Description of breast cancer risk program

The program assumes that there is a gene predisposing to breast cancer in addition to the BRCA1/2 genes. The woman's family history is used to calculate the likelihood of her carrying an adverse gene, which in turn affects her likelihood of developing breast cancer. The risks of developing breast cancer for the general population were taken from data on the first breast cancer diagnosis (ICD-10 code C50) in Thames Cancer Registry area (UK) between 2005-2009. The risk from family history (caused by the adverse genes) is modelled to fit the results in "Familial Breast and Ovarian Cancer: A Swedish Population-based Register Study, Anderson H et al., American Journal of Epidemiology 2000, 152: 1154-1163".

The risk from other classical factors including age at first child and benign disease are combined with familial risk.

The latest version of the model (v8) incorporates mammographic density.

Contact Details

Prof. Jack Cuzick
Centre for Cancer Prevention,
Wolfson Institute of Preventive Medicine,
Charterhouse Square,
London
EC1M 6BQ

email: riskevaluator@ems-trials.org

The program begins by opening the introduction screen. If you want to go directly to the evaluation screen the next time the program is run then you should tick the ‘Start from evaluation screen’ box.

The top part of the evaluation screen deals with personal information about the woman - her age at menarche, parity etc. The bottom part of the screen deals with her family history. The picture in the form gives the family history in the form of a pedigree diagram.

When entering details about a family member, if she developed breast cancer then the age at breast cancer should be put in the relevant age box, otherwise her current age should be used (or the age at which she died if applicable). If she developed ovarian cancer then a box is activated which asks for details at which age it occurred. Similarly if she developed bilateral breast cancer then a box is activated which asks for the age at which breast cancer developed in the second breast. If age is not known then ‘?’ should be entered in the age box. If it is not known whether a relative developed breast cancer then ‘u’ should be typed in the age box to represent no information. If the woman has an affected niece in the family then the affected niece button should be pressed to enter the details.

The effects of weight were only assumed to affect risk once the menopause is reached. If the height is not known then the BMI cannot be calculated and so weight is ignored. The risk from height is also assumed to be independent from the BMI. If you prefer to use imperial measurements for height and weight then the imperial radio button should be clicked.

Typing '?' represents missing data. If nothing is known about parity then the unknown radio button should be pressed. Similarly if nothing is known about the woman’s menopause the no information radio button should be pressed. If the woman has had atypical hyperplasia, LCIS or some other benign breast disease then the appropriate box should be ticked.

If genetic testing has taken place within the family, then the ‘Genetic Testing’ button can be pressed to give a form on which to enter the results. Finally to get back to the introduction screen the ‘Show start up screen’ button should be pressed.

The tools menu gives you options to choose the number of years to risk, whether to allow for competing mortality risks, and to choose population rates (UK, Sweden or Slovenia).

NEW v8! Information on mammographic density is given in the model under the Help menu. If entering mammographic density, then it is also very important to enter height and weight to obtain the most accurate risk assessment for an individual woman.

More documentation may be downloaded from the following links.

Original article on model (Stat Med)
Description of changes to v8 (DI Europe)
Amir et al.pdf
Program details
Risk program details
Risk program input file format (v6-8)

For further information on mammographic density please follow the following links.

• areyoudense.org (general information).
• densebreast-info.org (general information).
• Volpara® Volumetric Density (commercial software to calculate mammographic density that may be entered into v8)
• BI-RADS® ATLAS* Density Breast Composition Measure (radiological interpretation of mammographic density that may be entered into v8).
• Visual Assessment Score (radiological score of mammographic density that may be entered into v8)
  • Brentnall, A. R., Harkness, E. F., Astley, S. M., Donnelly, L. S., Stavrinos, P., Sampson, S., Fox, L., Sergeant, J. C., Harvie, M. N., Wilson, M., Beetles, U., Gadde, S., Lim, Y., Jain, A., Bundred, S., Barr, N., Reece, V., Howell, A., Cuzick, J., Evans, D. G., Dec. 2015. Mammographic density adds accuracy to both the Tyrer-Cuzick and gail breast cancer risk models in a prospective UK screening cohort. Breast Cancer Research 17 (1), 147+. URL
  • Sergeant, J. C., Wilson, M., Barr, N., Beetles, U., Boggis, C., Bundred, S., Bydder, M., Gadde, S., Hurley, E., Jain, A., Lim, Y., Lord, L., Reece, V., Evans, D. G., Howell, A., Astley, S. M., 2013. PB.17: Inter-observer agreement in visual analogue scale assessment of percentage breast density. Breast Cancer Research 15 (Suppl 1), P17+. URL
  • Sperrin, M., Bardwell, L., Sergeant, J. C., Astley, S., Buchan, I., Nov. 2013. Correcting for rater bias in scores on a continuous scale, with application to breast density. Statistics in medicine 32 (26), 4666-4678. URL
  • Harkness, E. F., Sergeant, J. C., Wilson, M., Beetles, U., Gadde, S., Lim, Y. Y., Howell, A., Evans, D. G., Astley, S. M., 2016. Should We Adjust Visually Assessed Mammographic Density for Observer Variability? Springer International Publishing, Cham, pp. 540-547.URL

*BI-RADS® ATLAS is a registered trademark of American College of Radiology. All rights reserved. The developer of this product is independently owned and operated, and is not an affiliate of the American College of Radiology. The American College of Radiology is not responsible for the contents or operation of this product or its associated software, and expressly disclaims any and all warranties and liabilities, expressed or implied, in connection therewith.

