|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 24, No 16 (June 1), 2006: pp. 2570-2575 © 2006 American Society of Clinical Oncology. DOI: 10.1200/JCO.2005.03.8349 Second Primary Tumors in Neurofibromatosis 1 Patients Treated for Optic Glioma: Substantial Risks After Radiotherapy
From the Department of Clinical Genetics, St Mary's Hospital; Cancer Research UK Paediatric and Familial Cancer Research Group, Royal Manchester Children's Hospital; Department of Clinical Radiology, Manchester Royal Infirmary; Christie Hospital, Manchester; Department of Clinical Neurosciences, Guy's Kings and St Thomas' School of Medicine, Guy's Hospital, London, United Kingdom; and private practice, Los Angeles, CA Address reprint requests to D. Gareth R. Evans, MD, FRCP, Department of Clinical Genetics, St Mary's Hospital (SM2), Hathersage Rd, Manchester M13 OJH, United Kingdom; e-mail: Gareth.evans{at}cmmc.nhs.uk
PURPOSE: Optic pathway gliomas (OPGs) are the most common CNS tumor in neurofibromatosis 1 (NF1) patients. We evaluated the long-term risk of second tumors in NF1-related OPGs after radiotherapy. PATIENTS AND METHODS: We reviewed 80 NF1 OPG patients from two NF1 clinics to evaluate the long-term risk of developing subsequent nervous system tumors, with or without radiotherapy. RESULTS: Fifty-eight patients were assessable for second tumors. Nine (50%) of 18 patients who received radiotherapy after their OPGs developed 12 second tumors in 308 person-years of follow-up after radiotherapy. Eight (20%) of 40 patients who were not treated with radiotherapy developed nine tumors in 721 person-years of follow-up after diagnosis of their OPGs. The relative risk of second nervous system tumor after radiotherapy was 3.04 (95% CI, 1.29 to 7.15). CONCLUSION: There is a significantly increased risk of second nervous system tumors in those NF1 patients who received radiotherapy for their OPGs, especially when treated in childhood. Thus radiotherapy should only be used if absolutely essential in children with NF1.
Histologically, optic pathway gliomas (OPGs) are pilocytic astrocytomas (WHO grade 1) that affect the optic nerve, optic chiasm, or both. OPGs are predominantly a childhood tumor and approximately 50% occur in people with neurofibromatosis 1 (NF1), a dominantly inherited tumor-predisposing disorder.1 The sequelae of NF1 OPG and its treatments are now becoming apparent in long-term survivors.1-3 An increased risk of non-OP gliomas has been reported, and there have been a number of reports of malignant peripheral-nerve sheath tumors (MPNSTs) after radiotherapy.4 In general, pilocytic astrocytomas maintain their WHO grade 1 status over many years. However, there are rare examples of malignant transformation, with most reported patients having undergone previous radiotherapy, potentially inducing the malignant change.5 The absolute risks of these complications have not been established in long-term longitudinal studies.
Fifty-eight patients with OPGs were identified for whom long-term follow-up data were available from two NF clinics (Manchester and London, United Kingdom), including a population-based sample of patients from the Manchester Children's Tumor Registry.1 These 58 patients formed the study group for second nervous system tumor development. Eighteen of these patients had received radiotherapy, whereas 40 had not. An additional 22 NF1 OPG patients were identified from the two centers but were excluded from the second tumor analysis, as detailed in Figure 1. All case notes of the 58 assessable patients were reviewed to obtain details of the diagnosis, treatments, and subsequent tumors. Twenty-nine of these patients were also seen as part of a long-term follow-up study, details of which are illustrated in Figure 1.
