Survival of childhood and adolescent/young adult (AYA) cancer patients in Ireland during 1994-2013: comparisons by age.

BACKGROUND: Some studies indicate that survival of adolescents and young adults (AYA) with cancer may be inferior to that of younger children with similar cancers, possibly related (in part) to differences in access to centralized or standardized treatment. AIMS: This study aims to evaluate differences in survival for AYA patients when compared with paediatric patients treated in Ireland over a 20-year time period. METHODS: This study compares relative survival for patients diagnosed in Ireland at ages 0-15 (paediatric group) and 16-24 (AYA group) during 1994-2013, followed to the end of 2014, for cancers defined by the International Classification of Childhood Cancer (ICCC) (Third Edition) group or subgroup. Five-year relative survival estimates, and excess hazard ratios (EHR) comparing excess mortality associated with a cancer diagnosis among AYA with that in the paediatric group, are presented. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. RESULTS: Significantly higher excess mortality was found for AYA with leukaemias, lymphomas, astrocytomas, malignant bone tumours, and Ewing and related bone sarcomas, soft tissue sarcomas and 'other/unspecified' epithelial cancers, rhabdomyosarcomas, and 'other and unspecified' carcinomas. In contrast, lower excess mortality was found in the AYA group for all cancers and intracranial/intraspinal tumours, and for gliomas other than astrocytomas or ependymomas. Comparing 1994-2003 and 2004-2013 cohorts, age-related survival differences narrowed for lymphoid leukaemias, but widened for all cancers combined and intracranial/intraspinal tumours combined. Centralization of services varied depending upon cancer subtype, with leukaemias, CNS tumours and bone sarcomas most centralized. Within these, improvements in survival for leukaemias and CNS tumours have been seen for the AYA population. CONCLUSIONS: Reasons for age-related survival differences, and differences in time-trend by age group, are not clear. The significant narrowing of survival differences by age in more recent years for lymphoid leukaemias reflects a more marked recent increase in survival among AYA. More work is required to explain and improve other age-related survival differences.