Practical Advice for Developing a Genetic Testing Program

Theodore Bosworth


November 30, 2015

Uncertainty about the logistics of genetic testing is clouding the general agreement that these tests should be far more commonly used in patients with cancer, judging from the consensus among experts at the 2015 European Cancer Congress. All four panelists in a symposium dedicated to this topic agreed that genetic testing is a mainstream service that is now essential in several forms of cancer, but they acknowledged a lack of standards for cost-effective applications, appropriate consent, and adequate counseling.

Several specific genetic tests are clearly and routinely indicated. For example, Susana Banarjee, MD, PhD, a consultant oncologist at Royal Marsden Hospital in London, described BRCA testing as "essential now for women with ovarian cancer," because the results "can and do lead to changes in therapy." Yet, Dr Banarjee agreed that clinicians require specific training to propose and gain consent for the test and to deliver the results. This is mandatory at her institution.

In the training at Royal Marsden Hospital, clinicians are instructed how to explain the goals of genetic testing and how results will be applied. Although some clinicians may require training so that they themselves are clear about the significance of genetic tests, the primary objective for certifying expertise is to ensure that they can convey accurate information in a format meaningful to patients.

This process, all acknowledged, is time-consuming. This is not the least of the logistical challenges. Busy clinicians can be saved from this burden by referring patients to a clinical geneticist, but these are in short supply. At Royal Marsden, clinical geneticists do not routinely counsel patients about BRCA testing. Rather, nurse specialists, who undergo the same certification process for counseling patients as physicians, are increasingly taking over this task. Dr Banarjee reported that they appear to be as effective for providing accurate information about genetic testing, but have so far been found more effective for securing consent to proceed to genetic testing.

gene testing should be conducted with narrow goals and limited genes.

At some institutions, the work has been divided, so that clinicians provide initial counseling and secure consent but patients are referred to a geneticist if results are positive. This, according to Helen Hanson, MD, a consultant in cancer genetics at the Royal Marsden, may be the best approach when geneticists are available because of their skills and knowledge in placing genetic information into context. This is particularly useful when a panel of genes is tested and there are incident findings, such as identifying a predisposition gene for colorectal cancer in a patient whose genetic testing was ordered to evaluate predisposition to breast cancer.

Owing to incident findings in patients without other risk factors for cancer, there was a consensus that gene testing should be conducted with narrow goals and limited genes. Although Dr Hansen said, "I think patients are very receptive to knowing all of the information relevant to their risk for cancer when it is presented with appropriate counseling," panels that include genes with incompletely validated associations may create unnecessary anxiety over findings for which there is no clear course of action.

"Many panels include genes I would not consider [as] conferring medium to high risk," cautioned Dr Hansen, who warned that ambiguous information could be unhelpful, particularly for patients who are not adept at considering relative risks.

The value of restricting genetic testing to "actionable" genes, meaning genes that will change clinical management either by screening or treatment, was the major message of Nicola Normanno, MD, chief of the pharmacogenomics laboratory at the National Tumor Institute—Fondazione Pascale in Naples, Italy. He particularly criticized offering genetic testing to patients who have cancer when there is no clear plan of action based on the results.

"If you want to activate a genomic testing program in your institution, this should really be done to match patients with a clinical trial. Then if you find a mutation, you will be able to offer a treatment tailored to their disease," Dr Normanno said. He cited data suggesting that only a minority of patients who undergo genetic testing then have their treatment guided by the results, "and we need to solve this problem."

Of note, a gene mutation associated with an actionable therapy does not guarantee benefit outside of the cancer in which benefit was initially shown, according to Dr Normanno. He cited recently published results of a phase 2 study in which molecularly targeted treatments were evaluated in tumors other than those where efficacy was initially demonstrated in a clinical trial.[1] In that study, the treatments provided no advantage in progression-free survival outside of their initial indications.

Other challenges to genetic testing highlighted by Dr Normanno include the difficulty of accounting for the changing molecular profile of tumors as they evolve over time, as well as the heterogeneity of tumors that may have multiple mutations with variable importance in driving progression. He noted that current classifications often divide patients into either having a mutation or wild-type disease even though a more detailed genetic profile will ultimately prove more useful in directing therapy. However, none of these unresolved problems suggests that genetic testing is not ready for routine application.

none of these unresolved problems suggests that genetic testing is not ready for routine application.

"Although there are all these challenges to using genetic testing that need to be solved, I really believe that genomics in oncology should be brought into the clinic as soon as possible, because this is going to increase the ability of patients to access recent developments in novel therapeutics," Dr Normanno said.

Still, like the other panelists, he contended that substantial thought must be invested in the logistics of developing genetic testing. For example, he believes that systematic protocols are needed for obtaining, evaluating, and storing tissue at the time of diagnosis. He also suggested that protocols for multidisciplinary management should be formalized.

"We need a direct interaction between pathologists, molecular biologists, and geneticists," Dr Normanno said. As therapies become more tailored on the basis of increasingly detailed genetic profiles, "there are likely to be cases in which we will need to go back to the laboratory to understand the meaning of specific alterations and to develop novel therapeutic approaches."

Once genetic profiles are in hand, one of the challenges is developing effective education programs so that patients feel fully engaged in the clinical choices affecting their prognosis, said Bettina Ryal, MD, PhD, founder of the Melanoma Patient Network Europe in Uppsala, Sweden. Although she said that patients know more about genetic testing as a result of the recent attention drawn to it by the actress Angelina Jolie and others, messages to individual patients must be tailored.

"Some patients may know very little, while others may know more than you," said Dr Ryal, cautioning against a rigid, one-size-fits-all approach. She suggested that a mix of educational materials, including materials in different media, such as print and video, generally help patients grasp the concepts with greater confidence.

"What we have found is that it takes repetition. Learning is about repetition," said Dr Ryal, citing her own experience in developing material for her work in patient advocacy. She suggested that developing broad and diverse educational materials cannot be underestimated for their role in making a genetic testing program successful from the patient's perspective.

As a patient advocate, she also suggested readable consent forms. Although she acknowledged that lawyers typically insist on long and detailed summaries, these will not be read by the average patient and are "effectively hiding important information." She suggested that longer forms be complemented with an executive summary that will help the patient understand the key issues.

Genetic testing programs cannot be undertaken casually, according to Dr Ryal. Among the unforeseen complications of genetic testing, for example, Dr Ryal said that a child might be unpleasantly surprised to find as a result of genetic testing that their father is not a direct relation. This can place the clinician in the center of an unanticipated "family disaster." These types of events and risks need to be considered at the outset when planning a genetic testing program, according to Dr Ryal, emphasizing the need for extensive planning.

Disclosure: Drs Banerjee, Hanson, Normanno, and Ryal each report no relevant financial conflicts.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.