Experts have seen the value of technology in the clinical trial space because it helps to remove natural bias that can leads to inaccuracies in data collection.
In part 2 of Pharmacy Times’ interview with Ivan Jarry, the CEO of ObvioHealth, Jarry explores the ways in which decentralized (digitized) clinical trials can improve the process of participant enrollment and management, data collection, and evaluation to potentially achieve accurate outcomes on a shorter timeline.
PT Staff: Now, this summer, you're working with Mi-Helper to conduct a study on migraine. This is just one example of a decentralized clinical trial that you conduct. Can you walk me through how the process works?
Ivan Jarry: Sure. So, as you know, a lot of people are suffering from migraine. So, we want it to be able to recruit a larger amount of population in a short amount of time. So, we are recruiting people online through advertising. And then people must fill in a few prescreening questionnaires to make sure that they qualify to participate in the study. And then the study has 2 phases; the first is to figure out what's the right dosage of that device—which is blowing air— in your nose. And so the first part of the study will be enrolling people [who], as soon as they have a migraine episode, will start using the device which will be shipped directly to their home with instructions. [They will] use the device at different levels, depending on the groups, and the first phase (in the first 24 hours) is just seeing the reaction of people who had just answered the questionnaire: “How long was your migraine? how intense was it?” And they answer this question every 2 hours. And that's going to be the first part of the study— to identify what's the intensity of the airflow that's going to be given to the participant.
Then once this will be entered, the second phase of the study is going to be randomizing participants some ways the actual device, some was a fake device, and people will use it. And we'll see the safety and efficacy profile of the device to help people with their migraine. So it's a short duration, because it's just on 24 hours, the device is very efficient and immediate alleviation of the migraine, and, and so the decentralized model allows us to recruit across the country, or ship the device very easily to the patient's home and people collected data in real time and get a conclusion and answers on the study very quickly.
PT Staff: When you conduct these studies, and you said It's in 2 parts, is that considered phase 1/2 like a normal centralized trial?
Ivan Jarry: It is the same. So the regulations, the rules, everything is the same. There's no specific guidelines or regulations for decentralized trial. We must follow the exact same rules as the traditional study.
PT Staff: That being said, as far as the patients go, is there a fear that the results could be the credibility of the results could be affected? Because you don't see them? And they're not there to be watched over?
Ivan Jarry: Sure. that was a question we had a lot at the beginning. A lot of sponsors— a few years back when they were looking at decentralized— were questioning the validity of the data that would be collected.
So I will go the other way around. What we've been trying to demonstrate is that the use of technology is that first, people can't rely on the time that they report events on an app. In the past you would give people a diary on paper, or they would go to a site, and you would ask them, “Okay. When did you use the product? At what time?” and they would tell it. [But] people would forget a lot of times to fill the diary. And so, you had what is called the parking lot effect. People will arrive at the site, they will open the diaries, and they will say, “Oh, I forgot to fill the last 3 days. Let me put something otherwis they're going to kick me out of the study.” So people will put whatever they could remember at that time. But if you're talking about migraine episode, it's impossible to remember 3 or 4 days ago. “How was your migraine? Was it very intense? how long did it last?” So, the information is actually—from a traditional study many times—very inaccurate in the format that it was being reported. And a lot of those episodes, like migraine pain or other, are very hard to recall. So, you need a tool so that—at the moment—you are able instantly to record the episode and the severity because you will not be able to remember in the future. And because there's a timestamp… if people forget to do a task, they can't go back and say they did it. We know—because there's a timestamp—that they did forget that task and so we can send them reminders to make sure that they're compliant. If they missed it, we know they missed it. In a traditional setting, we know that sometimes people would fill in those diaries after the facts.
PT Staff: Okay, thank you for explaining that. And when these trials happen, you're looking at safety/efficacy of a product or drug. Typically, the FDA must approve it but this takes a long time and you must collect data. Will going digital affect the timing and approval based on safety/efficacy outcomes?
