BRCA 1 of unknown clinical significance

BanR

Mutd: I am very very impressed!! You are a huge ocean of knowledge!!!! And thanks so much for sharing this with me. I have sent my reports for re-evaluation and i will surely let you know how things proceed. I wish this distant relative of mine who had ovarian cancer twice at age 27 and 30 and is now 31, could get herself tested. Maybe we get some valuable information. But at the same time its strange, since nobody else in the family or extended family got any kind of cancer and only we both ended up getting breast and ovarian cancer respectively! ( i got dx last year, close to my 35th birthday). Either its a coincidence or there could be some link.

thanks once again!

cymom

BanR - Can I ask which lab you used? Sounds like they made an error in the way they reported your results.  How unnerving...

I am currently waiting for BRCA test results.  My GC said I should use Myriad, because they have more data and are less likely to report a variant of uncertain significance than the competing labs because they have a lot of data that they have not shared.  I don't know if that's true, but I do worry about the accuracy of results from labs that lack data.  I also worry about whether to trust Myriad.  The reports I've heard about the steps they've taken to protect their bottom line are frightening.

BanR

Myriad is what I am considering too.

Anyway tomorrow I begin my 4 dose dense taxols. I am keeping my fingers crossed...white blood cells and haemoglobin is low. Wonder how will my first taxol be, in terms of side effects.

GraceBer

ctnymom - yes, this is true from what I've experienced.  I just got my results today and I'm also "variant of unknown significance," but I was pushed into going with Ambry labs because I was told that they'd get the results to us faster (and it sounds like my hospital has a relationship with Ambry) and considering that this was before Thanksgiving and I wanted surgery asap, we went with Ambry (and DID NOT get the results quickly - it took 5 weeks), but my GC did not highlight just how much smaller Ambry lab is and how little data in comparison it has to Myriad.  I'm now trying to find a way to get retested through Myriad without paying a huge sum of money (apparently my insurance won't cover it as I've already been tested).  

And for fun, here's my info: "p.T617A variant (also known as c.1849A>G or 1968A>G)....This variant was previously reported in the SNPDatabase as rs45564238...it is predicted to be benign...Since supporting evidence is limited at this time, the clinical significance of p.T617A remains unclear." 

cymom

GraceBER,

You should contact Sue Friedman at FORCE about this.  (suefriedman@facingourrisk.org)  I know she is in contact with the researchers at the University of Washington.  Many labs are trying to pool data to undo the damage that Myriad has done and give women answers if the data is, in fact, out there.  She mentioned to me that there are ways to determine the significance of mutations without going to Myriad.

Read this for more info: http://www.facingourrisk.org/messageboard/viewtop…

Hope this helps!

BanR

I believe, they have passed a new law and Myriad now needs to open up its own database. It seems Myriad does have have more data but it has yet not shared it with any other labs. Lets hope our mutations of unknown significances get clear. 

Got my report rechecked. They hope its benign. I will be in touch regularly with them now.

GraceBer

My genetic counselor got in touch with Myriad.  They confirmed that they also classify the variance that Ambry detected as "unknown significance."  That is, that they haven't upgraded (downgraded) to a deleterious variance/mutation.  However, Myriad won't reveal if they've classified it as benign.  I'd have to pay to have the test taken with them to get this result.  My counselor is confident that Myriad just also has this as an unknown significance.  And that someday it will be classified as benign and that I shouldn't worry about it.  Every doctor (surgeon and oncologists) have said the same thing.   But I just can't help but wonder...

cymom

They won't reveal if they've classified it as benign?  Grrrr…

If each lab classifies using different criteria, that's one thing, but if they have more data than other labs do to draw a meaningful conclusion, then what they are doing is essentially selling (at a grossly inflated price) fruit from a tree they never rightfully owned.  This is a particularly sleazy kind of profiteering.  I get that they are a for-profit company, but this is just wrong, and bad public relations too.  I would alert Sue Friedman.  

Try not to worry too much, but do make a stink if you can.

Mutd

GraceBer, there seems to be a confusion. If Myriad hasn't upgraded or downgraded the variant, and if they also classified it as "unknown significance", as they seem to have told your GC, then of course it's still uncertain. If it is uncertain, then they can't confirm or deny that it's benign - not because of some sinister plans, but because that's what "uncertain" means. It means "it may or may not be benign". 

