What happens in 2015 when the essential Myriad patents on BRCA 1 and 2 expire?
Think about it a while. What are you imagining? Lower costs, better tests, quicker turn-around times? A single panel covering all breast cancer susceptibility genes? An end to unhappy conversations with patients trying to explain why the process can’t be simpler and cheaper?
While you mull over your utopian fantasies of a patent-free universe, keep this in mind: Myriad has been thinking about it for a while now. After all, BRACAnalysis remains close to 90% of their total revenue, so it’s probably on the minds of Myriad executives. It certainly was on their minds at Goldman Sachs when they listed Myriad Genetics as a sell in February 2011.
Does Myriad have a plan? It sure looks like it. In 2006, Myriad ceased to publish BRCA variant information, and ended their relationship with the open-access Breast Cancer Information Core (BIC) database. Since then, they have assembled a private repository of genotype-phenotype information that seems to be a central component of its strategy for future earnings. How central? Well, Dan Vorhaus et al at the Genomics Law Report speculated back in 2011 that a strategy of relying on their “vast—and currently proprietary—database of BRCA test data, including VUS data” was behind the company’s decision not to even bother fighting patent infringement in Europe, citing Myriad CEO Peter Meldrum’s emphasis on “other competitive factors” as an alternate strategy for competitive advantage.
Forget about the ACLU lawsuit, and next-gen sequencing and all the other changes you have read about that may affect genetic testing going forward. Get the DNA sequence data wherever you want, and however you want, and as cheaply as you can, but as long as Myriad has sole control of the information needed to provide analysis, no other company will be able to challenge them on a competitive basis.
Is this fair? Well, it is an end run around the philosophical basis of patent protection, which is meant to provide a 20 year window of unfettered commercial use in return for a free and open sharing of information to stimulate further innovation. But it is not illegal. Myriad controls their database, and can’t be compelled to share. They own the database – but NOT the information. The information is out there – in report after report after report, languishing in the files of thousands of clinical cancer specialists. In other words, YOU HAVE IT.
So now, some exciting news. Dr. Robert Nussbaum at UCSF is spearheading an effort to collect BRCA 1 and 2 variant data in ClinVar, an accessible archive of anonymized genotype/phenotype information hosted by the National Center for Biotechnology Information (NCBI). While the goals of ClinVar are very broad – to aggregate information about sequence variation and its relationship to human health – Dr. Nussbaum’s goal are quite specific: to assemble a list of BRCA 1 and 2 variants found since 2005, along with information classifying them as benign, pathogenic, or unknown.
But THIS ONLY WORKS AS A COLLABORATIVE EFFORT. Dr. Nussbaum has contacted 600 clinicians involved in clinical care of HBOC patients (so far, 26 centers have contributed over 3000 BRCA 1/2 variants). With their cooperation and yours, ClinVar could amass a database to rival that of Myriad, ushering in an era of genuine access to unrestricted, competitively priced information for our patients. How great is that?
To get involved, contact Dawn Lee, a genetic counselor at Partners Center for Personalized Genetic Medicine who is working with Dr. Nussbaum. Here is her contact information:
You can get all the important specifics from Dawn, but for those of you who are interested, I’ve made a stab at some FAQ’s:
WHAT EXACTLY IS BEING COLLECTED?
This project is limited to collecting information on the variant, identifying it using cDNA and/or genomic numbering, and its classification in a 3-tier scale as benign, pathogenic/deleterious or unknown (some reports use a 5-tier scale including possibly benign and possibly pathogenic, which is also good). The goal is to capture each variant one time per family.
IS THIS OK? ISN’T IT A HIPAA VIOLATION?
Great question! HIPAA does not place any restrictions on the disclosure of information that is de-identified. For this reason, no names or other identifiers will be collected, including familial information or the name of the facility where the patient was seen. Does this mean that the clinician who orders the test has the right to use it in this way? “Ownership of Information’ issues are governed by state laws, and you can check out your state regulations in this 50-state survey of state laws governing the collection, storage and use of human tissue specimens by the National Cancer Institute but – spoiler alert – Dr. Nussbaum thinks the answer is yes, in all states.
DO WE NEED IRB APPROVAL?
Poor IRB’s! Everybody hates them so much. Don’t you think that is hurtful to their feelings? Sure, lots of people are doing this data collection without IRB approval. Those people are following Federal Regulation 46.101(4) from the Office for Human Research Protections, which specifies as exempt: “Research involving the collection or study of existing data,documents, records, pathological specimens, or diagnostic specimens, if these sources are publicly available or if the information is recorded by the investigator in such a manner that subjects cannot be identified, directly or through identifiers linked to the subjects.” But does anyone ask the IRB how this makes them feel?
IS THIS GOING TO BE A HUGE PAIN IN THE ASS?
First of all, if you have the reports in electronic form, it should be pretty straightforward. If you have them on paper, you have to make de-identified copies, which means masking the names in two places (Dr. Nussbaum suggests post-it notes). Dawn reports that there is a small stipend available for paying someone (contact her) – I suggest genetic counseling students (if you are in the NY area, contact me). And – wait this is exciting! – the next 20 centers to provide >200 variants will receive an Ipad mini.
WHY NOT COLLECT MORE PHENOTYPIC INFORMATION?
Why indeed? Why not report age of onset or bilaterality? And if you are going to do that, you might as well check for hormone status. It’s like that child’s book, “If You Give a Mouse a Cookie.” If you give a researcher laterality, he is going to want oncogene status. If you give her oncogene status, she is going to want response to treatment data. The limited goals of this project make it easier for more people to participate (see IS THIS GOING TO BE A HUGE PAIN IN THE ASS?, above), and thus to advance the primary goal.