A study published in the British Medical Journal [BMJ] on Feb 11, 2014, was featured on many of the news shows and the internet. The news commentators suggested that mammograms were not useful in detecting breast cancers in women under age sixty.
In order to gain an understanding of the issue, I did some investigating on my own. I read the Norway study, the Swedish study, and this newest Canadian study.
Women need access to reliable and dependable breast cancer screenings. Mammography has been criticized as outdated and unreliable. There is no doubt that mammography is insufficient for finding all breast cancers and it may lead to overtreatment and/or overdiagnosis. However, what is the alternative for early detection? Until an alternative for early detection is found, mammography screening for women age 40+ should continue.
Below is the Canadian National Breast Screening article excerpted from the British Medical Journal [BMJ] with my comments in yellow, the BMJ press release of the study, the Letters to the Editor, and my article about breast density. I hope you find them useful. AGEDOC
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Below is the excerpted article of the study with my
feedback in yellow.
Twenty
five year follow-up for breast cancer incidence and mortality of the Canadian
National Breast Screening Study: randomised screening trial
Published in
the British Medical Journal [BMJ] on 11 Feb 2014. Pages 348-366.
Authors and primary
investigators of this study:
1. Anthony B Miller, professor emeritus
2. Claus Wall, data manager
3. Cornelia J Baines, professor emerita
4. Ping Sun, statistician
5. Teresa To, senior scientist,
Abstract
Objective To compare breast cancer
incidence and mortality up to 25 years in women aged 40-59 who did or did not
undergo mammography screening.
Design Follow-up of randomised
screening trial by centre coordinators, the study’s central office, and linkage
to cancer registries and vital statistics databases.
Setting 15 screening centres in
six Canadian provinces,1980-85 (Nova Scotia, Quebec, Ontario, Manitoba,
Alberta, and British Columbia). Equipment was not controlled; some allegations of outdated equipment;
mammogram machines were not the same; no training for mammography screeners.
Participants 89,835 women, aged
40-59, randomly assigned to mammography (five annual mammography screens) or
control (no mammography). They
were not randomly assigned. Each woman
was given a physical exam at the beginning. Women with breast complaints were
placed in the mammogram group. Women with a history of BC were excluded. This
is NOT random assignment because it does not represent the population and they
were not divided randomly; therefore, findings cannot be extrapolated. This means
that the study has significant flaws and limited utility; findings/conclusions
are benign [no pun intended].
Interventions Women aged 40-49 in the
mammography arm and all women aged 50-59 in both arms received annual physical
breast examinations. Women aged 40-49 in the control arm received a single
examination followed by usual care in the community. Drop out rate: Over 93% of
participants in the control arm aged 40-49 returned their annual questionnaire,
whereas compliance with annual breast examination screening for those in the
control arm aged 50-59 varied between 89% (for screen 2) and 85% (for screen
5); only questionnaires were obtained for 3% to 7% of the women
Main
outcome measure Deaths from breast cancer. This is known as “problem-focused research.” Instead of measuring survival rates, they
measured death rates.
Results During
the five year screening period, 666 invasive breast cancers were diagnosed in the mammography arm (n=44,925
participants) and 524 in the controls (n=44 910), and of these, 180 women in
the mammography arm and 171 women in the control arm died of breast cancer
during the 25 year follow-up period. But how many survived because the mammogram found it early and they
didn’t die? The overall hazard ratio for death from breast cancer
diagnosed during the screening period associated with mammography was 1.05 (95%
confidence interval 0.85 to 1.30). The findings for women aged 40-49 and 50-59
were almost identical. No,
this is variance spread is not “almost identical.” During the entire
study period, 3250 women in the mammography arm and 3133 in the control arm had
a diagnosis of breast cancer, and 500 and 505, respectively, died of breast
cancer. How many survived?
That is what I want to know. Thus
the cumulative mortality from breast cancer was similar between women in the
mammography arm and in the control arm (hazard ratio 0.99, 95% confidence
interval 0.88 to 1.12). This hazard ratio confirms the
results were not “almost identical.”
After 15 years of follow-up a residual excess of 106 cancers was
observed in the mammography arm, attributable to over-diagnosis. This is flawed cause and effect.
Strengths
and limitations of this study
We believe that the lack of an impact of mammography screening
on mortality from breast cancer in this study cannot be explained by design
issues, lack of statistical power, or poor quality mammography. It has been
suggested that women with a positive physical examination before randomisation
were preferentially assigned to the mammography arm. Every study has limitations
and weaknesses and researchers are required to include them. In order to retain academic humility and
retain an unbiased stance, all known strengths and limitations are identified
and addressed. This is done so that next
time researchers conduct a similar study, the limitations are known. The researchers identified these flaws and
then explained why they were irrelevant.
