American Journal of Medical Genetics 122A:95–99 (2003)

Optic Pathway Gliomas in Neurofibromatosis Type 1:The Effect of Presenting Symptoms on Outcome

Allison King,1 Robert Listernick,2 Joel Charrow,3 Linda Piersall,4 and David H. Gutmann5*1Division of Pediatric Hematology and Oncology, St. Louis Children’s Hospital, St. Louis, Missouri2Division of General Academic Pediatrics, The Children’s Memorial Hospital, Department of Pediatrics,Feinberg School of Medicine, Northwestern University, Chicago, Illinois3Division of Genetics, The Children’s Memorial Hospital, Department of Pediatrics, Feinberg School of Medicine,Northwestern University, Chicago, Illinois4Division of Pediatric Genetics, St. Louis Children’s Hospital, St. Louis, Missouri5Department of Neurology, Washington University School of Medicine, St. Louis, Missouri

Children withneurofibromatosis type 1 (NF1)may present with optic pathway gliomas(OPG) that can progress to visual loss orother neurologic symptoms. These tumorsmay become evident either as a result ofpatient signs or symptoms or as an incidentalfinding on ‘‘baseline’’ neuroimaging studies.In an attempt to determine if there weredifferences between symptomatic and asym-ptomatic children with OPG, a retrospectivecohort study of ninety children with NF1 andOPG was performed using data from twolarge NF1 referral centers. Age at diagnosis,presenting symptoms, tumor location, asso-ciated features, and clinical response wereassessed for children who were initiallysymptomatic from their OPG (n¼51) andcompared to similar data of asymptomaticchildren whose tumors were incidentallydiscovered by MRI (n¼39). There were nodifferences in age at presentation, tumorlocation, NF1-associated symptoms, or clin-ical response between the groups. Initiallysymptomatic children were much morelikely to require treatment (OR: 14.8, 95% CI[1.9–116.7]) than those with incidentallydiscovered, asymptomatic OPG. Although36% of OPG were diagnosed in children overthe age of 6 years, none received prior neuro-imaging and only two children had previ-ously normal eye examinations, suggestingthat the vast majority of OPG in this groupwere longstanding, undiagnosed tumors.

Based on these findings, we do not advocate‘‘baseline’’ MRI in children with NF1, butstrongly recommend that all children of theage 10 years and younger with NF1 have com-plete annual ophthalmologic evaluations.� 2003 Wiley-Liss, Inc.

KEY WORDS: neurofibromatosis 1; opticpathway glioma; pilocyticastrocytoma; brain tumor

INTRODUCTION

Neurofibromatosis 1 (NF1) is a common autosomaldominant tumor predisposition syndrome that affects 1in every 3,500 people worldwide [Friedman et al., 1999].Individuals with NF1 are prone to the development ofbenign and, less frequently, malignant tumors [Matsuiet al., 1993]. Most adults withNF1will develop a benigntumor composedmainly ofSchwann cells andfibroblastsassociated with nerve sheaths, termed a neurofibroma.However, neurofibromas are less frequently encoun-tered in children with NF1. In contrast, 15–20% ofchildrenwithNF1will develop a brain tumor. Typically,these brain tumors are low-grade pilocytic astrocytomasarising in the optic nerve and chiasm, but may also beseen in the hypothalamus, brainstem, and cerebellum[Lewis et al., 1984; Lund and Skovby, 1991; Listernicket al., 1994, 1995; Pascual-Castroviejo et al., 1994;Habiby et al., 1995; Janss et al., 1995;Molloy et al., 1995;Vinchon et al., 2000].

Optic pathway gliomas (OPG) are classified by theWorld Health Organization (WHO) as grade I ne-oplasms that presumably arise from supporting astro-cytes in the optic nerve [Rubenstein, 1988; Kleihuesand Cavanee, 2000]. Microscopically, these tumorsare composed of proliferating glial fibrillary acidicprotein (GFAP)-immunoreactive astrocytes with char-acteristic eosinophilic Rosenthal fibers. NF1-associatedpilocytic optic gliomas are histologically identical topilocytic astrocytomas that occur elsewhere in the

*Correspondence to: David H. Gutmann, M.D., Ph.D., Depart-ment of Neurology, Box 8111; 660 S. Euclid Avenue, St. Louis, MO63110. E-mail: gutmannd@neuro.wustl.edu

Received 14 October 2002; Accepted 18 February 2003

DOI 10.1002/ajmg.a.20211

� 2003 Wiley-Liss, Inc.

