April 24, 2019 | Clinical trials are contingent upon investigational product being available at clinical sites on schedule, every time. Investigational product is highly regulated, and IRT (Interactive Response Technology) offers vendors and sponsors a way to effectively monitor a drug’s passage from depot to shipment to receipt at clinical sites while also managing multiple aspects of the clinical trial process.

With 16 years of IRT vendor experience, Jon Paras has a firm philosophy: keep it simple and keep the focus on patients. Paras leads the IRT team at Amgen, ensuring that IRT within Amgen clinical trials is fit for purpose, matches protocol design, and supports the clinical trial business process. The primary goal, he says, is to ensure that patients are randomized to treatment correctly and the supply chain is managed appropriately.

On behalf of Clinical Research News, Lee Yuan spoke with Paras about his role in IRT at Amgen, his research, as well as his personal philosophy regarding IRT.

Editor’s note: Lee Yuan is a Conference Producer at Cambridge Healthtech Institute. Paras recently spoke at SCOPE about clinical supply management, focusing on IRT, and joined a panel examining whether IRT is the optimal repository for end-to-end supply chain data. Their conversation has been edited for length and clarity.

Clinical Research News: Jon, what can you tell us about supply chain and IRT?

Jon Paras: My last role prior to coming to Amgen was focused on building synergy between IRT and the clinical supply chain groups at Almac. There are immense synergetic opportunities between IRT and clinical supply chain. As larger sponsor companies often have separate groups for those respective functions, it’s challenging to gain alignment across both groups. To be fair, at Amgen, we have many supply chain functions in-house, from manufacturer-of-product to distribution; many sponsors outsource those functions to a CMO. When you outsource to multiple vendors, you potentially add more complications, more people in the game—you increase the risk of misalignment.

I was in that role for about a year and a half, and it gave me a true appreciation for just how much IRT and supply chain need to work together from the start—from when a protocol is being developed. IRT and supply chain need to work together to consider the plans for the drug. How is that drug being packaged? How the drug is being packaged will impact your IRT design. Will you have one label? Will you have multiple labels? Is it a global study? Will there be a staggered release under countries in a trial? What are your main supply depots? Do we have to consider import licenses and delays there? That’s just on the drug side, because all of that will impact IRT design. Once a trial is underway, there’s the management of your drug supply using actual data from your clinical trial to update the Supply Chain Manager’s forecast and ensure that you’re not going to stock out, but also to ensure that you don’t have too much drug wastage. There are so many intricate pieces of how IRT data, or trial data in general, can really impact supply chain.

Where are the pain points in this space today?

There is more scrutiny from regulatory agencies to ensure that outsourced partners of sponsor companies are held to the same standard that sponsors are held to. Specific to IRT, the system design must match the intent of the sponsor protocol. In the past, once a requirement document of an IRT system was approved by a sponsor, the sponsor took on the accountability that the design matched their protocol. Today, it’s more of a partnership, and in some cases a contractual agreement, where the vendors need to ensure that they’re not just assuming what a sponsor says is true. They also need to ensure that they are looking at the sponsor protocol and really doing that double check to ensure that the IRT design is matching what is intended by the protocol.

Of course, that’s a discussion between sponsor-vendor, but the point of this is, if a vendor is working on a clinical trial, they’re not just the vendor, they are the partner of a sponsor, and they need to treat every study as such, which I agree with. We like to call our vendors "partners" because they are playing a part in helping us ultimately get a product to market and serving patients.

What about data and tracking?

Let’s talk about drug accountability data, for example. If that data is in a system, there is going to be extra scrutiny for the vendor to ensure that data is accurate. What does that look like on the vendor side? What kind of checks are they putting in place to ensure that the data is 100% correct?

When we look at supply chain and IRT, I know that there are different considerations on how those two groups can interact. At the SCOPE conference, we had a really interesting panel discussion on supply chain data, and who oversees certain types of data. The two things that came up were, when you look at supply chain forecasting, and when you want to update your forecast with actual data from your trial, is IRT the best system to source that? Or, do you want to look at any other system, like EDC (Electronic Data Capture) or CTMS (Clinical Trial Management Systems)? It’s a great discussion. It’s probably different for a sponsor based on business process, systems, structure, etc. There’s probably no one-size-fits-all answer there.

IRT is more than just keeping track of drug. How can IRT bring more value to the clinical trial process?

Something that is near and dear to my heart here at Amgen, is trying to solve the problems our clinical sites deal with in the daily life of clinical trials. One problem that currently exists is true temperature tracking of product once it arrives at a clinical site. The caveat is that this applies when you have a product that is refrigerated or frozen. Typically product is released and, when needed, shipped to clinical sites. Typically, when you have temperature sensitive product going to clinical sites, the shipment containing the product will have a temperature monitor in the package. The temperature monitor will show if that shipment had a temperature excursion by the time the shipment reaches the site. When the site opens the package, they’ll look at the shipment, look at that temperature monitor, and if the temperature monitor shows there was an excursion, the site will then put the product in quarantine or on hold through the IRT system. From there, what typically happens, is a supply chain professional will look at stability data and see if that drug can still be used, or if they need to damage it out.

