7 AI Healthcare Startups Automating Clinical Trials
Table of contents
It’s no secret that it’s extremely expensive to develop new drug therapies. One study estimates that pharmaceutical firms spend between $2 billion and $3 billion and more than a dozen years in R&D activities to get a new FDA-approved drug to market. An important step in the process is the clinical trial where human guinea pigs volunteers are used to test the toxicity and efficacy of a new treatment. One recent paper has argued that the median cost of the clinical trials needed to win FDA approval is only $19 million. Regardless, the price tag is still pretty high, especially for many mom-and-pop biotechs. That’s why startups like Qolty are trying to digitize medical research using big data and the Internet of Things (IoT). Of course, it was only a matter of time before AI healthcare startups began applying machine-learning algorithms to further automate clinical trials.
We’ve written quite a bit about how artificial intelligence is being used more generally in drug discovery. We even profiled one of the leading companies in this emerging sector, Insilico Medicine, which just announced its latest hookup with the pharmaceutical industry – a research collaboration with Pfizer (PFE). While Insilico Medicine has gained some fame and fortune (more than 70 published and peer-reviewed papers, along with $52 million in funding) for its work in biomarker and small molecule identification, the company also believes its AI-powered platform can accelerate clinical trial outcomes.
It isn’t alone. Below we profile 7 AI healthcare startups that are automating medical research, especially around clinical trials, in much the same way AI is being used to expedite other types of scientific research.
Getting Real World Evidence for Clinical Trials
As you would expect, designing better clinical trials begins with bigger and better data. That’s part of the value proposition behind a Boston area startup called TriNetX, which has reportedly raised $102 million since it was founded in 2013. It picked up $40 million in a Series D last March, led by healthcare giant Merck (MRK). The next month, TriNetX went out and bought a Belgian company, Custodix NV, which provides services around clinical trial design and patient feasibility. TriNetX itself reportedly uses its software to collect and analyze patient health records to identify ideal candidates for clinical trials, and the latest cash infusion will help expand its analytics platform in the areas of AI and machine learning. Here’s a look at one of its products:
TriNetX claims it provides real-world evidence from global clinical and claims data from more than 300 million patients to its customer base, which includes 29 industry players and consists of nine of the top 15 pharmaceutical companies including Novartis (NOVN), Sanofi (SAN), and Pfizer (PFE), as well as five of the top contract research organizations.
A Clinical Analytics Platform
AI has become synonymous with analytics in some circles. Such is the case with a clinical analytics platform from a Silicon Valley company named Saama that has been around since 1997. However, the company only started taking in venture capital since 2015, raising $75.8 million, including a $40 million round last March. In August of last year, Saama acquired rival and Silicon Valley neighbor Comprehend Systems for an undisclosed amount of money. Saama’s platform, Life Science Analytics Cloud, “seamlessly integrates, curates, and animates unlimited sources of structured, unstructured and real-world data to deliver more actionable insights.” Yes, we’ve heard this refrain before, and it even comes with the requisite graph showing data going into the black box of AI, which then spits out insights and strategy:
Customers include heavy hitters from healthcare, biotech, and pharma including Celgene (CELG), Gilead (GILD), and Roche (ROG).
Digitizing Clinical Trials Using Machine Learning
Irish startup Teckro is attempting to simplify clinical trials by using machine learning to digitize paperwork, enabling researchers to collaborate on smartphones from the golf course. Founded in 2015, the company raised $25 million in a Series C last February, to bring total funding to about $41 million. Basically, the company provides a mobile platform that allows clinical researchers to quickly retrieve information and control various protocols using the portal. Teckro claims to manage more than 100 clinical trials and is deployed at more than 12,000 active sites around the world. It is reportedly eyeing an IPO.
Drug Discovery for Designing Clinical Trials
New Yawk-based Owkin supports drug development in order to help companies eventually design better clinical trials using AI. Founded in 2016, the startup has raised about $18.1 million, including an undisclosed Series A in March. Before then, it picked up a cool $5 million in a round led by the Google (GOOG) venture arm in 2018. Owkin’s flagship platform, Socrates, uses machine learning to integrate biomedical images, genomics, and clinical data, among other sources, to discover biomarkers and mechanisms associated with diseases and treatment outcomes. It’s also working on an exit strategy:
Owkin recently published a paper in Nature Medicine about a deep-learning program trained on 3,000 patients with an aggressive form of lung cancer tumor that enabled Owkin to develop a prognostic model. The company is now working with partners in the biopharmaceutical industry to use those insights for clinical trials by identifying high-risk patients who have little to lose by trying might respond best to an experimental drug.
Pivoting from Pathology to Clinical Trials
We’ve written before about how companies are using AI to accelerate and improve pathology analyses, probably the most important step toward successfully diagnosing and treating cancer. That’s where Columbus, Ohio-based Deep Lens got its start. The startup has raised about $17.2 million since leaving stealth mode in 2018, a year after it was founded, including a $14 million round last April. The company’s AI platform, Virtual Imaging for Pathology Education and Research (VIPER), was originally a diagnostic tool for identifying cancer that Deep Lens gave away for free. Now the company is leveraging all of that cloud-based data and its network of partner institutions to flag eligible patients at the time of their diagnosis in order to fast track enrollment in clinical trials.
No Animals Were Harmed
We realize that we’re just one animal experiment away from Planet of the Apes, so it’s good to see a company like VeriSIM Life use AI to do away with the need to test new drugs on rats and orangutans. Founded in 2017, the San Francisco startup has raised a total of $6.4 million over a couple of Seed rounds, including $5.2 million last August. VeriSIM is developing AI-powered biosimulation models that can quickly predict how a drug will interact with an animal’s biological system, VentureBeat reported, which means pharma companies could accelerate the pre-clinical phase and move on to the zombie apocalypse human clinical trials.
VeriSIM said it plans to eventually create similar digital twins of humans, beginning with Arnold Schwarzenegger and Danny DeVito.
Combining IoT and AI for Clinical Trials
Austin, Texas-based Litmus Health is doing what an increasing number of AI healthcare startups are doing: taking data from wearables, sensors, and other smart devices and turning them into medically relevant insights. In this case, the startup’s platform uses machine learning to detect patterns in the data based on participant behavior and responses. There’s also a dashboard for real-time monitoring of the clinical study:
In one case study, a pharmaceutical company and the University of Chicago are conducting a three-year, 500-person study that is gathering data from smartphones and Fitbits to understand the relationship between sleep and flare-ups in Crohn’s and colitis patients. Eventually, the researchers can use the data to predict when flare-ups will occur based on heart rate, sleep, and physical activity.
Conclusion
AI healthcare startups are working to automate new drug development from the initial identification of biomarkers and therapies through the complexities of clinical trials. It’s a journey that until recently has been measured in years and billions of dollars. The pharmaceutical industry appears to believe that these companies can provide real value and legitimate shortcuts in discovering and testing new drug candidates. The big pharma players are both investors and customers – and, as some of these platforms reach maturity, likely buyers – of many of these startups. That gives us some confidence that while there’s a lot of buzzwords being bandied about by some of these companies, they’re also creating value.
Share