central nervous system in individuals without NF1(most commonly in the cerebellum) [Cummings et al.,2000; Gutmann et al., 2002]. NF1-associated OPG mayinvolve the optic nerve, chiasm, hypothalamus, andpost-chiasmal regions, including the optic tracts. Pre-vious studies have shown that, if all children with NF1undergo screening neuroimaging, 15% of themwill haveradiographic evidence of OPG [Listernick et al., 1989].However, only half of these children will ever developsymptoms or signs related to the OPG [Listernick et al.,1994]. Symptomatic OPG in children with NF1 gener-ally result in decreased visual acuity, diminished visualfields, or proptosis. There does not appear to be anyrelation between the size of the tumor and the severity ofclinical symptoms. Precocious puberty may be seen inas many as 40% of children with tumors involving theposterior chiasm and hypothalamic regions and is oftenthe presenting manifestation of the OPG [Habiby et al.,1995]. Lastly, chiasmatic tumors have also been asso-ciated with diencephalic syndrome and hydrocephalus[Russell, 1951; Packer et al., 1988].

Previous studies focused on the natural history ofOPG in children with NF1 have demonstrated thatclinical progression following presentation occurs in lessthan one third of the patients [Listernick et al., 1994].In addition, nearly all children became symptomaticbefore the age of 6 years. There are even reports ofOPGs in children with NF1 that spontaneously regress,suggesting that these tumors have limited growthpotential and may not always require treatment at thetime of diagnosis [Brzowski et al., 1992; Perilongo et al.,1999]. Based on these observations, most clinical NFcenters routinely perform detailed annual neurologicand ophthalmologic examinations on children duringthe first 6 years of life. MRI studies typically are notobtained unless neurologic or ophthalmologic abnor-malities are present. Despite a lack of clinical datasupporting theiruse in themanagement of childrenwithNF1, many clinical centers obtain ‘‘baseline’’ brain MRIevaluations to look for asymptomatic OPGs [Listernicket al., 1997], resulting in the discovery of asymptomaticOPGs. The purpose of this study was to determine iftherewere clinical differences between initially sympto-matic and incidentally identified asymptomatic childrenthatmight predict the behavior of NF1-associated OPG.

MATERIALS AND METHODS

This study was performed as an approved HumanStudies Committee protocol at Washington University.Only children diagnosed with OPG and who met diag-nostic criteria for NF1, as established by the NationalInstitutes of Health Consensus Development Confer-ence [National Institutes of Health Consensus Develop-ment Conference Statement, 1988], were included.

A retrospective chart review was performed on NFclinic charts of children with NF1 and OPG at St. LouisChildren’s Hospital and Children’s Memorial Hospitalin Chicago. These two pediatric NF centers are in largetertiary care pediatric hospitals in the Midwest withsimilar demographics; neither clinic was housed with-in a neuro-oncology program. St. Louis Children’s

Neurofibromatosis Center cares for approximately 400children, while Children’s Memorial Hospital Neurofi-bromatosis Clinic cares for approximately 900 children.

The respective NF1 databases were scanned for thepresence of OPG, and once identified, clinical data wereabstracted from each of the subject’s clinical presenta-tion. Data were collected from time of diagnosis to themost recent visit for each child.Date of diagnosis ofNF1,date of diagnosis of the OPG, characteristics of NF1,results of visual examinations and MRI evaluations,and the therapy for OPG were recorded for each child.Children were classified according to their status aseither being asymptomatic or initially symptomatic.The asymptomatic group represented children who hadbrainMRI either as a screening tool (‘‘baseline’’ MRI) orto evaluate a separate problem, such as a facial or neckplexiform neurofibroma. These MRI studies were notobtained to evaluate a specific neurologic complaint orphysical finding that suggested the presence of an OPG.Symptomatic children had brain MRI to investigate aspecific neurologic complaint or physical finding (e.g.,decreased vision, optic nerve pallor, headache, andprecocious puberty) that could directly be attributed tothe presence of an OPG.

Mean values were used for continuous variablesand frequency was used for categorical variables. Forcomparisons of continuous variables, the t-test wasused. Non-parametric tests and chi-square odds ratioswere used for comparisons of categorical data. A value ofP� 0.05 was determined to be the level of significance.All data analyses were performed with the StatisticalPackage for the Social Sciences (SPSS, Chicago, IL),version 11.0.

