Canadian brain cancer survival rates by tumour type and region: 1992-2008.

Yuan, Yan ; Shi, Qian ; Li, Maoji 等

There is little information available on brain tumour incidence and survival among the Canadian population. (1,2) Available information provides an overall estimate of brain cancer survival without an assessment of patterns by age, sex, region or tumour subtype; these patterns may be influenced by clinical and policy decisions regarding treatment. It is not well recognized by the general population that there is wide variation in the prognosis of patients with brain cancers, depending on the tumour histology type, patient and clinical features, all of which may influence diagnosis and treatment. Based on information from many other countries, (3-7) it is well established that early age at diagnosis is associated with better prognosis for patients with these tumours, largely because the histology types that occur most frequently in younger age groups have a less aggressive nature. Mao (8) demonstrated an improvement in brain cancer survival rates in the Canadian province of Saskatchewan between 1967 and 1986 that was primarily due to survival improvement in patients under the age of 65. The objective of our study is to investigate survival patterns by region and histology among Canadian patients with malignant brain tumours diagnosed between 1992 and 2008, while adjusting for age and sex. Due to the regional nature of health care systems and health care guidelines in Canada, some provincial differences are expected to emerge. This information will allow health care providers and researchers to explore the reasons underlying regional differences in survival rates and make evidence-informed decisions about clinical guidelines and health care policy with this patient population.

METHODS

Cohort selection

Data were acquired after the combined approval of Statistics Canada and the University of Alberta's ethics board. Administrative data were obtained from the Canadian Cancer Registry (CCR, 2012 release) for patients with brain tumours (International Classification of Disease for Oncology (ICD-O) 2nd and 3rd edition topography codes C700-C729 and C751-C753) who were diagnosed between 1992 and 2008 across all Canadian provinces and territories. We excluded patients whose brain tumours are non-malignant (ICD-O-2/3 behaviour code 0/1/2, n = 10,970) and whose chronologic sequence numbers of multiple primaries are not one (CCR code TD2, n = 2185). Thus our cohort consists of patients with first-ever primary malignant brain cancer diagnosis between 1992 and 2008.

Survival time is the outcome of interest and age, sex, geographical region, diagnostic method and tumour histology were independent variables. We collapsed some regions together due to small sample size: the Atlantic provinces (New Brunswick, Nova Scotia, Prince Edward Island, and Newfoundland and Labrador), the Territories (Northwest Territories, Nunavut and Yukon) and two Prairie provinces (Manitoba and Saskatchewan). We used CCR code T25 to categorize diagnostic methods broadly into "microscopically confirmed" and "non-microscopically confirmed" for patients diagnosed between 2004 and 2008. For patients diagnosed between 1992 and 2003, categories (CCR code T11) include "radiology or laboratory diagnosis other than histology, autopsy and cytology", "Surgery (without histology), or clinical diagnosis", "death certificate only" and "method of diagnosis unknown", all of which are classified as "non-microscopically confirmed". Histology types were grouped according to codes outlined by the Central Brain Tumor Registry of the United States (CBTRUS), (9) and the following seven histology types were selected for detailed analysis based on having 800 or more cases: glioblastomas (GBM), diffuse astrocytomas, gliomas (not otherwise specified)--hereinafter referred to as gliomas (NOS), oligodendrogliomas, anaplastic astrocytomas, oligoastrocytic tumours, and anaplastic oligodendrogliomas. All other histology types are collapsed into the "all others" category as an eighth histology type category.

Data analysis

The incidence of brain cancer and all-causes death was tabulated by sex, age groups, regions, diagnostic method and histology. Crude Kaplan-Meier survival estimates were calculated at one, two and five years for each histology group. We chose the earliest death clearance cut-off date among all provinces, December 31, 2008, as the censoring date. The observed survival (OS) estimates were reported instead of the relative survival (RS) estimates, mainly because these two rates are very close. The five-year survival estimates are 25% (RS) vs. 24% (OS) for brain cancer patients 15 years and older diagnosed in 2006-2008. (10)

Since the proportional hazard (PH) assumption does not hold for the variable age group for any histology type, age-stratified Cox PH models were fitted for each histology type to estimate the adjusted hazard ratios for regions. However, the stratification increases the uncertainty in the estimation of model-based survival rates. To obtain unbiased and efficient model-based age-, sex- and region-specific 5-year survival rate estimates, separate time-specific generalized linear models were used for each of the histology types. (11,12) Our primary analyses used all cases regardless of their diagnosis method.