Changes to version 8

Main changes

• Mammographic density added as a risk factor
• Mammographic density appears as white (radiopaque) areas on a mammogram. Dense breasts have more fibroglandular tissue and less fat than non-dense breasts.
• Three measures of mammographic density may be used in v8. These are
  • Volpara® Volumetric Density
  • Volpara® Volumetric Density is a fully-automated method to estimate percentage volumetric density that requires commercial software and digital mammography DICOM files ("FOR PROCESSING" type)
  • Visual Assessment Scale (VAS) Density
  • Assessed visually using a standard visual analogue scale from 0% to 100%. Requires training.
  • BI-RADS® ATLAS* Density Breast Composition Measures
  • The categories are defined:
  • a. The breasts are almost entirely fatty
  • b. There are scattered areas of fibroglandular density
  • c. The breasts are hetereogeneously dense, which may obscure small masses
  • d. The breasts are extremly dense, which lowers the sensitivity of mammography
  • *BI-RADS® ATLAS is a registered trademark of American College of Radiology. All rights reserved.
• Change to atypical hyperplasia model for women with substantial family history
• Risk given is based on the maximum 10y risk from atypical hyperplasia, or other factors than atypical hyperplasia.
• Definition of unknown benign disease updated
• `Unknown' benign disease now refers to when a biopsy has been taken, but the benign disease classification is unknown
• Option for polygenic SNP score added
• This treats a polygenic relative risk score (relative to the population) as independent of other factors.
• It requires the user to enter a relative risk associated with the SNP score.

Minor changes

• Addition of Slovenian baseline rates
• HRT intended use now defaults to two years
• Bugs with the GUI corrected
• Corrected error with BRCA1/2 estimates displayed when atypical hyperplasia selected
• Added ashkenazi info to print out
• Corrected bugs with family tree, drawing brother and affected nieces previously
• Changes to GUI
• Saves last used risk length, competing mortality option and country so that the user does not have to keep changing them if e.g. always using Swedish rates.
• Density, SNP risk and country added to print out; age added to x-axis label
• Reorganised layout of form to include density
• Updated input file format to include density and SNPs (MS Word .doc description).

Bug fixes

• v8.0b (13-Sep-2017): Fixed problem with HRT calculation.

Beta version changes

• beta6.rc2 (17-Jan-2017): Updated density box on form. Now defaults to the last density method chosen after selecting 'Add'.

Changes to version 7

Main changes

• The way risk is calculated for women with atypical hyperplasia and LCIS (no other risk factors used in calculation)
• Following Boughey, J. C., L. C. Hartmann, S. S. Anderson, A. C. Degnim, R. A. Vierkant, C. A. Reynolds, M. H. Frost, and V. S. Pankratz (2010). Evaluation of the Tyrer-Cuzick (international breast cancer intervention study) model for breast cancer risk prediction in women with atypical hyperplasia. Journal of Clinical Oncology 28 (22), 3591-3596.
• Updated and smoothed rates of invasive breast cancer (ICD-10 code C50)
  • UK: smoothed first breast cancer rates from Thames cancer registry, 2005-2009
  • Sweden: smoothed first breast cancer rates, 2006-2010
• Updates to the genetic model BRCA assumptions
• Following estimates from Antoniou, A. C., A. P. Cunningham, J. Peto, D. G. Evans, F. Lalloo, S. A. Narod, H. A. Risch, J. E. Eyfjord, J. L. Hopper, M. C. Southey, H. Olsson, O. Johannsson, A. Borg, B. Passini, P. Radice, S. Manoukian, D. M. Eccles, N. Tang, E. Olah, H. Anton-Culver, E. Warner, J. Lubinski, J. Gronwald, B. Gorski, L. Tryggvadottir, K. Syrjakoski, O. P. Kallioniemi, H. Eerola, H. Nevanlinna, P. D. P. Pharoah, and D. F. Easton (2008). The BOADICEA model of genetic susceptibility to breast and ovarian cancers: updates and extensions. British Journal of Cancer 98 (8), 1457-1466.
• Updated hormone replacement therapy component
• Added competing risks projection, to give risk allowing for death from other causes than breast cancer
• UK: smoothed rates from 2008
• Sweden: smoothed rates from 2006-2010
• Competing risks methodology following Gail, M. H., L. A. Brinton, D. P. Byar, D. K. Corle, S. B. Green, C. Schairer, and J. J. Mulvihill (1989). Projecting individualized probabilities of developing breast cancer for white females who are being examined annually. Journal of the National Cancer Institute 81 (24), 1879-1886.

