Association for Molecular Pathology vs. Myriad Genetics – Arti Rai

Association for Molecular Pathology vs. Myriad Genetics – Arti Rai


Arti Rai:
I will try to give you a relatively engaging and short discussion so that we can have some
time for questions. I will also, I’m afraid, use a few more words in my slides than you
may be accustomed to. Lawyers — and I’m a recovering lawyer — do like words, and so
there are more words than would be ideal. But the words end up being kind of important. So, as Eric has already indicated, the Supreme
Court, on June 13th, having heard argument on April 15th, decided that AMP versus Myriad
Genetics, the challenge was brought by the Association of Molecular Pathologists, represented
by the ACLU and the Public Patent Foundation. In its ruling, the court, the Supreme Court,
that is, held that not all gene patents are alike, and I’ll spend some time talking about
what specifically they meant by not all gene patents being alike. As it happens, they got
into the — for them, I think, what was really tough sledding with respect to molecular biology,
they had to distinguish between cDNA and gDNA, which I suspect for Justice Thomas was something
of a struggle. But he did an interesting job. So what I’m going to do today, for those of
you that don’t necessarily spend their time thinking a lot about patent law, and that
may be many of you, is provide a little bit of basic background on patent law, and also
on gene patents in particular, and how they emerged and why they emerged. Then I’ll discuss
the history of the Myriad case very briefly, and also, its ultimate resolution at the Supreme
Court. And finally, I’ll discuss what the case might mean going forward. That’s very
much a work in progress in terms of what it means, and there are subsequent lawsuits that
I will mention briefly as well. So, just in terms of background, this is Patents
101, and also, as it happens to be, Section 101 of the Patent Statute. That section has
typically been interpreted very broadly to cover basically anything under the sun made
by man. That’s a statement from the Legislative History of the 1952 Patent Act. And the provision
in the statute suggests as much — suggests that it’s pretty broad in its scope, so any
new and useful process machine, manufacturer, or composition of matter is supposed to be
encompassed within Section 101, which means, basically, any new process or product. That
said, there have been these long-standing common law exceptions that the Supreme Court
has enunciated, starting in the 19th century, and by “common law,” I mean as contrasted
with statutory law. Common law is the gloss that courts put on the statutory law. And this common law gloss encompasses exceptions
in the form of abstract ideas, laws of nature, and products of nature. The exception at issue,
specifically in the Myriad case, was products of nature. However, as we will see, all of
these three exceptions kind of run together a little bit, and that becomes important for
what the case means going forward. So the product of nature doctrine actually
has an interesting history beyond the Supreme Court, and in most ways, the most important
expositor of the product of nature doctrine was a judge by the name of Learned Hand — that
was really his name. Judge Hand, albeit only a district court judge at the time, rendered
what is one of the most important decisions in the product of nature space in a case called
Parke-Davis that was decided in 1911. In that case, he determined that isolated or purified
adrenaline represented patent-eligible subject matter, and in his view, this was not a product
of nature — isolated adrenaline represented one patent, purified adrenaline represented
another — this was not a product of nature because it was, and this was a quote from
the opinion, “For every practical purpose a new thing commercially and therapeutically,”
unquote. Because this language resonated with some of the commercial instincts that patent
lawyers have, they really have seized upon this language for more than a century as representing
what they determine to be the narrow exception that product of nature represents; in other
words, anything that is a new thing, commercially or therapeutically, for many in the patent
lawyer community, is not a product of nature. This view of the expansive realm of what is
within patentable subject matter, and therefore not a product of nature, was, to some extent,
reaffirmed in the patent lawyer community in the Supreme Court case of Diamond vs. Chakrabarty
that some of you may have heard of. That case involved a recombinant bacterium, a bacterium
that had two or more stable energy-generating plasmids put into it, each plasmid constituting
a separate hydrocarbon degredative pathway. In that case, the Supreme Court, in a divided
opinion, determined that this was not a product of nature because “It was markedly different,”
quote, unquote — again, these are the words that the court used — from anything found
in nature. Not quite as expansive as the Parke-Davis opinion which suggested that if what one had
done provided something commercially useful, it represented patent-eligible subject manner.