RESULTS

A total of 90 patients were identified with NF1 andOPG at the two medical centers (Table I). Thirty-ninechildren were asymptomatic and 51 were clinicallysymptomatic at the time of diagnosis. In the asympto-matic group of children, 16 patients (41%) were male,16 (41%) had a family history of NF1, and 32 (82%) wereCaucasian. The average age of diagnosis of NF1 was3.7 years, whereas the average age at diagnosis of theOPG was 5.4 years (range 0–14.4 years). The meanduration of follow-up after OPG diagnosis was 5.6 years(range 0–15.6 years). In the symptomatic children,22 patients (43%) were male, 20 (39%) had a familyhistory of NF1, and 42 (82%) were Caucasian. Theaverage age of these children at diagnosis was 4.8 years(range 0–15.8 years). The racial demographics of chil-dren with NF1 in the Neurofibromatosis Clinic of theChildren’s Memorial Hospital are known (Caucasian63%, African-American 17%, Hispanic 19%). Signifi-cantlymoreCaucasian childrendevelopedOPGcompar-ed to African-American children (chi-square,P¼ 0.024).

Presenting Signs and Symptoms

In the asymptomatic group of children, computedtomography (CT) or magnetic resonance imaging (MRI)scans were obtained following establishment of thediagnosis of NF1 (‘‘baseline’’ MRI). A few patients had

96 King et al.

head MRI for evaluation of a head or neck plexiformneurofibroma.

No single complaint was predominant at presentationin the symptomatic children (Table II). Twenty (39%)were found on ophthalmologic examination to havedecreased visual acuity, defined as worse than 20/20vision. Proptosis was also noted in 13 children (26%), allof who had tumors involving the optic nerve but not allhad decreased visual acuity. Precocious puberty wasnoted in 10 children (20%), five of whom had tumorsinvolving the chiasm and hypothalamus. Six patients(12%) presented with strabismus, all of which hadanother abnormality on ophthalmologic examination,such as optic nerve pallor or afferent pupillary defect.Five patients (10%) were found to have visual fielddeficits. Ten children had optic nerve pallor (20%); allbut one had another abnormality on ophthalmologicexamination as well.

Tumor Location

The tumor location in children whose OPG wereincidentally discovered did not differ significantlyfrom those of children who were initially symptomatic

(Table III). Most commonly, OPG were confined to theoptic nerve(s) alone. One child had an OPG involvingthe optic nerve and a glioma in the frontal lobe, whileanother had an optic nerve glioma and a separate tumorin the brainstem. Of the 10 children who had precociouspuberty, 7had tumors that clearly involved theposteriorchiasm or hypothalamus on brain imaging studies.

NF1-Associated Features

Common manifestations of NF1, including a positivefamily history, were evenly distributed between thetwo groups (Tables I, IV). Children who were fol-lowed for more than 5 years from diagnosis of the OPGdid not have a greater incidence of NF1-associatedcomplications.

Clinical Course

Treatment was initiated in children who had pro-gressive loss of vision, gross disfigurement secondary tothe OPG (proptosis), or poor weight gain (diencephalicsyndrome). One child (2.5%) in the asymptomatic groupwas treated for progressive disease. She developedprogressive worsening of visual acuity and receivedcarboplatin and vincristine. She has had no progressionof disease approximately 7 months after completingtherapy.

Eighteen (35.3%) of the initially symptomatic patientswere treated. Four (7.8%) received carboplatin alone astherapy for progressive disease. Nine (17.6%) receivedboth carboplatin and vincristine. Five (9.8%) underwentresection or debulking surgery for gross disfigurementin a blind eye. All of the resected tumors were WHOgrade I pilocytic astrocytomas; four involved the opticnerve (two on the right and two on the left) and oneoriginated from the hypothalamic area. None of thesechildren received radiation therapy. Whereas only 2.5%of the children with the initially asymptomatic OPGwere treated, over 35% of the children who initiallypresented with a symptomatic OPG eventually receivedtreatment (OR [95% CI]¼ 14.8 [1.9–116.7]).

