TY - JOUR
TI - Insulin-like growth FACTOR-I and binding protein-3 in relation to
childhood leukaemia
AU - Petridou, E
AU - Dessypris, N
AU - Spanos, E
AU - Mantzoros, C and
AU - Skalkidou, A
AU - Kalmanti, M
AU - Koliouskas, D
AU - Kosmidis, H and
AU - Panagiotou, JP
AU - Piperopoulou, F
AU - Tzortzatou, F
AU - Trichopoulos, D
JO - International Journal  of Cancer
PY - 1999
VL - 80
TODO - 4
SP - 494-496
PB - Wiley-Liss, Inc.
SN - 0020-7136
TODO - 10.1002/(SICI)1097-0215(19990209)80:4<494::AID-IJC2>3.3.CO;2-B
TODO - null
TODO - The aetiology of most cases of childhood leukaemia remains unknown, but
several studies have indicated that increased birthweight and height are
risk factors for the disease. Since insulin-like growth factor-I (IGF-I)
mediates the effect of growth hormone and has been positively associated
with prostate cancer, we have evaluated the role of this hormone and its
principal binding protein, IGFBP-3, in the aetiology of childhood
leukaemia. Incident cases of childhood leukaemia from those recorded by
a national network of childhood oncologists were enrolled in our study.
Controls were children hospitalised for acute conditions of no more than
moderate severity with matching for gender, age and maternal place of
residence. Blood measurements of IGF-I and IGFBP-3 were undertaken using
commercially available radioimmunoassays. Serum IGF-I values decreased
by about 1.7% per month, and the rate of decline was higher, though not
significantly so, among cases (2.1% per month) than among
controls(1.4%). There was no significant association between IGF-I and
the likelihood of childhood leukaemia, but an increment of I mu g/ml of
IGFBP-3 was associated with a substantial and statistically significant
reduction of childhood leukaemia by 28% (95% confidence interval 7%
to 45%). Because IGFBP-3 is essentially a binding protein, we interpret
our findings as indicating that bioavailable IGF-I may play an important
role in the aetiology of childhood leukaemia. The much smaller
quantities and the inherent instability of IGF-I in the blood in
comparison to those of IGFBP-3 are likely to hinder documentation of an
underlying positive association of IGF-I with the disease. (C) 1999
Wiley-Liss, Inc.
ER -