NEW ORLEANSWhen talking with women about their personal
risk of developing breast cancer, terms such as relative risk
are not very useful, Patricia Kelly, PhD, said at the American
Society of Breast Disease annual meeting, co-sponsored by the Ochsner
Medical Foundation, New Orleans.
Women can be easily misled by statistics from clinical research
trials, said Ms. Kelly, a specialist in breast cancer risk who serves
as a medical geneticist at Catholic Healthcare, San Francisco. She
called for investigators to translate their findings into a more
meaningful format and for clinicians to communicate such information
more effectively to patients.
Physicians should focus less on concepts such as relative
risk or percent increase or decrease in risk.
Instead, she said, they should speak in terms of the actual numbers
of women who will be affected and the patients risks for the
next year and for the next 5 years.
Relative risk is a way to keep biostatisticians and
epidemiologists from being criticized, she said. It is
time to move beyond taking care of the epidemiologists and get the
information we need for our patients. We need to demand from those
who publish data that they give us information that is useful to us
and to our patients. We should ask that the studies present the risk
as risk over time.
Risk of the BRCA2 Mutation
Women generally perceive their risk to be much higher than it
actually is because of the manner in which such information is
provided, she said.
As an example, she pointed to the risk of breast and ovarian cancer
conveyed by a BRCA2 mutation. Among carriers aged 31 to 39, the
lifetime risk of developing breast cancer is 11% and there is no
increased risk of ovarian cancer; at ages 40 to 49, the risk is 16%
for breast cancer and 0.4% for ovarian cancer; at ages 50 to 59, it
is 20% and 7%, respectively; and at ages 60 to 69, it increases to
36% and 20%.
Several major breast cancer trials illustrate her point, she said.
The Nurses Health Study, for example, found a 1.5-fold increased risk
of breast cancer among women who took hormone replacement therapy.
Such increased risk turned many women against the idea of
hormone replacement therapy, although their actual risk of developing
breast cancer is much lower than it sounds.
The 1.5 increase in risk in this study actually meant that
among 100 women who took hormones until age 60, there were 4.1 cases
of breast cancer, compared with 3.8 cases among women not taking
hormones. The difference was 0.3 cases. For women who took hormones
to age 70, there were 6.9 cases per 100 vs 6.3 for those not on
hormones, she said. Moreover, most studies have found no
increase in the risk of breast cancer in women taking hormone
She applied the same logic to the NSABP Tamoxifen Prevention Trial,
which found a significant 49% reduction in risk among healthy
high-risk women taking tamoxifen (Nolvadex). This number was
impressive to most people, she said, but its the
difference between 4.3 cases of invasive breast cancer vs 2.2 per 100
women over 5 years time.
Patients need to have the risk put into perspective, she said.
For comparison, what is the risk of choking while eating?
she asked. And they need to hear from us loud and strong what
degree of risk is reasonable and how small a risk of 1% a year, for
example, really is.
In considering tamoxifen chemopre-vention, she said, my
patients respond very differently to hearing 49% reduction
on the one hand and 2 women per 100 over 5 years on the
other. Were talking about very small numbers.
Additionally, she said, recent data indicating a 20% increased breast
cancer risk from oral contraceptives really translated into one case
out of 3,700 women, and a 40% increase in risk imparted by two
glasses of red wine daily amounts to one additional cancer per 1,500
She advised physicians to read critically and to look at the spread
of the confidence interval. If it is very wide, it is telling
you the numbers are too small on which to base a definitive (or even
approximate) clinical decision.
A 9-year breast cancer survivor and spokesperson for the Louisiana
Breast Cancer Task Force praised Dr. Kelleys efforts to
encourage physicians at the meeting to communicate more effectively
with women. Rather than empowering women, we are frightening
women to death with the statistics, said Cathie McMichael, of
Andrew Seidman, MD, of Memorial Sloan-Kettering Cancer Center, added,
There is an epidemic of fear. We physicians have a tremendous
responsibility to make our patients understand the statistics and how
they impact their own lives.
Patients can be easily misled by statistics, Dr. Seidman said.
Women considering taking tamoxifen to reduce their risk for
breast cancer need to know the absolute benefit in reducing their
risk this year and over the next 5 years, he commented.