This is arrogant, unprofessional, and biased.
Conclusion Annual
mammography in women aged 40-59 does not reduce mortality from breast cancer
beyond that of physical examination or usual care when adjuvant therapy for
breast cancer is freely available. Overall, 22% (106/484) of screen detected
invasive breast cancers were over-diagnosed, representing one over-diagnosed
breast cancer for every 424 women who received mammography screening in the
trial. This statement is
reckless. The study was seriously
flawed. The controls, hypothesis
[measuring death], and findings are overstated. I am being kind.
The article link:
Here is the BMJ Press
Release, a summary of the study:
Annual
screening does not cut breast cancer deaths, suggests Canadian study
Tuesday, February 11, 2014 - 13:56
Annual
screening in women aged 40-59 does not reduce mortality from breast cancer
beyond that of physical examination or usual care, concludes a 25-year study
from Canada published on bmj.com today.
Furthermore,
the study shows that 22% of screen detected breast cancers were over-diagnosed,
representing one over-diagnosed breast cancer for every 424 women who received
screening in the trial. Over-diagnosis refers to the detection of harmless
cancers that will not cause symptoms or death during a patient’s lifetime.
Regular
mammography screening is done to reduce mortality from breast cancer. Women
with small (non-palpable) breast cancer detected by screening have better long
term survival than women with palpable breast cancer. But it is not clear
whether this survival difference is a consequence of organised screening or of
lead time bias (when testing increases perceived survival time without
affecting the course of the disease) and over-diagnosis.
So
researchers based in Toronto, Canada decided to compare breast cancer incidence
and mortality up to 25 years in over 89,000 women aged 40-59 who did or did not
undergo mammography screening.
Women in
the mammography arm of the trial had a total of five mammography screens (one a
year over a five year period), while those in the control arm were not
screened.
Women aged
40-49 in the mammography arm – and all women aged 50-59 in both arms – also
received annual physical breast examinations. Women aged 40-49 in the control
arm received a single examination followed by usual care in the community.
During the
25 year study period, 3,250 women in the mammography arm and 3,133 in the
control arm were diagnosed with breast cancer and 500 and 505, respectively,
died of breast cancer. “Thus, the cumulative mortality from breast cancer was
similar between women in the mammography arm and in the control arm,” say the
authors.
At the end
of the five year screening period, an excess of 142 breast cancers occurred in
the mammography arm compared with the control arm, and at 15 years the excess
remained at 106 cancers. This, say the authors, implies that 22% of the screen
detected invasive cancers in the mammography arm were over-diagnosed – that is,
one over-diagnosed breast cancer for every 424 women who received mammography
screening in the trial.
They
stress that these results may not be generalisable to all countries, but say,
in technically advanced countries, “our results support the views of some
commentators that the rationale for screening by mammography should be urgently
reassessed by policy makers.”
While they
believe that education, early diagnosis, and excellent clinical care should
continue, they conclude that annual mammography “does not result in a reduction
in breast cancer specific mortality for women aged 40-59 beyond that of
physical examination alone or usual care in the community.”
In an
accompanying editorial, Dr. Mette Kalager and colleagues believe that long term
follow-up does not support screening women under 60.
They agree
with the study authors that “the rationale for screening by mammography be
urgently reassessed by policy makers,” but point out that this is not an easy
task "because governments, research funders, scientists, and medical
practitioners may have vested interests in continuing activities that are well
established."
Below are the BMJ journal Letters
to the Editor in response to this article.
These are GREAT! Please note their titles and
where they work. Here is the link for
the Letters to the Editor. These are
from 14 Feb, 2014:
Patrick
I. Borgen, Chairman of Surgery; Breast Surgeon Maimonides Medical
Center/Albert Einstein College of Medicine, 4802 Tenth Avenue Brooklyn New York
The real
tragedy is not that an article was reported which describes the results of a
deeply flawed trial - physicians can determine for themselves how to interpret
these results. The problem is that major media markets picked up the trial are
reported the results as a valid challenge to the routine use of mammography. It
is not. Most dramatically the New York Times (The Paper of Record) reported a
completely unbalanced view of the study, ignored larger, better studies and
left the reader with the impression that mammography just does not matter. This
is the height of irresponsibility. The BMJ simply reported the results of a
trial. It does make one wonder, however, if other journals demurred, realizing
that this paper advertises much more than it sells.
