Breast Cancer Risk Assessment Tools: A Clear Comparison of Models, Strengths, and Clinical Uses

A single risk number cannot capture the full landscape of breast cancer risk tools. Multiple validated models exist, each built for different clinical questions, data inputs, and populations. The expanded inventory below organizes every major tool currently available, including absolute-risk calculators, mutation-probability models, imaging-based tools, population-specific models, research frameworks, and referral screens. A comparison table then outlines the primary absolute-risk models used in everyday practice.

Categorized Inventory of Currently Available Tools

Absolute-risk models for invasive breast cancer

These estimate the probability of developing breast cancer over defined intervals and guide screening intensity, chemoprevention eligibility, and imaging pathways.

1. Gail Model / Breast Cancer Risk Assessment Tool (BCRAT). Developed from the Breast Cancer Detection Demonstration Project and updated with race- and ethnicity-specific data by the National Cancer Institute. Inputs include age, age at menarche, age at first live birth, number of prior biopsies (including atypical hyperplasia), number of first-degree relatives with breast cancer, and race/ethnicity. Outputs are 5-year and lifetime (to age 90) invasive risk. Available at bcrisktool.cancer.gov. Suitable for average-risk chemoprevention decisions. Not appropriate for women with prior invasive cancer, ductal carcinoma in situ, lobular carcinoma in situ, known high-risk mutations, or chest radiation.^1,2

2. Tyrer-Cuzick Model / IBIS (current versions include v8). Incorporates age, reproductive and hormonal factors, body-mass index, height, history of benign disease or atypia or lobular carcinoma in situ, detailed multi-generational family history (breast and ovarian cancer, ages at diagnosis, male breast cancer), Ashkenazi ancestry, BRCA status, and (in later versions) mammographic density. Outputs 10-year and lifetime risk plus mutation-carrier probability. Preferred when family history is complex or for younger women. Can overestimate risk in certain groups.³

3. Breast Cancer Surveillance Consortium (BCSC) Risk Calculator. Built from large U.S. mammography registries. Inputs include age, race/ethnicity, first-degree family history, history of benign breast disease or biopsies, and BI-RADS breast density. Outputs 5-year and 10-year invasive risk. Strength is explicit inclusion of density; requires a density assessment.⁴

4. BOADICEA / CanRisk. Multifactorial model using detailed family history, pathology, non-genetic risk factors, known pathogenic variants, polygenic risk scores, and density in recent versions. Estimates breast and ovarian cancer risks plus mutation-carrier probabilities. Implemented in the free CanRisk tool at canrisk.org. Flexible for high-risk and general populations.⁵

5. Claus Model. Relies on detailed family history of breast (and sometimes ovarian) cancer, including ages at diagnosis in first- and second-degree relatives. Primarily provides lifetime risk estimates. Useful for moderate family-history assessment. Does not incorporate most non-hereditary risk factors.⁴

6. Black Women's Health Study (BWHS) Breast Cancer Risk Calculator. Developed and validated specifically with data from Black women. Inputs include age, body-mass index, reproductive factors, biopsy history, family history, and related variables. Outputs 5-year and lifetime invasive risk. Addresses underrepresentation of Black women in earlier models; recent validation demonstrates better calibration than the Gail or Tyrer-Cuzick models in this population.^6,7

7. Rosner-Colditz Model. Derived from the Nurses' Health Study and focused on hormonal and lifestyle factors (age at menarche, menopause, and births; postmenopausal hormone use; body-mass index; height; alcohol; benign disease; family history). Provides absolute risk estimates and incorporates modifiable factors well. Less commonly available as a simple public online calculator; updates have added adolescent somatotype and predicted mammographic density.⁸

Mutation-carrier probability tools

These estimate the chance of carrying a high-risk genetic mutation rather than absolute cancer risk and are used mainly in genetic counseling.

8. BRCAPRO. Bayesian model focused on family history of breast and ovarian cancer (including ages and unaffected relatives) to estimate BRCA1 or BRCA2 mutation probability and associated cancer risks. Available via specialized software packages such as BayesMendel. Limited incorporation of non-genetic factors.⁴

9. Myriad Models (including Myriad II tables). Empirical models based on genetic testing laboratory data. Primarily estimate BRCA mutation probability using personal and family history (including Ashkenazi ancestry). Used mainly in genetic counseling settings.⁹

10. Penn II Model. Empirical model that predicts the probability of carrying a BRCA1 or BRCA2 mutation based on family history features (breast and ovarian cancers, ages, bilaterality, other associated cancers, ancestry). Available online. Focused solely on mutation-carrier probability.¹⁰

