Elsevier

European Journal of Cancer

Volume 39, Issue 18, December 2003, Pages 2622-2631
European Journal of Cancer

Incidence of malignant disease by morphological type, in young persons aged 12–24 years in England, 1979–1997

https://doi.org/10.1016/j.ejca.2003.08.006Get rights and content

Abstract

Cancer incidence data are generally presented in terms of primary site, but this method is inappropriate for cancers in young persons. We have used a morphology-based classification system to produce national incidence rates for cancers in persons aged 12–24 years by detailed diagnostic sub-type. The overall incidence rates for malignant disease in young persons aged 12–14, 15–19 and 20–24 years were 10.1, 14.4 and 22.6 per 100 000 population, respectively. The three most frequent cancer types in 12–14-year-olds were leukaemias, lymphomas and central nervous system (CNS) tumours. In 15–19-year-olds, lymphomas were most frequent and leukaemias second with carcinomas third. In 20–24-year-olds, lymphomas were again most frequent, but carcinomas and germ cell tumours were second and third. There was also variation with age in the ratios of rates in males and females. These changing incidence patterns have aetiological implications and provide clues for future hypothesis-based research.

Introduction

Cancer is most often diagnosed in the late middle-aged and elderly, and over 50% of all cancer registrations in England are for patients aged 70 years and above. In 1999, only 0.5% of all cancer registrations were teenagers and young adults aged up to 24 years [1]. However, cancer is the most common natural cause of death in this age group and is exceeded only by accidents [2]. Little is known about the aetiology of cancers in young persons. It is likely that environmental agents (in the wider sense, including diet) account for most late onset cancers, following chronic exposures over many years [3], but there is no such opportunity for such long-term exposures in teenagers and young adults. The mechanisms operating between exposure to a risk factor and clinical onset of a cancer may therefore differ either fundamentally or proportionally in young people compared with late onset cancers, and genetic susceptibility may play a greater role.

While a national co-ordinated approach to the treatment of cancers in younger children has been established for many years [4], teenagers and young adults with cancer have fared less well [5]. The teenage years and early twenties represent a critical period in terms of educational, social and career development. Disruption to education and vocational or professional training because of a prior diagnosis of cancer can have a profound impact on later life. Furthermore, the potential impact of cytotoxic treatment is of far greater importance in the young, than in the late middle-aged and elderly. Reduction in fertility, the possibility of developing treatment-induced second malignancies and organ failure are very important considerations in this age group. Furthermore, the spectrum of cancers encountered in teenagers and young adults differs markedly from both that seen in young children and in older patients [6].

All these considerations argue for the development of specialist services targeted towards teenage and young adult cancer patients. In order to develop those services, detailed and high quality statistics on the incidence and characteristics of cancers in this age group are required. To produce such statistics, a scheme for classifying cases into a manageable number of diagnostic groups is necessary. Such a scheme must be appropriate to the types of cancers that occur in the population group under study.

We have recently published a classification scheme which is specifically tailored to the adolescent and young adult cancer groups [6]. The scheme is largely based on morphology, and diagnostic groups are specified in terms of the International Classification of Diseases for Oncology (ICD-O) morphology and topography codes [7]. We proposed that our classification scheme should be used in future studies of cancers in adolescents and young adults to encourage a standard format for presentation of such data (‘a common language’) and thereby facilitate international comparisons and encourage research interest into these cancers. The scheme is available on our website at http://www.biomed2.man.ac.uk/crcpfcrg/CRUKPFCRG/PFCRG.htm.

We applied the scheme to national cancer registration data for England for the years 1979–1997 for patients aged 15–24 years. The main cancer types in this age range are lymphomas, leukaemias, bone tumours, central nervous system (CNS) tumours, germ cell tumours, soft tissue sarcomas and carcinomas. However, in contrast to older age groups, in which carcinomas of the lung, breast, large bowel and prostate account for over half of all cases [1], these carcinomas represent only 2% of malignancies in 15–24-year-olds. On the other hand, melanoma and carcinoma of the thyroid represent 8 and 3% of all cancers, respectively, in 15–24-year-olds, whereas across all age groups these cancers make up only 2 and 0.4% of the total [6].

The purpose of this study was to apply the classification scheme to an extended age range, including younger adolescents, and also to present incidence rates by detailed morphological type rather than by the site of origin.

