.^[1][8][15]
• Eyes — retinal hamartomas in 30–50 percent, usually asymptomatic.^[1]

The danger triad — three organs that kill

TSC is a multisystem disease, but three organ complications carry the patient out. Name them at every viva, in this order, because the order is the order of acuity.^[2][11][8]

• SEGA with obstructive hydrocephalus — the subependymal giant cell astrocytoma grows at the foramen of Monro, blocks CSF outflow, and presents as acute raised intracranial pressure. Dexamethasone, an external ventricular drain, and either neurosurgical resection or everolimus to shrink the lesion.^[2][5]
• Wunderlich haemorrhage from a ruptured renal AML — spontaneous retroperitoneal bleed from an angiomyolipoma that has exceeded 3–4 cm or grown an aneurysm of 5 mm or more: flank pain, shock, falling haemoglobin. Two large-bore cannulae, cross-match, urgent CT angiogram, and selective arterial embolisation (nephron-sparing) — nephrectomy is the last resort.^[7][11]
• Spontaneous pneumothorax in LAM — diffuse cystic lung destruction in (almost exclusively) adult women; roughly half will puncture. Intercostal drain, pleurodesis before any future transplant, lifelong sirolimus, and a hard rule against breath-hold diving and flying with a recent pneumothorax.^[8][14]

The two non-triad killers to keep in the same sentence: sudden unexpected death in epilepsy (SUDEP) is the commonest premature death in TSC, and status epilepticus in a known patient forces a repeat MRI to exclude SEGA growth or subependymal nodule transformation.^[1][2]

Infantile spasms — vigabatrin, now

The first seizure in TSC is usually infantile spasms, and vigabatrin is first-line in TSC. Onset clusters between 4 and 9 months: sudden flexion or extension jerks in runs of dozens, an EEG that is a chaotic mess of high-voltage slow waves and spikes (hypsarrhythmia), and, very often, the ash-leaf macules you find by Wood's lamp the same morning. This is West syndrome in a TSC child.^[1][13]

The reason TSC changes the algorithm is the response rate. In the only randomised comparison against a hormonal therapy, all 11 vigabatrin-treated infants with TSC became spasm-free versus 5 of 11 on hydrocortisone, and spasms disappeared faster (mean 3.5 versus 13 days).^[21] Start early: early epileptiform EEG activity predicts epilepsy and worse neurodevelopmental outcome.^[13]

^{[1] [13] [21] [22]}

The classic trap: vigabatrin works so well in TSC that its safety signals are easy to defer — visual-field loss and vigabatrin-associated brain MRI abnormalities are the reasons for baseline testing and structured follow-up. Document baseline fields where the child can co-operate, and re-check periodically in older children and adults.^[22]

mTOR inhibitors — the rapalogs

The molecular logic dictates the treatment: if TSC is a lost brake on mTORC1, give the brake back pharmacologically. Both everolimus (a 40-O-(2-hydroxyethyl) rapamycin) and sirolimus (rapamycin itself) are rapalogs that bind FKBP12, and the rapalog-FKBP12 complex allosterically inhibits mTORC1. They do, chemically, what the hamartin-tuberin complex used to do biologically.^{[5][7][8][9][10]}

The rule for which rapalog, which route, which lesion:^[2][4]

^{[5] [7] [8] [9] [10]}

Everolimus is the systemic workhorse. Titrate to a trough of 5–15 ng/mL. EXIST-1 (Franz, Lancet 2013) halved SEGA volume in 35 percent versus zero on placebo, with the response sustained at two years; EXIST-2 (Bissler, Lancet 2013) shrank renal AML by half in 42 percent versus zero, with renal function preserved over the long-term extension; EXIST-3 (French, Lancet 2016) halved refractory focal seizure frequency in 40 percent of the high-exposure arm versus 15 percent on placebo.^{[5][6][7][10][17][19]}

Sirolimus is the drug for the chest. In the MILES trial (McCormack, NEJM 2011), 12 months of oral sirolimus stopped the fall in FEV1 — the between-group difference was 153 mL, about 11 percent of the enrolment FEV1 — and improved FVC, functional residual capacity, serum VEGF-D and quality of life; when sirolimus stopped, the decline resumed.^[8] The serum VEGF-D biomarker tracks severity and treatment response: baseline levels were higher in patients needing supplemental oxygen and fell with sirolimus at 6 and 12 months (Young, Lancet Resp Med 2013).^[15]

Topical rapamycin is the trial-proven option for facial angiofibromas. In the TREATMENT trial (Koenig, JAMA Dermatol 2018), both 0.1 percent and 1 percent rapamycin applied daily at bedtime beat vehicle on the Angiofibroma Grading Scale within a month, with the higher concentration better still.^[9]