Imaging-derived risk tools

11. Clairity Breast. FDA-authorized (de novo authorization June 2025) artificial-intelligence tool that analyzes pixel-level features from a routine bilateral screening mammogram to generate a 5-year breast cancer risk score. Does not require questionnaire data. First commercially available AI-only imaging-based risk tool; referenced in updated National Comprehensive Cancer Network guidelines (2026) with risk thresholds (for example, 1.7% or higher as one criterion for increased risk).¹¹

Research or specialized population macros

12. Specialized National Cancer Institute and related models / macros. Includes the CARE model (for African American women), Asian/Pacific Islander American women risk assessment macro, Hispanic-American risk assessment macro, and versions incorporating mammographic density. These are research-oriented SAS macros or R packages tailored to specific populations.²

13. iCARE framework. Flexible research and software tool (R package) for building and validating absolute risk models that can combine classical risk factors, polygenic risk scores, and other data. Used in large multi-cohort validation studies. Not a single fixed public calculator but a widely used methodological platform.¹²

Family-history referral and scoring tools

These primarily identify people who may benefit from genetic counseling or further risk assessment rather than providing precise absolute risk percentages.

14. Additional family-history referral / scoring tools. Tools such as the Manchester Scoring System, FHS-7 (Seven-Question Family History Screening), Ontario Family History Assessment Tool, Pedigree Assessment Tool, and Referral Screening Tool. The Manchester Scoring System assigns points based on cancer type, age at diagnosis, and pathology to estimate the likelihood of a BRCA1 or BRCA2 pathogenic variant and remains widely used for testing thresholds.^13,14

Side-by-Side Comparison of Core Absolute-Risk Models

Sources for the table content include direct model documentation, head-to-head validation studies, and clinical reviews.^3,4,6,7

Model	Primary Inputs	Family History Detail	Density Included	Main Outputs	Strengths	Limitations	Primary Clinical Use	MRI Eligibility Support
Gail / BCRAT	Age, race/ethnicity, menarche, first live birth, biopsies, atypical hyperplasia	First-degree only	No (standard)	5-year and lifetime invasive risk	Simple, free online, validated across U.S. groups, established chemoprevention threshold	Limited family history; excludes prior cancer, known mutations, age under 35	Average-risk primary care and chemoprevention	No
Tyrer-Cuzick / IBIS	Age, BMI, height, reproductive/hormonal factors, biopsies including atypia/LCIS, Ashkenazi ancestry, genetic results	First- to third-degree; ages, bilateral, ovarian, male breast cancer	Yes (v8+)	10-year and lifetime risk; carrier probability	Most comprehensive hybrid model; strong for complex pedigrees	Can overestimate in some groups; more data entry required	High-risk clinics, MRI decisions, genetic triage	Yes
BCSC	Age, race/ethnicity, first-degree family history, biopsy history	First-degree only	Yes (core)	5-year and 10-year invasive risk	Built from large diverse U.S. screening cohorts; density is explicit	Narrower age range; limited family history	Density-aware imaging risk stratification	Limited
BOADICEA / CanRisk	Family history, high- and moderate-penetrance genes, polygenic risk scores, lifestyle factors	Full pedigree; multiple genes	Yes (recent versions)	Lifetime and age-specific breast/ovarian risk; carrier probabilities	Most flexible multifactorial tool; integrates genetics, density, and PRS	Requires detailed pedigree and online access	Hereditary clinics and comprehensive counseling	Yes
Claus	Number and ages of affected relatives	First- and second-degree	No	Lifetime risk tables	Simple, historically used for MRI thresholds	No non-familial factors; limited modern validation	Quick family-history estimate	Yes (historical)
BWHS	Age, BMI, reproductive factors, biopsy, family history, breastfeeding, oophorectomy	First-degree emphasis	No	5-year and lifetime invasive risk	Developed and validated specifically in Black women; better calibration than Gail or IBIS in that population	Less widely known outside specialized settings	Risk assessment for Black women, especially younger ages	Case-by-case

Key Practical Distinctions

Empirical models such as Gail and BCSC emphasize personal reproductive and biopsy history and suit average-risk women. Hybrid and genetic models (Tyrer-Cuzick, BOADICEA/CanRisk, Claus) capture multigenerational patterns and support lifetime-risk thresholds for supplemental magnetic resonance imaging under American Cancer Society and National Comprehensive Cancer Network guidelines. Density-inclusive models improve discrimination because dense tissue both elevates risk and reduces mammographic sensitivity. The BWHS calculator fills a documented performance gap for Black women. Clairity Breast supplies a questionnaire-free 5-year risk score derived solely from mammogram pixels and is now referenced in National Comprehensive Cancer Network 2026 guidance.^15,11

