[Senate Hearing 110-770]
[From the U.S. Government Publishing Office]
S. Hrg. 110-770
ROUNDTABLE DISCUSSION:
REGULATORY, SCIENTIFIC AND ETHICAL ISSUES RELATING TO GENETIC TESTING
=======================================================================
ROUNDTABLE DISCUSSION
before the
SPECIAL COMMITTEE ON AGING
UNITED STATES SENATE
ONE HUNDRED TENTH CONGRESS
SECOND SESSION
__________
WASHINGTON, DC
__________
JUNE 12, 2008
__________
Serial No. 110-30
Printed for the use of the Special Committee on Aging
Available via the World Wide Web: http://www.gpoaccess.gov/congress/
index.html
U.S. GOVERNMENT PRINTING OFFICE
47-953 PDF WASHINGTON : 2009
----------------------------------------------------------------------
For sale by the Superintendent of Documents, U.S. Government Printing
Office Internet: bookstore.gpo.gov Phone: toll free (866) 512-1800; DC
area (202) 512-1800 Fax: (202) 512-2104 Mail: Stop IDCC,
Washington, DC 20402-0001
SPECIAL COMMITTEE ON AGING
HERB KOHL, Wisconsin, Chairman
RON WYDEN, Oregon GORDON H. SMITH, Oregon
BLANCHE L. LINCOLN, Arkansas RICHARD SHELBY, Alabama
EVAN BAYH, Indiana SUSAN COLLINS, Maine
THOMAS R. CARPER, Delaware MEL MARTINEZ, Florida
BILL NELSON, Florida LARRY E. CRAIG, Idaho
HILLARY RODHAM CLINTON, New York ELIZABETH DOLE, North Carolina
KEN SALAZAR, Colorado NORM COLEMAN, Minnesota
ROBERT P. CASEY, Jr., Pennsylvania DAVID VITTER, Louisiana
CLAIRE McCASKILL, Missouri BOB CORKER, Tennessee
SHELDON WHITEHOUSE, Rhode Island ARLEN SPECTER, Pennsylvania
Debra Whitman, Majority Staff Director
Catherine Finley, Ranking Member Staff Director
(ii)
C O N T E N T S
----------
Page
Opening Statement of Senator Gordon H. Smith..................... 1
Panel of witnesses
Statement of Linda Avey, Co-founder, 23andme..................... 3
Statement of Thomas Hamilton, Director, Survey and Certification
Group, Centers for Medicare and Medicaid Services.............. 4
Statement of Steven I. Gutman, M.D., Director, Office of in Vitro
Diagnostic Device and Radiological Health, Food and Drug
Administration................................................. 5
Statement of Kathy Hudson, Ph.D., Director, Genetics and Public
Policy Center, Johns Hopkins University........................ 7
Statement of Judy Yost, Director, Division of Laboratory Services
and Clia Program, Centers for Medicare and Medicaid Services... 7
Statement of Elaine Lyon, Ph.D., Associate Professor of
Pathology, University of Utah School of Medicine, and Medical
Doctor, Molecular Genetics, Arup Laboratories.................. 8
Statement of Matthew Daynard, Senior Attorney, Division of
Advertising Practices, Bureau of Consumer Protection, Federal
Trade Commission............................................... 9
Statement of W. Gregory Feero, M.D., Ph.D., Senior Advisor to the
Director of Genomic Medicine, National Human Genome Research
Institute, National Institute of Medicine...................... 12
APPENDIX
Written Comment Submitted by Elaine Lyon, PhD, Medical Director
of Molecular Genetics ARUP Laboratories on Genetic Testing..... 41
(iii)
ROUNDTABLE DISCUSSION: REGULATORY, SCIENTIFIC AND ETHICAL ISSUES
RELATING TO GENETIC TESTING
---------- --
THURSDAY, JUNE 12, 2008
U.S. Senate
Special Committee on Aging
Washington, DC.
The Roundtable was Commenced at 2:16 P.M., in room G-11,
Dirksen Senate Office Building, Hon. Gordon H. Smith, Ranking
Member, presiding.
Present: Senator Smith.
Also Present: Christina M. Hinkle, Chief Investigative
Counsel.
OPENING STATEMENT OF SENATOR GORDON H. SMITH, RANKING MEMBER
Senator Smith. Welcome, everyone. We appreciate so much
your coming to this very important roundtable. We want it to be
somewhat informal, but make it helpful for this very important
topic. It's a topic that we had a hearing on in the Aging
Committee 2 years ago. It is obviously the issue of genetic
testing and direct-to-consumer sales and the impact that that
may have on consumers as they may or may not be exploited or
misled or defrauded.
My investigation in this area has revealed to me and to my
staff questionable clinical practices of laboratories
performing these tests, despite the fact that a number of these
labs purportedly were CLIA-certified. While at that hearing we
focused on a particular subset of DTC tests, the concerns
raised at the hearing apply across the board to genetic testing
and chiefly to these three areas:
No. 1, do consumers have adequate assurances of the safety
and accuracy and usefulness of genetic testing, be it the DTC
or physician-ordered tests?
No. 2, what protections exist for consumers' DNA, genetic
test results, and other sensitive information provided in the
course of genetic testing, particularly in the DTC arena?
No. 3, does CLIA provide adequate oversight standards for
genetic testing laboratories?
I have invited you all here as panelists today because from
what I have seen there remains much work for regulators and for
the Congress to do to protect consumers to ensure the privacy
and confidentiality of personal genetic information. CMS has
abandoned plans for a CLIA genetic testing specialty. To my
knowledge, the FTC has done little to pursue enforcement
actions, despite clear evidence of fraud in the marketplace.
Confusion abounds regarding the roles of CMS and FDA in
regulating certain aspects of genetic tests, especially DTC
tests.
For 2 years I have been sounding the alarm, calling for
more stringent oversight, a call that is echoed in the recent
report from the Secretary's Advisory Committee on Genetics,
Health, and Society. In the 2 years since this committee's 2006
hearing, genetic tests have continued to proliferate. According
to genetest.org, there are now available genetic tests for over
1500 diseases. This proliferation continues to fuel concerns
about the oversight of genetic testing and the protection of
consumers' health data.
I'm hopeful that today's panelists will address these and
other issues relating to genetic testing. Since we have a lot
of ground to cover in a short period of time, you'll have to
excuse me. I won't be here for the entire roundtable. My staff
will take over. I'll be here for just as long as I possibly
can, because this topic I think is very, very important and
it's a subject that is growing.
Our panelists today include. Tom Hamilton and Judy Yost
from the Centers for Medicare and Medicaid Services. Tom serves
as Director of the Survey and Certification Group. Judy serves
as Director for the Division of Laboratory Services and the
CLIA program.
Steve Gutman and Catherine Cook are here from the Food and
Drug Administration. Steve serves as Director of the Office of
In Vitro Diagnostic Device Evaluation and Safety in the Center
for Devices and Radiological Health. Kate is the Acting Senior
Associate Center Director of the Center for Devices and
Radiological Health.
Also we have Greg Feero, who serves as the Senior Advisor
to the Director of Genomic Medicine at the National Human
Genome Research Institute at NIH.
Kathy Hudson is the Director of there Genetics and Public
Policy Center at Johns Hopkins University.
Elaine Lyon is here on behalf of the American Clinical
Laboratory Association. She serves as an Associate Professor of
Pathology at the University of Utah School of Medicine and is
the Medical Director of Molecular Genetics for ARUP
Laboratories.
Finally, we have Linda Avey, Co-founder of direct-to-
consumer testing company 23andMe.
Thank you all for coming. I want you to be at home,
relaxed. This is not an inquisition. This is a serious inquiry
by the U.S. Senate to find out where we are and what this
means, what this means to consumer protection, what this means
to people's privacy, and what holes there are in our regulatory
system, to make sure that we're doing right by the American
people with this proliferating product.
So why don't we start. Any particular order here?
Why don't we start with--when we speak of regulatory gaps--
and here we go to CMS--what seems to me is in some instances
consumers are basically signing away their rights to their own
DNA information, the test results and any scientific
discoveries from research on their DNA. I understand that
23andMe require consumers to consent to the use of their DNA
samples in unspecific research.
My question is what aspects of DTC testing are regulated,
where are the holes, and what protections exist for consumers.
I throw it open. Would you like to answer that one?
STATEMENT OF LINDA AVEY, CO-FOUNDER, 23andME
Ms. Avey. The whole idea of research is something that
we're very interested in at 23andMe. We actually just launched
a new extension of our platform that we call 23andWe, because
what we're hearing from our customers, who voluntarily sign up
to be part of 23andMe, is that they're very interested in being
active participants in research, which really doesn't happen
right now. When someone is presented with the opportunity to be
a subject in a research project, they're not necessarily given
the option to have access to their own data that is generated.
23andWe is an extension of that idea of being part of a
research project, again very voluntarily. When people sign up
for 23andMe, yes, we say that we're going to do a limited
amount of research with their genetic data. But until they give
us additional information about their phenotype, we can't do a
study with them anyway. So it's all voluntarily submitted.
From what we're hearing people are really eager to be part
of research projects where they can actually have a voice. Just
like we see with autism, there are still parents out there who
believe that their child was affected potentially by their
vaccines. Those are the kinds of things we want to open up and
give parents and consumers and patients, cancer survivors,
anyone who's interested, the opportunity of being a participant
in research.
Senator Smith. Do they know when they sign this over that
it might go out more broadly as to them specifically?
Ms. Avey. One of the promises we make to each and every
customer is that we will never sell an individual's data. We
will never do that or make it accessible. If they want to
download their data and give it to a researcher, they can do
that. That's within their rights to do that.
But our intention is to keep the information very secure
and private within 23andMe, and then when outside researches
approach us and propose queries of our database, our
bioinformatics experts internally will do that query for them.
But the data will never leave 23andMe, again unless a consumer
says, ``I want my data to go to my doctor or a researcher I
know at UCLA who's doing a study and I want to be part of it.''
So they own that decisionmaking process.
Senator Smith. You feel like they have sufficient control
over it, then?
Ms. Avey. Absolutely. That's the whole design of the
platform.
Senator Smith. Any of our regulators have a different
opinion on that? [No response.]
If you don't that's fine; that's good news.
At the 2006 hearing we heard testimony regarding companies
providing DTC nutrigenetic tests that indicate various risk
levels for developing cancer, Alzheimer's disease, kidney
disease, macular degeneration, rheumatoid arthritis, high blood
pressure, and heart disease. In one instance a company even
claimed to repair DNA, a claim that the company is still making
as of this morning, even though we heard testimony that this
claim is scientifically unfounded.
Tom and Steve, at the hearing you both expressed concerns
about these health-related claims being made by these
companies. Given that these companies, along with many new
entrants into the market, are still in business and still
making similar, if not identical, claims, you have to wonder
how concerned the agencies are about consumer safety. Any
response?
STATEMENT OF THOMAS HAMILTON, DIRECTOR, SURVEY AND
CERTIFICATION GROUP, CENTERS FOR MEDICARE AND MEDICAID SERVICES
Mr. Hamilton. I think when you look at direct-to-consumer
testing it is extremely broad-ranging, and I count seven or
eight dimensions. There's the advertising and-or education,
there's the sales, the test ordering, the analytical validity--
was the test accurately administered--clinical validity, and
clinical utility, the interpretation of results, and then the
communication of those results to consumers.
With regard to any test that involves the analysis of human
specimens for health care purposes, those would fall under CLIA
and would be examined with regard particularly to the
analytical validity. Whether or not they've engaged in false
advertising on the front end transcends CLIA and I think that
one of the things that is quite obvious about this all very
challenging but very exciting field of ever-growing genetic
testing is that it's much larger than any of us. I think any
one agency or entity that tries to address all the public
policy issues here immediately confronts what our children may
have learned as Miss Piggy's fifth law. Never try to eat
something that you cannot lift. No one agency has the heft to
bring into solution, a solution to all of these issues.
So that's why I think the right tune here is not a solo,
but a choir, if you will. If we can get all of our agencies and
public-private partnerships to be a tune in harmony, a choir in
harmony, then I think that would be ideal, and I think that's
why the groundbreaking hearing that you held, Senator Smith,
was so important, and your continued leadership in this arena,
because I think it takes someone such as yourself to bring
everybody together to move this agenda forward.
Senator Smith. Do you know, Tom, in the last 2 years has
there been any greater coordination between CMS, FDA, the FTC?