Thirty-five (39%) childrenwere followed formore than5 years after the discovery of their OPG. This group of

TABLE I. Optic Pathway Glioma in Children with NF1

Asymptomatic(n¼ 39)N (%)

Symptomatic(n¼51)N (%)

Total(n¼90)N (%)

Age at diagnosis of OPG in years (range) 5.4 (0–14.4) 4.8 (0–15.6) 5.1Age at diagnosis of NF1 in years (range) 3.7 (0–14.2) 2.8 (0–15.8) 3.2RaceCaucasian 32 (82) 42 (82) 74 (82)African American 4 (10) 5 (10) 9 (10)Hispanic 3 (8) 3 (6) 6 (7)Other 0 (0) 1 (2) 1 (1)

GenderMale 16 (41) 22 (43) 38 (42)Female 23 (59) 29 (57) 52 (58)

Family history of NF1 16 (41) 20 (39) 36 (40)Length of follow-up from diagnosis of NF1 to diagnosis ofOPG (in years) (range)

1.7 (0–15.1) 2 (0–13.9)

Length of follow-up after diagnosis of OPG (in years) (range) 5.6 (0–14.4) 4.8 (0–15.6) 5.1

TABLE II. Symptoms and Signs in Symptomatic Childrenwith NF1

Children(n¼51)N (%)

SymptomsProptosis 13 (26)Headache 4 (8)Poor weight gain 2 (4)

Signs or symptoms of precocious puberty 10 (20)Signs on initial ophthalmologic examinationAbnormal visual acuity (VA) 4 (8)Optic nerve pallor 5 (10)Abnormal VAþpallor 5 (10)Abnormal VAþ afferent pupillary defect 6 (12)Abnormal VAþ 2 or more signsa 5 (10)Strabismus 6 (12)

aOptic nerve pallor, afferent pupillary defect, or optic nerve atrophy.

Optic Gliomas in NF1 97

children did not differ significantly from those childrenwho had been followed for a shorter period of time inrespect to either the need for therapy or the presence ofother NF1 associated complications.

Children Older Than 6 Years atTime of Diagnosis of OPG

Fourteen (27%) of the symptomatic children and 18(47%) of the asymptomatic children were older than theage of 6 years at the time of diagnosis of their OPG.Ninety percent of the childrenwith symptomatic tumorswere identified by the age of 10 years. However, of these32 children, none had had previously normal neuroima-ging and only 2 had had previously normal ophthalmo-logic examinations. One boy (age 7.8 years) presentedwith signs of precocious puberty, after having had anormal eye examination at 6.3 years. Head MRIdemonstrated an OPG involving the left optic nerve,chiasm, and left optic tract. Similarly, one 10.4-year-oldgirl was discovered to have abnormal visual fields,leading to thediagnosis of a left optic nerveand chiasmalOPG. She had had several previous ophthalmologicexaminations that had been normal. On follow-upevaluations, the visual fields returned to normal, hervisual acuity remained normal, and she never requiredany therapy.

DISCUSSION

In order to provide optimal care for childrenwithNF1-associatedOPG, itwouldbeuseful todefinea ‘‘high-risk’’subpopulation that ismost likely to progress and requiretreatment. We did not identify any factor specificallyassociated with the presence of a symptomatic OPGthat might serve as a predictor of the need for future

treatment. Previous studies have suggested that theanatomic location of the OPG might represent a factorthat influences morbidity. Balcer and colleagues foundthat children with post-chiasmal involvement had ahigher likelihood of visual acuity loss [Balcer et al.,2001]. Although not statistically significant, this wastrue in our study as well. In addition, it has been sug-gested that children with orbital plexiform neurofibro-mas have a poorer prognosis [Listernick et al., 1994;Pascual-Castroviejo et al., 1994]. In our series, neithertumor location nor the presence of a facial/neck plexi-form neurofibroma was associated with an increasedrisk of harboring anOPG that requires treatment. Sinceour patients are part of multidisciplinary NF clinicsrather than oncology programs, they provide an accu-rate representation of children with NF1 who developOPG.

Children with OPG who were initially symptomaticwere muchmore likely to eventually require treatment.This suggests that children who present with signs orsymptoms of OPG may require more frequent evalua-tions than children whose tumors were incidentallydiscovered. Janss and colleagues [Janss et al., 1995]suggested that tumors in children less than the age of5 years might also be more aggressive than those foundin older children. It should also be noted that youngchildren rarely complain of decreased vision and theirOPG are most commonly identified by formal ophthal-mologic evaluation.

