Monday, April 18, 2011

Medtronic Visit Update

I know I've been a total delinquent when it comes to writing about the Medtronic conference I attended (or writing about much anything else), but I'd like to share some of the things I took from the Medtronic North American Diabetes Advocacy Conference on April 1, 2011.

As I already disclosed, I travelled to Los Angeles as did a number of other diabetes advocates (that's how our host, Medtronic Diabetes, referred to us, although the Diabetes Online Community, or "D-OC" is the term many have unofficially adopted to describe ourselves collectively). Medtronic actually has their own blog post with photos from the event which can be visited HERE.

The agenda was not shared with us until the day before we travelled. I presumed it would be somewhat similar to what took place with Roche (another event I attended, see HERE for more on that event), where we introduced ourselves and what we've done in the diabetes space, then our hosts did the same, and perhaps had some conversation about issues impacting both us as patients, and Medtronic as a company operating in the space. That was mostly the case, although we as diabetes advocates did not have quite as much involvement (initially) as I might expect a medical device company would want from us. Not that it stopped anyone from chiming in; this group isn't (and wasn't) exactly shy about asking questions!

Karmel Allison from "A Sweet Life" already covered many of the items I was thinking of, so rather than simply rehash those, you may visit her excellent write-up of the conference at

Holding Onto Leadership In An Evolving Pump Market

I often write about "The Business of Diabetes" (which is collectively an industry valued at billions of dollars in sales per year on a worldwide basis). Medtronic (Minimed) Diabetes is pretty much the market share leader in the insulin pump space, as the first-to-market for commercialized insulin pumps (with about 70% share of the U.S. market), and while the company defends it's leadership position (and having longstanding contacts and processes with hundreds of insurance carriers does make the company a force to be reckoned with), there is no lack of competition eager to capture a slice of the growing insulin pump market, and some nimble startups have captured a slice of the market Medtronic has not really considered: people who hate pump tubing. (More on that in just a minute ...)

But the insulin pump market is evolving.

Notably, many consumers are moving from multiple daily injections, but the key to their making the move may not be the same as what it used to take to lead in this space, in part because those are confirmed expectations today. Specifically, pump reliability is expected, as is excellent customer service if there's a problem, and any company that wants to compete absolutely must work with insurance companies to get coverage. But today, other factors like product design and ease of use are becoming more important to newer buyers, so pump makers must evolve, too.

Growing Market for Insulin Pumps Lures New Competition, Innovation

Medtronic's insulin pump competiton includes longtime rival Roche which began as a Swiss company formerly known as Disetronic back in 1984 and entered the U.S. insulin pump market in 1991, but has since been expanded to include a startup firm planning to market a new, tube-less pump with some new features (like the ability to bolus without the remote control -- gee, has anyone ever misplaced a remote?!) known as Mendingo Solo (see for more) which Roche acquired last April ... I can only hope that Roche does NOT try to rebrand it AccuChek, since insulin pumps don't actually CHECK anything). Others include Animas/Johnson & Johnson which has considerable resources at its disposal to compete effectively; a Japanese-owned company named Nipro Diabetes Systems (which last year acquired a Florida-based company formerly known as Home Diagnostics, Inc. best known for making low-cost meters and strips sold under various pharmacy chain's brand names, see HERE for more details on that acquisition); a Korean firm named Sooil (banded as Dana in the U.S.) Diabetes Care -- perhaps best known for selling a relatively low-price traditional pump (the kind with tubing). Another firm known as Smiths Medical which previously marketed the Deltec Cozmo pump decided to quit the insulin pump business in March 2009. More recently, an innovative Massachusetts startup known as Insulet Corp. whose Omnipod is a disposable pump without tubing has arguably paved the way for others wanting to enter the market. Unlike the others, Insulet relies almost exclusively on contract-manufacturers to bring it's product to market.

