Medical Device Link.

CLIA Complexity

DCLD is now in its second year of undertaking complexity categorization as required by the Clinical Laboratory Improvement Amendments of 1988 (CLIA) simultaneously with its premarket reviews of new-product submissions. What's your impression of how that system is working?

We're making progress, but the issue of criteria for the category of waived complexity devices has proved to be a contentious one. Last August, about a month after I came on board at FDA, there was a very lively public meeting with strong supporters on both sides of the issue. Laboratory-based attendees—pathologists, technologists, and commercial laboratories—tended to have a restrictive view as to which tests should be waived under the terms of CLIA. On the other side, manufacturers of IVD tests and certain clinical groups representing physicians held the opposite point of view, favoring a more relaxed and open approach.

In a nutshell, we're trying to find a balance between concern for patient access to testing and concern for the quality of that testing. Often, we give access to physicians and patients who want to get their testing done in a doctor's office, knowing that such tests will not always be conducted by a medical technologist. But we recognize that there is an opposing concern for quality—concern that a nonprofessional should be able to perform a particular assay in such a way that it will be safe and effective enough to do the job. This is really the issue.

In the past few months, we have waived two influenza-type tests, and we have a number of other submissions that we're looking at right now. We continue to be in discussions with the other relevant agencies—the Centers for Disease Control and Prevention (CDC) and the Health Care Financing Administration (HCFA)—to see if we can all agree on what the next step should be.

Most recently, the Office of Device Evaluation has issued a draft guidance document on how to approach the whole business of CLIA waiver for a number of tests. We are now in the middle of the 90-day public comment period, when anybody can present their comments about the guidance. Copies of the guidance can be downloaded from the CDRH Web site at http://www.fda.gov/cdrh, and comments are due by May 30. We're very interested in knowing what the comments are and, if necessary, modifying what we've come up with.

The draft guidance requires manufacturers to undertake several relatively large agreement studies to demonstrate that their tests will have equivalent performance in the hands of untrained and professional users. What was the basis of ODE's thinking in requiring so many test participants?

I think it was a compromise number. Some wanted us to be more rigorous and look at a larger population, while others thought we could get by with fewer subjects.

The issue here is what type of comparison should be used. We've come up with the idea that the lay user, the nonprofessional, should compare his or her results with the professional in terms of relative accuracy and comparative values. And we've suggested that a spectrum of users should try this out. This approach tries to achieve a balance—opening the door for more CLIA-waived tests, but retaining the idea that we need to know how such tests will be performed in the hands of lay users.

There may be other alternatives to this approach, and I would welcome any suggestions. This is an excellent opportunity for industry to think of some alternatives, if companies feel that the proposed number is prohibitively large.

Because DCLD is now performing CLIA complexity categorization in tandem with its premarket reviews of product submissions, IVD products no longer have to undergo a double technology review, once for FDA and once for HCFA. Manufacturers in other medical device sectors argue that HCFA technology reviews are similarly unnecessary for their products, and that they are an obstacle to prompt reimbursement-coverage decisions. Would it be advisable to combine these technology reviews in some way?

When FDA conducts a product review, the key measures are expressed in the terms safety and effectiveness. At HCFA, the analogous terms are reasonable and necessary.

So FDA is looking at the clinical performance of a device in all of its aspects. To use an IVD example, the agency is looking at the analytical performance of the product to see whether it is what the company claims. That's the effectiveness piece. In IVDs, the issue of safety can arise when the consequences of false-negative or false-positive results are considered. So that's the safety piece.

HCFA's concerns, on the other hand, center on whether the use of such a test is reasonable and necessary for the screening, diagnosis, or monitoring of the patient. Those are different questions. And to make matters even more complex, such decisions are made both by HCFA at the national level and by regional carriers.

The effect is that companies have two hurdles to overcome. But there should be a way to combine what we want to do because it's very important to get these products out quickly. It's very important to get the reimbursement and coverage decisions made, and it should not be necessary for industry to do all that work twice.

In the future, perhaps FDA and HCFA will find a way of speaking with one another. We are doing more work with HCFA in terms of sharing some of our experiences. But maybe this is something that industry should present to the secretary of health and human services (HHS), since both agencies are part of HHS. FDA can't do anything about it alone. HCFA may not be able to do it alone. It's something that must be departmental rather than agencywide.

