Risk factors for malignant neoplasms in children

The clinical and epidemiological pattern of malignancies in children differs from that of adults in several aspects.

 

Epidemiological data

According to the International Association of Cancer Registries (IACR), childhood cancer is the second leading cause of death in children in developed countries. The incidence in children younger than 15 years old is estimated to be approximately 150 cases/million. Data show that the incidence of childhood cancer is increasing in the developed countries by around 1% annually [2]. Leukaemia (34%), central nervous system (CNS) tumours (23%) and lymphomas (11.5%) are the principal sites. There are types of cancer that are detected almost only in children and adolescents (very rarely in adults), such as neuroblastomas, nephroblastomas, rhabdomyosarcomas, retinoblastomas and certain primary gonadal cancers. Cancers that are common among adults and rare among children are those of the breast, lung and the large intestine.

The epidemiological pattern of diagnoses differs among the various age groups of children: in infants, neuroblastoma, retinoblastoma and nephroblastoma cover more than 50% of the cases, neuroblastoma being the diagnosis for 1/3 of the cases. In the age group of 1–4 years old, leukaemia represents around 44% of the diagnoses, while for the age group of 5–14 years old, leukaemia, CNS tumours and lymphoma represent 3/4 of the diagnoses. Adolescents (15–19 years old) suffer mainly from sarcoma and primary gonadal tumours (around 32%). Childhood cancer occurs more frequently until the age of 5 years and is more common in boys than in girls by 20% [2].

Only 5% of malignant neoplasms during childhood are attributed solely to genetic susceptibility. The increased incidence during the first 5 years of human life points to a prenatal onset of the disease. There may be a special genetic response to a common extraneous factor, or an aetiologic procedure stimulated by out-of-the-ordinary extraneous influences during the prenatal period [1].

Epidemiological studies that looked into the existence of special aetiologic agents leading to the manifestation of childhood cancer focus both on children being exposed to chemical substances, e.g. smoking of parents or air pollutants, and on the exposure to infectious agents [4].

 

Sites

Leukaemia: Acute lymphoblastic leukaemia (ALL) accounts for the highest percentage (around 25%) of childhood cancer, while acute myeloid leukaemia (AML) represents approximately 5% of the disease [7]. Chronic myeloid leukaemia is rare, while chronic lymphogenous leukaemia does not affect children. The average age of occurrence for ALL is around 4.5 years old and that for AML is around 6 years old. Aetiologic hypotheses refer to a protective influence of continuous immunostimulation caused, for example, by frequent infections at nurseries [8]. Correspondingly, situations associated with a delayed or limited stimulation of the immune system, e.g. history of missing vaccinations, younger birth line within the family and high socio-economic status, have been implicated as potential risk factors. Children who develop leukaemia are considered by some to be suffering from a congenital deficit in the immune response [1]. The only non-hereditary risk factor for leukaemia that has been documented is exposure to ionising radiation (during the prenatal period and early infancy). It has been hypothesized that AML may be correlated with alcohol abuse by mothers during pregnancy. More than 80% of infants with ALL and AML present with MLL-11q23 gene mutations or other chromosome abnormalities in leukaemia cells [7].

Lymphoma: Lymphomas are diagnosed around the average age of 10 years old. The most common types are the Non-Hodgkin lymphomas, largely represented by the Burkitt lymphoma, and the Hodgkin lymphoma which has been associated with infections caused by the Epstein Barr and HIV viruses [2].

Tumours of the Central Nervous System (CNS): In children, CNS tumours are mainly located under the cerebellar tentorium and they are essentially astrocytomas (40% of the CNS tumours). Boys are affected more often than girls. The studies of non-genetic risk factors show the aetiologic role of exposure to ionising radiation for therapeutic purposes, for both the mother and the child. Birth weight, parents who smoke, parents whose occupation concerns animal breeding, as well as parental infection by polyomavirus have been reported to be positively correlated with the risk of CNS tumours [1].

Neuroblastoma is one of the most frequent extracranial solid tumours. It originates from the sympathetic nervous system and is mainly located in the adrenal glands and the abdomen. Infants younger than 1 year old who do not carry the MYCN gene have an excellent prognosis [5,6]. Data about a positive correlation with maternal use of contraceptives or subfertility treatment are weak, as, in general, are data implicating environmental risk factors. The association with exposure to electromagnetic fields has not been documented [1].

Retinoblastoma is relatively rare; it occurs at an average age of 1.5 years old and, in the majority of cases, is due to the existence of a hereditary mutation of the RB1 gene. Some researchers have reported a positive association of retinoblastoma with HIV virus infection and in vitro fertilisation, and a negative association with maternal use of multivitamin supplements [4].

Wilms’ tumour: Wilms’ tumour or nephroblastoma usually occurs when children are between 2 and 3 years old. It has been associated to aniridia and abnormalities in the urogenital system. There are weak indications that implicate a large birth weight [5].

Malignant bone tumours: These occur more often in older children and adolescents and are represented by osteosarcoma and Ewing sarcoma. Children who have recovered from a previous neoplasm using radiation and alkylating agents bear the greatest risk. Certain studies associate the development of these tumours positively with the exposure to radon, as well as the taller stature of the child.

Soft tissue tumours: These include a group of tumours of heterogeneous origin, with rhabdomyosarkoma being the primary representative. Kaposi sarcoma seems to have a positive and strong association with the HIV infection [3].

Gonadal tumours: These constitute an heterogeneous group of malignancies. They appear in the gonads, intracranially, but also in other sites. The age distribution is bimodal: the first peak develops until the first 2 years of age and the second after 6 years of age, with a slight dominance of girls (boys/girls 0.8:1). There are not many studies about their aetiology, with the exception of studies about testicular cancer that has been associated positively with cryptorchidism, the intake of hormones during pregnancy, and maternal pre-pregnancy obesity [3].

 

References
  1. Bunin GR. 2004. Non-genetic causes of childhood cancers: evidence from international variation, time trends, and risk factor studies. Review Toxicology and Applied Pharmacology; 199:91-103.
  2. Kaatsch P, Steliarova-Foucher E, Crocetti E. 2006. Time trends of cancer incidence in European children (1978-1997): Report from the Automated Childhood Cancer Information System project; 42:1961-1971.
  3. Kaatsch P. 2010. Epidemiology of childhood cancer. Cancer Treatment Reviews; 36:277-285.
  4. McNally RJ, Eden TO, Alexander FE. 2005. Is there a common aetiology for certain childhood malignancies? Results of cross-space-time clustering analyses. European Journal of Cancer; 41:2911-2916.
  5. Smith MT, McHale CM, Wiemels JL. 2005. Molecular biomarkers for the study of childhood leukemia. Toxicology and Applied Pharmacology; 206:237-245.
  6. Terracini B, Coebergh J-W, Gatta G. 2001. Childhood cancer survival in Europe:an overview European Journal of Cancer; 37:810-816.
  7. Wiemels J. Perspectives on the causes of childhood leukemia. 2012. Chemico-Biological Interactions; 196:59-67.
  8. Zierhut H, Linet MS, Robison LL. 2012. Family history of cancer and non-malignant diseases and risk of childhood acute lymphoblastic leukemia: A Children’s Oncology Group Study. Cancer Epidemiology; 36:45-51.

 Anastasia Barbouni, MD, MS, PhD, Pediatrician, National School of Public Health