When an Eli Lilly executive pulled a can of chewing tobacco from the pocket of his blue jeans, it struck me as odd — partly because he didn’t look like the chewing tobacco type, and partly because the venue seemed… inappropriate.

This exec was Matt Clemente — Lilly’s Chief Technology Officer for the pharma giant’s Delivery, Device and Connected Solutions Division — and he’d just finished speaking to a dozen diabetes bloggers invited out to learn about the company’s vision for the future of insulin delivery. This gathering took place on May 3-4 at Lilly’s Cambridge Innovation Center, which is just over two years old and is located outside the MIT campus.

It seemed a strange time and place to take a dip of chewing tobacco… until I realized that the can Clemente held in one hand (with his smartphone in the other) actually had a few inches of clear tubing sticking out of it with a Luer lock on the end.

It wasn’t a can of chewing tobacco at all, it was Lilly’s top-secret new insulin pump, unveiled for the first time to the community! Hell, this was unveiled for the first time to practically anyone outside of the high-security product development lab.

Remember, Lilly first announced its D-tech plans in November 2017. This radical new-form insulin pump is a key component of a whole new D-tech approach the company has envisioned for the future of selling insulin, something the company calls the “Connected Diabetes Ecosystem.”

This gathering was meant provide an inside look at Lilly’s device pipeline, but naturally access and affordability were key issues on everyone’s minds too, along with questions about the “cost of innovation.”


Welcome to the Ecosystem

So starting with the technology unvieled: Just what the heck is a Connected Diabetes Ecosystem?

I don’t think the fact that the letters spell out CDE is a coincidence, because to some degree, Lilly wants to put a Certified Diabetes Educator in the pocket of every person with diabetes (PWD). Specifically, their Ecosystem, brainchild of the Innovation Center, will ultimately include either a closed-loop system using either a pump or smart pen, integrated CGM or blood glucose meters, a smart dosing/control algorithm, and a robust smartphone app to—like the One Ring in Tolkien’s Lord of the Ring—run the whole show.

OK, you say, there’s nothing particularly innovative about that. One might expect something more from a body in close orbit to MIT. 

Well, there are a couple of ways to think about this. Over the two-day Summit, the development team at the Innovation Center repeatedly emphasized that simplicity and seamless integration where their key goals. One way to look at this is to compare what Lilly is trying to do with the early home computer experience. PC systems were hobbled together by a variety of companies, and there was much finger-pointing when something didn’t work right. Apple computers, on the other hand, although more limited in their options, were controlled by a single source. The buck stopped with Apple, and damn if the things didn’t work better.

Today, most connected diabetes patients have insulin from one company, a delivery device from another, glucose monitoring systems from a third, and very possibly an independent app on top of all of that to try to track everything. The Lilly folks, who rightly say they know insulin better than anyone, think they can do better. They envision a fully integrated solution that’s simple for patients and highly customizable and scalable so it will work for nearly everyone — all with the convenience of one-stop shopping for ordering, delivery, customer service, and perhaps even featuring a single combined “system copay.”

Convenient, true, but at the same time it’s a potentially frightening example of vertical integration, where Lilly will own everything from the molecule to the patient data, and everything in between. If it works out, it will certainly be a win for Lilly and could greatly expand the use of their insulin, which under normal economic theory would be a win for PWDs too, as the economies of scale usually reduce cost. But that said, this isn’t what we’ve seen to date. Instead, with the increasing use of insulin over the last few decades we’ve seen nothing but upward spiraling costs. I’ll talk more about what the Lilly folks had to say about the cost issue in a moment, but first a bit on the group behind the Lilly Ecosystem. 


Inside Lilly’s Cambridge Innovation Center 

Big companies, in general, aren’t known for fast product development. And Lilly is a big company with 40,000 employees worldwide. Several years ago the top brass at Lilly recognized that startups are where innovation is at, and rather than continually buying promising startups, they decided to create an in-house startup of their own. 

Lilly chose to locate their new pet project in Cambridge Massachusetts’ Kendall Square, largely, they admit, to make it easier to recruit top talent. And they are not alone in setting up shop in the most innovative square mile on earth; literally just across the street from Lilly is arch rival Sanofi.

Lilly’s Innovation operation is located on the third and fourth floors of a modern but otherwise nondescript office building at 450 Kendall Street, on the cusp of the MIT campus. Inside, the Innovation Center feels much like a Silicon Valley startup, with exposed pipes in the ceiling, and an open floor plan with no offices (even for the on-site execs) to facilitate communication and openness. About 40 people—ranging from electrical engineers, to endos, to patent attorneys—work at the Center, which is designed and equipped to be able to quickly develop, prototype, and test ideas for the pharma giant’s pipelines in a nimble and fast-paced environment.

