Study assesses two ways to test patients' hereditary cancer risks

Up to 10% of cancers are caused by genes that can be easily detected by commercially available tests. These include such common cancers as cancer of the breast, ovary, colon, stomach, uterus and pancreas. 

We don't routinely screen for cancer susceptibility genes in primary-care settings because genetic testing is often considered too complicated and primary care doctors already have so many things they need to address. But it is an opportunity lost." 

Dr. Elizabeth Swisher, lead author, UW Medicine gynecological oncologist

In the JAMA Network Open study published today, Swisher and colleagues assessed two ways that primary-care practices could assess patients' hereditary cancer risks and deliver testing to those identified as higher risk. 

Patients with a family history of cancer are at the highest risk of carrying these genes. Even though they are an ideal group for testing, screening of these patients is rarely done as part of routine primary care, when it's most likely to detect the risk before a cancer has developed, Swisher said. 

In the study, one approach involved asking patients when they came to the clinic, or before a virtual visit, to fill out a questionnaire before they saw their primary-care provider. This was called the point-of-care approach.

The second approach involved sending a series of letters or emails to patients inviting them to fill out the questionnaire online from home. This was called the direct patient engagement approach. 

The questionnaires asked about the patients' cancer history and that of their first-degree relatives - parents, siblings, and offspring - and second-degree relatives such as grandparents, aunt and uncles, and nieces and nephews. The questionnaire also asked about relevant ethnic information, such as Ashkenazi Jewish ancestry, which is associated with a genetic risk of several cancers, including cancers of the breast, prostate, colon and pancreas. 

Patients whose completed assessment suggested they might carry a cancer-susceptibility gene were offered a test that screened for 29 such genes. 

The saliva-sample test was offered for free and could be administered at home. All patients who were found to have cancer-risk gene variants were provided genetic counseling. 

"The goal was to bypass the primary-care physician up until there was a positive test and then provide the physician with a care plan for each patient," Swisher said. 

The study was conducted in 12 primary-care clinics run by two different health care systems. Six were in Washington state, run by MultiCare, and six were in Montana and Wyoming, run by the Billings Clinic. The MultiCare clinics primarily serve a mixed ethnic and racial urban population. The Billings Clinic primarily serves a rural white population. 

The clinics were randomly assigned to use one or the other approach. During the study period, 95,623 patients were seen in the 12 clinics. Of these, 18,030 were approached during their visit in-clinic and 41,558 were sent email or letters. 

The researchers found that the point-of-care approach resulted in a higher proportion of patients (19.1%) completing the risk assessment than was the case in the direct patient engagement group (8/7%), relative to the total clinic population. 

But among those whose assessment indicated they were eligible for testing, a larger proportion of the direct engagement group got tested: 44.7% compared with 24.7% of the point-of-care group.

Swisher speculated the respondents in the direct engagement group may have already had concerns about their family genetic history. Although fewer from this group responded, those who did were more likely to test positive (6.6%) than were people in the point-of-care group (3.8%). 

"Both strategies had some utility," Swisher concluded, "but we need to develop ways to improve uptake of testing and further reduce barriers to the testing process."