Critical analysis of the new INCa information booklet

In August 2017 the National Cancer Institute (INCa) published a new information booklet on breast cancer screening entitled: "Breast cancer screening; be informed and decide". (1)
INCa presents this booklet on its website as follows: "This booklet is intended for women aged 50 who are invited for the first time to have a breast cancer screening test. It is intended to be joint to the invitation letter. By answering the main questions about breast cancer and screening tests, the booklet enables women to make an informed decision in response to the invitation. It can also be distributed at information meetings. »
In September 2017 we conducted a critical analysis of this booklet.
Three years later, in January 2021 we can make the observation that nothing has been changed in the content of this booklet.

________________________________________

September 17, 2017

Cécile Bour, MD
Marc Gourmelon, MD
Philippe Nicot, MD

The National Cancer Institute (INCa) has published a new information booklet on breast cancer screening entitled: "Breast Cancer Screening; Be Informed and Decide". (1)

The 2016 citizens' consultation had clearly pointed out the miscommunication and misinformation committed in the past by the institute, on page 85 of the full report available at the following link  http://www.concertation-depistage.fr/wp-content/uploads/2016/10/depistage-cancer-sein-rapport-concertation-sept-2016.pdf

"...the public consultation on organized breast cancer screening...highlighted the very limited information available to women on this screening. And for the information specifically intended for them, we note that this information does not include any mention of the controversy of which it has been the subject for several years, nor of the existence of real uncertainty regarding the benefit/risk ratio, nor of its limits...".

The criticisms on the communication provided by the INCa, fragmented, sometimes contradictory, tendentious, unclear on the over-diagnosis, continue from page 87 to 95.

Two studies published in the British Medical Journal (one Australian, the other by researchers from the Nordic Cochrane Collaboration) listed 17 key points that should be addressed in brochures on breast cancer screening, for a scientific and objective information. (2,3)

We used these 17 scientific criteria to evaluate the INCA booklet on breast cancer screening by mammography and to determine whether, in accordance with the requests of the citizen's consultation, the Institute has met the expectations for improving the quality of information.

For details of these two studies, see Dr. Nicot's article in Voix Médicales: http://www.voixmedicales.fr/2011/11/14/depistage-organise-du-cancer-du-sein-information-ou-communication/

ARTICLE PLAN
PART I: Analysis according to the 17 criteria
PART II: Other issues raised

PART I: The 17 key points to assess the quality of the information

These 17 key points are :

    1. Lifetime risk of developing breast cancer.

    2. Lifetime risk of dying from breast cancer.

    3. Survival after breast cancer.

    4. Relative risk reduction.

    5. Absolute risk reduction.

    6. The number of women who need to be screened to avoid death from breast cancer.

    7. Proportion of women who will be recalled.

    8. Proportion of breast cancer detected (sensitivity).

    9. Proportion of women without breast cancer who are screened positive (lack of specificity/false positive).

    10. Proportion of women with a positive test who have cancer (positive predictive value).

    11. Reduction of the relative risk of total mortality.

    12. Carcinoma in situ.

    13. Over-diagnosis.

    14. Over-treatment.

    15. Effect of screening on the number of mastectomies.

    16. Relative risk of radiotherapy, psychological distress in relation to false positives.

    17. Pain during mammography.

These 17 key points were reviewed in the new brochure issued by INCa in order to assess its informative value as objectively as possible.

1. Lifetime risk of developing breast cancer.

The brochure answers this first point: 3 out of 100 women aged 50 will develop breast cancer within 10 years. The different risk factors are detailed on page 4.

2. Lifetime risk of dying from breast cancer.

The question which, indeed, is of most interest to women is: what is the real risk of dying from breast cancer? "We can measure the risk of dying from breast cancer in France, in 2010 this risk was 4.1%...., of which 1.9% was between 50 and 79 years of age". (4)

This data is not present in the booklet of the INCa.