These are just the firms presently operating in the space, but several more have expressed interest to join the fray, including Tandem, Cellnovo and a joint-venture to be branded as "Jewel", all aiming to become players in the insulin pump space. I was not really surprised that Medtronic was unwilling and/or unable to share anything they might have in their development pipeline; technically, the FDA prohibits the marketing of devices that aren't yet approved, although discussing the wants and needs of patients is really not marketing, but market research). Karmel Allison correctly notes that Medtronic most likely has it's work cut out for it.

As I noted in my comment on Karmel's post, back in the days when I was still wearing an insulin pump (in the mid to late 1990's, see HERE for some more background), I decided to go with a Pennsylvania-based startup known as Animas which began in 1996. My reasons were entirely rational: back then, Minimed was still selling the 508 pump, and the smallest basal rates on that model, if you can believe it, was a whopping 1 unit/hour -- enough to make me hypoglycemic with all that insulin delivered every hour. I've always had minimal basal insulin requirements, as 90% of my total daily dosage was (and still is) prandial (mealtime) insulin. If I don't eat, I have almost no need for insulin except in the hours before I wake, sometimes called the Dawn phenomenon, but technically, even that's not entirely accurate. The notion that individuals require so much basal insulin isn't technically factual, because if I don't eat any protein, my basal needs decline precipitously. Proteins are metabolized into the bloodstream much more slowly than carbohydrates, typically they take 6-8 hours to be converted to glucose that the cells in the body can use. The notion that half of one's total daily insulin dosage should be basal actaully caused my first day on an insulin pump to be filled with hypos.

I called my CDE the day after starting on an insulin pump, shared my experience with her, and she admitted that the guidelines were only recommendations but that clearly did not work for me. My basals were cut to the lowest possible level (as I already mentioned, at the time, the Minimed 508's lowest basal rate as 1 unit/hour, but the Animas pump I chose enabled basals in 1/2 unit increments, and even that proved to me more than I required). Today, pumps enable delivery in tenths or even 100ths of a unit, so clearly, much has changed over time!

For the past decade, there have been a number of startups which have emerged, and likely more will come come given the size and appeal associated with this market (for companies, that is). At least 1 company changed it's mind (Deltec Cozmo, anyone?), but other newcomers have proven to be been pretty fierce competitors with innovative design.

Startups Expand The Pump Market Overall, Not Necessarily Steal Customers From Medtronic

In my assessment, the nimble startups haven't so much stolen market share from Medtronic, but have expanded the overall market to include consumers who might not otherwise consider an insulin pump.

Most notably, I'd point to Insulet Corp. whose tube-less Omnipod system now has an estimated 10% of the global market for insulin pumps. Unlike most of it's competition, Insulet relies almost exclusively on third-parties to make most of it's product. Notably, a Singapore-based company named Flextronics International Ltd. actually manufactures the disposable Omninpod pod-pumps in accordance with their design secifications at a partially-automated manufacturing line in China, and a Wisconsin-based firm named Phillips Plastic Corp. manufactures and supplies a number of injection molded components, while an Austin, Texas company named Freescale Semiconductor, Inc. makes and supplies the application-specific integrated circuit for the OmniPods. This business model has some clear risks, but also enables Insulet to operate as kind of a virtual company.

I should also use this opportunity to add that on March 31, 2011, one of Insulet's co-founders, John L. Brooks, III, was recently named President and CEO of the well-known Joslin Diabetes Center (see HERE for more). Mr. Brooks has been Joslin's acting CEO since Joslin's prior CEO retired in October. Mr. Brooks had not been working at Insulet for several years, but is once again back in the diabetes spotlight with his new role at Joslin.

Medical device giant (which is based in Minnesota) Medtronic pretty much bought it's way into the insulin pump business by acquiring Minimed, which was founded by Alfred Mann (the same guy behind MannKind which hopes to commercialize Afrezza inhalable insulin) back around 1980 or so, and Medtronic bought Minimed in August 2001. Medtronic recognizes the challenges facing it's business, and has brought in a number of new executives including Dr. Francine Ratner Kaufman as Vice President of Global Medical Affairs in 2008.

