Devices & Diagnostics

A Q&A with Verily’s CTO about the Galvani joint venture with GSK

Brian Otis, CTO of Google’s life sciences business, Verily, describes the technology at the heart of the GSK joint venture that seeks to understand and modify the language of the nervous system in patients with different diseases.

question and answer





Going “beyond the pill” has been an ongoing effort of Big Pharma, and an announcement Monday from the life sciences business of Alphabet (formerly Google) and GSK (GlaxoSmithKline) underscores that trend.

The U.K. pharma giant revealed that it is teaming up with Verily to create Galvani Bioelectronics, a joint venture to develop a miniature implant able to modify nerves signals as well as collect information about the body’s neural networks that will communicate with a hand-held receiver/programmer and leverage analytics and software to treat patients more precisely.

The technology is being billed as bioelectronic medicine that is more targeted than a drug that affects the entire body.

But if the technology sounds a little like neuromodulation —  other news headlines focused on the term “zapping nerves” an oft-used description of neuromodulation in the medical device world — it’s partly that. But my main takeaway after a conversation with Brian Otis, Verily’s chief technology officer, was that neuromodulation and Galvani’s technology are similar in the way a gorilla and humans are. That is  assuming, of course, that the technology works and finds its way to the market.

More on the phone conversation in a bit, but here are some details about Galvani Bioelectronics from the GSK press release:

  • The company will be based in Stevenage, U.K., where GSK’s global R&D center is based with a second research facility in Verily’s South San Francisco location.
  • GSK has 55% interest in the joint venture and Verily 45%.
  • Galvani will be led by Kris Famm, who was GSK’s vice president of Bioelectronics R&D.
  • The company will initially hire 30 scientists, engineers and clinicians, and will fund and integrate a range of collaborations with both parent companies, academia and other R&D companies.
  • The deal is expected to close by the end of the year.
  • The diseases that are being targeted are inflammatory, metabolic and endocrine disorders, including type 2 diabetes.

And now here’s a slightly edited version of the the Q&A with Verily’s Otis who is leading the charge from the Bay Area company to make bioelectronic therapies available.

Brian_Otis_VerilyMedCity: Is this neuromodulation on steroids or is it something else?

Otis: What we are trying to do is to investigate a different type of therapy in contrast to what people think about when they hear neuromodulation.

What we are trying to do is to get close to the body’s organs that could be causing or contributing to disease, understand how the signals traverse through the nerves to get to those organs, listen to those signals, listen to the body and provide a stimulation in a very precise targeted way close to the organs.

And that’s really the difference — it has to be a process of stimulating precisely near the organs that need it and making these devices extremely small to make the surgical procedures as non-invasive as possible.

The goal is to create a better map of the neural circuits of the body and this is something that GSK is bringing to the collaboration. We need to understand what this map looks like. What it looks like for someone who is healthy, what it looks like for someone who is diseased.

There are two different things that need to come out of this investigation. One is from an anatomical perspective, where on the body should you put this device to perform the recording and stimulation? Secondly, what are these signals, what do they look like, what signal processing do we need to do to extract meaning from those signals?

On the other end of the spectrum, and this is what Verily is bringing to the table, we need to actually need to make this device small. It has to have biocompatible precision nerve interfacing because we will be connecting to these very small nerves.

It’s the power of these integrated circuits that will allow us to do the computation necessary, the communication and the stimulation needed to do this.

At the end of the day, what we are trying to create is a closed-loop system where these devices are listening to the body, listening to the signals traversing through the nerves or doing real-time signal processing on them and for each individual patient, optimizing the parameters that we’re [sending] back to the nerve.

At the end of the day the goal is to restore healthy function to the person and whatever the organ we are treating.

MedCity: What are the limitations of traditional neuromodulation?

Otis: In traditional neuromodulation, you think of deep brain stimulation, or possibly blocking nerve signals to reduce pain.

We are not trying to simply block a signal. There are times when we may want to block a signal but what we are really trying to do is apply the correct signal to restore function, not just block a signal from traversing.