They're not at the table here, but we may have some FTC folks
in the audience who can participate in the roundtable.
Mr. Hamilton. I think there definitely has, and we've been
holding regular conference calls with CDC and the Food and Drug
Administration and have been particularly working with the
Centers for Disease Control. We've added a staff person at CMS,
acting on one of the GAO's recommendations, and she's sitting
behind me and that's Penny Keller, specifically focusing on
genetics, beefing up our capabilities to work in this area. She
is working with the Centers for Disease Control on a best
practices publication that will be coming out in their
morbidity and mortality weekly report.
We are working together with the Centers for Disease
Control to make a recommendation to the Clinical Improvement
Advisory Committee at their September meeting to form a work
group on proficiency testing to bring advances and more
proficiency tests, to make them more available and put them
into greater use.
We are convening a work group, a convocation amongst the
accrediting organizations in November to work on ways in which
we can promote the availability and the use of proficiency
tests.
Those are a few examples of things coming together. I think
Federal agencies are all pretty attentive to this now.
Senator Smith. The conference call you speak of, is that
the inter-agency task force that we spoke of at the hearing,
that there's going to be an inter-agency task force to tackle
this?
Mr. Hamilton. Well, there's multiple inter-agency groups
working on this. The proficiency testing would be not only
inter-agency amongst the Federal agencies, but involving the
accrediting organizations, someone from New York State since
they are pretty advanced in this arena, and public-private
organizations that can all collaborate to figure out the best
ways to move the agenda forward.
Senator Smith. The best practices the thank you speak of
and that we're striving for--this is a legitimate field of
medicine. I'm not suggesting that. What I am suggesting,
though, with the proliferation of this category of direct sales
to consumer, I wonder if there are best practices out there
that are actually being followed.
I wonder maybe, Steve, you've got a comment on that?
STATEMENT OF STEVEN I. GUTMAN, M.D., DIRECTOR, OFFICE OF IN
VITRO DIAGNOSTIC DEVICE EVALUATION AND SAFETY, CENTER FOR
DEVICES AND RADIOLOGICAL HEALTH, FOOD AND DRUG ADMINISTRATION
Dr. Gutman. Well, I think there are. I think that there are
best practices that are general, so I think a lot of what's in
CLIA applies directly to the testing of genetics. But I think
that the idea of the best practices is that there are enough
nuances in this particular area that there needs to be perhaps
some more specific guidelines or recommendations.
I can speak to the interaction, not on a formal basis, but
on an informal basis, and I can assure you there probably isn't
a week that goes by that there's not some kind of informal or
semi-formal interaction between our group and the CLIA group,
either asking them for help in our premarket compliance program
or offering them help in their inspection or compliance
program. So there's actually on an informal level, there is
perhaps some outsiders might view it as a disturbing amount of
communication and coordination.
Senator Smith. Is there a disturbing amount, Linda?
[Laughter.]
Ms. Avey. We're not disturbed.
Senator Smith. I'm curious, Linda. With your company, do
you have a sense that the industry is developing with some best
practices and standards that are designed to protect consumers
from what they're doing?
Ms. Avey. Yes, certainly we're very engaged with all the
agencies, I would say, as much as we can be, and open about
what we're doing and very open to the idea that we need new
regulation. I think trying to fit this round peg, if you will,
into the square hole of existing regulation is probably not a
good fit.
But that said, we're very open, and we applauded the
hearings you had 2 years ago because we recognize, and I think
coming from the scientific community, we saw what some of these
other companies were doing and were appalled by that. We feel,
very similarly, that this does need to be regulated in a very
strong way. It's just a matter of how can we do it and still
allow the United States to be a leader in this field, because
getting more genetic information and providing more people
access to their genetics will move the field of personalized
medicine forward, and if we block that we're going to be
stymied and stuck with the existing health care system that we
have.
Our overriding mission is to gather more data so that we
can make these connections between people's genetics and their
health outcomes and translate the great research that's going
on into the clinic more readily. The SACGHS writeup really
showed that there's a gaping hole in this translational aspect,
that we don't really have a way to take the results that are
now coming out in a flood from the research community of all
these genetic associations and move them into clinical
practice. How do we get those data into the clinics? If we
don't have a way to demonstrate clinical utility and validity,
we're never going to get there.
That comes from having many, many people involved. We think
the consumer has to be engaged in the process or it's just
never going to happen.
Senator Smith. Is yours a public company? I'm curious.
Ms. Avey. It's a private company now.
Senator Smith. A private company.
Ms. Avey. We're small. We're a startup.
Senator Smith. If I were to own stock in your private
company and I heard what you just said, that you were horrified
about some of the new entrants into this field, how do you
distinguish yourself from somebody selling snake oil versus
somebody putting out a legitimate product?
Ms. Avey. The bottom line, it comes down to what the
genetic associations are really--what's out there, what do we
know and what's real, and what is not real, what's been proven
and what has not been proven.
The research labs and our science team are working very
closely together to say, look, this is all new, we don't really
know what this means yet. We need to start asking our
customers. It looks like you're at higher risk for type 2
diabetes based on your genes. Do you have type 2 diabetes or
not? So that's where we're going to start putting out surveys
to our customers to say, let's start connecting the dots here.
So we just see some of those other companies as putting
their claims out way ahead of the science, where we don't know
yet, we have no idea what foods you should eat based on your
genes. We just don't know that yet. Some day maybe we will, but
until we get there we're going to be very true to the science
and very responsible about that.
Senator Smith. So you would be welcoming the agencies'
efforts to try and formalize some of this so you can
distinguish your company from some of the others that may just
be selling a lot of products to the injury of their customers
or of creating false hopes in their work product?
Ms. Avey. Any claims about these data we think are really
pretty premature. That's why we put that out there that this is
research data. We do feel like it's best to engage with
consumers through their own genetics because they get so much
more interested and they really want to be educated when you're
talking about their own DNA, what they're born with, what
they're going to die with, what does that mean for them.
But we put it out there that we just really don't know a
lot yet. We keep saying, we need to do more research, we need
to get more people involved.
Senator Smith. Kathy, I see you anxious to say something.
STATEMENT OF KATHY HUDSON, PH.D., DIRECTOR, GENETICS AND PUBLIC
POLICY CENTER, JOHNS HOPKINS UNIVERSITY
Dr. Hudson. Senator, you asked a very straightforward
question at the beginning, which is, are consumers protected,
are genetic tests safe? You asked that same question 2 years
ago when you had your hearing and had the GAO investigation. I
would argue that there have been no improvements in the
oversight of laboratory quality, in the oversight of genetic
tests themselves, and in the oversight of the claims made about
those tests in the intervening 2 years.
There are conversations and working groups and meetings and
conference calls, but we actually have not seen any actions.
Despite your call for actions and continued attentiveness,
which we appreciate, and the recommendations of now four expert
secretarial-level advisory committees that there be changes in
CMS, changes in FDA, changes in FTC, we actually haven't seen
so much as a notice of proposed rulemaking.
So it's great that the agencies are working together, and
talking is really great, but action is even better.
Senator Smith. Any proposed rulemaking coming out of the
advisory group?
Ms. Yost. Definitely, yes.
Senator Smith. Go ahead, Judy.
STATEMENT OF JUDY YOST, DIRECTOR, DIVISION OF LABORATORY
SERVICES AND CLIA PROGRAM, CENTERS FOR MEDICARE AND MEDICAID
SERVICES
Ms. Yost. On proficiency testing requirements, but we've
got to gather data, we've got to work with the experts in the
field, we've got to determine a negative, and to identify which
tests should have proficiency testing using scientific and
technical expertise. So it's going to take a little time, but
the plan is definitely to propose new proficiency testing
requirements, not only which tests are covered, but how they're
graded, how they're monitored, how the PT providers are
approved, and so forth.
So the entire scope of proficiency testing is being
reopened and reevaluated through our advisory committee
process.
Senator Smith. Can you give us a sense of a time line? How
much time do we need to gather all this information, all the
data that will make the difference to allow some rulemaking to
go forward?
Ms. Yost. Well, we've initiated a process starting in March
and we have a plan to meet with the proficiency testing
providers by November of this year. We've already gathered some
preliminary data on the most frequently performed tests. We're
going to also look at tests that are clinically relevant and
where proficiency testing materials might be available. We have
a meeting scheduled with our accrediting organizations and with
our advisory committee in September to develop a recommendation
and to convene a committee to begin the deliberations on that
process.
So it will take some time on the front end, but the idea is
that hopefully we will have a quality product then to be able
to move forward with.
Senator Smith. Kathy, do you have a comment, or Steve?
Dr. Gutman. Yes. You asked me the question about whether we
thought we had authority over these tests at the hearing.
Actually, there had been some ambiguity, so that was a well
placed, well chosen question. I answered that we did.
We haven't made a lot of progress. I won't pretend that we
have. We have, however--we haven't initiated rulemaking. We
have initiated guidance to--again, we're a risk-based
organization, so we start with things that are the most
worrisome to us. We've chosen a particular product line called
in vitro diagnostic multi-variate index assays. That's a
mouthful, but what that is is----
Senator Smith. Don't ask me to repeat it.
Dr. Gutman. What it is, you take a bunch of signals, you
put them in a non-transparent black box or black blender, you
grind them together and you come up with a magical score that's
somewhat not intuitive to the health care user. We thought that
would be an interesting place to start in terms of changing
that pattern in which we have applied enforcement discretion to
these tests and deferred entirely to CLIA, to perhaps becoming
partners in more active regulation.
Senator Smith. So would it help you to have Congress give
you more statutory direction in this?
Dr. Gutman. Yes.
Senator Smith. It would, OK.
Yes, Elaine?
STATEMENT OF ELAINE LYON, PH.D., ASSOCIATE PROFESSOR [CLINICAL]
OF PATHOLOGY, UNIVERSITY OF UTAH SCHOOL OF MEDICINE, AND
MEDICAL DIRECTOR, MOLECULAR GENETICS, ARUP LABORATORIES
Dr. Lyon. The issues we're discussing here will be
directing us in the laboratory community. We definitely have a
stake in this. I want to discuss CMS's comments on what is
being done. I disagree with Kathy's remark that nothing is
being done because there is significant activity.
We've been working with the CDC on a reporting initiative,
how to communicate genetic information back to physicians.
We've worked with the CDC in terms of getting reference
materials for the validation of tests and for the proficiency
testing. The CDC is also going forward with a study looking at
the clinical utility of these tests.
So I believe in the genetics community and the laboratory
community that there has been a lot of work moving forward to
address some of the issues, to make testing better and safer
for the public.
Senator Smith. I'd love to get the FTC up here too, if
there's anybody from the FTC here.
Mr. Daynard. Do you mind if I stay here?
Senator Smith. That's fine.
One of the questions, along with the clinical practices and
best practices, one of the other issues I'm concerned about is
obviously privacy. Linda's company says they only use it in a
very limited way, with their permission. But what happens if
somebody is selling the genetic information to somebody?
STATEMENT OF MATTHEW DAYNARD, SENIOR ATTORNEY, DIVISION OF
ADVERTISING PRACTICES, BUREAU OF CONSUMER PROTECTION, FEDERAL
TRADE COMMISSION
Mr. Daynard. Well, two points. I was going to answer the
first question about our coordination with other agencies and
what the FTC is doing. But we're also very involved in privacy
and security issues. That would be a serious problem. If there
is a privacy policy that says, for example, ``under no
circumstances do we sell or rent or give away your protected
health information,'' and if in fact they did or in fact they
didn't have well-established security measures the FTC might
take a very close look at it.
On the claims issue----
Senator Smith. You can see the damage that would be done to
an individual if their genetic information was sold and it's
out there in the insurance market----
Mr. Daynard. Certainly.
Senator Smith [continuing]. They have some test, whether
valid or not, that somehow says they have a predisposition to
cancer.
Mr. Daynard. Yes, absolutely. At the moment we don't have
evidence of that happening, but it may be happening.
But what I'm looking at, and what we looked at very shortly
after your hearing 2 years ago, was the claims issue. What we
did quickly, in conjunction with the Food and Drug
Administration and Steven Gutman's staff and the staff at the
CDC, was to issue a consumer alert warning them about issues
that they should be aware of when they consider direct-to-
consumer advertising.
That we could do quickly. A law enforcement investigation
can't be done quite as quickly. But I can confirm that we have
a couple of investigations in this area looking at the subject.
Senator Smith. So you've got all that you need?