Previous studies have demonstrated that the period ofhighest risk for the development of OPG in NF1 is thefirst 6 years of life [Listernick et al., 1994; Listernicket al., 1997]. Although we found that both 27% (14) ofchildren with symptomatic OPG and 16% (3 of 19) ofchildren treated for their OPG were older than the ageof 6 years, only two of these children had new physical

TABLE III. Location of Optic Pathway Gliomas in Children with NF1

Location

Asymptomatic(n¼ 39)N (%)

Symptomatic(n¼ 51)N (%)

Total(n¼ 90)N (%)

Nerve only 11 (28) 17 (33) 28 (31)Chiasm only 6 (15) 3 (6) 9 (10)Hypothalamus only 1 (3) 3 (6) 4 (5)Nerveþ chiasm 13 (33) 16 (31) 29 (32)Nerveþ chiasmþhypothalamus 4 (10) 8 (16) 12 (13)Chiasmþhypothalamus 4 (10) 2 (4) 6 (7)Nerveþ other brain location 0 (0) 2 (4) 2 (2)

TABLE IV. Associated NF1 Features in Children with Optic Pathway Glioma

Feature

Asymptomatic(n¼39)N (%)

Symptomatic(n¼51)N (%)

Total(n¼90)N (%)

Plexiform neurofibroma 13 (33) 16 (31) 29 (32)Head/neck location 7 (18) 9 (18) 16 (18)

Macrocephaly 23 (59) 25 (49) 48 (53)Learning disabilities 20 (51) 18 (35) 38 (42)Seizures 3 (8) 1 (2) 4 (4)Other brain tumors 0 (0) 2 (4) 2 (2)

98 King et al.

findings suggestive of either de novo appearance of atumororprogressionof anunknown, previously existingtumor. In fact, since the eye examination of the 10-year-old girl who had abnormal visual fields returned tonormal the following year without therapy, it could beargued that she had an asymptomatic tumor. None ofthese children had had prior normal neuroimaging.Since most of the older symptomatic patients probablyhad longstanding OPG that had not come to medicalattention until after the diagnosis of NF1 was estab-lished and formal ophthalmologic examination wasundertaken, these data reconfirm the recommendationthat children with NF1 should have annual ophthalmo-logic examinations during the first 6 years of life. Inaddition, these results support the need for continuedsurveillance in older children and suggest that the riskfor symptomatic progression extends beyond the age of6 years. As a result of our experience, we recommendthat all children who are of the age 10 years or youngerhave complete ophthalmologic examinations annually,beginning when the diagnosis of NF1 is suspected.Any child who has symptoms or signs suggestive of thepresence of an OPG should undergo immediate MRIevaluation.

A previous report found that all children with pre-cociouspubertyhadhypothalamic tumors [Habiby et al.,1995].While themajority of the patientswith precociouspuberty in our study had hypothalamic involvement,3 of 10 children did not have radiographic evidence ofhypothalamic or posterior chiasmatic tumors. Theseresults suggest that not all children with NF1 andprecocious puberty will have an identifiable hypotha-lamic tumor.

Themajority of childrenwithOPGwhowere asympto-matic didnot receive treatment.Although one childwithworsening visual function received therapy, she wouldhave been identified by ophthalmologic examination ona follow-up examination. The children with incidentallydiscovered OPG, whowere not treated, did not have anyprogression of visual loss or growth of the OPG. Thesedata reinforce our previous recommendation that ‘‘base-line’’ MRI of asymptomatic children with NF1 is notwarranted.

Although the experience of our centers representsthe largest study of children with NF1 and OPG, we didnot delineate a high-risk population of children withNF1 likely to develop progressive OPG, other thanyoung children who were symptomatic at presentation.Prospective studies of children with NF1-associatedOPG combined with molecular genetic analyses mayserve to identify specific risk factors that stratifychildren with NF1 into clinically important subgroups.Unless a high-risk group is identified, all children withNF1 should be examined yearly for signs and symptomsof OPG and established ophthalmologic screeningprotocols should be followed.

ACKNOWLEDGMENTS

We thank Dr. Michael R. DeBaun for advice andsuggestions during the execution of this project.

REFERENCES

Balcer LJ, Liu GT, Heller G, Bilaniuk L, Volpe NJ, Galetta SL, Molloy PT,Phillips PC, Janss AJ, Vaughn S, Maguire MG. 2001. Visual loss inchildren with neurofibromatosis type 1 and optic pathway gliomas:Relation to tumor location by magnetic resonance imaging. Am JOphthalmol 131:442–445.

Brzowski AE, Bazan C, 3rd, Mumma JV, Ryan SG. 1992. Spontaneousregression of optic glioma inapatientwithneurofibromatosis.Neurology42:679–681.

