Download Premium Content

First Name
Last Name
Country (Citizenship)
Opt-In to Future Emails
Thank you!
Error - something went wrong!

eCOA Best Practices - Licensing and Translation

July 22, 2019


So you’re on the kick-off call, and you’re going global, and you’re like, oh I need translations. So it’s typically kick-off meeting or after. However, we have seen improvement in sponsors and study teams thinking about translations prior to kick-off. But licensing is still a major afterthought. Where we should be thinking about it is up at the very top, when we are scoping a project. We always want to note if there are any timeline issues to make sure that we are not committing to a timeline that’s not feasible. And I can tell you from experience, sponsors appreciate this feedback during the proposal stage.

So what do we have to look at when we’re licensing clinical outcome assessments? Typically, we allow four to six weeks. Some of these tasks can be done in parallel with the eCOA implementation process. But first and foremost, we’ll need to identify which COAs are subject to licensing and reach out to the authors and developers to get the license agreement. We’ll also need to perform the legal review, and more often than not there are several rounds of legal review to negotiate the changes or any revisions to the contract. Then we’ll need to execute the licensing agreements and obtain the copyrighted COAs and any existing translations. So yes, we’ll to obtain those copyrighted COAs, and until this process is complete, we should not be beginning any work on those licensed COAs, until this process is complete. So some developers might let you start some tasks, they might provide you a watermarked copy so you can at least take those to your eCOA design meetings and try to talk about implementation. Also, you might be able to start your translation with that watermarked copy, if you’re doing that full linguistic validation. But as a general rule, we should not start any activities of this sort until that license agreement is in place and has been executed.

So, as mentioned before, we’re looking at four to six weeks, but factors can delay this process. Lack of consensus over legal review changes. Like I mentioned before, there are typically several rounds of review to go back and forth between the developer if required, and then the eCOA vendor, as well as the sponsors.

Another major industry issue that we’re facing is developer responsiveness, or lack thereof. I know that there are several developers in the room, or we’ve all worked with one. It’s no secret, it’s not hidden. It can be challenging, and you do require time to work with developers. But I know from experience, a developer could go on vacation for a month and not have a backup, and get back to you when they get back from their one-month vacation and sorting through thousands of emails, and I wish I was exaggerating, but that is a true story. So definitely want to take into consideration, you know, there are some things that you just can’t prepare for, but I think once you work with developers you start to notice trends, and if you see one of their COAs is required in a study, you can add in a little buffer time there.

Moving on to translating COAs. Typical timeline, ten to twelve weeks. This is for the full linguistic validation process. And a lot of these tasks can be done in parallel with the eCOA implementation process. So first and foremost, we’ll need to perform that translation gap analysis and finding out which COAs have existing translations or which require translations from scratch. Then we’ll need to confirm that that contract is in place for the licensed COAs, so everything I just went over on that previous slide. We’ll want to make sure that we’re working with those final source files. So we don’t want to start any translations on draft files or ongoing. You might think that you’re getting ahead by starting, but it’s actually more costly and complex to update an ongoing translation, rather than just to wait for the final source file. We’ll need to determine the translation methodology as well as any developer requirements. So is this going to be a simple forward and back translation, or is this a PRO that’s supporting a primary endpoint that’s going to be used for a label claim. We’re going to need to account for linguistic validation and cognitive debriefing. Then we’re going to need to actually perform the translations in eCOA screenshot review. So this step right here can take longer than expected as well. Typically, we include three rounds of screenshot review just to make sure that everybody’s satisfied with the end screens and that they meet all requirements.

So again, we said that typical timeline is ten to twelve weeks, but factors can delay this process. Again, we’re talking about developer responsiveness for translation review, so if that same developer goes on vacation for another two weeks during screenshot review, they get back to you and they have a ton of edits and you still have a couple of rounds of screenshot review to do. Another factor that can delay the process are these rare diseases. You know, we’re seeing an influx in the clinical trials for rare diseases, but that does—or it could—pose a problem when you are looking at linguistic validation and cognitive debriefing if you have to recruit patients with the disease. You might have a limited patient pool to pull from. You might have limited resources and linguists with that expertise as well. So all things to keep in mind. And as you can see, translations and licensing isn’t just a checkbox that you can tick at the end of the day. You have to follow a process, and you have to allow time to follow the right process.


So, keeping in mind everything that we have gone over so far, I’m not going to have you break out into sessions or anything, but I do want to go over a few examples to see—going back to that original question: Can we indeed meet the FPI or FPS for these two examples.