Mr. Daynard. Yes. I mean, the statute which declares that
unfair and deceptive practices in or affecting commerce are
illegal is pretty broad. We have done wonderful things with it,
and I think we have jurisdiction in this area as well.
Senator Smith. You're on the case.
Mr. Daynard. Yes.
Ms. Yost. I'd also like to mention, because the impetus for
your hearing last time was the four DTC laboratories that had
been identified by GAO. Since that time I'd like to mention
that we have collected a series of at least eight or nine of
them, so we have been very closely monitoring them, and in
addition, as the ACLS statement indicates that those that we
identified aren't doing testing that falls under CLIA. Some are
just marketing. All of them have very different various
functions and they're all organized very differently. Sometimes
they're interrelated.
But that said, those that are actually performing testing
that fall under CLIA have been required to obtain a CLIA
certificate and obviously assessed as to their compliance
status. So we have made a tremendous amount of progress, I
would think, since then, and we continue to do so. Now that we
have expertise and we are working much more closely with FDA,
we have the ability to look at the types of tests that they are
performing and verify their validity.
Senator Smith. The four labs we identified at the time,
you're on top of that?
Ms. Yost. Yes.
Ms. Avey. I'd like to add to that, Senator Smith. This is
something that we interacted with CMS on as well. When we first
started, it was unclear whether or not CLIA was the right
oversight for what we were doing, because we don't really claim
ourselves to be a specific clinical test. We look at 600,000
data points in your genome and then tell you what we know based
on the research. So it's not a test per se, but nevertheless we
were notified by CMS.
We were working with a very high quality laboratory, but it
was not CLIA-certified. But after the interaction with CMS, we
did switch to a CLIA-certified laboratory. Then we did tests
back to see how well the data correlated. We ran the same
people in the previous lab and then in the CLIA lab and got
over 99.9 percent concordance between the two labs. So we felt
very good about our previous lab, but now we're working under
CLIA oversight, which we do think is the right thing.
Senator Smith. I've got a question for you, Greg. But
before I get to it, Kathy and Elaine, you disagree on a point,
whether anything is being done or not. I wonder if we can
explore that just a little more.
Dr. Hudson. I applaud the efforts that CMS has undertaken
to identify what laboratories are offering these tests and to
bring them into CLIA compliance. The problem is that CLIA, as
it stands today, just isn't enough.
It's great that they're moving now toward proficiency
testing requirements. The Secretary's advisory committee made a
very straightforward recommendation, which was that if a
proficiency testing program exists, and you perform that test,
you must participate in that program. Proficiency testing is
like a pop quiz for a laboratory. You get a sample, you have to
do the test for the disease, cystic fibrosis, sickle cell
anemia, send the result back, and if you get it right you pass,
and if you don't get it right, authorities are alerted that
you're sub-optimal.
That's important. Getting the right answer is important,
and knowing the meaning of that test result is important as
well.
I'm encouraged that CMS is moving in the right direction,
but again, it's been 10 years since the first recommendation
for proficiency testing requirements.
In terms of the claims that are made about tests, I agree
with Linda that the platform that she is you're using is
incredibly, incredibly reliable. But the information that's
made available about the tests is more uncertain. Let me give
you an example.
Robert Green, who's a neurologist at Boston University,
spoke at an event earlier this week. He has had his genome
done, and he showed the results. He had his genome done--sort
of like you get your hair done and your nails done, you get
your genome done--at both 23andMe and a sister company,
Navigenics. For his variations for heart disease one company
said, ``You have a heightened, greater than population risk for
heart disease.'' The other company said, ``You have less-than-
population risk.'' So they gave absolutely contradictory
information. It turns out that he had an incredible risk for
heart disease because in fact he had a major cardiovascular
event and a triple bypass.
So if two companies testing the same variation's give you
two different answers, we've got a problem on our hands.
The head of the genome office at CDC, Muin Khoury, recently
published a paper in which he said there's insufficient
scientific evidence to conclude that genomic profiles are
useful in measuring genetic risk for common diseases.
Senator Smith. This is the point of the hearing to begin
with. The information that can come out of this is so explosive
and can be interpreted so differently. My concern with direct-
to-consumer sales, not that they can't be made appropriate, but
that they can be used in a very destructive way, and people
will go get the wrong treatment when they ought to be getting
the opposite treatment, and their information can be
disseminated in the insurance market in a way that would make
them uninsurable.
That's why this is an exciting new field in medicine and we
need to pursue it, but we've got to have--time is of the
essence to put up the kinds of standards and best practices and
regulatory barriers so that we're not leaving the consumer
behind in this.
Dr. Lyon. There is room for improvement and I won't deny
that. I think that what we do as a CLIA-certified laboratory,
we can really make improvements under CLIA and under CMS, and I
don't believe that there really needs to be additional
regulation. However, there can be additional improvements.
It is true that for these genome-wide associations, the
science is just coming out with them and it may be not be
strong enough right now for me to be able to interpret what the
tests mean. That doesn't mean we shouldn't go forward with the
science to get that understanding.
On the other hand, a lot of new genetic testing is being
thrown in with the mix which is really standard medical care.
That needs to be differentiated. We design, perform and
validate laboratory-developed tests. I brought some of our
validations and can show you what we do as laboratories to
validate these tests.
As new issues come into play, which is really where we
are--we're on the verge of going from single gene disorders--
such as we know the gene that causes cystic fibrosis; I know to
look for mutations in this gene and for the most part I know
how to interpret it--to looking for all of the genes that may
cause some type of lung disease. We need to go forward with
that. There are definitely reasons to do so.
But that shouldn't be confused with what we are doing right
now in the practice of genetic testing.
Senator Smith. Greg.
STATEMENT OF W. GREGORY FEERO, M.D., PH.D., SENIOR ADVISOR TO
THE DIRECTOR OF GENOMIC MEDICINE, NATIONAL HUMAN GENOME
RESEARCH INSTITUTE, NATIONAL INSTITUTES OF MEDICINE
Dr. Feero. I'd just like to first make the comment, I think
we all are sort of in a privileged position right now to be
dealing with a wealth of information that we don't exactly know
what to do with. I'd say that what we have right now is a lot
of data and somewhat less wisdom as to what we should actually
do with that information.
I think there are a couple of things that really need to be
thought about. One is the science is evolving very, very
rapidly, and I think the systems we have for dealing with that
rapid evolution don't--can't really keep pace with that
evolutionary process. Over the last few years, we've seen
discoveries for literally hundreds of new genetic associations
in I believe 40 common disorders. That pace is accelerating.
It's, I think, a natural outgrowth that various elements in
industry would see potential advantage in working with that
wealth of information. In many ways, some of those companies,
the responsible ones, may be on the vanguard of figuring out
how to exactly deal with that information at the end.
There's a really broad spectrum, of groups offering testing
some that are clearly doing things that are very egregious.
However, are folks on the other end of the spectrum that are
doing diagnostic testing that's well established.
Then there's this great grey zone in the middle that we
don't really have a good handle on how to deal with at this
point in time. Simple questions like what defines a medical
test are very difficult to answer when you start talking about
these things. Do you draw the boundary at Ann's asking
questions about Ann's ancestry? Do you draw the boundary when
you start to talk about things like eye color or your muscle
fiber types? Do you draw it at the boundary of predictive,
potentially predictive genetic testing for late onset
disorders? Clearly I think most people would agree that that's
where you start to think that, yes indeed, you're on the
medical end of the spectrum, But answering where that starts is
I think a real challenge.
We really need ways to be as nimble as possible as this
area rolls out to deal with these topics. Which is a great
challenge. The larger issue of measuring the utility of these
types of tests is going to be critical as we move down the
line, because we'll have the opportunity to do many things in
our health care system and it'll be less clear as to the added
value of each of those items.
Right now I think we're just sorting out what the pipeline
looks like for this area as it moves forward for determining
that utility.
Senator Smith. Greg, in February of this year there's a
company that started issuing, selling a genetic test to
determine bipolar disease. Mental health is an issue that
matters a lot to me and I'm trying to push it forward as a
coequal part of health care. I understand you're aware of this,
that you've lent tacit approval to the company's test. That's
what I've been told.
Dr. Feero. I would disagree. When that article came out, I
have to say I was chagrined. I was even more chagrined when it
came to me last night that that might potentially be on the
discussion today.
Senator Smith. Can you genetically test for bipolar
disorder?
Dr. Feero. Bipolar is a clinical diagnosis at this point in
time.
Senator Smith. Not a genetically determined----
Dr. Feero. Not a genetic--well now, it has a genetic
component to its causality certainly.
Senator Smith. OK.
Dr. Feero. But it is not diagnosed using genetic testing--I
was trying to make a fine point with the reporter in that case,
that using a test in a totally predictive manner in an
asymptomatic person is different than using a test in an
individual who already has symptoms that make them more likely
to have the disorder you're talking about. So sort of the
analogy might be to use a white blood cell count. In somebody
who's perfectly healthy interpret what an elevated white blood
cell count means, is very different than in the use of somebody
who has a fever and a cough which is productive of sputum. It
means a very different thing when you see a high white cell
count in that person.
I think they didn't quite get the distinction I was trying
to make.
Senator Smith. Does it concern anybody that there's a
company out there selling genetic tests to determine bipolar
disorder? Should that be done with a doctor or through the
mail?
Dr. Feero. I would say that if you look at their site
they've qualified it. But the question is do people understand
the qualifications that they've put into it. That's really the
big question when I look at this type of thing, is will the
consumer and the health care provider, frankly, be able to
understand this area sufficiently to make a determination as to
the value of the proposed application.
Senator Smith. Yes?
Dr. Lyon. When I heard this from Chris, I went to the
Internet site to see if I could find out what was going on and
what it was. In doing that, I needed to go back to the
scientific literature and read to see what the science really
was.
There is a modest association, according to the published
abstract. That does not mean that it's ready for prime time.
The unfortunate fact is that when I went to that web site I
couldn't get to the details of their test in order to make a
judgment on whether it was actually good or not.
Senator Smith. Yes, Kathy?
Dr. Hudson. That actually raises a really important point
about the transparency of information. So even if there's not a
heavy hand of regulation for these tests, we need to make sure
that consumers and physicians and others have access to
information about the scientific evidence on which you are
basing your claims, so that we can all tell the truth to one
another, at least know what the truth is.
Right now you can't. It's very difficult in many cases to
know what's the scientific evidence underlying the claims that
a company is making in selling a particular test.
Ms. Avey. I want to go back to what Kathy said earlier
about the seeming discontinuity between the heart disease
predictive markers between the two companies, Navigenics and
23andMe. That is something that we are going to address. The
good thing is both companies are very transparent about how we
do these risk calculations. You can do these in very different
ways. There are a lot of assumptions that we have to base it
on, whether it's average risk in a population, whether it's
lifetime risk.
Again, we're very clear about that, but that might be
partly why Bob Green ended up with two different end points
through these different services.
So we're meeting with Navigenics on the 17th of July to
start working together to develop our own internal standard so
that we're not confusing people, and that we're at least using
some of the same criteria for how we come up with these risk
assessments, if you can even call them that, because it's based
on research data.
Senator Smith. I have to--that buzzer you heard means I've
got to go vote. Unfortunately, the leadership doesn't check
with me on my schedule when they schedule a vote. But that's
life in the Senate.
I'm not doing anything today more important than this. This
is really exciting stuff. It's a whole new era that we're in.
But it can be dangerous. That's why I've asked you to come
together and to keep the focus on this, because we're focused
on it. If we need more legislation, if you need more authority
to get this right, we'll help you with that, because again it's
exciting, it can be troublesome too. We've got to get this
right. The sooner we do it, the better off the American people
will be and the better off health care will be.
So with that, if you'll excuse me, and with my apology for
having to go answer this vote, I'll turn it over to my staff. I
thank you all for being here. You've added measurably to moving
this forward in a constructive way and to the record of the
U.S. Senate. So I thank you all for your time.
Ms. Hinkle. Well, so we jumped right into the questioning
and I'm wondering if now, an hour into it, we might be well
served by backing out a bit and covering maybe some of the
broader aspects of genetic testing, and then we'll dive into a
few technical and nitty-gritty bits.
For the audience, there will be an opportunity for you to
engage in some Q and A during the last 15 minutes or so of the
roundtable. We've passed out some cards on your chairs. You can
hand those to our staffer in the back there who's waving her
hand if you have questions. Or you can just feel free to come
up to the middle here, where there was supposed to be a
microphone, but isn't unfortunately, and ask the question
yourself, whatever you prefer.
