'Our results provide clear evidence of an increase in cancer incidence in children and adolescents over the past decades, and of an acceleration of this trend', says Dr Eva Steliarova-Foucher, the project coordinator at the International Agency for Research on Cancer. 'However, cancer at these ages remains a rare disease,' she adds.
This is the first report of the European-Union supported ACCIS project (Automated Childhood Cancer Information System), steered by an international Scientific Committee composed of the authors of the paper, and is the result of collaboration between some 80 population-based cancer registries.
Dr Peter Boyle, Director of IARC, notes that 'This project once again highlights the central role of IARC in assembling data from many geographic regions and, employing a standardized approach to presentation and analysis, maximizing the use of such data to shed light on important issues in cancer control.'
Facts and figures
Based on 100,596 children with cancer, the average cancer incidence rate was 118 per million in the 1970s, 124 in the 1980s, and 139 in the 1990s, increasing significantly by 1% per year on average.
In adolescents aged 15-19, the rates increased by 1.5% per year on average: the corresponding incidence rates were 147 in the 1970s, 165 in the 1980s and 193 in the 1990s, based on 15,460 cases of cancer.
The study shows an increase for all types of cancer in both children and adolescents, leukaemias and brain tumours being the most common cancers in children in Europe, while carcinomas and lymphomas occur most frequently in teenagers. Dr Steliarova-Foucher adds: 'This increase in cancers is evident at all ages but is greater in children under four and in adolescents.'
Thanks mainly to improvements in diagnosis and treatment, children diagnosed with cancer during the 1990s had a 5-year survival of 75% in western and 64% in eastern Europe. Survival in adolescents was much the same. As in other developed countries, survival has improved dramatically since the 1970s, when 5-year survival was only 44% for children and 50% for adolescents. All tumour types benefited from the improvement of survival, including the major tumour groups of leukaemias, lymphomas and CNS tumours.
Although improvements in diagnosis and registration cannot be entirely ruled out to explain the higher figures, this can only possibly contribute partially to the increase in incidence rates.
'To follow-up our important findings, it will be necessary to identify the specific tumour types and population groups that are affected by these increases in incidence', Dr Steliarova-Foucher goes on.
Work on the ACCIS database will continue, in collaboration with cancer registries, focusing on general large-scale patterns to move towards more specific hypotheses. Identification of the risk factors is an expensive and long-term undertaking, because of the rarity of cancer in the young.
Dr Peter Boyle concludes: 'This is not the end of such a project and we need to find the means to extend collaboration to obtain a clearer insight into the situation of childhood cancer incidence and survival in lower-resource countries.'
The Lancet 2004; 364: 2097-105
Geographical patterns and time trends of cancer incidence and survival among children and adolescents in Europe since the 1970s (the ACCIS project): an epidemiological study
Eva Steliarova-Foucher(1), Charles Stiller(2), Peter Kaatsch(3), Franco Berrino(4), Jan-Willem Coebergh(5,6), Brigitte Lacour(6), Max Parkin(1)
(1)International Agency for Research on Cancer, Lyon, France
(2)Childhood Cancer Research Group, University of Oxford, Oxford, UK
(3)German Childhood Cancer Registry, University of Johannes Gutenberg, Mainz, Germany
(4)National Cancer Institute, Milan, Italy
(5)Dutch Childhood Oncology Group, The Hague, The Netherlands
(6)Department of Public Health, Erasmus University, Rotterdam, The Netherlands
(7)Registre National des Tumeurs Solides de l'Enfant, Faculté de Médecine, Vanduvre, France