For many years, CDRH resisted taking on the duty of CLIA complexity categorization because of the lack of resources to do so. Now that that system is working well, would it make sense for ODE to perform other types of technology assessment on behalf of HCFA?

Compared with HCFA, ODE has much less involvement with these very tough questions related to the criteria of reasonable and necessary—the financial benefit, the financial cost, the financial risks. In ODE, we can review any number of laboratory tests or therapies and determine that they are all safe and effective. By contrast, HCFA must often look at two, three, or four different modalities of diagnosis or therapy and then have to strategize its response. Should the agency agree to Medicare coverage for all of them, or just some of them? Just because a product clears the first hurdle—the FDA hurdle—does not mean that it will necessarily or easily clear the coverage hurdle.

Because I was a laboratorian for many years, paying attention to problems that affect turnaround times is second nature to me. Each year, the office receives thousands of submissions and issues thousands of decisions. And industry watches the office and judges its performance based in part on how long it takes to process a PMA or a 510(k). So having a mindset that focuses on the timely delivery of results will prove useful at ODE.

When you are considering how much rigor to require of an IVD manufacturer's product submission, does it make a difference whether the test is being used for screening, early detection, diagnosis, or monitoring? Does it make a difference whether the methodology is a cultured test or a molecular test?

I think so. The Medical Device Amendments talk about intended use, and I think it is important to know what the intended use will be. Sometimes tests are used for purposes that are not included in their intended-use statement. But intended use is important.

For instance, a test might have only 80% diagnostic sensitivity but still be ideal for use in certain settings because it has 98% diagnostic specificity. By contrast, a test that is being used purely for screening—where you want to be absolutely certain that you don't miss the disease—could be perfectly adequate with just so-so specificity if it's 99% sensitive.

The clinical setting is also very important. In certain situations, let's say for a home-use glucose test, one will not be as demanding as one might be in a hospital setting. In that case, having easy access to the test makes up for not having laboratory-quality results. It's so much better to have access to that test in the home setting. So these are important factors to take into account, and this is also where subjectivity plays a role. We hope that in DCLD we have the right people—those who have had a background in these issues and look at them from that perspective. That's why it's good to have experienced laboratorians there as well.

How important is clinical experience as a measure of a product's suitability for market? It sounds as though you would not be in favor of approving to market an IVD whose label simply stated that it would detect a certain analyte with certain levels of sensitivity and specificity.

Not necessarily. If you're looking at sodium, potassium, or amylase, all you need is a test that can verify a certain substance is really sodium, really potassium, or really amylase.

But when you have a particular test that's touted as a cancer marker or a test for heart attack, then it may be appropriate to see if the test really performs that function. That's the distinction.

And that's also where a lot of the debate is. Is it adequate for a company to say that a new test for heart damage correlates with CK-MB or troponin, or should the company also have to demonstrate that measuring the analyte with this test will diagnose a heart attack? In short, does the company have to be able to correlate its results to people with heart attack?

Depending upon the analyte, it probably is necessary to do that. One can imagine a scenario in which two tests compare very well with one another, but one is a good predictor of disease and the other is not.

U.S. and European Regulation

The European Union and the United States have quite different regulatory regimes. How would you describe the differences between them?

In the European Union, and in many other countries, the product approval process is only concerned with the issue of safety. So for some types of therapeutic devices, it may be easier for a product to get to market in Europe because all the company has to do is demonstrate the safety of its product.

But in the United States, FDA is also required to evaluate the effectiveness of devices. This additional criterion significantly changes the way that the agency reviews product submissions.

In the case of laboratory tests, we're right in the middle of dealing with this issue of how rigid our standards should be—how high should the bar be to evaluate effectiveness? Is it enough for a company to say that its test is as good as another one already cleared using a 510(k)? Or should we require the company to compare its test to a standard method?

Another consideration is how demanding we should be in requiring a company to compare its test results to some measure of clinical significance—to demonstrate the diagnostic sensitivity and specificity of the test. Is it enough for the company to say that its new test compares favorably with another test? Or should the company be required to show test results in relation to an accepted measure of a clinical disease or condition?

These are all elements of an internal discussion that we're having right now, and we expect that this discussion will also include industry. I do think we can come to some kind of an accommodation based upon protecting the public health, but also promoting the technology.

