26 / 32
Annals of Internal Medicine
Take-home message
•
The authors evaluated the relationship among breast cancer screening, detected tumour
size, and overdiagnosis rate using a cohort of Danish patients. Screening was not
associated with a lower incidence of advanced tumours; however, the incidence of
nonadvanced tumours increased with screening.
•
The overall rate of over-diagnosis was 48.3% in 2010.
Abstract
BACKGROUND
Effective breast cancer screening
should detect early-stage cancer and prevent
advanced disease.
OBJECTIVE
To assess the association between
screening and the size of detected tumors and
to estimate overdiagnosis (detection of tum-
ors that would not become clinically relevant).
DESIGN
Cohort study.
SETTING
Denmark from 1980 to 2010.
PARTICIPANTS
Women aged 35 to 84 years.
INTERVENTION
Screening programs offering
biennial mammography for women aged 50
to 69 years beginning in different regions at
different times.
MEASUREMENTS
Trends in the incidence of
advanced (>20 mm) and nonadvanced (≤20
mm) breast cancer tumors in screened and
nonscreened women were measured. Two
approaches were used to estimate the amount
of overdiagnosis: comparing the incidence of
advance and nonadvanced tum-
ors among women aged 50 to 84
years in screening and nonscreen-
ing areas; and comparing the
incidence for nonadvanced tumors
among women aged 35 to 49, 50 to
69, and 70 to 84 years in screening
and nonscreening areas.
RESULTS
Screening was not asso-
ciated with lower incidence of
advanced tumors. The incidence
of nonadvanced tumors increased
in the screening versus prescreen-
ing periods (incidence rate ratio,
1.49 [95% CI, 1.43 to 1.54]). The first
estimation approach found that 271
invasive breast cancer tumors and
179 ductal carcinoma in situ (DCIS)
lesions were overdiagnosed in
2010 (overdiagnosis rate of 24.4%
[including DCIS] and 14.7% [excluding
DCIS]). The second approach, which
accounted for regional differences in
women younger than the screening
age, found that 711 invasive tumors
and 180 cases of DCIS were overdi-
agnosed in 2010 (overdiagnosis rate
of 48.3% [including DCIS] and 38.6%
[excluding DCIS]).
LIMITATION
Regional differences com-
plicate interpretation.
CONCLUSION
Breast cancer screening
was not associated with a reduction in the inci-
dence of advanced cancer. It is likely that 1
in every 3 invasive tumors and cases of DCIS
diagnosed in women offered screening rep-
resent overdiagnosis (incidence increase of
48.3%).
Breast cancer screening in Denmark: a cohort
study of tumor size and overdiagnosis.
Ann
Intern
Med 2017 Jan 10;[EPub Ahead of Print],
KJ Jørgensen, PC Gøtzsche, M Kalager, et al.
Breast cancer screening:
tumour size and overdiagnosis
EDITOR’S NOTE
By Lee S. Schwartzberg
MD, FACP
S
ince mammographic screening for
breast cancer began 40 years ago,
our knowledge of the disease has
increased immeasurably. We know now
that there are different subgroups of
breast cancer with markedly different nat-
ural histories. Knowing this has given rise
to the concept of overdiagnosis, meaning
finding a tumour in a patient that is des-
tined to never be life-threatening. Hence,
working to diagnose this increases anxiety
without providing net health benefit. We
still do not know exactly how to predict
this on an individual tumour level.
Using the Danish staggered experience
with the introduction of screening mam-
mography and a relatively complete
national cancer registry, researchers pub-
lishing in
Annals of Internal Medicine
claim that screening does not reduce
the incidence rate of advanced breast
tumours. Their definition of advanced
tumour is somewhat arguable as it is lim-
ited to a size definition of >2 cm without
accounting for biologic subtype or posi-
tive lymph nodes. The advanced tumour
rate did not decrease over time, although
DCIS diagnosis rates increased substan-
tially. In another analysis, they estimated
the over-diagnosis rate to be 24% to 48%
of all tumours when considering both
invasive and DCIS detection.
The use of mammography continues to be
controversial because of data like these.
Yet, the message to the lay public should
continue to emphasise that mammogra-
phy, while not a perfect test, saves lives.
Clearly, we still have much to learn about
the optimal frequency of screening and
the best age to start and stop screening
from a population perspective. For an
individual woman, appropriate use of
screening should continue at intervals and
ages recommended by her physician.
Dr Schwartzberg is
a senior partner and
Medical Director
of the West Clinic,
a 30-physician
practice specialising in
oncology, haematology
and radiology
located in Memphis, Tennessee.
Screening was not
associated with a lower
incidence of advanced
tumours; however, the
incidence of nonadvanced
tumours increased with
screening.
BREAST
26
PRACTICEUPDATE HAEMATOLOGY & ONCOLOGY