In Dr. Kaufman's position, as I understand it, she is responsible for much of the research that the company requires to seek approval from regulatory agencies around the world -- which is no small task. In addition to her, the company has also added some expertise from outside of the medical field which I found impressive. Specifically, Lane Desborough was hired as a product strategist for Medtronic. He spoke about process control in a way much like my brother-in-law does, and Mr. Desborough has worked for a number of companies developing closed-loop systems for companies such as GE and Honeywell (unrelated to medical applications). But he has a personal connection to diabetes, with a young son trying to manage diabetes and sees the benefits of a closed-loop in this instance. Mr. Desborough's perspective brings expertise on how other, unrelated industries have responded to various regulatory and profound market changes of the sort now facing Medtronic's insulin pump business.

Dr. Kaufman Proves Non-Inferiority for Insulin Delivery Shut-Off Function; I Ask: When Will the FDA Buy Into It?

I've known about Dr. Kaufman as a well-published endocrinologist. However, like so many former ADA Chiefs, I find her "get on the diabetes control bandwagon" a bit too intense for my liking. I realize many PWD's do not adequately manage their condition, but that's as much a public policy issue as it is a medical issue, and placing most of the blame on patients for a pretty dysfunctional medical delivery system is a tad misplaced in my opinion.

Some of us in the community of people with diabetes would like to change the mindset in the medical profession to shift the focus away from ineffective and unsustainable scare-tactics which were so prominent in Dr. Kaufman's book Diabesity. For, example Riva Greenberg touched on that in an eloquent Huffington Post story (see HERE). Speaking of Riva, I really loved the quote from Diabetes Research Institute's Dr. Camillo Ricordi's quote when he told Riva Greenberg "It's easy to think diabetes is acceptable if you don't have it". OK, I've deviated off topic yet again. Anyway, back to Dr. Kaufman's presentation ...

I listened to Dr. Kaufman's presentation, and at first, I must admit that I was more disturbed by the the company's use (the term data mining readily comes to mine) of patient data for commercialization purposes (I have long been a privacy advocate, such as my fighting the NYC Department of Health and Mental Hygiene's involuntary registry of HbA1c results for patients with diabetes -- see HERE, HERE, HERE, and HERE for some background on that) -- I did not win that battle, and the guy responsible for that egregarious breach of personal privacy is now heading the U.S. Centers for Disease Control (we were on TV together, see HERE for that news clip from Canada's CBC). Ironically, the City of New York Department of Health and Mental Hygiene has yet to show any compelling link proving the benefits to that registry and invasion of patient privacy.

However, in the case of Medtronic, patients DID sign away their rights to the data (whether they realized it or not -- that's a separate conversation). However, as I listened, Dr. Kaufman presented a VERY compelling story with with all of that data. I listened closely and found her presentation on the pump suspend function when hypoglycemic to be absolutely riveting.

Dr. Kaufman really didn't need all of that data to convince me of the logic, yet what she was doing is walking us through the story they as a company are using to convince the FDA on why it makes sense and is really more of a safety feature. So far, the FDA isn't moving! The FDA's logic on this makes no sense whatsoever, at least in my mind.

Oath of Hippocrates: First, Do No Harm.

The Hippocratic Oath is first and foremost about doing no harm, yet the FatalFood and Drug Administration (FDA) which is staffed by so many doctors, seems oblivious to this. I find it troubling that regulators at the FDA could be so mirculously blinded to genuine short-term risks of intensive insulin therapy (namely, hypoglycemia), or at least they seem only too willing to overlook those risks over worry about longer-term dangers.

In my mind, having a suspend function on an insulin pump should only be considered as a safety feaure, not viewed as some kind of risk. This does not introduce a new risk, but does help eliminate an old risk of a drug category (insulin) that is responsible for sending some 56,000 people with diabetes to the Emergency Room each year, mostly due to hyPOglycemia, not hyPERglycemia (see HERE for the story I did about that referencing the Journal of the American Medical Association citation where these stats were attained). I was left wondering just how many reminders this agency needs of that?!

Strange Logic: Why the FDA needs to be Convinced that Delivering Even More Insulin to Someone who is Already Hypoglycemic is A Safety Feature

The FDA's fixation and senseless worry about anything that could remotely even slightly INCREASE glycosated hemoglobin (a.k.a. "HbA1c") seems almost like using a dirty word to this agency. The FDA sees that as artificial intelligence and has to be convinced that can be done safely.

