In general, the more data you enter into a risk model, the more accurate the risk calculation will be. On the other hand, the more data you ask users to enter, the less likely they will be to use the model for all the patients who need the calculation.
Thus, while Tyrer-Cuzick and BRCAPRO both use the age and vital status of every relative, this excessive need for data entry relegates these models to use only in high-risk clinics and only to the highest risk individuals.
To manage this problem, many people use what we have dubbed the “Lyte” version of the models. We have been using BRCAPROLyte for years on an informal basis and have recently undertaken a more formal study of this approach lead by Dr Biswas.
Biswas S, Atienza P, Chipman J, Hughes K, Barrera AM, Amos CI, Arun B, Parmigiani G. Simplifying clinical use of the genetic risk prediction model BRCAPRO. Breast Cancer Res Treat. 2013 Jun;139(2):571-9. Epub 2013 May 21.
We compared BRCAPRO using full family histories, to BRCAPROLyte, which used only the age of cancer diagnosis for each relative, stripping out the age and vital status of relatives. In addition, 2 new approaches were created, BRCAPROLYTE-Plus and BRCAPROLYTE-Simpl.
BRCAPROLYTE-Plus imputes the age of the relatives in the family structure without having to collect this data specifically. That is, if a patient is 40 you would assume her sister is also 40 and her mother is 60 and her child is 20.
BRCAPROLYTE-Simple further simplifies the process by imputing the entire family and their ages. That is, one does not even need to collect family structure. One could assume that a patient has one sister, one maternal aunt and one paternal aunt (in addition to mother and grandparents) and assume ages for all.
The ROC curves show that all of these modifications work pretty well.
When looked at it in terms of risk thresholds, the Lyte versions seem to do about as well as BRCAPRO. The table shows the number of mutation carriers detected (Numerator) over the number of referrals made by that tool (Denominator) at that cutoff (Threshold).
While this approach has not been tested for Tyrer-Cuzick, I would assume that similar findings would be expected. In fact, most people who use TC never enter unaffected relatives, de-facto using it as “TCLyte.”
In the context of a primary care practice or a breast imaging center, we feel the lyte versions are more than adequate as a “screening” family history. The significant advantage is that very large numbers of women can be screened with relatively minimal data entry (to see this approach, try CRA Health's Risk Express).
As with any screening study, if the threshold is tripped a more intensive approach can be undertaken by running the full versions of the models.
We can help you make the models more useful. You can quickly run the various models and compare their results at CRA Health's Risk Express.
About the Author: Kevin S. Hughes, MD, FACS
Kevin S. Hughes, MD, FACS is a co-founder and medical advisor to CRA Health, LLC. Dr. Hughes is the Massachusetts General Hospital’s Surgical Director of the Breast Screening Program, Surgical Director of the Breast and Ovarian Cancer Genetics and Risk Assessment Program, and Co-Director of the Avon Comprehensive Breast Evaluation Center, and serves as the Medical Director of the Bermuda Cancer Genetics and Risk Assessment Clinic. He is an Associate Professor of Surgery at Harvard Medical School. Dr. Hughes is actively involved in the establishment of standards and in research regarding the genetics, screening, diagnosis, and treatment of breast cancer. He is the author of numerous papers and book chapters on the subjects of breast cancer, screening, diagnosis and treatment, and risk assessment. More information can be found at: thebreastcancersurgeon.org.