After 20 Years of Watching Diabetes Tech, Kliff Eyes Smart Insulin Pens, CGM for Patients With Type 2 Diabetes

Although current medications can maintain healthy blood sugar levels in most patients who have type 1 (T1D) or type 2 diabetes (T2D), most patients don’t use them correctly and therefore suffer the expensive and unpleasant complications of both hyper- and hypoglycemia. Less than half of all patients with T2D achieve glycemic goals advocated by the American Diabetes Association, and about two-thirds die prematurely of heart disease.¹

Study results indicate that educational interventions can boost treatment adherence, at least among some patients,² but many experts believe the only hope for widespread improvement among real-world patients lies in new medications and novel technologies—products that dramatically reduce the pain and complexity of proper self-care.

David Kliff, who has run Diabetes Investor since he was diagnosed with the condition more than 20 years ago, is a leading spokesman for this view. His livelihood depends upon being able to predict how patients will behave (and, therefore, what they will buy), and he scoffs at the notion that any affordable intervention produces significantly better outcomes with traditional treatment tools. He believes that better tools can produce better outcomes, that several significantly better tools have recently hit the market, and that more are coming soon.

Kliff is not alone in his general optimism, but his specific predications differ from those of many others who think technology is about to revolutionize diabetes care. For example, Kliff said he doubts that any of the “artificial pancreas” devices currently under development will have any major impact on patients with T1D in the next decade. He thinks the real opportunity for improved outcomes lies in better glucose monitors and smart insulin pens. What’s more, unlike many, he thinks that better technology (rather than just better medication) will greatly improve outcomes for patients with T2D.

There is an underlying theme to nearly all of Kliff’s thinking about what will and won’t succeed in attracting customers, improving outcomes, and reducing costs for patients with diabetes: ease of use. The majority of people simply cannot or will not make proper use of any product that is physically or intellectually demanding. So, the more a system requires of patients, the less they will use it and the more mistakes they will make when they do use it. Any tool that makes life simpler for patients (at an affordable price) has a good chance of winning widespread adoption and improving outcomes.

Evidence-Based Diabetes Management™ (EBDM™) interviewed Kliff about what’s coming in diabetes technology and which innovations he believes have the greatest chance of success.

EBDM™: What new and pending technologies most excite you?

KLIFF: The next generation of continuous glucose monitors (CGMs). The current CGM from Dexcom is already a game changer. Although it must still be calibrated with a finger stick, the sensor was approved to allow patients to make insulin dosing calculations without verifying CGM readings with a finger stick.³ Annual worldwide sales of finger sticks have already fallen from $6 billion to $4 billion, and I think that the ongoing flurry in CGM innovation will bring that number down to $1 billion in the next 3 years. This is a major shift. CGMs will not only be faster and more comfortable than finger sticks, [they] will be cheaper and more accurate as well, and that will end finger sticks.

The next generation sensor from Dexcom, which seems to be the innovation leader in this market, is slated to be about the size of a dime and capable of self-insertion and self-calibration. All patients need to do is put it on a new piece of skin every 3 days, touch a button, and the device does the rest. That would be a game changer.

EBDM™: What does it do for patients, other than saving them finger sticks?

KLIFF: Saving finger sticks is a big deal on its own, as anyone who has ever done multiple finger sticks a day will tell you, but the real advantage is the extra information. The switch from finger sticks to [using a] CGM gives patients more information. More importantly, it also provides more information to their doctors, who can use it in real time, either to tweak a patient’s treatment protocol or to alert them that they’re failing to control their blood sugar. Perhaps most importantly, [continuous glucose monitoring] gives data to smart software that can then calculate exactly how much medication patients need and exactly when they need it.

The only FDA-approved software that currently uses CGM data to automatically calculate insulin doses is the software that operates Medtronic’s 670G system,⁴ but other companies are working on the same sort of intelligent software. Dexcom has partnered with Verily, which doesn’t sound like a huge alliance unless you know that Verily is a part of Google, which obviously has enormous experience with both: the necessary analytics and making user-friendly software.

