For molecular, a brand-new way

January 2008
Feature Story

Anne Paxton

Advances in molecular diagnostics have transformed laboratory medicine. Robotic DNA and RNA extraction, DNA microarrays, multiplex protein detection, and gene-expression profiling are among the technologies making it possible to diagnose earlier and to create more precise treatment options for patients.

The Center for Molecular Medicine, or CMM, launched in Grand Rapids, Mich., a year ago, is a melding of research expertise and hospital and laboratory resources that is using these technologies to probe cancer, heart disease, mental illness, and other conditions at the DNA, RNA, and protein levels.

A $6 million joint venture between a hospital, Spectrum Health, and a research institute, the Van Andel Institute, the CMM aims, as one of the three pillars of its business model, to accelerate the tailoring of treatment based on the molecular makeup of a patient or a patient’s tumor, or both.

In a CAP TODAY interview with writer Anne Paxton, the center’s executive director, Daniel H. Farkas, PhD, talked about the new center, test platforms, molecular diagnostics’ effect on point-of-care testing, reimbursement and economic studies, IVDMIAs, gene patents, and more.

What is the Center for Molecular Medicine’s mission?

The center is really designed to do three things. One is to be a 21st century molecular diagnostics lab. Two, on a fee-for-service basis we plan to court not only classical clinical trials from pharmaceutical companies but also less classical trials of new content and new platforms from diagnostic and biotech companies. Three, the center will serve as an outlet for the molecular diagnostics content being developed at the research institute.

We’re a CLIA-certified, CAP-accredited clinical laboratory, and so we can bring those tests online in a regulatorily appropriate environment.

What led you to join the center?

I’ve established three hospital-based molecular diagnostics laboratories in the past, most recently at The Methodist Hospital in Houston, but I was interested in coming back to Michigan.

I found out about the center and they worked to recruit me for a long time. I joined in December 2006. We have five technologists now, an office manager, and an administrative support person, and can gear up quickly to add new people when new projects come in the door.

I see the center as an opportunity for me to use my skills in laboratory medicine and my skills in business development to bring genomics to a new level here in Grand Rapids. I think it’s a paradigm for a future path for molecular diagnostics, and I was excited to be part of a forward-looking, entrepreneurial community that was interested in blazing trails.

What kinds of testing are you conducting?

In my previous hospital-based molecular diagnostics jobs, my main mission was to add new items to the test menus for the sake of our patients and physicians. Here, that’s not my main goal. It’s an important part of what we are doing but ancillary to our main objective, so I’m intentionally keeping a light test menu of cutting-edge tests.

We’ve been open for about a year and we have three tests on the menu. Roche Diagnostics’ AmpliChip CYP450 test, a microarray-based test to examine patient genotypes, was the first test we brought online, in March 2007.

In October 2007 we began offering the CellSearch system from Veridex to identify and count CTCs, circulating tumor cells, in metastatic breast cancer patients. And on December 14, we became one of the first laboratories in the nation to offer the CellSearch system to test for CTCs and guide treatment of metastatic colorectal cancer. The CellSearch tests represent an important advance in how breast and colorectal cancer can be managed over the long term.

I don’t think I’m going to bring on that many tests very rapidly, because I want to support the clinical trials part of the business, and patients always come first. If you have too many tests on the laboratory test menu, then that tends to trump everything else.

Xceed Molecular just launched its automated Ziplex system to perform complex molecular assays, and the center is the first member of the Xceed Molecular collaborator program. What kind of work will you be doing with Xceed?

Our cooperation with Xceed Molecular fits into our overall strategic mission. Xceed is a perfect example of a biotech company that eventually would like to become a diagnostics company. They see their relationship with CMM—which sits across the street from a hospital and a research institute—as a great way for them to gain experience in the clinical laboratory and simultaneously develop content for their eventual molecular diagnostics test platform.

We’re doing two things with Xceed. First, in our laboratory we are demonstrating the equivalence of their system to the gold standard method for gene-expression profiling. And second, in collaboration with the Van Andel Institute, we’re helping Xceed move new array-based content toward commercialization.

As translational medicine moves from the research laboratory to the clinic, technology that is affordable, reliable, and automated will be critical. We believe the Ziplex system is very user-friendly for the clinical laboratory and therefore uniquely positioned to fulfill these needs.

Which molecular diagnostic test platforms do you envision being the most robust or most relevant over the next few years?

Clearly, real-time PCR, which has re-revolutionized molecular diagnostics yet again in this decade, will remain the most robust platform that will generate the highest volume for the foreseeable future. It’s just a great platform.

