Risk Assessment

Jump To

Risk Model  |  Genetics  |  Risk reduction  |  Treatment options   |  Imaging   |  Surgery   |  Chemoprevention

For Professionals

Risk Assessment

1/8 American woman develop breast cancer. 

Most women diagnosed with breast cancer have no significant risk factors.

There are many risk factors for breast cancer.

The most common risk factors increase risk slightly. It is unlikely that women with these risk factors alone would be considered “high risk” as pertains to our discussion:

Increasing age.

Race: Caucasian women are at the highest risk, followed by African American women. 

Lifestyle: elevated BMI, moderate to heavy alcohol use, and the use of combination estrogen-progesterone hormone replacement therapy (HRT).  

Prolonged estrogen exposure due to early menarche (younger than 12) or late menopause (older than 55).  

Increased mammographic breast density. 

Age of first pregnancy: Women who are nulliparous or whose first pregnancy occurred at age 30 or later have double the risk of women whose first pregnancy occurred prior to age 20. 

Breastfeeding is associated with a small reduction in risk. 

Other risk factors are less common but may significantly increase a woman’s likelihood of developing breast cancer (ie, a strong family history, a known genetic mutation, or a history of atypia, lobular carcinoma in situ, or of mantle radiation), and women with these risk factors frequently are categorized as “high risk.”

Risk Models

Risk assessment for breast cancer has been important when ordering screening breast MRI. In fact, the main purpose of most of these models was to obtain insurance authorization for this screening modality. However, things are changing. Risk assessment is critical in lieu of new breast density legislation. Breast density is not a risk factor in and of itself, however, combined with risk assessment is critical in identifying individuals who are at higher risk, thus requiring high risk screening.

Patients with a known genetic mutation (see Hereditary Breast Cancers, above) or those who received mantle radiation at a young age are clearly at a high lifetime risk of developing breast cancer, but these patients are the minority. (Ref. 27-28)

For patients with a family history of breast cancer or with atypia or LCIS, risk assessment models can better quantify risk.

Risk assessment models incorporate various patient factors and family history and provide an estimation of a patient’s risk of developing breast cancer.

None of these models is perfect.

None of these models incorporates a history of mantle radiation.

Gail Model 

The most commonly used risk assessment model.

Originally described in 1989 and revised in 1999.

Incorporates patient age, age of menarche, age of first pregnancy, history of prior breast biopsies, previous diagnosis of atypia, and number of first-degree relatives with breast cancer.

Age at cancer diagnosis is not considered.

Affected second- and third-degree relatives are not considered.

Not validated for women under 35 or for women with LCIS.

Women with a 5-year risk of >1.7% are considered high risk (1.7% is the average risk of a 60-year-old woman being diagnosed with breast cancer). 

Claus Model 

Risk based solely on family history (including non-first-degree relatives) of breast and ovarian cancer and incorporates the age of diagnosis in those family members.

No personal or hormonal factors are included.


Estimates the likelihood of a patient having a BRCA mutation and provides an estimate of that patient’s risk of breast or ovarian cancer.

Incorporates the number of affected and unaffected family members and includes male breast cancer.

Non-BRCA genetic mutations are not addressed.

No personal or hormonal factors are included.

Tyrer-Cuzick Model (IBIS Model [International Breast Cancer Intervention Study]) 

Estimates a patient’s risk of breast cancer and of having a BRCA mutation or of having another autosomal dominant breast cancer gene.

Incorporates personal and hormonal factors as well as extended family history of both breast and ovarian cancer.

BOADICEA Model (Breast and Ovarian Analysis of Disease Incidence and Carrier Estimation Algorithm)

Predicts BRCA mutation likelihood and the risk of breast or ovarian cancer based on family history.

Family history includes breast, ovarian, prostate, and pancreatic cancers.

