Services on Demand
- Similars in SciELO
Print version ISSN 0021-7557
J. Pediatr. (Rio J.) vol.85 no.3 Porto Alegre May/June 2009
LETTERS TO THE EDITOR
Primary diffuse leptomeningeal gliomatosis: a rare disease in pediatric patients
Francisco Helder Cavalcante FelixI
IPediatric hemato-oncologist, Pediatric
Hemato-Oncology Service, Hospital Infantil Albert Sabin, Fortaleza, CE, Brazil
The recent report about a rare pediatric brain tumor by Val Filho & Avelar1 is interesting and deserves high consideration within the published literature about the topic, since it represents a very rare event. On behalf of its scientific accuracy, we must therefore make a major amendment about its diagnosis.
The authors made an adequate brief review about gliomatosis cerebri, a rare disease classified by the 2007 edition of the World Health Organization (WHO)2 as a grade III lesion with poor prognosis. It was first described by Nevin in 19382 and is defined as "a diffuse glioma growth pattern consisting of exceptionally extensive infiltration of a large region of the central nervous system, with involvement of at least three cerebral lobes, usually with bilateral involvement of the cerebral hemispheres and/or deep gray matter, ..."2 The main theory accepted to explain its origin is that it represents a subtype of otherwise ordinary diffuse glioma characterized by exceptional infiltrative capacity, classically without forming a tumor mass lesion.2 It is a rare kind of lesion, with only few hundreds of cases described, less than 30% of them occurring in children. Although 17% of gliomatosis cerebri cases were reported as having additionally leptomeningeal involvement, this finding is not sufficient to make such diagnosis when isolated.2 The WHO classification text further states that "gliomatosis cerebri should be distinguished from two other types of gliomatosis, i.e. leptomeningeal gliomatosis and gliomatosis peritonei. Leptomeningeal gliomatosis is the widespread infiltration of the subarachnoid space by a diffuse glioma, most commonly an intra-axial glioma that has invaded the leptomeninges (secondary leptomeningeal gliomatosis), or, rarely, leptomeningeal spread of a glioma originating in an ectopic leptomeningeal glial or glioneuronal rest (primary leptomeningeal gliomatosis)."2
These characteristics resemble the case reported by Val Filho & Avelar,1 in which initially a diagnosis of obstructive hydrocephalus was made followed by the finding of "lesions scattering throughout the subarachnoid space, mainly in the posterior cranial fossa. These lesions diffusely covered a large part of the encephalon ( )."1 Unfortunately, the image they pictured has no identification of the type of magnetic resonance imaging (MRI) sequence, but it is probably contrast-enhanced T1-weighted spin-echo. Although T2w/FLAIR images are the preferred studies for diagnosis of gliomatosis cerebri,2 one can clearly see in their sequence of images that there is not an extensive involvement of the brain parenchyma.1 Thus, their case report does not fulfill the criteria published by the WHO for the diagnosis of gliomatosis cerebri, but can be classified more properly as a very rare case of primary diffuse leptomeningeal gliomatosis (PDLG) in a young child with prolonged survival.
The first report of PDLG was published in 1954 by Moore. It is described more frequently along the spinal cord and associated with central nervous system (CNS) congenital dysraphism.3 It must be distinguished from secondary meningeal gliomatosis, where there is leptomeningeal invasion by neoplastic diseases of the CNS parenchyma, especially oligodendroglioma, medulloblastoma and gliomatosis cerebri. The diagnosis of true PDLG is exceptional and a report by Debono et al., in 2006, reviewed the literature so far and found 30 cases published, only nine of these cases in children aged 9-17 years old.3 Since then, we have found additional 10 cases including Val Filho & Avelar's (Table 1), and seven more cases reported recently by Dörner et al.4 in their review that Debono et al.3 did not cite. Of these, five were found in children aged 2-17 years old. This sums 47 cases reported in the literature, of which 14 in children. Definitive diagnosis of PDLG classically need postmortem detailed description,3 and this fact would reduce the number of cases reported to 36.
Applying less strict criteria (including patients without necropsy and those reported alive) and pooling the data collected by Debono et al.3 and Dörner et al.4 with ours, we found that more common initial presentation included symptoms and signs of intracranial hypertension (IH), like vomiting and headache (37/47 cases, 79%), multiple cranial nerve palsies (especially neuro-ophthalmologic) (25/47 cases, 53%) and seizures (11/47 cases, 23%). There was slight male predominance (25/22 cases = 1.13 ratio). There were 14 WHO low-grade tumors (30%) and 23 high-grade tumors (49%). The problem of diagnosing histological grade in PDLG is that the disease is frequently misclassified,2 making these numbers unreliable. Median survival calculated by the Kaplan-Meyer method using the reported survival rate of this group of patients is 5 months, and survival at 12 months is only 30%. This survival rate is biased by the fact that many of these reports were of postmortem diagnosis and, therefore, the patients did not receive treatment. Selecting only those patients that received specific treatment and reported survivals (19 cases), the Kaplan-Meyer median survival is 15 months, and the 1-year survival is 52%. Only four cases were reported to be alive at the time of their publication (1, 4, Table 1), all of them were children, and three of them were less than 3 years old. They received chemotherapy (CT) (three cases), or combined radiotherapy (RT) and CT (one case). All of the patients that survived 24 months or more in this pooled group are children (except for a 21 year old male), and all received CT or combined CT and RT. The more frequently used drug was cisplatin. Four cases reported using temozolomide (1, 5, Table 1), all of them had survival rates higher than the group median (range 15-96 months). Interestingly, the case reported by Gonçalves et al.5 had a remarkable survival rate (only matched by another reported case, see references).3,4 This child used valproate, a drug that has been recently reported to have antitumor effects that could enhance the effects of both RT and CT in brain tumor patients.6 Val Filho & Avelar reported that their patient had seizures but did not disclose which anticonvulsants the child had used. Other authors equally have not reported on anticonvulsant usage, so we cannot say if any of the other cases used valproate either.
