Child's Nervous System

, Volume 30, Issue 1, pp 47–64

Familial syndromes associated with intracranial tumours: a review


    • Department of Clinical Neurological SciencesWestern University
  • Yatri K. Patel
    • Department of Clinical Neurological SciencesWestern University
  • Navjot Chaudhary
    • Department of NeurosurgeryStanford School of Medicine
  • Ram V. Anantha
    • Department of Surgery, Schulich School of Medicine and DentistryWestern University
Review Paper

DOI: 10.1007/s00381-013-2309-z

Cite this article as:
Ranger, A.M., Patel, Y.K., Chaudhary, N. et al. Childs Nerv Syst (2014) 30: 47. doi:10.1007/s00381-013-2309-z



Most cancers of the central nervous system (CNS) occur sporadically in the absence of any known underlying familial disorder or multi-systemic syndrome. Several syndromes are associated with CNS malignancies, however, and their recognition has significant implications for patient management and prognosis. Patients with syndrome-associated CNS malignancies often have multiple tumours (either confined to one region or distributed throughout the body), with similar or different histology.


This review examines syndromes that are strongly associated with CNS cancers: the phakomatosis syndromes, familial syndromes such as Li–Fraumeni and familial polyposis syndromes and dyschondroplasia.


Central nervous system malignanciesFamilial syndromesPhakomatosis syndromesDyschondroplasia


Although a majority of central nervous system (CNS) cancers occur sporadically in individuals with no underlying familial disorder, a strong association exists between CNS malignancies and several inherited syndromes [1], including neurofibromatosis (NF) types 1 and 2, tuberous sclerosis, von Hippel–Lindau (VHL) disease, and Li–Fraumeni syndrome (LFS). Moreover, these syndromes are not uncommon: the prevalence of neurofibromatosis, for example, ranges from 1 in 2,000 to 1 in 5,000 [2], while the prevalence of tuberous sclerosis is approximately 1 in 10,000 births [3]. While most CNS malignancies are intracranial [4], extracranial CNS tumours may occur much more commonly in certain syndromes [5]. Additionally, patients with syndrome-associated CNS malignancies may have multiple tumours with varied histology, either confined to one region or distributed throughout the body [1, 6]. Despite these characteristics, however, the prognosis of syndrome-associated CNS malignancies may be better than the prognosis of sporadic CNS cancers with equivalent histology. Consequently, recognition of these syndromes may result in the earlier diagnosis of CNS tumours in predisposed individuals, thereby leading to earlier treatment and potentially improved outcomes. In addition, the prospective genetic analysis of affected individuals may enhance our understanding of the inheritance of familial risk factors in the development of CNS tumours, thereby improving screening for at-risk individuals [7]. In this review, we aim to discuss three groups of familial syndromes that predispose affected individuals to develop CNS malignancies: the phakomatoses syndromes; dyschondroplasia; and other familial syndromes (Table 1). The characteristics of the CNS tumours and the management of affected patients will be emphasised.
Table 1

Familial disorders associated with central nervous system (CNS) malignancies

Phakomatosis sydromes

 1. Neurofibromatosis (types 1 and 2 and segmental forms)

 2. Tuberous sclerosis

 3. Von Hippel–Lindau disease

 4. Basal cell nevus syndrome

Other familial syndromes

 1. Li–Fraumeni syndrome

 2. Familial adenomatous polyposis (Turcot syndrome)

 3. Rubenstein–Taybi syndrome

Dyschondroplasia syndromes

 1. Maffucci’s syndrome

 2. Ollier’s disease

Phakomatosis syndromes

The phakomatoses are a diverse group of disorders characterised by the presence of pathological lesions of ectodermal origin, including the skin, eyes and the central and peripheral nervous systems [8]. First described in 1920 [9], syndromes belonging to this group include neurofibromatosis, tuberous sclerosis and VHL syndrome; rarely, a patient may present with two distinct phakomatosis syndromes [10]. At a genetic level, all phakomatoses arise because of mutations within tumour suppressor genes. They are inherited in an autosomal dominant (AD) fashion, with variable expression and high penetrance.


Neurofibromatosis (NF) is the most common phakomatosis syndrome [11, 12] and comprises at least two distinct disorders, neurofibromatosis type 1 (NF-1) and neurofibromatosis type 2 (NF-2). The diagnostic criteria for NF-1 and NF-2 differ significantly (Tables 2 and 3, respectively; [12]): the characteristic lesion in NF-1, for example, is the neurofibroma [12, 13], whereas the characteristic lesion in NF-2 is a peripheral nerve Schwannoma or neurolemoma [12, 1416].
Table 2

National Institutes of Health (NIH) consensus criteria for the diagnosis of NF-1

At least 2 of the following within a given patient

 1. 6 or more café au lait spots with a maximum diameter

  a. 5 mm in pre-pubertal patients

  b. 15 mm in post-pubertal patients

 2. 2 or more neurofibromas of any type or one plexiform neurofibroma

 3. Freckling in the axillary or inguinal areas

 4. An optic glioma

 5. 2 or more Lisch nodules

 6. A characteristic skeletal lesion, such as

  a. Sphenoid wing dysplasia

  b. Thinning of long bone cortex, with or without pseudoarthrosis

 7. A first-degree relative (e.g. parent, sibling or child) with confirmed NF-1

Table 3

Modified NIH consensus criteria for neurofibromatosis type 2

Definitive diagnosis

 1. The patient has bilateral CN VIII Schwannomas on MRI or CT scan (no biopsy necessary)

 2. The patient has a first-degree relative with NF2 AND personally has either unilateral early-onset (age, <30 years) CN VIII Schwannoma or any 2 of the following

  a. Meningioma

  b. Glioma

  c. Schwannoma

  d. Juvenile posterior subcapsular lenticular opacity (juvenile cortical cataract)

Presumptive diagnosis

 1. Patient has early-onset (age, <30 years) unilateral CN VIII Schwannomas detected on MRI or CT scan and one of the following

  a. Meningioma

  b. Glioma

  c. Schwannoma

  d. Juvenile posterior subcapsular lenticular opacity

 2. Patient has >2 meningiomas and a unilateral CN VIII Schwannoma or one of the following

  a. Glioma

  b. Schwannoma

  c. Juvenile posterior subcapsular lenticular opacity (juvenile cortical cataract)

Additional clinical variants include milder and more severe forms of NF-2, respectively termed Gardner syndrome, and Wishart or Lee–Abbott syndrome, segmental NF-1 and NF-2 and mixed NF [14, 15]. All forms of the disease are inherited in an AD manner [17], although they display high phenotypic variability [12]. The latter can be observed both within a single family, and even between monozygotic twins, suggesting that other disease-modifying genes, or additional non-hereditary influences such as second-hit somatic events, environmental agents, epigenetic modification and post-zygotic mutations, are important in the manifestations of this disease [12].

Neurofibromatosis type 1

NF-1 is the most common form of neurofibromatosis, and the most common familial syndrome associated with CNS tumours [1, 6, 13]. It affects one in roughly 2,500 to 5,000 live births [1719] and occurs ten times more commonly than NF-2 [11, 17, 18]. Though AD, up to 50 % of cases arise spontaneously from a mutation in the gene encoding for a large protein called neurofibromin [18, 19], located on chromosome 17q11.2. Neurofibromin is a tumour suppressor gene, which, if lost or mutated in both alleles, can result in tumour growth. The disorder is diagnosed using National Institutes of Health (NIH) Consensus Criteria for the Diagnosis of NF-1 [1719]. Clinical signs of NF-1 include café au lait macules, Lisch nodules (iris nodules), axillary and inguinal freckling, neurofibromas, skeletal lesions (such as sphenoid wing dysplasia and thinning of long bone cortices), optic gliomas, and an increased risk of other CNS and systemic tumours [14, 17, 2022]. As certain features such as Lisch nodules are not present in infancy, only 50 % of children meet the diagnostic criteria for NF-1 by 1 year of age [11, 17]. However, 90 % are diagnosed by the age of 7, and virtually all cases are diagnosed by the age of 20 [23].