Let's back out. Let's take an overview of genetic testing.
Anyone who wants to jump in and set the stage, just very
succinctly tell us, where have we been with genetic testing,
what advances have we seen over the past couple of years? Who's
really using genetic tests and in what capacity?
We know that patients, doctors, researchers, even law
enforcement are using genetic testing. So maybe someone could
just give us a quick overview? Anybody? Greg?
Dr. Feero. A quick overview of what defines a genetic test.
I think there have been probably week-long arguments over what
exactly defines a genetic test. Is a genetic test simply a
molecular diagnostic assay that you might use to determine
whether or not there is hepatitis virus present in someone's
serum? Or is it only testing for rare genetic disorders in the
context of someone's germ line?
There aren't clearcut answers for that. I would say that in
the main, the vast majority of genetic testing with a fairly
strict definition of what you mean occurs in the realm of
relatively uncommon to rare diseases. For this type of testing
there is much less question, I think in general, about the sort
of systems that are in place to deal with the topic.
What we've seen in the last few years is a profusion of
sort of this new type of testing, which looks at variants
across the genome that are merely associated with the disease,
rather than necessarily causal, and then an attempt to use
those variants in sort of a predictive prospective manner to
inform people about various things, ranging from, as I
mentioned, eye color and ancestry to whether or not they're
going to develop Alzheimer's disease or cardiovascular
disorders.
I think the profusion of this sort of predictive testing
with associations is really one that has brought this question
to the fore. We've been waiting a very long time, I think, in
the scientific and medical community to finally crack into the
understanding of common chronic diseases that have a component
of both genetics and environment contributing to their
causality. We finally made that crack and--the rush is to try
to use that information in a way that's beneficial.
Ms. Hinkle. I know that clearly one of the frequently cited
concerns about genetic testing in the DTC arena is the absence
of a qualified medical provider to help patients interpret the
results. The American Medical Association is going to be voting
on a resolution at its meeting again I guess restating and
further enhancing, further expounding upon its opposition to
DTC testing.
I guess my question to you particularly as the provider at
the table is, who is qualified to interpret these tests? What
is a qualified medical professional?
Dr. Feero. I think that that is a question which is very
dependent on the context and the type of test that you're
talking about. There clearly, even now with the more rare
conditions, is a spectrum of who's qualified to interpret what
type of test.
So for example, a general internist might be quite capable
of interpreting a test for clotting abnormality in the form of
Factor V Leiden. It's a relatively straightforward test.
Internists and family physicians order these quite frequently.
However, I would say that the average internist at this point
in time is probably not adequately trained to fully interpret a
complex result from a BRCA-1 or BRCA-2 testing. In that case,
it would be best handled by someone who has formal genetics
training in the form of a medical geneticist or a genetic
counselor or a nurse geneticist.
Ms. Hinkle. Do most patients have access to that level of
care?
Dr. Feero. There is a great disparity, I think, right now
in the current genetics community between the sort of profusion
of new types of tests that will need to be dealt with and the
actual amount of manpower that's present to deal with the
information as it comes out. Hopefully I'm not misquoting this,
but this year in the match in the United States, which is the
process by which medical students select their residency, the
statistics for those people matching in medical genetics,
suggested that there was one U.S. medical graduate that matched
for medical genetics.
Ms. Hinkle. One?
Dr. Feero. Yes. So there's clearly a big manpower issue.
For example, I was in a talk earlier today where a family
physician was presenting some data from a survey that they did
that showed that 11 percent of family physicians that they
surveyed reported that their patients had to drive over 2 hours
to get to the nearest medical genetics professional. So there's
a great disparity.
Dr. Lyon. We have realized that what used to be ordered by
geneticists is now being ordered by other physicians, and
because of that it's very important to communicate well what
the results mean. So a laboratory's role is not just to give a
result, but we do need to give an interpretation of what that
result means.
One of the programs that I've worked with at the CDC is
their reporting initiative to get the information so it is
understood by a general practitioner or a family practice
physician.
Ms. Hinkle. So let's say I go on line, as I have here to
23andMe, and this is just their demo site here, so this isn't
any real health results. But I get a whole profile like this.
Let's pick something at random. I ordered it here on the web
page by what 23andMe has listed as the most established
research for. How about restless leg syndrome. So I'm going to
learn some genetic information.
As I'm logging in, tell me. So I go on line, I get these
test results. To whom do I take them if I want assistance in
understanding what these results mean?
Ms. Avey. We actually have some information now back from
our customer base, because we launched in November and we have
thousands of people who have signed up and who are now using
our service. We take all of their questions. Most of the
questions are about ancestry, or ``I forgot how to log in,'' or
``I don't know how to get to my data.'' It's been more about
the use of the web site.
We have to get better and better at how we direct people on
how they go through the service so they get the information
they're looking for. We have found that people are not really
asking questions like ``What do I do with these data, who do I
go to talk to? But we're following this closely.
In the mean time, we're aligned with other companies in
this space, and we agree that this information is going to
start getting into the hands of a lot of people, and the
medical community, especially the medical genetics community,
will not be prepared to really catch this on the other side if
a lot of people are having questions.
So we're having webcasts with the Genetic Counseling
Society. We're reaching out to physicians. We're working with
NCHPEC, we're talking to Muin Khoury at CDC. We think we have
to do this in a very broad way and a very scalable way, because
if we just have a few people who are educated about this
information it won't be a system capable of handling the load
of people who might start having questions.
We try to pack a lot of information about the fact that
most of the diseases you see are not just genetically
determined; there's a big environmental component. So for type
2 diabetes we put that out there. It's only about a 25 percent
genetic disease. Your environment plays a much bigger role.
So once people read through that information, a lot of
times they say, OK, it is what it is, but they may not have
increased risk for a lot of the diseases, because I think a lot
of our customers are, luckily, very healthy. Then they go and
they look at their ancestry information and that's where they
start having the questions, because it's something that they've
never seen before and it's not something they've really thought
about before.
Ms. Hinkle. Linda, earlier in speaking with Senator Smith
and actually in our earlier conversation as well, you certainly
distinguished what you offer from the nutrigenetic tests that
were the focus of the hearing 2 years ago. While I think
certainly maybe the type of testing or the labeling of the
testing is the same, I also find many things in common with the
results that you offer. That is, though, much like on those web
sites, you disclaim making any sort of diagnostic services and
that this is strictly for educational purposes, as a consumer
when I read something that is telling me what type of odds I
have of getting a condition, and as we see up here, an odds
calculator, when that's telling me increased risk, decreased
risk, that to me sounds like a health diagnosis.
This was exactly the same question that we posed to the FDA
and CMS with respect to the nutrigenetic testing and whether or
not these really were the type of health care claims that would
fall within the ambit of your jurisdiction. So are these types
of results to you really different than the nutrigenetic tests?
Is there really a distinction there? I guess where do we draw
the line in genetic testing, where one type of genetic test
falls under a different type of regulation, or the laboratory
would?
Dr. Gutman. Well, actually I would probably--I do think
Greg is right, it's nuanced when you start moving toward
whether you like particular apples or vegetables or where your
grandfather might be from. I don't think it's so nuanced when
you say, well, whether it's a definite possibility, a
reasonable possibility, a possible possibility. Are you
starting to associate--from FDA's--I'm not a lawyer, but from
FDA's perspective, it seems to me the definition is pretty
clear. We have a very broad interpretation of ``diagnostic.''
So that's clearly a health claim. You can put for informational
purposes, you can put not for real use, this is pretend. You
can't duck it. It's the old thing, if you quack and you have a
beak and you waddle, you're probably a duck; you're probably
not a horse.
Ms. Hinkle. So these look like diagnostic claims?
Dr. Gutman. Yes. But it begs the issue. I think it's an
incredibly complex issue. I think the fact that we at FDA
haven't moved faster or that others haven't moved faster or,
you know, we've only got the SACGHS report for 8 weeks. To
reinforce, as Kathy pointed out, it's not exactly a
fundamentally new construct that suddenly popped up.
Actually, what was striking about the SACGHS report from my
perspective is, one, it did reinforce what had been said before
at least a couple of times. Then it opened this incredible door
that somebody finally figured out that the utility was actually
important. I feel sorry for whoever gets stuck with that.
From the agency's standpoint, while we're from the
government and we're here to help, it is a big task, so we
would be careful about what we wish for. I remember saying that
at SACGHS as they struggled over whether to put FDA in or not,
is that we're not terribly worried about job security right
now, we're not sitting around. But we do think--I do think; I
shouldn't represent the agency in this comment--that there is
some value to independent review, whether FDA does it, whether
CMS does it, whether it turns over and 23andMe becomes an
accrediting agency for ACLA. There is some value to
independent, financially independent, intellectually
independent, absolutely agnostic in terms of technology and
claims--there is some value that somebody might want to step in
here and look at independent review.
I think that's hidden in the 200 pages of the SACGHS
report--not that well hidden, it's actually kind of obvious. So
I personally think that's a good idea. I don't know who should
do it, but that's a good idea.
I think on the issue of confusion--I thought the point was
made--certainly I do think the average doctor is having trouble
figuring out what a pro-time means, much less what a new
genomic marker for brain cancer or Alzheimer's disease or
bipolar disease, which I think is absolutely a medical claim
that's very confusing. I can't imagine a direct-to-consumer
patient actually sorting through that and starting to self-
diagnose bipolar disease. That's just absolutely frightening.
I think doctors have problems, maybe even the genetic
counselors. When you take the brand-new cutting edge molecular
signatures that have no literature, are brand-new, or
literature from a single source, they're going to have trouble.
I think that speaks to Kathy's request for transparency and
that the leitmotif both among people who swear at FDA, or swear
by FDA, is that nobody seems to speak against the idea of more
disclosure, of having transparency, of having registries.
Nobody says who'll fund it, whether it will be mandatory or
voluntary, who will administer it, how do you assure financial
and intellectual independence, how do you communicate the
results.
So there are some niceties that haven't been addressed, but
nobody seems to speak against that. The more you can get
information out there, then the more likelihood that, whether
they be a Ph.D. from Harvard in genetics or my mother-in-law,
that they might have some chance at least of reading and
getting a straight answer on whether there were three samples
or 3,000 in the credentialing of the test, whether the test was
established using well-established goals, standards, and
brilliant literature that map a course or using an article in
the ``Gutman Journal of Unusual Results.''
So I think that----
Ms. Hinkle. We'll get a subscription to that.
Dr. Gutman [continuing]. It's a timely topic to bring up
and we from the FDA, from the government, we are interested in
understanding how we can help and not hurt.
Ms. Hinkle. We've talked a lot about the genetic tests that
we sort of have a sense are not appropriate for direct-to-
consumer use--I guess I would ask you what genetic tests are
appropriate for direct-to-consumer use right now? Or are there
any?
Dr. Hudson. Can I respond?
Ms. Hinkle. Oh, absolutely, Kathy.
Dr. Hudson. I think ultimately there's going to be a large
number of tests that will be appropriate for consumers to
access, and there may be some now. If there was a test that
could tell me which over-the-counter medicine would work best
for me, why do I need to have a doctor order a test to tell
which over-the-counter drug to use?
Ms. Hinkle. We'll come back to pharmacogenetics and
pharmacogenomics in a bit.
Dr. Hudson. So I think that the tests that we need to worry
most about in terms of direct-to-consumer access are those
tests that are high risk. We can talk about what risk really
means. I think high risk means when you're going to make an
important health-related decision based on the result of that
test.
So if you're going to go to the gym a few more times, who
cares? If you're going to start buying more Q-tips because you
have sticky earwax and not dry earwax, who cares? But if you're
going to make important health-related decisions based on that
test, then I think that sort of moves it over to a more
important category. It would be impossible, and will become
even more impossible, for Judy and Tom and Steve and Kate to
oversee all tests. There were 40 associations last year, or
we're up to 40 validated? So this is going to become a tsunami
that's going to drown these guys. So they have to be able to be
selective about what's important. I think what's important is
what's important to the consumer and the consumer's health.
Ms. Hinkle. Yes?
Dr. Feero. A quick point. I'm a little reluctant, I think,
to paint the types of associations you see up there in the same
category as things that went on in the past with nutrigenomics,
because to some extent the fundamental scientific underpinning
of the actual association is actually very sound now. The
science----
Ms. Hinkle. In just 2 years it's evolved?
