Ineffectiveness of mammography screening to detect the most serious cancers
Summary by Cécile Bour, MD
October 1, 2020
https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2770959
Published on September, 25, 2020.
Saroj Niraula, MD, MSc1,2; Natalie Biswanger, BSc3; PingZhao Hu, PhD4; et alPascal Lambert, MSc2; Kathleen Decker, PhD2,5
- 1. Section of Medical Oncology and hematology, University of Manitoba, Winnipeg, Manitoba, Canada
- 2. Research Institute of Oncology and Hematology, CancerCare Manitoba, Winnipeg, Manitoba, Canada
- 3. Cancer Screening program, CancerCare Manitoba, Winnipeg, Manitoba, Canada
- 4. Department of Biochemistry and Medical Genetics, University of Manitoba, Winnipeg, Manitoba, Canada
- 5. Department of Community Health Sciences, University of Manitoba, Winnipeg, Manitoba, Canada
Study objectives
« Evaluate the differences and similarities in characteristics and outcomes of breast cancers detected by mammographic screening compared to those detected between screening mammograms (interval cancers) in women participating in a population-based screening program »
Results of study
« In this cohort study of 69 025 women, interval breast cancers accounted for one-fourth of breast cancers in routinely screened women, were 6 times more likely to be grade III, and had 3.5 times increased hazards of breast cancer death compared with screen-detected cancers. »
Meaning
« Heterogeneity in breast cancer defies assumptions necessary for screening mammography in its current form to be maximally effective; strategies beyond routine screening mammography are needed to prevent, detect, and avert deaths from the more lethal interval breast cancers. »
Explanation :
Breast cancer does not follow the linear and mechanistic pattern assumed.
Natural history of cancer
The theory that cancer can be treated because it has been diagnosed when very small, seems to be intuitive, flattering, yet contrary to observation (clinical cases, autopsy studies). In the case of breast cancer screening, we also have to deal with the true belief which is underpinned by frequently repeated mantra such as "cancer can knock on every door," "the smaller the better," "prevention is cure." Is it true?
These clichés are based on a linear and mechanistic theory of natural history of cancer. Cancer is believed to evolve in an ineluctable way, according to a set pattern. A cancer cell, then a nodule, then a large nodule, then a local invasion, followed by metastases and inevitable death.
But reality is a lot more complex than that.
Small does not mean caught on time, it can simply be a silent cancer, little or never progressive, even regressive, that would have been diagnosed during screening but would never have been killing the woman.
Or, on the contrary, at the time of diagnosis it may already be metastatic, while small or sometimes even occult.
Large does not mean being caught too late, but simply the case of a rapidly growing cancer that, due to its fast development, would be large at the time of diagnosis. In general, it is true that these lesions are on average more aggressive, but this is not absolute. In older women who give up to consult, large cancers may have significant local consequences, such as skin erosions or severe retractions, but without having spread to the distance.
We see these cases in consultation every day, which we consider "paradoxical".
Not all the cancers evolve and most of them do not become metastatic, they can stagnate, regress, grow so slowly that the patient will die of something else before.
As we can observe, the natural history of breast cancer does not follow the pre-established theory, nor the intellectual model that corresponds to what theorists have opportunistically imagined in order to fit in with their simplistic view. This study is useful to understand this topic : https://cancer-rose.fr/en/2020/12/17/are-small-breast-cancers-good-because-they-are-small-or-small-because-they-are-good/
The authors come here to this conclusion: due to intrinsically slow growth, many of the small tumors detected excessively by screening have a very good prognosis, which means that they are not expected to become large tumors and are inherently favorable. They are the ones that cause overdiagnosis, which results directly from the activity of screening. They will not develop enough to become dangerous.
In contrast, large tumors, responsible for deaths and most often with immediate poor prognosis, escape unfortunately to mammographic detection, due to too rapid kinetic growth.
A previous similar study:
Sarauj Niraula et al. cohort research remembers us a very important and comprehensive study of Pr. Autier. Mammographic screening: a major issue in medicine
https://www.sciencedirect.com/science/article/pii/S0959804917313850
One chapter in this major analysis deals with the specificity of cancers found through mammographic screening, which are less severe and with better prognosis cancers; they are those "selected" through screening, half of which would be overdiagnostics, meaning needless diagnostics that would never have killed the woman.
Mammography indicates for example a high sensitivity for ductal carcinoma in situ cancers, and a relatively low sensitivity for certain aggressive cancers such as 'triple negative' breast cancer.
Mammography basically does not detect lobular carcinoma in situ or invasive cancers that represent 8-14% of all breast cancers. Lobular carcinomas penetrate the tissues without forming masses, making it impossible to identify them by mammography.
Invasive cancers detected by mammography have the clinical and pathological characteristics of less aggressive tumors compared to interval cancers, i.e. those that progress rapidly between two mammograms, escape detection and have aggressive characteristics.
In addition, after analyzing the characteristics of these tumors and the expansion of the disease at the time of diagnosis, the risk of dying from a screened breast cancer is lower than the risk of dying from an interval cancer.
The authors of this study also reported that the interval cancers were similar to breast cancers diagnosed in the absence of screening.
So, if interval cancers are similar to cancers diagnosed in the absence of any screening, and if cancers screened have on average a better prognosis than interval cancers, it logically follows that a proportion of cancers screened are non-lethal cancers that would never have been symptomatic during a woman's lifetime.
These lesions have the microscopic morphological characteristics of cancer, but would have remained asymptomatic throughout the woman's life if the screening had not occurred.
The authors add: "Cancer overdiagnosis refers to cancer excess in women invited to screen divided by the total number of cancers which would be diagnosed in the absence of screening (on a population of the same profile, with the same age group, without screening)."
"If overdiagnosis is calculated using the number of screened cancers as a denominator, then for 100 screened breast cancers, 30 to 50 will be overdiagnosed."
Our conclusion
Breast cancer is clearly shown to be a very heterogeneous disease; indolent cancers with the probability of better healing outcomes are easily detected by mammography screening, abusively increasing the overall incidence of breast cancer, making believe that there are always more, but it is this mechanism of public health that produces them.
And they also give the illusion that the cure rate is improving because their host will never have been killed by all these cancers.
On the other hand, many of the aggressive and lethal types of breast cancers remain unnoticed or develop in the mammography interval.
Other strategies, particularly a deeper understanding of the natural history of cancer, which includes referring to fundamental studies on cancer growth models, are needed to improve the rate of breast cancer death and overall population mortality.