• No results found

Radiation therapy is associated with increased late mortality in long-term adult. survivors of childhood low grade glioma: A population based study.


Academic year: 2021

Share "Radiation therapy is associated with increased late mortality in long-term adult. survivors of childhood low grade glioma: A population based study."


Loading.... (view fulltext now)

Full text


Radiation therapy is associated with increased late mortality in long-term adult survivors of childhood low grade glioma: A population based study.

Rahul Krishnatry1,4,10*, Nataliya Zhukova1,4,5,8*, Ana Guerreiro Stucklin1*, Jason Pole8, Matthew Mistry4,5,7, Iris Fried9, Vijay Ramaswamy1,6, Ute Bartels1,6, Annie Huang1,4,6, Normand Laperriere1,10, Peter Dirks2,4, Shayna Zelcer11, Mariana Silva12, Donna L. Johnston13, Katrin Scheinemann14, Jason W An15, Paul C. Nathan1,6, Mark Greenberg1,6,8, David Malkin1,5,6,8, Cynthia Hawkins3,4, Pratiti Bandopadhayay16,17, Mark W. Kieran16, Peter E. Manley16, Eric Bouffet1,6 and Uri Tabori1,4,6,7.

*These individuals contributed equally to the manuscript. Affiliations:

Division of Hematology/Oncology1, Division of Neurosurgery2, Division of Pathology3, The Labatt Brain Tumor Research Centre4, Program in Genetics and Genome Biology5, Department of Pediatrics6, The Hospital for Sick Children and Institute of Medical Science7, University of Toronto, Toronto, Ontario, Canada; Pediatric Oncology Group of Ontario8, Toronto, Ontario, Canada; Hadassah Medical Centre9, Jerusalem, Israel; Radiation Medicine Program10, Princess Margaret Hospital, University of Toronto, Toronto, Ontario, Canada; Pediatric Hematology/Oncology, London Health Sciences Centre, London11, Ontario, Canada; Pediatric Hematology/Oncology, Kingston General Hospital12, Kingston, Ontario, Canada; Division of Hematology/Oncology, CHEO, CHEO Research Institute13, Ottawa, Ontario; Division of Hematology/Oncology14, McMaster Children's Hospital, Hamilton, Ontario, Canada; University of Alberta, Faculty of Medicine Undergraduate Program15, Alberta, Canada; Canada; Division of Pediatric Hematology/Oncology, Department of Pediatric Oncology16, Dana-Farber Cancer Institute, Boston Children’s Hospital, Boston, Massachusetts; Department of Cancer Biology17, Dana-Farber Cancer Institute, Boston, Massachusetts.



RK: is supported by the Society of NeuroOncology international research development fellowship. NZ: is supported by Pediatric Oncology Group of Ontario (POGO) fellowship.

This work is supported by grants from The B.R.A.I.N. Child Canada, The Pediatric Low-Grade Astrocytoma Foundation, Stop and Shop Pediatric Brain Tumor Program and The St Baldrick’s Foundation.

Corresponding author:

Dr. Uri Tabori, MD Staff Neuro-Oncologist

Division of Haematology/Oncology The Hospital for Sick Children The University of Toronto 555 University Avenue

Toronto, ON, Canada, M5G 1X8

Tel: (416) 813-7654; Fax: +1 416 813 8024 E-mail address: uri.tabori@sickkids.ca



PURPOSE: To uncover clinical risk factors and the impact of radiotherapy on children

and adult survivors of childhood low grade glioma (PLGG).

PATIENTS AND METHODS: We collected long-term follow-up information for all PLGG

patients diagnosed in all Ontario, Canada from 1985-2012 (n=1202) and determined factors affecting long-term survival. The impact of upfront (first line) radiation treatment on overall survival was determined in a discovery cohort of clinically relevant (incompletely resected) Ontario PLGG patients and an independent population based validation cohort from the Surveillance, Epidemiology and End Results (SEER) database (n=2402).

