Cancer rates in teens and early 20s rising – more research needed to understand causes
A new analysis of cancer figures for England shows that the overall incidence among teenagers and young adults is rising, with the biggest increase among 20 to 24-year-olds, particularly in lymphoma, melanoma and germ cell tumours, including testicular germ cell tumours.
A news briefing at Teenage Cancer Trust’s Third International Conference on Adolescent Cancer today (Monday 1 March) was told that although cancer is still rare in this age group – around 1,500 cases a year in England – it is the leading cause of death after accidents.
Professor Jill Birch, from the University of Manchester, [see note at end of this release] who analysed data for 13 to 24-year-olds between 1979 and 2000, said that the overall rate had risen from 15.4 to 19.8 per 100,000 over the 21 year period 1979-2000. This amounts to an average increase of 1.2% per year.
Different cancers dominated at different ages. Among 13 and14-year-olds, leukaemia is the most common accounting for over 22% of cancers, followed by lymphoma (nearly 21%) and malignant brain tumours (18%). By age 15-19 lymphoma dominates (27%) with leukaemia second (15%) followed by carcinomas and malignant brain tumours (around 11% each). Lymphoma also dominated in 20 to 24-year olds with carcinomas second (21%) and germ cell tumours third (17%).
Much less is known about the causes and risk factors for adolescent and young adult cancer than for cancer in older adults.
Professor Birch, who is Cancer Research UK Professorial Research Fellow at Manchester University, said: "The early age of onset and lack of opportunity for chronic exposure to environmental factors suggests that genetic susceptibility may be important. Highly penetrant genes or mutations probably only account for a small proportion of cases. What is more likely is that cancer develops as a result of exposure to a risk factor in a genetically susceptible individual."
She said that the increasing incidence of certain cancers over time would be consistent with environmental (exogenous) risk factors, which are becoming more frequent or occurring at increasing levels. Alternatively, lifestyle changes in recent years may also be contributing. For example, melanoma rates (linked to sun exposure) had almost doubled over 21 years and in the 20-24 year age group now accounted for one in 10 of all cancers.
"Exposure of a young child or fetus to environmental carcinogens during periods of rapid growth and development might lead to higher cancer risks in young adults than adult exposure at similar doses would to the risk of cancer in later life."
Infection in early life could be important in lymphoma with risks varying according to the individual’s HLA (histocompatability) system. Likewise, individuals’ genetic make up may make them susceptible to the ability of viruses to trigger cancer. Passive smoking may be a factor in some cancers, but exposure that happens before birth or in infancy may target different organs than those affected by exposure in later life. Again, the genetic make up of mother and fetus is likely to affect the level of risk.
Professor Birch said that good quality data on patterns and trends in cancer incidence were essential for researching and understanding causes and she called for a specialised system of cancer registration in this age group.
"We understand patterns of childhood cancer fairly well because there are specialised registries internationally. But childhood classifications don’t work for adolescents and young adults because many of the childhood cancers are not seen in adolescents. Adult data are presented by cancer site but that is not suitable for teenagers and young adults either. That’s an important reason why we don’t understand nearly so much about this age group.
"What we need is a classification that is tailored to the 13 to 24-year-olds with biologically similar cancers classified together, and diagnostic categories that allow maximum flexibility in analyses. A standard classification that is accepted internationally will ensure that results from different registries are comparable."
She concluded: "This would allow us to analyse by diagnosis, age, sex, geographical variations and to undertake international comparisons. Studies of familial patterns will also be important. Then we would have a proper basis for detailed molecular epidemiological studies that could really help us to identify the causes of the cancers that are affecting our young people."
 Data provided by National Cancer Intelligence Centre, ONS. Study included 32,000 cancers diagnosed in young persons aged 13-24 years between 1979 and 2000.
Note: Professor Jill Birch was to have given the presentation on incidence and aetiology to the conference at 09.00 Tuesday 2 March and to the news briefing at 10.30am Monday 1 March, but has had to withdraw for personal reasons. Professor Tim Eden, University of Manchester and Christie Hospital, Manchester, will now make the presentation to the conference and Mr Charles Stiller, senior research fellow, Childhood Cancer Research Group, Oxford University, will talk about incidence and aetiology at the news briefing.
Margaret Willson | alfa