A large part of the center’s work centers on devices in the broadest possible sense, ranging from simple pills packs to high-tech auto injectors for the full range of Lilly pharmaceutical products. To accomplish this, the third floor of center has its own machine shop with every tool imaginable from screwdrivers to laser cutters, a hydraulic analysis department, a complete electrical fabrication department that can make its own circuit board in-house in a day, 3D printers, and even a sewing machine. And, of course, there are dozens of computers, a roomful. The space, while open-feeling, is quite small. It’s easy to see how ideas can be easily and quickly shared. 

Above, on the fourth floor, Lilly has installed a complete drug formulation lab to help study how various containers affect a range of pharmaceutical products. Does a given plastic make a medication last longer? Or shorten its life? This lab is also the jumping-off point for studies on how medications interact with the human body.

Security is high. Doors and elevators require special badges for access and many visiting Lilly staff, both execs and PR folks, needed escorts to even get the elevators to work. Needless to say, we weren’t allowed to take photos in most of the Center, or not in the interesting parts anyway.

In addition to the Lilly team, there’s space at the center for employees of Lilly partners, called External Collaborators by the company, to temporarily set up shop.

The Center is run by T1D Exchange co-founder Marie Schiller, who now holds the title of VP of Connected Care and Site Head for the Cambridge Innovation Center, herself a 35-year veteran of type 1 diabetes.


Collaborating with Collaborators

Despite all of this home-owned and operated innovation infrastructure, when it comes to the Connected Diabetes Ecosystem, the Cambridge Center is functioning a bit more like an orchestra conductor than a composer. Rather than re-invent the wheel, Lilly is instead limiting their role to ring master and molecule master, partnering with others to bring their Ecosystem to life, chief among them Dexcom and Deka.

Never heard of Deka? Don’t feel bad, most people haven’t, but many have heard of their founder: Dean Kamen of Segway fame. What many people don’t realize is that he’s also the inventor of the 1980s AutoSyringe AS6C, the world’s first practical insulin pump, the invention of which resulted in Kamen’s induction into the National Inventors Hall of Fame three decades later.

Also on the collaborator roster (clearly the Lilly team isn’t in tune with the negative connotation of the word) are Rimidi, who are creating software to allow the Ecosystem to seamlessly export patient-gathered data into their doctor’s electronic medical record systems; and a team headed up by Dr. Ahmad Haidar of McGill University, who created the algorithm that will drive the closed-loop portions of the Ecosystem. Haidar retains his academic position, and his independent research, but 20% of his time is now tasked to the Lilly project.

Not surprisingly, missing from the collaboratorship is anyone from the payor side, but Lilly’s plan to tackle access is data-driven. In their clinical trails of the Ecosystem, the first of which just wrapped up, they say they’re going above and beyond what is required for simple FDA approval. Clearly the end game is to “prove” that the Ecosystem improves outcomes for patients, and therefore saves money for health insurance companies. 

Speaking of money and access….


Designing for Access

Many people in our community wonder how much of the money they shell out for insulin that they can barely afford today goes into innovation that they will never be able to afford tomorrow. While it wasn’t put that bluntly to the Lilly team by any of us, we did press them about the high cost of insulin and they responded by pretty much acknowledging the issue, but also not taking ownership of it. What they were willing to take ownership of is the recognition that they need to do all they can on their end to keep the tech cheap to build, so that there isn’t a huge investment that has to be paid off when the Ecosystem they’re designing is ready for market.

So while I can’t say what the end game from the Lilly boardroom is, the folks on the ground in Cambridge are sincere and concerned about access issues, and many of them are personally connected to our community by either having diabetes themselves or having loved ones with diabetes. They say cost and access are part of “virtually every discussion” on the ground, and part of every decision they make. Cost, we were told, is always part of planning, and also part of a technical culture of “anti-complexity” that aims to create products that are easy to build, with the goal of the lowest possible cost for the maximum possible patient access.

That’s a theme I heard a lot over the two days I was on-site. The Lilly team wants to reach the widest possible audience for the new Ecosystem. Rhett Alden, the Chief Digital Architect for Delivery and Device Connected Care (sorry, none of these Lilly people have short titles) said he doesn’t want to build something exclusively for the “high-skill, high-will, high-wealth” patient crowd. Rather, he wants to build something for everyone.