So the first of which is a global Phase 3 CNS trial. We’re looking to collect seven licensed COAs, they are required at screening. We’re looking at ClinROs and ObsROs. They will be collected on site, so we will provision tablets. All of the COAs are supporting key primary and secondary endpoints, so rater training is also required. This study also required a cognitive battery kit. We’re going to seven countries, and we will require nine languages, and about 60 sites, and a little over 2000 screened patients. So if we start in November, do you think it’s feasible to meet an FPS five months later in April?

Well, if you think five months sounds like a long time, plenty of time for study startup. But I can tell you from experience, starting in November, even if we could get a work order contracted and executed, December has its own issues with the holidays and out-of-offices. So again, going back to the developers, if you see any licensed COAs, add in some buffer for the holiday season. There are also multiple components. But we can handle a lot of these at the same time, once we get those licenses in place. So we can go ahead and start looking at the design, we can start the training development, getting the battery kits sorted and including those scoring sheets on the tablets. But the key piece of missing information here is that the final protocol is not expected to be approved until after the new year. So we’re right back down to 12 weeks. We can still start a lot of these tasks, at risk, but we would not want to perform UAT until we have that final protocol. So we did meet it, but it was tight.

Second case study. This is an opportunity where we had to develop a patient reported outcome, and then that PRO was going to be used in an upcoming clinical Phase 2 study. So a little background. We were going to develop the PRO directly in the eCOA platform. It did require content validation with patient interviews. And we were expecting FDA feedback in March. The Phase 2 study was targeting an FPI for the end of June. It was a US-only study, but Spanish was required. And it was going to be collecting two additional PROs in the Phase 2, but everything was going to be collected electronically. So, can we meet FPI for the Phase 2?

Well, a few areas that we were able to work on the timeline for the first part is, we developed directly in the eCOA platform, so we didn’t have to worry about paper migration. The patient interviews, we only had to recruit healthy patients, and it wasn’t a rare disease, so we were actually able to submit to the FDA sooner than anticipated. But while we were waiting for the FDA feedback, we could go ahead and get started on some of the tasks for the Phase 2. We could look into those two additional PROs, find out if there’s any licensing requirements, if they have translations or if we need to start that. We can also look at getting some backup devices provisioned, or looking at the logistics for that and setting up the Spanish helpdesk. And then we could go ahead and start discussing the build and design for the other two PROs. So we received FDA feedback earlier than anticipated, or right on target actually. And then as soon as we did, what really helped us was having that screenshot document available from the PRO that we developed directly in the eCOA platform. So we were able to build in the FPI for the Phase 2.

I’m sure we could spend all day going over case examples, but I just want to provide you with a few key takeaways from this presentation. First and foremost, plan ahead. You never want to blindside a sponsor after a project has been scoped for, awarded, and with committed timelines. You don’t want to say, oh by the way, I can’t meet those. So letting them know ahead of time so they can account for that time required.

Raise awareness. eCOA and all related activities are on the critical path for study startup. You know, a lot of people think about just eCOA, but there’s a lot more that goes into it before we can even start the implementation of it. So, as mentioned before, it’s typically an afterthought, so the more we can raise awareness hopefully we can all get on the same page that we can get the required time.


And last but not least, don’t skimp or don’t take shortcuts just to meet timelines. So say you didn’t plan ahead, and you now have an eight-week timeline, a lot of people will try to find ways to shave off time to still meet that eight-week timeline even though they might be taking shortcuts and impacting the data. One of those areas is linguistic validation, so if that PRO is going to be supporting a primary endpoint or label claim, they should have allowed time for linguistic validation and cognitive debriefing, but they might cut that stuff out to save a month. But in the end, their label might not be approved. So right then and there, if you just allowed that time at the front, you have better success. Using past versions of licensed scales or not doing your due diligence at the licensing stage and just pulling something from the internet, we all know what kind of problems that can cause. And UAT—user acceptance testing. A lot of people don’t see a value in doing a detailed UAT but I think it’s imperative that we do test the platform, whether it’s an application or the device itself. We can hopefully catch any errors or any things that’s not showing up on the device right before they actually get it in the hands of the patient when we are collecting data.

So with all of that being said, I hope this has been informative. And that concludes my presentation.


That was really insightful. I know, having worked at ICON, not in your team but I’ve seen what your team have been doing in terms of the processes you’re driving around eCOA implementation, and it really is awesome stuff. So thanks for sharing a picture of that with us today. So thanks Ashley.

[END AT 11:40]


Previous Video
Enhancing Data Quality and Rater Reliability with eClinROs
Enhancing Data Quality and Rater Reliability with eClinROs

Dan DeBonis discusses the value of rater reliability with eClinROs - eCOA Forum.

Next Video
Making eCOA Accessible for All Populations
Making eCOA Accessible for All Populations