Estimates for territories were not reported from the crude Kaplan-Meier and model-based analyses due to their small sample sizes. Survival estimates for the <20 years age group from the model-based analyses were not reported, in order to concentrate on the Canadian adult population. All analyses were performed using statistical software SAS 9.4 (SAS Institute, Cary, NC) and R 3.1.3. (13) All frequencies and proportions presented are subject to rounding in accordance with Statistics Canada requirements.

RESULTS

A total of 38,095 patients were diagnosed with first-ever primary malignant brain tumours between 1992 and 2008 in Canada. There is no improvement in survival rates over time during the study period (data not shown). The frequencies and the corresponding observed deaths are summarized in Table 1 for age groups, sex, regions and diagnostic method. Tumours occurred more frequently in males (56%) than in females (44%), and in older than in younger age groups. Patients over the age of 65 years accounted for 37% of the study population and 46% of deaths, while those under age 20 years made up 11% of the study population and accounted for less than 5% of the deaths. The frequency of patients by region reflected the population size in each province; the largest proportion of patients were diagnosed in Ontario (39%) and Quebec (26%), followed by British Columbia (12%). Manitoba and Saskatchewan combined accounted for 7.0% of brain cancers.

Histology type-specific incidence, death and 1-, 2- and 5-year survival estimates are shown in Table 2. The most common histology was GBM (37%) followed by diffuse astrocytomas (15%). In descending order, the estimated 5-year survival rates are: 65% (95% CI: 62.5%-67.4%) for oligodendrogliomas, 46% (95% CI: 42.6%-49.3%) for oligoastrocytic tumours, 41.5% (95% CI: 37.9%-45.0%) for anaplastic oligodendrogliomas, 33.9% (95% CI: 31.6%-36.2%) for gliomas (NOS), 26.6% (95% CI: 25.4%-27.8%) for diffuse astrocytomas, 18.2% (95% CI: 15.8%-20.7%) for anaplastic astrocytomas, and 4.0% (95% CI: 3.7%-4.3%) for GBM.

Figure 1 displays histology type-specific Kaplan-Meier survival curves for brain cancer patients. Long-term prognosis was best for patients with oligodendrogliomas and all other tumours, and poorest for patients with GBM. Survival curves show a poor survival experience within the first few years of diagnosis for gliomas (NOS), diffuse astrocytomas, anaplastic astrocytomas and GBM, and a better survival experience for oligoastrocytic tumours, anaplastic oligodendrogliomas, oligodendrogliomas and all other tumours.

[FIGURE 1 OMITTED]

The region-specific Kaplan-Meier survival curves are shown in supplementary materials (Supplementary Figure A--see ARTICLE TOOLS section on journal website). Using Ontario as the reference province, the adjusted hazard ratio estimates of regions are shown in Figure 2 for the eight histological types. The patterns of hazard ratios were not consistent by region across the histology types. Compared to Ontario, the estimated hazard rates were significantly higher in all other provinces for GBM and diffuse astrocytomas (p < 0.001), and higher for gliomas (NOS) in British Columbia, Alberta and the Atlantic provinces (p < 0.01). Hazard rate estimate was significantly higher for anaplastic astrocytomas in Quebec (p < 0.05), for oligoastrocytic tumours in British Columbia (p < 0.01), and for all other tumours in British Columbia, the Atlantic provinces and the Prairie provinces (p < 0.01) when compared to Ontario. Estimated hazard rates were significantly lower for anaplastic oligodendrogliomas in Quebec (p < 0.05) and marginally lower for oligoastrocytic tumours in the Atlantic provinces (p < 0.10) when compared to Ontario. When we restricted the analyses to microscopically-confirmed cases only, the regional survival patterns remain largely unchanged in all histology types except for gliomas (NOS) and "all others".

