How Dexcom Plans To Eliminate The Finger-Stick (And Bring CGM To The Masses), Part 2
By Jim Pomager, Executive Editor
In Part 2 of this Q&A, we continue our conversation with Dexcom CEO Terry Gregg, getting his insights on artificial pancreas technology, how Dexcom competes (and partners) with the major players in the diabetes device space, what startups should do to procure funding, and other topics. (Click here to read Part 1.)
Med Device Online (MDO): Last month, the FDA approved new Dexcom software that incorporates the algorithm being used in international artificial pancreas research. What does this mean for CGM technology?
Terry Gregg: When compared to self-monitoring of blood glucose (SMBG), or finger-stick measurement, our Dexcom Gen 4 Platinum has an accuracy of about 13 percent MARD (mean absolute relative difference). That was the best that’s ever been seen historically, even by researchers. In contrast, our predecessor product, the Seven Plus, had an MARD of 17 percent. With the new algorithm, we can achieve 9 percent MARD, so a tremendous improvement in accuracy.
We believe that this improvement in accuracy will help us achieve one of our long-term goals, which is to replace finger-sticks. We want patients to have the ability to dose their insulin without having to do a confirmatory finger-stick. In our opinion, 9 percent MARD is the threshold that achieves that. Right now, we are working through the protocol with FDA right now for a clinical trial sufficient to demonstrate that level of accuracy, and we are very confident we will achieve it.
We know — and again, it’s anecdotal and off-label — that many patients already dose off of their Dexcom. We don’t recommend it, but that’s the reality. We know it. FDA knows it. So we are trying to bring this technology forward in a way that makes sense, and that’s what this algorithm allows us to do.
MDO: What role do you expect Dexcom to play in the big picture of artificial pancreas development?
Gregg: An artificial pancreas has at least five component parts: a CGM, an algorithm, an insulin pump, an infusion set, and insulin. In some bihormonal systems, that number can reach nine, with another pump, another infusion set, another algorithm, and another drug. However, the whole system is triggered off the glucose values from the CGM, so achieving a MARD of 9 percent is huge for the artificial pancreas programs.
There are 22 or 23 artificial pancreas trial sites around the U.S., and Dexcom is in 20 of them with our G4 and G4AP CGM systems. We are highly invested in that particular arena, as the glucose component in those artificial pancreas projects. Everything we do in terms of product development, we ask ourselves, “Where is this going to fit within the artificial pancreas world? How can we help those people?”
Dexcom won’t drive the artificial pancreas initiative. We will remain focused on our mission to replace finger-sticks. One thing that I learned long ago is to be really good at one thing, and don’t stray too far outside of that area expertise, because it will dilute your efforts.
We are the market leader in CGM. There are only two approved CGM devices on the U.S. market; the other is from my former company. We have over 50 percent share in the market despite being a much smaller company. Medtronic has actually made the public statement that Dexcom spends more money on CGM than it does, and that’s true, which is why we are in this position.
Staying focused on glucose, without getting too far outside of that, has been very successful for Dexcom. We have grown quite rapidly. We just announced our earnings for the third quarter, and we grew about 60 percent compared to the third quarter of 2013. We had to up our guidance for the remainder of the year, whereas a lot of medical device companies — not just in diabetes — are actually doing the reverse. They are telling Wall Street that they are not going to meet their projected numbers. Here we are, on the other hand, saying we are going to beat the earlier numbers that we gave you.
MDO: So you will focus strictly on the CGM component of the artificial pancreas project?
Gregg: We will partner. We have a partnership with Johnson & Johnson’s Animus pump company, which last week received FDA approval for an insulin pump and CGM system that features Dexcom sensing technology. We also have a partnership with Tandem Diabetes, another pump company. We have a partnership with Insulet. We have a partnership with Asante. We want those folks to be highly successful, and we have an open architecture belief that they can access our technology.
MDO: Conspicuous by its absence in that list of partners was your former employer. Have you approached them about collaborating?
Gregg: Yes — that is their choice, not ours. We have offered to partner with them on more than one occasion, and they have elected to not take me up on my offer. But again, as I sit here today, if they wanted access to our technology, to incorporate it either into their artificial pancreas effort or just their pump and sensor combination, we are happy to have that discussion with them.
MDO: Obviously, Medtronic is also a competitor in the CGM space. What is your strategy for competing with large, established players like them in this space?
Gregg: We look at the competitive landscape all the time. I’m never arrogant enough to think we don’t have to stay focused to stay ahead. There are companies like Medtronic and even Abbott Diabetes Care, which brought forth a good product in CGM. It needed refurbishment, but any time Abbott makes the decision to go back into the CGM space, I consider them a very worthy competitor, from a technology standpoint.
It is very hard to compete with companies of that size. You need to have about $300 million of investment capital to get to a commercial product — that’s just what a Class III medical device, in this regulatory environment, is going to require. Many of these smaller, privately funded companies may or may not have innovative technology. We have a technical team that scours the Earth, looking for disruptive technology, but we haven’t found anything yet that is more accurate than what we are doing.
We stay very vigilant, to make sure there isn’t something bubbling up in some research lab at an academic center somewhere. If something came about, though, we would be aggressive in trying to acquire that technology.