For all patients, the diagnosis of NF1 was made according to criteria defined in the National Institutes of Health Consensus Statement of 1988,6 either directly at the time of long-term follow-up (29 patients) or indirectly through chart review (29 patients). Although there was no uncertainty in the notes regarding the diagnoses, given that this aspect could not be reviewed any further, the possibility that some were incorrectly diagnosed remains. The OPG diagnosis was based on clinical assessment, neuroimaging features, pathologic examination of biopsy material, or a combination of these methods. In each clinic, the incidence of second nervous system tumors in irradiated and nonirradiated OPG patients was compared. In Manchester, second primary tumors were ascertained from follow-up in the Genetic Register service and from the Regional Cancer Registry for North West England. All Manchester NF1 patients (including the OPG group) were checked on the North West cancer registry to achieve as complete a cancer ascertainment as possible. We compared the incidence of gliomas and MPNSTs in NF1 OPG patients versus the incidence in age- and sex-matched NF1 patients without OPGs (NF1 controls) from the Manchester NF1 Genetic Register (three to four NF1 controls per patient). Follow-up for controls was censored at the point when the matched NF1 OPG patients developed their second gliomas or MPNSTs, age of death, or last follow-up. Neither gliomas nor MPNSTs were screened radiologically in asymptomatic individuals. Most MPNSTs present symptomatically due to their aggressive nature and histologic confirmation is usually obtained. The gliomas may have been detected due to the presence of symptoms, but some may have been detected when neuroimaging was performed for another reason.
The average follow-up for the 58 patients with NF1 and OPGs was 18.6 years (median, 15.9 years; range, 0.03 to 58.4). Twenty-eight (48%) of these patients received treatment, of whom 18 (64%) had radiotherapy alone or in combination with other modalities. Details of radiation therapy were available for more than 70% of the irradiated group; total doses ranged from 25.5 to 50.0 Gy (Table 1). More patients received radiotherapy in Manchester, where this was a routine treatment. It is likely these irradiated patients were assessed as having more aggressive disease and other treatments were not believed to be appropriate. Seventeen of the 58 patients developed 21 second primary tumors (15 gliomas and six MPNSTs) during 1,029 person-years of follow-up (Table 2).
The location of the OPG was identified in 54 of the 58 patients. As expected, where there was just a single optic nerve (ON) involved, most patients (85%) did not receive radiotherapy. One child (ME12) with a single ON OPG received surgery followed by fractionated radiotherapy. Her tumor (OPG) recurred 8 years later and was again treated with radiotherapy. One adult received surgery and radiotherapy simultaneously. Table 2 shows the distribution of the OPGs in relation to age and treatment. Serial neuroimaging was reviewed by a neuroradiologist (J.G.) for 26 of the 29 patients seen as part of the long-term follow-up study. For the remainder, the OPGs and second tumor diagnoses were reliant on information from the clinical notes and cancer registry data. The median time after diagnosis of OPG for the development of second nervous system tumors was 12 years (range, 1 to 33 years). Nine of 18 patients who received radiotherapy after their OPG developed 12 second tumors in 308 person-years of follow-up (median time after radiotherapy, 14 years; range, 7 to 27 years). One patient (ME11) developed two separate MPNSTs in different regions on opposites sides of the face four years apart, and two patients (ME13 and ME15) developed an additional glioma and an MPNST. All nine were known to have had radiotherapy treatment in childhood (age range, 4 to 11 years). None of the three irradiated adults have developed second nervous system tumors. Eight of 40 patients who were not treated with radiotherapy developed nine tumors in 721 person-years of follow-up after diagnosis of their OPG (median time after diagnosis of OPG, 4 years; range, 1 to 14 years; Table 2). One (MS08) developed both an additional glioma and an MPNST diagnosed 4 and 5 years, respectively, after the initial OPG. The relative risk (RR) of a second nervous system tumor after radiotherapy was 3.04 (95% CI, 1.29 to 7.15). The RR of a second nervous system tumor after radiotherapy in those treated younger than age 15 years compared with the nonirradiated NF1 OPG group diagnosed younger than age 15 years was 5.53 (95% CI, 1.99 