Twenty five year follow-up for breast cancer incidence and mortality of the Canadian National Breast Screening Study: randomised screening trial
BMJ 2014; 348 doi: https://doi.org/10.1136/bmj.g366 (Published 11 February 2014) Cite this as: BMJ 2014;348:g366
All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
A lot of critical responses to the Canadian National Breast Screening Study focus on the critique of various medical and diagnostic aspects. While this could be valid and important, the critique proposed below focuses on mathematical analysis and logical conclusions made in the article that frankly are very worrisome due to a sensationalism and response in the general media to the conclusion of the study. This response culminated yesterday in clearly outrageous article “Cancer screening expert to radiologists: Stop lying about mammograms” by Karen Kaplan, published in LA Times
http://www.latimes.com/science/sciencenow/la-sci-sn-mammograms-breast-ca...
There are many objections to the statistical and logical approach used in the article, but I will focus only on one most important conclusion made by the study that seems to resonate the most in the general media and public – conclusion that mammography has no benefit for the screening of breast cancer. This conclusion is deeply flawed and not supported by mathematically accurate statistical and logical analysis. I discussed this matter with Dr. Miller during extensive private communication over the previous weekend. While he agreed with many of my statements and did not respond to mathematical arguments, Professor Miller still holds his position and stands behind the study conclusion about uselessness of mammography for breast cancer screening. I think that refusal to re-consider ones position in light of serious arguments is a wrong approach that will be seriously damaging to the health of millions of women. .
First of all, Dr. Miller acknowledged that diagnosis of breast cancer is made only after histological study of a suspicious area detected in breast study either by palpation or mammogram. What he wrote was specifically "That is the whole problem of over-diagnosis, the lesion appears to be a cancer when examined histologically. However, this does not necessarily reflect the behavior of the cancer. We now recognize that cancers can be detected following mammography screening that would never have grown to a size when they would have been detected clinically." Professor Miller also acknowledged:
a) "Unfortunately, at present there are no tests that can be applied to separate off the over-diagnosed cancers from the cancers that if not treated would kill”
b) “That is why, for ethical reasons, all are treated.“ meaning all histologically confirmed cancers are treated regardless whether they were initially detected by palpation or by mammogram.
However, despite these acknowledgements authors managed somehow to come to completely illogical conclusion because more cancers detected by mammogram than palpation and relatively fewer patients died AFTER TREATMENT in mammogram group than in the control (palpation detected) group, there was “over-diagnosis” of cancers using mammography. Such conclusion, obviously, does not withstand logical scrutiny. By this logic if no patients diagnosed with breast cancer in mammogram group would die, it would mean that they all were “over-diagnosed”.
On my suggestion that the term “over-diagnosed” is not appropriate in this context and rather terminal cancer term should be used Prof. Miller responded “Most pathologists do not like the term-overdiagnosis, but it is now in common use, at least in the epidemiology community, to which I belong.” This obvious mislabeling of outcome of both diagnosis and treatment as well as flawed logic in analysis makes the conclusion of uselessness of mammography for screening of breast cancer unacceptable. I urge authors of the study to correct this conclusion and try to remedy the damage done by the general media echoing this conclusion by printing editorial opinions at least in most important publications such as New York Times, Washington Post, LA Times and few others.
Competing interests: No competing interests
In his rapid response below Lazlo Tabar does not declare any conflict of interest. Tabar started Mammography Education Inc about 35 years ago, before he became the principal investigator of the Two-County Trial. This company is a very lucrative business. In the 1990s he made about 5 million Skr per year (1) which is close to USD 1 million. The internet portal Manta estimates an annual revenue for Mammography Education Inc of US$1•0—2•5 million per year in 2012 (2). He has made at least USD 25 million on mammography screening. Twenty-five million dollar is a minimum estimate. In addition Tabar is involved in several other companies (2).
1. Atterstam I. Svensk röntgenläkare tjenar miljoner på mammografi . SvD Feb 16, 2002
2. Riva C, Biollaz J, Foucraz P, Junod B, Nicot P, Spinosa J-P. Effect of population-based screening on breast cancer mortality. Lancet 2012; 379: 1296.
Competing interests: No competing interests
Although my colleague Dr Cornelia Baines has responded to many of the comments made by Lazlo Tabar on the Canadian National Breast Screening Study, I should like to add a few more from my own perspective.