Eva
Rubin,
Diagnostic radiologist (breast imager)
Montgomery
Radiology Associates, 2055 Normandie Drive Ste 108, Montgomery, AL 36111
Computer scientists have an excellent term for nonsense output
when input is faulty--GIGO, garbage in, garbage out. This is, unfortunately,
what characterizes the Canadian breast cancer screening trial. The cascade of
errors in this trial has been eloquently described by others--from substandard
mammography (even for the time), inadequate training of mammography
interpreters, poor follow-up of mammographically detected abnormalities, and
nonrandomized allocation of women entered into the trial. High numbers of
participants and long-term follow-up of a poorly executed trial cannot be
expected to address the deficiencies present at the outset of this trial.
One need only look at Table 1 to realize that something was
seriously awry. In no way do the numbers reflect expected (and observed)
results when screening is performed properly. Not only was the average size of
the cancers detected by screening nearly the same as that of palpable cancers
in the control group but there is no evidence of a decrease in average tumor
size over time as one would expect for incident vs. prevalent cancers.
Although the authors claim that randomization was blinded, the
significant excess of breast cancers in the screening arm at year 1 coupled
with the relatively large average size of the cancers does not favor an
explanation of overdiagnosis. And what happened to overdiagnosis in the
subsequent 15 years when the number of cancers detected in the screening arm
was equal to or less than in the control arm? Overdiagnosis unquestionably
occurs in breast cancer screening with mammography but in a study focused on
invasive cancers with average tumor sizes of 1.9 cm, the plausibility of
significant numbers of 'overdiagnosed' lesions has to be seriously questioned.
The primary lesson we should learn from the Canadian trial is
that screening mammography done poorly does not work. Basing decisions about
screening on this trial--the only outlier among eight other randomized
controlled trials--is tantamount to having the asylum run by the inmates.
Jacob
Levman,
Researcher
University
of Oxford, Institute of Biomedical Engineering, Parks Road, Oxford, UK, OX1 3PJ
The recent publication by Miller et al. indicating no mortality
benefits from mammography as part of the Canadian National Breast Screening
Study is fraught with bias and the conclusions deeply suspect.
The study ignores pre-invasive cancers which have a high
survival rate and are often detectable by mammography. It is reasonable to
assume that if they had been included in the analysis then the mammography arm
of the trial would have demonstrated higher survival rates than those reported.
It is not safe to assume that all pre-invasive cancers are indolent (i.e. that
they would never go on to harm the patient), especially since the standard
model of malignant tumour growth has invasive cancers first developing through
a pre-invasive stage.
The study’s results contradict the authors’ conclusion that
mammography is not assisting in saving women’s lives. The section of the
Results titled “Breast cancer survival” indicates that “The 25 year survival
was 70.6% for women with breast cancer detected in the mammography arm and
62.8% for women with cancers diagnosed in the control arm” which the authors
demonstrate to be a statistically significant difference. This demonstrates a
real benefit to women surviving breast cancer thanks to receiving mammographic
screening. Had pre-invasive cancers been included in this study the difference
in 25 year survival is liable to have been even larger. Concluding that
mammographic screening provides no benefit with respect to saving women’s lives
based on the analysis presented is unfounded and dangerous.
A. B. Miller, C. Wall, C. J. Baines, P. Sun, T. To, S. A. Narod,
“Twenty five year follow-up for breast cancer incidence and mortality of the
Canadian National Breast Screening Study: randomised screening trial,” British
Medical Journal, 2014;348:g366.
Keith
Ericson, Assistant Professor of Markets, Public Policy, and Law
Boston
University School of Management, 595 Commonwealth Ave, Boston MA 02215
Dear Editors,
The recent study by Miller et al. (2014) claims to show that
annual mammography screening does not reduce breast cancer mortality. An
accompanying editorial concludes that the evidence does not support screening
women under 60 (Kalager, Adami, and Bretthauer 2014).
Both conclusions are wrong. In fact, this study’s best estimate
is that mammography reduced breast cancer mortality by 11.5 deaths per 100,000
women screened. Based on the 95% confidence interval on the effect, their
results are consistent with it being cost-effective to spend $12,000 per person
on such a screening program.
The reduction in 25-year cumulative mortality from breast cancer
by 11.5 deaths per 100,000 people screened can be found simply by comparing
deaths in the treatment group (500 out of 44925, or 1113.0 per 100,000) to
those in the control group (505 out of 44910, or 1124.5 per 100,000). Another
way of presenting this result is that they find a hazard ratio of 0.99: the
death rate of treatment divided by death rate of controls.
Why then do Miller et al. claim that annual mammography
screening does not reduce mortality from breast cancer? Because the 95%
confidence interval on the effect of screening includes the possibility that it
has no effect: the 95% confidence interval on the hazard ratio ranges from 0.88
to 1.12. (A hazard ratio of 1 means no difference in deaths between the two
groups.) Nonetheless, the point estimate is that annual screening saves lives,
even while the 95% confidence interval indicates substantial uncertainty.