Ivan Jarry: There's 2 places where we can try to speed up the timelines. [First] is upfront in the recruitment because we can tap into a larger pool of population, especially when it's fully decentralized, and it enables us to have a larger pool of population. The second is we can recruit faster. We had another study on migraine that we launched not long ago. On day 1, we had over 200 people that consented. In a site it's hard to do it because you have the limitation of the physical side and the scheduling of those patients. When you do it online, people can fill a questionnaire and you can have millions of people filling the same questionnaire on the same day. So there's a speed at the beginning because of the size of the population and how fast you can enroll people in parallel in a digital format.
The second place where we can shorten the time is at the end. Normally—cleaning up data— it means you go to the physical sites, collect the data, do your monitoring, and then do queries and resolve them. So, there's little back and forth. Thanks to the digital format, we can make sure that the data we just collected is clean. I do not have to take data from a [physical] diary into an electronic system, since the patient is already putting it there. Now the patient typed the data and we can give it a check. [Say] they told me yesterday that they were 160 pound and today they put down 265 pounds. I can say “Okay, they must have typed something wrong. They may have meant to say 165.” So, we able at that moment to double-check with the patients if the information they put is correct. In a normal paper diary, they will put in the numbers and then a month later, we'll have to call them back and say, “Hey, we saw that you put 265 pounds on that day, is that correct or not?” So all that cleaning up of the data—as it comes because of the editing/checking and seeing the data in real time— allows us to close the study a lot faster at the end, and close the database a lot faster.
PT Staff: Do you see this becoming more of the trajectory of clinical trials?
Ivan Jarry: Yes, there is a data point [but] it's hard to find data—there's research [being done though]. Let's say about 90% of the pharmaceutical executives who have been asked say that they will have some kind of decentralized trial, or digitalized component, in their clinical trial in the future. So, it's a massive trend. Some [digitized elements] will be the low-hanging fruit. As I said, consent is one of them. And you can see it in standard of care (SoC) if you have the Epic app or whatever. Now before you go to your normal doctor appointment, the day before, you receive paperwork. They’ll say, “Hey, do you want to fill the paperwork on your phone?” So that's a trend that's going to continue, as in people also understanding the limitation of the paper format because, again, parking lot effect— no timestamp. So that's a second wave of adoption. And I think that long-term is the integration with the electronic medical record (EMR). Today, it's very siloed. We have the data from a clinical study that is very different, and we don't look at the data that was collected over years on that participant in the EMR by the doctor. I think now, to have digital format, we're going to be able to connect and get more information and try to get a better understanding of why a drug is working on one patient and not the other. If we have access to their historical medical data, other genetics, and everything that patients already have in their profile, we may be able to understand better why drug is working and fine-tune the target population, for example.
PT Staff: Very well put. Thank you. I am going to open the floor to you. Is there anything else you want to add? Is there anything you want patients or pharmacists or healthcare professionals to know about the importance and significance of digitized clinical trials?
Ivan Jarry: The last piece I would like to mention is again that, in the DCT/digitalization, we talk a lot about the process, which is recruiting patient and engaging with patients using an app to get them retained across the trials. I talked a little bit about the connectivity with EMR. The last piece that we do a lot of research about is how you collect endpoints. Electronic patient reported outcome (ePRO) on paper can be very inaccurate because there's no timestamp, but even when you ask a question to people, the answer may not be correct because they are not experts in evaluating—they have a natural bias. So wherever you can remove and replace questionnaire by digital instruments, we see there's a lot of value.
I'll give you an example. We worked with a client that works with children with autism. And 1 of the measurements they were doing is taking the child— traveling with the child every month— to a site, where the physician would assess that child. One of the assessments was [measuring] the ability to speak intelligently by the child. So it would take 9, 10, 11 visits (or 9, 10, 11 months) to see progress from the child because the physician is only listening to this child for a couple hours and then it's 1 month until you listen again. So it's hard to see progress over time. We've worked with that client to use technology to record the child at home every day and send a recording to Siri, for example, to see how many words Siri was able to understand and transcribe. Then you have an accurate measurement daily of the number of words that child can say. And so the improvement we believe we will be able to see is a question of a few weeks versus a question of months.
So those are places that we think very interesting— where you can remove the bias or the variability of the current assessment that we're using and replace it with tools, digital tools, that are way more precise in collecting data points. And so, you can see differences a lot faster. And that's something we're very passionate about and we believe that part of that change—which is not just a flaw—is how you collect those data points.