Large-scale population screening of pre-school children in Bavaria, Germany associated with reductions in DKA
The Fr1da trial, first launched in 2015, screened 90,632 children between two and five years old in Bavaria, Germany. Very notably, this is the first-ever study to screen this population of pre-school aged children for type 1 diabetes. This widespread screening at pediatric primary care centers revealed pre-symptomatic type 1 diabetes in 280 children (0.31% of study population). Of these children, 25% developed type 1. Very positively, the prevalence of DKA in the diagnosed population was <5% -- compared to 20% in Germany and 60% (!) in the United States. While direct causation between the screening protocol and these lower rates of DKA cannot be proven with this study (a randomized clinical trial would be needed), these results are extremely promising in terms of indicating the utility of such early and widespread screening practices. Early DKA events have been associated with poorer long-term A1c and cognitive outcomes, so preventing these events in young children with type 1 can be key in improving long-term health and reducing associated healthcare costs.
In an interview with our team, JDRF’s Senior Director of Research Dr. Jessica Dunne pinpointed (i) cost-effectiveness; (ii) implementation into existing healthcare systems; (iii) parent distress; and (iv) the risk of false positives as the biggest barriers to more widespread adoption of screening practices, with the first two being of the greatest priority. Previous research from Dr. Arleta Rewers has shown that diagnosis of type 1 in DKA has been linked to higher A1c levels even 15 years post-diagnosis, and while improvements in glucose have been associated with better outcomes in terms of complications, Dr. Dunne expressed that it still requires a “bit of a leap of faith” to link reductions in DKA to better long-term outcomes. As such, whether or not reduction of DKA will be enough to make a general population screening program cost-effective has yet to be determined. On healthcare systems, Dr. Dunne discussed how these types of screenings will likely be performed in the primary care setting, adding another layer of complexity given the already-existing time restraints for primary care physicians. Even past an initial screening, Dr. Dunne explained that time for education and follow-up, if an individual were to test positive, would also need to be allotted. Read on below for our full interview transcript with more details!
Interestingly, Dr. Dunne noted that ongoing JDRF and Helmsley Charitable Trust research on parent and physician preferences for screenings will help to understand the relative importance of study outcomes (i.e. avoidance of DKA, an intervention, etc.) versus how the test was done (finger prick, blood testing, location of test, cost, etc.). To this end, Dr. Dunne added, “Why I bring this up is because we spent a lot of time thinking that how kids got screened would be just as important as what the outcome is, but the outcomes may be the important thing. And while there may be challenges based on the age of children, we think those are really surmountable with respect to the outcomes we can offer
In terms of next steps, the study will continue as the “Fr1da-plus” trial and add nine and ten-year-old children to the study population. A cost-benefit analysis of the screenings will also be performed – we imagine that this analysis might show that the massive reductions in DKA, along with earlier interventions, could provide benefits that outweigh the costs associated with large-scale population screening. JDRF and Fr1da investigators will also use the data generated in this study to inform future screening protocols in type 1 diabetes. We’re eager to see how scalable these screening efforts may be in other parts of the world.
Interview with Dr. Jessica Dunne (Senior Director, JDRF Research)
Martin Kurian: What impact do you see this work having on the type 1 community? What do you hope the main takeaways will be for those involved in the field for patients, providers, and broader health systems?
Dr. Dunne: I think I’ll start off a little bit ahead of that, is that we’ve been supporting the Fr1da study since its inception in 2015. At JDRF, we’ve taken a very strong interest in promoting general population-based screening studies, so we’re supporting actually quite a few of these. The other one that is most similar to the Fr1da study in Germany is the ASK study in Denver, and there are some other iterations of this. There are a few reasons why we’re keenly interested in developing programs to transition these general population studies from clinical research to public health. In the absence of an approved therapy for delaying progression of T1D, the take home messages of this paper was, we can prevent DKA at diagnosis through careful screening and monitoring in the general population. In Europe, the numbers are a little bit better than they are here in the US. In the US, the latest numbers suggest that 58% of kids are diagnosed in DKA, while the numbers are a little bit lower in Europe.