Dr. Feero. It's evolved considerably. So these associations
are quite robust. The difficulty is that they provide
relatively meager predictive capabilities. So any one marker,
for example, associated with, say, type 2 diabetes, may elevate
your risk only by 1.2 or 1.4-fold.
Ms. Hinkle. Don't you think that's why most people are
getting these tests, though, because they want to know, am I
going to get cancer, am I going to get Alzheimer's?
Dr. Feero. The difficulty is in how do we interpret and
utilize those very small risk increases. There may come a time
ultimately where we have, say, 40 markers identified, we know
how to assemble them into a panel, we know how to compare that
to the environment the individual's in, and we can give them an
accurate risk prediction based on that.
But we're very early on in sorting out what to do with
those bits of information. But that doesn't mean that
ultimately there may not be utility in this. We're just so
early on.
Ms. Avey. Again going back to my earlier comment, what we
see lacking right now is the ability to translate this tsunami
into something clinically relevant and clinically useful.
That's really the role we see ourselves playing. If you go back
to the list of all 70 different traits that we're showing, we
have now the little icon for surveys that we have along these
disease lines. So what we're going to start doing is asking our
customers to give us information back, answering questions such
as ``Do you have type 2 diabetes?''
My brother-in-law has type 2 diabetes, but he wouldn't know
it from his risk profile through his genes. It likely means his
environment has had the most impact on his getting the disease.
So he looks at this and he feels empowered to say, no, these
genes, even though I have the disease, don't look like they
light up in me. So that means either we don't know enough about
the genetics of this or I've got the environmental gun loaded
and pointed at me and I have to do as much as I can to change
my environment. It's not a genetic thing for me personally.
So it's about gathering up all this other information and
empowering our customers to tell us and have them be active
participants in the continuation of these studies, rather than
having it go on in the research community and the Science and
the Nature papers come out and the scientific community reads
all of that, the New York Times picks it up, maybe sometimes
the Wall Street Journal writes about these studies, and people
think, ``Well, what does this mean for me? I want to know what
this means for me personally?'' That's the sentiment that we're
hoping will help drive additional studies and get more people
involved directly.
Ms. Hinkle. It's interesting you picked type 2 diabetes. If
I recall correctly, maybe just a week or so prior to the 2006
hearing NIH had issued a press release regarding genetic
testing and type 2 diabetes and, although they acknowledged
some of the predictive testing computers, they actually
discouraged consumers from undergoing routine testing, genetic
testing for diabetes.
Has the science evolved on that particular condition enough
in order to make routine genetic testing now more sensible?
Dr. Feero. You didn't ask me directly the question would I
subscribe to one of these services at this point.
Ms. Hinkle. I was going to actually ask everyone if anyone
at this table had undergone genetic testing, if anyone wanted
to actually share.
Dr. Feero. So if a patient came to me right now and asked
me this question in my office, what I would say is that there
is great promise in this area in general, but we are at a time
that's too premature to fully utilize this information. We're
really in a discovery phase, and that if you are looking to use
this to change your health care at this point in time, it's
probably not appropriate.
Ms. Hinkle. Then I'll ask you in public now, so hopefully
you won't feel too put on the spot. The question I asked to you
when we chatted on the phone is. Why should consumers spend
$1,000 on your test if you yourself admit that the scientific
claims are a little premature at this point? It seems like you
guys get a good deal out of it. You get lots of DNA for
research purposes, and even the disclosures on your web site
basically say that your DNA sample is ours once you give it
over, that you guys claim ownership in the spit sample that I
sent in to you, and I think that that's probably something that
most consumers maybe don't realize.
I mean, let's face it. How many people in the room read the
fine boilerplate? I'm an attorney and I don't half the time,
shame on me. But I did actually sit down and read all the
disclosures on your web page and I have to tell you, at this
point in time I wouldn't be prepared to submit a DNA sample to
you because I'd be very nervous about the research that was
being done on it.
So assuage my concerns. Tell me why my DNA is in good hands
when I turn it over to you?
Ms. Avey. Well, the first thing is that you don't have to
participate in extended research.
Ms. Hinkle. Any research?
Ms. Avey. The research requires us to collect additional
information from you. If you never fill out a survey, we're not
going to be able to have you be part of the studies that we
conduct, because we need the phenotypic information. We need to
know whether or not you've had breast cancer or if you've been
diagnosed with autism.
That's the information that we will use and that will be
voluntarily submitted by customers who are very interested in
participating in the research projects.
Similar to how you see people signing up for the Susan G.
Komen walks people who have a family member, co-worker or a
friend who has breast cancer are wanting to do these things. So
when we talk to people that say, yeah, I don't have breast
cancer, but my sister did, and if I'm a healthy control for a
study that I could be part of, sign me up.
So it's really more when someone has a disease and they're
very interested in sharing that information. That's partly who
we're seeing signing up for our service, because they get the
message that not only do you get this preliminary view of
what's going on in the research community, you will be able to
be enrolled in these projects, in these studies.
But hey, if you're healthy and your life's going along
great, this probably isn't for you. It's completely voluntary.
So we say that right off the bat, that this isn't for
everybody.
Ms. Hinkle. So I spit in the cup, I send the sample off to
the lab, and the sample's destroyed?
Ms. Avey. At this point, yes. All we do is we extract the
DNA out enough to do the genotyping and then we discard the
sample.
Ms. Hinkle. The results or the data from that, if I don't
want you to ever use it for anything other than just returning
the results to me, it will never be data banked, you'll never
use it for any further analysis or research, even if it's
anonymized and in the aggregate?
Ms. Avey. If you choose, you can get your initial snapshot.
So you could go in, sign in to 23andMe, print out all of the
information that we have on the web site that's pertinent to
your data, and then you can download your data. You can take
your whole data set, and then you can come to us and say,
delete my file, and we'll delete it.
So you have that choice. If you don't want to get the
continual updates--every month we have more and more
information coming out. So what we're finding is our customers
are saying, give us more, we want more information, what about
that study that came out about ``such and such'' that's why we
added the other category of preliminary research, because
things do end up in the New York Times and then our customers
say, well, why didn't you guys cover that one?
There was one study that came out about how well you learn
from your errors. It was in a major journal. But we didn't
think that was appropriate because it was a very small sample
size and we thought the study was way, way, way too
preliminary; not bad or good, just not enough information yet.
But because of that we put it in our preliminary research
category and we put one star on it. So we have a way now to
categorize what's preliminary. Maybe a study will prove the
test of time. Maybe we will actually find genes for whether or
not you can learn from your mistakes. But we don't want to be
the judge to what our customers are interested in learning. But
we just have to put the right caveats around the data.
Ms. Hinkle. Let me ask you a question a bit about maybe
sort of the future of genetic testing. I posed this question to
many of you already. Are we looking at a scenario where some
day, perhaps even soon, we'll be having genetic tests sitting
on the shelf next to, say, pregnancy tests or diabetes testing
strips? Where are we headed for genetic testing?
Or maybe I'm going to get the perfect genetic match in my
love life, so I should invest quickly and establish a genetic
sort of dating service. Danny and I, actually that's going to
be our million dollar idea, so nobody steal it from us. Well,
there you go. Someone already scooped me.
What do you guys think? Thomas, you've been terribly quiet
today. I'll put you on the spot.
Mr. Hamilton. Well, I don't think you want the government
making predictions about the future.
Ms. Hinkle. Everyone's on the edge of their seat waiting
for you to speak, the great oracle. Come on.
Mr. Hamilton. You were asking if we had ordered tests. I
haven't personally, but I probably will order the Alzheimer's
test because on one side of my ancestry it's not just rampant
Alzheimer's, but Irish Alzheimer's.
That's when you forget everyone you know except those
against whom you hold a grudge. [Laughter.]
I just haven't seen this particular phenotypical variant
show up in the questionnaires. But when it does, let me know.
Ms. Hinkle. Maybe we can get some grant funding for that
research for you, Tom.
Maybe we'll chat a moment about some ownership issues,
since I grilled you so mercilessly on the DNA samples. Who
should have rights to the DNA material when I do actually
either provide my--when I provide a blood test or saliva or a
cheek swab, be it a physician-ordered test or a direct-to-
consumer test? Who owns that DNA? Who should own it? What good
use should we be able to make of it as clinicians, as
researchers?
I have a rare health condition and there could be great
medical cures developed from my DNA. Do I have the right to say
no? Does the government have the right to come take it from me?
What are we looking at as far as the future there?
Dr. Lyon. I can talk about what we are doing right now.
With the samples that come in, the residual samples that we
have are very important for us as we continue to validate and
continue to work for new tests coming on. There is a way for
anybody who sends their sample in to us to opt out of using
their sample for validation and education purposes.
There needs to be a distinction between research and what
the laboratories need to do to keep tests validated. So we make
the assumption that we can use that DNA unless we are asked to
discard it. If we are asked to discard it, we do so.
If we want to do research on it, we work with an internal
review board, the IRB. We need to go through the entire IRB
process, which includes the informed consent and the
understanding of the study. So we have this dual process. One
is to be able to keep our tests validated in the clinical
laboratory--those are anonymous and we cannot go back and link
any new information or finding to the patient--or research,
where we go under a research protocol which does ensure
confidentiality and ensures what the internal review boards
require.
Ms. Hinkle. Yes?
Dr. Hudson. Elaine raises a really important point and it's
interesting, because I suspect that your laboratory is governed
by the privacy provisions of HIPPA because you probably bill
electronically. So while people can criticize the privacy regs
under HIPAA, they do govern the more traditional genetic tests.
I don't believe that HIPPA covers many of the direct-to-
consumer companies. So there's the issue then of what
assurances of privacy protections do you have, and how do they
compare to HIPAA, and should we expand HIPPA to follow the
information and not the entity? So that's one issue.
The second issue is that there are a set of accepted norms
of the ethics governing research in terms of whether a research
participant is informed and understands what they're getting
into, whether or not somebody other than the researcher is
looking out for the research participants' interests and making
sure that the benefits outweigh the harms, etcetera, etcetera.
But if I read the regulations right, these ethics
regulation also don't apply to pic genetic testing companies,
although you could voluntarily comply with those research
regulations.
So we have this interesting situation where we have sets of
entities that are subject to ethics standards, and sets of
entities that aren't subjected to those standards. In some
cases people are opting to meet the higher standard, but some
are opting not to.
Ms. Hinkle. From our regulator's perspective?
Mr. Hamilton. This isn't really a CMS question----
Ms. Hinkle. Right.
Mr. Hamilton. But we do a great deal of health care
purchasing and we do want managed care entities, health care
providers, to compete in the domains of good quality,
efficiency of services, effectiveness of services, not to be
competing to get unfair advantage to do favorable selection, to
use genetic information in a way to gain market share by
cherrypicking.
So we think that the protections in this area are quite
important. That means the protections that are put in place
amongst the providers of genetic information, because there
will always be some health care provider out there trying to
get--no matter how strong the laws are, there will always be
some provider trying to access some indication about potential
consumers so that they can get market share. We want them to
get market share if they're providing good services, but not
through the use of genetic information.
Ms. Hinkle. Well, since you're at the mike and talking
about finances, maybe you can talk about the converse of that,
reimbursement for genetic testing. What is CMS's policy right
now, particularly under Medicare, for reimbursement for genetic
tests?
Mr. Hamilton. I'll have to beg off on that question since I
represent the quality assurance side of Medicare, but not the
payment side.
Ms. Hinkle. Judy.
Ms. Yost. I'm sorry, I don't know.
Ms. Hinkle. That's all right.
Anybody at the table have some background or expertise on
that? [No response.]
Everybody just looks at you, Kathy, as the default.
Dr. Hudson. That's one area we haven't spent a lot of time
on yet, although let me make one quick comment. There is an
important piece that's been not really talked about here, which
is the translation into clinical practice and the need to
collect enough evidence. There's a weird imbalance right now
where we have tests for 1500 diseases, and then a zillion
variations that we can look at, but a tiny, tiny number of
health professional guidelines.
What the health care professionals say is that they need
evidence, and they need somebody to look at that evidence, and
weigh that evidence and then give it to them so they can
develop guidelines. There is a CDC-funded effort under way to
do that, where they are carefully looking at genetic tests and
then doing the evidence reviews, drawing conclusions, and
making recommendations. Hopefully, health care provider
organizations will be able to pick up that evidence base,
translate it into guidelines, and get it into the hands of
health care professionals, so they'll know what tests are good
tests and where the evidence base is strong and what tests are
a shrug. At the end of the day it's going to be the useful
tests that really benefit.