[FIGURE 2 OMITTED]

Table 3 shows 5-year region-, age group- and sex-specific survival rate estimates for selected histology types - i.e., excluding the gliomas (NOS) and "all others" - obtained from the time-specific generalized linear models using all cases. The estimates for gliomas (NOS) and "all others" were not provided because we observed large discrepancy between estimates from the all-case analysis and the microscopically-confirmed cases only analysis. Survival estimates were significantly higher for females than for males in all histology groups with the exception of GBM and anaplastic oligodendroglioma, where survival rate differences by sex were not statistically significant. Estimated survival decreased significantly as age-at-diagnosis increased from 21-44 years to 45-64 years, and to 65 years and over, in all histology groups studied. Compared to Ontario, the age- and sex-adjusted 5-year GBM survival estimates were lower in British Columbia, Alberta and the Prairie provinces (p < 0.01), and lower in the Prairies for anaplastic astrocytomas (p<0.05). Survival estimates were significantly higher for oligodendrogliomas in Alberta (p < 0.001), and for anaplastic oligodendrogliomas in Alberta and Quebec (p < 0.05). The patterns of these 5-year survival rates by province(s) were similar but not the same when compared to the patterns of hazard rates by province(s) estimated from the Cox proportional hazard model. This is to be expected, as the Cox model assumes proportion hazard throughout the entire follow-up time while the time-specific generalized linear model takes a snap shot of the survival status at 5 years post diagnosis. Repeating Table 3 restricted to microscopically-confirmed cases provided estimates that were similar to those reported here, except for diffuse astrocytomas of all age groups in Quebec and oligoastrocytic tumours for age group 45-64 in Quebec and the Prairie provinces (lower by 2 percentage points or more in the microscopically-confirmed cases only analysis).

DISCUSSION

As expected, observed brain cancer survival rates tend to be higher in females than in males and in younger than in older age groups. These data are limited to patients with malignant brain tumours. Information on non-malignant tumours has not historically been available in most provinces, (1) so survival rates for these tumour categories await the accrual of prospective data.

Survival and hazard rate patterns by region and histology type suggest that there may be a number of underlying reasons for variation. There are known differences in how Quebec vs. other provinces records date of diagnosis, (14) but we anticipate this survival underestimate to be rather small in the context of the long-term rates presented. It is possible that variation in clinical presentation, diagnostic classification, patient care and treatment decisions (by the patient, the physician and/or due to system differences) may explain some of the variation in provincial rates. No treatment variables were available in the CCR data, which is a limitation of the data sources. Thus we were not able to assess its effect on survival and whether it confounds the relationship of survival rate and region. Other clinical factors, including tumour size and other prognostic factors, have not been considered in this analysis. We know that the health care system is driven by provincial policy and that clinical guidelines and chemotherapy drug coverage vary across provinces, factors that will be indirectly reflected in these data. For example, in current clinical practice, Avastin is a relatively new drug used for GBM treatment that is covered by insurance in Manitoba but not in Alberta. We also know that factors in the cancer care continuum may be present, such as access to primary care, access to medical specialist and/or technology, delay in the referral process during cancer diagnostics and/or treatment phase, lack of care coordination, and adherence to treatment guidelines. Similar regional variation for other cancers in health care has been reported in Canada, the Netherlands and Italy, (15-17) indicating a need for routine monitoring across provinces/insurance providers.

We recognize that if histology is not classified consistently across provinces, then analyses by histology may be subject to measurement error and bias, contributing to the observed variable regional survival rates. The data did suggest possible variation in histology classification across regions. For example, British Columbia has a much higher proportion of diffuse astrocytomas and a much lower proportion of GBM than the rest of the country. It is possible that the BC survival rates reflect more advanced GBM tumours (classified as GBM) and less advanced GBM tumours (classified in the diffuse astrocytoma category) that would lower the estimated survival rates in both categories. Ontario has a much lower proportion of diffuse astrocytomas and a higher proportion of all other tumours. It is possible that some diffuse astrocytomas were classified into the all other tumours category, which could distort the true survival rates of both categories. The variation in microscopic confirmation, particularly for "all others" and gliomas (NOS) categories, also suggests some diagnostic variation which would be reflected in the histological classifications. Further assessment is needed to understand the variation in classification across regions. Going forward, surveillance data may need to incorporate the new molecular classification systems emerging for this type of information, which is expected to be increasingly clinically relevant. (18)

As the most populated provinces--Ontario and Quebec--tended to have the better survival rates, one might speculate that providing patient care for a rare disease is more effective when the number of patients is larger; but the better survival rates for the mixed tumours with oligo features outside of Ontario suggests that other health system factors may also be at play. Provinces with populations over large geographic regions may have urban/rural or socio-economic level factors that indirectly influence disease outcomes that have not been addressed in our analyses.