Gene-expression profiling is next. It is at the beginning, it is nascent, and we’re still gaining expertise as a laboratory medicine and medical care community in terms of how best to utilize gene-expression profiling. So that’s the future. But right now, the short answer to your question remains real-time PCR.

What impact do you think molecular diagnostics will have on point-of-care testing?

We’re actually working with a couple of companies to help them develop that technology.

Point-of-care molecular diagnostics used to be oxymoronic. But now, with real-time PCR and with advanced technologies that allow us to extract nucleic acids quickly and remotely, one can actually think about point-of-care molecular diagnostics. There are a good 10 or 12 companies I know about that, in various phases of development, are working to develop POC molecular diagnostics. They’re not ready for prime time—with the single exception of Cepheid and its platform called GeneXpert.

GeneXpert is a true sample-in, answer-out device. It’s not exactly point of care because it’s not portable, but it can be placed in a delivery room or emergency room or a surgical suite for the sorts of applications that make sense there. Two examples would be testing for group B strep in the delivery room and assessing surgical margins in tumor and lymph node excision in the OR—sort of a molecular frozen section.

Is the center involved in any studies related to reimbursement or economic issues?

CMM is working with two major payers, Priority Health and Blue Cross Blue Shield of Michigan. We’ve made more progress with the former, but I suspect we’ll have the same level of success with the Blues.

We’ve had a very active, very open dialog with Priority Health about the value of AmpliChip for certain subsets of patients. It’s a test that allows physicians to more rationally choose drugs and dosages to manage a patient. The disorder or disease that Priority Health, a local mental health facility, and CMM will be working on together includes depression and other mental illnesses for which drugs are prescribed. These drugs are metabolized by enzymes that are encoded by the genes interrogated on the AmpliChip.

So together we are designing a medico-economic study that will allow us to assess if the test has economic value. We already understand the medical value; we’re just trying to figure out where it fits in with the payment scheme of things.

Would there be a lot of demand for looking at metabolism of those drugs?

If you’re a depression patient, then yes. Any individual candidate for a drug metabolized by those enzymes which are interrogated indirectly by looking at the genes on the chip would be terribly interested in that.

I don’t want to bankrupt the system by suggesting every patient should be tested, because 25 percent of the drugs in any hospital’s pharmacopoeia are metabolized by the enzymes assessed by this chip. But we have to start somewhere, and we chose depression and other mental illnesses.

A lot of patients have unpredictable responses to antidepressants, and might have to try a whole series before finding one that works. Is that the kind of issue this study would address?

Exactly. So we’re going to try to do an economic study that finds real data to determine how many dollars did we waste in managing the patient because the drugs didn’t work and they’re on their third or fourth drug, how many days of work did they miss, what is the lost productivity—those sorts of economic questions.

And how far along is that project?

We are about a third of the way in. We hope that with 100 patients’ worth of data, medical economists at the insurance companies can draw some good conclusions.

How sophisticated are the economic models that would be needed to associate your research findings with payment?

The models are actually quite sophisticated. That’s exactly why I’m partnering with Priority Health and hopefully the Blues, because I know what I know and I know what I don’t know, and I certainly don’t know how to medically/economically model these sorts of questions. But I know from discussions with Priority Health that they have experts to do this sort of thing.

Would insurers have a strong incentive to pay more for the kinds of tests that could perform these kinds of studies?

There are two ways to answer that question. If economic data suggest that the test is beneficial at a certain price, then yes, it becomes easy to convince payers.

Even if the economic data don’t demonstrate that it’s financially advantageous, insurance companies do care about their patients, and if a $900 test, for example, saves $700 worth of drugs (in other words, there’s a net loss of $200), but their patient feels better sooner, I like to think that test is going to happen anyway. And I believe that it would.

Are there comparable types of studies going on throughout the country at payers or other research operations?

On the one hand, I hope so. On the other hand, from a selfish point of view, if we’re the first then that’s a feather in our cap. And if nobody else is conducting these studies, then I hope this article suggests to other parties/ insurers/laboratory partnerships to get on the stick.

How do you see CPT coding evolving to adapt to molecular diagnostics testing in the future?

That’s a hard question. The CPT code approach has to adapt to reimbursement for molecular diagnostic testing. The way by which dollar figures are associated with specific molecular testing CPT codes doesn’t work so well. Some tests are vastly over-reimbursed; some tests are wildly under-reimbursed.

I think the existence of array-based codes is a good step forward. The AMA, which of course puts out the codes, has demonstrated an understanding that arrays are a future part of laboratory medicine, specifically molecular diagnostics.