Two studies directly compared the accuracy of these multiple models:

Tyrer-Cuzick was found to be more accurate, based on ROC curve, than BRCAPRO and Gail Models

These latter 2 models were found to underestimate breast cancer risk  

Tyrer-Cuzick and BOADICEA were found to be more accurate than Gail or Claus Models, as Gail and Claus underestimated breast cancer risk 

Women estimated to have a lifetime risk of breast cancer of at least 20% are considered high risk.

Risk Reduction


Risk-reducing strategies such as the use of chemoprophylactic agents should be discussed with women whose estimated risk of breast cancer is 20% or greater.

Tamoxifen and raloxifene are selective estrogen receptor modulators (SERMs) and are FDA-approved as risk-reducing agents.

Tamoxifen (Ref. 42) 

NSABP P-1 documented the efficacy of tamoxifen as a risk-reducing agent.

Tamoxifen (20 mg daily for 5 years) decreased the risk of developing invasive breast cancer by 49% and reduced the risk of developing noninvasive cancer by 50% in high-risk women.

Preventative effect of tamoxifen was limited to estrogen receptor (ER) positive tumors.

Associated with an increased risk of endometrial cancer and of thromboembolic events, most significantly in postmenopausal women. 

Raloxifene  (Ref. 43-44)

STAR Trial (NSABP P-2) compared tamoxifen and raloxifene.

Risk reduction for invasive cancer was similar between tamoxifen and raloxifene.

There was a lower risk of noninvasive cancer in the tamoxifen group, but this was not significant.


Lower risk of thromboembolic events

Not associated with the increase in cataracts seen with tamoxifen

Raloxifene positively affects bone density

Only FDA-approved for use in postmenopausal women and is dosed at 60 mg daily

No ideal length of treatment established 

There is little or no evidence regarding the utility of chemoprophylaxis in women treated with mantle radiation or in women under 35.

There is limited evidence regarding the utility of chemoprophylaxis in women with BRCA1 and BRCA2 mutations.

Contraindications to these drugs include history of deep venous thrombosis, transient ischemic attack, thrombotic stroke, hypercoagulability, and pregnancy.

Common side effects of the SERMs include vasomotor symptoms, such as hot flashes.


An aromatase inhibitor currently used in the treatment of invasive breast cancer

Not yet FDA-approved as a risk-reducing drug

Recently published data on the MAP .3 Trial demonstrated a 65% risk reduction at 3 years follow-up in invasive breast cancer in high-risk postmenopausal women

Additional studies are ongoing 

Prophylactic Risk-Reducing Surgery

Prophylactic mastectomy reduces risk by at least 90% in high-risk women, including those with BRCA mutations.

Prophylactic mastectomy is reasonable in women without a cancer diagnosis but with:

A known BRCA1 or BRCA2 mutation or with another known breast cancer susceptibility gene

A strong family history (ie, multiple family members affected, multiple generations affected, male breast cancer, premenopausal breast cancer, bilateral breast cancer)

A history of atypia or LCIS

Prophylactic oophorectomy in women with BRCA1 and BRCA 2 mutations reduces breast cancer risk by half.



Mutations associated with breast, ovarian, and pancreatic cancers

Estimated lifetime risk of breast cancer 39-87%


Mutations associated with breast, ovarian, pancreatic, and prostate cancers

Estimated lifetime risk of breast cancer 31-84%

Other less common hereditary breast cancers

PTEN gene mutation (Cowden syndrome) 

25-50% estimated lifetime breast cancer risk

Associated with cerebellar ataxia, craniomegaly, facial trichilemmomas, oral mucosal papillomas, hamartomas, thyroid cancer, endometrial carcinoma, and other genitourinary cancers 

P53 mutation (Li-Fraumeni syndrome) 

90% estimated lifetime breast cancer risk, 56% by age 45

Associated with soft tissue sarcoma, osteosarcoma, brain tumor, adrenocortical carcinoma, Wilms tumor, phyllodes tumor, neuroblastoma, pancreatic cancer, and leukemia

STK11 mutation (Peutz-Jeghers syndrome) 

31% risk of breast cancer by age 60

Associated with small intestine cancer, colorectal carcinoma, pancreatic carcinoma, and ovarian tumors 

CDH1/E Cadherin mutation (Hereditary Diffuse Gastric Cancer)

39% lifetime risk of breast cancer

These are lobular breast cancers and are associated with diffuse gastric cancer.