Five percent of NF-1 patients have segmental NF-1, which involves one segment of the body, such as a single limb [24]. Segmental NF-1 may present with café au lait spots along one limb or one side of the body, and Lisch nodules in the ipsilateral eye [25, 26]. Another phenotype includes multiple localised cutaneous neurofibromas in the absence of all other findings [27]. Cases of segmental NF-1 occur due to genetic mosaicism, with mutations in the NF-1 gene occurring in the developing embryo after fertilisation [26]. Such mutations cannot be transmitted to offspring, and, if the gonadal progenitors are spared, the transmission risk is virtually non-existent. Depending on the fraction of gonadal cells involved, however, the risk of transmission to the next generation can increase to 50 %. Malignancies have been reported, but are exceedingly rare in segmental NF-1 [28].

The Committee on Genetics of the American Academy of Pediatrics (AAP) has published guidelines for the screening and follow-up evaluations of confirmed or presumed NF-1 [29]. The cornerstone of this follow-up is an annual examination, which should include thorough evaluations of the skin, eyes and nervous system and genetic counselling for parents at the initial consultation. Imaging, particularly magnetic resonance imaging (MRI) and spinal radiographs, as well as neuropsychological and developmental testing are indicated when children are diagnosed before they are 5 years old, or when children develop neurological deficits, vision loss, endocrinopathy, scoliosis, back pain, tract signs, impaired motor skills, or learning, speech or social difficulties [29].

Intellectual impairment affects up to half of all NF-1 patients [11, 12, 14, 20, 3033], ranging from speech disorders to mental retardation to learning difficulties [12, 14, 3336]. The latter occurs commonly, having been identified in one third to one half of all NF-1 patients [35, 37]. While the etiology of these impairments may be due to multiple factors, including seizures [38], it is unclear if they correlate with the number and size of focal hyper-intensities present on MRI [33, 36, 39]. These foci, known as unidentified bright objects (UBOs), are the most commonly visualised lesion seen on MRI in patients with NF-1 [32]. They present with increased signal on T2-weighted images in the absence of any mass effect and are accompanied by changes on T1-weighted images and contrast enhancement. They also exhibit a different appearance on MR spectroscopy compared with normal brain tissue and neoplasms. Although their clinical relevance remains unknown, UBOs are considered benign since a significant proportion regress spontaneously over time [32, 37]. More extensive work-up, however, is required for lesions that are diagnosed in older patients, correlate with neurological symptoms or have an atypical appearance.

Optic–hypothalamic gliomas are detectable in approximately 15 % of NF-1 patients and comprise the second most common imaging abnormality seen in NF-1. These lesions are classified into three categories, depending on the extent of tumour involvement [40]: lesions that only involve one or both optic nerves, sometimes representing a low-grade glioma or hyperplasia of the optic nerve sheath; lesions that also involve the optic chiasm, usually as globular thickening; and lesions that extend into the hypothalamus. Vision loss is characteristic of all these lesions [41] and usually progresses slowly [42, 43]. Other ocular signs include exophthalmos, optic nerve atrophy, painless ocular proptosis, papilledema, nystagmus, strabismus and conjunctival redness [41]. Optic nerve atrophy, papilledema, nystagmus and visual field defects are classically associated with lesions in the chiasm [43], although the visual field defects may be incongruous because the tumour may infiltrate the chiasm in an irregular manner. Lesions involving the hypothalamus may present with hemiparesis or ataxia, hydrocephalus (due to obstruction of the foramina of Monroe), endocrinopathy (especially growth hormone deficiency and precocious puberty) or diencephalic syndrome. The latter is characterised by nystagmus, profound emaciation, hyperkinesis and the subtle deterioration of muscles, and level of alertness [44].

The management of optic–hypothalamic gliomas is controversial, especially given the absence of guidelines [37]. Evidence suggests that these lesions are much more indolent and slow-growing in infants and children with NF-1, compared with patients without NF-1 [41, 45] although blindness is irreparable once vision is lost. As infants and toddlers may not complain of symptoms until their visual acuity has decreased significantly, screening this susceptible population is important. Surgery is reserved for patients with severe vision loss from a lesion that involves only a single optic nerve. It is contraindicated in patients with bilateral lesions or lesions involving the optic chiasm, because surgical resection usually results in total blindness. Radiation therapy, although effective, is contraindicated in young children because of the resulting cognitive and endocrine dysfunction [46, 47], increased risk of intracranial malignancies, and vasculopathies such as moyamoya disease. Chemotherapy using multi-drug combinations may be an effective alternative, especially in children [41].

Hemispheric and cerebellar gliomas are less common in NF-1 than lesions involving the optic tract [21] and are either benign or exhibit features of low-grade malignancy [21]. Unlike UBOs, hemispheric and cerebellar gliomas exhibit a mass effect on T2-weighted images, with increased signal on T1-weighted images. As most lesions are amenable to resection, surgery is the initial management for these gliomas. If total excision of a low-grade lesion is achieved, the patient may be monitored closely. Incomplete excisions require adjuvant chemotherapy and radiation therapy [21]. Younger children, however, exclusively receive the former because of the side effects of radiation.

Brainstem gliomas comprise a heterogeneous group of lesions, with at least three main subtypes: diffuse areas of brainstem enlargement; focal enhancing nodules with or without cystic areas; and peri-aqueductal gliomas. Most do not require treatment, although surveillance by MRI is indicated until they are confirmed to be indolent. Brainstem gliomas enhance in T2-weighted images, and exhibit a mass effect. They are more indolent than non-neurofibromatosis-associated brainstem gliomas, and some lesions may even regress spontaneously [48]. Adjuvant chemotherapy and radiation therapy is required for a minority of patients whose lesions enlarge progressively. Rarely, brainstem gliomas may progress to more malignant forms of astrocytoma, including glioblastoma [6, 21, 48].

Focal enhancing nodules, with or without cystic areas, are believed to represent pilocytic astrocytomas, and are generally indolent, although their course in the brainstem may be unpredictable. Consequently, ongoing surveillance is necessary when following these nodules. Small, focal intrinsic lesions may enlarge and regress spontaneously. Exophytic tumours are often more aggressive and require treatment. Periaqueductal gliomas represent low-grade gliomas or glial hamartomas and typically follow an indolent course. They manifest with late-onset aqueductal stenosis, leading to hydrocephalus that may require a shunt. Surgical resection is generally not required for brainstem gliomas seen in NF-1 [6, 21, 48].

Neurofibromas are the hallmark lesion of NF-1, occurring in peripheral nerves and paraspinal regions [14, 20, 29]. They are composed of Schwann cells, fibroblasts, mast cells, axons, and abundant extracellular matrix, with both myelinated and unmyelinated zones [14]. MRI is the best technique to visualise neurofibromas, facilitating the assessment of each lesion’s size, pattern of growth and proximity to adjacent structures. Neurofibromas may be abundantly distributed throughout the body, and may be highly variable in size, although there is no correlation between the number of cutaneous or external lesions and the number of lesions in deeper tissues.