And even at this point, it is more than just lip service. Lilly is already testing their Ecosystem among a wide range of patient populations, including, we were told, among migrant workers and the elderly. It’s clear that no matter what’s happening in the rest of Lilly, social determinants of health are in high awareness at Cambridge.


New “Hybrid” Pump

We got a sneak peak at Lilly’s new pump, which is fully controlled by a smartphone so the PWD wearing the pump doesn’t need access to the device to it to use it — allowing for the unusual form factor and small size. It’s a so-called hybrid pump that looks like a patch pump, but actually has short tubing that uses traditional infusion sets  although the Innovation Center team concedes that the infusion set is the “Achilles Heal” of pumping, and they are doing research on ways to improve this weak link).

Infusion sites for this new pump can be essentially anywhere on the body, and you can either opt to shove the little pump body in your pocket, tuck it in your bra, or simply stick it to your skin, using some sort of sticky pads that Lilly intends to develop for this purpose.

Like the Skoal can it resembles, with a tin and a lid, the new pump has two parts: the disposable bottom part holds a bladder-like insulin reservoir, and the durable top part holds all the electronics and a rechargeable battery. With the electronics and the algorithm living in three separate mini computers in the durable part of the pump, engineers say that a hybrid closed loop version of the pump will function just fine if you leave your smartphone behind or drop your phone and break it. The pump also features a single button, intended to be a backup bolus button that users could use to take extra insulin in the absence of their phones. 

In its first generation, the insulin reservoir must be filled by the user—at any amount between 1 and 3 ml, or 100 units to 300 units of today’s standard of U-100 insulin—and we were told that the Cambridge team spent a lot of time working on a user interface that makes it fast and easy to fill. That said, pre-filled insulin “cartridges” are the real goal for Lilly, and they hope to have them ready for the second generation of the pump. While they presumably want those cartridges filled with their product, we were told the team is working with insulins from other manufactures as well. 

It’s way too early in the development process to know how this will all play out, but there were hints that ultimately, the supplies could end up being covered as a pharmacy benefit, rather than DME. This is ostensibly all part of the goal for a simpler-for-the-patient system that also happens to be very, very good for Lilly. 

The pump is a single hormone design, although it wouldn’t take much imagination to add a second bladder. Hints were dropped that “future insulins” might make the need for a dual hormone system unnecessary.

The pump prototypes are being run on “de-featured” phones, but ultimately should be available for either IOS or Android, although it is impossible to say what the configuration at launch will look like. That will likely be dictated by the regulatory environment, rather than what’s technically possible. 

Of course, I wasn’t allowed to photograph the prototype pump, but I was able to hold it in my hands. In fact, I was the first of our group to touch it, largely because my hearing is getting worse with age and the abuse of airplane engines in my youth, so I was sitting in the front row and I can move pretty fast for an old guy when motivated to do so. My main impression was one of light weight. And while it looked like it should be able to reel in extra tubing like a tape measure, it does not. So even with short tubing, what to do with the excess will be an issue for those who choose to wear the pump on their bodies.

When will this new pump see the light of a prescription pad? Two to three years is the best guess at this point. The Lilly team is juggling a lot of balls right now with the system, and they are all up in the air.

There were few details provided on the smart insulin pen Lilly is developing, which supposedly will capture dosing data from existing disposable pens via Bluetooth or something. In fact, part of this meeting was a brainstorming session in which Lilly divided us advocates up in groups and picked our brains about pen features we most prefer.

Meanwhile, the on-site artist making a doodle poster of the meeting activity was a nice touch.



I left the Summit with mixed emotions. As always when visiting the Evil Empire, I found the individual Storm Troopers to be wonderful folks. It’s easy to hate Big Pharma on the whole, but it’s really hard not to love the individuals who make up these organizations.

As to the pump, well, Lord knows we need more pump choices, given the recent shrinking of the market with Animas’ withdrawal from the space. But having too many of my eggs in one basket makes me nervous. So on one hand, a vision of a pump that uses pre-filled insulin cartridges and everything conveniently run and serviced by a single vendor may well simplify life for PWDs (and hopefully result in fewer copays).

But it also seems that this is a highly effective way for Lilly to dominate the market, and monopolies have a way of creating negative outcomes. While I don’t doubt for a second the sincerity of the Wizards of Kendall Square, I can’t say the same for the pharma giant as a whole. We can only hope that their innovations bring real (also financial) benefits to patients in the trenches.


Note: DiabetesMine was invited to attend the Summit, and Lilly covered all travel costs including airfare, hotel, and food. Our reporting was not in any influenced by the host organization.