MDO: Earlier in the conversation, you mentioned a lesson you learned — namely, be wary of straying too far outside your area of expertise. Are there any other best practices or advice that you would share with other medical device makers, particularly those at small or emerging companies?
Gregg: Yes, and that is if you truly believe in what you’re doing, stick to it.
There were times in Dexcom’s history that were extremely challenging. Back in late 2008, the markets had turned very bad. We were struggling as a company. We had introduced the Seven Plus CGM system, but we were running out of money. A large shareholder had to dump 3 million shares of our stock in a few days. As a result the share price plummeted to $1.39. Luckily, I had the wherewithal to step in and buy a significant number of shares in the open market to indicate my confidence in the company. Today, we trade at $50.00 a share, but there were very lean times.
However, we never lost our mission view. I never told anyone how dire our situation was — only a handful of people, including the board, knew. But we rallied. I said, “I’ll figure out a way to get us money.” That pit bull mentality is critical. Once you latch on to something, don’t let go. You’ll find a way. It may be tough, but you’ll find a way.
As I mentioned, I talk to a lot of small, privately funded CGM companies that are working hard, and I encourage them to continue to try. They need to take it to another level, but some of these technologies are interesting. I don’t know that they are definitely better than what is out there, but I also don’t know that they aren’t. It’s too early.
My recommendation to these folks is take it to the next level. Find the funding that you need. If you get to a certain point, I will fund you, and so will others. But the fact is you have to stay the course.
You are in a tough world. Although we talked about how progressive this FDA has been, it is still their responsibility to make sure that products are safe and effective. You have to demonstrate that sufficiently to warrant their approval, which takes time and that takes money.
Again, that is why we are so proactive with the agency, because it reduces time and expense. We generally don’t have to do things twice, and that’s another lesson to be learned. Take the time to get the advice that you need, so that you don’t spend money unnecessarily.
MDO: You mentioned that Dexcom would be willing to fund CGM startups once they “get to a certain point.” What is that point, exactly? What would you need to see from them?
Gregg: Human data in a diabetes patient population. I see research published about curing diabetes in mice, and I have to laugh. There are a hundred strains of mice and some rats in which we have cured diabetes, but not one single human trial has worked.
I am associated with the Diabetes Research Institute at the University of Miami, the foremost research group in terms of islet cell transplants and other activities, and they have cured diabetes in mice. But their chief, Dr. Jay Skyler, a world-renowned endocrinologist, told me, “You know, we have 12 studies we have run lately on diabetes treatment modalities in humans. Ten of them have failed. The only reason probably the last two haven’t failed is they haven’t concluded yet. But each time we fail, we learn something.”
We saw C8 Medisensors go out of business. We saw Glumetrics go out of business. Look at Echo Therapeutics, which is on the ropes. Becton Dickinson had a fluorescence CGM in development, but they abandoned that earlier this year, because they couldn’t overcome biofouling. These companies found an answer, but they didn’t find the answer.
I tell our people that we have the answer. Now, we need to make it better, more cost effective, and easier to use.
MDO: What challenges lie ahead for Dexcom, and what are you doing to address them?
Gregg: The biggest challenge I see in front of us is a very rapidly changing healthcare landscape. Under the Accountable Care Act and other activities, you can see the writing on the wall for physicians — they are going to get paid for performance. I think the medical device industry is slightly behind that revolution in the way healthcare is delivered.
We are running a clinical trial right now that has an economic component to it. This is the first time that we have done this — all of our clinical trials heretofore have been about product approval. Now, we are running a very large trial in which we have to demonstrate that Dexcom CGM is cost effective, that we are taking more money out of the healthcare system than we are costing it.
I think that will be very important for the future, which is why I tell our folks every day, “We have to make this product less costly for a whole host of reasons.” One of those reasons is accessibility. The more we can reduce the cost, the greater number of patients has access to it. I think that we are going to see some pricing pressure that we don’t see today.
We need to be part of the solution. I don’t know if we can build them cost effectively or not, but you will certainly see us move back into the hospital arena with a subcutaneous sensor. That is another platform for Dexcom in the future, but it has to be done in a cost-effective way. That, for us, is the greatest challenge, but we have systems in place to address it.
MDO: The study you mentioned that has an economic component, did you work with CMS or other payers on your trial design?
Gregg: We are not working directly with CMS, but we certainly had counsel from the private payer system as to what they are looking for.
The CMS situation is interesting. There is a bill moving through the Senate for CMS coverage of CGM. We are lobbying for, along with Medtronic. CMS has said that it will cover CGM when we can obtain dosing claims for CGM, which gets back to why the artificial pancreas algorithm is so important. We know that 9 percent MARD is dosing-level accuracy, and we will be able to satisfy the CMS criteria that they have put forth to achieve coverage.
The sad thing right now is that there are 64-year-olds with private insurance who have had great success managing their diabetes with CGM, but they are facing the loss of reimbursement coverage as they migrate to Medicare. Dexcom is working on some initiatives to create opportunities for those truly needy patients. If they don’t have coverage, we will try to figure out something that can help them financially to acquire the product. For example, we have created a foundation that employees — including Dexcom executives — will spend their money to populate. It will be launched in the first part of 2015.