Can the path to equitable healthcare avoid insurers?

There are few challenges messier and more fraught than the U.S. healthcare system, but a growing number of startups are looking at ways to address shortcomings in standards of care through tech. We had three such companies share our virtual stage at TechCrunch Disrupt 2021 this year, including Cityblock Health president and co-founder Toyin Ajayi, Forward CEO and founder Adrian Aoun, and Carbon Health‘s Eren Bali.

Let’s just say this conversation got heated — fast.

The main point of contention arose around defining what constitutes customer-centric healthcare and Aoun’s stance that, regardless of what else is involved in a company’s approach, starting from a point of working with insurers disqualifies a company from making any consumer-centricity claims.

“We keep saying that these companies are kind of consumer-centric,” Aoun said, referring to the panelists. “But in many ways I think one of the things that you realize is that when you get in bed with the insurance companies, which, whether it’s a Carbon or a Cityblock, at the end of the day, [if] you get in bed with the insurance companies, unfortunately, your incentive is basically not to go build a good consumer product.”

“Your incentives are actually not the right thing — they’re not what the consumer needs,” he added. “So at the end of the day, you’re [referring to Eren and Carbon] launching a scheduling feature. We’re launching a heart health program that eliminates high blood pressure for 40% of our members. You’re launching a new way to bill; I’m launching cancer prevention.”

Ajayi took issue with the binary Aoun was trying to establish and explained why it’s actually not such a clear-cut division between working with insurers and having a real and meaningful focus on patient outcomes.

“Adrian has said, either you get reimbursed by insurance, or you build a consumer or patient-centered company. And you know, in parentheses, that only very wealthy people can afford. What we found is actually that’s not binary; there is another path, which is partner with insurers, but take risk on the total cost of care and outcomes. So we do not bill for a community health worker coming to your home, holding your hand, telling you that you matter and helping understand what goes on in your life. But we absolutely are incentivized to do that and to innovate in that space, because that allows us to earn the right to provide healthcare to people that make them healthier.”

StethoMe’s smart stethoscope lets your kid’s doctor listen to their lungs from afar

When you or your kid have any sort of respiratory issue, figuring out what’s happening minute-by-minute — and how well treatment is working — is a stressful, frustrating, and anxiety-filled process. I imagine it’s all of the above and more in the middle of a friggin’ RESPIRATORY DISEASE pandemic.

StethoMe, a team competing in this week’s TechCrunch Disrupt Startup Battlefield competition, is looking to help alleviate some of this for children with asthma and their parents. It has built a smart, connected stethoscope that can help parents perform lung examinations at home, sending high-fidelity recordings directly to their kid’s doctor and using machine-learning to help flag potential concerns.

This is the device:

Image Credits: StethoMe

Turn it on, use your phone to tell it what kind of exam you want to perform, and the built-in screen will walk you through the process. It’ll tell you where on the chest to place the device, whether or not the room you’re in is quiet enough, and more. After measuring across 6-8 points, it’ll provide you a report with details like respiratory rate, heart rate, and whether or not it detected any audio abnormalities — including wheezing, rhonci (gurgling sounds caused by fluid), or crackles.

From there, you’re able to send a link to the report directly to your kid’s doctor, where they can hear the recorded audio from each point on the chest. A scrubbable spectrogram, meanwhile, provides a visual overview of each recording and flags and labels any abnormalities detected by the system. That report looks like this:

A StethoMe screenshot showing a visual overview of each recording so doctors can listen to a kid's lungs from afar

Image Credits: StethoMe

This information is meant to help parents and their doctors detect asthma attacks earlier and more accurately, and to help determine how well long-term medications are working — is one medication better than another at alleviating harder-to-detect symptoms? Did bumping up the dosage slightly help?

Co-founder Wojciech Radomski tells me their product is already certified as a medical device in the EU, having obtained a CE mark for both the AI and the device; the FDA approval process in the U.S., meanwhile, is underway.

