HLA MHC Inf & Imm Genetics Evolution Epidemiology Genetic Epidemiology Biostatistics Glossary Homepage

Childhood Leukemias

M. Tevfik Dorak, M.D., Ph.D.

Leukemia is a malignancy of the hematopoietic system characterized by diffuse replacement of the bone marrow by neoplastic cells. In acute leukemias, the immature hematopoietic cells are increased in the blood, and chronic leukemias are characterized by an excess of well-differentiated blood cells. In children, the vast majority of leukemias are of the acute type, whereas in adults, chronic leukemias are more common ¹. Leukemias are the most common childhood malignancies, accounting for just above 30% of all cancer diagnoses in children under 15 years of age ^2-4. In this age group, approximately 75% of leukemias are classified as acute lymphoblastic leukemia (ALL) ^{2; 4}. The prefix 'acute' is superfluous but persists in the universal acronym ALL. The second most frequent leukemia type in childhood is acute myeloid leukemia (AML), and the second most common cancer in childhood is central nervous system tumours ^{2; 4; 5}.

Immature hematopoietic cells giving rise to ALL are not easy to distinguish morphologically. The modern classification of acute leukemias relies on the changes in the expression of cell surface antigens as a precursor cell differentiates. Using monoclonal antibodies, cell surface antigens (called clusters of differentiation (CD)) can be identified in cell populations; leukemias can be accurately classified by this means (immunophenotyping) ^{6; 7}. By immunophenotyping, it is possible to classify ALL into the major categories of 'common - CD10+ B-cell precursor' (around 50%), 'pre-B' (around 25%), 'T' (around 15%), 'null' (around 9%) and 'B' cell ALL (around 1%) ⁸. All forms other than T-ALL are considered to be derived from some stage of B-precursor cell, and 'null' ALL is sometimes referred to as 'early B-precursor' ALL. Etiology of childhood ALL is not known. The undisputed fact is the multigenic, multifactorial and multistep nature of its development ^9-11. A few recognized risk factors account for a small proportion of cases. The genetic abnormalities associated with the disease and the recognized epidemiological risk factors provide some clues to the etiology of childhood ALL.

Genetic background

Childhood ALL is not an inherited disease ^{12; 13}. A genetic background in childhood ALL is, however, suggested by a tendency to cluster in families that experience an excess incidence of leukemia or cancer ^13-17, increased risk for the siblings of a patients with childhood leukemia (one in five siblings develop leukemia) ^{18; 19}, and a high degree of concordance among twins ^20-24, although not in all studies ²⁵. There is, however, evidence for intrauterine single cell origin, with twin-to-twin transmission, of concordant leukemia in twins ^{22; 23; 26}. This is a more likely cause for concordance than genetic factors. It has recently been shown that leukemia may indeed arise in utero ^27-29. Several molecular studies found the same clonotypical MLL or TEL rearrangement in patients' blood samples taken at birth. This was shown for patients with infant leukemia and for those with cALL. These results provide unequivocal evidence for prenatal initiation of acute leukemia in most patients.

Acquired genetic changes

Cytogenetic or molecular biological techniques have revealed a number of clonal chromosomal changes in childhood ALL ^30-33. A major group of these changes consists of alterations in the number of chromosomes as a feature of genomic instability as in any malignancy. Alterations in ploidy is common in childhood ALL and have prognostic significance, patients with hyperdiploidy having better prognosis ^{9; 33-36}. Among the specific genetic changes, chromosomal translocations are common in childhood ALL. The t(12;21) translocation, barely detectable when searched by conventional cytogenetic techniques, is the most frequent genetic lesion occurring in childhood ALL ³⁷. Certain translocations have a negative influence on prognosis, particularly in cases of t(9;22) or t(4;11); whereas, t(12;21) confers a better prognosis ^{9; 31; 33; 36}. During the production of a translocation, the chromosome is broken and the gene at this site may be disrupted. The broken gene comes to lie adjacent to another gene as the partner chromosomes fuse. This reorganization can lead to the production of a fusion protein which can contribute to the development of leukemia. An example of this is the TEL-AML1 fusion protein resulting from t(12;21) ^{9; 32; 37}. Among infants with ALL, translocations involving 11q23 / MLL occur in about 85% of cases ¹⁰. Also more frequently detected by molecular analysis is deletions on chromosome 6q ^{38; 39}. This chromosomal change is present in 32% of (adult) ALL cases.