So that's where I think part of this European-U.S. issue is. We are in communication with the Japanese, we've had visitors from the Australian equivalent of FDA, and we're collaborating with our colleagues in Canada. Through the Global Harmonization Task Force, we're trying to harmonize different standards around the world. We may not always agree with one another, but at least we can see what each country is doing and compare notes with one another.

What do you think will be the long-term outcome of the discussions related to harmonization?

I think we're going to land somewhere in the middle, but I don't know exactly where. I do not think the United States can reduce its requirements to safety alone. The requirement to demonstrate effectiveness is very important, but the issue is how demanding we should be for that requirement. Without a demonstration of effectiveness, HCFA wouldn't know what to do. It would have to start its assessments from the beginning, and patients and physicians would be very concerned. I do believe we're going to find a middle ground somewhere.

What do you make of the IVD Directive's requirement for the traceability of calibrators? How will that requirement change practices at DCLD or for manufacturers?

My personal opinion is that one should ideally have a standard, traceable method or material. It doesn't have to be the best reference method or the most difficult reference material to get hold of. A company shouldn't have to use a gas chromatograph (GC) mass spectrometer analytical test to look at its new test. On the other hand, one should use the best reasonably available standard method as a basis for comparison with one's own tests. Companies shouldn't just use any old test.

And this is another area of contention. Is it adequate under FDA's 510(k) requirements—where the relevant measure is substantial equivalence to a predicate device—to compare a new test with just any other test that's on the market? This is another area where there is likely to be a fair amount of discussion and debate. And in the end I think we'll find a middle ground somewhere.

A lot of industry people have pointed out that this requirement of the IVD Directive is not very practical because there is no reference method or material for the vast majority of analytes that one would want to measure clinically.

In my opinion, it's not necessary to use a reference method or material so long as there's a reasonable standard method or procedure. Let's take blood glucose as an example. At one time that test had a reference method that involved GC mass spectrometry or isotope dilution or something along those lines, and every new test would have to be compared using that method. But that's all really unnecessary. Instead, the manufacturer can employ a good glucose oxidase or glucose hexokinase procedure, using off-the-shelf chemicals of a reasonable purity. Or the company can use another generally acceptable method that's recognized as accurate. And it's fine to compare the new test with that. I think that's a fair approach.

I don't subscribe to the idea of always having to use a reference method. However, I do subscribe to the need to have some sort of standard procedure that is used for comparisons.

This issue plays an important role in the case of tests for certain hormones that one might want to measure using an antibody-based procedure. Often, the analyte that is being measured is not a pure chemical by itself, not a free hormone, but one that is modified by being conjugated or combined with something else. The pure material would give you one answer, but patient sera would give you another. In this case, using a reference method could provide misleading results. So what one wants is a standard procedure that measures both the free and total amounts of the hormone, or one that measures the free and conjugated amounts.

The most important thing is to document what standard method is used, so that the end-user can later compare results. Otherwise, there's great confusion.

What is the role of organizations such as the National Committee for Clinical Laboratory Standards (NCCLS) in helping to formulate industry standards and related reference methods and materials? What should that role be?

They have a very important role, which I support. ODE is interested in having as many standards as possible so that manufacturers can use them to get through the premarket process more quickly and with less difficulty. To that end, we encourage volunteer organizations to develop standards.

FDA often serves on these various committees. As a matter of fact, I've been invited to be part of an NCCLS planning group, and I immediately said yes because I think that we can contribute. I'll be able to share what FDA’s thinking process is, and how the agency looks at things. In many ways, NCCLS serves as a crossroads for industry, the user, and government, and this is a very important place for us to compare ideas.

As an agency, we're very concerned with what our stakeholders have to say. In March, for instance, DCLD director Steven Gutman and I met with a group of IVD company representatives brought together by AdvaMed and the Association of Medical Diagnostics Manufacturers (AMDM) to discuss how to apply the “least burdensome” provisions of FDAMA to IVDs. Part of my goal as director of ODE is to have our organization be open and listen to what end-users have to say.

The NCCLS standards that FDA has recognized were originally designed for laboratorians rather than for industry. How could they be adapted or rewritten to meet the needs of industry?

Some of the standards are applicable. Those on how to evaluate the analytical performance of tests, for instance, or on how large a sample is required to assess the quality of a product.