At TechCrunch Disrupt, the company announced a deal wherein Poland’s Ministry of Health has purchased 1,000 devices to run a pilot test with over 100 doctors over the next half a year. “During this last month [alone],” Radomski tells me, “they’ve already made over 70,000 recordings.”

To maybe straddle the edge of too personal here: I freakin’ love this idea. I had asthma growing up. It dominated my life for a few years; even once the doctors got it under control (thanks science, love yooou), six-year-old me was always convinced I was having or about to have an asthma attack. The fear of being unable to breathe triggered crushing anxiety, which in turn convinced me I couldn’t breathe. While I can’t speak to how well this thing works at this point (that’s the FDA’s job), I wish I could package this thing up and stick it in a time machine and send it back to lil’ me in 1993 with a note that says “Use this, breathe easier.” (and maybe “p.s. buy bitcoin early” but I guess we shouldn’t screw with the timeline too much.)

StethoMe says it has raised a few rounds at this point (a $400K pre-seed, $2M seed, and $2.5M Series A) and received nearly $3M in grants from Poland’s National Center for Research and Development.

 

Cellino is using AI and machine learning to scale production of stem cell therapies

Cellino, a company developing a platform to automate stem cell production, presented today at TechCrunch Disrupt 2021 to detail how its system, which combines A.I. technology, machine learning, hardware, software — and yes, lasers! — could eventually democratize access to cell therapies. It aims to bring down costs associated with the manufacturing of human cells, while also increasing yields.

Founded by a team whose backgrounds include physics, stem cell biology, and machine learning, Cellino operates in the regenerative medicine industry. This space is currently undergoing a revolution, where new developments in gene and cell therapies could lead to breakthrough cures for a number of leading diseases. For example, the use of personalized human retinal cells could be transplanted to halt or reverse age-related macular degeneration, which can cause blindness. But today, such cell therapies are out of reach for most people because the process of cell production hasn’t been automated or made scalable and efficient.

Instead, human cells being used now in these clinical trials are mostly being made by hand by scientists who are looking at cells and evaluating — using their many years of training and expertise — which cells are low quality and need to be removed. They then scrape away those unwanted cells with a pipette tip. The process, as you can imagine, is time-consuming and produces only a small yield. In this manual process, you’d see a yield of about 10% to 20% of cells that would be able to pass the final quality assurance tests required for human transplant.

Cellino is working to improve this process in order to produce more cells of higher quality. Its goal is to push the yield to at least 80% over the next three years.

To do so, Cellino’s system is automating all the human steps in the production process using machine learning techniques.

To identify which cells are high quality or low quality, the company is collecting large training data sets where it’s teaching algorithms to make determinations about cell quality based on a variety of factors. This includes the cell morphology — meaning, the shape, size, and density of cells. Fluorescence-based surface markers can also be used to identify other factors of importance to the line of cells being produced, like the location of proteins on the cell, for example.

By using machine learning and AI to do the identification based on standard and well-accepted biological assays used by the FDA, the system could move away from human annotation and the variability that introduces into the process of human cell production.

After Cellino’s software has identified which low-quality cells need to be removed, it then uses a laser to target them. The laser creates large enough cavitation bubbles to kill the cell, but it’s done in a highly localized way where you’re not harming the neighboring cells, as thermal heat does not dissipate to the nearby cells. This is also a more precise technique than the manual method. (Cellino’s system has a 5-micron resolution, while cells are 10-15 microns in size). This results in a throughput of about 5,000 cells per minute, which is highly efficient compared with manual techniques.

Over time, this automation and efficiency could bring the cost down from nearly a million dollars per patient, which is what clinicians have to pay today to run a clinical trial, when outsourcing cell production. Cellino aims to get the cost down into the tens of thousands of dollars over time.