So Diamond vs. Chakrabarty was not quite as expansive, but the combination was taken by
patent lawyers and the nascent biotech industries to mean that basically anything that was isolated
or purified could be patent eligible, and that included gene sequences. So the first “gene patents,” quote, unquote
— and that is, to some extent, now a term of art because they mean different things
to different people, that term means different things to different people — covered cDNA,
in other words, DNA with introns excised that was intended to be used to generate therapeutics
and cover therapeutics. One of the reasons, in part at least, these particular cDNA patents
were not controversial was that they were intended to cover a scope of genetic research
and development, an area of genetic research and development, i.e. therapeutics, that looked
very much like what had been patent eligible in the past in terms of small molecules. So
these were just large molecules as opposed to small molecules. And so when, for example, when some of these
patents began to issue in the early ’80s, and Amgen got one of these classic patents
on DNA sequences encoding erythropoietin, cDNA encoding erythropoietin in 1987, a patent
that subsequently allowed it to make tens of billions of dollars in revenue over the
course of about 23 years, that wasn’t considered particularly controversial, at least certainly
not in the patent bar community, and even among the larger community, the idea of DNA
sequences that would be patentable but would correlate therapeutics wasn’t particularly
controversial. And this patent was a subject of litigation; however, none of the litigation
involved the issue of whether or not this constituted patent-eligible subject matter,
probably in part because everyone in the litigation had their own gene patents. So Amgen sued
Genetics Institute, and Transkaryotic, and Roche, and, of course, all of those entities
had their own gene patents to defend, and so there was no incentive on the part of anyone
to say that gene patents were not patent-eligible subject matter. Controversies began to emerge however when
patents issued — and some of these patents issued relatively early on, including the
Myriad patents — that were later interpreted to cover not just therapeutics but also diagnostics.
Obviously, like all patents, patents that covered diagnostics increase costs, or have
the potential to increase costs, and restrict access. The argument, or one of the reasons
people were more concerned about patents covering diagnostics was, in particular, with the respect
to laboratory-developed testing of the sort that Myriad does. That’s not currently FDA
regulated, so patents were seen as less necessary for purposes of providing incentive to get
a particular product to market. If one had the relevant gene, it wasn’t considered that
expensive to then start doing testing on the gene, and in point of fact, in a lot of the
cases, testing started before the patents emerged. And the patents ended up being used
to shut down certain testing — diagnostic testing laboratories. In addition there was a substantial amount
of federal funding involved with some of these patents, including in the Myriad case itself,
and we’ll talk about that a little bit more — I’ll talk about that a little bit more
towards the end. So the access questions in particular, relative
to the diminished need for an affirmative incentive provided by patents, were the focus
of this very prominent report issued by the Secretary’s Advisory Committee on Genetics,
Health, and Society, and our own Jim Evans was at the helm of that particular report.
The focus there was not only on initial access but also on problems for women who wanted
to get second opinion testing, with the respect to BRCA-1 and BRCA-2. The Secretary’s Advisory
Committee was — relatedly was concerned about whether sole providers of genetic diagnostic
testing, such as Myriad, but also other providers, like Athena, had the optimal incentive to
work aggressively with all types of insurers, in particular insurers that might cover less
well off populations, such as state Medicaid providers. So the report discussed not only the BRCA-1
and BRCA-2 patents, but also patents on genes implicated in certain types of colon cancer,
Alzheimer’s disease, spinocerebellar ataxia, and long QT syndrome. This report was issued
in 2010 after Myriad had been sued by the Association of Molecular Pathologists, represented
by the ACLU in district court. So it came out as this debate was, for the first time,
the Section 101 debate, was, for the first time, unfolding in litigation. Again, never
before had a Section 101 challenge been brought to gene patents of any sort. So we get to the Myriad gene patent litigation.