The international literature on overall brain cancer survival comes from a large number of regions: Europe, the UK, the Nordic countries, Australia, the US and Korea, and tends to be consistent in reporting improved survival rates beginning in the 1970s (3-8) through to the present. (19) In Canada, an improvement in survival rates for brain cancer patients between 1967 and 1986 was reported in the province of Saskatchewan which was attributed to patients under the age of 65 years. (8) In contrast, the UK reported an unexpected decline in survival during the late 1980s and 90s, with better survival rates emerging in males and younger age groups a decade later. (20) Histology-specific analysis such as that reported here has begun to explain some of these patterns by age and is needed to monitor changes in the histology-specific treatments now available. A recent analysis of GBM survival in the US suggests that the survival rates for these tumours did not progress much in the past three decades, (21) which agrees with our observation of no overall survival improvement for brain cancers from 1992 to 2008. The benefits of clinical trials, which changed the standard of practice for GBM, may now become apparent at the population level. US patients diagnosed in 2005-2006 had a 30% 2-year survival rate for GBM compared to 18% in 2000-2001. (21) Survival data on a histology-specific basis have not been available for the Canadian population previously. (1)

The ranking of these Canadian survival rates by histology is similar to that reported in CBTRUS for US data between 2005 and 2009. (22) The survival rates reported here are lower than those reported in the US for all histology groups. As noted above, there are differences in patient access to health insurance, access to care, clinical care guidelines and socio-cultural factors that may influence physician and patient decisions, reflecting these between-country survival differences. These data within Canada suggest the need to prioritize how to best approach implementing current evidence to improve brain cancer patient survival within Canada.

DISCLAIMER

This research was supported by funds to the Canadian Research Data Centre Network (CRDCN) from the Social Sciences and Humanities Research Council (SSHRC), the Canadian Institute for Health Research (CIHR), the Canadian Foundation for Innovation (CFI), and Statistics Canada. Although the research and analysis are based on data from Statistics Canada, the opinions expressed do not represent the views of Statistics Canada.

REFERENCES

(1.) Davis F, Nagamuthu C, Ross J, Megyesi J. Current status of brain tumour surveillance in Canada and why it matters. J Registry Manage 2015; 42(4): 139-45. PMID: 26938650.

(2.) Shaw A, Woods R, Semenciw R, Megyesi J. CNS tumours in Canada: Who are we missing? NAACCR 2014 Annual Conference: Capitalizing on Cancer Surveillance Data for Improved Cancer Control. Ottawa, 2014.

(3.) Shugg D, Allen BJ, Blizzard L, Dwyer T, Roder D. Brain cancer incidence, mortality and case survival: Observations from two Australian cancer registries. Int J Cancer 1994; 59(6):765-70. PMID: 7989116. doi: 10.1002/(ISSN)1097-0215.

(4.) Sant M, Van Der Sanden G, Capocaccia R. Survival rates for primary malignant brain tumours in Europe. Eur J Cancer 1998; 34(14):2241-47. doi: 10.1016/S0959-8049(98)00336-0.

(5.) Davis FG, Freels S, Grutsch J, Barlas S, Brem S. Survival rates in patients with primary malignant brain tumors stratified by patient age and tumor histological type: An analysis based on Surveillance, Epidemiology, and End Results (SEER) data, 1973-1991. J Neurosurg 1998; 88(1):1-10. PMID: 9420066. doi: 10.3171/jns.1998.88.1.0001.

(6.) Bray F, Engholm G, Hakulinen T, Gislum M, Tryggvadottir L, Storm HH, et al. Trends in survival of patients diagnosed with cancers of the brain and nervous system, thyroid, eye, bone, and soft tissues in the Nordic countries 1964-2003 followed up until the end of 2006. Acta Oncol 2010; 49(5):673-93. PMID: 20192877. doi: 10.3109/02841861003610200.

(7.) Jung K, Yoo H, Kong H, Won Y, Park S, Lee SH. Population-based survival data for brain tumors in Korea. J Neurooncol 2012; 109(2):301-07. PMID: 22660961. doi: 10.1007/s11060-012-0893-5.