Neurofibromas can be categorised into fusiform and plexiform lesions, based on their growth pattern. Fusiform lesions are discrete lesions that involve a well-circumscribed portion of a single nerve, and they are usually easy to resect. Plexiform lesions are often found in nerve trunks, extend over long distances, and invade nerve tissues diffusely [8]. They are highly vascular tumours that induce diffuse hypertrophy of the adjacent connective tissue. As they are so invasive, plexiform lesions are virtually impossible to resect without causing major neurological deficits [8, 17, 20]. Consequently, resection is reserved for malignant tumours, or lesions that cause intractable pain or severe disfigurement. Several chemotherapy protocols are also being tested for patients with progressive neurological impairment. At least 2 % of NF patients experience malignant transformation of a neurofibroma into a neurofibrosarcoma [49], while the rate may be as high as 10 % for plexiform neurofibromas [17].

Neurofibromas can also be categorised by their anatomic location, into subcutaneous, peripheral nerve, plexus, paraspinal, craniofacial, and visceral lesions. Subcutaneous neurofibromas can be fusiform or plexiform [17] and are associated with an overall increased risk of mortality [8, 50]. Both types of subcutaneous lesion, but especially plexiform lesions tend to recur after resection. Moreover, there is the risk of neurological injury and significant haemorrhage with resection, the latter potentially catastrophic in areas like the face where a tourniquet cannot be applied to stop bleeding [51]. As such, surgical removal of subcutaneous neurofibromas is reserved for lesions that are cosmetically intolerable, cause intractable pain, grow rapidly or undergo malignant change [17]. Peripheral nerve neurofibromas may be asymptomatic, but can produce pain, paresthesias or other neurological sequelae if they irritate the involved nerve [52]. As they tend to be fusiform, resection may be feasible without resulting in major neurological deficits, with the size of lesion and the presence of malignant transformation major predictors of surgical outcome [52]. As long as the nerve fascicles are minimally involved, lesions may be successfully resected by starting at the proximal and distal poles of the tumour, and identifying all the fascicles from which the tumour arises. En bloc resection often results in major deficits if all the fascicles are involved; therefore, subtotal resection often is performed. Stringent surveillance is necessary in the latter cases, because approximately 15 % undergo malignant transformation [13].

Paraspinal neurofibromas are usually fusiform or nodular lesions that involve nerve roots as they enter the spinal canal. They occasionally involve multiple nerve roots, and cases in which all the nerve roots were involved have been reported. Generally, these lesions do not grow, or they grow very slowly. Monitoring is often all that is required, but there is considerable debate as to whether routine imaging is required in addition to clinical examination alone [53]. Since some lesions cause progressive spinal encroachment, however, surgical resection is indicated for these cases [54]. Moreover, these lesions, especially if plexiform, are considerably more likely to undergo malignant transformation than lesions on the skin [55]. When resection is necessary, various surgical approaches are used, depending on the spinal level and the relationship of the tumour to surrounding paraspinal anatomy, to access and resect these tumours [56].

Craniofacial neurofibromas are typically plexiform lesions that involve peripheral nerves of the face [5762]. The lesions can grow to large dimensions, resulting in grotesque cosmetic deformity and/or facial paralysis [58, 61]. Many involve the orbit, where they can be extremely invasive and destructive, and even extend intracranially [60]. Resection of orbital lesions often requires enucleation. Stable lesions are monitored closely. Recently, facial transplantation has been used in the treatment of particularly disfigured facial neurofibromas [63].

Lastly, visceral neurofibromas also tend to be plexiform and, consequently, may not be resectable. They may occur in almost any tissue (such as the bladder or the gastrointestinal tract), and related symptoms depend on their location and size [6466]. Malignant transformation is a recognised, long-term risk for visceral neurofibromas.

Neurofibromatosis type 2

Much less common than NF-1, NF-2 has an estimated prevalence of one in 25,000 to 50,000 [11, 15, 18]. Though it shares a common name with NF-1, NF-2 is an entirely different clinical entity in terms of its underlying cause, presentation, characteristic lesion, and course [67, 68]. In fact, the somewhat elusive nature of both NF-1 and NF-2—given that lesions may be small and subclinical, and family history may be unobtainable—make it difficult to estimate the true prevalence of these disorders. Typically, NF-2 is caused by a mutation affecting chromosome 22q12 and the gene product merlin (a moesin-, erzin- and radixin-like protein), which is also called schwannomin. Merlin encodes for a polypeptide that may affect cell growth and motility. Interestingly, it is a tumour inhibitor that is conspicuously absent in many brain tumours [69, 70]. In addition, the same chromosomal abnormality is found in spontaneous spinal Schwannomas, suggesting that a single location causes Schwann cell tumour growth [71, 72].

Clinically, NF-2 is a combination of features that always entails at least one vestibular Schwannoma, in addition to a variety of other tumours (such as neurofibromas, meningiomas, gliomas and neurilemomas), juvenile posterior sub-capsular cataracts, and, occasionally, café au lait spots. Like NF-1, it is diagnosed using NIH Consensus Criteria, initially proposed in 1988 [7], and subsequently modified in 1997 [2]. These consensus criteria are listed in Table 3.

The diagnosis of NF-2 may be delayed, given that vestibular Schwannomas may present unilaterally, with the other side only becoming affected considerably later. Roughly ten percent of individuals with a unilateral vestibular Schwannoma will ultimately be diagnosed as having NF-2 [73]. Similarly, juvenile cortical cataracts may predate the confirmation of any other lesions. The AAP Committee on Genetics has published guidelines for the baseline screening and follow-up evaluations of confirmed or presumed NF-2 [29].

As with NF-1, NF-2 might also present in a segmental form, although it is less well defined. Segmental NF-2 has been defined as multiple discrete neurilemomas in peripheral nerves within an extremity, without any central features of NF-2. In general, segmental NF-2 is less problematic than systemic disease [74], although some cases of segmental NF-2 can be extremely disfiguring, with reports of malignancies in the literature [28].

The most common neurological lesions seen in NF-2 are vestibular and non-vestibular cranial nerve neurolemomas, intracranial meningiomas, meningioangiomatosis, intraparenchymal gliomas, extracranial neurolemomas, meningiomas and neurofibromas [69].

Vestibular Schwannomas occur in roughly 95 % of patients with NF-2 [19, 31], and are considered diagnostic for NF-2 when they are present bilaterally. Typically, these lesions occur in late adolescence or adulthood, but they may also be observed in children. NF-2 should be suspected in any patient with a posterior cataract and multiple spinal cord or peripheral nerve tumours, and an MRI of the head is indicated to rule out the disease. Symptomatic tumours during childhood may also suggest a more aggressive course. If a vestibular Schwannoma is present, regular hearing tests are indicated, because of the risk of developing progressive bilateral hearing loss.

The primary goals of treatment in NF-2 are to preserve hearing, and to prevent brainstem dysfunction or damage secondary to compression by tumour [7580]. The latter is an urgent indication for surgical resection. Debate arises, however, if the vestibular Schwannoma is small and not compressing the brainstem. Some recommend early surgery to remove small tumours and consequently lower the risk of post-operative hearing loss; others recommend a wait-and-see approach, with the risk that surgery may be required eventually and increase the risk of post-operative hearing loss or facial nerve defects [76, 7880]. The management of vestibular Schwannomas, both within and outside of neurofibromatosis, is changing as enhanced imaging technologies facilitate the earlier detection of lesions, more targeted therapies are available to deliver focussed treatments to neurofibromas [77, 81], and the options for hearing restoration expand [80, 82].

Stereotactic surgery may also reduce the risk of deafness or delay the onset of deafness for years [83], by reducing tumour growth. However, the benefits of this approach still remain unproven. Another alternative approach is partial (sub-capsular) resection, leaving residual tumour still adherent to the seventh and eighth cranial nerves, to reduce the risk of injury to both [84]. The caveat, however, is that residual tumour may grow and become problematic at a later date. Finally, in recent years, gamma knife surgery has emerged as an option that may improve outcomes in some patients, and especially in those with large lesions for which traditional surgery carries heightened risks of hearing loss and facial nerve dysfunction [85, 86]. Irrespective of the treatment plan, especially in patients with bilateral lesions, learning sign language or lip-reading may ultimately prove beneficial. The placement of a cochlear or auditory brainstem implant may also preserve some hearing, even in patients who have undergone bilateral resections [80, 82].

