PURPOSE: To describe the survival rates among Canadian children and teenagers with cancer diagnosed between 1985 and 1988 using population-based data, specifically for the more common forms of childhood cancer, and to assess the effect of age at diagnosis and sex as prognostic factors for selected childhood cancers. DESIGN: Retrospective survival study based on incident cases of cancer identified by the National Cancer Incidence Reporting System and follow-up ascertained by computer record linkage to the Canadian Mortality Database. SUBJECTS: A total of 4409 patients with cancer first diagnosed at 19 years of age or younger between 1985 and 1988, and followed up to Dec. 31, 1991. MAIN OUTCOME MEASURES: Survival rates calculated at 1, 3 and 5 years according to the actuarial life table and the proportional hazards models. RESULTS: The 5-year survival rate for all cancers combined was 71%. Females with acute lymphoblastic leukemia and astrocytoma had markedly higher survival rates than their male counterparts (p < 0.05). Age at diagnosis was a significant predictor of survival among children with acute lymphoblastic leukemia or acute nonlymphoblastic leukemia (p < 0.01), infants having a substantially poorer prognosis than older children. Conversely, the survival rate among infants with neuroblastoma was higher than that among older children, 87% surviving for 5 years after diagnosis. CONCLUSIONS: The survival rate among Canadian children and teenagers with cancer is favourable in relation to the rate among adults with cancer. Nonetheless, the 5-year survival rates for several childhood cancers remain poor (i.e., less than 65%). The survival rates among Canadian children with cancer are similar to those among children with cancer in other developed countries.
PURPOSE: To describe the survival rates among Canadian children and teenagers with cancer diagnosed between 1985 and 1988 using population-based data, specifically for the more common forms of childhood cancer, and to assess the effect of age at diagnosis and sex as prognostic factors for selected childhood cancers. DESIGN: Retrospective survival study based on incident cases of cancer identified by the National Cancer Incidence Reporting System and follow-up ascertained by computer record linkage to the Canadian Mortality Database. SUBJECTS: A total of 4409 patients with cancer first diagnosed at 19 years of age or younger between 1985 and 1988, and followed up to Dec. 31, 1991. MAIN OUTCOME MEASURES: Survival rates calculated at 1, 3 and 5 years according to the actuarial life table and the proportional hazards models. RESULTS: The 5-year survival rate for all cancers combined was 71%. Females with acute lymphoblastic leukemia and astrocytoma had markedly higher survival rates than their male counterparts (p < 0.05). Age at diagnosis was a significant predictor of survival among children with acute lymphoblastic leukemia or acute nonlymphoblastic leukemia (p < 0.01), infants having a substantially poorer prognosis than older children. Conversely, the survival rate among infants with neuroblastoma was higher than that among older children, 87% surviving for 5 years after diagnosis. CONCLUSIONS: The survival rate among Canadian children and teenagers with cancer is favourable in relation to the rate among adults with cancer. Nonetheless, the 5-year survival rates for several childhood cancers remain poor (i.e., less than 65%). The survival rates among Canadian children with cancer are similar to those among children with cancer in other developed countries.