Objectives

This study describes up-to-date cancer incidence and survival in Italian paediatric and adolescent patients, based on data collected by the network of Italian cancer registries (AIRTUM). It updates the monograph published on the same topic in 2008. The main objective of this monograph is to present the statistics according to standard rigorous epidemiological methods and disseminate them to a wide range of readers, including the lay public. Given the deep impact of the 2008 monograph on the general public, in this update we complement descriptive statistics with additional data and commentaries on issues of importance for public health, in order to provide unambiguous criteria on how to interpret the statistics. The study is the result of the collaboration between AIRTUM and AIEOP (Italian Association of Paediatric Haematology and Oncology) with contributions from interested parties, including representatives of parent associations. The monograph is divided into three parts. The first part presents incidence rates, survival probabilities, and time trends, by sex, age, geographical area, and cancer site or type, by means of tables and graphs as in the previous monograph, to facilitate direct comparisons. Four articles summarize and comment the results. The second part uses data from AIRTUM and AIEOP to outline patient management and health care issues; it includes estimates of the number of new cases in the next decade and of young adults living after a paediatric cancer diagnosis. Health organizational aspects of treatment services for paediatric patients, based on the AIEOP database, are also discussed, along with long-term complications in cured patients. The third section describes the changes in mortality trends due to improving therapies and healthcare services, and discusses risk factors and prevention of childhood cancer, late adverse events in cured patients, and other related issues.

Material and methods

Data herein presented were provided by AIRTUM population-based cancer registries, covering 47%of the Italian population below age 20 years, in the period 2003-2008. Quality of cancer registration in Italy is elevated, with high proportions of microscopically verified diagnoses (91%in the 0-14 years age group and 96% between 15 and 19 years of age) and a very small proportion of cases collected through death certificate only (0.1%).The proportion of cases in diagnostic groups XI (other malignant epithelial neoplasms) and XII (other and unspecified neoplasms) of the International Classification for Childhood Cancer (ICCC), based on the third revision of the International Classification of Diseases for Oncology (ICD-O-3), were 7.0% in the 0-14 years age group and 26.0%in the 15-19 years age group.The ratio between mortality and incidence was 17.7% in both children and adolescents. Detailed results are presented in 24 fact sheets for the 12 major ICCC-3 diagnostic groups and 10 sub-groups of special interest; the series is completed by a sheet on all malignant tumours and one on all tumours including non-malignant neoplasms of the central nervous system. All sheets include results for three age groups (0-14, 15-19, and 0-19 years) and are followed by two commentaries on incidence in the recent period, one on trends and the other on survival. Incidence rates were age-standardized on the European population and presented per million children. Incidence rates are also presented by age group, sex, and geographical area. Incidence trends were evaluated for two periods, 1988-2008 and 1998-2008, using estimated annual percent changes, and survival estimates were calculated by age and period. Indicators and corresponding 95% confidence intervals are shown in forms of graphics and tables at the end of the monograph and online at http://www.registri-tumori.it. Geographical analyses were conducted rearranging cancer registries into four macroareas (North-West, North-East, Centre, and South and Islands). Age groups were the same used in descriptive studies on children worldwide (0, 1-4, 5-9, 10- 14 years for paediatric tumours and 15-19 years for adolescents). Incidence trend analyses included cancer registries with three or more years of registration in the 5-year period, using Poisson regression models. Observed survival was computed according to the Kaplan-Meier method. The estimate of expected cases in the next decade was based on observed incidence rates in the most recent period, extended to the Italian estimated population of children and adolescents in the periods 2011-2015 and 2016-2020. The AIEOP database (Modello 1.01) allowed us to compare the number of patients treated and followed-up in specialized centres with expected cases based on AIRTUM estimates. The AIEOP database also provided information regarding health care migration throughout Italian regions and the number of foreign (immigrated) children treated in Italian AIEOP centres.