Dr. Feero. I would extend that point, that the EGAPP
process is a very valuable process for reviewing evidence, but
you actually have to have evidence to review. Developing the
sort of translational evidence that's needed is going to be a
challenging and costly endeavor, and will be a moving target.
That's what I was trying to refer to previously. This concept
of developing a pipeline in this area to figure out how to
develop that evidence in a cost effective way is a major
challenge, I think, moving forward.
Ms. Avey. That's something that I can't expect that the
government would take on solely. That's where I think the
public-private partnerships are really going to make a huge
difference, and that's exactly the role we want to be playing
with our partners, with the technology providers, with the labs
that have CLIA oversight.
So I think all the pieces are in place and then it's just a
matter of us working together and not halting something that
could be very positive for health care.
Then I just wanted to mention, going back to the privacy
issue, we are a for-profit company and we would not have a
business model if we did not protect the privacy of our
customers. We know that. That's a fundamental thing and an
element of this company. So protecting that privacy is, of
penultimate importance to us.
So we do that, but we also give our customers the option of
sharing their data. If they want to download it and give it to
a researcher at their favorite institution to do studies, they
can do that. But we would say that we are even beyond HIPPA
compliance without the unintended consequences of HIPPA which
we hear about all the time from physicians, who say they can't
look at the same medical record in their office that they can
see at the hospital, and it really hinders their ability to
take care of their own patients.
So we don't think HIPPA is perfect, but, that said, we
totally believe in the privacy of this information. When you
see what UCSF did when they sold the record of 6,000 patients
recently--it was in the papers everywhere. HIPPA does not
prevent those types of things from happening.
Ms. Hinkle. What resources currently are available to
consumer or perhaps even medical providers who have patients
coming in either wanting genetic testing or wanting help in
interpreting genetic tests? Where are the resources? Do the
agencies have information on their web sites that's digestible?
A third party, or is there a hole there right now as far as
public education?
Dr. Feero. A number of years ago NHGRI, in collaboration
with a number of other entities, created or helped to create an
organization called the National Coalition for Health
Professions Education in Genetics, to try to begin to create an
infrastructure for delivering information on genetics to the
health professional community.
That organization has been working diligently with a number
of groups, most recently physician's assistants, prior to that
in the nursing organizations, to try to deliver the
information.
There are a wide variety of efforts on various web sites
that are funded by the government at least. A good example is
Gene Tests and Gene Clinics that provide information for the
less common disorders.
Ms. Hinkle. I have to say, I don't find that a particularly
helpful web site from a consumer's perspective.
Dr. Feero. It's not designed for consumers. It's designed
for the health professional.
At this point in time, the resources that are available in
the public space to help with the interpretation of these
association studies are much more scarce. In fact, the NIH is
working on this right now. There's a trans-NIH communications
group on the genetics of common chronic disease, that is
working to formulate a response to that sort of vacuum of
publicly available information on this topic.
But it's clearly something that I think a number of
different agencies, the CDC included, have their eye on.
Ms. Avey. Just to mention that we're in discussions with
them as well, because if we're already aggregating this
information and creating our own tools, we're happy to share
those and work with NCHPG directly if that makes sense, if the
information we're providing is something that they're also
interested in.
Ms. Hinkle. Kathy, you offered your perspective much
earlier in our conversation that there has been a real lack of
progress with respect to genetic testing oversight. Of course,
not everyone at the table shares that opinion. So I guess my
question to you is, what particular progress would you like to
have seen the agencies be undertaking at this point? What would
you really like to say this must be your top priority over the
next 1, 2 years in order to ensure adequate consumer
protections?
Dr. Hudson. I appreciate you pulling this roundtable
together because I learned a lot from my fellow panelists. I
was stunned to hear Steve say that he welcomes new authority,
and that'll be interesting to follow up on, exactly what
authority is he welcoming.
Ms. Hinkle. I was going to ask in a moment what authorities
he would like from Congress and we're just going to get that
written up.
Dr. Hudson. I'm excited about the efforts that CMS is
making in terms of moving forward with proficiency testing,
despite their lengthy rejection of our petition requesting
proficiency testing last year. So I think we need to move
quickly on proficiency testing. I don't think it's that
difficult. I don't think it should take that long. My colleague
Gail Javitt put together a notice of proposed rulemaking to
implement the Secretary's committee's recommendation and it
took her about 4 hours in her office alone. Of course, she's
extraordinary, but it's not that hard.
The most important thing second to proficiency testing is
creating a genetic testing registry and creating that
transparency. Nobody's against it. Yes, the details are hard.
It absolutely must be mandatory. It absolutely must be at an
agency that has proven informatics and genetics sophistication.
But I think that that could move forward quickly and help a
lot.
Ms. Hinkle. Where should that be housed?
Dr. Hudson. I personally believe it should be either housed
at FDA or NIH, because both have a track record of success in
making information instantly publicly available and easily
searchable. CMS, sadly--and I understand their resource
constraints--2 years ago Tom testified that you could find out
the CLIA certification status of labs on their web site. I
checked yesterday and still can't. So I think----
Ms. Yost. It's coming.
Dr. Hudson. It's coming. It's coming.
Ms. Yost. It's almost there. I saw it yesterday. It's
almost there.
Ms. Hinkle. So it's not online?
Ms. Yost. It's almost there.
Mr. Hamilton. It's almost there. The wheels of government
grind finely but slowly.
Dr. Hudson. At different rates in different agencies.
Ms. Hinkle. So we're not quite there yet.
Mr. Hamilton. In contrast to Gail's 4 hour piece, we worked
for a year to get permission to get exceptions to our hiring
freeze and a year to fill the position. So government--the
Federal Government is purposely designed with the famous checks
and balances. But notwithstanding, we're making slow progress.
I think we had the web site ready to go and then----
Ms. Yost. For like a day.
Mr. Hamilton. For that cameo time. Then the directives came
down to make it all 50 compliant, which it should be, of
course. But if you've ever had the pleasure of trying to make a
large enterprise 50 compliant so that individuals who are blind
can read it----I don't use charts. That's all I can say. At any
rate----
Ms. Hinkle. So could we anticipate that this year some
time?
Ms. Yost. I'm saying weeks.
Mr. Hamilton. Kathy is right in so many ways. We support
the registry idea and it ought to be within an agency----
Ms. Hinkle. So long as somebody else houses it.
Mr. Hamilton. If we were good at it, we'd jump at the
chance, frankly.
Ms. Hinkle. Well, Steve, Greg, here's your opportunity to
take up the mantle and go to bat at your respective agencies.
So who wants it?
Dr. Feero. That's above my pay grade. [Laughter.]
Ms. Hinkle. Show some initiative. Get that promotion you've
been bucking for.
Dr. Feero. I don't think that's what I'd get.
Ms. Hinkle. Is FDA well equipped to handle a registry like
that?
Dr. Gutman. Well, as I said before, we're from the
government and we're here to help. So we're always anxious to
do what's appropriate. I think when I was talking about
authority it might be really not so much authority, because I
believe we have the authority. I think Congress could perhaps
express some interest in direction. If you look at the SACGHS
report, it's a very complex report. It offers more than one
path to Rome. So a path to Rome could be more reliant on CMS,
it could be more reliant on FDA, it could be more reliant on a
public-private partnership, or it could be reliant on other
out-of-the-box thinking, all of which might fall within
existing authority.
It would be a matter of how you emphasize that authority.
So I think that there might be a role for, after a factfinding
by Senator Smith or by anyone in the Congress, to express
advice and interest. There are no easy answers here. These are
very complex problems. I think what Thomas said at the
beginning is that really, in order to do it right it really
does require the involvement of multiple stakeholders, not just
FDA, not just CMS.
I think the idea of a public-private partnership has to be
carefully thought out and carefully orchestrated, but it is
probably necessary.
Ms. Hinkle. Judy, I'll put you on the spot for a second,
since I know you frequently speak in public about the CLIA
genetic testing specialty and PT. So I think everyone in the
room is very familiar with the agency's history in abandoning
the genetic testing specialty. So can you tell us a bit more
about your plans for PT, because even just as recently as this
December when my boss met with Administrator Weems and
questioned him as far as the agency's plans for enhancing
oversight over genetic testing, I would say we got what was a
less than specific response about the agency's enthusiasm
toward mandating proficiency testing for genetic tests.
Maybe you could just----
Ms. Yost. Well, actually there's a very simple answer. You
don't need a specialty in order to change the proficiency
testing requirements. So we feel it gives us actually an easier
way to proceed to use the existing infrastructure to develop
proficiency testing, because currently it depends on how you
define a genetic test and, as you heard, no one has the same
definition, no two people.
But even so, that said, the genetic tests that could be
potentially genetic tests if you define them that way are
interspersed among all the current specialties now. So we just
envision that that would just continue, with the change that we
would make. So it makes that part of the process a lot more
simple.
As we progressed through the discussions with the SACGHS
oversight review and thought about what the options were and
what were the things that were most important and what were the
things that we could do and could not do, and what were things
better done through other vehicles, we felt that PT was
something that not only for genetic testing but overall the
regulations needed to be updated, because they hadn't been done
since 1992.
So that has become our priority, that's next if we can ever
get the cytology PT out. We are well under way with a plan and
players in order to accomplish that.
With regard to some of the other concerns about CLIA, I
know there's discussion about personnel and quality control and
some other areas. We feel that there are probably ways that we
can accomplish that through use of professional guidance and
incorporating it into our guidance documents and still get
there from here.
We've found in the past, because we have experience in
that, in doing that, that once you start to put a professional
standard, as long as it's within the scope of our authority,
into our guidance, people tend to follow it because it's an
easy route to compliance, and then it standardizes the standard
of practice as well. It becomes the standard of practice.
We saw it in microbiology with cut points for
susceptibilities and using a consensus standard, and it has
worked extremely well.
Ms. Hinkle. So you're confident that if you issue those as
sort of sub-regulatory guidance that you're still going to have
adequate enforcement authority or there's just going to be
adequate enough compliance, that you're not going to have to
worry about bringing down the stick against people?
Ms. Yost. You would always have outliers. There will always
be an aberration. There will either be circumstances or
individuals who either willfully or unknowingly will not be in
compliance. That's a fact of our lives on a daily basis. So you
can never say never. But we believe that for the most part it
does become----because in many of these cases too we're talking
about competition. So the idea is that if you're in compliance
you can use that as added value to your services to be able to
work in the marketplace as well.
Ms. Hinkle. Is there anything you can share as far as
specifics about this guidance or when we might look for that to
be issued?
Ms. Yost. I think again this is an area where we have to--
there is a lot of material currently available that a number of
the professional organizations have already developed. It's a
matter of vetting that information, working with the experts in
the field to provide us perhaps some additional guidance with
their expertise.
Ms. Hinkle. Wouldn't most of that work have already been
done, though, with the 6 years that you spent laboring on the
CLIA genetic testing specialty? I mean, how much more up to
speed does CMS really need to get at this point?
Ms. Yost. Well, I'm not talking about up to speed. I think
we have to decide what we really want and what are the most
important things. I don't think we can do it all yet tomorrow.
But clearly we've agreed that proficiency testing is important.
We have the work that's been done with CDC on the best
laboratory practices document that will be published next year.
In addition, I believe we should work with partners. Rather
than we being the selectors as the Federal Government, I think
we need to have private entities provide additional advice to
us to----
Ms. Hinkle. Well, Kathy and--they've been very free with
their advice.
Ms. Yost. The invitation's been open for 2 years. We've
asked--we don't claim to be the be-all, end-all. So we want to
have folks buy into whatever it is that we select to use,
rather than send--picking something and then have people not
happy. So it's better, again, to work on the front end rather
than the back end.
Mr. Hamilton. Well, if we can think about this
strategically, we've got 1500 genetic tests.
Ms. Hinkle. The number increasing every day.
Mr. Hamilton. Expanding every day. How many proficiency
tests do we have? 50, 60, 70? The number of tests is growing
faster than the proficiency tests. So we will always have this
enormous gap.
Into that gap come two things. One is alternative quality
control, alternative assessment, that needs to be pretty robust
and needs to be adequately enforced. That's our piece under
CLIA, to make sure that it is. There are current requirements
that apply when a proficient test is not being used, that the
laboratory twice a year verify. So that needs to continue to be
applied and made as robust as we possibly can.