EBDM™: So, the CGM will send a constant stream of data to some program, located either in a phone or a standalone device, and the software will tell the patient how much insulin to take and when to take it?

KLIFF: Yes. We’re not there yet, but that’s almost certainly where it’s going.

EBDM™: Wouldn’t it be easier for patients if the software sent the info to a pump, which delivered the insulin automatically?

KLIFF: In theory, yes. In practice, not for a long time. Pumps just aren’t the right solution for most people with type 1 diabetes. Pumps require more work for the patient initially, and these devices can, and do, malfunction, which can be dangerous, as they deliver a lethal drug. And that’s why only a tiny fraction of the people who could use a pump actually do use a pump. Significantly more than 90% of insulin users deliver their medication with insulin pens, which haven’t improved nearly as much as glucose monitors in the past decade, but they’re on the cusp of a big leap forward with the first generation of smart pens starting to hit the market.

EBDM™: What do smart pens do, and how do you expect them to improve outcomes?

KLIFF: Functionality varies from maker to maker. Some of them are just cases or caps for standard insulin pens. Those are able to keep track of when patients use insulin and tally their total usage. Patients who use traditional pens are supposed to keep detailed manual records of all that stuff, which most of them don’t, so it’s pretty much impossible for their doctors to evaluate their treatment adherence. Automating all that record keeping will give doctors the first accurate record they’ve ever had concerning patient behavior, and that’s certainly a first step toward improvement. If patients keep failing to take proper doses at the proper times, the pens can alert doctors, who can nip problems in the bud rather than waiting 3 months for the patient’s next appointment.

A few of the smart pens go further, though. They’re fully engineered devices that, once doctors have programmed in a treatment protocol, will automatically size doses and notify patients when it’s time to use them. The only time patients need to do anything more than insert new needles and insulin is at meal times, when patients provide information about their food and the pen software calculates how much extra insulin they need.

There are a bunch of companies working in this space. Companion Medical has a complete pen that is already FDA-approved.⁵ There’s a company called Common Sensing in Cambridge, Massachusetts, that is doing clinical trials at Joslin.⁶ Bigfoot (Biomedical) just acquired a company called Timesulin that’s working on another.⁷

EBDM™: Do these smart pens use the information from CGMs to calculate how much insulin you use?

KLIFF: Not yet, but once again, that is almost certainly where this will end up. Dexcom has said that it has reached out to the companies that are making these pens about integrating the data from both devices. Presumably, all the other CGM makers have done likewise, and presumably, some company will make the software that combines the basic protocol that each doctor prescribes with the data from both pen and monitor to make the pens dial up the proper doses for each patient. All patients will need to do—other than adding new needles and insulin cartridges—is hold the pens in the right places when their smartphones tell them to do so. The whole thing will be a cheaper and simpler version of the artificial pancreas idea that most people are talking about, like an artificial pancreas light.

The interesting question is, who will write the software that controls everything? It may be a company that makes some actual piece of equipment, or it may be a big technology company that wants to get into the healthcare market. Apple, Google, Amazon, and Samsung have all expressed interest in diabetes, and I think they will probably end up making the big competing platforms that individual devices work with and support.

EBDM™: How much of this must come together before the new technology begins improving outcomes?

KLIFF: None. The new toys that are already on the market do more than enough to change outcomes—if they’re used properly. Existing smart pens and CGMs connect to the internet, so they can already tell doctors exactly which patients do and don’t comply, more or less, with prescribed treatments. This alone can inform doctors when poor results require new protocols and when they require better patient compliance. That’s a leap forward. Up until now, it has been very hard to tell whether poor results stem from the shortcomings of the protocol or the patient.