By scaling cell production, personalized cell therapies could also help a broader range of patients compared with other techniques relying on banks of stem cells. These aren’t always genetically diverse samples, leaving smaller ethnic groups out of the progress being made in this space. Banked cells also require recipients to take immunosuppressants, as the cells aren’t your own and the body may reject them.

The use of lasers is an idea developed by Cellino co-founder and CEO Nabiha Saklayen, who patented an invention in cellular laser editing while at Harvard earning her Ph.D. in Physics. She was encouraged to turn the technology into a startup by her collaborators, who included had leading biologists like George Church and David Scadden.

“Not all scientists become entrepreneurs, and I became an entrepreneur because I had an amazing support network around me,” notes Saklayen, of the push to join the startup space. She immediately recruited Marina Madrid, an applied physicist she had worked with for years on the co-invention of laser-based intracellular delivery techniques, as her other co-founder. To gain more mentorship about growing a startup, Saklayen turned to the Boston area startup ecosystem.

“I didn’t know anything about startups. I wanted to work with people who knew how to build companies, how to commercialize technology, how to build instruments —  and the Boston ecosystem is fantastic in that way. So I started connecting with lots of people in those early weeks — anybody that was in the biotech realm or Harvard Business School,” Saklayen explains.

This led her to Cellino co-founder and CTO Mattias Wagner, who had built companies before in the optics and instrumentation space.

“That’s how the founding team came together. It was very complimentary because Marina and I were co-inventors of the original technology that inspired the platform and Mattias brought this tremendous background in semiconductors and optical instrumentation,” says Saklayen.

Since its 2017 founding, Cellino has gone on to raise $16 million in seed funding in a round co-led by The Engine and Khosla Ventures, with participation from Humboldt Fund and 8VC.

The company is now collaborating with the NIH on compatibility studies. Currently, that means Cellino is making stem cells on its system which it’s then comparing with the ones made at the NIH that are already being tested in humans for personalized cell therapies for retinal diseases. Cellino later hopes to use its system to address areas like Parkison’s, muscle disorders, and skin grafts, among others.

The company wanted to present at TechCrunch Disrupt to share more about what it’s building and to source new talent.

“For me, it’s about talking about this idea around democratization and industrialization of cell therapies. I really want to get that message out because that is the movement we need to drive over the next decade for all of these cell therapies to be accessible to all patients,” says Saklayen.

“Cellino’s angle is also very unique in the sense that, because we have this automated system to manufacture human cells, our system could make cells for every human being — in this country, in the world,” she continues. “And there are a lot of cell therapy approaches that are looking to use off-the-shelf cells and off-the-shelf therapies, which will only work for certain parts of the population. As the U.S. becomes more diverse, ethnically, we need personalized solutions for everybody.”

 

The Blue Box is betting on the future of at-home breast cancer tests

You can take a pregnancy test or colon cancer test from your bathroom, or, these days, a COVID-19 test from the comfort of your living room. You might one day be able to get a breast cancer screening at home, too, if you have a urine sample and an artificial nose. 

That’s the vision behind The Blue Box, a startup competing this week at TechCrunch Disrupt’s Startup Battlefield. The company, founded by Judit Giró Benet while pursuing her Master’s at the University of California Irvine, is developing an at-home handheld device designed to screen urine samples for breast cancer. 

The company, founded in January of this year, is in the process of scientifically validating The Blue Box – which includes both hardware and artificial intelligence components. The Blue Box has been awarded equity-free prizes from Argal (€2,000), the 2020 James Dyson Prize (£35,000), a grant from the Tarragona region of Catalonia (€4,000), and a prize for winning the pitch.tech competition ($10,000). 

Benet imagines a product where you might be able to slip a urine sample into an $80 box, have your sample analyzed by a machine learning algorithm (that algorithm is being trained right now), and have test results sent to your phone in about 30 minutes. 

“You would have a Blue Box at home and the whole family could use it at home with the frequency your doctor tells you [to],” Benet tells TechCrunch. 

