Episode 17: Autoimmune Type 1 Diabetes: Challenge for our patients and clinicians

(upbeat music) - Hello, and welcome to Check Point Now, the podcast at the intersection of immunotherapy and toxicities. This is your host, Dr. Frane Sharif, endocrinologist, assistant professor of medicine, and an associate director at the Center of Cancer Immunotherapy at Duke Cancer Institute. I'm joined today by my wonderful co-host, Dr. Tian Zhang. Hi, everyone, I'm Dr. Tian Zhang, GU medical oncologist and associate professor at UT Southwestern Medical Center, Simmons Comprehensive Cancer Center. Before we begin today's discussion, just a reminder that the content discussed in this podcast is not a substitute for direct professional medical care and diagnosis. The opinions expressed here represent our own. Today, we have Dr. Ole Petr Hammondek and Dr. Le-Min, both expert endocrinologist from Brigham and Women's Hospital. This is a particularly special moment for me to welcome you, OP, since you were my supervising resident many years ago when I was a sub-interim. Dr. Hammondek is an endocrinologist at Brigham and Women's Hospital and an assistant professor of medicine at Harvard Medical School. He directs the Center for Onco Endocrinology, a joint clinical center between Brigham and Women's Hospital and Dana-Farber Cancer Institute. He is also an educator who directs the Endocrinology Fellowship Program at Brigham and Women's Hospital and serves as the education editor for the New England Journal of Medicine Group. Thanks for joining us, OP. - Thank you so much, Tiana, for having me on the podcast. I'm so glad that our paths crossed again. - Well, thanks, OP, for joining us. We also have today with us another expert endocrinologist, Dr. Min, who is joining us from the Division of Endocrinology, Diabetes and Hypertension at Brigham and Women's Hospital. He is the Associate Clinical Director for often neuroendocrine program and an assistant professor of medicine at Harvard Medical School. Dr. Min is a physician scientist. He has been doing patient care as well as medical translational research. His research interests focus on immune checkpoint inhibition related endocrinopathies, neuroendocrine disorders and neuroendocrine regulation of reproduction. He has served on the NCI investigational drugs during committee immunotherapy working group. Welcome, Ali, to the podcast today. - Thank you for having me. - Welcome to Checkpoint now, OP and Lee. We're both so delighted to have you join us today. OP, can you share with our audience about your professional journey into finding your niche with being an expert in managing endocrine disease and cancer patients? - As for my interest in the field, really started during my endocrinology fellowship. I had a kind of a transformative, I guess, patient encounter of a patient who came to see me in the thyroid clinic and had a history of multiple myeloma had been feeling kind of miserable during his multiple myeloma treatment. They've had to reduce the dose of his and any plastic agents. And he was just not great. And then finally someone sent a TSH. And the TSH was sky high. So he had profound hypothyroidism. And that really shouldn't have been a surprise. He was on Lenelidimide, which has a known side effect of hypothyroidism, but just no one had thought of kind of sending the TSH. So that got me sort of thinking that there is a role here for the endocrinologist in helping manage some of these endocrine side effects of cancer therapies. I gave a grand round on the topic. I wrote a review paper on the topic and then sort of started my career in oncoendocrinology that way. - That's wonderful to hear. Now, Le, how was your interest in this area ignited? - Yeah, similar to OP actually. I had in my first case immune checkpoint inhibiting induced endocrine toxicity. It was like a rare toxicity. The patient suffered from myeloma and she was started on epilimumab. You know, it's an anti-CTL for antibody. And it's developed grace of the mother's say was really bad. And we, you know, she was treated with steroids. And subsequently, you know, that was the medication was in a clinical trial and we did not know a lot about side effect. We see more and more patients than with endocrine toxicity initially like a hub of societies. And I did some study to look for the steroids roles in terms of improvement of checkpoint inhibiting induced hub of societies or, you know, and the tumor outcome. So we published a paper. We do not recommend high dose steroids which is routinely recommended as a standard treatment for immune checkpoint inhibiting induced toxicity. Subsequently, we look at thyroid toxicity and we were invited to write a review, publish endocrine review to have a comprehensive review on the immune checkpoint induced endocrine toxicity. So that brought my interest in this area and I'm still working on this area doing some research on the patient care. - Well, that's wonderfully. I've read many of those papers and that certainly shaped my interest as well. So wonderful, exciting work coming from you. Now let's get into the focus of our discussion here today which is autoimmune type one diabetes from immune checkpoint inhibitors. So Lee, how often do you see the side effect in patients? - So the short answer is not often. Actually, since this immune checkpoint inhibitor treatment we graduate this kind of patient. We kind of collected the patient in our database. We have about 20 patients so far. I'm pretty sure it's a little more than that but by study, the incidence is about 0.5%, which means if you have 1,000 patient exposed to