RESULTS: At a median follow-up of 12.73 (0.02-33) years; only 93 deaths (7.7%) were

recorded with 20 years overall survival (OS) of 90.1% + 1.1%. Children with NF1 had excellent survival and no tumor-related deaths during adulthood. Adverse risk factors included pleomorphic xanthoastrocytoma tumor type (p<0.001) and thalamic location (p<0.001). For patients surviving >5 years post-diagnosis, upfront radiotherapy was associated with three-fold increased risk of overall late deaths (p=0.001) and four-fold increased risk of tumor-related deaths (p=0.013). On multivariate analysis radiation therapy was the most significant factor associated with late all deaths and tumor-related deaths (HR 3 [1.3-7.0; p=0.012] and HR 4.4 [1.3-14.6; p=0.014] respectively). Similar association of radiotherapy and late deaths was observed in the validation cohort (p<0.001). In contrast to early deaths, late mortality was not associated with PLGG progression but rather tumor transformation (54%), secondary malignancy (11%) and other non-oncological causes (35%).

CONCLUSION: The chronic clinical course of PLGG is associated with excellent

long-term survival and is hampered by increased delayed mortality in patients receiving upfront radiotherapy. These observations should be considered while deciding treatment options for these patients.



Pediatric low grade gliomas (PLGG) are the most common brain tumor of childhood1

accounting for more than 35% of pediatric central nervous system malignancies. Children with PLGG may experience tumor progression multiple times and the resulting

morbidity remains a concern.1,2 Short (<5 years) and intermediate term (5-10 years)

overall survival is reported to be excellent and most children survive to adulthood.3-7

Since follow-up is often erratic for adult survivors of PLGG, convincing data are lacking

on very long-term outcome (>10 years post diagnosis).5 Specifically, information is

difficult to gather on the exact timing, causes of death and associated risk factors in the survivors during adulthood.

Multiple risk factors have been identified to be associated with progression-free and

overall survival of PLGG.3-7 These include patient factors (age at presentation or

diagnosis, gender and a preexisting diagnosis of neurofibromatosis type 1),8,9 and

tumor-related factors (presence of disseminated disease and certain pathological

subtypes).6,10,11 There are disparities between these studies partially due to differences

in pathological classification of PLGG and the emergence of new entities such as pilomyxoid astrocytoma and pleomorphic xanthroastrocytoma which were not

recognized in older studies.10-13 Furthermore, studies incorporating patients imaged

prior to the widespread use of magnetic resonance imaging (MRI)5 or from

non-population based referral centres lead to inherent selection bias.14,15

The impact of surgical resection and the use of radiation therapy have been extensively

studied.16-19 While complete surgical resection is consistently associated with improved


majority of non-resectable symptomatic PLGG patients were irradiated as the first line treatment (upfront treatment) and existing reports suggest excellent tumor control in a

significant proportion of these patients.16-19 However, in the context of a low-grade

tumor, especially in young children, long-term sequelae associated with radiation

treatment is a concern16,21,22. Follow up studies are lacking to document survival data in

PLGG patients during adulthood and the impact of risk factors and treatment on long term outcome.

In order to address these issues, we constructed a unique population-based database including all children diagnosed and treated for PLGG since widespread use of MRI in the province of Ontario, Canada; we validated our observations with a second large

independent cohort of patients for whom clinical outcomes are available.5 Our findings

present important insights which may affect future management and treatment approaches to PLGG in view of long-term outcome information.

Patients and Methods Patients

For this population-based, retrospective study, we assembled clinical data from a cohort of pediatric patients (0–18 years of age) diagnosed with a grade I/II glioma at any of the five pediatric cancer centers in Ontario between January 1, 1985 and October 1, 2012 (n=1202). The detailed characteristics for the entire cohort are shown in table 1. These are the only institutions that treat children with brain tumors in Canada’s largest province (population >12 million) and virtually all patients are exclusively treated at their residential location, therefore no selection bias is expected.


Data collection

Data on the primary tumor, treatment and follow-up information (childhood and adult) were available through the Pediatric Oncology Group of Ontario Network Information System (POGONIS). POGONIS was launched in 1985 and collects detailed demographic, disease, treatment and outcome data on all patients with a malignancy treated at any of Ontario’s five pediatric cancer centers prospectively. Through its team

of trained data managers, POGONIS captures 98% of cancer patients in Ontario23. The

Cancer Care Ontario (CCO) health system data resource - Ontario Cancer Registry (OCR) tracks clinical data related to important life events, including date and causes of death on all residents; this information is updated every month by health workers, with

accuracy of 95%24 and is also captured by POGONIS. These resources enabled us to

collect accurate long-term follow-up data for >95% of patients in real time. Methods of data procurement and the study design were approved by Research Ethics Board of The Hospital for Sick Children.

Each individual patient data were recorded until the last update in October 2012. Survival was defined as years from initial diagnosis to last date of follow-up or death as

on or before 31st Oct 2012. Factors such as gender, age at diagnosis, histopathology,

tumor location and upfront radiation treatment were used to stratify patients.