I have already written about how the FDA rushed-to-market biosynthetic human insulin back in 1982 which has almost killed countless patients including me. Indeed, an entire organization based in the UK called the Insulin Dependent Diabetes Trust (IDDT) has made a mission of addressing a number of the facts the FDA has conveniently swept under the rug (and has worked very hard to keep them there) to enable the biotechnology industry under it's watch to flourish.

Notably, a former FDA employee named Henry I. Miller actually admitted (OK, he bragged) in his book that he single-handedly ushered in the U.S. biotech era by waiting for his boss to go on vacation, and then took Humulin's (rDNA origin) approval to his boss' boss, who then approved the drug without a thorough review of the drug's adverse impacts in clinical trials. True, a vast majority of patients do not experience these problems, but some 10% do -- isn't doing harm to a minority of patients a violation of the Hippocratic Oath? But this post isn't about Henry Miller, it's about Medtronic and what the company is trying to convince the FDA of, and what the company shared with those of us who attended their recent conference.

Having a suspend insulin delivery during function during hypoglycemia (especially nocturnal hypoglycemia, which keeps many parents of kids with diabetes up many nights, not to mention spouses and partners of adults with this disease): sounds perfectly logical -- except to the FDA decision-makers.

Based on the FDA's sordid history on that not-so-little issue, I find the agency's stance on the artificial pancreas and the suspend insulin delivery function to be nothing short of miraculous.

Yet, in the strange world the FDA operates in, the reviewers see this functionality as crossing some sort of mysterious line for artificial intelligence on drug delivery devices, and they've been unwilling to specify what criteria they will use to evaluate such functions.

As I've noted, ordinarily, I woundn't be rushing to defend the so-called artificial pancreas because I feel all of the time, money and effort spent on this initiative might have been better allocated to an area like autoimmunity research, which remains starved for research funding. I was also not crazy about a company data-mining patient's results, but after seeing the FDA's position on this, I was actually quite glad Medtronic had access so much patient data and could use it, because Dr. Kaufman and colleagues have so much, rich data to build a compelling case for the FDA on why the FDA needs to abandon it's almost exclusive use of surrogate endpoints (meaning HbA1c) to make what is such a common-sense decision: to suspend insulin delivery when a patient is hypoglycemic.

This should NOT be viewed by the FDA as being some sort of ethical dilemma, yet the agency is depicting it that way. Today, the technology already exists to combine a continuous glucose monitoring system (CGMS) and an insulin pump to to this, yet the FDA remains fixated on anything that might contribute to an HbA1c increase. Nevermind how many more lives could be saved during the middle of the night while the remote risk of complications which may (or may not be; there's already some data suggesting that glycemic variability, as well as inflammation associated with autoimmunity in addition to hyperglycemia may be behind these complications that are basically presumed to be due to sustained hyperglycemia) due to hyperglycemia decades from now. So Medtronic first built a case that the suspend function is not inferior (a noninferiority study) to user-controlled functionality by itself.

Dr. Kaufman's presentation eloquantly validated what is already approved by the European Medicines Agency (EMEA) and is already permitted for patients in Europe: for an insulin pump's delivery of insulin to be suspended automatically (without patient intervention) during hypoglycemia.

If a continuous glucose monitor detects a hypo, the device should instruct the pump to stop delivering insulin, which would make a person's blood glucose level drop even further. It all makes perfect sense, unless you are the FDA. The FDA is charged with keeping patients safe, or at least that's what the agency claims. So one should be asking what logic they're using when they argue that suspending insulin delivery during a hypo would adversely impact a patient's HbA1c, and that it crosses a line on artificial intelligence for drug delivery?

