NEW ORLEANS--Cytologic and biomarker assessment can be used to identify
a cohort of women at extremely high risk for short-term breast cancer development,
Carol Fabian, MD, said at the American Society of Preventive Oncology annual
This, in turn, raises the possibility that such a cohort could be enrolled
in breast cancer chemoprevention trials using biomarker changes as surrogate
endpoints rather than development of cancer.
The researchers, from the University of Kansas Medical Center, Kansas
City, previously reported that evidence obtained by random fine needle
aspiration (FNA) of epithelial hyperplasia with atypia and biomarker abnormalities
was more prevalent in women at increased epidemiologic risk for breast
cancer than in low-risk controls.
The current study included 335 women (median age, 44) who were at high
risk for breast cancer because of a first-degree relative with breast cancer,
prior precancerous biopsy, prior invasive cancer, or some multiple thereof.
They underwent FNA and were followed for subsequent cancer development
between 1989 and 1996. These results were compared with FNA findings and
cancer development in a group of paid volunteers who were at low risk for
The study found that cancer was more likely to develop in women with
FNA evidence of hyperplasia with atypia. In turn, hyperplasia with atypia
was predicted by the presence of multiple biomarker abnormalities--DNA
aneuploidy, overexpression of p53, epidermal growth factor receptor (EGFR),
and Her-2/neu overexpression.
FNA evidence of hyperplasia with atypia was present in 18% of women
with risk factors for breast cancer. An additional 13% had evidence of
hyperplasia without atypia but with multiple abnormal biomarkers. Thus,
in a population at increased epidemiologic risk, largely because of positive
family history (69% of study subjects), 31% had random FNA biomarkers associated
with a marked short-term risk of breast cancer, Dr. Fabian said.
Significant differences were found in low-risk vs high-risk subjects
in the prevalence of normal, hyperplastic, or atypical pathology. And except
for ploidy, there was a stepwise increase in all biomarkers with the degree
of cytologic abnormality.
Twice the Expected Rate
At 34 months' follow-up, 10 women had developed ductal carcinoma in
situ or invasive cancer; an additional four women developed lobular carcinoma
in situ. This is twice the expected rate, Dr. Fabian said.
Cancer developed in 5 of 40 subjects with both atypical hyperplasia
and multiple biomarker abnormalities; in 2 of 21 women with atypical hyperplasia
alone; in 2 of 70 with multiple biomarkers alone; and in only one of 203
subjects with neither atypical hyperplasia nor multiple bio-marker abnormalities.
All of these cancer cases occurred in women in the high-risk group;
however, Dr. Fabian pointed out that the low-risk volunteers were not followed
with the same intensity.
According to this analysis, 5% of high-risk women with atypical hyperplasia
or multiple biomarkers will develop breast cancer within three years, Dr.
Fabian said. The strongest predictors of breast cancer were atypical hyperplasia
and multiple positive biomarkers, especially EGFR and p53. One third of
subjects in the study had one or both of these findings.