For histopathology stratification, only patients treated at The Hospital for Sick Children (SickKids) were considered where 683 out of 803 were diagnosed by biopsy. Neurofibromatosis type 1 (NF1) status was also available only for the Sickkids cohort. Detailed cause of death was collected for the Sickkids cohort (n=803). Further information is available in the supplementary methods.


To assess the impact of upfront radiation (defined as the primary medical treatment with or without surgery within one year of diagnosis) on long term survival, we used a

discovery cohort of clinically relevant (not completely resected)25 Ontario PLGG

patients. To focus on long term implications, only patients surviving the first 5 years post initial diagnosis and treatment were considered. An independent validation cohort of PLGG was available through the SEER cohort recently described by Bandopadhayay et

al.5 Similarly, only patients surviving at least 5 years were included for this analysis.

Tumor-related deaths were defined as glioma progression or transformation excluding secondary malignancies outside radiation field and other non cancer related mortality. Further information on methods used in this study is available in the supplemental materials and methods.

Statistical analysis

Statistical analysis was performed using Statistical Package for the Social Sciences (SPSS) v20. Survival was calculated using the Kaplan-Meier method and log-rank test. The hazard ratio was calculated by Cox regression analysis. In all cases p<0.05 was considered significant.


Long-term survival of PLGG

Complete clinical and survival data were available for 1202 patients with a median follow-up of 12.73 years (range, 0.02 – 33.33 years). Patient demographic distribution, tumor location, and follow-up duration were similar between participating centers (Supplemental Table S1).


Overall, there were 93 deaths (7.7%). Ten, 20 and 30-year overall survival (OS) were 93.1%+0.8%, 90.1%+1.1% and 88%+1.5% respectively (Figure 1A). Although excellent overall outcome was observed, late deaths occurred up to 23 years post initial diagnosis.

We did not observe significant difference in survival based on gender (Supplemental Figure S1) or age (Figure 1B), with the exception of children with optic pathway glioma (OPG). Young children with OPG diagnosed at age < 3 years had an inferior outcome as compared to older children (20-year OS was 80.4+6.1% and 97.9+1% respectively; p=0.007, Supplemental Figure S2).

Survival for NF1 children with PLGG

We identified 125 children with NF1 and PLGG of which 116 (92.8%) had OPG. Children with NF1 and tumors affecting the optic pathway had excellent survival outcome (20-year OS was 95.1+2.4%; Supplemental Figure S3A). All deaths occurred within the first 12 years following diagnosis and none after the age of 18. A total of four deaths noted: two unrelated to tumor or treatment, and two due to malignant transformation with upfront radiation treatment at 10.8 & 11.2 years after treatment. A total of four NF1 patients received upfront radiation treatment, of these two died of tumor transformation (as above) and two patients are alive at 16 and 26 years. Overall survival for patients with optic pathway glioma without NF1 was lower (20-year OS of 88.7+3.6%) but this did not reach statistical significance (p=0.06, Supplemental. Figure S3B).


Pilocytic astrocytoma was the most common subtype and exhibited excellent 20-year OS of 95.5%±1.7% (Figure 1C). No significant differences in survival were observed for most histological subgroups (ganglioglioma, mixed glioma and low grade glioma NOS). Although the number of patients was small and follow up was shorter for the more recently described subgroup of pilomyxoid astrocytoma, no deaths were observed at the median follow-up of 6.83 years (n=10, range 2.47-13.39 years). In contrast, children with pleomorphic xanthoastrocytoma (n=10) had significantly lower survival with a 10-year OS of 42.9 ± 20.8% (p<0.001; Figure 1c);

Tumor location and survival pattern

Posterior fossa tumors had an excellent 20-year OS of 96.8% ± 1.1%, similar to Cerebral hemispheric tumors (89.1+2.7%), OPG (90.6+2.2%) and spinal tumors (89.7+2.7%); while tumors of the brainstem (83.1+3.4%) and thalamus (74.2+6.3%) had significantly lower 20-year overall survival (p<0.001; Figure 1D). Initial (<5 years post diagnosis) high number of deaths in brainstem and thalamic tumors may be due to biopsy bias. However, long-term (10-20 years post diagnosis) follow-up revealed higher late mortality for thalamic tumors (Figure 1D). Examining other risk factors revealed that these thalamic tumors did not differ from tumors at other locations with respect to pathological subtype, age of presentation or gender. Furthermore, biopsy bias, cannot

explain these late deaths.

