NEW YORK (GenomeWeb) – NorthShore University HealthSystem is testing whether incorporating an in-house-developed method to assess patients' genetic risk for breast, prostate, and colorectal cancer can lead to better prevention for individuals who do not yet have cancer.
The researchers will analyze 300 SNPs that genome-wide association studies have shown to be associated with cancer risk in 500 consenting patients between the ages of 40 and 70 who have not been diagnosed with cancer. Tests will be delivered through the patients' primary care physicians. About 150 physicians will participate in the study.
"Typically, we use family history to stratify patients into high or low risk," which is "good, but insufficient," Jianfeng Xu, director of the Program for Personalized Cancer Care at NorthShore, told GenomeWeb.
The goal of this study will be to evaluate whether incorporating a genetic risk score can identify individuals who would have been missed by traditional family history-based screening. The researchers will also establish the clinical utility of such a test and study how patients react to the information and how physicians incorporate such testing, Peter Hulick, medical director for NorthShore's Center for Personalized Medicine, told GenomeWeb.
Genetic testing firm Counsyl will run the NorthShore designed next-generation sequencing-based assay out of its clinical laboratory.
The researchers chose the 300 SNPs through various GWA studies. They are all well-validated SNPs associated with an increased risk for either breast, prostate, or colorectal cancer, but fall outside of the traditional high-penetrance cancer genes, like BRCA1, BRCA2, and the Lynch syndrome genes, Xu said.
Xu's team then developed the testing algorithm to take the SNP results and convert them into a numerical score to predict a person's genetic risk for developing breast, prostate, or colorectal cancer.
He said that NorthShore partnered with Counsyl to run the actual test because it has a large clinical laboratory with the ability to scale up if, following the initial 500-patient trial, NorthShore decides to offer the test more broadly.
Along with the genetic risk score, the NorthShore researchers will provide the primary care physician information about recommended screening protocols for patients based on their risk. For instance, said Xu, it is recommended that individuals at an average risk for colorectal cancer receive a colonoscopy every 10 years. But, depending on a person's other risk factors like genetics or family history, they may benefit from starting screening at an earlier age and at an increased frequency, he said.
"We will provide simple tools for the primary care physicians to make management recommendations," Xu said.
Similarly, women with a 20 percent or greater lifetime risk due to family history for developing breast cancer qualify for increased screening, such as MRI or mammagrams, Hulick said. In the trial, the researchers hope to catch women who may be at an increased risk due to their genetics, but don't have the family history, Hulick said. Although BRCA1 and BRCA2 testing is not part of the trial, women who would qualify for such testing based on other indicators, like family history, would still be offered that testing.
Also as part of the trial, the researchers will "monitor uptake of the recommendations and make sure that they are in line with the patients' level of risk," Hulick added.
Patients will be recruited from NorthShore's Genomic Health Initiative — a research program within the health system's Center for Personalized Medicine in which patients agree to donate a blood sample for genomics research. Researchers will send consent forms to patients who are enrolled in that program and meet the trial's eligibility requirements.
The risk score will also be incorporated into the patients' electronic medical records. A key component of the trial will be to provide results that are easily understood by physicians with clear guidelines, Hulick said. "Any successful precision medicine initiative has to be integrated into the system such that physicians and patients are both comfortable using it," he said.
If this initial trial is successful, the researchers hope to both roll out this type of testing more broadly, but also to expand the trial to include testing of the major cancer genes like BRCA1, BRCA2, and Lynch syndrome genes.
"Everyone should have knowledge of their inherited risk from family history, a genetic risk score, and high-penetrant cancer genes to design a cancer screening protocol," Xu said. "That's the future." But, he added, the group decided to start with the SNP-based test as a first step to evaluate uptake and use. Also, Hulick noted, with a SNP-based test as opposed to full gene sequencing, physicians will not be confronted with trying to interpret variants of unknown significance.
The NorthShore team is taking a different approach than others in the field who are advocating for incorporating genomics when assessing individuals for cancer risk.
For instance, a number of researchers, most notably Mary-Claire King of the University of Washington, have been advocating for broader screening of cancer predisposition genes due to several studies that have identified significant numbers of individuals with mutations in those genes who do not have a strong family history.
In addition, companies such as Color Genomics and Counsyl now offer relatively inexpensive (under $400) NGS-based panel testing of cancer predisposition genes. And, the HudsonAlpha Institute for Biotechnology is nearing the end of its one-year-long offer of a free 19-gene cancer screening panel to 30-year-old women.
Hulick and Xu said that while such panels may be incorporated into a future trial, they wanted to take this initial step of using a genetic risk score along with the standard family history-based information and incorporating that into patients' electronic medical records. Aside from helping to stratify patients into high- or average-risk groups, the process will be important for physicians and patients to feel comfortable with genomic information. "No matter how good a test is, it will fail if patients and physicians are not comfortable using it," Hulick said.