Question

Which risk factors exist for developing breast cancer?

Recommendation

Several factors for developing breast cancer are shown in the Evidence part of this module.

Conclusions

It has been shown that the following risk factors give an RR ≥ 4 for developing breast cancer:

  • carrier of mutations in genes with high penetrance, including BRCA1 or 2;
  • very high risk due to family history, without a proven mutation in BRCA1 or 2;
  • chest radiation therapy before the age of 40;
  • atypical benign breast laesions consisting of atypical (ductal or lobular) hyperplasia, flat epithelial atypia, lobular carcinoma in situ, papillary laesions and complex sclerosing laesions (radial scars);
  • history of ductal carcinoma in situ;
  • history of breast cancer;
  • high density mammogram at advanced age.

Level 2: B Dumitrescu 2005, Morrow 1999, McPherson 2000, Santen 2005

Literature summary

There are various known risk factors that play a role in breast cancer. For a summary of the literature search, based on reviews, see appendices on Oncoline. The table below gives a global overview of the risk factors named in these reviews. The decision was made to state the risks in terms of relative risks (RR). It is not always possible to convert RR to lifetime risk (LTR), since the information required for populations is not always known. For the Netherlands, an RR of 1 corresponds to an LTR of 10%.

 

Risk factors for developing breast cancer

Factor

Relative risk

Reference

Older age (over age 45 versus under age 25)

< 10

Dumitrescu 2005

McPherson 2000

Mutations in BRCA1/2

6 – 8

Dumitrescu 2005

McPherson 2000

Geographic region (North American and Northern Europe versus the Far East, Africa and South America)

5 - 10

Dumitrescu 2005

 

High density mammogram

4 - 6

Boyd 2010

Atypical benign breast laesions:

Atypical (ductal or lobular) hyperplasia, flat epithelial atypia, lobular carcinoma in situ, papillary laesions and complex sclerosing laesions (radial scars)

4 - 5

Dumitrescu 2005

McPherson 2000

Morrow 1999

Santen 2005

Prior history of radiation; chest and/or axillary radiation, e.g. due to Hodgkin's lymphoma before age 40

3 - 20

De Bruin 2009

Van Leeuwen 2003

Aleman 2003

Breast carcinoma or DCIS in medical history

2 - 4

Morrow 1999

Late age at the time of first child, over age 35 vs. before age 20

2

Dumitrescu 2005

McPherson 2000

High postmenopausal bone density

2 - 3.5

Dumitrescu 2005

Diethylstillbestrol (DES) use during pregnancy

2

McPherson 2000

Late menopause, after age 54

≤ 2

Dumitrescu 2005

McPherson 2000

Morrow 1999

Nulliparity

< 2

Dumitrescu 2005

McPherson 2000

Morrow 1999

Hormone replacement therapy (HRT) use for over 10 years

1.4 - 3

Dumitrescu 2005

Alcohol intake, risk is dose-dependent, 2-5 units per day vs. no alcohol intake

1.2 - 1.5

Brennan SF 2010

Key 2006

Li 2010

Oral contraception        Recent use

                                   Past use

1.2 -2.4

1.0-1.2

Dumitrescu 2005

Cibula 2010

Mutations in other highly penetrant genes; p53, PTEN

1 - 6

Dumitrescu 2005

Early menarche, before age 11

1 - 3

Dumitrescu 2005

McPherson 2000

Morrow 1999

Physical exercise 5x per week vs. inactivity

0.85

Patterson   2010

Bernstein 2009

In vitro fertilisation

Not clearly elevated

Salhab 2005

Dor 2002

Zreik 2010

Obesity

            Premenopausal, body mass index > 35

            Postmenopausal, body mass index > 35

 

0.7

2

McPherson 2000

 

General population

People with two risk factors – age over 50 and of the female sex – are screened under the national screening for breast cancer. Regarding geographic region, note that for people from low-risk areas (such as Asia), the difference decreases the longer they live in a high risk area (such as North America).

 

Genetic risk factors

The gene mutations in the BRCA1 and 2 genes are the most significant genetic risk factors, with an RR of 6-8. In addition, there are several rare tumour syndromes caused by highly penetrant genes including breast cancer. The most significant are Li Fraumeni (P53), Cowden syndrome (PTEN), Peutz-Jeghers (STK11) and hereditary diffuse gastric cancer (CDH1). For more information on these syndromes, go to http://www.vkgn.org.