Myriad, I think, was chosen by the ACLU for a number of reasons, one of which was it was
the provider that was most aggressive in terms of asserting its patents against providers
who considered themselves to be doing clinical research in addition to providing patient
care. In this particular case, the ACLU/AMP challenged 15 claims in seven patents. Many
of these patents, at it happens, were initially owned and exclusively licensed by the University
of Utah, which had received a fair amount of NIH funding, particularly early on. NIH
was actually the co-owner of one of the patents. So this is — the federal funding issue was
intimately involved with this particular case. The litigation, notably, however, the litigation
strategy by the ACLU didn’t focus on access at all. And I think the ACLU determined that
they were going to kind of talk the typical patent lawyer’s line of access — not access,
but innovation. And so the arguments that were used very much were along the lines of
these patents were diminishing innovation in the area of genomic research and development. In point of fact, at the federal circuit,
Judge Lourie stated what I think all the judges in this litigation agreed with, at least implicitly,
by not saying anything about access. He’s stated explicitly, and he wrote the majority
opinion, “This appeal is not about whether individuals suspected of having an increase
risk of developing breast cancer are entitled to a second opinion.” Similarly, even the
dissent in the federal circuit case, before it got to the Supreme Court, really focused
on innovation, and the dissent drew a distinction between cDNA and gDNA that had been alluded
to in the Secretary’s Advisory Committee report, but for the purposes of saying that cDNA patents
wouldn’t impede research and development in genome sequencing, whereas gDNA patents might.
So, again, very much focused on innovation. The Department of Justice’s intervention in
this case at the federal circuit level also drew the cDNA/gDNA distinction with a focus
on follow-on innovation, and Eric has mentioned that NIH was influential in crafting the U.S.
government’s opinion. I think that’s an understatement in many ways; NIH was a critical player, and
NIH, as it happens, has had a long history in helping to shape genomic patent policy
starting in the late — the end of the 20th century with the so-called utility and written
description guidelines that were very significantly shaped by NIH, and I think had a very positive
influence in terms of making sure that genomic innovation was not impeded by patents. All
of this, by the way, is drawn out in exhaustive detail in many more words in a Duke Law Journal
article; the role of NIH over the last 15 years or so is drawn out in exhaustive detail
in this Duke Law Journal that were published last year, so if you’re interested, you can
go to that. But, again, the focus on innovation, and the fact that cDNA patents are unlikely
to impede innovation whereas gDNA patents might well impede whole genome sequencing. At the Supreme Court only claims at issue
were the so-called product claims; the process claims had been taken out of the litigation,
so there were only nine composition of matter or product claims at issue. The Supreme Court,
like Judge Bryson in dissent below, like the solicitor general, and like Eric Lander, who’s
also been previously mentioned, adopted a cDNA versus gDNA distinction. Unfortunately,
the court’s opinion was not as clear as to why it was adopting this distinction as it
might have been. It did say that cDNA was not naturally occurring, and thereby implicitly
suggesting that gDNA could be naturally occurring, but it didn’t really specifically say that,
and we can only infer based upon the fact that Eric Lander’s brief was mentioned about
six times in oral argument, that they relied upon Eric Lander’s brief’s argument about
gDNA being found — isolated gDNA being found in fetal DNA and also upon cell death. So,
unfortunately, though, the opinion doesn’t explicitly say that, so that a little bit
hard to parse on the why gDNA is naturally occurring versus cDNA is not. Another piece of the opinion focuses — and
this important going forward for purposes of what the opinion means for future patenting,
and also existing patents outside the cDNA/gDNA context — focuses on the so-called information
versus chemical distinction. This was a distinction that was drawn by the district court judge
in this case, and the Supreme Court picked it up again. Again, however, it didn’t completely
connect the dots as to why cDNA versus gDNA mapped onto that distinction. In point of
fact, the court even suggested — the Supreme Court, that is — that cDNA and gDNA both