this immune checkpoint inhibitor treatment only four patients will develop this toxicity in terms of type one diabetes. - That's great to hear that it is on the rare side. Is there a difference in the incidence based on which agent the patient is treated with? Is it more common like other adverse events with combination therapy with CTLA-4, with PD-1 inhibitors than with monotherapy? - Yeah, there's a study. They look at WHO database by individual port adverse effect. What the funding from that paper is 70% of this autoimmune diabetes are related to anti-PD-1 treatment and some of them relate to anti-PD-AR1 and the combination treatment actually, the incidence was not high. There's one possibility because combination treatment was introduced later patient initially just on monotherapy, that's my interpretation, but it's a very rare in patients who only receive the anti-CPI-4 treatment. - Interestingly, how do patients often presents with type one diabetes? - Yeah, so the typical presentation and I say typical a little bit here tongue-in-cheek because it is so rare that it becomes a little bit difficult to say what's typical and what's not. But I'll say that the patients that we have seen have typically presented very acutely to the point where we've had normal glucose values on a Friday and then admission in diabetic ketoacidosis on Sunday. So a very rapid onset of hyperglycemia with diabetic ketoacidosis that has been the classic presentation and sometimes, and most of the time I would say the patients have symptoms that go with the DKA. So things like, you know, polydipsia, polyuria, nausea, vomiting, orthostasis and so on. Now, if you go back to the literature, there are certainly cases that have been described of more insidious onset of hyperglycemia that really hasn't been the experience in our center. - That's really helpful, OP. And when would you recommend a hospital admission for patients? Is it only for patients who present with DKA or what other symptoms or signs are you looking for? - Yeah, so typically with these patients because they present in DKA almost invariably, they are admitted to the hospital, both for appropriate treatment for the DKA, but also to get all the training needed for the new medication in their life, which will be insulin. So we typically will admit them. Now, there are, you can certainly manage DKA if it's mild in the outpatient setting, but that's usually more in people who have established the A/B that's where you don't have the very high educational needs that a new, a patient with new diabetes would need. Now, if the patient is presenting in a more insidious manner, you don't necessarily have to admit them purely for a glucose number. So if they don't have diabetic ketoacidosis, you can certainly manage this as more of an outpatient. Of course, if the presentation is not the acute diabetic ketoacidosis, I think there's a differential diagnosis here of just simply a presentation of type two diabetes. And so you would probably want to evaluate more for those types of things if they have that somewhat more atypical presentation. - I couldn't agree more OP with what you're saying as far as the management is concerned. The patients I have managed as an outpatient were mostly in mild DKA or those who have had type two diabetes now have suddenly transitioned to type one diabetes or autoimmune, where you see a higher severity and grade of hyperglycemia in these patients. So these are patients who are highly knowledgeable about their disease to begin with and have some exposure to diabetes education and how to test their blood sugars and administer insulin. But an insulin naive patient who has a no preexisting history of diabetes is very challenging, even if it's in mild to manage this with available resources at such an acute setting as an outpatient. So I couldn't agree more. It should be decided case by case. Now Lee, we are often checking antibodies, autoimmune markers for specific for type one diabetes once patients are diagnosed with the side effect. In my practice and in prior studies we've seen clinically and biochemically confirmed cases of type one diabetes with negative antibodies. What do you see in your experience and what have you seen at Brigham? - Yeah, so we did a small study including 11 patients. We look for the antibody actually, the positive rate is about 50%. And it's pretty consistent with other literatures. So it's unlike classic type one diabetes, you're going to see like a 90% of the patient who will have a positive antibody. But this unique immune checkpoint induced diabetes see only about 50% with positive antibodies. So if patient with a negative antibody doesn't have this or mean they do not have autoimmune diabetes. - That's really helpful. And in that 50% of patients who don't have positive antibodies how would you confirm the diagnosis of autoimmune, immune mediated type one diabetes from our checkpoint inhibitors? - Yeah, so first of all, I like to know the history actually the most common hyperglycemia from oncology patient population is steels induced hyperglycemia. So you ask the question. And in the meantime, we check C-Peptide. If it's a high, it's unlikely autoimmune diabetes. If it's low, I like to repeat again because sometimes you have glucose toxicity if glucose levels are very high. So you like to repeat the later just to confirm. So C-Peptide is a very important lab we use to help with distinguishing type one and type two in this population. - Thanks so much. That's an excellent point. And I can't emphasize more what you just said that in the absence of antibodies that