Lazlo Tabar appears to downplay the importance of efficacy trials, which enroll volunteers with informed consent. Screening is a voluntary activity, a randomized trial without informed consent is regarded as unethical in North America, and increasingly in Europe. Randomization by individual enables the collection of detailed information by risk factor on each enrollee, we have shown our trial was equally balanced by all criteria considered. Not so the Two-County trial, where the controls were only contacted after screening ceased in the active study population. Tabar claims that Duffy et al (2003) demonstrated balance in the Two-County trial, however the only relevant data published by Duffy et al related to age, and the balance was not good.
Tabar makes much of the exclusion of our trial in women age 50-59 from the IARC (2002) evaluation; what he fails to acknowledge is that the Working Group was evaluating in this age group the comparison of mammography only vs. no screening, to which our trial can not contribute as in that age we were seeking an answer to the question “how much does mammography contribute to the reported benefit of mammography plus physical examination of the breasts”. The answer as we have demonstrated, is no benefit, but much over-diagnosis. However, our trial in women age 40-49 was included in the IARC evaluation of women in that age group.
Tabar claims we made a flagrant error in our paper in the headings of Figures 2-4. In fact he is in error, the curves are based upon mortality, we used the term survival in the legend to draw attention to the high proportion of women who were still alive.
He also repeats the fact some of our external consultant radiologists resigned. The reason they did so was not the one he gave, it was because they wished to take over the role already being ably performed in our trial by the late Douglas MacFarlane as reference radiologist, and that to transfer that role to external consultants in the USA was not logistically feasible. Fortunately we found another external expert, who took over from them in the later years of the trial.
Tabar repeats some of Norman Boyd’s comments on our randomization. Norman worked with me during the pilot phase of the trial in Toronto, he agreed with me at the time that telephone calls to secure random allocation would not be feasible, given the wide geographical spread of the centres with up to 4 hours time differences, and at that time, outsourcing computer-based allocations was not possible in the absence of the internet. He also fails to mention that, as I have already stated in these columns, our allocation was evaluated by Bailar and MacMahon (1997) and found fully acceptable.
His comments on “advanced” breast cancer have already been answered by me in my response to Kopans. To repeat, mammography will initially increase the detection of advanced breast cancer, but if it does not reduce breast cancer mortality, then the numbers will equalize as we demonstrated.
I leave it to others to judge which of the two trials we are comparing is more likely to inform policy now. Cornelia Baines has nicely set out the differences. A trial in rural Sweden with no adjuvant therapy for breast cancer available and with far too little detail published on the quality control procedures applied is not a good basis for policy decisions.
References
Bailar JC, MacMahon B. Randomization in the Canadian National Breast Screening Study. Report of a review team appointed by the National Cancer Institute of Canada. CMAJ 1997; 156: 213-5.
Duffy SW, Tabar L, Vitak B, Yen MF, Warwick J, Smith RA, Chen HH. The Swedish Two-County Trial of mammographic screening: cluster randomisation and end point evaluation. Annals of Oncology 14: 1196–1198, 2003
Competing interests: No competing interests
We do not want to go back to the Dark Ages of Breast Cancer Screening
László Tabár, MD, FACR (Hon) and Tony Hsiu-His Chen, DDS, PhD
The publication by Miller and colleagues in the British Medical Journal used substandard and outdated mammography technique to claim that mammography
screening would add nothing beyond what physical examination can accomplish. It
takes much more than poor mammography to reach such a conclusion, many other
things would have to go wrong and they did go wrong in the Canadian trials.
The Canadian trials were a failure from the beginning
It is a well-known fact that the Canadian trials were not population based. Instead, these were trials using motivated volunteers, who represented only 26-27%
of the invited population. It is particularly important to note that the Canadian trials could not evaluate the independent impact of mammography because of the confounding effect of physical examination. For these reasons the Canadian trials were excluded from the meta-analysis of the impact of early detection using
mammography upon mortality from breast cancer by the World Health Organization’s
IARC workshop in 2002, even though AB Miller chaired the section. It is quite
surprising that three authors of the current BMJ article (Miller, Baines, Narod) are on record as having agreed with the following conclusions of the WHO-IARC publication (IARC Handbooks of Cancer Prevention, Vol. 7: Breast Cancer Screening. Lyon:IARC; 2002.):
“There is sufficient evidence for the efficacy of screening women aged 50-69 years
by mammography as the sole screening modality in reducing mortality from breast
cancer”.
“There is inadequate evidence for the efficacy of screening women by clinical breast
examination in reducing mortality from breast cancer”.