Given this uncertainty, is continued annual mammography
screening worth continuing? Put another way, is it cost-effective? A very rough
calculation shows that it might well be. (A full cost-effectiveness calculation
is certainly worth doing). Suppose the value per life saved was set at $9.1
million (the value used by the U.S. federal government in 2013.) Then, a
mammography screening program for 100,000 people that reduced 11.5 deaths—the
point estimate in Miller et al. 2014— has a benefit from life saved of $104.7
million. Thus, a screening program with a cost $1047 per person would be cost-effective.
Moreover, the 95% confidence interval from Miller et al. also consistent with a
reduction of 135 deaths per 100,000 (i.e. a hazard ratio of 0.88). This would
entail a benefit from life saved of $1.23 billion, and justify a program cost
of $12,300 per person.
While Miller et al. could not reject the hypothesis that annual
mammography screening had no effect, they can also not reject the hypothesis
that screening saves 135 lives per 100,000 screened, which would justify a
screening program substantial costs. We still don’t know for sure—and there are
other costs to consider, such as the risk of over treatment— but this study’s
best estimate is that annual mammography screening reduces breast cancer
mortality.
References
Miller, A. B., C. Wall, C. J. Baines, P. Sun, T. To, and S. A.
Narod. 2014. “Twenty Five Year Follow-up for Breast Cancer Incidence and
Mortality of the Canadian National Breast Screening Study: Randomised Screening
Trial.” BMJ 348: g366–g366. doi:10.1136/bmj.g366.
Kalager, M., H.-O. Adami, and M. Bretthauer. 2014. “Too Much
Mammography.” BMJ 348: g1403–g1403. doi:10.1136/bmj.g1403.
Carmen
Vidal,
Breast Cancer Screening Program Coordinator, Montse Garcia
Catalan
Institutet of Oncology - IDIBELL, Av. Gran Via 199-203, 08908 Hospitalet de
Llobregat (Spain)
Dear Editor,
The Canadian National Breast Screening Study has serious sources
of distorsion (methodological issues) that have been already mentioned in
previous responses to this article. In addition, we have some concerns regarding
how the authors presented their findings that can lead to misleading decisions
regarding screening.
We wondered why the authors did not stratify the analysis
according to age groups. Evidence indicates that there is more harm and less
gain for women aged 40-49 years in terms of the number of deaths prevented.
Another reason to stratify according to age groups is that they were not
comparable in terms of the control arm. Younger women (40-49 years) allocated
to the control arm were told to remain under the care of their family doctor
whereas older women (50-59 years) were offered annual physical examinations.
The Independent UK Panel on Breast Cancer Screening stated that
although there is no single optimum way to estimate overdiagnosis the two most
useful estimates are: from the population perspective, the proportion of all
cancers ever diagnosed in women invited to screening that are overdiagnosed and
from the perspective of a woman invited to screening, the probability that a
cancer diagnosed during the screening period represents overdiagnosis.
The authors missed the opportunity to estimate overdiagnosis
using the population perspective. That is, the excess of cancer as the
proportion of cancers diagnosed over whole follow-up period in women invited for
screening (Table 1).
They did not clarify how they estimated overdiagnosis, there is
no explanation why they measured the excess cancer cases at 15 years -instead
of 25 years of follow-up- and why they considered as the denominator 484
screening detected cancers -not reproducible data- .
References
1. Independent UK Panel on Breast Cancer Screening. The benefits
and harms of breast cancer screening: an independent review. Lancet.
2012;380.1778-86.
About breast density by Jan Vinita White
Ultrasounds and MRI’s
in addition to mammograms are recommended for women with dense breasts. Studies show that women with dense breasts
have a higher risk for breast cancer, and that 60% of cysts are missed with
mammography alone. However, ultrasounds
and MRI’s add cost. In 2009, Connecticut passed a “notification” bill whereby
women with dense breast tissue must be informed of their increased risk and
insurance companies MUST pay for an ultrasound screening.
Jean Weigert, who is
an executive with the Radiological Society of Connecticut, lobbied against the
notification bill on the grounds that it would increase costs and anxiety
without much benefit. However, she reversed her opinion later when women with dense
breast tissue were given ultrasound tests after the law passed.
Reviewing data on
78,000 women in six different practices, Weigert found that about half of the [8,651]
patients with dense breast tissue went on to have ultrasounds which found 28
cancers that were not visible on mammograms. “According to Weigert's analysis,
the additional screening for those 8,652 women cost $2.15 million, or $110,000
for each additional cancer found. Finding cancers early saves many times that
amount compared with the cost of late-stage cancer treatment, she notes” (WSJ
Health Journal, 2012, para. 8). I agree.
From the Wall Street Health
Journal dated Aug 6, 2012. Here is the link:
