Using test results in the clinic

Results independent of other risk factors

Because deCODE’s tests’ risk results are largely independent of other risk factors, they can either stand alone as a measure of the genetic contribution to the cause of common diseases, or be used to modify the patient’s risk as assessed by conventional risk assessment tools, such as the ten year Framingham risk score for heart attack or the Gail score derived risk for breast cancer. This simply happens by multiplication of the two relative risk values, resulting in a more complete risk assessment.

Impact and integration

In order to evaluate the effect a test may have on a patient’s overall assessed risk and on subsequent clinical care, it is useful to look at the range of the risk results and the distribution of the riskvalues in the population: that is, what proportion of the population will receive a result putting them at lower, average or higher risk.

This information tells the physician what the likelihood of a given result is before the test is ordered. This should be considered together with information on disease prevalence (population risk), age, other risk factors and their degree of effect, as well as clinical criteria and guidelines on intervention or screening. By this physicians can better understand and explain to their patients the significance of the test in clinical context and the likelihood the results will have an impact on treatment and/or preventive measures and recommendations.

deCODE’s breast cancer test offers an instructive example. The average lifetime risk of breast cancer among US women of European descent is 12%. The deCODEBreastCancer™ risk distribution curve shows that while only about 1.5% will receive a relative risk result of 2.0 or higher (representing double the average risk), 5% will have a result of 1.65 or higher and 12% will get a result of 1.4 or higher. While a result of 2.0 brings a woman’s lifetime risk to 24%, a result of 1.65 will bring her over 20% lifetime risk, the threshold at which the American Cancer Society recommends an annual breast MRI in addition to mammograms. For women who are 55 and older, a result of 1.4 or higher, which 12% of women will have, corresponds to at least a 1.7% likelihood of developing breast cancer within the following five years. This is the threshold used by the American Society of Clinical Oncology (ASCO) for consideration of the use of tamoxifen (or similar drugs) as preventive therapy.

Into the above calculations may also be taken the risks associated with conventional risk factors as individually assessed, or as combined in common risk assessment tools such as the Gail score. This provides a more complete and personalized risk assessment for the patient.

deCODE test results and family history

The common variants measured by deCODE’s genetic risk tests appear to be largely independent of family history of disease. That is, they do not appear to account for higher risk in patients with close relatives who have been diagnosed with the disease in question. This means that the relative risk detected by deCODE tests can be used to multiply the risk due to family history for calculation of a more complete lifetime risk. It also means that there are rarer genetic risk factors, of the sort that account for family history, still to be discovered.

deCODE risk profiles are therefore valid both for patients with and without family history of disease, and are independent of the risk conferred by such history. At the same time, deCODE’s risk results do not test for or exclude the possible presence of rare Mendelian genetic causes of the same diseases, such as BRCA mutations for breast cancer or variants causing hereditary blood-lipid disorders leading to early-onset cardiovascular disease.

This content was last reviewed on January 25, 2011.