Acute Leukemias - Republican Scientific Medical Library

a 5.3 · Etiology 83
5.3.1.2 Cytogenetic Abnormalities
Cytogenetic abnormalities frequently found in ALL
cases include germ-line karyotypic abnormalities, somatic
karotypic abnormalities, translocations, and deletions.
The germ-line abnormalities associated with
childhood leukemia include Down syndrome (trisomy
21) [64, 118, 144], Bloom syndrome [64], Fanconi anemia,
Klinefelter syndrome, and ataxia-telangiectasia
[37]. The somatic abnormalities associated with childhood
leukemia include aneuploidy (in one form or another
in 92% of childhood ALL cases), pseudodiploidy
(in 41.5% of ALL cases) and hyperdiploidy (in 20–30%
of pre-B ALL and about 90% of early pre-B ALL) [108,
127].
Translocations frequently found in ALL cases include
the TEL-AML1 translocation (found in about
20–25% of B-lineage childhood ALLs) [11, 34, 91, 95,
125]; MLL translocations (found in about 70–80% of infant
leukemias [16, 53, 94, 109, 126], but less commonly
in other leukemias, both childhood and adult); MLL-
AF4 gene fusion (very common in infant ALL and also
found in ALL of older children); and other translocations
occurring in childhood ALL including t(9,
11)(p22;q23) [16, 53] and t(11, 19) [46], and CDK6-MLL
[112]. TEL-AML1 is found in about 1% of cord blood
specimens, yet only about 1% of those with this translocation
will develop ALL in childhood.
Deletion of 6q occurs in 11% of childhood ALL cases
[108]. Among childhood ALL cases with t(12, 21), 77%
also have 12p12–13 deletions [15].
In short, the chromosomes that are known to be involved
in karyotypic abnormalities found in childhood
ALL are 1, 4, 6–9, 11, 12, 14, 19, 21, and 22. Neither X nor
Y is known to be involved with childhood ALL. Translocations
are especially common in childhood ALL.
Triggers for molecular anomalies may be inherited
during pregnancy, and may develop during infancy or
early childhood [13].
5.3.1.3 Infectious Etiology
The most widely accepted current theory of causation of
childhood ALL is based on an infectious etiology associated
with decreased immune function. Three variations
on this theme of the “infection” that have been
put forward are (1) exposure to a specific infectious
agent postnatally, proposed by Kinlen [56], (2) exposure
to a specific infectious agent prenatally or around the
time of birth, proposed by Smith [137], or (3) a delay
in the initial exposure to infectious agents in general beyond
the first year of life, proposed by Greaves [39]. A
recent review of this topic is provided by McNally and
Eden [88], but does not resolve the controversy. We provide
a brief review of some of the literature supporting
each of these hypotheses.
5.3.1.4 Postnatal Infection by a Specific
Leukemogenic Pathogen
According to Kinlen’s hypothesis [56, 57], “outbreaks” of
ALL follow epidemics of some common (and perhaps
subclinical) infection, of which ALL is a rare outcome.
These outbreaks tend to occur when infectious and susceptible
populations come into close proximity or intermingle
(“population mixing”), thus facilitating the
spread of the pathogen. In relatively rural, isolated populations
it is more likely that a sizeable portion of the
population has not previously been exposed to the infectious
agent, and thus is susceptible.
Kinlen and colleagues have published many studies
that are consistent with this theory. For example, he has
documented high rates of leukemia mortality among
preschool children in Kirkcaldy, following a rapid population
increase due to the construction of the new town
of Glenrothes [56], and excess leukemia mortality in five
rural British New Towns founded between 1946 and
1950, which brought together new residents from a variety
of isolated (i.e., low-density) rural settings. The statistically
significant excess leukemia mortality was seen
in rural towns for the time period 1946–1965 but not
1966–1985, consistent with the population mixing hypothesis.
The excess leukemia mortality was not seen
in overspill towns, which were less rural and had smaller
rates of in-migration, also supporting the population
mixing hypothesis. The relationship between population
influx and childhood ALL received further support
from a Cumbria-based study of children of “incomers”
(both parents born outside Cumbria), who were at higher
risk of common ALL than children of “local residents”
(at least one parent born inside Cumbria) [21].
Studies in other countries found similar effects, such
as extraordinarily high childhood leukemia mortality
rates in Italy and Greece during 1958–1987, which have
been attributed to high levels of population mixing associated
with massive rural-to-urban migration in the
years following World War II [59]; a study in Hong Kong