Intracranial meningiomas are found in approximately 50 % of patients with NF-2 [15], and often are multiple in number. As with spontaneously arising meningiomas, they tend to be more aggressive in children compared with adults [87, 88], although surgery may be deferred if there are multiple lesions.

Meningioangiomatosis is a rare, benign, focal lesion of the leptomeninges and underlying cerebral cortex, which is characterised by leptomeningeal and meningovascular proliferation [89, 90]. Histologically, it appears as a hamartomatous proliferation of capillary-sized vessels, meningothelial cells, and fibroblasts within the cerebral cortex. It may also occur in the presence or absence of an adjacent meningioma. Although sporadic cases have been reported, meningioangiomatosis most often occurs in patients with NF-2. As with most neurofibromatosis lesions, they can be multifocal, especially in NF-2. Though they appear benign on histopathological sections, meningioangiomatosis can produce seizures [91], albeit more commonly in sporadic cases rather than NF-2-associated lesions [90]. Surgical resection may be indicated in the case of intractable seizures that are resistant to anti-epileptic therapy [91]. At least one case of sudden death, presumably secondary to a fatal seizure, has been reported involving a previously asymptomatic 13-year-old boy [92].

Non-vestibular cranial nerve neurolemomas are found in up to 50 % of NF-2 patients [93]. While they are benign like vestibular Schwannomas, neurolemomas have a more unpredictable course [78, 79]. Some grow very slowly, if at all, and cause no problems; others grow rapidly and are quite problematic, both due to damage of the involved nerve and because of their mass effect. Also, on occasion, malignant transformation of a non-vestibular neurolemoma has been reported [94]. As non-vestibular cranial nerve neurolemomas are frequently multifocal, surgery is often deferred unless they are unacceptably symptomatic, endangering function or growing rapidly [76, 77].

Spinal lesions are typically more common in NF-2 than in NF-1, while intracranial lesions are less common. Ependymomas are the most common malignancy in NF-2, whereas astrocytomas predominate in NF-1 [6, 15, 68]. As they are usually well-circumscribed, ependymomas are often surgically resectable. Their post-operative management, including the use of adjuvant therapy, is similar to treatments for intramedullary spinal tumours in non-neurofibromatosis patients [78, 79].

Benign intraspinal tumours of the nerve sheath or meningeal cells are also more common in NF-2 compared with the general population. In NF-2, however, multiple tumours can form at different spinal levels. Similar to intracranial nerve sheath or meningeal tumours, both neurilemomas and meningiomas are slow-growing, well-circumscribed tumours that displace, rather than invade, adjacent tissues [15, 95]. As the onset of neurological symptoms occurs when more than ninety percent of the spinal canal is occluded, patients can be asymptomatic for years. Once deficits start to appear, however, even minimal expansion can significantly reduce function. Surgical outcomes are better if these tumours are excised before the onset of neurological dysfunction [15, 67]. Management, therefore, consists of close monitoring with repeat neurological exams and MRI, and prompt surgical resection before the deterioration of neurological function. Intractable pain is another indication for the resection of neurilemomas and meningiomas; because they generally involve a single nerve fascicle, surgery can be performed without significant neurological deterioration [67, 95]. Neurofibromas occur less commonly in NF-2 compared with NF-1 [67, 95]. Given that neurofibromas, especially plexiform lesions, are not as well-encapsulated as neurilemomas [96], surgical resection is more difficult and leads to worse postoperative outcomes compared with the latter.

Tuberous sclerosis (Bourneville’s disease)

Tuberous sclerosis (TS), also called tuberous sclerosis complex (TSC) or Bourneville’s disease, is the second most common phakomatosis syndrome after NF-1. Like neurofibromatosis, it was initially described by von Recklinghausen in 1862, and affects between 1 in 6,000 to 1 in 30,000 people [9799]. However, estimates of incidence are somewhat unreliable because of marked variations in penetrance. TS was named after Désiré-Magloire Bourneville, a French physician who coined the term ‘sclerose tubereuse’, likening the cerebral lesions he detected at autopsy to small potatoes. The physical manifestations of TS are due to the formation of hamartia (malformed tissue and like cortical tubers), hamartomas (including facial angiofibroma and subependymal nodules) and, very rarely, malignant hamartoblastomas. These various lesions can cause seizures, developmental delay, and behavioural problems in patients with TS. Although the classical triad of epilepsy, low intelligence, and skin lesions is described, TS is now recognised to involve numerous organs, including the kidneys, lungs, heart and eyes. Many cases of ‘low intelligence’ have since been recognised as learning disabilities, autism, and pervasive development disorders [100, 101].

TS is AD, but up to 60 % of cases arise from spontaneous mutations [98, 99, 102]. Two tumour-suppressor genes, TSC 1 and 2, are responsible for TS. Roughly 80 to 90 % of mutations involve TSC-2, while only 10 to 20 % of mutations involve TSC-1 [98, 102]. The genetic locus for TSC-1 is chromosome 9q34, and the TSC-1 gene product is called hamartin. The genetic locus for TSC-2 is chromosome 16p13.3, and the TSC-2 gene product is called tuberin [98, 102]. Both hamartin and tuberin appear to have roles in cell differentiation, proliferation and migration. The disorder affects cellular differentiation, proliferation and migration during early development, leading to various diffuse hamartomas and neoplastic lesions affecting virtually every organ [102, 103]. It can present at any age, but most commonly appears during childhood or early adolescence [102].

TS has numerous non-neurological manifestations. Skin lesions are the most common and recognisable feature of the syndrome [102, 104, 105]. These lesions are numerous and varied, and include ash-leaf spots, facial angiofibromas, lumbosacral angiofibromas (Shagreen patches), café au lait spots, periungual fibromas (Koenen tumours), forehead plaques, skin tags (molluscum fibrosum pendulum), confetti macules and poliosis. Ash-leaf spots are the most characteristic lesion, seen in 97 % of TS patients (compared with only 5 % of the general population) and are usually the only visible sign of TS at birth. The spots are hypomelanotic macules up to several centimeters in length, and found on the trunk or buttocks. The axis of the leaf tends to be perpendicular to the axis of the spine. As they are hypo-pigmented, a Wood’s lamp may be required to see ash-leaf spots in fair-skinned individuals.

Facial angiofibromas, or adenoma sebaceum, are red to dark-brown macules or papules that are usually clustered around the nose and cheeks in a butterfly pattern. They may be mistaken for acne in teenagers, but, unlike the latter, angiofibromas do not contain purulent material and cannot be drained. Histologically, they comprised blood vessels and fibrous tissue. They can be removed by dermabrasion or laser treatments for cosmesis [106]. Lumbosacral angiofibromas, also called Shagreen patches, are areas of thick leathery skin that are dimpled like an orange peel and usually found on the lower back or nape of the neck. They, like ash-leaf spots, are of variable size, but can grow quite large. Periungual fibromas (Koenen tumours) occur rarely in childhood, but become more common by middle age. They are small, fleshy tumours that grow around and under the toenails or fingernails, and may require surgical excision if they cause symptoms such as bleeding. Forehead plaques are raised, discoloured areas on the forehead. Skin tags (molluscum fibrosum pendulum) are non-specific, as they may also occur in healthy individuals. Confetti macules are clusters of small, punctate, hypopigmented macules that look like confetti. They are specific for TS, and are distributed symmetrically on the limbs. Poliosis is a tuft or patch of white hair on the patient’s scalp or eyelids.