Association of upfront radiotherapy with long term survival

In order to examine other causes of late deaths in patients with thalamic tumors we specifically analyzed survival for children who survived >5 years post-diagnosis. Patients who received upfront radiation treatment showed significantly lower survival


than those who did not (OS at 20 years of 76.9+11.7% and 96.9+3.1%; p=0.039, Supplemental Figure 4).

To further explore this observation we carefully constructed a discovery cohort of clinically relevant Ontario PLGG patients for which total resection was not performed and survived the first 5 years (discovery cohort, n=607). Long-term OS for patients who received upfront radiation was significantly lower than patients who did not (20-year OS 84+3.9% versus 95.6%+1.2%; p<0.001, Figure 2A). Importantly, the use of upfront radiation treatment was associated with threefold increase in risk of all deaths (HR 3.3, 95% CI 1.6-6.6; p=0.001). Furthermore, for the Sickkids cohort (n=385) where comprehensive data on causes of death was available, tumor-related survival was significantly inferior for patients who received upfront radiation compared with those who did not (20-year OS 87.4+4.9% versus 97%+1.1%; p=0.007, Figure 2C) and upfront radiation treatment increased the risk of tumor-related death by four-fold (HR 3.9, CI 1.3-11.9; p=0.013). On multivariate analysis, upfront radiation treatment was the most significant factor associated with late overall deaths and tumor-related deaths (p=0.012 and p=0.014 respectively; Table 2).

To test whether the observed late deaths in patients treated with radiation is applicable to other population based cohorts we performed similar analysis on the SEER

database, recently reported by Bandopadhayay et al5 (validation cohort). Survival

patterns for the validation cohort mimicked the findings for our discovery cohort as shown in figure 2B for all deaths (20 year-OS of 84.4+1.9% and 95.3+0.8% for children treated with and without radiation respectively; p<0.001) and Figure 2D for tumor-related deaths (20 year-OS of 89+1.72% and 94.3+0.8% for children treated with and


without radiation respectively; p=0.001). The risk of late death for children undergoing radiation increased threefold for all deaths (HR 3.3, CI 2.3-4.7; p<0.001) and tumor-related deaths (HR 3.4, CI 2.1-5.5; p=0.001).

Early and late causes of death in PLGG

Overall, 55/803 patients from SickKids died during the study period. The most common causes of death were transformation to malignant glioma (n=26) and tumor progression (n=10). Three patients developed secondary malignancies and 16 died due to non-oncological causes. Complete information of cause of death is shown in supplementary Table S2. Half of all deaths occurred in the first five years from diagnosis (n=29/55; 52.7%) and 17 (30%) deaths occurred in adulthood (> 18 years age).

Causes of death differed significantly with time from diagnosis (Figure 3). While 81% of early deaths (<5 years after diagnosis) were due to tumor progression or transformation, none of the late deaths (>5 years after diagnosis) were due to tumor progression. Other non glioma related causes accounted for 46% of late deaths.


In this large population based study of children with PLGG; we were able to demonstrate excellent survival for these patients into adulthood, provide valuable information on patient subpopulations such as children with NF1 and identify various factors affecting survival which may be important in the management of these individuals. This study has the longest reported median follow-up time (12.73, 0.02-33 years) and to our knowledge is the first to utilize large population based inclusion of all patients diagnosed and followed up in MRI era with accurate information regarding treatment timings and causes of death.


Most large PLGG studies report 5-year progression free survival of 30-40% following

most treatment regimens.7,26-30 Since many of these patients experience multiple

progressions during the course of their disease and hence various lines of treatments, it may be tempting to apply a more aggressive regimen at progression to avoid patient demise. Despite multiple progressions, we observed >90% survival at 20-25 years post diagnosis in absence of aggressive modalities such as radiotherapy and no late mortality due to tumor progression. This observation adds a new dimension to recent studies which demonstrate excellent long-term outcome for these children and supports the notion that PLGG may be different in its clinical and biological behavior as

compared to adult counterpart.3,5,7

We did not observe any significant impact of the most common pathological subtypes on long-term overall survival. Important findings are the lack of detrimental outcome for pilomyxoid astrocytomas and the poor survival associated with pleomorphic xanthroastrocytoma. Pilomyxoid astrocytoma has been described as an aggressive type

of PLGG with young age of onset and poor prognosis.13,31,32 Since then several groups

have shown that survival for older patients diagnosed with this pathological subtype is

not necessarily poor.33,34 Our current study demonstrating excellent long-term survival

of these patients further supports these encouraging findings. In contrast, less than half of patients with pleomorphic xanthoastrocytoma in our study are long-term survivors. This worrisome observation warrants further confirmation.