For women with a 5-year estimated risk of >1.7% or a lifetime estimated risk of >20%, the following surveillance is recommended: 

Annual screening mammography

Clinical breast exam every 6–12 months

Consider the addition of annual screening MR

In our practice, when we include annual MRI into a patient’s screening regimen, we stagger the annual mammogram and the annual MR by 6 months so the patient is undergoing some form of imaging every 6 months.

For women with a known genetic predisposition to breast cancer (eg, a BRCA mutation), the following is recommended:

Annual clinical breast exam for women under 25

Starting at age 25, annual mammography and annual MR are recommended in addition to a clinical exam every 6-12 months

For women without a known genetic predisposition but with a strong family history of breast cancer, annual mammography, annual MR, and annual or semiannual clinical exam should begin 10 years prior to the youngest age of breast cancer diagnosis in the family. 

While breast self-exam has not been shown to affect breast cancer-related mortality, women are encouraged to be familiar with their bodies and to have any changes evaluated.









Tomosynthesis/3D Mammography





Ultrasound guided core

Stereotactic Core

MR guided core

Benign Biopsies

Atypical Ductal Hyperplasia (ADH)


  • Atypical ductal hyperplasia is a proliferative lesion of the terminal duct lobular unit characterized by both cytologic and architectural abnormalities that are similar to ductal carcinoma in situ (DCIS); however, the extent of the lesion is smaller (only 1 or 2 ducts and <2 mm) such that it does not meet criteria for DCIS.
  • Therefore, core needle biopsy fragments that appear to be ADH could actually represent a small portion of a DCIS lesion. Studies of surgical excision of ADH lesions diagnosed by core needle biopsy confirm that cancer is found in approximately 10-20% of cases.  
  • Surgical excision is therefore recommended for core needle biopsy lesions demonstrating ADH.

Atypical Lobular Hyperplasia (ALH)


  • Atypical lobular hyperplasia is a proliferative epithelial lesion that enlarges the terminal duct lobular unit and is characterized by a uniform confluent population of epithelial cells that distends the acini. ALH is similar in appearance to lobular carcinoma in situ but lesser in extent.  
  • Atypical lobular hyperplasia on core needle biopsy has a more controversial upgrade rate to cancer at excision, ranging from <5% to 50%.  Because of the lack of consistent upgrade rates for ALH, performing surgical excision in a routine fashion is also controversial.  Surgical excision is generally considered a safe approach so that a cancer is not missed. However, ALH is often an incidental histologic finding in addition to the originally targeted lesion that is proven to be benign. For this reason, some surgeons excise all ALH lesions, while others excise only lesions of higher suspicion based upon other coexisting high-risk pathology or clinical and imaging features, such as those cases of ALH associated with ADH, mass lesions, or larger areas of calcifications that were not completely removed.  

Papillary Lesions


  • Papillary lesions of the breast can range from completely benign (intraductal papilloma) to malignant (intraductal papillary carcinoma or invasive papillary carcinoma) and can be difficult to distinguish based on tissue fragments obtained with core needle biopsy.
  • Papillary lesions may also be heterogeneous, such that the entire lesion must be evaluated in order to rule out atypical hyperplasia or cancer.
  • Upgrade rates to cancer are in the range of ~10-35%. When papillary lesions with atypia are excluded the upgrade rates are lower, with some recent reports supporting observation in selected papillary lesions without atypia.  

Flat Epithelial Atypia


  • Flat epithelial atypia (FEA) is a recently characterized columnar cell lesion of the breast with epithelial hyperplasia containing cytologic atypia. This lesion is rare, occurring in <5% of breast core needle biopsies.  
  • Due to its recent recognition and low frequency, data on upgrade rates is limited. However, reports suggest that cancer is found in approximately 10-15% of FEA cases at surgical excision.