The most common cardiac manifestation of tuberous sclerosis is the intraventricular rhabdomyoma, a benign tumour of striated muscle [107]. Close to half of these tumours in children are associated with tuberous sclerosis [108]. In the appropriate clinical context (positive family history of TS), diagnosis of this tumour is strongly suggestive of TS [109]. Multiple rhabdomyomas affecting multiple chambers are common, although most tumours are found within the ventricles [110]. The incidence of these tumours in tuberous sclerosis ranges from up to 90 % in newborns to as low as twenty percent in adults; this is because they generally grow during the second and third trimesters and spontaneously regress after birth, sometimes disappearing completely [111]. Most tumours are detected on prenatal ultrasound after twenty weeks of gestation, and rarely manifest symptoms [110, 111], such as a heart murmur, valvular abnormalities (because the tumour impedes proper functioning of the valve), and arrhythmia. Most of these cardiac complications occur before the age of 1 year [110, 111].

Renal involvement is the second most common cause of morbidity and mortality in TS after neurological disease, often only manifesting during and after adolescence [112]. Numerous types of renal lesions exist, including angiomyolipomas, renal cysts, renal cell carcinomas and oncocytomas. Renal cell carcinomas can affect up to 3 % of TS patients [113], significantly higher than the general population; but most TS-related renal cell cancers are diagnosed in adulthood [114]. Oncocytomas are benign adenomatous hamartomas that are very uncommon in TS patients. Angiomyolipomas, on the other hand, are found in 45 to 70 % of TS patients [113] (compared with 0.3 % of the general population) and are composed of vascular tissue, smooth muscle and fat [114, 115]. Patients may have multiple, bilateral angiomyolipomas, and may complain of hematuria. Although they are benign, lesions larger than 4 cm in diameter are at risk for catastrophic haemorrhage, either spontaneously or with minimal trauma [116]: one near-fatal case has been reported in a child as young as 12 years of age [117]. Renal cysts can be found in twenty to thirty percent of TS patients, but they are usually not clinically concerning, although life-threatening haemorrhage can occur in a small percentage of cases [118]. Moreover, in the two percent of TS patients with AD polycystic kidney disease, renal failure is a common sequelae [119].

Lung involvement in tuberous sclerosis usually affects women in their third and fourth decades of life, implicating hormonal influences. Lung disease may also manifest during adolescence, and therefore must be evaluated even in young patients [112]. The classic lesion is lymphangioleiomyomatosis (LAM), which ultimately manifests itself in roughly thirty percent of female TS patients [120]. LAM consists of progressive replacement of lung parenchyma with multiple cysts, generally measuring two to twenty millimetres in diameter, with equal involvement of all lobes [121]. These cysts are formed by hyperplastic smooth muscle. Recent genetic analysis suggests that this proliferative bronchiolar smooth muscle represents monoclonal metastases, potentially from a coexisting renal angiomyolipoma [120, 122]. There have been several documented cases of TSC-related LAM recurring following lung transplantation [123], confirming the concept of an ‘external’ source. The overall prevalence of LAM in patients with TS has been estimated as two to forty percent [113], rendering it vastly more common than the prevalence of 0.0001 % observed in the general population. Nonetheless, LAM is often misdiagnosed as asthma, emphysema, or some other pulmonary disease because plain radiographs reveal the superimposition of cysts that produce the reticulo-nodular pattern seen in interstitial lung disease. High-resolution CT is more specific for the diagnosis and better at assessing the degree of pulmonary involvement. LAM is slowly progressive, and, interestingly, it tends to be associated with less severe TS, with fewer seizures and less intellectual impairment. As LAM is seen almost exclusively in menopausal females, it has been presumed to be highly dependent on estrogen, such that anti-estrogen therapies have been promoted for its treatment [122], despite their lack of proven effectiveness [124]. As the mutations in the TSC1 and TSC2 tumor suppressor genes result in hyperactivation of the mammalian target of rapamycin (mTOR) signaling pathway and subsequently cause abnormalities in numerous cell processes, mTOR inhibitors like sirolimus and everolimus have been tested as targeted therapy for a variety of TS-related complications, particularly LAM [125]. To date, however, no mTOR inhibitor has been proven effective or approved for the treatment of LAM [124]. Fortunately, the course of LAM is generally slowly progressive, with roughly 90 % of patients with LAM remaining alive 10 years after diagnosis [126].

Retinal phakomas, gray or yellow retinal plaques that may be single or multiple, are the classic ocular lesion seen in TS, being present in up to 87 % of patients [105]. They rarely affect vision and, hence, usually do not warrant any treatment.

Neurological involvement is the most common cause of morbidity and mortality in TS, and the most common cause of death in patients under 30 years old [103, 104]. Various forms of intellectual, social, and behavioural impairment are caused by cortical tubers, including autism (in twenty to 60 % of patients), learning disabilities (in about 50 % of patients) and mental retardation and self-mutilation (in 10 % of patients for each) [113, 127]. These effects are often progressive, with many neonates appearing neurologically normal [127]. Problems stem not only from cortical tubers, however, but also from other lesions that include subependymal nodules and subependymal giant cell astrocytomas (SEGA). Various imaging techniques, including contrast and non-contrast CT, and MRI are used to visualise TS-associated CNS lesions. Gadolinium-enhanced MRI is the most sensitive radiographic study. Electroencephalograms are useful to document and identify the foci of seizures. The diagnosis of tuberous sclerosis as a syndrome, however, is made using previously-established criteria in Table 4 [109]. Management is essentially supportive, but includes regular monitoring that consists of MR imaging of the head every 1–2 years to assess tumour growth, mass effect and CSF outflow [105, 127].
Table 4

Modified NIH consensus criteria for tuberous sclerosis [112]

Major features

 1. Facial angiofibromas or forehead plaque

 2. Nontraumatic ungual or periungual fibroma

 3. Hypomelanotic macules (more than three)

 4. Shagreen patch (connective tissue nevus)

 5. Multiple retinal nodular hamartomas

 6. Cortical tuber

 7. Subependymal nodule

 8. Subependymal giant cell astrocytoma

 9. Cardiac rhabdomyoma, single or multiple

 10. Lymphangiomyomatosis

 11. Renal angiomyolipoma

Minor features

 1. Multiple randomly distributed pits in dental enamel

 2. Hamartomatous rectal polyps

 3. Bone cysts

 4. Cerebral white matter migration lines

 5. Gingival fibromas

 6. Nonrenal hamartoma

 7. Retinal achromic patch

 8. ‘Confetti’ skin lesions

 9. Multiple renal cysts

Definite TSC, either two major features or one major feature with two minor features. Probable TSC, one major feature and one minor feature. Possible TSC, either one major feature or two or more minor features

Of the various tumours seen in TS, cortical tubers are the most common, identified in more than eighty percent of patients [127, 128]. They are formed from aberrant neuronal migration during the development of the cerebral cortices, and primarily affect the frontal and parietal lobes. Most patients with cortical tubers present with intractable, lifelong seizures, almost universally if the lesions are large in size [129]. Many such patients present with intractable seizures before they are 1 year old, which is an ominous prognostic sign [130]. Unfortunately, resection of these tumours generally is not effective for seizure control. Nonetheless, resection may be warranted if the lesions cause mass effects. Some have also advocated for earlier and more aggressive medical management of seizures to prevent or limit the seizure-induced neurocognitive decline [130], given that diagnosis of these lesions can be made earlier due to advanced imaging techniques.

Subependymal nodules are the hallmark lesion for TS on computed tomography (CT) scans [127]. They may be calcified or non-calcified, and may number from one to as many as twenty lesions. On CT, they can appear like candle-wax drippings. They are composed of abnormal, swollen glial cells and bizarre multinucleated cells that are indeterminate, in terms of glial versus neuronal origin. There is no interposed neural tissue. These nodules have a tendency to calcify as the patient ages. More ominously, however, any nodule that markedly enhances and enlarges over time should be considered suspicious for transformation into a SEGA.