We and others have reported high frequency of the BRAF/KIAA1549 fusion35,36 and

favorable outcome in PLGG.25 Recently, BRAF V600E mutations and CDNK2A


outcomes.36,37. These molecular signatures may help in accurate prediction of long term outcome and determine which patients may benefit from a less and more aggressive treatments accordingly.

Radiotherapy is still commonly used for the treatment of PLGG and has shown superior

progression-free survival (PFS) as compared to chemotherapy.17-19 However, long-term

survival post radiation is reported in very few studies and have shown associated

increased risk for vascular events and secondary radiation induced malignancies.22,38,39

Studies comparing patients who were irradiated upfront to non-radiation approaches are subject to selection bias at several levels. Firstly upfront radiation treatment is currently reserved for more aggressive PLGG. This is highlighted by the early deaths (1-3 years

post diagnosis) observed in these cohorts5. Second, these are usually deep seated

tumors for which only a small biopsy sample is available leading to biopsy bias where presumably a low grade glioma may actually represent a heterogeneous tumor with high & low grade components. This may also explain large number of tumor transformation related deaths in first few years of diagnosis (<5 years). Analysis of patients which survived 5 or more years bypasses many of these limitations and enables a more accurate analysis of late deaths possibly related to radiotherapy.

Our study raises major concerns regarding the association of radiotherapy with increased tumor-related late deaths. The unique pattern of late deaths in both the Ontario and SEER populations (US) support the validity of these findings. Since no plateau has been reached, the impact of vascular events and tumor related deaths may be even higher with longer follow-up. These observations further question the role of radiation in the management of this chronic childhood disease with excellent survival.


One of the most interesting findings in this study is the lack of late deaths associated with tumor progression (Figure 3). Indeed, none of the NF1 patients died in adulthood and none of the other PLGG adults survivors in this study died of tumor progression. These intriguing observations support the current biological hypothesis that in RAS/MAPK driven tumors, oncogene induced senescence causes spontaneous growth

arrest and tumor burnout in adolescent and young adults with PLGG.40,41 This

information supports the concept that tumor progressions in the first few years of PLGG course are not necessarily associated with worse overall patient outcome and one should try to avoid toxic treatments with long term sequelae.

This study has the classic limitations of a retrospective study spanning decades of treatment strategies, pathological diagnoses and the inherent limitations of the POGONIS and SEER data collection. Nevertheless, the consistency of results between centers and cohorts strongly supports the robust observations, unique features and natural history of PLGG.

In summary, this study reinforces the concept that PLGG is a chronic disease which extends into adulthood and the need to manage these children with view of long-term effects of various treatment options. Since most patients will survive at least into mid-adulthood, treatment options such as radiation therapy should be considered with caution. Specifically, although Radiation therapy may be associated with better tumor control in early years it may lead to significant devastating sequelae affecting not only quality of life but also long-term survival during adult life.


Future studies detailing biological characteristics may enable better risk stratification of patients and allow use of less toxic therapies in low risk patients to reduce long-term morbidity and mortality.



1. Ostrom QT, Gittleman H, Farah P, et al: CBTRUS statistical report:

Primary brain and central nervous system tumors diagnosed in the United States in 2006-2010. Neuro Oncol 15 Suppl 2:ii1-56, 2013

2. Rickert CH, Paulus W: Epidemiology of central nervous system tumors in

childhood and adolescence based on the new WHO classification. Childs Nerv Syst 17:503-11, 2001

3. Benesch M, Lackner H, Sovinz P, et al: Late sequela after treatment of

childhood low-grade gliomas: a retrospective analysis of 69 long-term survivors treated between 1983 and 2003. J Neurooncol 78:199-205, 2006

4. Armstrong GT, Conklin HM, Huang S, et al: Survival and long-term health

and cognitive outcomes after low-grade glioma. Neuro Oncol 13:223-34, 2011

5. Bandopadhayay P, Bergthold G, London WB, et al: Long-term outcome of

4,040 children diagnosed with pediatric low-grade gliomas: an analysis of the

Surveillance Epidemiology and End Results (SEER) database. Pediatr Blood Cancer 61:1173-9, 2014

