Originally Published IVD Technology March 2002
In Person
Evolving alliances
Academic researchers are playing an active role in advancing the development of IVD product offerings.
In the past, academic
researchers with ideas that could be developed into new IVD products have had
to go to great lengths to seek out industry funding. But no more.
According to Jim Wittliff, PhD, MD hc, director of the hormone receptor laboratory
at the James Graham Brown Cancer Center and professor of biochemistry at the
University of Louisville (KY), IVD companies are today aggressively seeking
out new technologies—and they aren't waiting for academic researchers
to come to them.
A pioneer in tests for breast cancer treatment and prognosis, Wittliff is a
proponent of developing relationships with industry to expedite end results.
Currently serving as a visiting industry professor at Arcturus Applied Genomics
(Carlsbad, CA), Wittliff represents the new research laboratorian, acting as
a liaison between academia and industry.
In this interview with IVD Technology editor Steve Halasey, Wittliff talks about
the dynamic relationship between industry and academia, his current project with
Arcturus Applied Genomics, and why it's such an exciting time to be involved in
the genomics field. The complete text
of this interview is also available.
IVD
Technology: What is the relationship between academic researchers who
are studying the basic biochemistry or molecular biology of disease markers
and the IVD industry? How is knowledge transferred from one realm to the other?
Jim Wittliff: I can respond to that from my own experience with New England
Nuclear (NEN), with whom I developed the first FDA-approved tests for estrogen
and progestin receptors. In the past, the principal manner in which companies
learned about academic research was that they simply read contributions in the
field from the scientific and clinical literature, and then approached the researchers
about their product ideas.
In my case, I had already approached Kodak when I was at the University of Rochester
(NY), but they said if it wasn't a $5 million–a-year product, they weren't
interested. At the time, NEN wasn't a part of DuPont (Wilmington, DE), so they
were interested in smaller projects. Of course NEN later went on to become a
much bigger entity.
But the point is that companies located many of us through the literature, at meetings, going to workshops, or at presentations we gave. There was a much greater span of time between what we identified in the laboratory as product-worthy and when industry was ready for development.
How has that
process changed?
Within the past couple of years, industry has become much more aggressive. Companies literally have scouts out in the field, looking at the early progress of young investigators, while keeping an eye on the findings of seasoned researchers. A lot of manufacturers are getting tips from their representatives in the field, who are much more informed. I think that's a compliment to the diagnostic manufacturers.
In addition
to scouting for up-and-coming research and researchers, are companies also funding
that kind of research?
They are. Instead
of the investigators going to industry with a result and seeking support, in
many cases companies are going to the investigators first. That's how some
of my studies brought me into contact with Arcturus Engineering Inc. (Mountain
View, CA) The company learned that I was active in the field of tumor markers,
and they wanted to apply their laser capture microdissection (LCM) technology
to detection of tumor markers in human carcinomas.
That effort has expanded into the genomics area, where we are applying LCM to the capture of human breast carcinoma cells and developing, for the first time, molecular signatures of pure cell populations. Then we are correlating these with patient characteristics—their clinical response, prognosis, and other parameters of major importance.
When companies
talk with researchers, what are they looking for? Do they have in mind a critical
mass of experience or a level of expertise?
Absolutely. And most of them have a focus. They'll have a diagnostic focus in
cardiovascular medicine, or cancer, or transplantation and immunology. So they're
very well informed.
I am really impressed
with the attention that manufacturers pay to posters presented at scientific
meetings such as those of the Clinical Ligand Assay Society (CLAS), the American
Association for Clinical Chemistry (AACC), the Endocrine Society, and the American
Association for Cancer Research (AACR). Meetings such as these are where manufacturers
are getting ideas and learning who some of the newcomers are.
Once an investigator has a marker or a product, it can be quite expensive to
obtain rights. By that time, the researcher probably has a relationship with
a company. But if a company can catch someone early—someone who may have
made good progress but doesn't have national or international recognition—then
the company and the investigator have a greater opportunity to be first—and
that's what it's all about.
How early are
we talking? Are we talking about doctoral candidates?
Well, yes. But postdoctoral fellows have become key players in the intellectual transfer of information and technology. Some of these people go on to be courted by and work with industry. It's a win-win situation, because there is earlier recognition for young people—particularly talented ones—and technology transfer is quicker. In the long run, patients benefit because we bring to the marketplace more products with greater utility for clinical management.
Intellectual Properties
 |
|
Components
of the laser capture microdissection instrument by Arcturus Engineering
Inc. (Mountain View, CA).
|
In the past,
academics have been freer with information about intellectual properties than
most companies are. Are universities controlling the release of information
more than they used to, and how does that affect access to such information
in meetings like the ones you're talking about?