Inherited genetic changes

In addition to the acquired genetic abnormalities, a number of inherited genetic syndromes are associated with childhood leukemia, although they lead to a small number of cases ^{3; 10; 40}. The best-known ones causing ALL are Down’s syndrome, neurofibromatosis, Shwachman syndrome, Bloom syndrome and ataxia-telangiectasia. Children with Down’s syndrome are estimated to have an approximate 10- to 15-fold increased risk for the development of acute leukemia (ALL or AML), the most common subtype being M7 (megakaryoblastic) variant of AML ^{41; 42}. The familial occurrence of leukemia mentioned above also suggests a genetic component in the etiology.

Epidemiology

Childhood ALL is a heterogeneous disease. Significant geographic variations in its incidence exist, with rates ranging from 9 to 47 per million ². Rates are highest in Costa Rica, low among US blacks, and lowest in Kuwait ^{2; 43}. There are variations in the incidence among whites, with rates being higher in New Zealand and Australia than in Europe. In England and Wales, there seems to be geographical clustering ^{44; 45}. This may be due to environmental exposures, infectious agents or unknown factors. The incidence of childhood ALL (below the age of 15 years) is approximately 30% higher among boys and male gender is a poor prognostic factor ^{36; 46-50}. In the case of T-cell disease, the male:female ratio is nearly 4:1 ⁵¹, and in infant leukemia there is a female predominance ³.

The distribution of childhood ALL in age groups is not homogeneous. A peak in incidence occurs between the ages of two and five years but only in common ALL ^{40; 43; 52; 53}. The age peak is absent in many developing countries ³, leading some to postulate that it may reflect environmental exposures associated with modernization ⁵⁴. In Africa, ALL is relatively rare before the age of five years ³ although this may be due to under registration. The absence of the age peak in other subtypes of childhood ALL underlines the importance of taking the heterogeneity of the disease into account in epidemiological studies. A number of epidemiological risk factors have been identified in childhood ALL (Table 1). Some of these are discussed below.

Table 1. Risk factors for childhood ALL identified in epidemiological studies

_______________________________________________________

Male sex

Down's syndrome and other genetic disorders

Sibling with leukemia, brain tumour or Down's syndrome

Middle and upper socioeconomic class

Miscarriage(s) in the maternal reproductive history

Advanced maternal and paternal age

Parental smoking

Parental or household exposure to pesticides

Parental history of autoimmune disorders

High birth weight

Being the first-born or the only child

Delayed exposure to common childhood infections

Prenatal ionizing (diagnostic) radiation exposure

Nitrous oxide administration during delivery

Post-natal use of chloramphenicol

Electromagnetic field exposure (?)

__________________________________________________________

Data compiled from Refs 3; 10; 55-57

See also the SEER childhood leukemia report (Table I.5) and Table 1 in Linet, 2003.

Socio-economic status

The speculation that the peak incidence of ALL in early childhood may reflect socio-economic factors prompted epidemiological studies to investigate this possibility. The epidemiological data have generally shown a consistently increased risk of ALL in children of the middle and upper socio-economic classes ^{58; 59} but there is also evidence against this ^{53; 60; 61}. Although socio-economic status can be confounded with race, personal habits, life style, access to medical care, maternal age, occupational exposures and parental education, there are also several etiologic hypotheses that attempt to explain this phenomenon, including delayed exposure to infectious agents associated with smaller families, less crowding, and later interaction with other children ^{3; 61-65}.

Maternal reproductive history

Since noted in the Oxford survey of childhood malignancies by Stewart et al in 1958 ⁶⁶, numerous studies have examined the association between miscarriages in maternal reproductive history and childhood leukemia ^{61; 67-73}. Prior fetal loss appears to be one of the most consistent risk factors for childhood ALL in different populations such as UK ⁶⁶, USA ^{68; 69; 73}, Holland ⁷⁰, and Germany ⁶¹. Only one study has found a lower risk associated with prior fetal loss in a Chinese population ⁷¹. One study failed to find any association in ALL ⁷⁴, and another one in infant leukemia ⁷⁵. It is particularly important that the ongoing US Children's Cancer Group case-control study has so far reported only the maternal history of fetal loss as a risk factor for childhood ALL ^{73; 76}. In that study, this association is significant only for those patients diagnosed before four years of age and most significant in those patients diagnosed before two years of age. In the latter group, one previous fetal loss is associated with a five-fold increased risk (P < 0.001), whereas, two or more fetal losses are associated with a relative risk [RR] of 24.8 (P < 0.001)