6. Fisher PG, Tihan T, Goldthwaite PT, et al: Outcome analysis of childhood

low-grade astrocytomas. Pediatr Blood Cancer 51:245-50, 2008

7. Gnekow AK, Falkenstein F, von Hornstein S, et al: Long-term follow-up of

the multicenter, multidisciplinary treatment study HIT-LGG-1996 for low-grade glioma in children and adolescents of the German Speaking Society of Pediatric Oncology and Hematology. Neuro Oncol 14:1265-84, 2012


8. Khanani MF, Hawkins C, Shroff M, et al: Pilomyxoid astrocytoma in a patient with neurofibromatosis. Pediatr Blood Cancer 46:377-80, 2006

9. Binning MJ, Liu JK, Kestle JR, et al: Optic pathway gliomas: a review.

Neurosurg Focus 23:E2, 2007

10. Fernandez C, Figarella-Branger D, Girard N, et al: Pilocytic astrocytomas

in children: prognostic factors--a retrospective study of 80 cases. Neurosurgery 53:544-53; discussion 554-5, 2003

11. Fisher PG, Breiter SN, Carson BS, et al: A clinicopathologic reappraisal of

brain stem tumor classification. Cancer 89:1569-1576, 2000

12. Colin C, Padovani L, Chappe C, et al: Outcome analysis of childhood

pilocytic astrocytomas: a retrospective study of 148 cases at a single institution. Neuropathol Appl Neurobiol 39:693-705, 2013

13. Rao AA, Laack NN, Giannini C, et al: Pleomorphic xanthoastrocytoma in

children and adolescents. Pediatr Blood Cancer 55:290-4, 2010

14. Gajjar A, Heideman RL, Kovnar EH, et al: Response of pediatric low grade

gliomas to chemotherapy. Pediatr Neurosurg 19:113-8; discussion 119-20, 1993

15. Kumar RA, Khandelwal N, Sodhi KS, et al: Comparison between

contrast-enhanced magnetic resonance imaging and technetium 99m glucohepatonic acid single photon emission computed tomography with histopathologic correlation in gliomas. J Comput Assist Tomogr 30:723-33, 2006

16. Erkal HS, Serin M, Cakmak A: Management of optic pathway and

chiasmatic-hypothalamic gliomas in children with radiation therapy. Radiother Oncol 45:11-5, 1997


17. Marcus KJ, Goumnerova L, Billett AL, et al: Stereotactic radiotherapy for localized low-grade gliomas in children: final results of a prospective trial. Int J Radiat Oncol Biol Phys 61:374-9, 2005

18. Merchant TE, Kun LE, Wu S, et al: Phase II trial of conformal radiation

therapy for pediatric low-grade glioma. J Clin Oncol 27:3598-604, 2009

19. Saran FH, Baumert BG, Khoo VS, et al: Stereotactically guided conformal

radiotherapy for progressive low-grade gliomas of childhood. Int J Radiat Oncol Biol Phys 53:43-51, 2002

20. Shaw EG, Wisoff JH: Prospective clinical trials of intracranial low-grade

glioma in adults and children. Neuro Oncol 5:153-60, 2003

21. Grabenbauer GG, Schuchardt U, Buchfelder M, et al: Radiation therapy of

optico-hypothalamic gliomas (OHG)--radiographic response, vision and late toxicity. Radiother Oncol 54:239-45, 2000

22. Merchant TE, Conklin HM, Wu S, et al: Late effects of conformal radiation

therapy for pediatric patients with low-grade glioma: prospective evaluation of cognitive, endocrine, and hearing deficits. J Clin Oncol 27:3691-7, 2009

23. Greenberg ML, Barr RD, DiMonte B, et al: Childhood cancer registries in

Ontario, Canada: lessons learned from a comparison of two registries. Int J Cancer 105:88-91, 2003

24. Robles SC, Marrett LD, Clarke EA, et al: An application of

capture-recapture methods to the estimation of completeness of cancer registration. J Clin Epidemiol 41:495-501, 1988


25. Hawkins C, Walker E, Mohamed N, et al: BRAF-KIAA1549 fusion predicts better clinical outcome in pediatric low-grade astrocytoma. Clin Cancer Res 17:4790-8, 2011

26. Ater JL, Zhou T, Holmes E, et al: Randomized study of two chemotherapy

regimens for treatment of low-grade glioma in young children: a report from the Children's Oncology Group. J Clin Oncol 30:2641-7, 2012