Results

In the period 2003-2008, 31 cancer registries reported 4,473 incident malignant neoplasms, 2,855 in children and 1,618 in adolescents. Cancer incidence rates were 164 cases per million in children aged 14 years or below and 269 cases per million in patients aged 15-19 years. Limited geographical variations emerged. In children (0-14 years) a significant increase in malignant cancer incidence was observed until 1997 (APC: +3.2%), followed by a plateau (APC: -1.1%not statistically significant).Until the late Nineties, a statistically significant increase was also observed in the incidence of all leukaemias in males (APC: +5.7%), lymphoid leukaemias (APC: +5.6%), representing 80% of all leukaemias, Hodgkin and non- Hodgkin lymphomas (APC: +6.3%). A significant decrease emerged for lymphoid leukaemia starting in 1995 (APC: -1.9%), while no substantial change in cancer incidence rates was observed in the last decade of observation for all malignant neoplasms and lymphomas. In addition, no variation emerged for malignant (according to the most recent classification) central nervous system (CNS) neoplasms, while an annual increase of 1.8% (significant) was observed in the period 1988-2008, when non-malignant tumours were included. Increases in cancer incidence were observed throughout the study period for neuroblastoma (APC: +1.9%) and epithelial tumours or melanoma (APC: +4.1%). In the period 1998-2008, in addition to lymphoid leukaemias, a significant decrease was observed for all malignant neoplasms, lymphomas in girls, CNS tumours (males and females), and renal tumours in girls, while no increases were observed in this age group. In adolescents (15-19 years) between 1988 and 2008, a significant increase in incidence rates was observed (APC: +2.0%) for all malignant neoplasms, all lymphomas (APC: +2.9%; in particular Hodgkin lymphoma, APC: +3.6%), thyroid cancer (APC: +6.1%), and melanoma (APC: +8.1%). Conversely, lymphoid leukaemia is the only neoplasm showing a long-term decrease in adolescents. Recent trends (1998-2008) confirm the long-term increases only for all malignant neoplasms in girls and thyroid cancer (APC: +7.9%, boys and girls), while a decrease in bone tumour incidence emerged in girls, albeit based only on 46 cases. Cancer mortality in children showed a persistent decrease for all neoplasms and even for more frequent cancer sites or types, and mortality rates for cancer were three-fold higher in the early Seventies than in 2008. In addition, five-year survival after cancer diagnosis increased in the last three decades and was still increasing in the period 2003- 2008, reaching 82% in children and 86% in adolescents. In the period 2008-2010, 4,488 children (0-14 years) were treated in one of the AIEOP clinical centres and we estimate, based on the above-presented incidence rates, that they represented 92% of all cancer cases in Italy. However, in adolescents, the proportion of patients treated in AIEOP centres was only 25%. A migration of patients living in the South of Italy to Central and Northern Italy emerged from AIEOP information. The expected number of cancer cases in children aged between 0 and 14 years of age is approximately 7,000 in the period 2016- 2020, while the corresponding figure for adolescents between 15 and 19 years of age is 4,000, with no relevant variation in comparison with the previous five-year period.

Comments

The present findings update descriptive cancer epidemiology in children and adolescents in Italy based on data provided by an extensive network of general and specialized population-based cancer registries. Data obtained from cancer registries are supplemented by additional information collected by specialized clinical AIEOP centres and mortality reports collected by the National Institute of Statistics (ISTAT). Incidence rates reported in Italy were slightly higher in comparison to other developed Countries, but relatively consistent between different Italian areas. Our results also showed that the significant increase in cancer incidence observed until the end of Nineties has halted, with the exception solely of thyroid cancer in adolescents. Efficacy of therapeutic protocols has improved constantly since the Seventies, and recent findings confirm this trend in all age groups and, in particular, for rarer tumours and cancer types that have very poor prognosis. Findings derived from cross-analysis with AIEOP data suggest that it is possible to further improve the efficiency of our healthcare system, in particular for adolescents; migration can be reduced with a more rational use of hospitals throughout Italy.