What the Fr1da study and other ongoing studies have shown is that we can largely prevent DKA at diagnosis through screening and monitoring. This is important for a number of reasons, not the least of which is that there was this paper about three years ago from Dr. Arleta Rewers in Denver showing that diagnosis in DKA vs. not in DKA had a significant impact on A1c levels 15 years later. We can sort of make the conclusion in our heads that if we can reduce DKA, we can improve glucose control, and it might lead to better outcomes in terms of complications. Obviously, that study doesn’t exist, so it’s a little bit of a leap of faith, but we all believe that better glucose control is good for better long-term outcomes.
The other important piece that has come to light recently is that avoidance of DKA has improvements in cognitive functioning, both in the short term and long term, which again is a really important outcome – to make sure that we avoid DKA to manage cognitive function in children, so we’re really thrilled about the work that Dr. Anette-Gabriele Ziegler has been able to do in Germany. We’re in conversations with her about expanding the study to other ages and other populations. This first initial study was for screening at one point in two and five-years-olds, now we’re expanding the study to do a secondary screen for those children who were negative in the initial screen at nine to ten, because we know children also develop autoantibodies and autoimmunity later in life.
To address your question on what impact, it was shown for the first time in the general population through screening and monitoring, we can reduce DKA diagnosis, which can show that screening is feasible in the general population.
The other main takeaway for those involved in the field is that this is really an important step as we transition studies from clinical research to public health.
Rhea Teng: What are the biggest barriers to widespread adoption of these screening practices? How well can this be scaled? How do you envision such screening efforts in the US being carried out?
Dr. Dunne: All very good questions and very timely because I just got off a call about half an hour ago discussing this very thing. I think one of the biggest challenges is the cost-effectiveness of implementing some of these programs, and whether or not avoidance of DKA will be enough to make a general population screening program cost-effective, and making sure that as we are focused on avoidance of DKA, we’re also focused on disease-modifying therapies to delay or stop the progression to type 1 in this population.
When we think about intervention in children that are two and five years old, interventions obviously have to be not only efficacious but also safe, and so this is something we think [can be done], but it’s also a challenge in the space.
The other things they touch on in the paper that I think are important are the psychological impact of a screening test, and how it raises some levels of distress in the parent as well. They make the point at the end to make sure that there is careful education of the parent of what the real risks are. Oher studies like TEDDY are showing similar effects. The education and really understanding what the true risk are can mitigate some of the negative outcomes of screening, and what I mean by negative outcomes is the increase in stress and anxiety in parents, and obviously the kids are a little bit too young to understand that.
One of the other things we talked about too is the idea of false positivity – so children who may seem positive for autoimmunity but may never progress to true type 1. Based on some of the early data from Dr. Ziegler and from others, showing that the lifetime risk of developing type 1 once you have multiple autoantibodies approaches 100%, but we don’t actually have that lifetime data, so there is a possibility that you could get some false positives, either through the antibody test itself of because these kids might just never progress. I think that’s a relatively low risk here.
I think the second part in terms of how do we transition it to public health care, we spent a lot of time earlier talking about US healthcare systems and how the screening would likely be done in primary care settings. Knowing how limited the time primary care physicians have with patients is, where the education and follow-up then happens if somebody tests positive, are challenges as we think about transitioning it to a public healthcare. I think cost and also implementation into our healthcare system are probably the number one challenges, and we are actively trying to figure out what risk mitigation strategies are available to do in those areas.
Rhea Teng: What surprised you most with the findings of the study?
Dr. Dunne: I have to say that I’m not surprised by anything that was in the study. As I said, we’ve been working with Dr. Ziegler for about five years, in designing and executing this study, so a lot of it was based on prior data in some of the high-risk kids, modeled on studies like TEDDY, where they screened high risk genetic kids, or TrialNet, where you’re screening for family members. What we didn’t really know was what type 1 diabetes in the general population was like. Compared to some of the previous studies in some of the high-risk populations, despite some small differences, what we’re learning is that it’s largely similar. Again, I think nothing super surprising, but just further highlights what we have to address in terms of cost and psychological impact for implementation.