to 15.36; Table 2). We did not include a patient (MS01) who has become quadriplegic from an apparent tumor affecting both cerebral peduncles. Although the neuroradiologic opinion is that this is a second primary, extension from the original OPG could not be eliminated. If included as a second primary glioma, the relative risk of second tumors would have been even higher for the irradiated group. No deaths from second primary tumors occurred in the nonirradiated group, but five of the nine irradiated patients who developed second tumors died from their high-grade second tumors. To determine if the incidence of subsequent non-OP gliomas was any greater than that of gliomas expected in an NF1 population, we compared the incidence in the Manchester population-based NF1 cohort. Thirteen (31%) of the 45 Manchester NF1 OPG patients developed a subsequent non-OP glioma (median time after diagnosis of OPG, 7 years; range, 1 to 33 years; median for irradiated patients, 13 years; median for nonirradiated patients, 4 years). All of the non-OP gliomas were anatomically distinct from the optic pathway and there was no evidence of direct spread. There were significantly fewer gliomas in the control group (NF1 patients without OPGs; one of 166 patients; P < .0001). Excluding the 45 Manchester NF1 patients with an OPG from the 788 Manchester NF1 patients overall, seven (0.9%) of 743 NF1 patients had a second CNS glioma, despite the fact that more than 150 had had cranial imaging. More tumors were high grade in the radiotherapy group, but no aggressive gliomas arose within the OPG tumor itself. All five MPNSTs in NF1 OPG patients with radiotherapy arose in the radiation field, but the only MPNST in the nonirradiated group arose distant to the cranium (MS08 MPNST in ischiorectal fossa). The RR for MPNST after radiotherapy in NF1 OPG patients was 8.89 (95% CI, 1.07 to 74.0). There were four MPNSTs in the 12 irradiated Manchester NF1 OPG patients, compared with three MPNSTs in 48 NF1 controls (RR, 5.33; 95% CI, 1.37 to 20.7). There were three MPNSTs in 166 matched controls for all NF1 OPG patients (from the Manchester NF1 data set), when these controls were observed for the same duration as the OPG group. The risk of MPNST was not significantly increased in the nonirradiated NF1 OPG patients.
This longitudinal study demonstrates that there is a high incidence of second nervous system tumors in NF1 OPG patients, and this risk is even higher in patients whose OPGs have been irradiated. Although there is a radiologic classification system for OPGs,7 this is rarely used in practice. The retrospective blinded review of 26 patients' serial neuroimaging only excluded the diagnosis in three patients. In each of these, the diagnosis had been questioned previously. This provides some reassurance that the diagnosis in the great majority of patients is likely to be accurate. The cumulative risk of developing cancer after radiation exposure has been estimated at 0.6% to age 75 years, and the reports of therapeutic radiotherapy and subsequent benign and malignant tumors within the radiotherapy field are increasing.8-10 This risk is increased in NF1, where loss of the NF1 gene product, neurofibromin (a tumor suppressor gene), results in the presence of activated ras. Active ras inhibits apoptosis and stimulates cell proliferation, and NF1-related gliomas and MPNST have been show to lack neurofibromin expression. NF1 mice recently have been shown to be susceptible to radiation-induced malignancies, with sarcomas and breast cancer particularly prevalent.11 The tumors showed a typical pattern on comparative genomic hybridization with loss of the normal copy of NF1. It has also been shown recently that oncogenic RAS activation appears to predispose cells to genomic instability through both mitogen-activated protein kinase–dependent and –independent pathways that affect critical checkpoints in the cell cycle (G2/M).12 Therefore, this would increase the likelihood of accumulation of mutations and of second tumor induction. The reason for the baseline increase in other gliomas in NF1 OPG patients is not clear, but there may be genotype-phenotype correlations for the development of glioma in NF1. Ars et al13 found a clustering of truncating mutations in the 5' end of the NF1 gene in OPG patients. This 5' clustering was also observed in our cohort of 29 NF1 OPG patients, in whom 14 pathogenic changes were identified (manuscript in preparation). The resultant NF1 protein, if functional, would lack the ras–guanosine triphosphate activating protein domain. Gliomas are inducible with radiation,14 and the increased glioma risk in irradiated