Some more minor changes have also been made.

• Allowed father or brother to have breast cancer. They are treated the same was as an affected woman aged 35. The output file has been ammended to allow this, but the program will still read in version 6 files.
• Standardised lifetime risk: charts and predictions to 85
• Allow user to enter llbs, without giving stones
• Allow user to specify number of years to risk in results
• Cosmetic changes, including that the results graph runs to 85 and is plotted yearly to make it smoother. The woman's ID and date are included on the print out
• Added a Tools menu for competing risks, number of years to risk, and country population rates.

Version 7 bug fixes and updates

• v7.01: Fixed ovarian cancer older than 80 evaluation bug.
• v7.02: Cosmetic changes to form, including making the competing mortality option easier to access.

Changes to version 6

The program now allows you to save patients’ details. Also more information about the woman is given in the print preview screen.

A number of other options are available in the toolbar. These are

Add – Add a patient to the file.
Del – Remove a patient from the file.
Risk – Create a file giving risks for the patient in 10 years time.
Sort – Sort the patients in alphabetical order
Find – Find a patient (from the id code)

Format of text file used for saving patients’ details (only needs to be known if using the program for batch input)

Version number (in this case “v6”)

number of patients

For each patient

patient_id    age    menarche    parous (2 not known)   age_at_first_child
menopause_status (3 not known)   menopause_age    height (in m)    weight
(in kg)
hyperplasia(without atypia)    atypical_hyperplasia    LCIS
ovarian_status    age_at_ovarian_cancer    ashkenazi_inheritance
hrt_use    hrt_type    hrt_length    intended_use_length    hrt_last_use
genetic_test    father_genetic_test
mother_affected_status    mother_bilateral_status    mother_ovarian_status
mother_age    mother_bilateral_age    mother_ovarian_age
mother_genetic_test_result
no_of_sisters    for each sister defined (same seven variables as defined
for mother (affected_status...genetic_test_result))
paternal_gran variables (7 variables as before)
maternal_gran variables
no_of_paternal_aunts    variables for each maternal aunt
no_of_maternal_aunts    variables for each paternal aunt
no_of_daughters    variables for each daughter
no_of_nieces (= number of sisters + number of brothers with daughters)
for each sister: no_of_daughters    variables for each daughter
no_of_paternal_half_sisters    variables for each half_sister
no_of  maternal_half_sisters    variables for each half sister
no_of_paternal_aunts    for each paternal aunt: number of daughters
variables for each daughter
no_of_maternal_aunts    for each maternal aunt number of daughters
variables for each daughter
no_of_paternal_brothers_with_daughters    for each paternal brother: number
of daughters    variables for each daughter
no_of_maternal_brothers_with_daughters    for each maternal brother: number
of daughters    variables for each daughter

For missing data -99 should be generally be used.

Special cases

The patient_id is a string variable containing the code for the patient
(underscores should be used instead of spaces if necessary).
parous - 0 nulliparous, 1 parous, 2 status not known.
menopausal_status - 0 premenopausal, 1 perimenopausal, 2 postmenopausal, 3
status not known
hrt_use -  0 never,  1  use >5 years ago, 2  use <5 years ago, 3  current
user
hrt_type -  0 oestrogen only,  1 combined (relevant for hrt users, but 1 by
default)
hrt_length - how long someone has been using hrt in the past (relevant for
hrt users, 0 by default)
intended use length - how long someone intends using hrt in the future
(relevant for current users, 0 by default)

For each relative

affected status - 0 no breast cancer, 1 breast cancer
bilateral_status - 0 no bilateral breast cancer, 1 bilateral breast cancer
ovarian_status - 0 no ovarian cancer, 1 ovarian cancer
age - age of relative
bilateral_age - age at bilateral cancer (relevant if woman had bilateral
breast cancer, -99 default)
ovarian_age - age at ovarian cancer (relevant if woman had ovarian
cancer, -99 default)
genetic test - 0  no test,  1  no BRCA gene found,   2  BRCA1 gene found,  3
BRCA2 gene found

Changes to version 5.1 

Added “Copy to Clipboard” option for print preview screen so that user can paste results into Word or PowerPoint.

Changes to version 5

• Added the buttons “Affected cousins” and “Half sisters”. These activate dialog boxes, which ask for information about cousins and half sisters respectively.

• Added a check box for Ashkenazi inheritance to be ticked for someone from an Ashkenazi Jewish family.