“There is inadequate evidence for the efficacy of screening women by breast selfexamination
in reducing mortality from breast cancer”.
The scientific community with good reason and with the assent of Miller and colleagues excluded the Canadian studies from the assessment of mammography screening.
Nevertheless, twelve years later AB Miller and colleagues revert to their earlier opinion that mammography does not have an effect on mortality from breast cancer.
The Canadian trials are a total failure for many reasons The Canadian trials stand apart as the only ones among the published screening trials that
failed to show any effect on mortality from breast cancer. The flagrant professional error in Figures 2-4, claiming that survival curves are actually “breast cancer specific mortality”
curves does little credit to the peer review process carried out by the British Medical Journal as well as the authors, all of whom should know better. The current BMJ article does not provide news for anyone, since these trials have been plagued with serious quality assurance problems that have been well known and carefully documented over the past few decades. After Drs. Wende Logan and Stephen Feig resigned as outside reviewers of the Canadian trials, I (LT) was invited by AB Miller to review their image quality. Dr. Cornelia
Baines was assigned to record my evaluation of the relevant mammography quality
assurance parameters for 50 selected cases from the Canadian trials. At case 15 she
terminated the review because of the overwhelmingly negative ratings given to every case.
The next day at the Committee meeting I should have recommended immediate termination of the Canadian trial, and I regret that I did not do so. My protests would most likely have been to no avail, as were the previous protests of Drs. Wende Logan and Stephen Feig, followed by the vehement protests, among others by Drs. Dan Kopans from Harvard University, and Martin Yaffe and Norman Boyd from Toronto. It is worthwhile mentioning the publications of Dr. Norman F. Boyd and colleagues in Radiology (1993;189:661-3) and in
the Canadian Medical Association Journal (1997;156(2):207-9). He and his co-authors
wrote: “Questions about randomization in the NBSS however, have subsequently been
raised because of the study’s findings that showed an imbalance in the number of women
with advanced breast cancer. In particular, the mammography arm had an excess of
patients with involved axillary lymph nodes whose breast cancer had been detected by
physical examination at baseline (i.e., before randomization)”. Boyd’s early warning about the imbalance in the number of advanced cancers in the two arms of the Canadian trials was right on target, because it is a well-known fact that the most sensitive predictor of the forthcoming mortality results in a screening trial is whether the screening in question
decreases the advanced cancer rate or not. The Canadian trials are the only ones among
all the trials that did not report any decrease in the advanced cancer rate. This is not surprising because of the above mentioned imbalances, but also because the radiologists were not trained for the demanding task of finding early breast cancers. Boyd and coauthors also wrote: “We now know that names were altered and that there were more alterations in the mammography arm. Although 78% of these changes could be accounted for in some way (e.g., clerical errors), the remaining 22% (representing 101 names) also now know that a coordinator at one of the NBSS centres was suspected of assigning her friends to the mammography arm and that the suspicion was strong enough to remove her from her position. We do not know the method of subversion thought to have been used by this coordinator, nor do we know whether an examination of name alterations revealed what she was alleged to have done”.
“Comparison with other trials”
There is no comparison since the Canadian trials are uniquely poorly designed and carried out and therefore have been much criticized for decades. Miller and his colleagues make an attempt to compare the Canadian trials only with the Swedish Two-County Trial, although their title is “Comparison with other studies”. This comparison is inappropriate for many
reasons and the statements published in the BMJ article are contrary to fact: the
randomization did not take place on “county level” as the authors claim, since a Swedish county corresponds to a province in Canada or a state in the US; the average tumor size in the control group was not 28 mm but 22 mm; their false claim that the significant reduction
in mortality observed in women invited to screening in the Two-County Trial is due to an initial imbalance has already been shown to be unfounded and AB Miller and his colleagues should be aware of this publication (Duffy et al, Ann Oncol 2003;14:1196-8); furthermore, the fact that mortality from breast cancer was compared among women invited vs. not invited women implies that the true impact of screening among those who actually attended
was considerably higher than 31%; the Swedish Two-County Trial was population based, the Canadian trials were not.
