When the authors of a report include “Highly Anticipated” in the title you know that either: 1) it is highly anticipated, or 2) they’re trying to get some attention.  In the case of the recent ACMG guidelines ACMG Recommendations for Reporting of Incidental Findings in Clinical Exome and Genome Sequencing Report (www.acmg.net) it’s both.  And what do these recommendations say?  For one thing, they say that if you get a sequencing test you can’t choose to not hear about anything bad that is discovered.

As most people are aware, recent advances in genome sequencing technologies have made it possible to sequence most or all of a patient’s genome for clinical diagnostic and predictive testing.  Here’s a couple of open-access reviews to get you up to speed on the details:

Genomics Reaches the Clinic: From Basic Discoveries to Clinical Impact
Teri A. Manolio and Eric D. Green
https://www.sciencedirect.com/science/article/pii/S0092867411010701

Next generation sequencing—implications for clinical practice
Eleanor Raffan and Robert K. Semple
http://bmb.oxfordjournals.org/content/99/1/53.full

So, now that we can do it, everyone is starting to concentrate on the next questions.  Should we do it and, if so, how and under what circumstances?  There are a number of big issues with this type of testing because it deals with genetics.  You can’t escape or change your DNA, your parents gave it to you and your kids will get half of it, and a myriad of privacy and legal issues.  Another concern is that during testing an unexpected discovery will be made: a so-called incidental finding.

Incidental findings have been an issue in other areas of medicine for some time: the phrase incidentaloma has been used in radiology to describe a tumour that is discovered during imaging for an unrelated condition (“my back hurts”, “OK let’s give you a chest x-ray”, “I’m sorry to say you have a lung tumour”).  In genetics they were not common because the testing was so targeted that it would usually only find what it was looking for.  (Incidental findings of non-paternity, consanguinity and sex-chromosome abnormalities are possible, among other things.)  But once we start sequencing entire exomes or genomes, it is increasingly likely that testing will discover medically-relevant changes in genes completely unrelated to the purpose of the test.

Incidental findings in research have been discussed for a few years now.  The concern there is that researchers will discover something in a patient’s genome but be unable to inform the patient, either due to legal and ethical restrictions or to sample anonymization.  See the following for an overview:

Disclosing pathogenic genetic variants to research participants: Quantifying an emerging ethical responsibility
Christopher A. Cassa, Sarah K. Savage, Patrick L. Taylor, Robert C. Green, Amy L. McGuire and Kenneth D. Mandl
http://genome.cshlp.org/content/22/3/421.full

In the clinical realm, the patient is known and there is a mechanism to inform the patient of genetic findings.  However, the patient might have had testing to see if they were a cystic fibrosis carrier, and perhaps the testing discovers that they carry a pathogenic BRCA1 variant.  Should they be informed that they are at high risk for developing breast or ovarian cancer?  Probably.  But they might not want to know.  And it means that their siblings, their children and other relatives should also be informed.  What if you’re testing a child and discover the same thing?  Should a child be told they carry a risk for an adult-onset disorder? (These are emphatically not even tested for currently).  There are a lot of scenarios but you can see the problem.  It’s potentially Pandora’s box and you don’t get to choose whether it’s opened.

These ACMG recommendations are brand new and everyone is taking some time to digest and consider them.  But they have some strong and possibly surprising rules for the labs doing the testing.  With the caveat that I’ve only given them a quick skim, here’s my quick interpretation of of the important bits:

1) Mutations found in a provided list of 56 genes (mostly cancer-associated) must be reported by the laboratory, regardless of the clinical indication (what the test is looking for).  If the labs generate the data they can’t mask or ignore it to avoid this requirement.
2) The clinician ordering the sequencing test is responsible for informing and consenting the patient about the possibility of these incidental findings.  About 1% of patients can expect to get one.
3) The patient can’t decline to have incidental findings returned – if they don’t want them their option is to not have the testing.  “Duty to warn” trumps “patient autonomy”.
4) These recommendations are still valid if the patient is a child.  “Parent’s right to know” trumps “child’s right to not know”.

The gene list has been carefully chosen to “prioritize[d] disorders where preventative measures and/or treatments were available and disorders in which individuals with pathogenic mutations might be asymptomatic for long periods of time.”  The stated assumption is that any reasonable person would want to know about these.  But that’s not a very nuanced position.

As I said, everyone is trying to think about what these might mean and how they’ll be implemented.  This is really big news in the medical genetics community and it has important implications for everyone.  Take a look at the link at the top (it’s not very technical) and think about it.  Everyone’s opinion is important.