As a kid, Mary Lou Jepsen’s intrigue for science was ignited by a specific scene in Star Wars where R2D2 projects a hologram of Princess Leia saying, “Obi-wan Kenobi, you’re my only hope.” Over the next few decades, Jepsen would study studio art and electrical engineering, holography, and optical sciences (all while fighting off a brain tumor in her twenties) and then go on a tear, co-founding One Laptop per Child (OLPC), founding Pixel Qi to make low-power LCDs, founding and leading two moonshots at Google X, and advancing virtual reality as an executive at Facebook/Oculus VR. But in 2015, she decided to leave her “cushy job” with Mark Zuckerberg to start an entirely new venture in a completely new industry—healthcare.
“I thought that with the chips and technology we’re developing, I knew that there was a strong chance we could make something that could potentially leapfrog pharmaceuticals or at least work in tandem with them because drugs aren’t the only things that penetrate our body—infrared light does, ultrasound does, electromagnetism does,” Jepsen told Inside Precision Medicine.
In 2016, Jepsen started Openwater, which today is in the process of commercializing a hospital-grade technology platform that integrates semiconductor physics, light, and sound to diagnose and treat diseases at the cellular level.
While the technology itself is something to write home about, with preclinical and clinical studies showing promising results like shrinking glioblastoma tumors in mice and treating severe depression in humans, perhaps the most innovative part of Openwater is its business model, which is to open-source its technology to accelerate medical innovation by reducing complex medical device regulatory time and costs.
“Smartphones are available in refugee camps, but they don’t have cancer therapy,” Jepsen said. “With this ability, for example, of camera chips shipping in smartphones with pixel sizes the size of the wavelength of light that can see in the near infrared, we can… perhaps create better diagnostics and therapeutics that can be made in the factories that make our smartphones at those kinds of prices. We can use the factories to make things that are IRB ready that can enable us to jump through clinical trials in parallel for all kinds of aggressive cancers, mental diseases, cardiovascular diseases, and more.”
There’s lots of buy-in to the idea. Openwater announced last week that it has secured $100 million in total funding from new and existing investors to advance its mission of making advanced medical care accessible worldwide. New and existing investors include Plum Alley Ventures, Khosla Ventures, BOLD Capital Partners, Esther Dyson, and Peter Gabriel, among others.
Cheap camera chips for medical devices
The time and resources it takes to develop therapeutic complex medical devices are estimated to take around 20 years and require $94 million from concept to approval, according to a report from the U.S. Department of Health and Human Services (HHS). Jepsen, who cited a different source that puts these values at 13 years and $658 million, thinks that the open-source approach could cut the concept-to-approval process for therapeutic complex medical devices down to three years and $10 million.
Beyond coming from a technology background, one of the reasons that Jepsen isn’t trying to create a One Laptop per Child version for drugs is that the cost of the latest and greatest in therapeutics, like CRISPR-based cell and gene therapies that have limited, if any, safety data and aren’t priced the same way as cell phones and computers—they’re going for millions of dollars.
“The cost of [Casgevy] is a million dollars, and it’s an underestimate to say the vast majority of humanity cannot afford a million dollars for [Casgevy],” said Jepsen. “So, why are government dollars funding that when it’s just for the rich? And if that research succeeds, the rapid distribution and impact on humanity is part of the equation for funding it in time scale. [For instance], sub-Saharan Africa is still waiting for the next round of antiretrovirals for HIV. Can we use consumer electronics in this way?”
Even when it comes to therapeutic complex medical devices, Jepsen says that research institutions and companies can spend millions of dollars and many years building new rigs based on new technology with very slow life cycles. Or if they’re going to get one made for them, there’s the whole process of working with a manufacturer, getting parts shipped into the lab, and then sending a technician to put it all together.
“Consumer electronics is motivated on scale,” said Jepsen. “For example, Apple could have sold that first iPhone for $5M a piece. People would have bought it. They had decided the value is the network. The value could instead be saving lives. Technology is largely driven by open-source.”
So, she’s convinced that the best way to slash costs and time from concept to approval is not by figuring out how to use their own technology to drive one product to the finish line but instead to make Openwater’s platform based on “a camera chip that costs a dollar” open-source. Pointing to the decades of research and countless papers that have established how to use ultrasound and biophotonics therapeutically, Jepsen thinks that people working with these modalities could simultaneously be making use of Openwater’s miniaturized and inexpensive technology to develop new therapeutic complex medical devices.
The open-source approach puts the scientific spirit of testing and retesting into action. As Jepsen put it, part of open-sourcing is to engender the trust to enable this to go through more iterations.
Jepsen said, “If you can look at how something works and change it, you can go through more iterations to make a better product faster. So, why don’t we do that in healthcare?”
