Y. Remvikos et al., GENETIC EVOLUTION OF BREAST CANCERS .3. AGE-DEPENDENT VARIATIONS IN THE CORRELATIONS BETWEEN BIOLOGICAL INDICATORS OF PROGNOSIS, Breast cancer research and treatment, 34(1), 1995, pp. 25-33
The influence of age on the occurrence of phenotypic features of progn
ostic significance was studied in relation to the DNA index values, me
asured on DNA histograms from a series of 1019 breast cancer patients.
Globally, the distributions of all parameters showed variations with
age, the most prominent being the decreases in the percentage of estro
gen receptor-negative and high proliferative activity cases with incre
asing age. When analyzed according to the DNA index classes, all param
eters were found to some extent linked with the stage of genetic evolu
tion. However, the associations varied with age, defining two extreme
groups. The younger patients (less than 40 pears) presented a more com
plete acquisition of the 'aggressive' phenotype and near-triploid tumo
rs from this group were very frequently steroid hormone receptor-negat
ive, high proliferation, and grade III. By contrast, near-triploid tum
ors in patients above 65 presented relatively frequently as receptor-p
ositive, low proliferative activity, and even grade I. The correlation
of the proliferative status with steroid hormone receptor content led
to similar conclusions, high proliferation being more strongly correl
ated with the absence of estrogen and progesterone receptors in younge
r patients. Interestingly, the association between high proliferation
and negative progesterone receptors was much weaker in patients above
55. Our results suggest that the currently established biological prog
nostic factors, including DNA profile, steroid hormone receptors, and
histopathological grade, show patterns of association which vary with
age. Of these, only progesterone receptor could be influenced by menop
ausal status. These findings have to be taken into consideration for f
uture prognostic factor-related treatment decisions, but also for futu
re methodological improvements of multivariate survival analyses.