Clinical genome sequencing. While exome and genome sequencing is widely used as a research tool, these technologies are also routinely applied in a clinical setting. As with many other data-rich diagnostic tests in medicine, there is an ongoing question on how to deal with potentially relevant findings that turn up indicentally. Now the American College of Medical Genetics and Genomics (ACMG) has released their long-expected recommendations on data return of incidental findings in clinical exome and genome sequencing. Their recommendations provide an interesting basis for discussion on what to do with genetic findings that are found by chance.
The spectrum of opinions. Clinical exome and genome sequencing will soon become a standard technology in genomic medicine with an increasing number of patients undergoing high-throughput sequencing. The widespread application of these technologies raises the question of how to deal with findings that clinicians weren’t looking for in the first place, but that cannot be ignored as they might point towards a potentially treatable disease in the patient. There is a wide range of opinions regarding this matter and the ACMG recommendations are very clear that there is no absolute truth. On the one hand, there are genetic liberitarians who feel that any potentially pathogenic variants should be returned to patients, as the patient has the right to know. On the other hand, there are so-called genetic empiricists who claim that little is known about the penetrance of most pathogenic variants and that an overly liberal mode of data return might place an unrealistic burden on a patient, turning him or her into a “patient in waiting” for a disease that was found incidentally. The ACMG recognizes that there is a spectrum of opinions regarding the return of indicental findings in high-throughput genetic data. Therefore, they try to find a middleground.
The ACMG solution. As a compromise between both extremes the ACMG makes five suggestions.
(1) Germline mutations found on a list of ~60 genes for ~20 disease should be returned in any case, independent of the initial indication of the sequencing or the age of the patient. When sequencing patient-parent trios, this information should also be returned for the unaffected parents.
(2) Only a certain class of variants per gene should be reported. This range of variants differs from gene to gene, including known pathogenic mutations and predicted pathogenic variants.
(3) The burden of pre- and post-test counseling is on the clinician ordering the test.
(4) Given the existing technical difficulties, only diseases are included where point mutations play the main role. For disorders due to copy number variants, repeat expansions or indels, the technology is not ready yet to deliver high quality data.
(5) These guidelines should be revisited every year given the dynamics of the field. The ACMG is well aware that we are currently in a transitional period where these technologies can be applied to some, but not all patients and that perspectives may change in the future. There are a few points in the recommendations that I would like to point in particular.
The diseases to be reported. The list of conditions in the ACMG recommendation is heavily biased towards genetic cancer syndromes and heart diseases. There is only one neurological disorders on this list, namely Tuberous Sclerosis Complex, which may be the result of the broad phenotypic range of some of variants as GLUT1 deficiency and/or the fatal course that does not allow for intervention as in the case of Huntington’s disease.
Patient preferences and data return to children. The ACMG discusses whether patients should be given a choice of whether incidental findings are returned. They also discuss the case of exome/genome sequencing in minors. In both cases, they suggest that no exception should be made for the return of incidental findings. One of their reasons is that such stratification might put an unrealistic burden on the sequencing laboratory and –in the case of minors- parents might only learn about a potential relevant finding applicable to the parents themselves through the information of the child. By placing the burden of data return on the referring physician, ACMG issues a relatively industry-friendly guideline that relieves the sequencing industry of the burden of dealing with two potentially controversial scenarios, namely (a) returning data to a patient who hasn’t asked for it in the first place and (b) returning data to a minor who might have made a different choice as an adult. I feel that ACMG does not believe that there is a responsibility for these issues on the side of the sequencing industry. I am curious to see how this discussion will evolve over the years.
Questions not asked. The ACMG recommendations emphasize that they did purposely not consider several issues that are also relevant to the field of incidental findings. They did not include diseases that are part of the newborn screening, and they voluntarily did not consider issues related to data ownership, patenting, secondary costs to the health care system and the fact that some of these incidental findings may become part of the patient’s permanent health record. They suggest that the estimated 1% of patients who are expected to receive reports on incidental findings should be monitored centrally.
The burden of the all-knowing clinician. The ACMG recommendations place a significant responsibility on the clinician ordering the test, suggesting that the person in charge of counseling the patient prior to exome/genome sequencing should be well trained in understanding the intricacies of genetic incidentalomas. Even though in an ideal world, there is sufficient time to discuss all the pros and cons of these investigations with patients, this is hardly the case in real life. Even though high throughput genetic investigations require a lengthy pre-test counseling, this is hardly reimbursed or valued. We are in the strange situation where health care systems are willing to pay several thousands of dollars/euros for performing these tests, while resources for translating these finding to the patients before and after the test are hardly valued. It does not take much fantasy to anticipate that this imbalance will create a wide range of conflicts in the near future. Maybe after “big tobacco” and “big pharma”, “big omics” will be the next industry to be hit with class action lawsuits.