The following data can inform the reader about the true difference between the CNBSS and the Two-County Swedish Trial: the Swedish Trial did decrease the advanced cancer rate in
women invited to screening, the Canadian studies had an excess of advanced cancers on the MP (mammography) arm; The Swedish Trial has accomplished a significant decrease in mortality from breast cancer, the Canadian trials did not! According to the original report of the CNBSS the relative rate of node positive cases in the MP (mammography) arm versus the UC (control) arm in the age group 40-49 and the age group 50-59 were 1.53 (95% CI:
1.11 – 2.09) and 1.07 (95% CI:0.80-1.44) respectively and 1.27 (95% CI:1.02-1.57) for the two age groups combined. The failure of reducing advanced breast cancer therefore resulted in the lack of benefit in reducing breast cancer mortality in the CNBSS. By contrast, the corresponding figure of reducing advanced cancer for women aged 40-74 was 0.75 (95% CI:0.65-0.87) that was reported in the first article of the Swedish two-county trial
(Tabar et al, 1985). The reduction in advanced breast cancer also led to 31% reduction in mortality from breast cancer in women invited to screening.
The current publication in the BMJ confirms Dr. Boyd and colleagues’ conclusion published 21 years ago: “Taken at face value, the results of the NBSS argue for abandoning mammographic screening as a population-based means of controlling death rates from breast cancer. We believe such a conclusion to be unjustified findings of the NBSS.” “…the results of these trials should not be used to change the
prevailing scientific view of the potential benefits of screening with mammography” (Norman F. Boyd and colleagues in Radiology (1993;189:661-3).
Do people really believe the misleading information that new treatments render early detection unnecessary as Drs. Kalager and Miller wants you to believe? The facts clear the issue: follow-up of 9040 patients with breast cancers diagnosed between 1998–2003 showed that even in the era of adjuvant systemic therapy, node-negative cancers still have
considerably better survival than node positive. While survival of women with node-positive tumors has improved in the era of adjuvant systemic therapy compared with earlier times, it clearly is a worthy goal to have the cancer diagnosed before it spreads to the lymph nodes.
Screen detected node positive cases had a better survival than clinically detected node negative cases. Thus, improvements in therapy have not rendered early detection
redundant.
Competing interests: No competing interests
Department of Public Health 18 February 2014
University of Copenhagen
Letter to Editor of BMJ
Miller et al. recently published the 25 year follow-up of the Canadian National Breast Screening Study (1). It is very useful that the initiators of this large trial have made the effort to assess the long-term outcome.
The authors concluded that “annual mammography for women aged 40-59 does not reduce mortality from breast cancer beyond that of physical examination and usual care” and that “22% of screen detected invasive breast cancers were overdiagnosed”. It might though be useful to consider the results in some details.
The authors state that their “primary analysis includes only deaths from invasive breast cancers diagnosed during the screening period”. Out of the 666 breast cancer cases diagnosed in the mammography arm during the five intervention years of the trial, 180 died from breast cancer during the follow-up period. Out of the 524 breast cancer cases diagnosed in the control arm of the trial, 171 died from breast cancer during the follow-up period. Using Cox proportional hazards model this resulted in a hazard ratio (HR) of 1.05 (95% confidence interval (CI) 0.85 to 1.30). It should be noted, however, that an analysis including only deaths from breast cancers diagnosed during the screening period will be bias, because the pool of breast cancers from which deaths can occur will due to the lead time be larger in the mammography arm than in the control arm. The breast cancer mortality over the 25 years of follow-up will not be affected by lead time, and the authors may argue that the HR for this outcome of 0.99 (95% CI 0.88 to 1.12) supports the conclusion of no effect. However, during the 25 years a total of 3250 breast cancers were diagnosed in the mammography arm of which only 15% could have benefitted from the intervention in the trial, and an increasing proportion of the remaining breast cancers in both trial arms will have been detected in the later service screening.
Overdiagnosis was calculated as excess number of breast cancer cases in the mammography arm as compared with that in the control arm fifteen years after enrollment divided by number of screen detected breast cancers in the mammography arm. This gave 106 divided by 484, equal to 22%. We agree that overdiagnosis should be based on the excess number of breast cancers in the mammography arm as compared with the control arm, but to compensate for lead time, women should be followed for some time after end of screening, at least eight years according to our experiences from Denmark (6), in order to ensure that the entire compensatory dip is included. This is not the case in the Canadian data, not fifteen years after enrollment, neither 25 years after enrollment, as the women will gradually have been covered by service screening after end of the trial.
On this basis the firm conclusions from the authors of no effect of annual mammography beyond that of physical examination and usual care and of a 22% overdiagnosis seem overstated.