27. Bouffet E, Jakacki R, Goldman S, et al: Phase II study of weekly

vinblastine in recurrent or refractory pediatric low-grade glioma. J Clin Oncol 30:1358-63, 2012

28. Jakacki RI, Bouffet E, Adamson PC, et al: A phase 1 study of vinblastine

in combination with carboplatin for children with low-grade gliomas: a Children's Oncology Group phase 1 consortium study. Neuro Oncol 13:910-5, 2011

29. Ronghe M, Hargrave D, Bartels U, et al: Vincristine and carboplatin

chemotherapy for unresectable and/or recurrent low-grade astrocytoma of the brainstem. Pediatr Blood Cancer 55:471-7, 2010

30. Scheinemann K, Bartels U, Tsangaris E, et al: Feasibility and efficacy of

repeated chemotherapy for progressive pediatric low-grade gliomas. Pediatr Blood Cancer 57:84-8, 2011

31. Ajani OA, Al Sulaiti G, Al Bozom I: Pilomyxoid astrocytoma of the

cerebellum. J Neurosurg Pediatr 7:539-42, 2011

32. Forbes JA, Mobley BC, O'Lynnger TM, et al: Pediatric cerebellar


33. Omura T, Nawashiro H, Osada H, et al: Pilomyxoid astrocytoma of the fourth ventricle in an adult. Acta Neurochir (Wien) 150:1203-6; discussion 1206, 2008

34. Toyoda K, Tsutsumi K, Ono T, et al: [Pilomyxoid astrocytoma presenting

in an elderly patient: a case report]. No Shinkei Geka 37:1123-8, 2009

35. Jones DT, Hutter B, Jager N, et al: Recurrent somatic alterations of

FGFR1 and NTRK2 in pilocytic astrocytoma. Nat Genet 45:927-32, 2013

36. Zhang J, Wu G, Miller CP, et al: Whole-genome sequencing identifies

genetic alterations in pediatric low-grade gliomas. Nat Genet 45:602-12, 2013

37. Matthew Mistry, Nataliya Zhukova, Daniele Merico, et al: BRAF mutation

and CDKN2A deletion define a clinically distinct subgroup of childhood secondary high grade glioma. JCO (In Press)

38. Dirks PB, Jay V, Becker LE, et al: Development of anaplastic changes in

low-grade astrocytomas of childhood. Neurosurgery 34:68-78, 1994

39. Bowers DC, Liu Y, Leisenring W, et al: Late-occurring stroke among

long-term survivors of childhood leukemia and brain tumors: a report from the Childhood Cancer Survivor Study. J Clin Oncol 24:5277-82, 2006

40. Jacob K, Quang-Khuong DA, Jones DT, et al: Genetic aberrations leading

to MAPK pathway activation mediate oncogene-induced senescence in sporadic pilocytic astrocytomas. Clin Cancer Res 17:4650-60, 2011

41. Bartkova J, Rezaei N, Liontos M, et al: Oncogene-induced senescence is

part of the tumorigenesis barrier imposed by DNA damage checkpoints. Nature 30:633-7, 2006


Figure Legends

Figure 1: Long term overall survival of children with pediatric low grade glioma (PLGG) (A) All study patients; (B) All PLGG stratified by age at diagnosis; (C) Sick Kids

PLGG stratified by pathology subtype. (D) All PLGG stratified by tumor location. Pilocytic astrocytomas (PA), Pilomyxoid astrocytoma (PMA), Pleomorphic

xanthoastrocytomas (PXA), Low Grade Gliomas Non specified (LGG), gangliogliomas and Mixed glioma including oligodendroglioma.

Figure 2: The effect of radiation therapy on long term survival of PLGG.

(A) Ontario discovery cohort including all deaths; (B) SEER validation cohort including all deaths; (C) SickKids discovery cohort tumor-related deaths; (D) SEER validation cohort including tumor-related deaths.