3. Survival after breast cancer

The brochure talks about it and highlights the 99% survival rate for cancer detected "early". But without explaining what this means.

" Survival" is rather a measure of a cancer lifetime

It gives an optical illusion: by anticipating the date of "birth" of the cancer detected during a screening, one has the impression of a longer life after cancer, whereas the lifetime of the person is in no way changed. Let's take a life expectancy for a given woman of 73 years, if this woman is diagnosed with cancer by screening at the age of 67, she will enter the 5-year survival statistics. If the diagnosis is made later, around the age of 72, when a symptom occurs for example, then this same woman will not be included in the survival statistics. Survival at 5 years leads to an illusion of success, whereas the life expectancy of women in France has not changed at all since screening is performed.

4. 5. 11. Relative risk reduction. Absolute risk reduction.

On page 8 it is written: "International studies estimate that these programs prevent between 15% and 21% of breast cancer deaths. "

This is an indication of relative risk only. The editors are ignoring the claim of women citizens to stop being fooled by numbers that do not mean what they seem to say. The 20% fewer deaths does not mean that 20 less women out of 100 will die of breast cancer, if they get screened.

This 20% corresponds only to a relative risk reduction between two comparative groups of women.

In fact, according to a projection made by the Cochrane Collective (6) based on several studies, 4 out of 2,000 women screened over 10 years, die of breast cancer; 5 out of a 2000 women group not screened over the same period of time, will die of breast cancer, the change from 5 to 4 mathematically corresponds to a 20% reduction in mortality, but in absolute terms only one woman's death will be avoided (absolute risk of 0.1% or 0.05%).

Actually, this corresponds to an absolute risk reduction of 0.05% (1 woman in 2000) to 0.1% (1 woman in 1000) at the end of 10 to 25 years of screening, according to the estimates used (American journals, Prescrire magazine, US TaskForce). (5)

This is why the citizens had requested during the citizen consultation a presentation based on real data, and not on percentages that embellish the situation.

6. The number of women needed to be screened to prevent a breast cancer death

We search unsuccessfully for this indication. According to the Cochrane Collaboration synthesis, for example, 2000 women must be screened over 10 years in order to find a benefit of a life saved through screening. (6)

7. Proportion of women who will be recalled.

One of the causes of recalling women is the examination judged "technically insufficient" by the second reader who receives the images validated by the first reader, but which he considers insufficiently well done. This causes additional anxiety for the patients, and of course, additional irradiation, due to the sometimes very formalistic questions of incidence.

The second cause of recall and not the least are false positives, i.e. a positive screening test for a woman who does not present a cancerous lesion (which overlaps with criterion n°9).

8. Proportion of breast cancer detected (sensitivity).

9. Proportion of women without breast cancer who have a positive detection (lack of specificity/false positives).

10. Proportion of women with a positive test who have cancer (positive predictive value).

The brochure announces well in " To know" section on page 11: "The false positive - In most cases, it turns out that the abnormalities discovered are benign and that it is not cancer. This is referred to as a false positive", but without giving the proportion, the criterion is therefore not met.

We provide you with the estimate according to Prescrire magazine, Cochrane and US Task Force: out of 1,000 to 2,000 women screened over a period of 10 to 25 years, depending on the synthesis, between 200 and 1,000 women will receive a false alarm, leading to at least 200 unnecessary biopsies. (5)

This precise data is not included in the brochure.

We recommend that our readers refer to the explanations provided on our site: https://www.cancer-rose.fr/cancer-du-sein-un-peu-de-technique/.

12. Carcinoma in situ.

This point is not addressed at all.

In the cancer statistics posted on INCa site, only invasive cancers are considered. However, in situ cancers, which are the main cause of over-diagnosis, are estimated to account for 15 to 20% of breast cancers. They are all the more frequent, as participation in screening is important. Most of the cancers qualified as "early detected" are in situ cancers for which it is known that treatment does not change much the prognosis, since most of them would not have developed.

When a non-fatal disease is treated unnecessarily, it is normal that survival (see criterion 3) is high.