Cummings TJ, Provenzale JM, Hunter SB, Friedman AH, Klintworth GK,Bigner SH,McKlendon RE. 2000. Gliomas of the optic nerve: Histologic,immunohistochemical (MIB-1 and p53), and MRI analysis. Acta Neuro-pathol 99:563–570.

Friedman JM, Gutmann DH, MacCollin MM, Riccardi VM. 1999. Neurofi-bromatosis: Phenotype, natural history, and pathogenesis. 3rd edn.Baltimore: Johns Hopkins University Press, 381 p.

Gutmann DH, Hedrick NM, Li J, Nagarajan R, Perry A, Watson MA. 2002.Comparative gene expression profile analysis of neurofibromatosis1-associated and sporadic pilocytic astrocytomas. Cancer Res. 62:2085–2091.

Habiby R, Silverman B, Listernick R, Charrow J. 1995. Precocious pubertyin children with neurofibromatosis type I. J Pediatr 126:364–367.

Janss AJ, Grundy R, Cnaan A, Savino PJ, Packer RJ, Zackai EH, GoldweinJW, Sutton LN, Radcliffe J, Molloy PT. 1995. Optic pathway andhypothalamic/chiasmatic gliomas in children younger than age 5 yearswith a 6-year follow-up. Cancer 75:1051–1059.

Kleihues P, CavaneeWK. 2000.World Health Organization classification oftumours, International Agency for Reseach on Cancer (IARC). Pathol-ogy and genetics of tumours of the nervous system. Lyon: IARC Press,308 p.

Lewis RA, Gerson LP, Axelson KA, Riccardi VM, Whitford RP. 1984.Von Recklinghausen neurofibromatosis: Incidence of optic gliomata.Ophthalmology 91:929–935.

Listernick R, Charrow J, GreenwaldMJ, Esterly NA. 1989. Optic gliomas inchildren with neurofibromatosis type 1. J Pediatr 114:788–792.

Listernick R, Charrow J, Greenwald M, Mets M. 1994. Natural historyof optic pathway tumors in children with neurofibromatosis type I: Alongitudinal study. J Pediatr 125:63–66.

Listernick R, Darling C, Greenwald M, Strauss L, Charrow J. 1995. Opticpathway tumors in children: The effect of neurofibromatosis type 1 onclinical manifestations and natural history. J Pediatr 127:718–722.

Listernick R, Louis DN, Packer RJ, Gutmann DH. 1997. Optic pathwaygliomas in children with neurofibromatosis 1: Consensus statementfrom the NF1 optic pathway glioma task force. Ann Neurol 41:143–149.

LundAM, Skovby F. 1991.Optic gliomas in childrenwith neurofibromatosistype I. J Pediatr 150:835–838.

Matsui I, Tanimura M, Kobayashi N, Sawada T, Nagahara N, Akatsuka J.1993. Neurofibromatosis type I and cancer. Cancer 72:746–754.

Molloy PT, Bilaniuk LT, Vaughan SN, Needle MN, Liu GT, Zackai EH,Phillips PC. 1995. Brainstem tumors in patientswith neurofibromatosistype 1: A distinct clinical entity. Neurology 45:1897–1902.

National Institutes of Health Consensus Development Conference State-ment: Neurofibromatosis. 1988. Arch Neurol 45:575–578.

Packer RJ, Bilaniuk LT, Cohen BH. 1988. Intracranial visual pathwaygliomas in children with neurofibromatosis. Neurofibromatosis 1:212–222.

Pascual-Castroviejo I, Martinez Bermejo A, Lopez Martin V, Roche C,Pascual Pascual SI. 1994. Optic gliomas in neurofibromatosis type 1(NF1). Presentation of 31 cases. Neurologia 9:173–177.

Perilongo G, Moras P, Battistella A, Clementi M, Laverda A, Murgia A.1999. Spontaneous partial regression of low-grade glioma in childrenwith neurofibromatosis-1: A real possibility. J Child Neurol 14:352–356.

Rubenstein LJ. 1988. Pathologic features of optic nerve and chiasmaticgliomas. Neurofibromatosis 1:152–158.

Russell A. 1951. A diencephalic syndrome of emaciation in infancy andchildhood. Arch Dis Child 26:274.

Vinchon M, Soto-Ares G, Ruchoux MM, Dhellemmes P. 2000. Cerebellargliomas in children with NF1: Pathology and surgery. Childs Nerv Syst16:417–420.

Optic Gliomas in NF1 99