NF1 patients is consistent with a doubled glioma risk after radiotherapy in the general population,15 but the NF1 patients have a higher background risk. The median time to second tumor in the irradiated group of 13 years is consistent with radiation-induced tumors (compared with the median time of 4 years in nonirradiated patients).16-20 The increased risks of new primary tumors in NF1 are probably common to many tumor predisposition syndromes, although an additive effect of involvement of the ras pathway is still possible.20 It is possible that the irradiated group had more aggressive gliomas, and that these could have been associated with an increased risk of gliomas elsewhere. However, there was no evidence of spread from the original OPG. All six deaths in the irradiated group potentially were a result of sequelae to radiotherapy rather than the primary disease. An inherent complication of this study was the inability to confirm histologically the OPG diagnoses in most patients. Similarly, there are difficulties in confirming the second gliomas and in determining whether these and the secondary MPNSTs truly are induced by radiation therapy. Radiation-induced sarcomas have been reviewed in detail by Cahan et al.21 In their patients, they suggested that radiation-induced tumors should be considered when there was no previous malignancy in that area, and the new malignancy was in the field of radiotherapy, was confirmed histologically, and arose after a significant time period (average, 11 years; range, 5 to 22 years), which they believed suggested total cure of the previously treated tumor. In our study, the MPNSTs that we assume occurred due to radiotherapy can be shown to fulfil most of these criteria. All occurred in the radiotherapy field, mean duration post-treatment was 21 years, and tumors were histologically confirmed. It is unlikely MPNSTs were present at the time of OPG diagnosis because their aggressive tendency is unlikely to have remained dormant for so long. It is possible that they may have occurred anyway in these NF1 patients, but the increased frequency to that observed in the un-irradiated NF1 OPG patients (RR, 8.89) and in matched NF1 controls not known to have OPG (RR, 5.33) is significant. Although the radiation field was not evaluated, all four patients were irradiated in childhood. Two MPNSTs occurred in the facial region (ME11), two in the ipsilateral brachial plexus to OPG (ME13, ME15), and the radiation field in the last patient (GE03) was said to cover both optic pathways. Because of the small size of the patients and scatter of the radiation, it is likely that these areas did receive some irradiation. For second gliomas, the criteria suggested21 are more limited. As far as is possible to establish in those with OPGs, none of these second gliomas were present at OPG diagnosis. Three patients in whom an OPG occurred after a glioma were present in our initial cohort but were excluded from this second tumor analysis (Fig 1). Not all controls had neuroimaging, so we cannot exclude the possibility of under-reporting of asymptomatic tumors, as discussed. Histology was seldom obtained and the differential diagnosis of OPG was dependent on the radiologic review in most cases. However, the duration after radiotherapy was consistent with the suggested criteria (average, 14 years after radiotherapy; range 1 to 27 years)21 and that seen in radiation-induced tumors.16-20 MPNST is the most serious malignancy in NF1, with a lifetime risk of approximately 10%.4 Five-year survival in NF1 patients with MPNST is as low as 20%.4 In the present study, the risk of MPNST was increased more than five-fold in the NF1 OPG group, most significantly in those receiving radiotherapy. Even as recently as 2000, some clinicians were advocating the use of radiotherapy for OPG in general.22 The occasional reports of major benefits from radiotherapy in NF1 OPG are substantially outweighed by the risks. Radiotherapy may have a place in the treatment of the more aggressive sporadic OPGs, but this is not the case for NF1 OPG.23 Indeed, our data cause some concern about the exposure of NF1 children to high doses of diagnostic radiation. Repeated use of computed tomography scans in children probably should be avoided given the alternative modality of magnetic resonance imaging. In conclusion, the generally benign course of OPGs in NF11-3 compared with sporadic OPGs and the risk of vascular or endocrine complications2 argue against the use of radiotherapy. Thus, in children with NF1, radiotherapy should only be used if absolutely essential.
The authors indicated no potential conflicts of interest.
Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
1. Singhal S, Birch JM, Kerr B, et al: Neurofibromatosis type 1 and sporadic optic gliomas. Arch Dis Child 87:65-70, 2002 2. Listernick R, Darling C, Greenwald M, et al: Optic pathway tumors in children: The effect of neurofibromatosis type 1 on clinical manifestations and natural history. J Pediatr 127:718-722, 1995[CrossRef][Medline] 3. Guillamo JS, Creange A, Kalifa C, et al: Prognostic factors of CNS tumours in neurofibromatosis 1 (NF1): A retrospective study of 104 patients. Brain 126:152-160, 2003[CrossRef][Medline] 4. Evans DG, Baser ME, McGaughran J, et al: Malignant peripheral nerve sheath tumours in neurofibromatosis 1. J Med Genet 39:311-314, 2002 5. Burger PC: Pilocytic astrocytoma, in, Kleihues P, Cavanee WK (eds): World Health Organization Classification of Tumors, Pathology and Genentics: Tumors of the Nervous System. Lyon, France, IARC Press, 2000, pp 45-51 6. National Institutes of Health Consensus Development Conference Statement: Neurofibromatosis. Bethesda, Md., USA, July 13-15, 1987. Neurofibromatosis 1:172-178, 1988[Medline] 7. Dodge HW Jr, Love JG, Craig WM, et al: Gliomas of the optic nerves. AMA Arch Neurol Psychiatry 79:607-621, 1958[Medline] 8. Ron E, Modan B, Boice JD Jr, et al: Tumors of the brain and nervous system after radiotherapy in childhood. N Engl J Med 319:1033-1039, 1988[Abstract] 9. Brada M, Ford D, Ashley S, et al: Risk of second brain tumour after conservative surgery and radiotherapy for pituitary adenoma. BMJ 304:1343-1346, 1992 10. Sadetzki S, Flint-Richter P, Ben-Tal T, et al: Radiation-induced meningioma: A descriptive study of 253 cases. J Neurosurg 97:1078-1082, 2002[Medline] 11. Chao RC, Pyzel U, Fridlyand J, et al: Therapy-induced malignant neoplasms in Nf1 mutant mice. Cancer Cell 8:337-348, 2005[CrossRef][Medline] 12. Knauf JA, Ouyang B, Knudsen ES, et al: Oncogenic RAS induces accelerated transition through G2/M and promotes defects in the G2 DNA damage and mitotic spindle checkpoints. J Biol Chem 281:3800-3809, 2006 13. Ars E, Kruyer H, Morell M, et al: Recurrent mutations in the NF1 gene are common among neurofibromatosis type 1 patients. J Med Genet 40:e82, 2003 14. Salvati M, Frati A, Russo N, et al: Radiation-induced gliomas: Report of 10 cases and review of the literature. Surg Neurol 60:60-67, 2003[CrossRef][Medline] 15. Ron E: Cancer risks from medical radiation. Health Phys 85:47-59, 2003[CrossRef][Medline] 16. Schneider AB, Favus MJ, Stachura ME, et al: Salivary gland neoplasms as a late consequence of head and neck irradiation. Ann Intern Med 87:160-164, 1977 17. Friedman JM, Birch PH: Type 1 neurofibromatosis: A descriptive analysis of the disorder in 1,728 patients. Am J Med Genet 70:138-143, 1997[CrossRef][Medline] 18. de Vathaire F, Hawkins M, Campbell S, et al: Second malignant neoplasms after a first cancer in childhood: Temporal pattern of risk according to type of treatment. Br J Cancer 79:1884-1893, 1999[CrossRef][Medline] 19. Korones DN, Padowski J, Factor BA, et al: Do children with optic pathway tumors have an increased frequency of other central nervous system tumors? Neuro-oncol 5:116-120, 2003[Abstract] 20. Evans G, Birch JM, Ramsden R, et al: Malignant transformation and new primary tumours after therapeutic radiation for benign disease: Substantial risks in certain tumour-prone syndromes. J Med Genet 43:289-294, 2005[CrossRef][Medline] 21. Cahan WG, Woodard HQ, Higinbotham NL, et al: Sarcoma arising in irradiated bone: Report of eleven cases—1948. Cancer 82:8-34, 1998[CrossRef][Medline] 22. Grabenbauer GG, Schuchardt U, Buchfelder M, et al: Radiation therapy of optico-hypothalamic gliomas (OHG)–radiographic response, vision and late toxicity. Radiother Oncol 54:239-245, 2000[CrossRef][Medline] 23. Adams C, Fletcher WA, Myles ST: Chiasmal glioma in neurofibromatosis type 1 with severe visual loss regained with radiation. Pediatr Neurol 17:80-82, 1997[CrossRef][Medline] Submitted August 11, 2005; accepted March 7, 2006.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2006 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
