Originally published in DPHARM: Disruptive Innovations to Advance Clinical Trials Newsletter
physIQ won DPHARM Idol Disrupt in 2017 for their cloud-based artificial intelligence platform that combined continuous streams of data from wearable sensors with algorithms to provide physiological insight in clinical research and predict potential adverse events in patients. Chris Economos presented in 2017 on behalf of physIQ and is the company’s Chief Commercial Officer.
We specialize in collecting and analyzing continuous data that streams from wearable biosensors. There has been a long history in clinical trials of collecting data to support efficacy and safety endpoints. Most of that data over the years has been taken sporadically in a clinic. Wearable sensors enable the opportunity to passively collect that data, hour after, day after day, month after month, to develop a much richer story of patients’ responses to the drugs they’re taking in the context of a clinical trial -- responses related to both efficacy and safety.
Now, in the world of IT, these continuous data sets are unique. If you’re collecting every heartbeat, or every breath or every subtle movement, you need extraordinarily sophisticated analytics to make sense of it. That’s what we specialize in.
So while we work extensively with wearable sensors, we’re not a hardware company. We don’t make the sensors. Rather, we partner with the companies that make the sensors. You have different sensors for different jobs. For one type of therapeutic area, you might want one type of sensor and another therapeutic area, you want a different type of sensor.
What we specialize in is a clinical-grade digital platform that ingests the data from these sensors and extracts insight using FDA-cleared analytics. As people are wearing them – we have studies going on in over 17 countries – the data streams to the cloud and then, through our platform, our portfolio of FDA-cleared physiology analytics transforms the raw data into insight…insight that tells that story related to the clinic endpoints that are the focus of a given clinical trial.
2017 was an interesting year in the lifecycle of our company and DPHARM played an important role. We originally launched the company to develop a solution not for clinical trials, but rather, to help health systems better care for their at-risk patient populations. Every day in every hospital, people are being discharged. These are sick people who are still at-risk of deterioration. So we founded the company to provide a solution for these providers who needed better ways to proactively take care of their at-risk patient populations.
The idea was, you identify an at-risk patient, put a sensor on them, stream their data to the cloud, and then use really sophisticated analytics to tell the clinician that this patient is deteriorating. So we built this really clinically robust platform and we learned along the way that there was a need for this to support clinical trials as well.
So it was right around that time that we made that realization and decided, “Okay, we’re going to invest in the pharma space and the clinical trials space.” I think one of our very first investments in pursuing the clinical trials space was DPHARM 2017.
What we specifically presented at DPHARM in 2017 was data about how we were using this particular approach to monitoring an individual’s health in the context of chemotherapy. Anytime you’re giving somebody some sort of anti-cancer treatment, whether it’s chemotherapy or immunotherapy, you’re usually incurring a high risk of adverse events. These are really amazing, powerful therapeutics, but they also tend to carry a high risk of side effects.
We ran a study with the University of North Carolina where they put wearable sensors on their patients who were undergoing some sort of anti-cancer treatment, for example chemotherapy, and we monitored them using our proprietary approach to characterizing physiology.
The goal was to see if we were able to predict these adverse events that might be associated with a patient getting chemo.
I shared a couple of examples of actual patient data where, using this personalized approach to analyzing the data from a wearable sensor, we were actually able to show and predict adverse events that were experienced as a result of the chemotherapy. But the important thing is we were able to show it days in advance.
In this observational study there were patients who were hospitalized or had to go to the emergency department because of side effects related to the chemotherapy. And our analytics actually detected these issues a couple of days ahead of time which, in an active monitoring scenario, would have meant there could be more proactive measures taken to ideally avoid the hospitalization.
We thought that the chemotherapy population and use case would be particularly interesting to the DPHARM audience. As it turns out, it was because we won.
Nobody had ever seen anything like this before. It truly was a game changer in terms of the type of innovation that’s possible in clinical trials. They just never had seen a technology that could measure what our analytics could using wearable biosensor data.
At the time, I think we had one clinical trial customer in one therapeutic area. Now we’re working with, I want to say, about five of the top 10 pharma companies, across a whole slew of therapeutic areas. We’re running Phase III, Phase IV clinical trials across 17 different countries.
There’s two really significant developments, that I think not only speak to physIQ maturing as a company but that really speak to digital medicine maturing within pharma.
First, we’re now selling enterprise licenses to our customers. It used to be that as they were innovating or they would find innovative solutions, they would try it in a study here or study there. That’s a tough model for everybody. But now we have enterprise contracts whereby the pharma mothership is licensing our platform such that they can use it across all of their different studies that they’ve got going on. So irrespective of therapeutic area or irrespective of a wearable device, they’re basically buying or licensing our platform at enterprise-level to use it across the organization.
So to me, I think that speaks volumes for where pharma is, in terms of adopting wearable sensors within their clinical trials. They’re no longer just doing it on a one-off exploratory basis, They’re saying this is now going to be part of the standard and [they] need an enterprisegrade solution and enterprise-grade partner in physIQ to enable that vision.
Another significant development is that we’re now actually partnering with these same companies to develop new digital biomarkers. This algorithm codevelopment model is very exciting where you’ve got these really big corporations who recognize they need novel digital biomarkers to support new endpoints, but also recognize that they don’t have this expertise in house. In order to move as fast as they want, they need to work with experts like physIQ.
What we’re going to see over the next 12 months is a dramatic acceleration in these enterprise platform licenses. That really seems to be where the industry is going; they want a platform that they can use across all their trials, irrespective of the algorithm that they’re going to apply, irrespective of the therapeutic area, irrespective of the device.
I think we’re also going to see a significant acceleration in the number of novel digital biomarkers that are accepted by the FDA. To date in most clinical trials, most endpoints related to wearable sensors have been secondary or exploratory endpoints. I think we’re going to see them become the primary endpoints.
The other key trend is we’re going to see a shift, whereby it’s not so much the internal innovation groups within these pharma companies that are going to be driving this. But rather, it’s going to be the clinical teams themselves.
In other words, you’ve got these clinical teams who are responsible for designing and implementing a study. And then you have these separate innovation groups, whose job it is to find opportunities to innovate. Most of the clinical teams have traditionally looked through the lens that they’ve always looked through: “What are the traditional endpoints? What are the traditional measurement tools?”
They’re being sold, if you will, on the value of digital from their internal innovation groups. And so I think we’re going to see that innovation is going to start coming more from within the clinical teams as they get more and more comfortable with these novel methods for collecting and analyzing data.
It’s really important to be really clear on the problem that you solve. Understand the value of the problem that you solve. There’s nothing profound in that; that’s just general good business. But technology can have potentially a lot of different applications, and you’ve got to be really, really clear on what your application is solving for your customers.