The real question is how aggressive we, as a society, will be about using the data to motivate patients to change their behavior. We already have auto insurers that use technology to monitor how their policy holders drive and then offer very cheap rates to people who drive safely and very high rates to those who drive aggressively. Would we be willing to use CGMs and smart pens to monitor diabetic patients and offer discounted premiums to those that complied religiously with treatment protocols?

EBDM™: Do you think that would be a good idea?

KLIFF: I think it would be far more effective in changing behavior than any intervention that’s been studied to date or any app that’s designed to motivate people to change their behavior. There are at least companies that are working on changing behavior, and I don’t think any of them will have any real success. People won’t change their behavior unless there is a real reason to do it, unless you make it significantly easier to do the right thing or significantly harder to do the wrong thing. Charging higher premiums to nonadherent patients and lower premiums to adherent patients would improve outcomes for a huge number of people, and it’s certainly something we could do now, if we decided the benefits justified the invasion of privacy.

EBDM™: How else can the data from CGMs improve outcomes for people with T1D?

KLIFF: This isn’t just for people with T1D. Monitors have already gotten to the point that they probably make sense for people with T2D, and if they’re not there yet, then they certainly will be by the next generation, which, in addition to its other advantages, should be cheaper than the current generation.

The same data that can tell doctors whether type 1 patients do or don’t comply with their treatment protocols can tell doctors whether type 2 patients are complying with their treatment protocols. It doesn’t matter whether the medication is insulin or something else. The data will show an experienced eye when patients take medicine. It will also give doctors more data than they’ve ever had about how narrow a range of blood sugar each individual patient can maintain with any protocol, and that will give doctors more feedback and more ability to fine tune than they have ever had before.

All CGM makers want to expand to the type 2 market because it is so much bigger. Really, all companies that can conceivably make their products helpful for type 2 patients will make such patients a priority. It’s simple math.

References

1. Garcia-Perez LE, Alvarez M, Dilla T, Gil-Guillen V, Orozco-Beltran D. Adherence to therapies with type 2 diabetes. Diabetes Ther. 2013;4(2):175-194. doi: 10.1007/s13300-013-0034-y.

2. Zullig LL, Gellad WF, Moaddeb J, et al. Improving diabetes medication adherence: successful, scalable interventions. Patient Prefer Adherence. 2015;9:139-149. doi: 10.2147/PPA.S69651.

3. FDA expands indication for continuous glucose monitoring system, first to replace fingerstick testing for diabetes treatment decisions [press release]. Silver Spring, MD: FDA; December 20, 2016. fda.gov/newsevents/newsroom/pressannouncements/ucm534056.htm. Accessed July 23, 2017.

4. Tilleskjor S. Breaking news: FDA approves the MiniMed 670G system, world’s first hybrid closed-loop system. Medtronic website. medtronicdiabetes.com/blog/introducing-the-minimed-670g-system/.Published September 29, 2016. Accessed July 23, 2017.

5. Companion Medical receives FDA clearance for smart insulin delivery system [press release]. San Diego, CA: PR Newswire; July 27, 2016. prnewswire.com/news-releases/companion-medical-receives-fdaclearance-for-smart-insulin-delivery-system-300304933.html. Accessed July 23, 2017.

6. Common Sensing begins study with cloud-based insulin-monitoring solution to address $5B treatment management challenge [press release]. Cambridge, MA: PR Newswire; October 18, 2016.

prnewswire.com/newsreleases/common-sensing-begins-study-with-cloud-based-insulinmonitoring-solution-to-address-5b-treatment-managementchallenge-300344709.html. Accessed July 31, 2016.

7. Bigfoot Biomedical acquires Timesulin [press release]. Milpitas, CA: Bigfoot Biomedical; June 8, 2017. bigfootbiomedical.com/bigfoot-acquirestimesulin/. Accessed June 23, 2017.

Download PDF: After 20 Years of Watching Diabetes Tech, Kliff Eyes Smart Insulin Pens, CGM for Patients With Type 2 Diabetes