The organ-by-organ ladder

The classic trap: stop everolimus after an AML shrinks and it rebounds within months — mTOR inhibition is a brake, not a cure. Continue unless toxicity forces a pause, and hold it two to four weeks before any major surgery (it wrecks wound healing). Watch for stomatitis, hyperlipidaemia, myelosuppression and pneumonitis.^[7][19]

Two genotypes, one response

TSC2 is the bigger, meaner gene. TSC2 mutations bring earlier seizure onset, more cortical tubers, larger SEGA, heavier seizure burden, more autism, and more renal AML than TSC1. Two-thirds of cases are TSC2; about a quarter are TSC1; the remainder are mosaic or unclassified.^[1][16]

The consolation that earns viva marks: response to everolimus is independent of which gene is mutated and where the mutation sits within it — SEGA and AML shrink in TSC1 and TSC2 alike, because the convergent endpoint is always mTORC1 activation.^[16]

The TSC2-PKD1 contiguous gene syndrome is the special case to name unprompted. A large deletion on chromosome 16p13 takes out both TSC2 and the adjacent PKD1 gene, so the patient has TSC and autosomal dominant polycystic kidney disease — early hypertension, progressive cystic kidneys, declining eGFR. Check the blood pressure, control it tightly, consider tolvaptan in selected adults, and run combined nephrology-genetics care.^[1][11][18]

Mosaic TSC is the other special case: the mutation is present only in a subset of cells, so the phenotype is segmental or attenuated — a unilateral band of angiofibromas along Blaschko lines, minimal systemic disease. Blood genetic testing may come back negative; test the affected skin.^[1][12]

Surveillance — the lifelong rhythm

TSC is a clinic for life, and surveillance is how you stay ahead of the danger triad. The 2012/2021 consensus sets the rhythm, and the mTOR inhibitors add a trough to check.^[2][4]

• Brain MRI — every 1–3 years in children with SEGA or subependymal nodules; every 5–10 years in adults if previously normal.^[2]
• EEG — at diagnosis in infants; serial EEG in infants with normal imaging to catch pre-symptomatic epileptiform activity, because early EEG activity predicts epilepsy and neurodevelopmental outcome.^[13]
• Renal MRI (or CT/MR angiography) — every 1–3 years to track AML growth and aneurysm formation.^[11]
• Echocardiogram — at diagnosis in infants (cardiac rhabdomyoma); repeat only if symptomatic, because most regress spontaneously through childhood.^[2]
• High-resolution chest CT — adult women at diagnosis and every 5–10 years for LAM; men only if symptomatic.^[8]
• Serum VEGF-D — adult women with suspected LAM; 800 pg/mL or higher confirms the diagnosis and avoids biopsy.^[15]
• Everolimus/sirolimus trough — every 1–3 months, target 5–15 ng/mL.^[5][7][8]
• TAND screen — at diagnosis and annually: autism, intellectual disability, ADHD, anxiety, mood, behaviour. Structured neuropsychology is the standard of care.^[1][2]

The mimics — and the trap in each

Differential diagnosis turns on the index lesion. Hold the index lesion in mind and the list collapses to a handful, each with a one-line discriminator.^[1][12]

^[1]

How TSC patients come to harm — the preventable list

• SUDEP in a child with uncontrolled epilepsy — the commonest preventable death; aggressive seizure control is prevention.^[1][13]
• A Wunderlich haemorrhage from an AML nobody imaged because the patient "felt well" — surveillance MRI catches the 4 cm lesion before it ruptures.^[7][11]
• A SEGA missed on a thin-slice MRI, presenting days later as obstructive hydrocephalus.^[2][5]
• A first pneumothorax in a woman with TSC-LAM who was never told she must not fly or dive.^[8]
• A TSC2-PKD1 contiguous deletion missed on standard sequencing, with uncontrolled hypertension and declining eGFR.^[1][11]
• An everolimus rebound when the drug was stopped after AML shrinkage.^[7][19]
• A vigabatrin visual field that was never checked, lost over years.^[1]
• First-degree relatives of an apparently de novo case who were never screened.^[1]

Special populations

The neonate usually presents through the heart — a cardiac rhabdomyoma on the 20-week fetal echocardiogram — or through the skin at birth (ash-leaf macules) or the brain at 4–9 months (infantile spasms). Work up with brain MRI, echocardiogram, Wood's lamp, abdominal ultrasound, ophthalmology and EEG; test the genes, because early diagnosis opens preventive epilepsy management.^[1][2][12]

The adult woman presents through the chest: progressive dyspnoea, a chylous effusion, or a first pneumothorax — LAM. Baseline HRCT, serum VEGF-D, spirometry, and a decision on sirolimus. Counsel against scuba diving and abrupt barometric change.^[8][15]