covered information suggesting that, however, that, nonetheless, cDNA was less problematic
in terms of the informational claims it made. All we can assume, and this is what I would
assume and have argued should be the interpretation of the case, is that Justice Thomas had been
impressed by Eric Lander and the solicitor general, and their statements that cDNA, even
though it claimed information or could be seen as claiming information, didn’t — those
sorts of things wouldn’t impede future follow-on innovation in the same way as gDNA. And Bob
Cook-Deegan and I put that interpretation on the case in a Science article that just
came out a couple of months ago, and so we’re hoping that the lower courts will take that
as a plausible interpretation that said the Supreme Court unfortunately wasn’t as clear
as it could have been. So in the immediate aftermath of the case,
I suspect some of you know this already, on the day of the case being decided itself,
a whole bunch of providers, a whole array of providers, announced that they would be
testing for BRCA-1 and BRCA-2 mutations. Ambry in particular announced a price that was about
a 50 percent reduction from the Myriad price, and perhaps not surprisingly, Ambry was the
first provider that was sued. So Myriad, as it happens, had been alleging that it had
all these other patents it could use even if they lost the case to the Supreme Court,
and nobody necessarily took them that seriously, but it turned out they were planning to sue
on those other patents if they lost to the Supreme Court, and they did, in fact, sue
on 10 other patents involving dozens of claims. None of these claims, with the exception on
one that we can talk about, was at issue in the Supreme Court case. Both these suits — they
also brought a suit against Gene-by-Gene — both these suits, for them, have a bit — for Myriad,
have a bit of a home-court advantage in that they’re being brought in Utah district court,
and Myriad is obviously based in Utah. So assessing the suits, I won’t say too much
here for fear of being proven wrong in the next few months when the Supreme — the district
court decides the cases, but these claims all cover basically either methods for sequence
amplification, and then sequencing, and then comparison with wild-type, a lot of method
claims, and then a few primers plus PCR claims. The determination of how these claims will
be adjudicated, I think, is anyone’s guess, but I think that what the court will have
to do — the district court at the initial level, and obviously, this case will go to
the federal circuit which hears all appeals in the patent cases, and then maybe even back
to the Supreme Court — the determination of what the court — district court will do,
I think, will be based on a combination of the Myriad case, and then also a case — another
Section 101 case the Supreme Court decided last year, called Mayo versus Prometheus. Some would say that the combination of those
two cases means that these claims that Myriad is bringing are likely invalid because they
don’t require any real inventive activity — that’s the term of art that was used in
the Mayo case — beyond a law of nature or a product of nature. That said, one could
argue, and Bob and I make this point in our Science article, that Myriad could be seen
as walking back Mayo versus Prometheus a little bit because cDNA, arguably, is nothing more
than non-inventive routine activity applied to a product of nature, i.e. DNA. So whether
the lower courts will — how they’ll interpret the combination of Myriad and Mayo versus
Prometheus, I think, is going to be very interesting to see. And the key question will be, is inventive
activity beyond a law or product of nature always required, and I think the Supreme Court
has, frankly, left that question a little bit open. Ambry has counterclaimed. Frankly, I think
that its counterclaims on validity are quite plausible. It also argues a variety of anti-trust
violations by Myriad. I think those allegations are less clear to me, at least, or the basis
for those allegations is less clear to me. They use Section two of the Sherman Act and
talk about monopolization, however, patents, by definition in some cases, represent legitimate
monopolies. So I’m not sure that the anti-trust claims, based upon the patents themselves,
are necessarily valid. Another piece of the puzzle which is very interesting and important
is all of this data that Myriad has, mostly secret since 2006, however — and I’ll talk
about that in a moment — however, the Ambry brief doesn’t specifically link that data
issue with the anti-trust question, which I think would be a very interesting argument
to make, and presumably, they will, in further briefing on that question, draw that link.