(8.) Mao Y, Desmeules M, Semenciw RM, Hill G, Gaudette L, Wigle DT. Increasing brain cancer rates in Canada. CMAJ 1991; 145(12):1583-91. PMID: 1742695.

(9.) Ostrom QT, Bauchet L, Davis FG, Deltour I, Fisher JL, Langer CE, et al. The epidemiology of glioma in adults: A "state of the science" review. Neuro Oncol 2014; 16(7):896-913. PMID: 24842956.

(10.) Canadian Cancer Society's Advisory Committee on Cancer Statistics. Canadian CancerStatistics 2015. Toronto, ON: Canadian Cancer Society, 2015.

(11.) Zhou QM, Zheng Y, Chibnik LB, Karlson EW, Cai T. Assessing incremental value of biomarkers with multi-phase nested case-control studies. Biometrics 2015; 71(4):1139-49. PMID: 26195245. doi: 10.1111/biom.12344.

(12.) Uno H, Cai T, Tian L, Wei L. Evaluating prediction rules for t-year survivors with censored regression models. J Am Stat Assoc 2007; 102(478):527-37. doi: 10.1198/016214507000000149.

(13.) R Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria: The R Foundation for Statistical Computing. 2015. Available at: http://www.R-project.org/ (Accessed May 12, 2016).

(14.) Ellison LF, Gibbons L. Five-year relative survival from prostate, breast, colorectal and lung cancer. Health Rep 2001; 13(1):21-34.

(15.) Canadian Partnership Against Cancer. Prostate Cancer Control in Canada: A System Performance Spotlight Report. Toronto, ON: CPAC, 2015.

(16.) Wouters MWJM, Siesling S, Jansen-Landheer ML, Elferink MA, Belderbos J, Coebergh JW, Schramel Fm. Variation in treatment and outcome in patients with non-small cell lung cancer by region, hospital type and volume in the Netherlands. EurJSurgOncol 2010; 36:S83-92. PMID: 20598845. doi: 10.1016/j.ejso.2010.06.020.

(17.) Sant M, Minicozzi P, Allemani C, Cirilli C, Federico M, Capocaccia R, et al. Regional inequalities in cancer care persist in Italy and can influence survival. Cancer Epidemiol 2012; 36(6):541^7. PMID: 22770694. doi: 10.1016/j.canep.2012.06.006.

(18.) Louis DN, Perry A, Burger P, Ellison DW, Reifenberger G, von Deimling A, et al. International Society of Neuropathology-Haarlem Consensus Guidelines for Nervous System Tumor Classification and Grading. Brain Pathol 2014; 24(5):429-35. PMID: 25345897. doi: 10.1111/bpa.12171.

(19.) Kohler BA, Ward E, McCarthy BJ, Schymura MJ, Ries LA, Eheman C, et al. Annual report to the nation on the status of cancer, 1975-2007, featuring tumors of the brain and other nervous system. J Natl Cancer Inst 2011; 103(9):714-36. PMID: 21454908. doi: 10.1093/jnci/djr077.

(20.) Rachet B, Mitry E, Quinn MJ, Cooper N, Coleman MP. Survival from brain tumours in England and Wales up to 2001. Br J Cancer 2008; 99(suppl 1): S98-101. PMID: 18813276. doi: 10.1038/sj.bjc.6604603.

(21.) Koshy M, Villano JL, Dolecek TA, Howard A, Mahmood U, Chmura SJ, et al. Improved survival time trends for glioblastoma using the SEER 17 population-based registries. J Neurooncol 2012; 107(1):207-12. PMID: 21984115. doi: 10.1007/s11060-011-0738-7.

(22.) Dolecek TA, Propp JM, Stroup NE, Kruchko C. CBTRUS statistical report: Primary brain and central nervous system tumors diagnosed in the United States in 2005-2009. Neuro Oncol 2012; 14(suppl 5):v1-49. PMID: 23095881. doi: 10.1093/neuonc/nos218.