Table 1: Study population and characteristics. Ontario cohort (n=1202) Male 637 (53%) Female 565 (47%) Age at Diagnosis(years) Median (Range) 7.26 (0 – 17.87) < 3 222 (18.5%) 3 - < 8 421 (35%) 8 - < 14 407 (33.9%) 14 - 18 152 (12.6%) Location Posterior fossa 374 (31.1%) Optic pathway 301 (25%) Lateral/peripheral brain 211 (17%) Brainstem 158 (13.1%) Thalamic 78 (6.5%) Spinal cord 49 (4.1%) Midline other 15 (1.2%) Other 16 (1.3%) *Pathology N=683 Pilocytic astrocytoma 284 (41.5%)

Low grade astrocytoma NOS 277 (40.5%)

Ganglioglioma 61 (8.9%)

Pilomyxoid astrocytoma 10 (1.5%)

Pleomorphic xanthoastrocytoma 10 (1.5%)

Oligodendroglioma (mixed glial) &

others 41 (6.1%)


Alive 1109 (92.3%)

Dead 93 (7.7%)


Table 2: Cox regression model for multivariate analysis in discovery cohort. All Deaths Covariable Hazard ratio 95% Confidence limit p value Age (<3 v >3 years) 1.1 0.4 to 2.7 0.920

Location (Thalamic v other) 2.3 0.9 to 6 0.092


Grade 1 astrocytoma v other NF1 not biopsied v other

1.6 0.3 0.5 to 5.1 0.04 to 2.4 0.380 0.259 Upfront Radiation treatment

(Yes v No) 3 1.3 to 7 0.012 Tumor-related Deaths Covariable Hazard ratio 95% Confidence limit p value Age (<3 v >3 years) 0.6 0.1 to 2.9 0.547

Location (Thalamic v other) 2.4 0.6 to 9.4 0.207


Grade 1 astrocytoma v other NF1 not biopsied v other

2.8 0.9 0.7 to 11.2 0.1 to 8.1 0.152 0.910 Upfront Radiation treatment

(Yes v No)


0.00 0.25 0.50 0.75 1.00 0 5 10 15 20 25 30

Overall Survival (years)

OS probability 1202 922 723 504 266 56 1 No. at risk p = 0.393 0.00 0.25 0.50 0.75 1.00 0 5 10 15 20 25 30

Overall Survival (years)

OS probability 0-3 >3-8 >8-14 >14-18 223 166 123 84 45 11 0 421 338 270 195 109 24 1 405 313 248 172 81 17 0 153 105 82 53 31 4 0 0-3 >3-8 >8-14 >14-18 No. at risk p < 0.0001 0.00 0.25 0.50 0.75 1.00 0 5 10 15 20 25 30

Overall Survival (years)

OS probability Ganglioglioma LGA Mixed Glioma PA PMA PXA 0.00 p < 0.0001 0.25 0.50 0.75 1.00 0 5 10 15 20 25 30

Overall Survival (years)

OS probability Brainstem Cerebral Hemispheres Midline,other Optic Pathway Posterior Fossa Spinal Thalamus A) D) C) B) Figure 1


A) C) Figure 2 p = 0.000375 0.00 0.25 0.50 0.75 1.00 5 10 15 20 25 30

Overall Survival (years)

OS Probability No Radiation Radiation 497 382 255 125 27 1 No Radiation 110 94 72 50 11 0 Radiation No. at risk p < 0.0001 0.00 0.25 0.50 0.75 1.00 5 10 15 20 25 30 35

Overall Survival (years)

OS probability No Radiation Radiation 1909 945 539 244 113 51 1 518 398 334 239 148 66 3 No Radiation Radiation No. at risk p < 0.0001 0.00 0.25 0.50 0.75 1.00

Tumour Related OS probability

No Radiation Radiation B) D) p = 0.00718 0.00 0.25 0.50 0.75 1.00 No Radiation Radiation


Related documents

Examples of enhanced translation of mRNAs containing uORFs con- comitant with eIF2 α phosphorylation include the induc- tion of the transcription factors ATF4 in mouse

In this study, the impact of science fairs, which is one of the informal learning environments, on the scientific epistemological beliefs and scientific perceptions of adults

Implementation of land value taxation almost immediately increased lo- cal revenues (since 2005, in Russia 100 percent of land tax is assigned to municipal budgets), despite the

Figure 12: RGB false color visualization of first three eigenimages from Pavia hyperspectral image: (a) classical PCA on spectral bands, (b) scale- decomposition MPCA, (c)

Specifically, the reason- able investor would likely want to know, in this instance, about the extent that the companies in question are producing or planning to produce

Some of the solutions can include (but are not lim- ited to) steps such as image preprocessing, face detec- tion, facial feature point detection, face preprocessing for illumination

ns you should given to th will stop y ore the resul on descriptio This allows asks | Run B ulations con o “Beginning lts page. To alter th ge. Any chan however yo e length of t