Note that the FDA has pretty much been on autopilot when it comes to ALL diabetes drugs that should reduce HbA1c and that's all that has ever needed to be proven. I would also remind my readers that the FDA approved, then changed it's mind, on type 2 drugs like Avandia which indeed reduced HbA1c (never mind that it increased cardiovascular events, that wasn't the "surrogate" endpoint they needed to look at to evaluate diabetes drugs of any type). But the FDA claims that only human beings should be making decisions on insulin delivery, and that machines are not ready to do so and should therefore NOT be permitted to do so, all in the half-baked name of patient safety.

Under what criteria is it EVER safe to permit a device to continue deliver a drug (insulin) that the agency knows will continue reducing a patient's blood sugar when it's already too low, especially when the agency has data to show it should temporarily stop doing that?

While awake, for example, a patient CAN override such a recommendation if they want it to continue delivering insulin. What device makers and the JDRF want is to enable the device to stop delivering insulin when they have data to show that it would be harmful, such as when the patient is sleeping.

Using data from a boatload of real patients, Medtronic actually has data to prove noninferiority for the suspend function during hypos.

Now, we just need to convince the FatalFood and Drug Administration they need to go along with it. Medtronic is quite confident they can convince the FDA to go along, and they have a LOT of data to prove it. Now they just need the FDA, which is a bit gun-shy thanks to scrutiny for approving dangerous drugs under the former leadership of Andrew von Eschenbach, now wants to prove it's looking out for the best interests of patients. But don't do that at the cost of making WISE decisions, Ms. Hamburg (who is the present Chief of the FDA) -- please!

The JDRF's Government Advocacy Website actually had a page to address Artificial Pancreas Guidance that's still worth visiting.

I've always been only luke-warm to the idea of an artificial pancreas because the JDRF has repeated failed to disclose how much of fundraisers' money they've spent on the project which will ultimately benefit for-profit companies as much (if not more) than it will patients who still struggle to get insurance coverage for fancy new devices. The FDA has yet to say what it will be looking for when they are submitted for approval. The JDRF's new CEO Jeff Brewer has told the press he donated $1 million of his own money for the project, but why the organization cannot disclose this remains a mystery to me, unless they've gone way over-budget on this project and are now embarrassed to disclose that information.

BUT ... I do think that the FDA owes it all Americans for clear guidance on exactly what criteria they'll be using to evaluate such a device so that device makers can actually meet that criteria. When the criteria is nebulous, hazy and vague, it's almost impossible for any company to adequately address that. It's a bit like trying to study for a test when you don't know what the subject matter will be!

The JDRF's advocacy recently asked lawmakers to write to FDA commissioner Margaret Hamburg to provide clear guidance on what things they'll need device-makers to provide to agency with in order to get an approval. Visit where there's more info. The great news is that thanks to JDRF Advocacy efforts, on Wednesday April 13, 2011, the House letter closed, but 240 House Representatives had signed on (see HERE to see which members signed)! That number is well over a majority for that chamber. The Senate letter to the FDA remains open, so JDRF Advocacy is asking everyone to please call their Senators today and urge them to sign on.

It's not asking for the world here, only for some clarity on the criteria the FDA will be using to make it's decision on these devices so applicants may address these things. Is that really asking for too much?

My voice alone isn't enough to move the FDA, but collectively, we CAN convince them! We have influenced this organization in the past, such as on the guidelines for diabetes drugs that came out a few years ago, and I am certain we can do so again, but we need to act collectively!

I hope to address some of the other highlights from that conference in another post (such as the tour of their sensor manufacturing facility), but I hope that my readers -- whether they support the artificial pancreas project or not -- will still see the logic in advocacy to help push the FDA to outline what they will be reviewing on these devices and why it is important!


Kathy said...

I have mixed feeling about the artificial pancreas. I just don't completely believe that an organ can be mechanized. I would rather see research dollars channelled to a biological cure. On the other hand, the data shows that it does reduce the amount of hypoglycemic events. This is certainly helpful.

I didn't realize that the FDA holdup for the project was because of the potential rise in glucose. This does not show much empathy on their part. They don't seem to get the one step back for two steps forward concept.

Karmel said...

Thanks for the shout-out :)

And, more importantly, thanks for your rabble-rousing-- you are absolutely right that we can collectively move the FDA.