Received: July 29, 2015

Accepted: November 19, 2015

Yan Yuan, PhD, Qian Shi, MSc, Maoji Li, MMath, Chenthila Nagamuthu, MPH, Ellie Andres, MPH, Faith G. Davis, PhD

Author Affiliations

School of Public Health, University of Alberta, Edmonton, AB

Correspondence: Yan Yuan, PhD, School of Public Health, University of Alberta, 3-299 Edmonton Clinic Health Academy, 11405 87 Avenue, Edmonton, AB T6G 1C9, Tel: 780-248-5853, E-mail: yyuan@ualberta.ca

Acknowledgements: This research was conducted with funding from Brain Tumor Foundation of Canada.

Conflict of Interest: None to declare. Table 1. Frequency of malignant first-ever primary brain tumour diagnosis and death by sex, age, region and diagnostic method during 1992-2008 in Canada N Death Overall 38,095 28,460 Sex Male 21,260 16,140 Female 16,835 12,325 Age (years) [less than or equal to] 20 4080 1315 21-44 7370 3875 45-64 12,660 10,240 [greater than or equal to] 65 13,985 13,030 Region Ontario 14,700 10,265 Quebec 9920 7570 British Columbia 4705 3775 Alberta 3210 2415 Atlantic provinces * 2950 2340 Prairie provinces ([dagger]) 2550 2060 Territories ([double dagger]) 60 35 Diagnostic method Microscopically confirmed 28,720 21,440 Non-microscopically confirmed 9375 7020 * Nova Scotia, New Brunswick, Prince Edward Island, and Newfoundland and Labrador. ([dagger]) Manitoba and Saskatchewan. ([double dagger]) Northwest Territories, Nunavut and Yukon. Table 2. Histology type-specific incidence, death and Kaplan-Meier survival estimates at one, two and five years post diagnosis for brain cancer patients Histology N Death Kaplan-Meier survival estimate 1-year % (95% CI) Glioblastoma (GBM) 14,120 13,340 26.5 (25.7-27.2) Diffuse astrocytoma 5680 4475 48.9 (47.6-50.2) Glioma, NOS 1845 1245 49.5 (47.1-51.8) Oligodendroglioma 1715 785 87.5 (85.8-89.0) Anaplastic astrocytoma 1100 900 47.9 (44.9-50.9) Oligoastrocytic tumour 1030 590 78.0 (75.3-80.4) Anaplastic 885 580 76.5 (73.6-79.2) oligodendroglioma All others 11,725 6545 59.9 (59.0-60.8) Overall 38,095 28,460 47.0 (46.5-47.5) Histology Kaplan-Meier survival estimate 2-year % (95% CI) 5-year % (95% CI) Glioblastoma (GBM) 9.5 (9.0-10.0) 4.0 (3.7-4.3) Diffuse astrocytoma 36.1 (34.8-37.4) 26.6 (25.4-27.8) Glioma, NOS 40.5 (38.2-42.8) 33.9 (31.6-36.2) Oligodendroglioma 80.7 (78.7-82.5) 65.0 (62.5-67.4) Anaplastic astrocytoma 33.1 (30.3-36.0) 18.2 (15.8-20.7) Oligoastrocytic tumour 63.0 (59.9-66.0) 46.0 (42.6-49.3) Anaplastic 61.1 (57.7-64.3) 41.5 (37.9-45.0) oligodendroglioma All others 54.6 (53.7-55.6) 46.8 (45.8-47.7) Overall 35.2 (34.7-35.6) 26.9 (26.5-27.4) Note: NOS = not otherwise specified. Table 3. Estimated five-year survival probability (95% confidence interval) for selected histology types by region, age group, and sex when sex effect is significant (p < 