If one has a first-degree relative with breast cancer, the RR is 1 to 4, depending on one's age and other family history. In women with breast cancer in the family, the RR depends greatly on the number of relatives, whether it is first- or second-degree kinship, and at what age the breast cancer occurred. If there is only third-degree kinship with breast cancer, the RR is not elevated enough to justify screening outside the national breast cancer screening programme. See the decision tree after section 1.3.4.

 

Hormonal risk factors

Risk factors with an RR of 2 or higher are present when the woman is over age 35 at the time of having her first child, and in postmenopausal women with high bone density. Since estrogen can contribute to high bone density, estrogen use as a part of hormone replacement therapy can have a direct relationship as a risk factor in developing breast cancer. As a result, it may not be the high bone density, but estrogen use that may be the risk factor that gives an RR of 2 or higher.

DES use during pregnancy gives an RR of 2, as does postmenopausal overweight.

An RR of up to 2 has been published for menopause after age 54. Menarche before the age of 11 gives an RR of 1-3. Long-term hormone replacement therapy gives an RR of 1.4 to 3. Use of oral contraceptives gives an RR of less than 2 in most studies. It is notable that there is currently no obvious evidence that in vitro fertilisation increases the risk of breast cancer.

 

Many other risk factors are indeed associated with a statistically significant increase in risk in large populations, but have little practical significance for an individual woman.

An exception are women who underwent chest or axillary radiation before the age of 40, usually as part of treatment for Hodgkin's Lymphoma.

There are no prospective studies on this group. In a retrospective study of 91 patients with an average age of 42, treated for Hodgkin's, 10 cancers were found in a period of 10 years; 4 by MRI only, 3 with mammography in addition to MRI, and 3 only with mammography (based on microcalcifications) [Sung, 2011].

Based on a risk estimate, beginning 8 years after the radiation therapy these women are offered the same screening program as gene mutation carriers. See section 1.3.2, Table 2.

Another exception is women who receive radiation in the breast region for other forms of childhood cancer, including Wilms tumour, sarcoma, neuroblastoma or non-Hodgkin lymphoma. For information on the definition of risk groups and the associated screening policy for breast cancer after treating childhood cancer, see the guideline "Follow-up after childhood cancer," sections 1, 2 en 3 (www.skion.nl).

For women who underwent chest radiation therapy after the age of 40, screening may be started 10 years after radiation therapy. This means that the national breast cancer screening programme is adequate for these women.

Considerations

In the texts below, a very high risk is roughly equivalent to RR 6-8, a high risk RR 3-4, a moderately increased risk RR 2-3 and a slightly increased risk RR <2.

There is no consensus on how to define the degree of increased risk. Different risk factors are usually studied in different populations, so adding them together is not possible. However, there are models that combine some risk factors, such as menarche, age at the time of first child, and first-degree relative with breast cancer [Gail, 1989; Tyrer, 2004].

The epidemiologically proven relationship between density of glandular tissue and an increased risk of breast cancer applies to both premenopausal and postmenopausal populations [MacCormack, 2006]. It seems paradoxical that the percentage of glandular tissue reduces with age, while the cancer incidence increases. But this paradox can be explained: it is mainly a question of exposure to hormones, growth factors and effects of menarche, pregnancy and menopause on glandular tissue. Dense glandular tissue is also associated with atypical benign breast laesions. The density of the breast tissue has a hereditary component.

Since evidence of the relationship between dense glandular tissue and breast cancer has mainly been found in screening populations, no recommendations can be made for other screening modalities [Boyd, 2010].

The increased incidence of breast cancer in general and the high frequency of mild risk factors, such as low number of pregnancies and late age at first child, increase the demand for screening outside the national breast cancer screening programme. This calls for a good information campaign. If all women with mild risk factors would go to a hospital radiology department outside the national breast cancer screening centre, this would heavily overcrowd these departments. Furthermore, it is questionable whether those departments are adequately equipped for this screening role, and whether this could be in conflict with the national breast cancer screening Act [Wet op het bevolkingsonderzoek (WBO)]

The following points are important in the information for women who are worried about their risk of breast cancer: most women will not get breast cancer. Most of those who do get breast cancer have no family history of it. For most women older age is the main risk factor for getting breast cancer.

Authorization date and validity

Last review : 13-02-2012

Last authorization : 13-02-2012

Initiative and authorization

Initiative : Nationaal Borstkanker Overleg Nederland

Authorized by:
  • Nederlandse Internisten Vereniging
  • Nederlandse Vereniging voor Heelkunde
  • Nederlandse Vereniging voor Pathologie
  • Nederlandse Vereniging voor Radiologie
  • Nederlandse Vereniging voor Radiotherapie en Oncologie

Method of development

Evidence based