SEGA develop in five to fifteen percent of TS patients [131], typically developing near the Foramen of Monro, where they frequently cause obstructive hydrocephalus. SEGAs are slow-growing and rarely undergo malignant transformation, but are problematic because of their location and relative inaccessibility for resection. As for cortical tubers, complete surgical resection of SEGA often is difficult because these intra-ventricular tumours originate in the caudate nucleus and septum pellucidum. In addition, though relatively avascular, the tumours invade surrounding tissue and can be quite large. Incomplete resection, meanwhile, places the patient at risk for local recurrence and re-growth. Despite this, partial or complete surgical resection of these lesions is still widely considered to be the standard of care [130, 132], especially for the relief of increased intracranial pressure, focal deficits, and obstructive hydrocephalus. This being said, a recent observational study noted that the post-operative prevalence of seizures, hydrocephalus, headaches, and stroke or hemiparesis increased significantly compared with preoperative time period [133].

Overall, the long-term prognosis among TS patients is poor, with the rate of mortality markedly increased compared with the age-matched general population. The primary cause of death is renal disease, and the primary cause of morbidity is CNS disease causing seizures, cognitive impairment, and obstructive hydrocephalus. Recently, the mTOR has demonstrated promising results in recent clinical trials. In both case reports and clinical trials, treatment with mTOR inhibitors has resulted in significant reductions in SEGA volume and improved or resolved ventriculomegaly [134, 135]. In the USA, this led to the 2010 Federal Drug Administration approval of everolimus for the treatment of SEGA in tuberous sclerosis patients who are not candidates for surgery [136]. Other CNS tumours, including astrocytomas, have been described with increased frequency in TS patients [137]. For such patients, the immunosuppressant drug rapamycin has demonstrated some promise as a means to shrink tumours [137].

Von Hippel–Lindau disease

VHL disease is named after the German ophthalmologist Eugen Von Hippel, who discovered retinal angiomas in 1904, and the Swedish pathologist Arvid Lindau, who noted the association between these angiomas and CNS tumours (which Lindau described as ‘angiomatosis of the central nervous system’) in 1927. Also known as retinocerebellar angiomatosis, the incidence of VHL has been estimated as roughly 1 in 36,000 live births [6], though it usually does not manifest in infancy. The average age at first presentation is 26 years, and the average age at diagnosis is 31 years [6]. Nonetheless, paediatric cases are not uncommon, and certainly seen by the paediatric neurosurgeon as a challenging disorder to treat.

VHL disease is AD, with 97–99 % of familial cases and only 1–3 % occurring as the result of spontaneous mutations. It is associated with inactivation of the tumour-suppressor gene VHL, which is found on chromosome 3p25 [6, 138]. Decreased levels of the VHL protein (pVHL), which is important in a critical pathway helping cells to adapt to hypoxic stress, lead to over-expression of a hypoxia-inducible transcription factor 1 which, in turn, results in increased cell proliferation, and the over-expression of several growth factors. Ultimately, the disease manifests as multiple, multi-systemic benign and malignant tumours, which are sometimes bilateral [139]. These tumours include haemangioblastomas of the cerebellum, spine, brainstem and retina (the most common tumour identified), renal clear cell carcinomas, pheochromocytomas, pancreatic and renal cysts, endolymphatic sac tumours of the petrous bone at the cerebellopontine angle [140], papillary cystadenomas of the epididymis or broad ligament and haemanigiomas of the adrenal glands, liver and lungs. The diagnosis is made clinically using established criteria (Table 5).
Table 5

Classification criteria for von Hippel–Lindau disease

Two or more haemangioblastomas, or

1 haemangioblastoma plus some visceral lesion, like pheochromocytoma, pancreatic or renal cyst or renal cell carcinoma; or

1 retinal or CNS haemangioblastomas or some other visceral lesion plus a positive family history of CNS or visceral manifestations of the disease

VHL is subdivided into various clinical subtypes, based upon genotype-phenotype correlations within families. In type 1, the VHL gene has deletion or nonsense mutations (resulting in the absence of protein), and members of this group manifest hemangioblastomas and clear cell renal carcinomas [139]. In type 2, the VHL gene has missense mutations (creating aberrant VHL protein): type 2A are at risk of hemangioblastomas and pheochromocytomas [141]; type 2B patients are at risk of hemangioblastomas and pheochromocytomas and a higher risk of clear cell renal carcinomas [141]; and type 2C patients are only at risk for pheochromocytoma [141].

Central nervous system manifestations are highly prevalent in VHL [142, 143], with CNS haemangioblastomas occurring in sixty to eighty percent of patients. Moreover, they are more likely to be multiple and present at an earlier age than when they occur sporadically, being a presenting feature in roughly sixty percent of VHL patients [144]. These lesions may occur anywhere along the cranioaxial axis, but only one percent of these tumours are supratentorial [144]. The site of lesion determines the symptoms with which the patient presents. The cerebellum and brainstem are the most common sites of haemangioblastomas in VHL [144]: patients present with headaches, vomiting, lethargy, dysmetria, ataxia, papilloedema, polycythemia from tumour production of erythropoietin and/or enlarging cysts that may cause brainstem compression (solid tumours generally do not cause such compression in VHL).

Conversely, spinal haemangioblastomas present with neck, chest and back pain, sensory losses, and various signs and symptoms of cord compression depending upon tumour location [144]. Patients with cervical haemangioblastomas typically present with neck pain, signs and symptoms of cord compression and, sometimes, severe infratentorial and supratentorial subarachnoid haemorrhage. Finally, retinal haemangioblastomas present with vision loss secondary to haemorrhage, exudation and retinal detachment [145].

Haemangioblastomas are typically cystic and enhance with contrast. Contrast-enhanced T1-weighted MRI is considered the diagnostic test of choice to detect and monitor CNS lesions. The optimum treatment of CNS haemangioblastomas is complete surgical excision when possible, since residual tumour may cause severe bleeding [145]. Pre-surgical endovascular embolisation may reduce operative complications and morbidity, while small asymptomatic lesions may be monitored with repeat MRIs. Alternative treatments include gamma knife radiosurgery [146, 147], which appears to be effective for small to medium-sized nodular lesions, but not cystic lesions which can bleed. Linear-accelerator-based cranial stereotactic radiation therapy has also proven effective with some tumours. And multiple agents that target gene products downstream from pVHL and HIF-1, such as platelet-derived growth factor, have recently become available and are being tested [6]. Despite these surgical advances, however, there is a significant mortality by the age of 50 [148, 149], typically due to metastatic renal cell carcinoma and/or cerebellar haemangioblastomas [148, 149].

Basal cell nevus syndrome (Gorlin–Goltz syndrome)

Basal cell nevus syndrome (BCNS) is an AD disease. The causative gene PTCH1 is located in chromosome 9q31 in about 85 % of cases [150]. The PTCH1 gene product is a trans-membrane receptor that binds to and regulates a protein called Sonic Hedgehog Homolog (SHH), one of three proteins in the mammalian signalling pathway family called ‘hedgehog’,and one which plays a key role in the regulation of organ development in vertebrates, including the growth of fingers and toes and the organisation of the central nervous system [150]. It also controls cell division in adult stem cells and has been implicated in oncogenesis. Mutations in the PTCH1 gene result in uncontrolled SHH activation [150].