0.05), * based on all cases (including non-microscopically confirmed) Region Age (years) Histology GBM DiffuseAstro Male Ontario 21-44 17.3 (15.2-19.6) 61.0 (57.2-64.5) 45-64 3.5 (2.9-4.2) 24.5 (20.7-28.5) [greater than or 1.9 (1.5-2.4) 7.7 (5.4-10.4) equal to] 65 Quebec 21-44 18.7 (16.4-21.1) 56.4 (52.6-60.1) 45-64 4.1 (3.4-4.8) 19.7 (16.4-23.1) [greater than or 2.3 (1.8-2.8) 5.1 (3.5-7.1) equal to] 65 British 21-44 7.7 (4.0-13.0) 37.3 (33.6-41.0) Columbia 45-64 0.7 (0.2-2.0) 6.1 (4.8-7.5) [greater than or 0.3 (0.1-1.0) 0.6 (0.3-1.0) equal to] 65 Alberta 21-44 9.0 (6.4-12.0) 44.6 (39.7-49.3) 45-64 1.0 (0.5-1.7) 10.1 (7.4-13.3) Atlantic [greater than or 0.4 (0.2-0.8) 1.5 (0.8-2.6) provinces equal to] 65 ([dagger]) 21-44 15.0 (12.0-18.4) 53.9 (48.7-58.8) 45-64 2.7 (1.8-3.7) 17.3 (13.2-21.8) Prairie [greater than or 1.4 (0.9-2.1) 4.0 (2.3-6.4) provinces equal to] 65 ([double 21-44 11.7 (8.7-15.2) 53.2 (47.4-58.7) dagger]) 45-64 1.7 (1.0-2.6) 16.6 (12.1-21.8) [greater than or 0.8 (0.4-1.4) 3.8 (2.0-6.3) equal to] 65 Region Age (years) Histology DiffuseAstro Oligo Female Male Ontario 21-44 66.4 (62.8-69.7) 69.6 (64.9-73.8) 45-64 31.2 (26.9-35.5) 47.6 (41.5-53.5) [greater than or 11.9 (9.0-15.2) 15.9 (9.9-23.2) equal to] 65 Quebec 21-44 62.2 (58.5-65.7) 72.6 (67.3-77.1) 45-64 26.0 (22.2-29.9) 51.8 (44.3-58.9) [greater than or 8.5 (6.3-11.1) 19.6 (12.2-28.4) equal to] 65 British 21-44 44.2 (40.2-48.0) 69.0 (62.2-74.8) Columbia 45-64 9.8 (7.9-11.9) 46.7 (37.6-55.3) [greater than or 1.4 (0.9-2.2) 15.2 (8.2-24.2) equal to] 65 Alberta 21-44 51.2 (46.3-55.9) 82.5 (76.7-87.0) 45-64 14.9 (11.4-18.9) 67.4 (58.2-75.0) Atlantic [greater than or 3.1 (1.8-4.9) 37.6 (24.7-50.5) provinces equal to] 65 ([dagger]) 21-44 59.9 (54.9-64.5) 75.9 (66.4-83.1) 45-64 23.3 (18.5-28.5) 56.9 (43.3-68.3) Prairie [greater than or 7.0 (4.5-10.2) 24.7 (11.8-40.0) provinces equal to] 65 ([double 21-44 59.3 (53.7-64.4) 75.5 (67.9-81.5) dagger]) 45-64 22.6 (17.3-28.3) 56.1 (45.4-65.6) [greater than or 6.6 (4.0-10.0) 23.9 (13.4-36.2) equal to] 65 Region Age (years) Histology Oligo AAstro Female Male Ontario 21-44 76.6 (72.3-80.4) 37.3 (30.3-44.2) 45-64 58.0 (51.7-63.7) 7.7 (4.6-11.7) [greater than or 25.9 (18.4-34.1) 3.3 (1.4-6.6) equal to] 65 Quebec 21-44 79.0 (74.4-82.9) 35.7 (28.5-42.9) 45-64 61.7 (54.4-68.1) 6.8 (4.0-10.8) [greater than or 30.2 (21.3-39.6) 2.9 (1.1-5.9) equal to] 65 British 21-44 76.1 (70.2-81.0) 51.0 (36.1-64.1) Columbia 45-64 57.1 (48.2-65.1) 17.4 (7.4-30.9) [greater than or 25.0 (15.9-35.1) 9.9 (3.0-21.7) equal to] 65 Alberta 21-44 86.8 (82.0-90.4) 43.9 (31.3-55.7) 45-64 74.8 (66.6-81.3) 11.7 (5.1-21.4) Atlantic [greater than or 48.8 (35.5-60.8) 5.9 (1.7-13.8) provinces equal to] 65 ([dagger]) 21-44 81.7 (73.8-87.4) 38.9 (27.3-50.4) 45-64 66.0 (53.7-75.8) 8.6 (3.6-16.4) Prairie [greater