The booklet does not address this issue.

13. Over-diagnosis.

14. Over-treatment.

Over-diagnosis is addressed on page 7, but only briefly.

The problem of over-treatment is not mentioned.

15. The effect of screening on the number of mastectomies.

This issue is not addressed. Large scale international studies (Harding (7)) as well as meta-analyses of the Prescrire review (8, 9 ,10) and the Cochrane Collaboration (6) ) show a significant increase in mastectomies, radiotherapy and chemotherapy since screening, while the detected lesions are more and more smaller in size.

16. The relative risk of radiotherapy, psychological distress in relation to false positives.

This point is not addressed.

17. Pain during mammography.

This point is discussed on page 9.

Let's now do the math: of the 17 key points for informed and objective information, only 5 are addressed in the INCa brochure (no. 1, no. 3 although imperfectly, no. 4 but without explanation, no. 13, no. 17).

PART II: Other points of concern

A-Studies

The INCa booklet mentions the scientific controversy over screening, which it refers to as being based on "few studies and difficult to compare".

According to INCa, the debate is not about whether lives are saved, but about how many lives are saved.

However, since the 2000s, numerous studies have been regularly published in the major independent international journals (New England Journal of Medicine, British Medical Journal, JAMA, Lancet, etc.).

The studies cannot be described as "few" when, on the contrary, they abound.

For example, we will cite only three of the major studies published in recent years: Bleyer Welch, 2012 (11), A. Miller, 2014 (12), Harding, 2015 (7).

The first shows the rise in breast cancer incidence since the introduction of screening without a decrease in the most severe forms. The second shows the absence of significant impact of screening on mortality, revealing identical survival rates regardless of the stage of cancer in two groups of women compared, one group screened and the other not, with a follow-up at 25 years.

Finally, the third shows that a 10% increase in participation in screening leads to a 16% excess of diagnosis without any impact on mortality, but with a surge in less aggressive forms, stability of advanced forms, and a significant increase in the most severe treatments.

It is not clear which studies enable INCa to assert that the debate would not be about saving lives because it is not possible to identify any significant decrease in mortality.

B-The radiation delivered

Concerning X-rays during a mammographic examination, for which the dose required after 50 years would be lower (page 12), the problem is not the low dose theoretically delivered, but the reality of the repetition of doses, no matter how low they are.

Moreover, quite frequently, several incidences per breast are necessary, their dose being added to those of the routine examination.

The real problem, however, is radiation for therapeutic purposes in the case of over-treatment, which is itself a consequence of over-diagnosis by mammography, with real and serious consequences on health (coronary heart disease, radiation-induced cancers of the oesophagus, skin, mediastinum and lung).

C-Natural history of the disease

On page 3, we find :

"In general, the earlier breast cancers are detected, the greater the chances of a cure.
"

There is confusion between the smallness of the cancer detected and its early detection.

A small cancer is not synonymous with early detection, it can be small and be present for many years without ever manifesting itself. A large lesion is not necessarily late either, it may have developed in a very short period of time, which is often the case with what are called interval cancers. These rapid cancers are often the most aggressive, occurring after a normal mammogram and before the next one scheduled two years later. Screening cannot anticipate them.

The problem is the insufficient knowledge of the natural history of cancer, which the booklet itself admits on page 2, where it says that there are several types of cancer evolution and kinetics. This would justify treating all cancers, according to the booklet (page 7) which threatens that a woman who is not screened regularly, will inevitably have a more advanced cancer, a heavier treatment and a reduced chance of recovery.

"If you develop breast cancer and you have not been screened regularly, the cancer will be diagnosed at a more advanced stage. This will reduce the chances of a cure and have a greater impact on your quality of life. Indeed, heavy treatment, sometimes dangerous, and its consequences are significantly greater in the presence of cancer diagnosed at an advanced stage".