What has to change is the other half of the equation. Simply publishing a code without associated rational reimbursement is only half the battle. And if you don’t take it all the way across the goal line, then you haven’t scored a touchdown. We need rational, intelligent, data-derived discourse about what an appropriate dollar figure is for the CPT codes. There needs to be more dialog between those who peg a dollar figure to the codes and the practitioners of tests, namely medical professionals, so the right dollar figure—not too high, not too low—is set.

There has been a certain amount of critique about the Food and Drug Administration’s guidance on IVDMIAs (in vitro diagnostic multivariate index assays). Can you comment on that?

I think the FDA is appropriately concerned about new classes of tests coming into the marketplace. IVDMIAs are array-based tests, but the data goes through a “black box” software component so that I as a laboratory director don’t necessarily understand the algorithms or what’s going on with the data to generate the laboratory values that I then provide to my physicians.

And if one subscribes to the notion that the pathologist is the “doctor’s doctor,” that the pathologist acts as a consultant to the physician who orders the test, then I’m a little bit concerned about IVDMIAs, because I don’t feel I can give my physicians a complete story, a complete consult, the complete information.

I think that’s where the FDA is coming from. They want to put the tests through their paces to make sure they’re as robust as they should be.

Some bills on Capitol Hill right now, such as the Genomics and Personalized Medicine Act, address genetic discrimination. What impact might they have on your operations?

I have been involved in discussions on genetic information non-discrimination since 1997 when Rep. Louise Slaughter of New York introduced legislation. It’s 10 years later, and we still don’t have a bill that’s been signed into law to protect individuals against genetic information discrimination. So I’m not concerned [about impact on operations], because when these things become law, hopefully they will be well vetted and there will be lots of input from organizations like the CAP and the Association for Molecular Pathology, so that a bad law isn’t put into effect. In general, what I worry about are onerous, overly burdensome regulations being put on the laboratory when CLIA regulation, combined with CAP’s Laboratory Accreditation Program, provides excellent oversight already. I hope Congress doesn’t send anything up to the president that reinvents the wheel for laboratories that already know exactly what they’re doing and have been doing it for many, many years.

When those laws hit the books I will of course comply with them, but right now I’m interested in providing as much input as possible to make sure they do become the best possible laws.

Frankly, with 2008 being an election year, I can’t imagine you’d have to worry about it before 2009 at the very earliest.

The College is concerned about gene patents that restrict the types of tests pathologists can perform as diagnostic medical procedures. Can you comment on bills now pending in Congress that would prohibit patents from being obtained under certain circumstances?

It’s okay for folks to disagree, but it’s a complicated issue. Patents are legal; patents are the bedrock of our economy. It’s perfectly appropriate to prohibit patenting a gene sequence for all sorts of reasons, but if there is a particular polymorphism or signature or particular misspelling in the code or the gene-expression profile that a company has invested significant dollars and effort and resources into turning into a diagnostic test, I believe they have earned the right to enjoy patent protection.

Does the center have any plans for expansion?

We just doubled the size of the lab in anticipation of all the work we’ll be bringing in. So we have about 3,300 square feet of ultra-modern laboratory space.

We will see how we can do with five med techs and scientists supporting the work, but I expect that with the amount of business development I’ve been doing, and the amount of interest in the marketplace, we’ll have many more than the eight or nine projects I have right now, and the center is only going to be getting bigger.

What is the future of molecular testing?

Molecular diagnostics is poised for continued, excellent double-digit annual growth because of fields like pharmacogenomics, companion diagnostics, and personalized medicine.

I think molecular diagnosticians need to think about the future very carefully. There are a handful of cash cows on the molecular diagnostics laboratory test menu, including HIV viral load, Chlamydia/gonorrhea, cystic fibrosis, HPV, and maybe a couple of others. But beyond those, things move into low-volume, high-cost esoterica.

And if, selfishly, we’re not going to see our tests lost to the microbiology laboratory, then laboratories can look to CMM as something of a model for the future, whereby we are divesting ourselves of the higher-revenue, high-volume tests in exchange for relatively modest and light test menus. That way we can concentrate on developing new tests in conjunction with our academic colleagues, our research institute across the street, or with companies looking for opportunities to partner.

I would tell the readers of CAP TODAY that we’re being very, very successful at that at CMM, and if they want to copy the model, they can. I don’t mind because I think there’s plenty of business to go around, and if molecular laboratories think about those sorts of collaborative opportunities, then they’ll continue to thrive.

And if they can figure out a way to hold on to the profitable tests and still develop new collaborations, then more power to them.