Freed: This is Steve Freed and we’re here at AACE 2018. We have a special guest with us who had a presentation and maybe we can start off with you telling us about yourself and the type of practice you are in.
O’Malley: Sure, so I am currently an endocrine fellow at Mount Sinai Hospital and I am going to stay on as faculty next year there where I’ll be doing a mix of clinical and research, focusing on artificial pancreas technology and diabetes in pregnancy.
Freed: You are also working on the artificial pancreas, is that correct?
Freed: I’ll probably ask you about that also. Let’s start off with you telling us a little bit about the study that you are presenting at AACE.
O’Malley: Sure, so I had a poster this morning and basically, we studied women with type 1 diabetes during pregnancy. We looked at how their insulin pump settings changed throughout pregnancy and we’re trying to get more information about the specific changes that happened during pregnancy – what times of day the really big changes happened – and we did this to get more information for, eventually, an artificial pancreas.
Freed: And why did you choose this population?
O’Malley: So, pregnant women are historically understudied and, specifically, in type 1 diabetes, there is a lot of expert opinion about their care but there is very little data. So, in the US, there is only one study looking at the details of insulin pump therapy in these patients and that was only with 9 patients. It is also a really important time for the patients themselves because hypoglycemia and hyperglycemia and likely, just variability, all are associated with poorer outcomes and so they not only have to improve their control to really strict goals, but they also have to do it while being pregnant. So, it’s a population that really needs more help from technology.
Freed: Using a pump and using a syringe vial, for a person who is pregnant, is it always recommended they use an insulin pump?
O’Malley: It hasn’t been very well established if it’s really that much improvement. Sensor use has definitely been shown to improve outcomes but the pump really helps specifically with what we were looking at which were the times of day. So, with a pump, you can get much better control of dawn phenomenon which can be exaggerated in women in pregnancy and so it has those benefits that with the injections, you just can’t do without having the risk of lows.
Freed: What were the findings of the study of your poster?
O’Malley: What we found was that the basal rates had a slight decrease in the early pregnancy which corresponds to when women are at highest risk of hypoglycemia and then around halfway through the pregnancy, they started to rise. We saw the biggest change in overnight settings. The most dramatic change was in the carbohydrate ratios which became less aggressive in early pregnancy but then had a steady increase to become much more aggressive. Those changed by about 50 percent.
Freed: What questions did your study arise?
O’Malley: So, there is still a lot more to know especially with the recent data like from the CONCEPT trial, which shows that glucose variability may be just as important as episodes of hypoglycemia or hyperglycemia. So, in populations where we try to limit our patients to people who have relatively good control going into their pregnancy, because we figured if some are very uncontrolled and then become motivated, some of those changes could be just optimization. However, even in those women who have an A1C of 6, we don’t know what all their variability was and how that could affect how then they need to have changes. So, the different tailoring – the changes to the specific woman depending on what their risks are and depending on what their pre-pregnancy BMI was – those are all things we have to learn.
Freed: What are the clinical findings of the study?
O’Malley: Well, I think that this kind of data that really puts more concrete numbers on what happens during pregnancy will help clinicians become more proactive in adjusting the settings. So, if you know what you think is going to happen in the next month, then that can affect what your recommendations are instead of only being reactive.
Freed: So, what can you actually tell people from the results of your study? So, you have someone who is pregnant come in, they are a type 1 diabetic, what did you learn that will help that type of patient?
O’Malley: Well, we learned we could predict that they may have the high risk of lows in the beginning of their pregnancy, which is why glucose sensors are going to be so important. But, looking at what their baseline control is – because we didn’t see that dramatic of a decrease in the early pregnancy – so really, tailoring that portion to the patient and then being comfortable enough to get a lot more aggressive with carbohydrate ratios as the pregnancy progresses.
Freed: Sometimes studies come out with results and they say, “Well we should have done this.” So, where did the results lead you to the next study that you would like to do?
O’Malley: So, we’re definitely going to try to get more prospective data. In this study, we worked with Mount Sinai and Mayo Clinic but we are going to expand that to even more sites because with limited numbers of patients, there is a bias of what your providers’ habits are and what they tend to do. We’re also looking at the CGM data to combine with the insulin pump data to get a more robust picture of the entire picture. And really just to get the same weeks, patients and more structured data to really look at the numbers in a more unified way.
Freed: Has the CGM made a big impact on pregnancy in diabetes?
O’Malley: I think definitely. Luckily, Mayo Clinic and Mount Sinai are good referral centers so a lot of patients came pre-pregnancy for planning and they were already on a CGM and tried to get as optimized as possible. But a lot of women only go on a CGM once they are already pregnant, especially with the newer sensors coming out that won’t have Tylenol implications, which for a woman who is pregnant, when they are limited to that, it can be really frustrating to not have that. And also, their limited sites, so a lot more study has to be in women to see what the real utility is of it how does this actually play out in them?
Freed: It also says that you’re participating in a large multiple trial for the artificial pancreas and it says it’s a first of its kind. What makes it the first of its kind?
O’Malley: I am involved in two trials with the in-control group, which is out of the University of Virginia and one is in a hybrid closed group trial that is using pumps. Then there’s another trial that is the first of its kind which tries to take the technology and the algorithms and all the benefits of artificial pancreas but apply it in a way that benefits people who are using multiple daily injections. So, it is a system where the algorithm can actually talk to the insulin pens and see how much insulin is on board so that they can then make suggestions for how to adjust doses, work with the CGM data to try to predict lows and highs, and give more of that information and more of those benefits to people who may not want to use a pump.
Freed: So, in your study, is it a true closed system or do you still have to use the carb ratios?
O’Malley: In that trial, it’s a hybrid system so they’re still using your own carbohydrate ratios. I think in the future, one of the things that I think from this poster is that since in pregnant women their biggest changes are in carbohydrate ratios and ideal artificial pancreas system in them, would have some sort of way to automatically adjust carbohydrate ratios.
Freed: And what’s the future for studies with the artificial pump?
O’Malley: So, what we hope to do with this data is have enough as well as collect more data on pregnant women to see what happens in pregnant women who are well controlled and do this in order to develop algorithms for an artificial pancreas in pregnancy. And I think this would not only be really helpful for pregnant women who have such strict controls and have such high risk of hypoglycemia, but also it would be a proof of concept that we can get to tight goals. If we can get pregnant women down to 90 or below 90 in the morning, then that means that it is good for the entire field and that, that’s possible. And then also having something that can actually change with predictions of what week in pregnancy – that will also benefit women who have menstrual cycles that need different adjustments – something that can actually adjust or at least predict changes in anyone.
Freed: Well, I want to thank you for your time, I thought it was very interesting and enjoy the rest of your stay here.
O’Malley: Thank you, I will.