The data, I think at this stage, are arguably more important than the soon-to-expire patents,
these particular patents, even the ones that Myriad is suing upon are going to expire in
the next few years. So the data is going to be the competitive advantage that Myriad has
going forward, at least for the longer term. And that brings us to that data issue. Myriad
itself, in its argumentation against Ambry and Gene-by-Gene, is actually using the data
as a bit of a club to argue for a permanent — a preliminary, and then a permanent injunction
against Ambry and Gene-by-Gene, saying that Ambry is likely to have a variants of unknown
significance rate of 25 to 30 percent, much worse than Myriad’s 3 percent rate, and therefore,
it would actually harm the public to have Ambry and other providers testing. As you
probably know, Myriad, however, has stopped contributing to public databases, it stopped
contributing to the Breast Cancer Information Core in 2005, and more generally, the larger
scientific community hasn’t had access to the Myriad databases since 2006. So one question is, what will happen in respect
to this data? And there are efforts already afoot, and our own Robert Nussbaum is intimately
involved with one of them, to quote, unquote, “free the data.” So the Sharing Clinical Reports
Project out of UCSF, working with the Genetic Alliance, is asking patients and clinicians
to submit to de-identified data that has been received from Myriad after Myriad’s testing
to create an alternative to the Myriad — to create additional alternatives to the Myriad
database. So we’ll see how that goes. Larger impacts. So Greg Graff did an article
in Nature Biotechnology a few months ago that suggest that 8,700 U.S. gene patents with
quote, unquote “naturally occurring sequences” are still in force, about 41 percent of those
human. Unfortunately, Greg, it’s a great article, and I would commend it to your attention,
but his definition of naturally occurring doesn’t really map on to cDNA and gDNA distinction,
so we don’t really know — and I’ve been in communication with him about this — what
percentage of the 8,700 are gDNAs that would be knocked out and what percentage of cDNAs
that remain valid. However, he does make the important point in the article that the percentage
of both cDNA and gDNA patents relative to entirely synthetic patents began to decline
precipitously around the time, I should note, that NIH became heavily involved in persuading
the PTO to issue much stricter examination requirements for gene patents.
This is not the remit of this particular group, but in terms of the larger patent bar, I think
that the biggest concern is not over gene patents but over what the statements in Myriad,
and also the Mayo case, mean for quote, unquote “purified” large and small molecule therapeutics
because that is an area where patent protection, I think, the conventional wisdom has been
patent protection is extremely important, and the patent bar is particularly concerned
about claims to proteins and antibodies that aren’t explicitly designated as synthetic.
Going forward, I think most patent attorneys will designate every patent claim that they
can possibly think of as synthetic, but retroactively, those claims that are not quote, unquote “synthetic”
may be in peril. So, that’s it. Any question either now, or
you can email me at [email protected] Eric Green:
Questions? Please step to the microphone, anybody in back just step to a microphone,
turn the microphone on. Female Speaker:
[inaudible] Eric Green:
Wait, wait, wait — go to — they won’t hear you unless you are at a microphone and the
microphone is on. Female Speaker:
[inaudible] Ohio University. We recently had a debate about this in a journal club, and
I — the point you mentioned about — a cDNA is really just a copy of a naturally-occurring
molecule. It’s just a common technique. Even if you could use it for a therapeutic — if
you don’t change it, okay, and you found it from a naturally-occurring messenger RNA or
other non-coding RNA, it is still just a copy; there is nothing novel about it. So I would
argue that the only way a cDNA could be patentable is if they’ve modified it in a way for a unique
purpose, for a unique application. I don’t even know if you identify an mRNA, clone it
into a cDNA, and it has a novel mutation for a disease. That’s still a naturally-occurring
molecule that was in a tumor, or whatever disease state the person was in. So I really
am having a hard time wrapping myself around what the — why there is such a big misunderstanding
between the genomic versus the cDNA, as opposed to novel manipulations or natural mutations,
maybe, that you can patent. It really — so I’m glad you hit on that point, and I’m glad
you — I know that somebody is out there saying, what a minute, it’s really still a natural
product, the cDNA, if you don’t manipulate it. All you’re doing is copying and putting