Why a Woman's Risk Estimate Changes

Absolute risk rises with age as baseline incidence increases. New diagnoses in relatives alter family-history inputs. Biopsy results showing atypical hyperplasia or lobular carcinoma in situ multiply risk in several models. Density can shift with age, weight, or hormone use. Reproductive milestones, hormone therapy changes, body-mass-index shifts, and new genetic or polygenic results all move the estimate. Guidelines therefore recommend periodic reassessment.¹⁵

Operational Implications and Leadership Question

Match the tool to the clinical question, available data, and patient ancestry. Use Gail or BCSC for short-term average-risk stratification and chemoprevention. Use Tyrer-Cuzick or CanRisk when family history is complex or lifetime risk for magnetic resonance imaging eligibility is required. Prefer the BWHS calculator when assessing Black women. Consider density-inclusive or image-based tools when mammography data are already present. Document the specific model, exact inputs, and absolute risk output so that imaging pathways and navigation remain consistent over time.

Leadership question: How will your organization standardize model selection by clinical scenario and patient ancestry, capture inputs and outputs in the record, and trigger scheduled re-assessment so that imaging intensity tracks each woman's changing absolute risk rather than the first number calculated years earlier?

All tools remain accessible through the cited public or institutional platforms. Results are probabilistic estimates that require clinical interpretation in context.

---

References

1. Gail MH, Brinton LA, Byar DP, et al. Projecting individualized probabilities of developing breast cancer for white females who are being examined annually. J Natl Cancer Inst. 1989;81(24):1879-1886. doi:10.1093/jnci/81.24.1879
2. National Cancer Institute. About the Breast Cancer Risk Assessment Tool. bcrisktool.cancer.gov/about.html
3. Hillas J, Hans M, Park KU. A review of current breast cancer risk calculators and their recent advances. Curr Breast Cancer Rep. 2025;17:15. doi:10.1007/s12609-025-00581-6
4. Vachon CM, et al. Performance of breast cancer risk-assessment models in a large mammography cohort. J Natl Cancer Inst. 2020. View source
5. CanRisk documentation. University of Cambridge. canrisk.org
6. Palmer JR, et al. A validated risk prediction model for breast cancer in US Black women. J Clin Oncol. 2021. doi:10.1200/JCO.21.01236
7. Performance of three breast cancer risk assessment tools in US Black women. Breast Cancer Res. 2026. View source
8. Rice MS, Tworoger SS, Hankinson SE, et al. Breast cancer risk prediction: an update to the Rosner-Colditz breast cancer incidence model. Breast Cancer Res Treat. 2017;166:227-240. doi:10.1007/s10549-017-4391-5
9. Lindor NM, Johnson KJ, Harvey H, et al. Predicting BRCA1 and BRCA2 gene mutation carriers: comparison of PENN II model to previous study. Familial Cancer. 2010;9:495-502. doi:10.1007/s10689-010-9348-3
10. Lindor NM, Johnson KJ, Harvey H, et al. Predicting BRCA1 and BRCA2 gene mutation carriers: comparison of PENN II model to previous study. Familial Cancer. 2010;9:495-502. doi:10.1007/s10689-010-9348-3
11. OncLive. Clairity Breast is added to NCCN guidelines for breast cancer screening and diagnosis. 2026. View source
12. Choudhury PP, Maas P, Wilcox A, et al. iCARE: An R package to build, validate, and apply absolute risk models. PLoS One. 2020;15(2):e0228198. doi:10.1371/journal.pone.0228198
13. Evans DGR, Eccles DM, Rahman N, et al. A new scoring system for the chances of identifying a BRCA1/2 mutation outperforms existing models including BRCAPRO. J Med Genet. 2004;41(6):474-480. doi:10.1136/jmg.2003.017996
14. Evans DG, Harkness EF, Plaskocinska I, et al. Pathology update to the Manchester Scoring System based on testing in over 4000 families. J Med Genet. 2017;54(10):674-681. doi:10.1136/jmedgenet-2017-104584
15. American Cancer Society. Recommendations for the early detection of breast cancer. View source
16. National Comprehensive Cancer Network. Breast Cancer Screening and Diagnosis guidelines (2023-2026 versions). nccn.org

---

About the Author

Richard D. Lippert Jr. is the founder and CEO of Mammologix LLC. He has more than thirty years in breast imaging program operations, is clinically trained in radiologic technology and mammography, and has tracked FDA MQSA National Statistics monthly since December 2002.