Use this URL to cite or link to this record in EThOS:
Title: A study of the aetiology and epidemiology of cancers in teenagers and young adults
Author: Arora, Ramandeep
ISNI:       0000 0004 2708 9607
Awarding Body: University of Manchester
Current Institution: University of Manchester
Date of Award: 2011
Availability of Full Text:
Access from EThOS:
Access from Institution:
Introduction: Little is known about the aetiology of cancer in teenagers and young adults (TYA) aged 15-24 years, although in England, cancer is the most common cause of disease-related mortality in this age group. The most common cancers at this age are lymphomas, central nervous system (CNS) tumours and germ cell tumours (GCT). The commonest carcinomas seen at older ages including lung, breast, large bowel and prostate account for only 3-4% of TYA cancers. In this thesis I describe the incidence patterns of selected cancers in TYA and the variation seen with geography, time and in population subgroups. The focus is on CNS tumours, GCT and bone tumours as they either peak in incidence in TYA and/or contribute disproportionately to cancer related mortality in TYA. This will allow formulation of hypotheses regarding aetiology of cancer in this age group which can then be tested by further research. Methods: For the majority of the analysis, anonymised national cancer registration data from England on individual patients of all ages with newly diagnosed cancer between 1979 and 2003 were used. To contrast the incidence patterns in England with that of India, data from five Indian urban population based cancer registries were used for part of the analysis. Age, sex, site and histology specific incidence rates were calculated and expressed per million person years. All rates, where appropriate, were adjusted to the world standard population using direct methods. To explore the link of growth with development of osteosarcoma and Ewing sarcoma, a random-effects meta-analysis was undertaken on studies which investigated an association of these tumours with height at diagnosis. Results: The incidence of cancer in TYA overall in England exceeded that of India. This was also true for most individual sites including epithelial cancers of lung, colon/rectum, breast, ovary and cervix, and non-epithelial cancers including melanoma, Hodgkin lymphoma and testicular cancer. Notable exceptions to this pattern were cancers of the mouth, gall bladder and stomach (females only) where incidence was higher in India. In England, CNS tumours in TYA were a composite of pilocytic astrocytomas and embryonal tumours (representing tail end of childhood CNS tumours), pituitary tumours, nerve sheath tumours, high grade astrocytomas and meningiomas (representing early-onset of CNS tumours that peak in incidence in the 6th and 7th decade of life), and of CNS GCTs, pleomorphic xanthoastrocytomas and neurocytomas which show a peak incidence in TYA. Irrespective of site or histology, GCT in England showed a peak in incidence between ages of 10 to 39 years which was more marked in males. This however varied by site and the peak incidence was seen at 10 to 14 years in the CNS, 15 to 19 years in ovary, 25 to 29 in mediastinum & thorax and abdomen & pelvis, and 30 to 34 years in testicular tumours. Osteosarcoma and Ewing sarcoma were the predominant bone tumours in TYA in England and showed a distinct peak of incidence at 10 to 14 years age in females and a larger peak at 15 to 19 years age in males. The peak incidence of osteosarcoma of long bones of the lower limb was six times more than that at any other site while the peak incidence of Ewing sarcomas located in the bones of the central axis exceeded those in long bones of the lower limb. The average height of patients with osteosarcoma at diagnosis was found to be significantly above the average height of the reference population, at the 95% level. The association of greater height at diagnosis with Ewing sarcoma was also significant at the 95% level but much weaker. Conclusion: In this thesis I have explored the epidemiology of cancer in TYA using some of the established methodologies which have previously been used in advancing our knowledge of childhood and older adult cancers. These studies provide some clues to aetiology. Variation in environmental exposures and lifestyle factors between England and India can explain the majority of the differences in incidence patterns observed. Genetic predisposition to cancer along with carcinogen exposure could lead to early onset of some cancers generally seen in older adults. Regardless of site, the similarity in age-incidence patterns of GCT, suggests a common initiation of these tumours in embryonic/foetal life with variable rates of tumour progression as a result of local factors or events during postnatal and pubertal period. The incidence patterns of osteosarcoma along with the strong and consistent association with a greater height at diagnosis indicate that bone growth is important in the development of this tumour while different biological pathways which may be unrelated to growth could also be relevant for Ewing sarcoma.
Supervisor: Birch, Jillian Sponsor: Not available
Qualification Name: Thesis (M.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: Cancer ; Aetiology ; Epidemiology ; Incidence ; Teenagers ; Adolescents ; Young Adults ; England ; India ; Central Nervous System Tumours ; Germ Cell Tumours ; Bone Tumours ; Osteosarcoma ; Ewing Sarcoma