But beyond its role in developing documents and standards, NCCLS offers some great benefits as a communication vehicle. The role of NCCLS will evolve as needs evolve and as its constituencies demand it to do so. I think the organization already has an increasing interest in looking at things from an IVD industry perspective, not just from the lab-user perspective.

Genetic Testing

In the area of molecular testing, there has been a fair amount of controversy over what regulatory requirements should apply to labs, as compared with manufacturers of kits. Where do you think the agency should stand on this issue?

FDA has the statutory authority to regulate devices in interstate commerce and also to regulate their manufacturing processes to ensure that they are conducted appropriately. There's been no explicit understanding that FDA has the same authority over so-called home-brew testing.

But if large laboratories will be doing home-brew genetic testing in one place while receiving specimens from all over the country, we have a dilemma. If a business involves sending specimens from one state to another, shouldn’t it be regulated as interstate commerce? Who will regulate what happens in a lab where such genetic testing is performed? Certainly, someone should inspect these laboratories. Some organization should ensure that we know what tests are out there. And someone should see that physicians and patients are informed about the implications of the test information.

Genetic testing is a very hot topic at FDA. We're very involved with the Clinical Laboratory Improvement Amendments Committee (CLIAC), the National Institutes of Health (NIH), HCFA, and other groups in discussing a wide variety of concerns. Such concerns include informed consent, how to specify what a test means analytically, what a test result means in clinical terms, the consequences of providing that information, and what kinds of genetic counseling will be necessary before and after testing. And there are additional issues related to ensuring the confidentiality and privacy of test results with regard to employers, insurance companies, and even other private parties. Many of these issues are terribly complex from the bioethical, scientific, and medical perspectives. So it’s very important for the agency to look at all of them very carefully.

There is a debate going on within FDA over what the agency’s role should be. We're trying to sort out whether to play a consultative or collaborative role in relation to those groups that normally inspect laboratories, or whether we should broaden our responsibilities.

I’m only speculating, but I think that the regulation of molecular testing will probably come together something like the following. Without a doubt, manufactured devices for molecular or genetic testing will remain subject to FDA scrutiny and regulation. For molecular-testing laboratories, I think inspections will probably be carried out by some combination of groups including CDC, HCFA, the College of American Pathologists, and certain states' regulatory groups. However, I think FDA will also play an important role because of the agency’s expertise in looking at tests, their labeling, and the information they provide to the patient.

In addition to raising ethical questions, the regulation of genetic testing also raises some significant marketplace issues. The manufacturers of genetic tests incur significant expenses in order to meet FDA requirements, while labs using home-brew tests escape such requirements. Such a regulatory imbalance would seem to create a disincentive for development of the market in molecular diagnostic test kits.

That's also a concern that we have. Is it truly a level playing field when any lab can create a home-brew test and avoid a lot of the scrutiny, premarket approval time, and detailed analysis required for a device?

I think the difficulty is a practical one. ODE approves about 45 premarket approval (PMA) applications each year, of which perhaps 10 are in the clinical laboratory area. The office also clears about 10,000 premarket notifications (510(k)s) each year, of which maybe 800 are in the laboratory area. To bring home-brew testing under FDA regulation, we would probably need to add about 900 different genetic home-brew tests that are performed in various laboratories around the United States—and there may be a similar number of nongenetic home-brew tests.

So how could the agency find the resources to evaluate all such tests, and how should we do so? Should they be considered PMAs or 510(k)s? Should individual laboratories that make home-brew tests be required to undertake the same kind of extensive studies that manufacturers perform? And if a home-brew laboratory changes its test over time, should it be required to file a PMA supplement or 510(k) modification, as a manufacturer would be required to do?

In terms of equity, there does appear to be a discrepancy between the ways that home-brew laboratories and IVD manufacturers are regulated. But resolving such an issue requires us to consider both how to do it, and how to do it within existing resources.

Again, I think that we can find a middle ground. It might consist of a system that would require laboratories to register all of their home-brew tests—or perhaps only their genetic home-brew tests. We might also have to devise an approach that includes a PMA Lite or modified 510(k), where the inspecting agencies would take charge of collecting specific information about each laboratory’s home-brew tests. For instance, labs might be required to provide the name of the test, the basis of the test, what is known about the genetic nature of the test, how the lab ensures informed consent, what kinds of information are presented to the patient, and so on. Even such a general registration would offer some degree of scrutiny over what these laboratories are doing.