Benet says the device is modeled after a series of studies showing that dogs are able to pick up distinctly cancerous smells. 

For instance, early-stage studies have shown that specially trained Labrador Retrievers can accurately identify early-stage colon cancer in both breath and stool samples. Urine, the particular bodily fluid of interest to The Blue Box, also has proved to be useful for cancer sniffing dogs. In one study, German Shepherds were able to identify prostate cancer patients by sniffing out “volatile organic compounds” present in urine samples. 

One June 2021 trial on 40 breast cancer patients, 142 patients with non-breast malignant disease, and 18 healthy people found that a trained Labrador Retriever could accurately identify the breast cancer patients’ urine samples 40 out of 40 times in double blind-tests. The authors concluded that a screening method based on detecting compounds in urine warranted further study. 

The Blue Box is designed to help sniff out breast cancer, sans the dog component. 

You might think of The Blue Box itself as a replacement for the dog, and the AI component as a digital brain. 

The key for The Blue Box will be clinically validating both parts of the equation. Benet declined to share the specific cancer biomarkers that The Blue Box will test urine samples for – though she noted that they are pulled from scientific literature. 

So far, Benet says The Blue Box has a minimum viable hardware product that’s “fully functional.” 

The next piece of the puzzle is training the machine learning algorithm to recognize late state breast cancer. So far, the company reports a 95 percent classification rate for their algorithm on metastatic breast cancer (a very late stage) – which means if can accurately categorize 95 percent of those samples. 

That’s a first step for the company, but the goal is to be able to detect cancer before it reaches that especially dangerous stage. On that front, The Blue Box is still in the throes of clinical validation. The Blue Box, she says, is currently being studied at University Hospital Joan XXII in Tarragona, Catalonia, and University Hospital Sant Joan in Reus, Catalonia, in a study led out of the University of California, Irvine. So far, they’ve collected more than 40 urine samples. 

Should The Blue Box be able to prove that their technology can accurately detect early-stage breast cancer, there’s evidence that at-home cancer screening tests can gain regulatory approval. 

In 2014, the FDA granted premarket approval to Cologuard, a prescription stool test that is designed to detect colon cancer in average-risk individuals. For many screening tests, the critical measure of success is sensitivity or the ability of the test to accurately detect a disease when it is present  –  the sensitivity of colorguard for the detection of colon cancer was 92.3 percent, per a study in the New England Journal of Medicine. 

To follow in those footsteps The Blue Box would also need to have rigorous clinical data to earn pre-market approval and will have to demonstrate high sensitivity. 

The Blue Box is also technically a medical device that delivers oncology-related information, which means it will need to work with the FDA to demonstrate its validity before going to market. 

“We are a medical device so we will need to go through the FDA and the MDR [the European equivalent of the FDA]. We will start this phase by 2023,” she says. 

One of the many arguments made in favor of at-home cancer testing is that it could help close a screening gap. 

In the context of colon cancer, fear, whether over the procedure itself or the result, have consistently been identified as barriers to testing in studies, but other barriers — cost, lack of insurance or transportation, or skepticism about screening guidelines — also remain powerful. 

Breast cancer screenings, argues Bénet, face some of the same barriers that colonoscopies do. There is room for improvement in terms of screening. In 2019, 76.4 percent of women in the US aged 50 and over had gotten a mammogram in the last two years. This number has remained relatively stable since 1998, per the National Cancer Institute, though the pandemic also resulted in a spike in missed screenings that will likely affect more recent statistics.

A 2014 study on underserved women found that the largest barriers to obtaining mammograms was fear of cost (even if the service was free, the fear persists), mammogram-related pain, and the fear of receiving bad news. 

The Blue Box, perhaps like Cologuard, seems poised to tackle fears of mammogram-related pain (urine samples are painless). Benet has also thought about the process of receiving the bad news her product might deliver.