Martin Kurian: What are the next steps for these researchers and JDRF? You mentioned the Fr1da-Plus Study and the cost-effectiveness analysis that the whole team will be working on in the future, but more broadly, what do you hope this research spurs in terms of progress over the next five years? And over the next decade as well?
Dr. Dunne: It’s always hard to put time frames on where this is going to lead. I think again, ultimately, the goal with screening studies is to move these out of clinical research into the public sector. I already touched on some of the things that will be necessary to address to drive those there.
I think a large piece of it is now that we have a large body of preliminary data showing some feasibility, to replicate this in other areas of the world as well. I already talked about the Denver study and understanding how different healthcare systems would be able to implement these types of studies, and it would need to be different in different parts of the world. That’s something we need to get a better handle on. Obviously, in the US, it’s a very complicated healthcare system, and so some place in Europe like Germany might be a more attractive place to really begin to have those conversations. We’re not abandoning the US at any level there though.
In terms of next steps, we already talked about Fr1da-Plus and extending the age range for screening, and it’s really just continuing to build the data needed to be able to transition this to clinical care, and so we had some initial discussion with Dr. Ziegler about expanding the network to other parts of Germany and other parts of Europe. I think, without putting a time frame on it, in terms of what the research and progress over the next few years is, however you want to interpret few, is building the case of how to implement this into clinical care.
There’s a lot of different pieces of the puzzle that fall into some of those concerns in terms of cost, understanding the psychology, understanding the role of both parents and physicians, being able to look at some of the data to understand when the optimal age is for screening to get the highest positive predictive value and highest sensitivity, so those are some of the activities we are considering as we move these programs forward.
Martin Kurian: Were there any unique challenges with doing this screening in preschool age children vs. school-aged children?
Dr. Dunne: You know, and I think that’s a difficult question for me to answer, I think a lot of this has to do with the local healthcare system, the parents’ willingness to screen, and also the pediatricians’ willingness to screen. Certainly, parents have large control over their children, but as children get older, something that may appear optional may be a little more difficult to do in older children.
When we talk about how the test is done, it’s currently a blood test, and when you talk about doing blood tests in young children vs. school-aged children, there’s obviously a little bit of a different dynamic there.
One of the things we are working on here at JDRF is trying to reduce the amount of blood volume that would be needed for these types of antibody screening tests, and so just a simple finger prick with a small drop of blood to be able to do this, is something we think that would be acceptable for all parts of the population.
The other thing that we are also working on that’s relevant to this are both the parent and physician preference studies for screening. We talked about two major things in this reference study, it was both (i) How the test was done: is it a finger prick, blood draw, urinary sample, salivary sample? And I just want to make it clear that the latter two are not yet available – it’s all blood testing right now. And how much it costs and whether it can be performed at a doctor’s office or needs to be done somewhere else? And the other attribute we studied was (ii) What was the outcome: was it avoidance of DKA, was it an intervention? Why I bring this up is because we spent a lot of time thinking that how kids got screened would be just as important as what the outcome is, but the outcomes may be the important thing. And while there may be challenges based on the age of children, we think those are really surmountable with respect to the outcomes we can offer.
Rhea Teng: Are there any other questions you wished we had asked?
Dr. Dunne: The other point Dr. Ziegler also made in the publication about the availability of intervention trials. I think, while prevention of DKA is important, we want to make sure that we are developing safe and efficacious therapies to offer to this population to delay or stop the progression of type 1, and obviously, that’s a challenge.
We addressed a lot of things about how to advance this into clinical care. More engagement with health authorities would always be welcome, and I hope this publication spurs some interest in that area.
--by Martin Kurian, Rhea Teng, and Kelly Close