On the one hand, we have to understand the language of the nervous system. And on the other, we have to develop technology that is small enough and precise enough to allow us to use that disease biology understanding and practice in patients.

MedCity: You talked about developing a small implant. How small are we talking about?

Otis: This is the really exciting part about what we’re doing. We are starting out with a cubic centimeter — half the size of a sugar cube — and progressing down to something that could be as small as a grain of rice.

There are lots of technical challenges in doing that. One of that is power consumption. It takes power to do the recording, it takes power to do the computation, and the communication, and the stimulation. If we can reduce the power consumption, we can reduce the size of the batteries and reduce the encapsulation and make the whole system smaller.

The smaller that we make the implant — and this is important — the less invasive the surgical procedure will be and the larger the number of patients that we can help.

That implanted device is a really key part of the system.

MedCity: How do you gather the data from the implant?

Otis: A hand-held device would be able to provide updates to the implant and get data from the implant. But I want to emphasize that the implant itself will have intelligence as well. This amounts to a closed-loop system where the implant is listening to these signals, it will make decisions, it will optimize the stimulation parameters for those signals and it will stimulate.

So you will be able to communicate with the implant from outside, but it will be able to work autonomously as well inside the body. The hand-held receiver will be look at configuration data and be able to program the stimulator as needed.

We want to look at other data too. What other physiological signals are important and those depend on the disease state that we are talking about whether it be rheumatoid arthritis or metabolical disorders. Those signals will be different and we need a way to get those signals and the data from those into the optimization process as well.

MedCity: How is the data from the receiver viewable because that hand-held device will be in the patient’s home, correct? 

Otis: Patients will have a hand-held device, and how the software, the apps work will depend on the therapeutic area in question. But across the board, it goes without saying that we need the data analytics to analyze all of the data, from the actual stimulator — which will be the actual electrical signals themselves — or other types of sensors that we might want to have on the body.

So there will be that computational engine running on the implant and maybe outside the implant as well to be able to fine-tune these parameters.

Beyond that, the actual user experience and the composition of the app — those are things that Verily and Google are very good at — will depend on the therapeutic area.

MedCity: What stage of development are you in?

Otis: At this point, GSK has been working for a couple of years with a lot of outside collaborators on trying to identify the disease biology. They are quite far along on that path. On our side, for the past four-plus years at Verily and Google, we have been working with miniaturized electronics to interface with the human body.

We have been able to do things like miniaturize a radio, a small computer, a battery and antennas down to a size that we can put in a small contact lens.

We have a program with Dexcom where we are working on a small, patch-based continuous glucose monitor. So we are quite far along in being able to create and manufacture low-power, small devices that are meant to interface with the human body. We are at an extremely good point to launch this collaboration off of.

MedCity: What will your first market be in terms of treating a disease? Will it be type 2 diabetes or something else?

Fortunately, there are a lot of diseases that look promising. Rheumatoid arthritis impacts more than a million people in the U.S. — it affects their quality of life, their ability to work and even their life span.

Many of those people are not responding to current pharmacological treatments, which is extremely unfortunate. So rheumatoid arthritis is an interesting one and we are paying a lot of attention to that. To add to that, there was a paper in last month’s PNAS (Proceedings of the National Academy of Sciences) which showed the ability to stimulate the nervous system can result in an improvement in rheumatoid arthritis markers. So that is extremely encouraging.

Another one is Type 2 diabetes and the worldwide impact of the disease is astounding. We are doing a lot already in the continuous glucose monitoring on the diabetes management side. This would give us possible another really important tool to help people with Type2 diabetes.

It’s too early to talk about market size and first product. We have work to do to establish clinical proof, to finalize the hardware, validate it and understand what the surgical procedure is going to be. But the early results look good.

MedCity: Earlier on you mentioned sensors. Do you mean to say that the implantable device will work with some body-worn sensors?

Otis: It’s a possibility and completely dependent on the therapy area that we are working on. We want to make sure that this implant can truly listen to the body. That it’s not just setting a stimulation parameter and being done.