The number of breast cancer cases diagnosed during the five years of the trial has changed slightly across the reports from the Canadian National Breast Screening Study. In the first follow-up including observations to 1988/1990 627 breast cancer cases were reported from the mammography arm and 506 from the control arm (2-3). In data followed up to 1993, the numbers were 639 and 520 (4-5), respectively. In the present report, the numbers are 666 and 524, respectively (1). This might be due to better tracing in registers, but as the increase is 6% in the mammography arm and 4% in the control arm, it would be desirable, if the authors had commented on it.
Elsebeth Lynge
Sisse Njor
REFERENCES
1) Miller AB, Wall C, Baines CJ, Sun P, To T, Narod SA. Twenty five year follow-up for breast cancer incidence and mortality of the Canadian National Breast Screening Study: randomised screening trial. BMJ. 2014 Feb 11;348:g366. doi: 10.1136/bmj.g366.
2) Miller AB, Baines CJ, To T, Wall C.Canadian National Breast Screening Study: 2. Breast cancer detection and death rates among women aged 50 to 59 years. CMAJ. 1992 Nov 15;147(10):1477-88. Erratum in: Can Med Assoc J 1993 Mar 1;148(5):718.
3) Miller AB, Baines CJ, To T, Wall C. Canadian National Breast Screening Study: 1. Breast cancer detection and death rates among women aged 40 to 49 years. CMAJ. 1992 Nov 15;147(10):1459-76. Erratum in: Can Med Assoc J 1993 Mar 1;148(5):718.
4) Miller AB, To T, Baines CJ, Wall C. The Canadian National Breast Screening Study-1: breast cancer mortality after 11 to 16 years of follow-up. A randomized screening trial of mammography in women age 40 to 49 years. Ann Intern Med. 2002 Sep 3;137(5 Part 1):305-12.
5) Miller AB, To T, Baines CJ, Wall C. Canadian National Breast Screening Study-2: 13-year results of a randomized trial in women aged 50-59 years. J Natl Cancer Inst. 2000 Sep 20;92(18):1490-9.
6) Njor SH, Olsen AH, Blichert-Toft M, Schwartz W, Vejborg I, Lynge E. Overdiagnosis in screening mammography in Denmark: population based cohort study. BMJ. 2013 Feb 26;346:f1064. doi: 10.1136/bmj.f1064.
Competing interests: No competing interests
The national breast screening programmes around the world have provided us with a natural experiment of the greatest historical importance, not because of their success in reducing breast cancer mortality but because the reality failed to live up to expectations. The scientific explanation for these disappointing outcomes became clear following the recognition of the phenomenon of over-diagnosis of the disease [1, 2]. By way of illumination, let me propose that the pathological diagnosis of cancer at screening is seriously flawed because it is based on a syllogism (a syllogism is a logical argument in three propositions, two premises and a conclusion, the conclusion being specious). A simple example might be that people who die from meningitis, harbour meningococci in their nose. This does not mean that harbouring meningococci in the nose will lead to your death from meningitis; in fact about 20% of the population carry these bacteria in their nasal cavities [3].
Cancer was defined by its microscopic appearance about two hundred years ago. The 19th century saw the birth of scientific oncology, with the discovery and use of the modern microscope. Rudolf Virchow, often called the founder of cellular pathology, provided the scientific basis for the modern pathologic study of cancer [4]. As earlier generations had correlated the autopsy findings observed with the unaided eye with the clinical course of cancer one hundred years earlier [5], so Virchow correlated the microscopic pathology of the disease. However, the material he was studying came from the autopsy of patients dying from cancer. In the mid 19thC, pathological correlations were performed either on cadavers or on living subjects presenting with locally advanced or metastatic disease that, almost always, were predetermined to die in the absence of effective therapy. Since then, without pause for thought, the microscopic identification of cancer according to these classic criteria has been associated with the assumed prognosis of a fatal disease if left untreated. The syllogism at the heart of the diagnosis of cancer, therefore, runs like this; people frequently die from malignant disease, under the microscope this malignant disease has many histological features we will call “cancer”, ergo anything that looks like “cancer” under the microscope, will kill you. I would, therefore, like to reinforce the argument, that some of these earliest stages of “cancer”, if left unperturbed, would not progress to a disease with lethal potential. These pathological entities might have microscopic similarity to true cancers, but these appearances alone are insufficient to predict a life threatening disease.