than or 35.8 (20.8-51.0) 3.9 (1.1-9.8) provinces equal to] 65 ([double 21-44 81.3 (74.9-86.2) 23.5 (12.4-36.7) dagger]) 45-64 65.4 (55.6-73.6) 2.3 (0.4-7.4) [greater than or 35.0 (22.8-47.4) 0.7 (0.1-3.5) equal to] 65 Region Age (years) Histology AAstro Oligoastro Female Male Ontario 21-44 44.4 (36.8-51.8) 56.8 (49.4-63-6) 45-64 12.1 (7.8-17.4) 29.8 (22.5-37.5) [greater than or 6.1 (3.2-10.3) 1.4 (0.3-4.5) equal to] 65 Quebec 21-44 42.9 (35.0-50.5) 63.8 (55.8-70.6) 45-64 11.0 (6.9-16.3) 38.2 (29.0-47.2) [greater than or 5.4 (2.7-9.6) 3.3 (0.8-9.0) equal to] 65 British 21-44 57.5 (42.6-69.9) 54.8 (46.0-62.7) Columbia 45-64 23.7 (11.4-38.5) 27.6 (19.9-35.8) [greater than or 14.9 (5.5-28.6) 1.0 (0.2-4.0) equal to] 65 Alberta 21-44 50.8 (37.8-62.4) 61.3 (52.5-69.0) 45-64 17.2 (8.4-28.6) 35.1 (24.5-45.8) Atlantic [greater than or 9.7 (3.5-19.7) 2.5 (0.4-8.1) provinces equal to] 65 ([dagger]) 21-44 46.1 (34.0-57.3) 67.3 (55.5-76.5) 45-64 13.3 (6.4-22.8) 42.8 (28.2-56.6) Prairie [greater than or 6.9 (2.5-14.6) 5.0 (0.8-15.2) provinces equal to] 65 ([double 21-44 30.4 (17.6-44.3) 61.6 (49.3-71.7) dagger]) 45-64 4.5 (1.1-12.0) 35.4 (21.9-49.2) [greater than or 1.7 (0.2-6.3) 2.5 (0.3-9.6) equal to] 65 Region Age (years) Histology Oligoastro AOligo Female Ontario 21-44 63.6 (56.3-69.9) 56.9 (50.2-63.0) 45-64 37.9 (29.7-46.1) 27.1 (21.1 -33.4) [greater than or 3.2 (0.9-8.1) 2.2 (0.6-6.0) equal to] 65 Quebec 21-44 69.7 (62.2-76.0) 68.2 (61.8-73.8) 45-64 46.2 (36.5-55.3) 41.3 (33.6-48.7) [greater than or 6.5 (2.2-14.2) 7.6 (3.1-14.9) equal to] 65 British 21-44 61.7 (53.2-69.2) 49.8 (38.2-60.3) Columbia 45-64 35.6 (26.9-44.4) 19.9 (11.4-30.1) [greater than or 2.6 (0.6-7.4) 0.9 (0.1 -4.2) equal to] 65 Alberta 21-44 67.5 (58.8-74.8) 71.5 (59.9-80.3) 45-64 43.2 (31.6-54.2) 46.0 (30.8-59.9) Atlantic [greater than or 5.1 (1.3-13.3) 10.5 (2.7-24.4) provinces equal to] 65 ([dagger]) 21-44 72.8 (62.2-80.8) 61.1 (49.9-70.6) 45-64 50.6 (36.0-63.5) 32.0 (20.1 -44.5) Prairie [greater than or 9.0 (2.2-21.8) 3.6 (0.6-11.5) provinces equal to] 65 ([double 21-44 67.8 (56.2-76.9) 62.5 (46.3-75.1) dagger]) 45-64 43.5 (29.2-57.0) 33.7 (16.7-51.7) [greater than or 5.3 (1.0-15.1) 4.2 (0.5-15.6) equal to] 65 Note: "Glioma, NOS" and "all others" are not included because the survival estimates are different between all-cases analysis and microscopically-confirmed only cases analysis. GBM, glioblastoma; DiffuseAstro, diffuse astrocytoma; Oligo, oligodendroglioma; AAstro, anaplastic astrocytoma; Oligoastro, oligoastrocytic tumour; AOligo, anaplastic oligodendroglioma. * Based on the time-specific generalized linear models. All numbers reported are in percentages. ([dagger]) Nova Scotia, New Brunswick, Prince Edward Island, and Newfoundland and Labrador. ([double dagger]) Manitoba and Saskatchewan.