Let’s say product is good (not temperature compromised), and it is now at the site and available to be given to patients. But how do you know that that product has never been temperature compromised once it is at the site? In a perfect world, your product will remain in your refrigerator until it’s taken out to be given to a patient. Your refrigerator will probably have a temperature gauge to ensure that temperature doesn’t fluctuate, or there’s no change within simple parameters. But I think all of us who have refrigerators know that there are going to be some hot spots and cold spots. It’s not going to be consistent throughout the entire refrigerator, because depending on where your product may lie, it may be a hot spot or cold spot. So, does that impact the make-up of the product that you are trying to give to a patient?

If you look at more real-world scenarios, let’s say you have the product in the fridge and you have a patient coming in for a visit and the site coordinator takes out a box, or a vial of drug and starts to prep it. Let’s say it’s been sitting out for about five hours, and the patient just couldn’t come because there was a snowstorm, and they just couldn’t get to the site. That product is now placed back in the refrigerator. That’s five hours it was sitting out. Was there an excursion? Is the product still acceptable to give to a patient? The answer is we don’t know, because the IRT loses all temperature tracking on that product once it is at the site.

Another scenario is a site that has a pharmacy three floors away where all the product is kept. What if someone is retrieving the product and keeping it out of the fridge and at the clinical site to avoid going back to the pharmacy floor again? There are so many possibilities for the product to be temperature compromised in unintended situations.

I think that when we look at how things have operated, we assume that the sites are perfect and will follow process to a T. But we know that in clinical trials, not everything runs according to plan. We know that not everything is perfect, and these are probably real-life scenarios that need to be considered. I would challenge folks to think about, for that particular product that they’re trying to get to market, whether a temperature excursion could impact the make-up of the drug and impact efficacy on a patient, if consumed?

Have you found a solution?

Amgen had been working for years with different vendors looking for a solution to that problem. What we’ve come up with is looking at ‘kit-level’ temperature monitors instead of the typical shipment level temperature monitors.

When you look at the typical flow of drug, getting from depot to site, you’ll have a monitor in each of these boxes. But even before that, when you look at the process for getting the shipment packed, these monitors can be programmed with stability information for the product that you’re shipping. Then each monitor is affixed to the inner packaging of the product. Once that drug or that shipment gets to a site, an IRT mobile application can talk to those kit-level temperature monitors in the shipment via Bluetooth technology.

Typically, at the point you try to confirm a product shipment at a site, if the monitors are working correctly, either all will be "good", or all will be "bad". Now, if for whatever reason one of the kits, or one of the boxes in that shipment, has been temperature compromised, the mobile application will then automatically update the status of that product in IRT based on the temperature read. That’s helpful once you get to the site for many reasons. One, you’re eliminating any manual intervention. Essentially, if the drug is not acceptable based on stability, whatever is programmed into that monitor, it’s damaged. Then if there is a need for a resupply, we’ll send an automatic resupply. But once the drug is at the site, this is where we’re adding value, because of those situations that I described. Could the drug have been outside of the fridge for a couple of hours? Well, your site could build into their process doing temperature checks every, let’s say, three to four weeks. It’s as simple as going to your mobile app and picking your study and doing a scan of the product at your site. And if something is temperature compromised, the system will note the damage and resupply will be sent if needed. You can also do that check right before a patient is dosed. That’s something that you can build into your IRT mobile app functionality. That’s just ensuring that we are giving good product to patients. We do have this system in place on a current trial , and we hope to expand to more trials in the future.

How should we be approaching IRT design?

You’ll find that many IRT vendors will say that they can build anything and everything into their IRT, and that is a true statement because you really can build anything. You really can match an IRT system to the protocol to a T if you want to. But the main question that I ask when I look at different designs is, even though you can build everything, do you want to? If you build every bell and whistle into an IRT system—and protocol changes are inevitable—when you have a protocol change, is there an impact on your IRT system design? The more complex you make your IRT, the higher the likelihood of impact on your design. So, when you look at the problem or the problem statement that I have in my head for every trial, what it comes down to is, am I introducing risk to the patient? Our main goal is to ensure that our patients are getting randomized correctly, getting the right treatment, while also ensuring that we’re managing supply chain appropriately.

How do we want to approach IRT design? I’m taking my stance here for trying to keep things simple; let’s keep it basic while ensuring I’m considering our business process. I’m looking at every stakeholder in our company. I’m looking at the folks who run our clinical trials and their process. I’m looking at our system and what kind of data integrations, or data flow, do we have across systems. I’m taking all of that into consideration. With all those factors, I’m trying to figure out how do we build IRT design optimally, while ensuring that we’re not adding extra risk to a patient. If you can keep your design efficient, yet keep it effective, you can have time efficiency and cost efficiency. This is especially important as a clinical protocol evolves because it is a virtual certainty that there will be protocol amendments and each one could have impact on an IRT build! Time and cost efficiency are added bonuses, but the top-aligned, primary focus for me is looking at impact to patient.

Every sponsor company will have different needs based on their protocols, their business processes and their systems. From my perspective, if you can just keep the IRT "fit for purpose"—don’t overcomplicate, don’t add a data integration just because it’s new and exciting—do it because there is a real need, and that real need will add efficiency and/or effectiveness to your trial.