If we stand back and take a broader look at nature, this shouldn’t be surprising. Conventional mathematical models of cancer growth are linear or logarithmic, in other words, completely predictable at the outset; promising transition from in-situ phases to early invasive and from early invasive to late invasive over time. These mathematical formulae may be appropriate for designing theme park rides, but cannot begin to explain the exquisite organisation of cell proliferation and the complex inter-relationships of cells of different progeny
Most natural biological mechanisms are non-linear or better described according to chaos theory [6]. The beauty of the tree in full leaf and the symmetry of a sprig of broccoli, reflect their fractal geometry that looks remarkably similar to the appearance of the mammary ducts and lobules under the microscope. The rate of growth and the development of the lung, along with the fingers and toes in the foetus, cannot be described in linear terms. Wound healing starts with the knife and ends when it needs to, although rarely wound healing carries on too long to leave an ugly keloid scar. Prolonged latency followed by regression should not be all that surprising. We accept the case for prostate cancer; men will die with prostate cancer in situ and not of prostate cancer. [7].
Further support for this contention comes from other sources. For example, there has been an epidemic of bilateral mastectomies in the USA following the uncontrolled proliferation of MRI scans in the routine work up of women presenting with a single focus of early breast cancer [8,9]. The MRI scan is guilty of unveiling not only latent foci of pseudo-cancers outside the index quadrant, but also latent foci harboured in the contralateral breast. This is heartbreaking when one considers all the work, over three decades, and all the volunteers in trials of breast conservation [10, 11]. We now know, with the utmost confidence, that breast-conserving surgery (BCS) is a safe alternative to more radical surgery, yet that hard won knowledge is brutally ignored when the surgeon is induced to treat the MRI image rather than the patient.
Next, it is worth noting that, contrary to all common sense predictions, the increased rate of detection of duct carcinoma in situ (DCIS) has lead to an increase in the mastectomy rate for the screened population [1, 2]. Up to 45% of screen detected cases of DCIS end up having mastectomy because of the multi-centricity of the disease [12]. Yet the paradox is that clinically detected multi-centric invasive breast cancer is relatively uncommon [13,14]. Surely that is proof enough that at least half of these foci of DCIS regress if left alone; which half, of course, remains the problem.
In conclusion, therefore, we can state, with a great deal of conviction, that a large proportion (in the order of 50%) of screen detected (pre-clinical) foci of breast cancer are not programmed to progress if left unperturbed. Whereas those cancers that slip through the screening net and announce themselves as “interval cancers”, are the ones programmed to kill if left untreated. These observations are of seismic importance, and could set the agenda for breast cancer research for the next decade. For a start, we should consider a trial of active surveillance versus conventional therapy for screen-detected cases of DCIS. Using this platform, we might then learn the nature of the clinical or biological characteristics of the disease that allow it to leave its dormant phase and enter the transition to early invasion. After all, this is good enough for men. [15]
If we choose to ignore these observations, either because they do not support our prejudices, conflict with our financial interests or for some sleazy political agenda, then we will have missed an opportunity of a lifetime, and that would be unforgivable. “There’s none so blind as them who will not see”.
References
[1] Jørgensen K J, Gøtzsche P C
Overdiagnosis in publicly organised mammography screening programmes: systematic review of incidence trends BMJ 2009;338-341
[2] Welch G H, Black W C Over-diagnosis in cancer J Natl Cancer Inst 2010;102:605-13
[3] http://www.meningococcal.org/
[4] Rudolf Ludwig Karl Virchow Cellular Pathology: As Based Upon Physiological and Pathological Histology.