This concept, Jepsen said, was a struggle for her investors to swallow. But she’s run the numbers, and her approach of hoping to treat millions of people for a few bucks instead of a few patients for $1 million comes out being more profitable—10–100 times more profitable.
“The business model is, ‘How do we get more of this brilliant work to be used by people and give the regulators a hundred times more data than they’ve ever had before?’” said Jepsen. “So, we’re letting everyone use our platform that’s very highly reconfigurable because we’re able to use new chip technologies.”
Medical resonance
There are now a few case studies for how Openwater’s healthcare platform—which combines high-resolution infrared imaging, precision-tuned ultrasound, and targeted electromagnetic fields for visualization, monitoring, and treatment of biological issues—can quickly go from concept to approval.
For example, Openwater developed technology for focusing ultrasound for therapeutics by developing chips that can focus ultrasound frequencies and rhythms at safe and effective levels. Jepsen said that one application of this is much like when an opera singer can ping a wine glass and then match that note with her voice to break the glass without harming anything else in the room, but swapping in a chip for the human and a target cell for the glassware. Because cell types have different mechanical properties, specific parameters of the ultrasound can be “tuned” to the cells for a variety of outcomes, such as targeting the brain cancer that once almost got the best of Jepsen. By running through different frequencies and rhythms over many octaves and cycles, they’ve been able to kill glioblastoma in brain organoids and in mouse models without harming healthy tissue.
In an effort to quickly validate the ultrasound technology in humans, Jepsen and Openwater switched from ablating cancer to a new modality upon finding an institutional review board (IRB) document for neurostimulation, shifting ion channels to inhibit or excite neurons.
“You can turn on and off neurons, and you don’t need Elon’s one-inch hole in your head—we’re doing it noninvasively,” said Jepsen.
So far, Openwater has done a small study with their ultrasound on people with severe treatment-resistant depression, which in recent years has seen an uptick in the use of transcranial magnetic stimulation (TMS) with moderate results. Part of the problem is that TMS can’t be focused very well and doesn’t have very deep penetration, getting blocked by the skull and limiting brain targets, both challenges that can be overcome by the ultrasound approach, which Jepsen proposes to be a “form of precision medicine.”
They’ve also used the same fundamental technology and applied it to infrared to see blood flow at far better resolution, by some metrics, than a multimillion-dollar MRI or CT scan. Jepsen also wants to see Openwater’s technology applied to treating blood vessel occlusions and stroke, the fifth leading cause of death in the United States.
Jepsen said, “That’s just a lot for us to do. We have a platform, and everywhere I go to talk about it, everyone asks, Can I get one of your units? So we decided to change our business and make them for all. Maybe we can do more good and make more revenue in a way that’s more impactful.”
Nevertheless, what really matters to Jepsen is that by providing the chip architectures and software as open-source, there’s no reason for Openwater to go the long and winding (and expensive) route of developing a completely new medical device approval, which could bankrupt the startup. Instead, they can be the seed for anyone to start a clinical trial with the Openwater platform in their field of expertise.
“How can we, as a startup, take a thousand shots on goal?” asked Jepsen. “This enables us, with our partners, to take that number of shots. We should be in one hundred clinical trials next year. I’m hoping it’s a thousand.”
Making fire
The existing global manufacturing capabilities for electronics, especially in Taiwan and China, are what Jepsen thinks will drive Openwater to success. And it goes beyond the physical infrastructure and factories—it has to do with capturing the hearts and minds of people everywhere. It also doesn’t hurt that it’s a bit easier to explain the science behind the modalities being powered by Openwater’s tech than something like CRISPR.
“I have to explain infrared light, and I say, the campfire was really important,” said Jepsen. “They can see the light. They feel the heat—it penetrates your belly and warms you. Infrared is around us all the time and we know the safe levels of it. So why not use it, given that we have these fantastic results and we stand on the shoulders of these giants with these fantastic results? We’ve lowered the barrier to entry while using the consumer electronic supply chain.”
All that said, Openwater hasn’t entirely abandoned the idea of commercializing its own devices. You can reserve one of two devices—Open-Motion 3.0 and Open-LIFU 2.0 (low-intensity focused ultrasound)—for $100. Each comes with a small box called a console and a device with either sensors or transducers, respectively, that looks like any wrist-wearable (e.g., FitBit, iWatch). But it doesn’t seem like Openwater is focused much, if at all, on these. The manuals have an almost sardonic quality, with pictures of the devices strapped to a model’s head or abs, drawing a parallel to those belts for shocking your stomach into a 6-pack.
The real selling point of these products, which is described at the very beginning of the manual, is to provide an example of how Openwater is filling the gap between expensive research hardware available for proof-of-concept studies and commercially manufacturable hardware suitable for regulated medical devices. They are to encourage anyone to take the next great idea from the bench to bedsides worldwide.