That's right. As recently as five years ago, most small universities didn't
pay a great deal of attention to many of the intellectual properties that their
faculty were developing. There were patent policies and so on, but there were
few restrictions on principal investigators, when they worked with industry
or with foundations, to prevent them from discussing technology development
or to ensure the protection of intellectual properties.
As a result, the
rights to many valuable intellectual properties were lost. And so many editorials
were written and every university began to set up technology transfer and intellectual
property offices to protect these rights. In addition, most of us get helpful
advice about what we can discuss and what level of presentation can be made
at meetings. That's really been helpful.
Where is that
advice coming from?
It's coming
from within the university, from the intellectual property office. According
to policy here in the Commonwealth of Kentucky, all university-based grant applications
have to go through a multilevel review before they are submitted.
I'll give you an example. I'm professor of biochemistry, and a member of the Brown Cancer Center. My grant applications go through the cancer center director's office, to the chairman of biochemistry, and then to the dean of the school of medicine, where a fiscal officer looks at them. They finally go to the office of sponsored programs, where intellectual property potential may be reviewed. When there is any indication that a grant application is going to be funded, that office immediately develops the appropriate contracts and agreements to ensure the protection of IP. Not one dollar exchanges hands until all of those agreements have been signed.
It sounds as
though the research community is becoming more savvy and knowledgeable about
intellectual property protection.
We are. Many of
us have worked with industry from the days when it was not considered attractive.
I started collaborating with industry as an assistant professor at the University
of Rochester. My chairman warned me that this was not a particularly good thing
for a young investigator to do, and that I should be addressing fundamental
research and not worrying about the applications. And yet, the university received
huge amounts of support from Xerox and Kodak.
But the culture
has changed. I appreciate it when my students and fellows meet people from industry.
We're very good about interacting with our company representatives, so
that they tell us about technology. We've even found out about the focus
of other investigators who might be our competitors or collaborators.
There's a
lot more communication. Because the Internet makes it so easy to communicate—to
send ideas and diagrams—there's a really exciting evolution going
on in laboratory medicine and related technology.
Research Properties
How were the tissue-based biochemical tests that you devised in the 1970s different
from traditional immunohistochemical tests?
Before that time,
virtually all laboratory medicine testing was performed on serum or on soluble
marker proteins. The only tissue-based tests that were done were immunohistochemistry
tests using fixed paraffin-embedded tissue.
But to perform quantitative measurements of the estrogen receptor, progestin receptor, Her-2/neu oncoprotein, and other markers, the laboratorian actually takes a piece of the tissue biopsy, grinds it, and makes an extract. Then the extract is examined using a biochemical test. It may be a radioligand binding test, or an enzyme immunoassay, or an ELISA, or some other highly quantifiable procedure to measure the analyte concentration.
How does laser capture microdissection advance this method of testing?
When one grinds a tissue biopsy, there are many types of complicating cells that are also homogenized. In addition to the carcinoma cells, there are lymphocytes, red blood cells, neutrophils, macrophages, and connective-tissue cells.
There are some
techniques that can reduce such complications—such as being careful to
trim off connective tissue from the biopsy being homogenized. But even so, when
we measure an analyte in such a homogenized preparation, we don't really
know the contribution of each cell type to the final answer.
By contrast, LCM
enables the operator to physically separate the cells of interest, with complete
biochemical integrity. Right now, I'm using this technique to separate
the human breast carcinoma cells in a biopsy from the normal cells present in
the heterogeneous tissue specimen.
Then, with my collaborators at Arcturus Applied Genomics, we are performing gene expression profiling of very small numbers—on the order of 200 to 500 cells. And it is almost routinely possible to produce, for one of the first times ever, complete gene expression profiles of 12,000 genes. That's extraordinary, and it's why I'm convinced that this kind of technology will soon be used in laboratory medicine testing. Imagine carrying your gene expression profiles of key tissues on a chip in your wallet or on a finger ring. Then when you exhibit a symptom or go for a checkup, those chips can be read and compared with a molecular signature of the tissue in question.
How did this
partnership with Arcturus develop, and where is it headed?
Arcturus Engineering
approached me to discuss the development and the use of laser capture microdissection
for measuring small amounts of analytes in laser-captured cells. In the meantime,
the company also developed an applied genomics program that functions as a separate
research and development entity.