Twenty Lectures Delivered in the Pathological Institute of Berlin during the months of February, March and April, 1858, Translated by Frank Chance, 1924, Robert M. De Witt publishers, New York
[5] Rokitansky. Karl. Freiherr von. (1804-1878)
A manual of pathological anatomy London, Printed for the Sydenham society (1849)
[6] Baum M, Chaplain M, Anderson A, Douek M,
Vaidya J S Does breast cancer exist in a state of chaos?Eur J Cancer 1999; 35: 886–91
[7] Donovan J, Mills N, Smith M, et al
Quality improvement report: Improving design and conduct of randomised trials by embedding them in qualitative research: ProtecT study. BMJ 2002;325:766-70
[8] King T A, Sakr R, Patil S, Gurevich I, Stempel M,
Sampson M, Morrow M Clinical management factors contribute to the decision for contralateral prophylactic mastectomy.J Clin Oncol. 2011 Jun 1;29(16):2158-64
[9] Itakura K, Lessing J, Sakata T, Heinzerling A, Vriens, E, Wisner D, Alvarado M, Esserman L, Ewing C, Hylton N, Hwang E S
The impact of preoperative magnetic resonance imaging on surgical treatment and outcomes for ductal carcinoma in situ Clin Breast Cancer. 2011 Mar;11(1):33-8
[10] Fisher B, Anderson S, Bryant J, et al Twenty-year follow-up of a randomized trial comparing total mastectomy, lumpectomy, and lumpectomy plus irradiation for the treatment of invasive breast cancer N Engl J Med, 2002;347:1233-41
[11] Veronesi U, Cascinelli N, Mariani L, et al
Twenty-year follow-up of a randomized study comparing breast-conserving surgery with radical mastectomy for earlybreast cancer N Engl J Med 2002;347:1227-32
[12] Schwartz G F, Patchefsky A S, Feig S A, Shaber G S,
Schwartz A B Multi-centricity of non-palpable breast cancer Cancer, 45:2913-2916, 1980
[13] Gump F E, Habif D V, Logergo P, Shikora F, Kister S,
Estabrook A The extent and distribution of cancer in breasts with palpable
primary tumours Ann Surg 204: 384-388, 1986
[14] Holland R, Veling S H, Mravunac M, Hendriks J H
Histologic multifocality of Tis, T1-2 breast carcinomas:implications for clinical trials of breast-conserving surgery Cancer 1985;56:979-90
[15] Cooperberg M R, Cowan J E, Hilton J F, Reese A C, Zaid H B, Porten S P, Shinohara K, Meng M V, Greene K L, Carroll P R Outcomes of active surveillance for men with intermediate risk prostate cancer J Clin Oncol. 2011 Jan 10;29(2):228-34
298
Competing interests: No competing interests
Please explain me why this study is accepted in the BMJ. After all those years technology is changed. Not only technically but also the interpretation, position of the breast etc. I do not understand that average tumor is 1.9 cm. I do not understand this type randomisation. I do not understand why the researchers focussed on 40-50 knowing the evidence for screening is between 50-69. I do not understand.I do not understand etc. If there are a lot of MD's and patient who "do not understand": please retract this article.
Competing interests: No competing interests
Dr Tabar writes eloquently to defend his study and condemn the Canadian study. However, comparisons are illuminating.
• Trial A has informed consent and individual randomization. Trial B has no informed consent and uses cluster randomization.
• Trial A maintains consistent numbers of participants and deaths over years of follow-up. Trial B does not.
• Trial A has 100% compliance at first screen; not so for Trial B.
• Trial A used 2-view mammography, Trial B single-view mammography.
• Trial A screens women every 12 months. Trial B screens every 33 months.
• Trial A has an external audit of mammography based on stratified sampling. Trial B does not.
• Trial A has a higher cancer detection rate with smaller tumor size at first screen than Trial B.
• Trial A has external pathology reviews to confirm all biopsies performed. Trial B does not.
• Trial A has an external death review panel to determine cause of death in all cases of deaths in participants known to have breast cancer during the trial or suspected of having breast cancer after linkage with a national data base. Not so for Trial B.
Trial A is the Canadian trial, and B the Swedish Two-county. The fact that the Canadian trial yielded a null result has led to fervent criticism by screening zealots with loud voices. (1)
Dr. Tabar writes: It is a well-known fact that the Canadian trials were not population-based. Instead, they used motivated volunteers who represented only 26% to 27% of the invited population. It is particularly important to note that the Canadian trials could not evaluate the independent impact of mammography because of the confounding effect of physical examination. For these reasons, the Canadian trials were excluded from the meta-analysis of the impact of early detection using mammography on mortality from breast cancer by the World Health Organization's International Agency for Research on Cancer (IARC) workshop in 2002…
False – the reason it was excluded was because it was comparing single to double modality screening unlike the other trials.
Dr. Tabar writes that radiologists W. Logan and S. Feig resigned from the trial – and indeed they did because they misunderstood their roles – they thought they were conducting the trial and Dr. Logan wanted mammograms exported to her in the US. They represent well the loud voices in the screening controversy - those described in the New England Medical Journal as being in clear conflict of interest (2). These loud voices are many, as is illustrated in some of the Rapid Responses.
As for my curtailing Dr. Tabar’s condemnation of CNBSS mammograms, I have no recollection of doing so but if I did it would have been well warranted. Canadian radiologists did not need to hear Dr. Tabar condemn 50 mammograms. They would have understood his “overwhelmingly negative” ratings after 15. We were not in the business of humiliating credentialed professionals from the best health care institutions across Canada. Dr. Tabar had his own agenda. Amazing to think he actually believes he could have shut down a Canadian study.