In any case, it's very clear that for the foreseeable future there will be two camps on this issue. Clinical laboratories will argue that the regulation of home-brew testing has never been the domain of FDA, and shouldn’t become so. And IVD manufacturers will argue that industry needs a level playing field. And once again, we'll have to weigh the pros and cons of both sides—remembering also to consider the issues in terms of patients and in terms of promoting new technology.

The secretary's advisory committee on genetic testing (SACGT) has recommended that FDA regulate genetic home-brew labs. Would the agency prefer not to take the lead role, but only to take on certain specified responsibilities as part of a coordinated multiagency effort?

We're right in the middle of those discussions, which have mostly been attended by CDRH director David Feigal and DCLD director Steven Gutman. All I can say is that there's a lot of give and take and a lot of internal discussion about what FDA’s role should be. We're trying to find one that's both reasonable and workable.

FDA is located in the neighborhood of most of the developments in this arena, and we can readily see that this whole field is just going to explode.

The greatest concern I have is not about how these tests are done—because I think the labs can be monitored one way or another—but about the public perception of such genetic information. If a woman with a family history of breast cancer tests positive for both the BRCA1 and BRCA2 genes, for instance, what course of action should be followed? Should all the female members of the family be tested, even if they do not have breast cancer? How far should those with positive results go? Should they get a prophylactic oophorectomy? Should they go on Tamoxifen? Should they get a bilateral mastectomy? Should they live in fear every day of their life, get a mammogram every six months, and get an MRI once a year? And what happens if the lab screws up?

With such serious options in the balance, what kind of information should be given to the public? If we decide that such information should be shared with the public, then it’s very important to ensure that genetic counseling is available both before and after the results.

All of these problems are made more difficult by the fact that they involve probably dozens of laboratories. There are scores of academic institutions and research labs, and very few of these have traditionally been involved in QA/QC or TQM activities in the ways that large commercial laboratories have been.

The BRCA gene tests are a good example of home-brew testing that is not currently regulated by FDA. Looking at that example, some manufacturers might wonder whether they would be better off to buy a laboratory and avoid the regulatory headaches.

The same thing goes for toxicology testing, or vitamin testing, or any other kind of testing that can be done in a large, centralized commercial laboratory for esoteric testing.

But where do we draw the line? So many tests are conducted as home-brew or in-house testing that it becomes a very difficult issue to sort out. The issues also relate to the fact that FDA has not traditionally been involved in the practice of medicine, and the agency has tended to view the performance of home-brew laboratory testing as analogous to the practice of medicine.

It is certainly a concern that current procedures might tend to inhibit the growth of new technologies, because manufacturers might decide that they would be better off simply buying a commercial lab. On the other hand, there are many inherited diseases that affect only a small proportion of the population. For those, there might not be enough financial incentive for a manufacturer to justify producing a kit, but still enough financial incentive for a particular laboratory to justify offering the test. If such a home-brew testing laboratory had to undergo the same scrutiny as a device manufacturer, however, it's doubtful whether the potential profit stream would justify investment in such testing.

When all of the discussions have been completed, will there be a set of guidances or regulations that delineate responsibilities and codify procedures?

I think that's the intention. In the beginning, we may not be able to be as specific as SACGT would like us to be. We may need to initiate a pilot study or some pilot programs to see how certain proposals work. We're still in the early stages of discussion and development, and we're trying to bring together as many knowledgeable people as we can—people from academia, from commercial laboratories, from various organizations and associations—to weigh in on this issue.

We're on the cusp of an exciting genetics revolution. Take the whole field of pharmacogenetics. Using DNA-chip technologies, one can find out before surgery whether a patient is a fast or slow metabolizer of certain types of drugs—information that can be very important in administering anesthesia or postsurgical analgesics.

How should such products be regulated? If they're sold on a disk or a chip they would be classified as devices, and they'd be regulated by FDA. But what if that same methodology were to be home brewed? There probably should not be a difference in how those two products are regulated, but that's where we're going to have some very difficult questions to deal with.

If an FDA-regulated kit already exists, perhaps the makers of equivalent home-brew tests should be required to demonstrate some correlation to the commercial kit. Then we'd have to find some other way of dealing with home-brew tests for which there are no commercial equivalents.