Pregnancy in TSC needs pre-conception review: mTOR inhibitors are teratogenic (stop at least 12 weeks before conception), vigabatrin continues where seizure control demands it, and fetal echocardiography at 20–24 weeks screens for cardiac rhabdomyoma. Plan the airway early — facial angiofibromas and gingival fibromas can distort it.^[2]

The trials that changed practice

^{[5] [6] [7] [19] [8] [9] [10]}

The 2012 Consensus (Northrup/Krueger and Krueger/Northrup, Pediatric Neurology 2013) set the diagnostic and surveillance framework; the 2021 update refined it and split out the fibrous cephalic plaque; the 2015 strategic planning conference (Sahin et al.) set the research agenda, including preventive vigabatrin trials.^{[2][3][4][20]}

The mantra, and the mnemonic

^[1]

The mantra: Lose the brake, grow the hamartoma — put it back with a rapalog.^[5][7][8]

Ward-round test — three stems, thirty seconds each

Stem 1 — the baby at the top of the topic (answer)ShowHide

The 7-month-old with flexion-spasm clusters, hypsarrhythmia on EEG, and four ash-leaf macules on Wood's lamp. What is the diagnosis, the likely gene, and the first drug? Model: This is tuberous sclerosis complex — infantile spasms (West syndrome) on a background of hypomelanotic macules, already meeting several major criteria. The likely gene is TSC2: TSC2 variants show earlier epileptiform EEG activity than TSC1.^[13] The first drug is vigabatrin, first-choice in TSC — in the pivotal randomised trial, all 11 infants on vigabatrin became spasm-free versus 5 of 11 on hydrocortisone, treated at 150 mg/kg/day.^[21] Then arrange the consensus baseline workup: brain MRI, echocardiography, renal imaging — and visual-field monitoring once vigabatrin continues.^[1][2]

Stem 2 — the flank pain and the falling haemoglobin (answer)ShowHide

A 28-year-old woman with known TSC on everolimus for SEGA presents with acute left flank pain, hypotension, and a haemoglobin that has dropped 30 g/L. CT angiogram shows a 6 cm left renal angiomyolipoma with active extravasation. Walk the team through the next hour. Model: This is a Wunderlich haemorrhage from a ruptured renal AML — the feared complication of an angiomyolipoma that has outgrown surveillance. Resuscitate first: large-bore access, cross-match, treat the shock, and bring urology and interventional radiology in urgently. Everolimus has a continuing role for the residual lesions and the contralateral kidney — mTORC1 inhibitors shrink angiomyolipomas in trials, but tumours regrow once treatment stops, so therapy is not simply discontinued after one bleed.^[7][11]

Stem 3 — the cosmetic papules that are not acne (answer)ShowHide

A 19-year-old is referred for "acne that does not respond" — bilateral, smooth, pink-red dome papules over the nasolabial folds and malar eminences, sparing the upper lip, with two fleshy fibromas emerging from under the great-toe nails. What is the lesion, what is the diagnosis to exclude, and what is the first-line topical? Model: The facial lesions are angiofibromas (the historical name adenoma sebaceum is a misnomer — they are fibrovascular hamartomas, not sebaceous adenomas), and the toe lesions are Koenen tumours (periungual fibromas). Two major cutaneous criteria (3-plus angiofibromas, 2-plus ungual fibromas) put tuberous sclerosis complex firmly on the table — examine for ash-leaf macules under Wood's lamp and image the brain and kidneys. The topical with trial evidence behind it is rapamycin applied daily at bedtime (the TREATMENT trial), which targets the mTOR-driven lesion directly.^[9][1][12]