With regard to the proficiency tests themselves, we need an
adequate strategy. These things do not just magically appear.
We had a 17-year history trying to get national proficiency
tests for cytotechnology. That's a very long saga, but in 2005,
17 years after CLIA was passed, we finally got national
proficiency testing there.
So what we really need for proficiency testing is a real
strategy. Instead of thinking about this as an all or nothing,
either proficiency testing exists or it doesn't, if we think
about it perhaps as a ramp, where you start in sequence with
the education, then with possibly quasi-volunteer, quasi-
voluntary participation, and then ultimately the legally
mandated national participation.
That's why in that sequence we want to work with
particularly the accrediting organizations and the professional
societies, both of which do a great deal of work on the
educational side of things. But on the accrediting
organizations, they have the ability to require their members
to do proficiency testing where they make the proficiency tests
available. So on the one hand, that's short of a universal
government mandate, and it's voluntary for an organization to
decide whether or not to be accredited. But having made that
voluntary decision, the accrediting organizations can make the
participation in PT mandatory.
So that's a step up from just education. From that ramping
up it's a shorter step to us either doing sub-regulatory
guidance, which is just short of a full mandate. We can make it
a full mandate by changing the regulation and adding these
various analytes to the regulation. So Ithink that's the
strategy. That's why Judy's got planned a November meeting with
the accrediting organizations, to really focus on this, and we
want that CLIA work group.
Ms. Hinkle. Why wait until then?
Mr. Hamilton. Excuse me?
Ms. Hinkle. Why wait until then?
Kathy, I'm sure you have some thoughts on this. My question
to you is, Steve indicated earlier that some direction from
Congress might be welcome or needed. Do you guys need a little
help from Congress, some direction here steering the ship in
this respect? You know, people are just chomping at the bit.
They're just waiting for CMS to do something. Clearly you heard
my boss express his frustrations at the very glacial pace at
which CMS seems to be moving here.
So there's certainly some frustration, I think, about the
stakeholder community as well as on the Hill, just kind of
waiting for CMS to do something.
Mr. Hamilton. Again, look at the nature of the proficiency
tests. We do not have a group of sallow scientists hidden in
the basement concocting proficiency tests. These come from the
professional community. So we want to work with the
professional community to develop those.
Now, which comes first, a mandate for proficiency testing
or the proficiency test itself?
Ms. Hinkle. Well, you know if CMS required proficiency
testing there'd be plenty of people willing to provide
proficiency testing to capitalize on all the people now who
have to undergo that testing. It's the whole ``build it and
they will come'' philosophy that we talked about.
Ms. Yost. That is very marketplace-driven, too. The
proficiency testing providers currently will not develop the PT
material if in fact they don't have a market to sell it in. So
that you can see that by the difference of----
Ms. Hinkle. But if they know CMS is going to require it,
then----
Mr. Hamilton. Well, you asked what Congress could do. If
Congress provided up-front funding and said don't worry about
any marketplace uncertainty, here's reimbursement, the
marketplace is guaranteed.
Ms. Hinkle. I've got 20 dollars in my wallet over here.
We'll start with that.
Mr. Hamilton. I'll match you and----
Ms. Hinkle. Yes, right, matching funds.
Yes, Elaine?
Dr. Lyon. Can I comment with that? As Mr. Hamilton has
said, what CLIA requires is to demonstrate the accuracy of your
tests. When there are proficiency programs available that we
can be a part of, we are a part of it. Some of the tests we do
are not in those proficiency programs. However, that doesn't
eliminate the lab's responsibility for doing proficiency
testing.
We will then get with other laboratories doing the same
test. We will work together. We will exchange samples. We have
these set up to be able to do.
If that doesn't work, we still have the responsibility to
make sure they are accurate, and we do it by what we call an
internal proficiency. So one cannot say that the proficiency
testing isn't being done; it is being done. For some tests, it
may not be under a formal program right now.
Would we be in favor of expanding that program? Yes,
because it makes it easier for us. We don't have to find
another laboratory to work with. But the proficiency testing is
being done. The requirements from CLIA to show that your test
is analytically accurate is being met for every test that we do
in our laboratory.
Ms. Yost. That's a requirement, that's a regulatory
requirement currently in place that we very closely monitor
because, interestingly enough, it's one of the most frequent
findings we have on surveys. So that obviously means we're
actually looking at it.
The thing to remember here, though, as folks are losing
perspective, is everything's hanging on proficiency testing.
CLIA is a package deal. CLIA has quality control. That means
every day you do a test you have to check to see that it has
worked before you report any results, and you have to have
qualified people, you have to have trained people. You have to
check their competency on an annual basis. You have to have a
quality monitoring program in place that you monitor every test
and every system in place in your laboratory to ensure that
your test results are accurate, that they're reliable, and they
get to the person who's going to use them reliably, and
confidentially, by the way.
So those other requirements, as well as having
recordkeeping requirements for every step of the testing
process, so that if somebody needed to go back and check to see
what happened from the time the specimen was collected through
the analysis until the result is reported, you could track that
test. All of that is part of CLIA and, together with
proficiency testing or alternative assessment, as this is
called, that is what makes an accurate result.
You have to really look at the whole picture and not just
focus on one requirement. Yes, it's a lovely measure of
outcome, it's a lovely measure of test accuracy. But frankly,
you don't get PT results until 3 months after you reported that
patient's result. What good does that do you? You've got to
know on a daily basis that your test is working.
Ms. Hinkle. I appreciate you making that point. I think
that's a point well made. I think we could probably spend the
entire 2 hours together talking about proficiency testing, but
it would be incredibly remiss of me not to talk about a couple
of the SACGHS task force report recommendations. I see that Tom
has brought his trusty copy there.
Instead of me sort of picking through various
recommendations, I think people would probably be much more
interested to sort of hear what your perspectives were on the
report recommendations. Was there anything that surprised you
by being included or not included? Is there anything from the
agencies' perspectives you think particularly feasible--I know
you've already talked about PT--particularly feasible to
implement or just completely, can't do it, that's crazy talk?
Anyone jump in. I know you touched on a couple points as
well, so anyone just jump in. [No response.]
All right, we'll go around the table. Kate, you've been
very quiet today. How about you? Give you a chance to get on
record and some camera time.
Ms. Cook. I'm actually here to carry Steve's bag.
Dr. Gutman. Well, I will reiterate what I said before.
Actually, I'll be more personal than perhaps I should, because
it was an incredible report, so I'm just in awe that they had
30 people who worked, beat their time line, argued over every
word and paragraph and sentence, and came out with something
that was reasonably intelligible. So I'm extraordinarily
impressed.
Ms. Hinkle. That's credit where credit is due.
Kathy, you were on that.
Dr. Hudson. Yes. Steve was there as well.
Dr. Gutman. Yes. I wasn't trying to praise myself.
I was actually quite peripheral.
There were a couple of things that amazed me. It amazed me
that after the report was written when you looked at the press
it was as though there was only one paragraph in the report and
that was the paragraph describing FDA. So I found that
disheartening, that we are so fearsome. You could have had that
written in one afternoon. You didn't need the whole report. So
you look at the press and it just completely missed the point.
The report was rich in that it actually wasn't just the FDA
issue. It was the education, the reimbursement. The only thing
that was--everything was reiterating from--you got a snapshot,
but I think if you want to look at the big problem, the whole
field hasn't quite addressed maybe all of these complex issues,
education, reimbursement, analytical validity, clinical
validity.
The new one in town--I don't remember, but I certainly
don't remember it in the SACGHS report; it wasn't in the tax
report for genetic testing--was this notion it is 2008, so
we're a little bit further along than we were even 8 years ago.
People are worried about evidence base of medicine and they are
worried about value base of tests. How could they not when 16
percent of health care dollars are being spent on--16 percent
of GNP's spent on health care and it looks like in my lifetime
it will be 20 percent and my kids will be paying 25 or 35
percent. We're a very rich country. I don't think we're that
rich.
So I think the epiphany in the report was that it's not
just FDA, it's not just CLIA. There's a problem here. AARP's
trying to work on the problem. Blue Cross-Blue Shield
technology evaluation center is trying to work on the problem.
But you've got to figure out a problem. Do we actually have
money to spend on these tests? Which tests should we be
spending the money on? How should we be deploying them and how
should we be making sure that they are improving the quality of
performance?
So it's a rich report. I hope everyone in the community
doesn't shelve it. I hope that--they can agree with it, they
can disagree with it, but I hope they respond to it and it
stimulates. It's worth not being ignored and not being shelved.
Mr. Hamilton. I was not in the group, so I can be
completely fulsome. It is an awesome, awesome report.
Congratulations, Kathy, Steve, and Chairman Toich, and before
him Reed Tucker, and everyone who participated. It created a
forum to bring very disparate points of view together to
fashion a very constructive set of recommendations that do make
it very clear that this is going to take all of us working
clearly together and for quite some time to make progress.
But it's an optimistic report because I think it charts a
pathway. We can all have our particular viewpoints about what
the solution or a technique might be, but I think the report
keeps the vision before us of, regardless of the approach,
here's the destination. So that's why I am carrying it around.
I think it ought to be required reading.
Ms. Hinkle. They should have that in the pocket size, like
the Constitution that the members carry.
Judy, what was your take-away from the report?
Ms. Yost. It was actually----
Ms. Hinkle. By the way, I'm assuming that you've all read
it from cover to cover.
Ms. Yost. It was actually a very enlightening experience to
work with such wonderful people. Truly I learned, I learned as
much as I complained. So I did find it highly valuable. Clearly
we are doing our very best to look at those things that we can
accomplish within the scope of our authority and to work
through them. We've begun in a number of different areas.
They're already under way.
We'll do our best. But again, it is pretty awesome because
it shows you a full picture of what's happening. It's not just
isolated to one aspect of genetic testing because it's a
continuum, it truly is. So it was an honor to be able to work
on it.
Again, I always learn more than I give. So it's always
worthwhile time.
Ms. Hinkle. Kathy.
Dr. Hudson. I already shared my top priorities. The one
area that didn't get addressed--and a lot got addressed by this
committee but the one area that I think may have gotten less
attention than it should have was how we need to move nimbly to
get information into drug labels about diagnostics. This is
already saving lives. There's a genetic test that can tell if a
person will suffer severe reaction if they take a particular
antiviral for HIV and it's not yet, as I understand it, on the
drug label. So how can you move really fast? How can we move
really fast so that we can get the benefits as fast as
possible?
Dr. Gutman. Yes, I do want to respond to that, because the
agency is cognizant of that and we think that a policy
statement we need to clarify is that when you do have a--when
you do have a diagnostic and a drug linked so tightly that the
selection of the drug or the avoidance of selection of the drug
is related to use of the diagnostic, we at least in my neck of
the woods would say that the drug becomes a slave to the
diagnostic, so that the cross-labeling becomes critical, that
cross-labeling has to be based on data. In fact, Dr. Woodcock
has been fairly explicit in suggesting that those tests do
require pre-market review if they're going to make drug
decisions.
We have a very colorful, inconsistent past history. I
suspect the transitional period will be rocky, but I do think
we are going to move in the direction of having higher
expectation when you have a really tight link. When you have a
weaker link, then I think--I still think we ought to be looking
at that policy, but I don't think it's as compelling.
Ms. Hinkle. Elaine.
Dr. Lyon. What I saw in this report was, first, they still
couldn't define a genetic test. It makes it very hard for us to
move forward if genetic testing is defined so broadly that it
goes into every aspect of medicine.
Then the question that they had brought up is whether there
is an exceptionalism about genetic testing that is different
than the rest of medicine. I don't think that there is a
complete agreement with that, either. I think overall the tone
was that there is not an exceptionalism, that genetic testing
does fall into the realm of other types of clinical laboratory
testing.
Those were the two ideas that need to be clarified for us
to really move forward.
Ms. Hinkle. Greg.
Dr. Feero. So first, going back to my comment about being
above my pay grade, what is not above my pay grade to say in
NHGRI is that we are very supportive of the idea of a registry
and think that there is great potential value in that.
Particularly with regard to this report, I think that the
introduction of the concept of utility as part of that registry
is something that is very valuable. Clearly, genomic
applications span the health care system. They're coming out at
a tremendous pace and, as much as any technology, they have
great promise for improving the health of the general
population. But they also have a cost associated with them and
I think we have to be cognizant of that as we move forward.