The Canadian study encompassed more than 50 center-years of operation – and Dr. Tabar proposes that one coordinator in a short span of time was able to corrupt the results from 90,000 women by improper randomization of five, ten, even a hundred? of her so-called friends. Absurd. As for the randomization sheets categorized for age by quinquennium, it is hardly surprising that a number of women might first have been entered on sheets designated for women of a different age group. The mistake discovered, the woman would be re-entered on another page. Does Dr. Tabar lives in a world where mistakes are never made - 90,000 correct entries with nary an error?
If Dr. Tabar were truly interested in the well-being of women and the spectrum of consequences of screening, he might have mentioned over-diagnosis and increased risk of mastectomy. How many of those committed to breast screening realize that since 1987 more women die of lung cancer than breast cancer in the US?
A Google search for Mammography Education Incorporated provides interesting reading.
1. Baines CJ. Rational and Irrational Issues in Breast Cancer Screening. Cancers 2011,3,252-266;doi:10.3390/cancers3010252
2. Quanstrum KH, Hayward RA. Lessons from the mammography wars. N Engl J Med. 2010; 363:1076-70
Competing interests: No competing interests
I did not see any reference to breast size, shape or "lumpy, bumpy breast" as possible confounding issues that nullified by these data.
In addition, can these data be extrapolated to all postmenopausal women?
Competing interests: No competing interests
Re: Twenty five year follow-up for breast cancer incidence and mortality of the Canadian National Breast Screening Study: randomised screening trial
We are pleased to see our trial continues to incite comment. In the additional response provided below, we concentrate on the points made by some of the authors which we feel indicates they did not fully understand our paper nor the study design.
In spite of rather extensive prior e-mail discussions, Mr Leonid Winestein (RR February 21) appears still not to understand the analysis of our study. The computation of over-diagnosis we made was based upon comparisons of cumulative incidence between our two study arms. It will be clear that the vast majority of women who were diagnosed with breast cancer in our study did not die of their disease. The computations of over-diagnosis relate to this group, so his comments on death from breast cancer are irrelevant in this context.
Elsebeth Lynge (RR February 19) also comments on our computations of over-diagnosis. She states that women should be followed for 8 years after cessation of screening to compensate for lead time. In fact we did, the excess at 15 years is 10 years after cessation of screening in our trial. Although she is correct that some women may have been screened subsequently in provincial programmes, the breast cancer cumulative incidence curves largely remained parallel after the end of the lead time period indicating that such screening was not differential between the two arms. Her concern about lead time leads her to criticise our presentation of mortality results at 5 years, but in fact, recognising this issue, we presented in the text results at 6 and 7 years, and they were almost identical to the 5-year results. In previous reports on the 11-16 years follow-up of the study we extended the time interval to 9 years with a similar result. In the present report we did not feel that it was necessary to report the findings at 9 years, as it is now recognized that previous estimates of the lead time gained by mammography have been too long. When corrections are applied for over-diagnosis, lead time approximates to 1-2 years.
John D Keen (RR February 17) requests estimates of over-diagnosis including in situ lesions. For DCIS, 93 were diagnosed in the mammography arm and 20 in the control arm, the corresponding numbers for LCIS were 27 and 8. A paper regarding the outcome of the patients with these diagnoses will shortly appear in the International Journal of Cancer (To T, Wall C, Baines CJ, Miller AB: Is Carcinoma in Situ a Precursor Lesion of Invasive Breast Cancer?). If these lesions are included in estimates of over-diagnosis on screening the percentage becomes 33% (198/604), and the equivalent in terms of mammography alone detected lesions 60% (198/332).
Peter Sasieni (RR February 17) sees little difference statistically between our results and those of others. However that ignores the difference between our trial and others in the control group, and the availability of adjuvant therapy in our trial, whereas it was not in the others, also ignored in the Marmot evaluation. Further, the fact that there was an initial excess mortality from breast cancer in the mammography compared to the control arm does not argue against over-diagnosis, but is compatible with theories over the potential immune balance between primary tumours and metastases, first identified by the Fishers and re-enforced by Retsky et al (1).
1. Retsky M, Demicheli R, Hrushesky W. Premenopausal status accelerates relapse in node positive breast cancer: hypothesis links angiogenesis, screening controversy. Breast Cancer Res Treat. 2001;65:217-24. [PMID: 11336243]
Competing interests: No competing interests