Other Lesions with Atypia


  • On occasion, the pathologist may identify features of atypia (either cytologic or architectural) on core needle biopsy that do not meet the established criteria for the specific lesions described above. In these cases there may be a comment of an atypical proliferation. Although a second pathology review may be helpful, in general this is a situation where a surgical biopsy is indicated to provide a larger tissue sample for definitive diagnosis.

Rationale for Surgical Excision


  • When atypical ductal hyperplasia and papillary lesions are identified by core needle biopsy, they are associated with a significant rate of upgrade to cancer if the biopsy site is surgically excised.  
  • This is primarily due to (1) sampling error of the core needle biopsy technique (removing only a portion of the lesion) and (2) the fact that these lesions are histologically difficult to discriminate from cancer when only fragments of the lesion are present in the biopsy tissue. 
  • The goal of surgical excision is to remove the biopsy site and the original mammographic lesion that prompted the core needle biopsy in order to obtain a definitive diagnosis.

Importance of Concordance


  • Concordance of core needle biopsy results refers to the combined assessment of clinical, imaging, and pathologic findings that are all internally consistent. Radiologic-pathologic concordance is mandated in current practice utilizing core needle breast biopsy.  
  • Core needle biopsies are subject to error based upon sample volume limitations and inaccurate targeting. It is critical for the surgeon as part of the multidisciplinary breast care team to evaluate these possible sources of error by assessing the concordance of clinical, imaging, and pathologic findings associated with a breast lesion. 
  • To assess the targeting of the lesion, the breast surgeon should personally review the original diagnostic mammogram demonstrating the lesion, as well as the postbiopsy mammogram to assess the position of the biopsy marker and how well it marks the location of the original lesion.
  • Surgical excision is strongly recommended when:


      1. There is concern that the core needle biopsy did not sample the target lesion seen on imaging.
      2. Core needle biopsy histology demonstrates atypia or a papillary lesion in the presence of a palpable or imaging mass lesion.
      3. Findings are judged to be discordant.

Considerations for Avoiding Excision of Atypia on Core Needle Biopsy


  • In general, surgical excision is favored because the morbidity of surgical breast biopsy is judged to be lower than the cost of a missed cancer.
  • If surgical excision is not performed, short-term mammographic follow-up is necessary.
  • Clinical risk factors of the patient should be considered; eg, omitting surgical excision would not be recommended in a patient with a BRCA mutation or strong family history of breast cancer.
  • Some favorable subgroups have been identified in whom close surveillance rather than excision might be considered, with informed consent that there is a small ~5% or less chance of a missed cancer.


        1. ADH: Lesions that might be observed: no mass lesion, and all or >95% of calcifications removed by core biopsy.  Caution is advised because other recent studies have been unable to identify low-risk subgroups.  
        2. ALH: The primary criterion to omit excision is that the ALH finding is incidental, ie, that the targeted lesion was explained by another benign pathology that does not require excision and is concordant with the imaging appearance. One recent study found that incidental ALH on core biopsy with fewer than 3 foci of ALH has a very low likelihood of upgrade. 
        3. Papillary lesions: Recent data suggests that papillary lesions without atypia may be followed with imaging surveillance, especially if imaging demonstrates that the lesion has been completely removed by needle biopsy. However, caution is advised for this approach, considering multiple previous reports demonstrating upgrade rates of 20-30%. 

Long-Term Risk of Breast Cancer Risk Associated With Atypical Hyperplasia


  • Women with atypical hyperplasia (ADH or ALH) have an approximate fourfold relative increase in risk for developing a subsequent breast cancer, with absolute cumulative risk estimated at ~20% at 20 years.  
  • In the NSABP P-1 trial, tamoxifen reduced breast cancer risk by 86% in women with atypical hyperplasia, and raloxifene reduces risk of invasive breast cancer similar to tamoxifen. For this reason, women with atypical hyperplasia should be counseled about the risks and benefits of tamoxifen and raloxifene.  (


In Situ








Skin sparing

Nipple sparing


Oncoplastic Surgery








Side Effects