This rare condition, which affects roughly one in 60, 000 live births [6], is characterised by multiple basal cell cancers, often presenting in adolescence [18, 151]. BCNS also presents with a wide range of non-neurological tumours (Table 6), including melanomas, leukaemias, lymphomas, lung and breast cancers, medulloblastomas and meningiomas [152]. Odontogenic keratocysts (jaw cysts) are often the first sign of the syndrome, commonly manifesting in early childhood. These cysts lined with keratinised epithelium originate in the dental lamina, locally erode through teeth, and can cause significant dental displacement and loss if they are not completely excised. Medulloblastomas are the most common CNS tumour, and they too present early, in roughly 3 to 5 % of children with BCNS [153]. The medulloblastomas seen in BCNS tend to occur earlier than in sporadic cases and often are histologically distinct from classic medulloblastomas, being defined by the presence of several prominent nodules, or ‘pale islands’, of tumour. These areas are of lower cellularity, which are reticulin–free, exhibit nuclear uniformity, and are in a background of collagen-rich, highly-proliferative tumour. Desmoplastic medulloblastomas also tend to be more discrete than the classic variety and often are located in the cerebellar hemispheres. Given their location in the posterior fossa and the fact that they can become quite large, hydrocephalus is a common complication of medulloblastomas in BCNS, and may be a presenting sign.
Table 6

Clinical characteristics of basal cell nevus syndrome

Multiple nevoid basal cell cancers

 Odontogenic keratocysts (jaw cysts)

Other bone cysts

Calcified falx cerebri/dural calcifications

Pitting of plantar or palmar surfaces

Congenital skeletal abnormalities including:


 High scapulae (Sprengel’s deformity)

 Frontal bossing (that also may involve parietal area)

Synostosis of various cranial sutures

Bifid ribs

Cleft lip and/or palate

Eye disorders




Non-CNS benign and malignant tumours


 Chronic lymphocytic leukaemia

 Non-Hodgkin’s lymphoma

 Lung cancer

 Breast cancer

 Myocardial fibromas

 Lyomesenteric cysts

 Ovarian fibromas and dermoids


CNS tumours



Like virtually all the familial cancer syndromes, BCNS is diagnosed using established clinical criteria (Table 7) [151]. Advanced imaging, like MRI or CT, is necessary to assess CNS tumours. Jaw cysts, other bony cysts, and medulloblastomas may present before nevi on the skin [154]. Interestingly, a recently-diagnosed 10-year old child was found to have café au lait spots [151], emphasising the considerable phenotypic variability evident in virtually all of the phakomatosis syndromes [6].
Table 7

2003 Diagnostic Criteria for Basal Cell Nevus Syndrome

Major criteria

 1. Calcification of the falx cerebri

 2. Bifid or fused ribs

 3. Jaw cysts

 4. Palmar and plantar pits

 5. First-degree relatives with the same syndrome

Minor criteria

 1. Medulloblastoma

 2. Ovarian fibroma

 3. Macrocephaly

 4. Congenital facial or skeletal abnormalities like cleft lip or palate, hypertelorism, frontal bossing, syndactyly and radiological bone abnormalities like bridging of the sella turcica

The diagnosis is established in the presence of two or more major criteria or one major criterion plus two or more minor criteria

Treatment of BCNS is largely supportive and as for others with similar tumours, of the CNS and elsewhere. However, it is complicated by current failures to accurately predict the course of medulloblastomas, in general and in BCNS, since no histological grading system has yet been identified that accurately predicts prognosis. This is confounded further by concerns as to the potentially-increased risk of radiation-induced secondary malignancies, especially in children who have the potential for very long-term survival (such as the child who develops secondary osteosarcoma in the radiation field for a previously treated medulloblastoma). The general impression that medulloblastomas have a more indolent course in BCNS than otherwise has led to some to suggest that the dose of radiation can be reduced [155]. As with many of the familial cancer syndromes described in this paper, better understanding of the underlying genetics of cancers like medulloblastoma has led to more directed chemotherapies that hold some promise [156].

Other familial syndromes associated with paediatric CNS malignancies

Li–Fraumeni syndrome

LFS is another rare AD disease caused by a germ-line mutation of chromosome p53 in roughly 70 % of families in which the syndrome is diagnosed [157]. Patients exhibit an increased risk of variety of carcinomas and sarcomas, including premenopausal breast cancers, osteosarcomas, soft tissue sarcomas, acute leukaemia, cancer involving the adrenal cortex and primitive neuroectodermal tumours (PNET) like medulloblastoma. This increased predisposition to a wide variety of malignancies likely stems from deactivation of p53, which normally controls apoptosis and the repair of damaged DNA.

Patients present not only with a variety of cancers, but with cancers at a very early age. The mean age at presentation in LFS patients with brain tumours is about 25 years. The diagnosis of so-called classic LFS is made in any patient under 45 years of age who presents with a bone or soft-tissue sarcoma, plus one first-degree relative who presents with any cancer before age 45, plus one further first- or second-degree relative of the same lineage who has had any cancer before age 45, or a sarcoma at any age [158]. More recently, a related syndrome, called LFS, has been described, defined as a proband with any childhood tumour or any sarcoma, brain or adrenocortical tumour before 45 years of age, who has a first- or second-degree relative with any cancer before the age of 60 [159, 160]. Interestingly, whereas p53 germ-line mutations are found in seventy to eighty percent of families with classic LFS, they are only identified in 20 to 40 % of families with LFS-like syndrome [41]. The CHK2 checkpoint homolog gene, CHEK2, which is located on the long (q) arm of chromosome 22, also has been implicated in some families with classic LFS. Recently, mutation of another gene, which encodes for the breast cancer 2 (BRCA2) susceptibility protein, has been found with increased frequency in the non-classic syndrome [160]. It should be noted that p53 mutations are rare in sporadically-occurring medulloblastomas.

Overall, about ten percent of LFS patients will develop a glioma before the age of 45, and another five percent will develop a supratentorial PNET, such as a medulloblastoma or choroid plexus carcinoma [161]. Since LFS is so rare, no clinical trials document the optimum treatment; but it generally is agreed that incident tumours should be treated as for sporadic cases, albeit with increased vigilance for additional tumours, both within the proband patient due to the increased risk of second cancers [159], and within the family.

Familial Adenomatous Polyposis or Turcot Syndrome

A Canadian surgeon, Jacques Turcot, is credited with characterising Turcot syndrome, one of the several familial polyposis syndromes associated with autosomal recessive inheritance and the presence of multiple colonic adenomas and adenocarcinomas [162]. An additional feature of Turcot syndrome is its association with several different neuroepithelial tumours of the central nervous system, including astrocytomas, medulloblastomas, pineoblastomas, gangliogliomas and ependymomas.

Turcot syndrome has been categorised into types 1 and 2: type 1 is characterised by glioblastoma, absence of familial adenomatous polyposis, and hereditary non-polyposis-related colorectal carcinoma. Germ-line mutations in a few DNA mismatch repair genes– PMS2, MLH1 and MSH2– are associated with type-1 Turcot syndrome. Type-1 Turcot syndrome is also associated with café au lait spots [41]. Conversely, type-2 Turcot syndrome manifests with medulloblastomas and multiple adenomatous polyps that often undergo malignant transformation [163]. Medulloblastomas, glioblastomas and anaplastic astrocytomas are the most common CNS tumours observed in Turcot’s syndrome, accounting for 95 % of CNS tumours in these families [163]: the latter two tumours are inevitably fatal. Additionally, they tend to occur early, with medulloblastomas typically diagnosed in children less than 10 years old and gliomas in those under age 30 [41, 164166].

As such, and because some die of metastatic colon cancer that sometimes presents quite early in childhood or in the second decade of life, many die as adolescents or young adults. In one example, doctors in Pittsburgh reported the case of a girl who developed a medulloblastoma at the age of 5 years. Ten years later, she developed adenocarcinoma of the colon. Seven months after resection of a locally invasive colonic adenocarcinoma, she presented with a second primary CNS tumour, this time a glioblastoma multiforme [165], of which she died shortly thereafter.

