Childhood Cancer
Childhood cancer in Scotland, 1975-1999
Childhood cancer in Scotland 
[1.9MB]: Incidence, mortality and survival data specific to children with cancer in Scotland (1975-1999).
Cancer in adolescents and young adults in Scotland, 1976-2000
A classification system for adolescents and young adults with cancer has recently been developed (Birch et al, 2002), and is summarised in Table 1 
[25KB]. This classification has been used to show incidence of and survival from cancer in adolescents and young adults in Scotland. Almost 160 cancers arise in adolescents and young adults (aged 15-24 years) in Scotland every year (Table 2 [19KB]), accounting for approximately 0.7% of total (all ages) malignant neoplasms, excluding non-melanoma skin cancers (see notes). Lymphomas account for the highest proportion of cancers in this age group, almost 24%. Carcinomas, germ cell tumours (mostly testicular), melanomas (mostly cutaneous) and leukaemia and lymphoma in total account for almost 80% of all cancers in this age group.
The breakdown for adolescents and young adults diagnosed in 2000 is shown in Figure 1 [18KB] for the most common cancers. Overall, 62% of all cancers occur in the 20-24 year age group and 38% in the 15-19 year age group.
The age-specific incidence rates (Table 3 [20KB]) show the same trend for all cancers combined, with higher incidence in the older age group, and they show that, overall, incidence is higher in males than females. When looking at all cancers individually, for both sexes combined, leukaemia, CNS tumours, bone tumours and soft tissue sarcomas have a higher incidence rate in the younger age group, with the remaining cancers having a higher incidence in the older age group. Incidence is higher in males than females for the majority of cancers except melanomas, carcinomas and miscellaneous specified tumours.
Incidence rates of all cancers combined in adolescents and young adults have increased over time (Table 4 [20KB]), rising from 178.3 to 237.4 per million population between the periods 1976-1980 and 1996-2000. This increasing trend is seen for most of the individual cancer groups, although there has been a decrease in the incidence of the miscellaneous specified group. The most striking increases in incidence are seen for germ cell tumours and melanoma.
One year observed survival from all cancers in adolescents and young adults has increased by 11%, from 81% for those diagnosed during 1976-1980 to 92% for those diagnosed during 1996-2000 (Table 5 [19KB]). This increase is also generally apparent when cancers are examined individually, except for soft tissue sarcoma where one year survival has decreased (from 93% to 83%). Overall, five year survival has increased by 19% (from 60% to 79%), and as with one year survival, five year survival has increased generally for each cancer group, with the exception of soft tissue sarcoma.
Fluctuations in survival over time are likely to be explained, at least in part, by random variation due to small numbers of cases, but the apparent decrease in survival from soft tissue sarcoma seems to be explained by a change in case-mix, specifically an increase over time in the relative proportion of rhabdomyosarcomas which often have a very poor prognosis. Note that unspecified malignant neoplasms (not otherwise specified) (group 10) are not shown in Table 5 due to small numbers of cases.
Reference:
Birch JM, Alston RD, Kelsey AM, Quinn MJ, Babb P and McNally RJQ (2002). Classification and incidence of cancers in adolescents and young adults in England 1979-1997. Br J Cancer; 87: 1267-1274.
Notes
1. Results are presented for cases diagnosed in the period 1976-2000 for the 10 main categories of cancers in adolescents and young adults. The period was chosen to provide an analysis covering the most up-to-date time period and to cover a 25 year period. Note that the follow-up period for survival analyses was to 31 December 2002.
2. Details of the access database developed by Birch et al for applying the classification to registry data can be found at the following web link: www.biomed2.man.ac.uk/crcpfcrg/Crukpfcrg/project2.html#Classification (accessed 13 November 2003). This website also includes results of their analyses of cancer incidence in adolescents and young adults in England & Wales.
3. To keep the Scottish analyses consistent with the Birch et al paper, our analyses were restricted to all malignant neoplasms excluding non-melanoma skin cancers (ICD-10 C00-C96 excluding C44). Only those patients aged between 15 and 24 years at the time of diagnosis were included.
4. Patients whose vital status is unknown were excluded from the survival analyses (there were 7 such cases). One patient was also excluded from the survival analyses because it was evident that the patient had moved out of Scotland, although no date of embarkation could be found. Patients registered from death certificates only (DCOs) were excluded from the survival analyses (there were 14 such cases). DCO registrations relate to patients for whom the presence of cancer was recorded on the death certificate, but for whom no other records mentioning cancer could be traced.
5. Mid-year population estimates used for calculating the rate per million population were supplied by GROS and are published, for 2000, in their annual report on their website at: www.gro-scotland.gov.uk/grosweb/grosweb.nsf/pages/00annrep (accessed 13 November 2003).