The only reason Myriad would have asked you to order a test for a variant which is still uncertain is to add your results to their dataset in order to accelerate reclassification of the variant. To upgrade or to downgrade, it is important to quantify your risk level, and to offer free follow-up tests to informative family members.

Conversely, if (not you, but somebody else) had a variant which Myriad classified as benign, and asked Myriad to run a test to confirm it, then the lab would refuse. They can't take insurers' money for tests which do not have clinical significance. No clinical testing is possible for benign variants!

cymom

"The only reason Myriad would have asked you to order a test for a variant which is still uncertain is to add your results to their dataset in order to accelerate reclassification of the variant."

Mutd -- the first thing you said makes sense, but I don't get the second paragraph.  Why would Myriad offer the test if the only purpose would be to add to her result to their own data set?  Why would it even offer the testing in this situation?  Now that there are so many other labs doing this testing, it seems like reclassification would be best accelerated if everyone reports their results to a shared database.  I seem to remember reading somewhere that people are doing this, but I can't locate where I read about it...

Mutd

Let me try spelling it out. If the variant is Uncertain Significance then (in most cases anyway) it has no clinical utility at the moment ... but it may become clinically useful later (especially if it is upgraded into a pathogenic mutation). So testing for the variants sometimes makes sense for the people, and for their doctors.

What about the testing labs like Myriad? When does testing people for an uncertain variant makes sense for a lab? Usually, the value for a lab is that additional tests bring more data, and as the data accumulates, the variant is eventually reclassified. It may be a pretty big deal for the companies in today's competitive environment, to push the variant rate down, to tell about at professional society meetings and in publications, to make the physicians and genetic counselors aware of the work being done. Just imagine, what if your test result is the last little building block needed to complete reclassification? Once it happens, then not just yourself, but 5, 10, sometimes dozens of your genetic sisters with the same variant will receive news that the significance of their variant has been clarified. It feels good for the patients and it feels good for the lab people.
The catch is that your test result doesn't have much statistical value in itself, without clinical family data. Consider the 3 examples: 

1. Jane has a BRCA1 variant, she is 65 and cancer-free, and her maternal cousin had breast cancer at 70. The whole picture just doesn't look like BRCA1 mutation family. Statistically, it adds to the pile of evidence for benign-ness.

2. Deanne has a BRCA1 variant, she had both breast and ovarian cancer, and there is a scary pattern of early cancer in her sisters and older relatives. This really does look like a BRCA1+ family, and it adds to the body of evidence for pathogenicity of the variant.

3. Clarissa skips all the details about her health condition and family history. The test result doesn't add much to the variant classification data.

Then there are more things to consider, like family testing etc. Like if someone in your family is healthy in her old age, yet shares the variant, then adds to "benign" evidence. But if someone with cancer shares the variant, then it adds to "pathogenic" evidence. It could be quite valuable, which is why Myriad offers family follow-up for free. Often, the "family variant tree" data doesn't even need to be solicited, because in any big cancer family, there are, already, multiple relatives who already got tested. All it takes is to connect the dots to see that, say, Deanne's second cousin once removed also had the variant (but probably never told her). With just a little addition of genealogy work, one get incredibly valuable insights about the role of variants in such families.

A shared database would have been wonderful ... and there are two of them already (BIC and ClinVar), but both suffer from the same fatal weaknesses. Because of patient privacy concerns, they ask very little (or nothing at all) about health and family, and there is no follow-up family testing either. And of course sharing is voluntary and data quality isn't enforced, which breeds mistrust and a feeling of unfairness among the parties. "Can I trust my sworn enemies' data", "I contribute a ton and some freeloaders get it all in exchange for nothing", and so on. Until data-sharing is mandatory, and until privacy concerns are solved, all this data sharing remains just a feel-good exercise. BIC and ClinVar have an additional function too. There, one can report how a lab such as Myriad classified this or that variant, so it is no longer a "secret". This way, Myriad advantage gradually (very gradually!) erodes, but alas, it doesn't accelerate reclassification of presently uncertain variants. For that, only data sharing and pooling would help.

cymom

Wonderful!  Thank you for this. 