I guess my overall comment, and perhaps this is because I
haven't been in government quite long enough to understand
this, it is a report, and that ultimately a report is only good
as its implementation and how people's toes are held to the
fire over time for its a product. That would be my comment,
that I think it is an excellent report; the proof will be in
the pudding.
Ms. Avey. For us, we broke it up into chunks and everybody
got assigned their reading, and then we came together and had a
meeting about it, which was great. Our scientists all reported
back what they got out of the section they read. I think at the
end of it we all thought this is such a huge opportunity for
23andMe to be involved, to be part of this public/private
partnership with all the different agencies, because it was
very clear that genetics education is really lacking for the
medical community and for lay people. If that's a role that we
can play, we're very excited about that. That's really one of
the missions of the company, is to just get genetics out into
the mainstream.
But it was very clear that, even on testing that isn't
genetic, there's still a huge disconnect between what comes out
of a testing lab and what a doctor is able to report back to
their patient. So it's not just us--I don't think we've created
this problem by coming out and creating this new industry. I
think it existed before we came along. But certainly as far as
genetics goes, that's where we're going to be focusing and
we're very excited about the educational opportunities.
Ms. Hinkle. Since I promised everyone in the audience an
opportunity for Q and A, is there anyone in the audience who
has any questions for any of the panelists?
Ms. Fomous. My name is Cathy Fomous. I thank you for all
the wonderful things that you said about the report. I really
appreciate it and I will bring it back to the staff.
My question is for Linda, because I know the least about
what your company does. You have mentioned a few times during
the conversation today that you are very enthusiastic about
doing research with the database that you have and the
voluntary nature of the additional information. My question is,
do you think that you'll voluntarily follow guidelines for
human subjects research protection, such as an informed consent
process?
Ms. Avey. Absolutely. This is something that's coming up a
lot in our discussions with researchers and with other
organizations about the stifling stranglehold IRBs have over
the research process. We have a consent form and we've met with
a commercial IRB to have them kind of bless it. Because we
break the paradigm, we don't know what questions we're going to
ask yet. Frankly, coming from Perlegen and Affymetrics and
Applied Biosystems and working with researchers over the last
25 years, I can tell you that the current research paradigm
requires that you know what questions you're going to ask in
advance most of the time, other than things like the Framingham
study, where it's prospective and you learn information over
time that you might not have even thought to ask in a study.
So we're very open-ended in that we don't know what
questions we're going to ask and we don't know even what
information we're going to find. So the IRB scratched their
head like, ``This doesn't fit our mold.'' You have to be able
to say what is the end point of your protocol and what is your
protocol.
So for us it's frustrating. They've been very honest to
say, this just doesn't work the way we review protocols. What
was interesting to me, though, is at the end of the meeting one
of the people on the executive team of this IRB came up to us
and said, ``I have a daughter with a genetic disease and she's
going to love what you're doing.''
So we really need to work together with the IRBs, I think,
as well to see how do we do this, how can we make these non-
protocols, if you will, fit into an IRB paradigm.
Ms. Hinkle. Any other questions from the audience?
Mr. Daynard. I may be biased, but since I have all the
experts here and we're talking about this, Greg, I think you
mentioned that the associations seemed to be robust over the
last few years depending upon the polymorphism you're looking
at. Linda, you said that you may not disagree about the
associations, but you still can't tell people what to eat. I
think what we may be seeing in terms of claims with some of
these companies is the claim that having a susceptibility and
gaining information about your genotype and maybe even getting
recommendations about lifestyle and diet changes is going to
impact your health outcome.
What kind of evidence supporting the claim would you
require if it's not a clinical trial?
Ms. Avey. That's where we want to work with the agencies to
make those definitions. What does become information that has
clinical utility? I think as a society we have to come to an
agreement of how we get to those end points. I don't think we
have all the answers. I don't think FDA has all the answers or
CMS. But I think together we can come to some kind of
conclusions, because doctors are going to need to know this.
They're going to have patients coming to them time and again
with the reports and say, here, doc, what do I do? I don't
think we know what to say yet.
That's why we sound like a broken record. We keep going
back to the fact that we need more data, and that's really what
the mission of 23andMe is to get many, many, many people,
hundreds of thousands of people hopefully, involved answering
the same questions, that we can then make those clinical
connections and connect the dots.
But we don't know the numbers we'll need for that. It
depends on the condition, it depends on the genetic
association. It's really complicated.
Mr. Daynard. Exactly.
Ms. Avey. But that doesn't mean we shouldn't do it.
Mr. Daynard. What are respondents going to say to me when I
present them with a draft consent agreement to comply with the
Federal Trade Commission Act is, boy, this is complex, and
we're not quite there yet. We don't know. Maybe these
association studies are good enough. Then someone will say,
well, even a strong association is not good enough because you
don't know that the recommendations are genotype-specific. But
here's the FTC that's going to tell someone, you need this kind
of evidence and you don't have it; your association studies
aren't good enough if the claim is that this is going to impact
your health outcome, not just that there's an association. That
may be right, and that's a distinction that we're continually
trying to hone.
Yes, we'd love to talk to you.
Ms. Hinkle. Anyone else?
Dr. Feero. I would just add the comment that certainly that
whole question is an active area that the NIH is looking at
carefully, in conjunction with other groups. It is a huge
challenge. It's going to be dependent on the particular
disease, the particular markers, the particular invention
you're proposing. But that really is I think the $64,000
question for any registry that is going to use utility as one
of the bars that you set. A, what is the bar; and B, who's
going to set it, is really going to be a challenge.
Mr. Daynard. Well, I hope the FTC doesn't become premature
in this area, because the FTC very definitely, although it
enforces the law against unfair and deceptive practices, it
doesn't want to chill an emerging science, and that's our
standard. It's a rigorous standard, but it's flexible. So
either I see it the way it is or I make it up. Just kidding.
Ms. Hinkle. Any additional questions from the audience? [No
response.]
I have received one question via email from John Rockoff
with the Baltimore Sun and he was just curious to know about
the volume of genetic testing. Does anyone here at the table
know how many patients each year undergo genetic testing? [No
response.]
No one?
Or how might we get that answer? Is that something that's
even really tracked or identifiable?
Dr. Feero. The fundamental problem is we haven't defined
exactly what we mean by ``genetic tests.'' So it's very hard to
count when you don't know what you're trying to count.
Dr. Lyon. One point that I could say is that most of the
genetic testing that we do right now, is for the relatively
common genetic diseases, the Factor V Leiden, the cystic
fibrosis. I had the numbers pulled from our laboratory and
about 70 percent of them in volume have FDA-cleared reagents to
do them.
Ms. Hinkle. Well, I promised you all that it would be about
2 hours and in lieu of opening statements we would afford the
panelists an opportunity to make closing remarks if you wish.
So would anyone care to make any type of closing statement
before we adjourn?
Dr. Lyon. I just wanted to emphasize that the knowledge
that we've gained from the Human Genome Project has resulted in
meaningful discoveries in our understanding of diseases and its
care. I want to emphasize that genetic testing does have value.
It has tremendous value to diagnose difficult and rare
diseases. It has value in detecting people who are highly
susceptible to disease before symptoms begin and hopefully that
we can do something to intervene. With pharmacogenetics we are
able to target the appropriate medication.
Genetic tests are being regulated, as are all clinical
laboratory tests. We appreciate CLIA's stand to allow us to do
laboratory-developed tests. I believe they are critical for
many of these rare genetic diseases for which manufacturers
will not making reagents and the laboratories are responsible
to make sure that their tests are valid. We do participate in
the proficiency testing and we follow the CLIA and professional
standards.
There are some guiding principles that I think that we
should include. The tests do need to be developed based on
medical knowledge and should be ordered by a knowledgeable
health professional that can help in their use and
interpretation. All laboratories performing these tests should
be CLIA-certified for high complexity testing; and that any
exaggerated or unsubstantiated claims then could be
investigated by the FTC.
The innovation in molecular testing is extremely sensitive
to regulation and reimbursement. Too much of the former and too
little of the latter could prevent or delay the hoped-for
medical advances. I believe that we can target the problems of
marginally useful testing while allowing testing based on good
science to accomplish its promise in diagnosing, treating, and
preventing disease.
Thank you.
Ms. Hinkle. Anyone else?
Ms. Avey. I don't have a formal response, but I just want
to thank everyone for being invited. It's been great to be
here. It's been really illuminating. We're very excited about
the future. We think the United States can play a big role in
personalized medicine and can be a leader in this. I think
we've fallen short in the last few years in things like stem
cell research. I really think it's time for us to step back up
to the plate again.
The great thing about the U.S. is, especially in
California, we have a very diverse population. We're already
talking to places like the South Asian Heart Center. At a
hospital in Sunnyvale they're looking into why do Indian men
who have vegetarian diets and take care of themselves, die of
heart attacks? If we can do a study in the U.S., but that
information can translate to India, we can start a global
effort that could help other countries that may not have the
same resources as the U.S.
So we're very interested in signing up a very diverse
population of people and having them be enrolled in studies,
because when you look at these gene journals that we have,
there's a drop-down menu that you have to select which
population does it apply to, and a lot of times it's just
Caucasian. Research is done in a very Caucasian-centric way and
we want to change that. We want to enroll people of all
backgrounds.
The other issue we have is that people are of mixed race
and won't know how research results apply to them--if they're
like Barack Obama and he's looking at his results, does he pick
Caucasian or African? He doesn't know because we haven't done
studies in multi-ethnic people.
So we're very excited about the opportunity and we just
hope by working with the regulatory agencies that things don't
get stifled, but that we do it in a very responsible way.
Thanks.
Ms. Hinkle. Yes?
Ms. Yost. I just wanted to quickly say that the ``I'' in
``CLIA'' stands for ``Improvement,'' and so we realize that
that goes for us too. We know that we can do better and we have
intentions; and I think mostly just to extend again the offer
to please work with us to help us get the things that you need
to improve and enhance the oversight over genetic testing.
Dr. Gutman. I'll just reiterate the message that I conveyed
at the very beginning. This is an area that we all have a role
to play. We all have a responsibility, and sometimes when
everyone has a responsibility that can sometimes be a recipe
for no one. So I particularly appreciate the role of you,
Chris, and Senator Smith in bringing us together and the SACGHS
for performing the same function; and to say the ``I'' in
``CLIA'' does not stand for ``you.'' We look at it as being
``I'' as well. So we want to be part of the solutions here.
Thanks.
Ms. Hinkle. FDA?
Ms. Cook. We heard today about the fact that there isn't an
identified definition of genetic test. But I think that people
can agree that there are some tests that clearly are. Even
though the borders might not be clearly identified, there are
some that are.
Steve mentioned earlier that FDA has taken a risk-based
approach in some of its regulatory efforts with regard to
laboratory-developed tests. So I think it's worth observing
that here, and even though we may not have really the full
definition of genetic tests, there are identified genetic
tests, there are identified risks, and bearing those two things
in mind, it might really help this effort move forward to
address the most significant risks.
Ms. Hinkle. Well, I would extend my thanks personally and
certainly those of Senator Smith to everyone here. Senator
Smith is very engaged in this issue. We had a difficult time
getting him out of his seat. If you saw us frantically passing
notes to him behind, it was like, the vote's coming, you have a
meeting, you're missing your meeting, the vote's coming. He did
not want to leave.
So clearly this topic is very engaging. As you know, it can
be very difficult to engage the interest of a Senator. So I
think that speaks to how just compelling this topic is and
really how engaging you all are as panelists as well.
I have to say I was a little surprised when I first sort of
bounced this idea off of some of my colleagues here on the
Hill. Everyone was quite skeptical about the concept of getting
everybody together at a table. They were like, you're never
going to get the agencies to sit down with all these
stakeholders and have a conversation and do a roundtable on
this topic. I was completely surprised at that sentiment.
Certainly you're all very agreeable people, very well
intentioned, very knowledgeable, and it is always such an
education. Every time I have the true privilege of speaking
with any one of you, you are all always very generous with your
time with respect to staff and the members. This is the second
time you've come before this committee and you've given up
hours of your time today to share with us and the public, and
we very much appreciate that. You have added immeasurably to
the record today.
For those of you who are interested, an archive of this
will be webcast on the Aging Committee's web site and a
transcript will be available at some point in time, too.
I look forward to following up with all of you, and thank
you so much.
With that, we are adjourned. Thank you.
[Whereupon, at 4:25 p.m., the roundtable was adjourned.]
A P P E N D I X
----------
[GRAPHICS NOT AVAILABLE IN TIFF FORMAT]