Should primary health physicians order cancer-related genetic tests?

(This post is reproduced from an article submitted to The Medical Observer on 7th June, 2016.)

GPs are skilled at assessing and treating patients and, in the case of complex problems, referring them to the right specialist. What’s different about genetic testing?

There is nothing truly difficult about being a genetic oncologist or clinical geneticist: much of it is pattern recognition based on acquired knowledge and experience.

And there isn’t anything intrinsically difficult about a ordering a genetic test. Such testing involves a blood or saliva sample to provide the white cells from which the germline DNA can be extracted.

The patient’s DNA is then screened for changes in the DNA code. These may be common and usually harmless (polymorphisms), disease-causing (pathogenic mutations), or the change may not yet be classified or fully understood (a variant).

As with any test, it’s the interpretation that can be is problematic. Think of trying to interpret an MRI scan without a radiologist’s report. For this reason, many providers of genetic testing, particularly direct-to-consumer testing, emphatically state that they don’t provide a medical service, diagnosis, treatment, or even advice.

This shifts the expensive part of the test — informed consent and explaining what the result actually represents in term of cancer risk — to the doctor who orders the test.

To continue the imaging analogy, the best way to investigate a sore knee might be via ultrasound, X-ray, CT, MRI or even blood test (if septic arthritis is considered). Similarly, genetic testing comes in different forms: single gene or genes, targeted panels, general panels, arrays or whole exomes or genomes.

Choosing the correct test is important to maximise health resource dollars and to avoid creating more uncertainty. For example, using breast MRI when a mammogram would do.

Ordering more tests is not necessarily better. Panel testing is not usually used in a clinical setting, as the more genes tested, the greater the chance of finding variants, each of which needs to be researched and considered.

Not all genetic labs provide the same quality of testing or interpretation. And, just as with imaging, not finding a mutation doesn’t mean there isn’t a problem.

When it comes to heritable cancer syndromes, a positive family history is still a positive family history whether a mutation is identified or not. That’s why the term uninformative rather than negative is employed when a mutation search doesn’t turn up a mutation. Such individuals and families remain at increased risk and therefore may require earlier onset and/or more intensive cancer screening.

GPs are often highly skilled at managing these families or co-ordinating their care.

Not identifying these individuals, or worse, providing false reassurance, might mean the patient doesn’t seek adequate or early screening or might present late for investigation of a symptom.
I would argue that if you are going to refer the patient to a genetics service anyway — because of a high-risk family history or if a variant is detected — it is much better to do so before ordering genetic testing. Long waiting lists don’t get any shorter when your patient is already anxious about an unexpected or unexplained result! This would reduce the risk of an incorrect or insensitive test being ordered.

Like surgery, whether it’s a small procedure performed in your practice or a total knee replacement, informed consent is required. This involves discussing the purpose of the genetic test (which requires an understanding of the syndromes associated with each gene to be tested), the benefits (based on a detailed family history) and risks (including on life insurance) and other options such as research opportunities and not having the test at all.

With genetic testing, I believe it should also form part of informed consent to discuss how the lab will handle the patient’s personal and family information and how the DNA sample will be treated and stored. For example, my usual lab is based in Australia and stores the DNA sample indefinitely.

This allows it to be used as a positive control in predictive testing of other family members and also allows future testing, even if the patient is deceased, if permission has been provided.
The lab also uses the samples for internal quality control and notifies me as information becomes available regarding a particular variant or when further testing should be performed due to changes in technology.

Some labs use the DNA to test other parts of the genome for independent research and development.

Some people baulk at the idea that their DNA may be used for commercial gain. I am not just talking about large corporations: public hospitals and universities have enthusiastically applied for patents and aggressively defended them. The commercial application of medical knowledge is big business.

One reason public clinics may have long waiting lists is because the consenting process, and explaining the result, takes time.

If something goes wrong, it is you, not the test provider, who is in the firing line. And doctors have been sued. They have been sued for failure to warn of genetic predisposition to cancer, premature utilisation of a genetic test to tailor treatment and incorrect interpretation of a test result, to name a few instances.

Just as you wouldn’t recommend a do-it-yourself knee replacement kit, long waiting lists shouldn’t be used as an excuse to provide inferior genetics services.

Public familial cancer clinics do triage patients, providing more rapid access for those needing to make cancer-related decisions. Outreach services exist for non-metropolitan patients. In my practice I provide bulk billed telehealth consults for rural or regional patients, usually to the GP’s practice.

Although GPs and registrars may sit in for only part, if any, of this consult, the detailed summary letters, complete with management guidelines, allow GPs to build up their experience of and familiarity with hereditary cancer syndromes. They also improve the recognition and management of individuals who are actually at low risk.

Moreover, by encouraging patients to attend with blood relatives, whole communities can be educated about cancer risk reduction.

So, while GPs can order cancer genetics tests, I would argue that most shouldn’t. This is not gate keeping, it’s about ensuring GPs, like all doctors, practice within their level of competence, keeping their patients safe and themselves out of the newspapers.

Posted in: For Drs