Benet says the company is working on incorporating a “virtual doctor” within The Blue Box app that can communicate with a user once they’ve received a diagnosis. “We will try to train this bot so that it can sense the mental state of the patient,” says Benet. “If she’s processing the news correctly, if she needs help from a medical professional.”

That feature, she says, should be unveiled in the next few months. 

At-home testing isn’t a universal salve to the multifaceted reasons people don’t get cancer screenings. But there is evidence that these at-home tests do reach people who might otherwise forgo a screening. 

At-home screening tests, for instance, have allowed healthcare systems to mail out tests to people who might otherwise miss an appointment. A 2018 review paper found that when people received colon cancer tests by mail, colon cancer screenings went up about 22 percent.

The Blue Box is still in its early validation phase, but if the company can make a similar dent in the breast cancer screening world, Benet hopes it will lead to more people catching breast cancer during early, critical stages. 

“I believe that with The Blue Box we will be able to finally create a change that should have happened many, many years ago,” she says. 

 

Prenome could help pregnant women better predict and manage gestational diabetes

“We’re always trying to stick something up a women’s something.”

Stevie Cline, the co-founder of Prenome, is tired of how invasive diagnostic processes are for women, even with modern technology and broader healthcare advancements. So, she teamed up with her former co-worker, Sarah Brozio, to launch a startup all about providing actionable insights to women – no poking or prodding needed.

Prenome, going through TechCrunch Disrupt Startup Battlefield this week, is launching a saliva-based genetic test, paired with a behavioral app, to help women navigate gestational diabetes, a common condition that comes up while pregnant.

The startup’s success is rooted inextricably to data. In May, Prenome began recruiting patients who have had gestational diabetes for a clinical study, OneInSeven. The co-founders are using this data to build a more accurate pipeline of information about women’s health conditions, one that represents the actual diversity of women in the United States.

Prenome could help pregnant women better predict and manage gestational diabetes

Image Credits: Prenome

Once Prenome extracts a genetic risk score, it combines it with a risk score built off of social determinants, with factors like previous trauma, disposable income, proximity to grocery stores. By combining these two factors, Prenome claims it can accurately predict gestational diabetes at a 90% accuracy rate, compared to the 80% baseline today.

The co-founders picked gestational diabetes over the plethora of other women’s health conditions because it has a strong genetic component. Cline explained how gestational diabetes is predictable and preventable, which gives Prenome a chance at screening high-risk individuals and also providing interventions that will help them prevent getting gestational diabetes early on.

Once Prenome is able to develop technology that helps women navigate gestational diabetes, it can use that insight to better address associated conditions, such as infertility and polycystic ovary syndrome.

“People think that genetic diseases are usually caused by a single gene or a single chromosomal disorder, like Down Syndrome or Cystic Fibrosis,” Crozio said. “[They] wanted this idea that one gene equals one problem so we can fix that – but that’s actually really untrue for most of our health.” 

She explained part of the driver for the biobank is to be able to mine it for more and more indications on triggers within women’s health conditions. Similar to hormonal health, genetic health is complicated yet key in understanding the common threads between these conditions.

Prenome’s biggest stance may be that it doesn’t want to test for any condition that it can’t intervene or improve on.

“We’re never going to do a test [for] risk [of] Alzheimer’s in your 80s because there’s nothing really right now that we can do to prevent that from happening, women don’t need more stress” Cline said. “If we are telling you [that you are] high risk, it’s something that you can improve on, we can intervene on, we can provide for.”

Over a year since launch, Prenome has so far filed patents on both its test and methodology. It also plans to publish research with data from its OneInSeven Study.

The saliva-based test, which it hopes to release in 2022, is working to get CLIA-certified, which would allow physicians to order it for their patients. Eventually, Prenome will pursue FDA-approval, which would allow Prenome to go direct-to-consumers.

“Having doctors feel like they are cut out of the loop and that they are not important and part of the women’s health cycle is detrimental for us,” Cline said. “We really want to focus on how we engage people [and] get them to want to get their patients to use our tests.”