References22ShowHide

1. [1]Henske EP, Jóźwiak S, Kingswood JC, et al. Tuberous sclerosis complex. Nature Reviews Disease Primers, 2016.PMID 27226234
2. [2]Northrup H, Aronow ME, Bebin EM, et al. Updated International Tuberous Sclerosis Complex Diagnostic Criteria and Surveillance and Management Recommendations. Pediatric Neurology, 2021.PMID 34399110
3. [3]Northrup H, Krueger DA, International Tuberous Sclerosis Complex Consensus Group. Tuberous sclerosis complex diagnostic criteria update: recommendations of the 2012 International Tuberous Sclerosis Complex Consensus Conference. Pediatric Neurology, 2013.PMID 24053982
4. [4]Krueger DA, Northrup H, International Tuberous Sclerosis Complex Consensus Group. Tuberous sclerosis complex surveillance and management: recommendations of the 2012 International Tuberous Sclerosis Complex Consensus Conference. Pediatric Neurology, 2013.PMID 24053983
5. [5]Franz DN, Belousova E, Sparagana S, et al. Efficacy and safety of everolimus for subependymal giant cell astrocytomas associated with tuberous sclerosis complex (EXIST-1): a multicentre, randomised, placebo-controlled phase 3 trial. The Lancet, 2013.PMID 23158522
6. [6]Franz DN, Belousova E, Sparagana S, et al. Everolimus for subependymal giant cell astrocytoma in patients with tuberous sclerosis complex: 2-year open-label extension of the EXIST-1 study. The Lancet Oncology, 2014.PMID 25456370
7. [7]Bissler JJ, Kingswood JC, Radzikowska E, et al. Everolimus for angiomyolipoma associated with tuberous sclerosis complex or sporadic lymphangioleiomyomatosis (EXIST-2): a multicentre, randomised, double-blind, placebo-controlled trial. The Lancet, 2013.PMID 23312829
8. [8]McCormack FX, Inoue Y, Moss J, et al. Efficacy and safety of sirolimus in lymphangioleiomyomatosis. New England Journal of Medicine, 2011.PMID 21410393
9. [9]Koenig MK, Bell CS, Hebert AA, et al. Efficacy and Safety of Topical Rapamycin in Patients With Facial Angiofibromas Secondary to Tuberous Sclerosis Complex: The TREATMENT Randomized Clinical Trial. JAMA Dermatology, 2018.PMID 29800048
10. [10]French JA, Lawson JA, Yapici Z, et al. Adjunctive everolimus therapy for treatment-resistant focal-onset seizures associated with tuberous sclerosis (EXIST-3): a phase 3, randomised, double-blind, placebo-controlled study. The Lancet, 2016.PMID 27613521
11. [11]Lam HC, Siroky BJ, Henske EP. Renal disease in tuberous sclerosis complex: pathogenesis and therapy. Nature Reviews Nephrology, 2018.PMID 30232410
12. [12]Staley BA, Vail EA, Thiele EA. Tuberous sclerosis complex: diagnostic challenges, presenting symptoms, and commonly missed signs. Pediatrics, 2011.PMID 21173003
13. [13]De Ridder J, Verhelle B, Vervisch J, et al. Early epileptiform EEG activity in infants with tuberous sclerosis complex predicts epilepsy and neurodevelopmental outcomes. Epilepsia, 2021.PMID 33778971
14. [14]Gupta N, Lee HS, Young LR, et al. Analysis of the MILES cohort reveals determinants of disease progression and treatment response in lymphangioleiomyomatosis. European Respiratory Journal, 2019.PMID 30846465
15. [15]Young LR, Lee HS, Inoue Y, et al. Serum VEGF-D a concentration as a biomarker of lymphangioleiomyomatosis severity and treatment response: a prospective analysis of the Multicenter International Lymphangioleiomyomatosis Efficacy of Sirolimus (MILES) Trial. The Lancet Respiratory Medicine, 2013.PMID 24159565
16. [16]Kwiatkowski DJ, Palmer MR, Jozwiak S, et al. Response to everolimus is seen in TSC-associated SEGAs and angiomyolipomas independent of mutation type and site in TSC1 and TSC2. European Journal of Human Genetics, 2015.PMID 25782670
17. [17]Bissler JJ, Budde K, Sauter M, et al. Effect of everolimus on renal function in patients with tuberous sclerosis complex: evidence from EXIST-1 and EXIST-2. Nephrology Dialysis Transplantation, 2019.PMID 30053159
18. [18]European Chromosome 16 Tuberous Sclerosis Consortium. Identification and characterization of the tuberous sclerosis gene on chromosome 16. Cell, 1993.PMID 8269512
19. [19]Bissler JJ, Kingswood JC, Radzikowska E, et al. Everolimus for renal angiomyolipoma in patients with tuberous sclerosis complex or sporadic lymphangioleiomyomatosis: extension of a randomized controlled trial. Nephrology Dialysis Transplantation, 2016.PMID 26156073
20. [20]Sahin M, Henske EP, Manning BD, et al. Advances and Future Directions for Tuberous Sclerosis Complex Research: Recommendations From the 2015 Strategic Planning Conference. Pediatric Neurology, 2016.PMID 27267556
21. [21]Chiron C, Dumas C, Jambaqué I, et al. Randomized trial comparing vigabatrin and hydrocortisone in infantile spasms due to tuberous sclerosis. Epilepsy Research, 1997.PMID 9095401
22. [22]Golec W, Sołowiej E, Strzelecka J, et al. Vigabatrin - new data on indications and safety in paediatric epilepsy. Neurologia i Neurochirurgia Polska, 2021.PMID 34541635