It is an awfully tangled web. The kicker seems to be that though patients are paying for this test for it's clinical value, the testing labs are getting something extremely valuable in return: exclusive data.  How many people understand what they are giving away when they are consenting to the test?  Did they really consent to having their genetic information used for exclusive commercial purposes?  As a patient, the use of my genetic data for profit makes me more much more uncomfortable that the privacy loss I might risk by sharing my information anonymously, with pedigree, in a public database.  Seems to me that especially in the case of uncertain variants, a standard disclosure and consent form is needed that gives patients the right to make decisions about sharing information.

Mutd

The patient data are used to improve medical care (in aggregate form, as data from multiple patients, but of course without disclosure of personal health identifiers) across the board in medicine / public health. Currents stats on infectious diseases to plan vaccination and prevention (even though you paid for the office visit). Data on drug adverse reaction to make sure that it stays safe (even after you paid for the prescription). And so on. 

The data is in the system because of the way the system works: your Dx, your address and SSN, your test results must flow between the physician, the lab, and the insurer in order for the insurance to approve, and to pay for, your care. But sharing any of these data pieces is forbidden. If you sign up for, say, free followup testing from Myriad, then it comes with a separate consent document for very limited disclosure (maybe your name, because someone has to tell to the family members who signed them up, and the variant name, because they need to know what is being tested, but nothing more than that).

Likewise, if a university research project pays for your care, then you may need to disclose something to them, but you can keep the lab in the dark now (no insurance, no disclosure necessary). Or if you pay for your own test then you may need to share your health data with the doctor who orders the test and receives the test results, but then you don't have to tell Myriads and Ambrys of this world anything at all.

Just remember that the primary use of the patient data is always the same, to improve the clinical outcomes and to make patient care better for everyone. Of course it translates to a commercial value, because a better product sells better. But the goal is always to improve the product, not to jack up the sales in some dubious way.

cymom

So Myriad can share this information in aggregate?  Has there ever been a situation like this, where a lab can legally share information in aggregate but has a financial incentive (and a legal obligation to its stockholders) not to?    

I worry about these rare variants.  How many of them are there?  On how many different genes?  If multiple labs offer initial testing and then offer free testing for family members when they find a variant, then you could easily see a situation where families with rare mutations land in different labs where they can't do any good for each other.  For rare variants, the companies may never reach a conclusion.

Mutd

cymom - did I get it right that you're in the process of being tested at Myriad, but your mother hasn't been tested? And that so far, there hasn't been an uncertain significance variant result in your family? If true, then it may be too early to worry about these variants, because the chances are quite small, something on the order of 2%. It's a pretty high certainty test.

Just to compare with the kind of a futuristic test I got (for a constellation of cancers in my family which did not look like any well-known syndrome). It was for all 25,000 genes there are, and there were thousands of uncertain variants in my report. Just the subsection on the cream-o-cream variants about which something suspicious has been known ran like 30 PDF pages long. There are TONS of poorly-known genetic variation out there, everywhere, and the special case of the BRCA1/2 genes at Myriad is a remarkable exception.

Of course most of the variants are rare. Many are unique to one person (every child is born with thousands of novel DNA changes, not present in either parent, and if these variants aren't lethal or highly disabled, then they may get passed to one child, and maybe one grandchild, but you just may not find them anywhere else). Fragmentation of the data between labs is a huge deal indeed, as it really robs the geneticists of their chances to classify variants. It isn't such a raw deal, yet, for the BRCA genes in the US, because most of the data is still kept in one database at Myriad. But in Canada, UK, or France, BRCA testing is done by a hodgepodge of regional and local labs, and although they started sharing "some" data within national single-payer health systems, it's kind of too little too late. Or in the US in the field of hereditary colon cancer, those genes are patented too, but licensed out non-exclusively, and therefore there are dozens of labs, big and small, and the data is split between multiple providers all the time. I mean competition in genetic testing may not be all bad news ... without some competition, innovation may slow down even more, as if it wasn't terribly slow already in the medical industry with all the mammoth regulation, stone-age physicians, and cozy relations between insurers and providers. But fragmented data is a real concern, and the American College of Medical Geneticists (which provides guidelines for the variant work) takes a totally defeatist position with respect to rare genetic variation, signaling that it'd be perfectly OK if these variants remain uncertain forever.