Section snippets

Materials and methods

Anonymised individual level national cancer registration data for the years 1979–1992 were obtained on CD-ROM [8]. More recent data up to 1997 were supplied directly by the Office for National Statistics (ONS). From 1979 to 1994, data are coded by the ICD-O first edition [9] and ICD ninth revision [10] and from 1995 onwards by ICD-O second edition [7] and ICD tenth revision [11]. National population estimates by single year of age, gender and calendar year were supplied by the Population

Results

The study included 28 573 cases of malignant neoplasms, 3673 in the 12–14 year age range, 9307 in 15–19-year-olds and 15 593 in 20–24-year-olds. The study population encompassed a total of 170 263 000 person-years at risk, including 36 479 000 in the 12–14 year age range, 64 753 000 in the 15–19 year age range and 69 031 000 in the 20–24 year age range.

Table 1 shows the incidence of malignant disease among the study population by age group, gender and main diagnostic group. The incidence of all

Discussion

Worldwide cancer incidence data are generally presented mainly by primary site, grouped according to the ICD 11, 17. While this is acceptable for cancers in older adults—which are mainly carcinomas. ICD cannot distinguish non-epithelial cancers from carcinomas or different types of non-epithelial cancers from each other. Data on cancers in young people are more appropriately presented in terms of histopathological type (morphology). Morphology-based classification systems for analysis and

Acknowledgments

Data used in this study were contributed by the nine regional cancer registries in England. Jillian M. Birch is a Cancer Research UK Professorial Fellow. Robert D. Alston is supported by Cancer Research UK.

References (54)

  • M.J. Quinn et al.

    Cancer 1971–1997. Registrations of Cancer Cases and Deaths in England and Wales by Sex, Age, Year, Health Region and Type of Cancer [CD-ROM]

    (1999)
  • International Classification of Diseases for Oncology

    (1976)
  • International Classification of Diseases, Injuries and Causes of Death, 9th revision

    (1975)
  • International Statistical Classification of Diseases and Related Health Problems, 10th revision

    (1992)
  • B. Francis et al.

    The GLIM System, Release 4 Manual

    (1993)
  • R.J. Lukes et al.

    Report of the nomenclature committee

    Cancer Res.

    (1966)
  • S.W. Weiss et al.

    Enzinger & Weiss's Soft Tissue Tumours

    (2001)
  • J.M. Birch et al.

    A classification scheme for childhood cancer

    Int. J. Cancer

    (1987)
  • E. Kramárová et al.

    The international classification of childhood cancer, 1996

    Int. J. Cancer

    (1996)
  • M.A. Smith et al.

    Cancer in adolescents 15–19 years old

  • L. Fritschi et al.

    Incidence of Cancer among New South Wales adolescentswhich classification scheme describes adolescent cancers better?

    Int. J. Cancer

    (1995)
  • M.F. Greaves

    Speculations on the cause of childhood acute lymphoblastic leukaemia

    Leukaemia

    (1998)
  • L.J. Kinlen

    Epidemiological evidence for an infective basis in childhood leukaemia

    Br. J. Cancer

    (1995)
  • M.A. Smith et al.

    Evidence that childhood acute lymphoblastic leukaemia is associated with an infectious agent linked to hygiene conditions

    Cancer Causes and Control

    (1998)
  • Cited by (58)

    • Head and neck sarcomas in adulthood: current trends and evolving management concepts

      2020, British Journal of Oral and Maxillofacial Surgery
      Citation Excerpt :

      Sarcomas are rare, heterogeneous, malignant tumours.1 In adults, about 10% of all sarcomas affect the head and neck,2 but in children 21% of all malignancies are sarcomas,3 of which 7%-10% affect the soft tissues.4 Unlike carcinomas, which are epithelium-derived, sarcomas are of mesenchymal origin, and are broadly classified as bone sarcomas (BS) and soft-tissue sarcomas (STS).

    • Cryopreservation of testicular tissue in children

      2013, Bulletin de l'Academie Nationale de Medecine
    • Changes in the Patterns of Care of Central Nervous System Tumours Among 16-24 Year Olds and the Effect on Survival in Yorkshire Between 1990 and 2009

      2013, Clinical Oncology
      Citation Excerpt :

      The overall lack of risk reduction may, in part, be explained by the excess of low-grade tumours. Although the most common subgroup was astrocytoma [8], as expected, the excess of low-grade astrocytomas (1.72:1) conflicts with national data showing equal proportions [9]. Our survival data for independent diagnostic groupings is similar to that previously published for the UK, except for the significant relative reduction in risk seen for the ‘other intraspinal and intracranial neoplasms’ and for medulloblastoma [15].

    • Paediatric, teenage and young adult cancers

      2011, Specialist Training in Oncology
    • CML in pregnancy and childhood

      2009, Best Practice and Research: Clinical Haematology
    • Soft tissue sarcoma: Recognizing a rare disease

      2022, Cleveland Clinic Journal of Medicine
    View all citing articles on Scopus
    View full text