Prenome early strides have been noticed, and bet on. So far, Prenome has attracted $1 million worth of interest from investors, including Cleo Capital, OnDeck’s Runway Fund, Johns Hopkins University, Wing VC, First In Ventures and A-Level Capital.

 

EyeGage is building a database of eye scans for drug testing

LaVonda Brown developed an interest in eye-tracking during her time at Georgia Tech. The fascination with all the information that can be derived by scanning the so-called Windows to the Soul formed the foundation of EyeGage, one of the 20 companies competing at this year’s Disrupt Startup Battlefield.

The startup’s entry into the TechCrunch competition arrives as EyeGage launches its first product: an app designed to let users know if they’re sober enough to drive. If not, they’ll get a big red “Do Not Drive” warning and a link to call either an Uber or Lyft. The application is free and serves a dual purpose. In addition to the obvious consumer-facing purposes, it doubles as an opt-in for EyeGage’s growing dataset of eyes.

“Consumers can download the app, take pictures of their eyes and then we can suggest whether or not they should use rideshare. Essentially not driving, based on their eyes,” explains Brown. “That is free. I like to call it a barter subscription service. They give us pictures and videos of their eyes and we give them access to the technology so they can make a responsible decision.”

The app is the most forward-facing aspect of EyeGage’s business at the moment – and like most of what the company does, it will go toward building out its dataset of eyes. The company will start with measuring the impact of alcohol on various aspects of the eyes, including studies it’s currently conducting with participants in a federally approved testing facility. Those who sign forms to participate will drink booze, while the company collects images and videos of their eyes, along with a blood sample.

Marijuana is next on the list, given its current legal status in certain states. Other drugs like opioids, amphetamines and benzodiazepines will be more difficult to gather, though hospitals and clinics that mete out legal versions of these substances could prove a good source for collecting that data – with the proper consent.

Brown says workplace environments are a logical next step, as well. Law enforcement is also on the list, though there are various hurdles to attaining those sorts of partnerships. “We’re targeting high-risk workplaces like construction, manufacturing and transportation. In those industries in particular they have a high rate of drug and alcohol use,” she tells TechCrunch.

There may also be a potential use for the company’s dataset beyond its immediate use for detecting substances in the body.

“Monitoring eye behavior can be used for so many domains,” Brown adds. “And, of course, you can identify someone by their eyes. It can be used to diagnose certain illnesses, concussions or diabetes, and it can be used in different market segments. Your eyes are so informative about what’s going on in your body. They can tell if you’ve had caffeine, depending on how it responds to light. If it’s too fast, it’s some kind of stimulant. If it’s too slow, it’s some kind of depressant.”

EyeGage has raised $142,455, to date. That includes $42,455 in pre-seed from friends and family, as well as a recent award of $100,000 from Google Black Founders Fund.

For BioNTech, the COVID-19 vaccine was simply the opening act

BioNTech’s founding story dates back to the late 1990s, when CEO and co-founder Uğur Şahin, his wife and co-founder Özlem Türeci, and the rest of the seven-person founding team began their research.

Focused specifically on an area dubbed “New Technologies,” mRNA stood out as one area with tremendous potential to deliver the team’s ultimate goal: Developing treatments personalized to an individual and their specific ailments, rather than the traditional approach of finding a solution that happens to work generally at the population level.

Şahin, along with Mayfield venture partner Ursheet Parikh, joined us at TechCrunch Disrupt 2021 to discuss the COVID-19 vaccine, his long journey as a founder, what it takes to build a biotech platform company, and what’s coming next from BioNTech and the technologies it’s developing to help prevent other outbreaks and treat today’s deadliest diseases.

“At that time, mRNA was not potent enough,” Şahin recalled. “It was just a weak molecule. But the idea was great, so we invested many years in an academic setting to improve that. And in 2006, we realized ‘Wow, this is now working. Okay, it’s time to initiate a company’.”