Now, here's
a bit of serendipity. What Arcturus didn't realize when they first spoke
with me is that I have a large biorepository of biopsy specimens from many different
kinds of cancers from patients I have served over the past 25 years. Arcturus
introduced me to its LCM technology with the goal to study a few proteins. But
then with the addition of a new genomics R&D program, using the same LCM
technology and a microarray approach, one could examine the molecular signatures
of cancer biopsies from my biorepository.
When we applied
the Arcturus LCM and genomics technology to the biopsy samples the results were
revolutionary, in that expression of thousands of genes was measured on pure
populations of human breast carcinoma cells. Briefly, we showed that we could
capture these cells from 5-µm sections of frozen tissue biopsies, some
of which were 15 years old. After fixation and dehydration the captured cells
yielded enough RNA—1 to 6 ng from 200–1000 cells—that the RNA
could be amplified and gene profiling for 12,000 genes could be performed reproducibly.
That's extraordinary
and exciting, and it opens the possibility that gene expression profiles, so-called
molecular signatures, may identify various subtypes of carcinomas, leading to
a more clinically relevant classification and treatment selection. I can envision
a time when gene-expression profiling will become a part of routine medical
practice, and baseline profiles will be part of a person's record. And
then when a person experiences symptoms of a disease, the clinician can request
the laser capture of cells, and the gene expression profile of a tissue will
be compared with that of the control to note changes in gene expression. After
bioinformatic analysis the results will then give the clinician a better method
of selecting an appropriate drug or even predicting the patient's clinical
course and response to various drugs. That's a new type of laboratory medicine.
And I predict that's where we're going.
When do you think the LCM product is likely to be commercialized?
The data on the first 100–200 biopsies will be ready by this fall, so we're moving very quickly. We're hoping to present some of this work at the International Cancer Congress in Oslo, Norway, which takes place from June 30 to July 5 this year. And before that, in May, CLAS is sponsoring a biotech session at its meeting in Houston, where about 10 manufacturers, including Arcturus Applied Genomics, will be presenting.
Routes and Roadblocks
Based on your observations, where do you think the field is headed? What
are the most exciting developments, and the potential roadblocks?
Some of the roadblocks
that I foresee include the necessity for addressing first a kind of culture
change in the surgery and pathology suites. To perform gene expression profiling
intact, RNA is needed. But operating rooms are full of hot lighting, so the
minute a biopsy is excised, it can become dehydrated and the tissue and RNA
can be damaged. From such a sample, one might be able to get good structural
information to determine that the sample contains carcinoma. But to perform
gene expression profiling, the biological integrity of these molecules has to
be retained.
So one thing that
is going to have to happen in pathology is greater cooperation with the proteomics
and genomics lab, which I see as an integral part of the laboratory medicine
complex of this century.
Right now, we need to be building those components and working with surgeons and pathologists to show them the value of genomics and proteomics in clinical medicine, and the new protocols for treating the tissue. We have surgeon and pathologist collaborators in Louisville who work with us to do that now. But I think we're among the few places that really address this important issue. And that's mostly because the doctors here appreciate what we're doing and want to be a part of it—and they want their patients to be a part of it.
There is certainly
commercial interest in forwarding such research.
Yes, and it's not for just one technology. There's room to develop new kinds of gene chips, for instance, to find particular patterns of genes involved in a signal-transduction pathway rather than looking at 10,000 genes. For example, for certain diseases such as endometrial cancer, you may only need to look at 1000 genes or 100 genes. Already, miniaturized systems are being developed to focus on various pathways.
And it's also necessary to develop methods to record and display all this information for a doctor to read. The bioinformatics community deals with very sophisticated algorithms and projections and graphs. And those are not easy for doctors to read. There's a tremendous amount of development that must occur in that field.
These are all
interesting technologies that have strong potential for improving patient care.
How do you see their effect on the field as a whole?
We are literally moving to a new level of laboratory medicine and its application
to human disease. And it's not just proteomics, genomics, and bioinformatics,
but the idea that we will have a detailed insight—gene and protein profiles
of individual types of cells as reflected by the influence of their environment
in the body. By comparing the gene and protein expression profiles of a patient's
normal cells with those of diseased cells, one gains powerful knowledge. Can
you imagine the benefits if we do that?
We must seriously prepare our medical students and graduate students for this new culture, and instruct them in the applications of this type of reporting. Who knows what is around the corner about to be discovered, or what further developments will complement or enhance those discoveries?
Jim Wittliff, PhD, MD hc, is director of the hormone receptor laboratory at the James Graham Brown Cancer Center and professor of biochemistry at the University of Louisville (KY). He is currently a visiting industry professor at Arcturus Applied Genomics (Carlsbad, CA). He can be reached via e-mail at jim.wittliff@louisville.edu.
Copyright ©2002 IVD Technology
