Send letter to journal:
Re: Giant cell astrocytomas in Tuberous Sclerosis Complex

June 30, 2008

To the Editor: We read with interest the study by O’Callaghan et al.1 The authors’ observations largely confirmed what was already known about giant cell astrocytomas (GCA) in tuberous sclerosis (TSC). However, we have serious concerns about the methodology and several of the conclusions drawn. Most importantly, they do not provide plausible evidence for their main clinical conclusion that screening for GCA should not be undertaken - the direct opposite of current international consensus guidelines.2 At an NIH-funded consensus conference, an international expert panel recommended periodic cranial imaging every 1 to 3 years in children, generally up to the age of 21 years.2 The recommendations were based on evidence that neuroradiological surveillance, early detection and early surgical intervention for GCA in TSC was associated with better neurological, cognitive and behavioural outcome than children with TSC who did not have surveillance for GCA.3,4,5 The consensus panel felt that surveillance was of particular importance given that those most at risk of GCA in TSC would be children, many of whom would have global intellectual disability (mental retardation) thus posing significant challenges for clinical evaluation for and monitoring of possible features of raised intracranial pressure.6 It has never been recommended that a presumed GCA be resected solely on the basis of contrast enhancement, size > 1 cm, MR spectroscopy, or any characteristics other than clinically-concerning serial growth, or the presence of CSF obstruction. This is because, as O’Callaghan et. al. correctly note, these attributes cannot reliably identify a GCA. The clinical diagnosis of a GCA can be extremely difficult. They often present insidiously with subtle changes in behavior, affect, cognitive function, or seizure frequency long before overt symptoms such as headache, vomiting, ataxia, or visual loss are apparent. By the time clear-cut symptoms of increased intracranial pressure are evident, the likelihood of reversing them is markedly diminished. Figure 1 shows the extent of a GCA 36 hours after first clinical signs of increased intracranial pressure emerged. Operative resection of a large giant cell astrocytoma is fraught with difficulties, often rendering gross total resection impossible. Partially resected GCA almost invariably recur. Surely, O’Callaghan and colleagues are not advocating that we wait until it is almost impossible to be clinically helpful or until a child becomes blind, or suffers irreversible neurologic deficits? There is clear international consensus that the best radiological predictor of transformation of a subependymal nodule (SEN) to a GCA is serial growth, hence the recommendation that subependymal lesions that demonstrate growth on serial neuroimaging be evaluated for resection, or followed more closely. This was exactly the question that this paper did not examine. Only 1 of O’Callaghan et al’s 179 patients had serial scans.1 They argued that this adult patient who had neuro-imaging between the ages of 35 and 41 showed no change in their GCA. This was used as evidence of stabilization of serial growth. Stabilization of a GCA that has previously shown serial growth has never been reported in the literature, neither has spontaneous regression, as has been occasionally reported in gliomas associated with neurofibromatosis type 1.7 Interestingly, examination of figures 1 and 2 of the paper1 shows a SEN located superiorly in the ventricle and away from the foramen of Monro, suggesting that this lesion may have been static and not of sufficient size to produce hydrocephalus or other symptoms. The authors correctly point out that it is not known how rapidly a GCA might grow. It is also not known how slowly a GCA might grow. We have identified cases in whom GCA demonstrated growth, remained stable for intervals of many years, and then began to increase again. Finally, the authors concluded that ‘screening’ was not warranted for GCA in TSC because the rates of asymptomatic GCA were higher than the rates of symptomatic GCA. Apart from the fact that they did not use the internationally agreed criteria to define GCA, and excluded most of the population who would be most at risk, they also made a conceptual error. ‘Screening’ studies as outlined by them are aimed at general population public health screening such as for cervical cancer, for instance. The WHO criteria for screening are of little relevance to high-risk groups in specific disease populations. Medical literature predating the widespread availability of MRI and CT scanning confirms the invariably poor outcome of GCA patients who come to medical attention only when symptomatic.8 The cost of surveillance of this ‘at risk’ population and the risk of operating on a patient prior to the onset of overt neurological symptoms, are significantly less than the cost to society, family, and the individuals with TSC of an emergency operation, often carried out under less than optimal conditions, long-term neurologic sequelae such as blindness, new or worsened cognitive defects and physical disability. Neuroradiological monitoring for GCA every 1-3 years into adulthood in those with diagnosed TSC should remain the internationally agreed standard of care.

David Neal Franz, MD Pediatric Neurologist Cincinnati, Ohio, USA

Petrus J de Vries, MRCPsych, PhD Developmental Neuropsychiatrist Cambridge, UK

Peter B. Crino, MD Adult Neurologist Philadelphia, PA, USA

The following members of the TS Alliance Professional Advisory Board have reviewed and agree with the views outlined in this letter: Peter M. Black, MD, PhD, Candida Brown, MD, Harry T. Chugani, MD, PhD, Francis DiMario, MD, Kevin Ess, MS, PhD, David Frim, MD, Ajay Gupta, MD, Gregory Holmes, MD, Mary Kay Koenig, MD, PhD, Michael Kohrman, MD, William McClintock, MD, Hope Northrup, MD, PhD, E Steve Roach, MD, Mustafa Sahin, MD, Steven Sparagana, MD, Howard Weiner, MD, and Joyce Wu, MD.

References

1. O’Callaghan FJK, Martyn CN, Renowden S, Noakes M, Presdee D, Osborne JP. Sub-ependymal nodules, giant cell astrocytomas and the tuberous sclerosis complex: a population based study. Arch Dis Child 2008; May 13 [Epub ahead of print]

2. Roach ES, DiMario FJ, Kandt RS, Northrup H. Tuberous Sclerosis Consensus Conference: recommendations for diagnostic evaluation. National Tuberous Sclerosis Association. J Child Neurol 1999;14(6):401-7.

3. Torres OA, Roach ES, Delgado MR, Sparagana SP, Sheffield E, Swift D, Bruce D. Early diagnosis of subependymal giant cell astrocytoma in patients with tuberous sclerosis. J Child Neurol 1998;13(4):173-7.

4. Nabbout R. Santos M, Rolland Y, Delalande O, Dulac O, Chiron C. Early diagnosis of subependymal giant cell astrocytoma in children with tuberous sclerosis. J Neurol Neurosurg Psychiatry 1999; 66(3): 370-5.

5. Goh S, Butler W. Thiele EA, Subependymal giant cell tumors in tuberous sclerosis complex. Neurology 2004; 63(8): 1457-61.

6. de Vries P, Humphrey A, McCartney D, Prather P, Bolton P, Hunt A; TSC Behaviour Consensus Panel. Consensus clinical guidelines for the assessment of cognitive and behavioural problems in Tuberous Sclerosis. Eur Child Adolesc Psychiatry 2005;14(4):183-90.

7. Listernick R, Ferner RE, Liu GT, Gutmann DH. Optic pathway gliomas in neurofibromatosis-1: controversies and recommendations. Ann Neurol 2007;61(3):189-98.