Just as in the case of genetic tests, manufacturers that want to sell products regulated as general-purpose reagents or analyte-specific reagents must meet regulatory requirements greater than those for companies that sell bulk chemicals. But even CLIA-regulated labs are not required to use such products. Why is this?

I think a lot of such inequalities are historical, in that device manufacturers have traditionally had the bar set higher. But this doesn't occur just in the laboratory field; there are similar examples in the regulation of devices used by internists as opposed to the requirements imposed on an internal medicine practice. As soon as one takes a generic product and makes it into a device by selling it for specific medical purposes, the regulatory requirements become much greater.

The best way to deal with inconsistencies in the regulation of device manufacturers and clinical laboratories may be for us to become more demanding of labs, especially for those tests that have great clinical implications. But at the same time, we have to watch that everything we are demanding of device manufacturers does not become too stringent. We have to have realistic expectations.

Products of the Future

What do you see as the very hot areas that ODE will need to be prepared for over the next two to five years?

One of the things about working in government is that we can see and hear, but we can't speak about what we've done. So I can only base my comments on devices that have already been approved or whose information is already part of the public domain. But in some particular areas, we have seen some very interesting products, and I can assure you that the products we are now looking at will be equally exciting.

In cardiology, there has been a real revolution in the use of stents. About 75–85% of patients who undergo angioplasty now receive stents. Today, we're seeing different types of stents using many different types of materials and a variety of therapeutic coatings. We've also approved some fascinating electrical stimulation devices—including various types of defibrillators—and we're watching to see how they will do.

The world of neurology is also very exciting. In the past, we've had electrical stimulation products that are intended to relieve pain. We've also seen electrical stimulation used as a treatment for Parkinsonism.

Orthopedics is a very important area, with growth stimulated in part by sports injuries and degenerative diseases. We've seen a variety of prosthetic devices that can be placed between vertebral bodies, whether in the neck, cervical vertebral column areas, or elsewhere in the back. We've also seen a lot of total joint replacement applications for a variety of different products.

Ophthalmology has been very, very hot, and we think that area will continue to grow. Everyone probably knows about LASIK surgery. In the future, the progeny of LASIK surgery will also likely come to market.

Wound healing is a very big area, encompassing medical problems ranging from decubitus ulcers to postsurgical complications. We've seen applications for a variety of different devices used for wound healing. There are also devices used to inhibit the scar formation that can occur after surgery. And you've probably heard about the robotic surgery system that we approved last year.

We are also seeing applications for more and more combination devices. These include certain kinds of syringes, or stents, or other devices that are associated with one type of a drug or another.

In terms of the IVDs that are part of the public domain of knowledge and information, the most important growth area is that of genetic testing. There must be close to a thousand different genetic-testing assays used in the United States today, many of them in very small research laboratories, some in larger, specialized laboratories. We expect more and more genetic testing in the future.

We also have to be prepared in the area of testing for bioterrorism. We hope that such an event will never happen, but we have to be prepared if it does.

Physician's office testing, or near-patient testing, is also an area in which I expect a lot of growth. We are probably going to see more and more home-use testing of one type or another. On a more macro level, as people age and are at home more—and not necessarily in nursing homes—we're going to have various ways of monitoring older patients. Such monitors may present information to a data-processing module so that a physician or healthcare worker will be able to watch a patient from a distance to see what different results are and how the patient is doing.

With monoclonal antibodies, DNA technology, genetic testing, miniaturization, microelectrodes, ion-selective electrodes—the list goes on and on—we'll see more and more technology. Putting it in another way, I think we're going to see two things going on simultaneously: the drug field will continue to grow—but for some problems drugs will only get us so far—and then devices will probably be necessary to fill in the gap.

You mentioned home-use tests and the growth of remote-monitoring capabilities. Does the explosive growth of information technologies used in conjunction with devices present a particular challenge for the center?

Yes. There are a number of considerations that must be taken into account. We need to know whether a system's programs are written in such a way that they can be evaluated. The agency has to determine how much of a system's software we are obligated to evaluate. And we need to find ways of ensuring that there are not problems in the software, and that there are not difficulties in the transmission of the signal to the software.

Today, almost everything is hooked up with some kind of data-processing module and interpretive aspect, so in that regard we need to look at the total system. I think we've become more efficient in doing that as well.