We conclude by stating that the report by Val Filho & Avelar can be reclassified as a very interesting case of PDLG in a young child, with one of the longest reported survival rates for this rare and grim disease. This exceptionally rare disease affects all ages, is uncommon in children in whom it seems to bear a better prognosis (only children were reported alive) and is responsive to cisplatin-based radiochemotherapy or temozolomide (that could be used as second choice). Even with multimodality treatment, it has a very poor prognosis with short reported median survival rate (15 months) comparing with other high-risk brain tumors such as glioblastoma multiforme and diffuse pontine glioma. One compelling report is that of a prolonged survival in a child who used CT and valproate, suggesting that epigenetic inhibition can have a role in its treatment in the future.
No conflicts of interest declared concerning the publication of this letter.
José Aloysio da Costa Val Filho;I Lucília Graciano Silva AvelarII
IOficial médico. Chefe, Serviço
de Neurologia, Hospital da Polícia Militar, Belo Horizonte, MG, Brazil. Coordenador,
Serviço de Neurocirurgia Infantil, Biocor Instituto, Nova Lima, MG, Brazil
IIOficial médica, Serviço de Pediatria, Hospital da Polícia Militar, Belo Horizonte, MG, Brazil
We would like to thank Dr. Francisco Helder Cavalcante Felix for his comments about our case report "Gliomatosis cerebri with favorable outcome in a child: a case report."1 Dr. Felix has conducted an extensive research and a very elegant scientific discussion about the theme, which helped us to better understand this rare disease.
However, we would like to clarify some of Dr. Felix's comments.
As mentioned in his letter, the image showed was a magnetic resonance (MR) contrast-enhanced T1. According to our point of view (as a surgeon), this acquisition time shows better the findings in the subarachnoid space, as well in the parenchyma, in contrast to what was stated in the description.2 Anyway, the child was exhaustively studied by MR imaging after the diagnosis, in all sequences, with and without contrast, and the images were analyzed by many radiologists, who agree with the diagnosis.
During some time, before the oncologic treatment, the patient seemed to have some involvement of the frontal and cerebelar lobes, as well the midbrain, what could explain the obstructive hydrocephalus (Figure 1). She also has, to this date, a cystic involvement of the spinal cord (syringomyelia), which resembles a regression of the tumor originally inside the organ (Figure 2).
Initially, she used phenobarbital and carbamazepine for the seizures. Currently, she is not using any drugs. She never used valproate.
Therefore, we conclude that this can be a mixed form, with characteristics of gliomatosis cerebri and primary diffuse leptomeningeal gliomatosis (PDLG), as perfectly pointed by Dr. Felix. In our daily clinical practice with patients, we observed that pure forms, as described in our reference sources, are less common than mixed presentations.
Again, we thank Dr. Felix for his contribution to the topic.
No conflicts of interest declared concerning the publication of this letter.
References (Letter to the editor)
1. Val Filho JA, Avelar LG. Gliomatosis cerebri with favorable outcome in a child: a case report. J Pediatr (Rio J). 2008;84:463-6 [ Links ]
2. Fuller GN, Kros JM. Gliomatosis cerebri. In Louis DN, Ohgaki H, Wiestler OD, Cavenee WK, editors. WHO Classification of tumours of the central nervous system. Lyon: IARC; 2007. p. 50. [ Links ]
3. Debono B, Derrey S, Rabehenoina C, Proust F, Freger P, Laquerrière A. Primary diffuse multinodular leptomeningeal gliomatosis: case report and review of the literature. Surg Neurol. 2006;65:273-82. [ Links ]
4. Dörner L, Fritsch MJ, Hugo HH, Mehdorn HM. Primary diffuse leptomeningeal gliomatosis in a 2-year-old girl. Surg Neurol. No prelo 2008. [ Links ]
5. Gonçalves AL, Masruha MR, Carrete Jr H, Stávale JN, Silva NS, Vilanova LC. Primary diffuse leptomeningeal gliomatosis. Arq Neuropsiquiatr. 2008;66:85-7. [ Links ]
6. Wolff JE, Kramm C, Kortmann RD, Pietsch T, Rutkowski S, Jorch N, et al. Valproic acid was well tolerated in heavily pretreated pediatric patients with high-grade glioma. J Neurooncol. 2008;90:309-14. [ Links ]
Referências (Resposta dos autores)
1. Val Filho JA, Avelar LG. Gliomatosis cerebri with favorable outcome in a child: a case report. J Pediatr (Rio J). 2008;84:463-6. [ Links ]
2. Fuller GN, Kros JM. Gliomatosis cerebri. In: Louis DN, Ohgaki H, Wiestler OD, Cavenee WK, editors. WHO Classification of tumours of the central nervous system. Lyon: IARC; 2007. p. 50. [ Links ]