it into a vector so you can sequence it. Arti Rai:
So I think that the cDNA/gDNA distinction has been one that is not universally subscribed
to in terms of the doctrine, that it’s not entirely clear to a lot of people that, as
a matter of doctrine, it makes a huge amount of sense. That said, as I indicated, the cDNA
patents are considered much more commercially valuable, and I think that there’s a certain
element of economic policy also in patent law that is, you know, inevitably apart of
patent law. Male Speaker:
So could you clarify, when you are talking about cDNA patents, are you talking specifically
about Myriad cDNA patents, or the set of cDNA patents that — because there are a lot of
cDNA patents, right? Arti Rai:
The set of cDNA patents are — those are all still valid. Male Speaker:
Right, but okay — Arti Rai:
Yeah. So, yeah, absolutely — sorry, follow up. Male Speaker:
No, no, so my question was, is your feeling that, you know, because there are so many
cDNA patents out there, outside of this — the ones that Myriad has, that the Supreme Court
didn’t want to take them all down at the same time that it was taking down — you know,
by restricting it to the gDNA and leaving the cDNAs open, it was a sort of Solomonic
decision, or whatever — Arti Rai:
Yeah, it was I think it was a Solomonic decision, I’m not sure they fully understood all the
implications. But I think it was clear to them that drawing that distinction would leave
the biotech industry largely satisfied, and also satisfy the concerns of researchers about
gDNA claims. So it was a way to try to balance the interests of the commercial community
and research — the research community. Male Speaker:
I wonder if you could comment a little bit further on this distinction of diagnostic
versus therapeutic, especially in cases of oncology where that line is getting blurred,
or in rare diseases where oligonucleotides and others are — actually, the diagnostic
is intrinsically linked to the therapeutic — Arti Rai:
Yes, so I think it is going to be very interesting because, yes, certainly with companion diagnostics,
I think it’s going to — there’s going to be pressure brought to bear on that distinction,
because with companion diagnostics in particular, some of those patents could be quite commercially
important. And — so query how that will all work out. I think some people are rightly
concerned about that question. Male Speaker:
Arti, I was wondering, do you have any sense of what Ambry, GeneDx, and the others, what
their response has been since the suits? I mean, specifically, are they still offering
the tests, and if any of them are still offering it, do you — you know, are you aware of any
reports of sort of the quality of the annotation going on since they don’t have access to the
Myriad databases? Arti Rai:
I believe they are still offering the tests, I do not know about — I know they dispute
the Myriad numbers. I don’t know specifically about — as I understand it, they put their
VUS rates significantly lower than the Myriad numbers. And, in fact, one of the briefs argued
that Myriad was unfairly maligning their tests, and that was part of the monopolization claim,
although I’m not entirely clear as to how that fit into the monopolization question. Male Speaker:
I can confirm that they’re still offering the tests. Many of us are sending it to Myriad’s
competitors. I’d be interested in what Bob Nussbaum might think. You know, that claim
of 20 to 30 percent VUS is, I think — my perception is, is grossly inflated. Certainly,
what Ambry has on their website is a estimate that they’ll have — I think it’s around a
5 percent VUS rate, which seems more realistic to me, but I was wondering what Bob thought. Male Speaker:
Well, there’s a paper published in 2009, first author Easton, which includes four authors,
I believe, from Myriad that have a lower VUS rate than they’re claiming now Ambry would
have, and based on public databases back in 2009. So although Myriad may have improved
their VUS rate, it is sort of hard for anyone to know because it is a closed shop. But I
think that it’s hard for me to imagine why the VUS rate would have jumped in the last
four years. [laughter] So that’s kind of my feeling about it. Arti Rai:
Thank you. Eric Green:
Any other — thank you, Arti. That was terrific. So we’re breaking? Rudy Pozzatti:
We are breaking for lunch for the rest of the audience, but for the council members,
we are going to have your photograph taken. So if you could please stay at the table,
the photographer will come in here — I think we have a room reserved for the picture. And
then after that, you can go to lunch, and can we please come back at 1:15 to resume.
Okay.

1 thought on “Association for Molecular Pathology vs. Myriad Genetics – Arti Rai”

Leave a Reply

Your email address will not be published. Required fields are marked *