There are two problems with this statement. It is needlessly alarmist and is not based on any studies, the real question being "What are the benefits of not getting tested? "(The brochure asking this one, page 7: "What are the risks of not being screened for breast cancer?"). The assertion that the burden of treatment is reduced thanks to "early" detection is contradicted by the Harding study (7) in the United States, by the meta-analyses of Cochrane (6) and the Prescrire Review (8 9 10), which report an increase in chemo- and radiotherapy, with consequences in the form of thrombo-embolic, cardiac and radiation-induced neoplasia complications not even mentioned in the booklet. These studies and meta-analyses also objectivize the increase in mastectomies, which logically should drastically decrease since the lesions discovered are smaller in size. Hence the second subsequent question: "What is the situation in France with regard to the promised decrease in total mastectomies, and what is the status of the evaluation?"

D-Over-diagnosis

Overdiagnosis is quickly addressed in the booklet (page 7), we will not argue about its extent, as the latest studies put forward the figure of 50%. However, justifying the treatment of all cancers with the same aggressiveness, because we do not know how to apprehend, it is a rather lapidary approach to the problem.

The third source indicated by INCa at the bottom of the document refers to the "Handbooks of Cancer Prevention", which is presented here: https://www.youtube.com/watch?v=4ejAL_pzLGI

The CIRC (International Agency for Research on Cancer, Lyon) itself (from about 7:54 minutes) opts for a red coding ('missing or incomplete data') of over-diagnosis in the context of breast cancer screening.

IN CONCLUSION

The information quality score is therefore 5/17, i.e. 6/20.

Click here to enlarge:

In view of all these elements, this new INCa brochure does not seem to meet the expectations of the citizen consultation.

P.S. In the meantime, at the beginning of April 2018, the INCa has added references at the bottom of the document, but the sources added are only reports or legislative texts or sources from the INCa itself...

References

http://www.e-cancer.fr/Expertises-et-publications/Catalogue-des-publications/Livret-d-information-sur-le-depistage-organise-du-cancer-du-sein

Slaytor EK, Ward JE How risks of breast cancer and the benefits of screening are communicated to women: analysis of 58 pamphlets.http://www.bmj.com/content/bmj/317/7153/263.full.pdf

Jørgensen KJ, Gøtzsche PC. Presentation on websites of possible benefits and harms from screening for breast cancer: cross sectional study. bmj.com 2004;328:148. Sur : http://www.bmj.com/content/328/7432/148.full.pdf+html

http://www.unaformec-idf.org/documents/lagny/KseinHill2014.pdf

https://cancer-rose.fr/en/2020/12/15/the-over-diagnosis-in-a-graph-and-a-table/

https://www.cochrane.org/CD001877/BREASTCA_screening-for-breast-cancer-with-mammography

http://jamanetwork.com/journals/jamainternalmedicine/fullarticle/2363025

8 Dépistage des cancers du sein par mammographie Deuxième partie Comparaisons non randomisées : résultats voisins de ceux des essais randomisés. Rev Prescrire. 2014 Nov;34(373):842–6.

9 Dépistage des cancers du sein par mammographie Première partie Essais randomisés : diminution de la mortalité par cancer du sein d’ampleur incertaine, au mieux modeste. Rev Prescrire. 2014 Nov;34(373):837–41.

10 Dépistage des cancers du sein par mammographies Troisième partie Diagnostics par excès : effets indésirables insidieux du dépistage. Rev Prescrire. 35(376):111–8.

11 http://www.nejm.org/doi/full/10.1056/NEJMoa1206809#t=article

12 http://www.bmj.com/content/348/bmj.g366

Cancer Rose est un collectif de professionnels de la santé, rassemblés en association. Cancer Rose fonctionne sans publicité, sans conflit d’intérêt, sans subvention. Merci de soutenir notre action sur HelloAsso.


Cancer Rose is a French non-profit organization of health care professionals. Cancer Rose performs its activity without advertising, conflict of interest, subsidies. Thank you to support our activity on HelloAsso.

Leave a Reply