Rubenstein–Taybi syndrome

Rubenstein–Taybi syndrome is an AD disorder associated with numerous anatomical/functional abnormalities that include abnormal facies, microcephaly, various dental problems, broad thumbs, big toes and moderate to severe intellectual impairment [167]. There also is an increased incidence of neuroepithelial tumours; in particular medulloblastomas, meningiomas and oligodendrogliomas [157], though other CNS tumours have been described [168]. A germ-line mutation in one allele of CRE-binding protein (CBP; a transcriptional co-activator for several c-AMP-regulated genes) has been implicated in many cases. CBP binds to the activated form of GLI, a transcription factor that is important in the regulation of the SHH that, as stated earlier, controls cell division in adult stem cells and has been implicated in oncogenesis. The GLI gene is downstream of the PTCH1 gene that is mutated in BCNS. To date, management tends to be supportive, focussing on the various deformities, disabilities and developmental issues that occur in these patients, with malignancies treated in the same way as patients without the syndrome [167]. Recent work in the field of epigenetic drugs, drugs that target specific gene mutations, may provide more directed therapies in the future [169].

Dyschondroplasias (Ollier’s disease and Maffucci syndrome)

Enchondromatosis, also called dyschondroplasia, is a hamartomatous proliferation of chondrocytes within bony metaphyses [170]. Enchondromatosis is often asymptomatic, and diagnosed as an incidental X-ray finding. Conversely, it can lead to significant deformities, reduced bone length [170, 171], and occasional pathologic fractures [171]. Moreover, enchondromas appear to have an association with malignancy. This includes both chondrosarcomas that result from sarcomatous transformation of the enchondromas themselves, and other histologically-distinct malignancies, including angiosarcomas, osteosarcomas, CNS tumours, ovarian tumours, and various leukemias [172]. This association with malignancy appears to be particularly true in diseases with multiple enchondromatoses, as in Ollier’s Disease and Maffucci’s Syndrome [173].

In 1881, an Italian pathologist named Angelo Maffucci first described a patient with enchondromatosis and venous hemangiomas on the skin [174], although others followed suit within a decade [175]. In 1889, a French surgeon, Louis Léopold Ollier, described a patient with enchondromatosis in the absence of any evident vascular abnormalities [176], and termed it Ollier’s Disease (OD). Both conditions are rare: Ollier’s Disease has a reported prevalence of roughly one in 100,000 [177], while fewer than two hundred cases of Maffucci’s Syndrome (MS) have been reported in medical literature [178]. It is unclear whether these two entities are clinically distinct or variations of the same disease [179], especially since some patients who were diagnosed with Ollier’s Disease were later discovered to have hemangiomas [180, 181].

In 1904, Boinet diagnosed a chondrosarcomatous lesion of the skull base in a 37-year old French gentleman with Maffucci’s syndrome [182]. Since that time, 45 additional cases of CNS malignancy (ranging from 6 to 58 years of age) have been reported in patients with OD or MS [172]. The combination of enchondromatosis and an intracranial malignancy is extremely rare, because of the rarity of the dyschondroplasias themselves [177, 178]. Patients with OD manifest their CNS neoplasm earlier than patients with MS [183]: in one study, patients with OD and malignancy were 10 years younger than their MS counterparts (25 versus 35 years), as were patients with OD and chondrosarcoma compared with patients with MS and chondrosarcoma [183]. Among those with non-sarcomatous neoplasms (NSNs), there were no differences in average age. Additionally, seven of twenty-four patients with OD were under 18 years old, compared with just two of twenty-two patients with MS. These findings suggest that underlying enchondromatosis is associated with an increased lifelong risk of CNS malignancy [184].

The increase in malignant potential associated with OD or MS is not known. Single enchondromas are associated with an elevated risk of malignant change. Altay et al. [185], for example, conducted an 18-year retrospective analysis of 627 cartilage-forming benign bone tumours and found that 32 patients had experienced malignant transformation, with fourteen of these patients initially having had a solitary osteochondroma, ten multiple osteochondromas, six a solitary enchondroma, one Ollier’s disease and one Maffucci’s syndrome. The single patient with OD had two chondrosarcomas, and the patient with multiple osteochondromas had three chondrosarcomas. The overall rate of malignant transformation for cartilage-originating tumours was 5.1 % (4.2 % for solitary osteochondromas, 9.2 % for multiple osteochondromas and 4.2 % for solitary enchondromas).

A variety of chromosomal abnormalities have been reported in isolated cases of OD or MS and chondrosarcoma [180]. These abnormalities include the interstitial deletion, del(1)(p11p31.2), as the only chromosomal abnormality identified in a low-grade chondrosarcoma in an OD patient [186]. In a tibial chondrosarcoma associated with OD, Bovée et al [187] identified a loss of heterozygosity on chromosome bands 13q14 and 9p21, and over-expression of the p53 gene. Chang et al. [188] identified identical male twins with OD who both developed astrocytomas within their cerebral cortex during their early 20s; and Robinson et al. [109] found evidence of mitogenic neurotransmitters within both enchondromas and soft tissue hemangiomas in a patient with Maffucci’s syndrome. These studies imply that the bony, vascular and malignant lesions might be related to an underlying neural abnormality. To date, however, no consistent chromosomal abnormalities have been identified in these patients.

Similar to the phakomatoses and familial syndromes such as LFS and Turcot syndrome, enchondromatosis appears to confer a substantial increased risk of a variety of CNS and other malignancies, at least through the sixth decade of life and as early as the first decade; and children, adolescents and adults appear not to differ substantially in this risk. These two points have implications for both primary physicians and specialists, including surgeons, because it means that the risk of intracranial malignancy should not be ignored in any patient with enchondromatosis, whether they have accompanying vascular lesions or not. In fact, those without vascular lesions may have higher risk over the first few decades of life than those with. Further research is needed to clarify the causes of increased risk. As our understanding of the genetics underlying various familial disorders improves, more effective therapies may be developed to successfully treat associated malignancies once they arise, and may even prevent them in patients who are at increased risk.


Most CNS tumours occur sporadically. However, the extremely strong association that exists between a variety of CNS tumours and a number of familial syndromes warrants vigilant attention, for several reasons. First, the tumours often behave differently, often being multiple in both number and type, and demonstrating atypical responsiveness to treatment, some associated with better and others worse outcomes than their sporadic counterparts. Second, other bodily systems and tissues frequently are involved, oftentimes with tumours of their own. Third, because so many of these syndromes are associated with cutaneous or other visibly apparent abnormalities, these tumours can be anticipated and thereby detected earlier in their course, potentially aiding in their treatment. Lastly, the association of both CNS and non-CNS tumours with familial syndromes for which genetic/chromosomal abnormalities have been established is resulting in enhanced understanding regarding the underlying pathology behind these tumours, which holds the promise for future, more-targeted therapies to both prevent and treat them.

NF-1 is the most common and classic familial syndrome associated with CNS tumours, but a wide range of both similar and highly dissimilar syndromes exist, ranging from neurofibromatosis type II and tuberous sclerosis to BCNS to dyschondroplasia syndromes. As scrupulous vigilance is the cornerstone of patient care in these patients, neurosurgeons, both paediatric and adult, need to be familiar with all these syndromes if they are to optimise their management. Recognising these syndromes also will help the surgeon to decide when surgery is in fact indicated and likely to be of benefit to the patient, and when not. In addition, as targeted therapies become available, like everolimus for SEGA and potentially other manifestations of tuberous sclerosis, they too may start to play a huge role in patent management.

Conflicts of interest

The authors do not have any conflicts of interest.

Copyright information

© Springer-Verlag Berlin Heidelberg 2013