In the context of the practice of medicine, how do you see the field of IVDs changing over the next decade? Will the importance of IVDs change? What are the major trends that manufacturers should watch for?

First, patients are going to become much more involved in the treatment of their own diseases. It's been estimated that more than half the time that people go to the Internet, they're looking up information relating to their own disease or one that their family members have. This trend will play a very important role for all of medicine, but especially for the IVD area because the first thing that patients will want to know is what tests are available to them. And after that, they will want to know what the information means, and so on. I imagine that commercial laboratories and IVD companies will more and more seek to communicate to patients the benefits and implications of certain kinds of testing, and patients will be highly aware of that information.

Second, there will be more and more near-patient testing of one type or another. I don't know what it's going to look like. In some cases, patients might put a drop of blood or a bit of urine on a pad that they then bring to their doctor's office or send to a commercial laboratory. Some types of testing will be done in nursing homes, and perhaps also in larger business offices.

Third, regulation and reimbursement will continue to be major factors. On the regulatory side—beginning, I hope, with the current draft guidance on waived complexity tests under CLIA—we’re likely to see a more realistic view of what those regulations should be. And manufacturers should be able to take advantage of this regulatory path by creating tests that are easier to perform and kits that are designed to be virtually foolproof. On the reimbursement side, however, matters are likely to remain difficult for some time. No company will sell a test kit, and no laboratory will perform testing, unless there's some sort of reasonable reimbursement. It’s difficult to know how that might be brought about, but I think it's going to happen.

Fourth, the patients of the future will be very different from those of the past. Patients will be much more likely to be suffering from chronic conditions such as arthritis, renal disease, Alzheimer's disease, stroke, neurological problems, or sequelae related to anticoagulation therapy. New test features will be designed to meet the demands of this kind of patient population, including the use of information technologies to monitor patients remotely and over extended periods, and greater capacity for data handling. Such a different patient population may also stimulate the development of home-use tests that can be performed by the patients themselves, or at least by a visiting caretaker, without the need for an office visit or hospital admission.

Fifth, there will be a lot more genetic testing. And these tests will not be directed solely at inherited diseases, but will also be used to understand the genetic bases of pharmacokinetics in individual patients so that therapies can be tailored precisely to the needs of each patient. Along the way, such tests will give us a greater understanding of how genetic inheritance and environmental conditions interact to shape an individual.

How will the business environment change for the IVD industry in the foreseeable future?

I think there's going to be more of a global marketplace, and we'll see a lot more connectivity among the various countries of the world. We are living in a global marketplace, and global companies—very large multinational organizations—are playing an increasingly important role.

It seems likely that there will be more consolidation among a few major IVD companies. And the larger ones, of course, will buy out some of the smaller companies that are doing interesting, cutting-edge work. Some of the companies with a specialized interest in women’s health might be getting to the point where they would be logical targets for such acquisition offers.

Certain types of testing are undergoing a renaissance of sorts. I think there will be more and more interest in microbiology, which is the last area to be automated after chemistry and hematology. And I also think we're going to see more such automation in the future. I'm certain that in the quiet rooms of the research labs today there are more and more developments going on. I'm very optimistic and very positive about the potential for those technologies.

It's hard to know precisely what's going to occur, but it’s possible that clinical laboratory testing will become unified with diagnostic imaging in certain centers. The driving idea would be to offer a wide range of diagnostic modalities—IVDs, CAT scans, MRIs, and so on—all in a single location. And those centers will also try to do more with their diagnostic information as time goes on.

One problem that may occur is a shortage of medical technologists, which will probably affect some specialty areas more than others.

How economic issues will affect the sector is very uncertain. A great deal depends upon the health of our economy and what kind of support healthcare is given in the future. Those factors are unknown. But every IVD organization will have to evaluate whether there's enough potential for a profit margin that will enable it to continue forward, so economics will always play a role.

What roles will FDA play in that business environment?

We at FDA hope to be able to promote the new technologies I’ve mentioned. I do not want the agency to be looked upon as the last big obstacle that companies have to overcome. I don't want it to be viewed as an organization that makes it more difficult for companies to invest in, develop, and market their devices.

The agency needs to be certain that tests and other devices work reasonably well. We need reasonable assurance that such products are safe and effective. At the same time, we want to work collaboratively with our stakeholders. That's a very important goal for me personally and for DCLD.

Copyright ©2001 IVD Technology