The goal is that it will be a closed-loop system. A lot of that data will come from listening to the nervous system itself, but we need to be open to the possibility of listening to other signals as well.

What Galvani is trying to do is extremely precise interventions both spatially in terms of where we are targeting the therapy and in terms of the actual data and signal.

We want to be able to speak the language of the nervous system. we need to listen, we need to do an interpretation and we need to apply the proper signals to restore proper organ function.

MedCity: What’s the regulatory pathway in the U.S.?

Otis: It’s probably not prudent to speculate too much time at this point on the regulatory pathway. But just in terms of looking at the expertise we are bringing to the table — from the GSK side, it’s extensive experience with drug development, disease biology and establishing clinical proofs of different types of therapies. On the Verily side, we have a lot of experience on the medical devices side, on the manufacturing of those devices, understanding the regulatory pathway there.

Between the two of us, we clearly have the expertise and the institutional knowledge to make that happen.

MedCity: And just to clarify that there is no drug component to this bioelectronic therapy that Galvani is developing?

That’s a really interesting point. The field of bioelectronic medicine is intended as a new form therapy and the thing that is really important to note here is that this could be a precise way of interfacing with an organ through electrical signals to avoid some of the side effects that would result from taking drugs.

But that doesn’t mean that this couldn’t be extremely complementary to different types of pharmaceutical therapies either the ones that are out there or ones that could be co-developed with bioelectrical intervention. So that opens up a completely new dimension to trying to help patients.

If you look at what GSK has been doing for the past three years — and this is why we’re excited to have them as a partner — is that they are truly trying to build new maps of these neural circuits and at a very deep level, understand disease biology and understand what these signals look like for people that are diseased and what these signals look like for people that are healthy.

On the Verily side, yes, there is an important device component and we are trying to define the frontier of medical device technology. The part that’s interesting is that the feedback results from us developing new technology, GSK having this new biological understanding, feeding back to new specifications for this technology and looking at that feedback loop to create new therapies. That’s really exciting.

MedCity: How will your product address the cost and reimbursement issue that we are currently contending with in healthcare?

So these areas that we are going after are really dictated by where we can have a huge effect in improving people’s lives. It’s early to speculate on market size at this point, but if you look at other things that are happening in the healthcare landscape  for example the move to paying for outcomes, that’s something that fits in very nicely with what we are talking about here.

The whole mission of Verily is to improve outcomes and the program with Galvani is really in line with that — trying to have more precise therapies that target exactly that diseased organ and leaves the rest of the body alone. This is precisely what we need to be doing and inventing to improve outcomes.

That’s our mission – we’re improving outcomes, we’re improving outcomes, we’re improving outcomes. That’s it. Everything else will follow that.

MedCity: Verily has faced some criticism over projects that may not be realistic, and employees leaving. Do you think that creating a joint venture with an established healthcare company will give Galvani some cover?

I have been at Google for almost five years. Before that, I was a tenured professor at the University of Washington for seven years. This is really my dream job.

I think we can have a diverse portfolio. We can have projects that are much more early stage that clearly would have a benefit. We can also have other projects that have much less execution and timeline risk and different things in between. I personally think we need to be doing both of those.

The projects that we have in Verily span a range. And of course we are not afraid to take on projects that are risky if they have a benefit to the patient at the end of the day.

We have a program with Dexcom with continuous glucose monitors and this is going to have a huge impact. This is something that is quite far along and things are looking good. Execution wise it’s a well-oiled machine. That’s just one example.

I love the portfolio approach of Verily and it’s just an amazing place to be doing this type of work.

MedCity: Is there anything else you would like to add?

Otis: In my mind, there is a kind of virtuous cycle that’s going to result from what the GSK is doing and what Verily is doing. The devices that we create will allow better understanding of the disease biology, better understanding of the disease biology will allow us to better understand what kind of devices we need, and once we get that feedback loop going, that is really what’s going to help drive this to something that’s going to help people. That’s the part that I am extremely passionate about.

Photo: Getty Images, Paul Bradbury