At A Glance
Name: David Link
Position: Executive vice president, Expertech Associates, Concord, Mass.
David Link, a vice president of the medical devices consulting firm Expertech, talks of the US Food and Drug Administration with the knowledge of an insider. Having spent 10 years there, starting in 1976 as manager of the agency’s medical device program, he helped implement its initial efforts in the regulation of medical devices.
Today, the agency is just beginning to consider how it will handle a new genus of devices, microarray-based multi-analyte tests. On April 21, the FDA issued the first draft of a guidance document on which it is seeking an unusual two rounds of public comment (see SNPtech Reporter, May 23, 2003).
On June 10, the pharmacology and toxicology subcommittee of the agency’s Advisory Committee for Pharmaceutical Science will meet at the Center for Drug Evaluation and Research in Rockville, Md., to review issues relating to the format and content of genome-scale gene-expression data generated during non-clinical pharmacology and toxicology investigations.
Clearly, it is a time of transition and transformation at the FDA, when it comes to genomics tools like microarrays.
BioArray News, SNPtech Reporter’s sister publication, spoke with Link to get an insight into how the agency might approach regulating in vitro diagnostic devices that are based on microarray technology.
You were there when the FDA first began to regulate medical devices. What was that like?
Up to that time, FDA regulation of devices was minimal. Before 1976, you could make a device and not have to clear anything through FDA; you didn’t have to tell them you existed. As of 1976, you had to register with them, you had to list the products you made, and you had to clear most devices through FDA in one way or another; and you had to manufacture them in accordance with fairly rigorous quality systems concepts.
Does the FDA change?
No, not much. FDA is driven by statute and regulation; they get their authority from statutes, and they take that authority and convert it into specific regulations and tell the regulated industry what it has to do, and what obligations it has to meet. Those things don’t change overnight; they are in place for long periods.
As technology changes, and as new kinds of products come into the marketplace, the agency will try and figure out which of the regulations they want to apply to the new products and how to do that. For instance, in the device business, if you offer an IVD and it’s going to tell you if you have cancer or not, the FDA would say that’s a pretty important product. It has significant risks associated with it. If it gives you a false negative and you come down with cancer, well that is going to spoil your day.
They would expose that particular product to a high level of review, one that would require some pretty elaborate clinical trials, and all kinds of information on how it is going to be made. They would scrutinize the labeling that goes with it, and all of those things would have to be done to their satisfaction before they would approve it and permit you to sell it. If it doesn’t have that high level of risk — let’s say an IVD where you take a blood sample and it analyzes it and tells you your blood level of calcium — that’s a simple, well-understood analyte. It probably wouldn’t be exposed to that high level of review.
Do you think the FDA sees microarray-based diagnostics as a different kind of diagnostic?
In IVDs, this ill-defined field of molecular diagnostics is going to force the agency to think hard about how to control these [new tests]. The applications are broader than a traditional IVD. The FDA is feeling its way through it.
This is essentially a new technology, and the FDA is relatively unsure of how they are going to control these specifically. That is, are they going to put these in the most rigorous level of review or are they going to feel comfortable enough to go through the agency in a less rigorous form of review?
When you read the guidance that [the] FDA put out recently, applicable to these types of products, it doesn’t take long to realize that they are seeking help in the way of comments to help them understand these new products, these new technologies and where they lie in the scheme of things. It is unusual at the outset to say they are going to make two rounds.
Why is that unusual?
All guidance invariably involves one round of comment. FDA will write it, and then will announce it followed by a period of comment. Then it will take those comments, review them, analyze them, and incorporate what they think are the persuasive ones and put out a final version. To seek comments twice is very unusual. That indicates to me they are not sure what they have on their hands, and they are not sure how they want to regulate it.
Some of those developing these technologies are labeling them as ‘for research purposes only.’
I am sure that a lot of people out there conclude that the product they are offering doesn’t even come under the FDA’s jurisdiction. They may think, ‘Hey, these are just a research product. We don’t have to do anything with the FDA.‘
That depends on the claims of the product. In the FDA’s jargon, there is a term called ‘indications for use’. ... Whatever you claim your product will do will dictate whether or not you are telling the outside world, ‘Here is a diagnostic product, it is useful for this purpose.’ If you go so far as to make those kinds of claims or indications; then you are probably a product, an IVD subject to FDA’s jurisdiction. Once you reach that point, you are well advised to find out what the rules of the game [are] for that kind of product.
FDA doesn’t like to see a product out there having escaped their jurisdiction and having the manufacturer make some kinds of claims for it where it could be, or will be, useful in any sort of a diagnostic mode. If those claims are made, they would consider that product an IVD. That comes under their jurisdiction and they would want to pass judgment on it.
These are variations among the broad field of IVDs that perhaps distinguish them from the more traditional, conventional medical devices. The FDA has a couple of terms for those kinds of products — they are called IUO and RUO — investigational use only, and research use only — and there are regulations that apply to them. Then, they have another category of IVDs — ASRs. Those are devices that may not make any claim, but they offer to analyze a specific analyte. There are regulations that pertain to those too.
What kind of advice would you offer to companies creating this type of technology?
If you are going to be in this business, you have a partner in Washington called the FDA. You would be well advised to fully understand what your obligations are as you move through the development process, through commercial distribution. New, small companies ought to think long and hard about what their regulatory strategy should be and what that is going to involve so that it doesn’t take years to get there.
The first piece is to know full well what the FDA’s position is on these kinds of products, and know full well what are the regulations and guidance documents that apply to these kinds of products. [The company] ought to become familiar with FDA policies on the products and say that ‘if something changes, does that apply to my product?’ You can call somebody at the FDA, they have a new office of IVDs, not more than three or four months old. A lot of people won’t do this because they don’t want to reveal themselves to the regulator.
What are the international issues?
Internationally, no matter where you manufacture a medical device, if you are offering it for sale in the US, you are subject to US rules and regulations. The FDA routinely makes a number of foreign inspections. You register with the FDA, you tell them what you make. They decide on [the] basis of that and any past history, whether or not they want to inspect your plant. The quality system regulations apply in China, Taiwan, the UK, Italy, and South Africa, everywhere.
In the case of IVDs specifically, the situation in Europe is different. The Europeans have published a directive, an IVD directive ó one of three directives that apply across the medical device spectrum. So if you sell in Europe, you will have to be familiar with [the] directive and meet its requirements; otherwise you won’t be able to put [a] CE mark on the product. If you don’t have it, you won’t be able to import it into the EU. So, there are fairly comprehensive, clear rules and regulations in the US and Europe, which are somewhat different. Both parties want to set down these requirements and make sure that manufacturers selling in their part of the world fulfill them.
The FDA has tons of authority. They are just trying to figure how best to apply it and when to apply it.
Molecular diagnostics has applications in the drug industry, and that is going to require working closely with the pharma companies. As soon as you get a product where the device and the drug people will want to have their fingers in the pie, then you have an added degree of complication.
I can say this: The difficulty of dealing with the agency goes up as a square of the number of centers involved. If you have to deal with devices and drugs, you are four times worse off than just dealing with one of them. The question will always be: Who has the primary jurisdiction?