should not change the diagnosis and treatment of someone you suspect strongly suspect of having autoimmune type one diabetes. And C-Peptide and other supportive labs is stimulated C-Peptide when the sugar is elevated is actually very, very helpful in getting to that diagnosis. Now, since we were talking on the lines of treating mild cases of DKA or mild cases of hyperglycemia or patients with suspected autoimmune diabetes who are not severe like presenting in an acute setting. How should oncologists and other providers approach someone who's not in DKA? - Yeah, so certainly the sick patient who is in DKA moderate to severe DKA that patient needs admission. And there are very good algorithms for how to manage DKA in that setting with intravenous fluids with insulin and usually intravenous insulin and electrolyte monitoring and repletion. In patients where you have mild diabetic ketoacidosis or perhaps even just hyperglycemia without ketoacidosis you can manage it with some subcutaneous insulin and intravenous fluids. Again, not typically what we would do in someone who's brand new to this diagnosis because of all the that you have to tackle which is difficult to do at a patient clinic. And if you have an infusion abilities or in the emergency room. So most of these patients do end up getting admitted. Now, if you have a patient with hyperglycemia who you can sort of readily teach how to do injections I have had patients who are nurses and or who have experienced injecting low molecular weight heparins. You can manage this very mild or hyperglycemia without diabetic ketoacidosis in the clinic. It does require some coordination here between the oncologist and either an endocrinologist or a primary care clinician. And ideally also the diabetes educator. And that can be tricky to make all of that happen in sort of the oncology clinic setting. But this is sort of where it's helpful to sort of have a go to endocrinologist that is used to managing some of these things and can help guide the oncologist with getting treatment started. - Great points OP on that. And I think when we look at these patients it's less of a clinical dilemma on how to manage this and more of a logistic issue of how to coordinate all of this as an outpatient which is readily and easily available in a very short span when the patients get admitted to access and the logistics around getting a diabetes educator, injections, teaching them and getting all of this done and coordinated as from an outpatient perspective can be the most challenging part but managing less severe presentations of decay or even mild hyperglycemia. And maybe I can just interject here that one of the things that we see hyperglycemia not just with immune checkpoint inhibitors we see it with a variety of other anti-neoplastic agents. So we have set up now such that one of our diabetes educators is essentially on call for these sorts of cases and can see patients same day for diabetes teaching and initiation of insulin. And it's not just for the oncology patients this is sort of something that's available for acute hyperglycemia in general but has been mostly rolled out with the cancer patients for a variety of reasons. So that really is a key part of the team in managing diabetes. - Absolutely, that's wonderful to hear. I hope we can do something like this at Duke as well. Opie and Lee, I'm gonna open this question to both of you. My patients who develop type 1 diabetes as a consequence of cancer therapy are often very distressed about this diagnosis. I've collected anecdotal data when they come to clinic and see me with diabetes distress scores on them and I see a heightened level of stress around having to co-manage their cancer, the medications, the multiple injections and how to eat with their diabetes. Now, I'm sure you see this as well and how do you counsel such patients and what resources do you have at Dana-Farber and Drigen to support them? So I'll open it first to Opie. - Yeah, I think obviously any chronic disease is going to be a stressor but they'd be the same particular because of the impact that it has, maybe not every second of every day but certainly a large proportion of the day is spent in considering the impact of your activities, of your food intake, of your medications, on your glucose values. And then you have the added stress of maybe hypoglycemic events or symptoms of uncontrolled hyperglycemia. And that's challenging for anyone. Any patient with new type 1 diabetes will have that stressor. But most of the time this is the one chronic disease that they have to tackle and they don't have, you know, oftentimes metastatic cancer that they also need to consider and tackle and think about at the same time. Not to mention that occasionally there are impacts of the cancer treatment on the glycemic control, things like if they switched to a different therapy, maybe they need dexamethasone for their nausea management or, you know, there are other agents that can raise the blood sugar too. So it becomes kind of a full-time job for these patients to manage their diabetes while their cancer, perhaps they only have to come in every three weeks for their infusions. So we do have a psychosocial oncology service at Dana-Farber, which is extremely helpful in managing some of these challenges. And it'll also say that our certified diabetes educators can be really helpful, not just in kind of helping patients tackle the stressor of their new condition, but also in thinking about ways to maybe lessen the impact of the diabetes on their lifestyle. And that has been really helpful in many of my patients. - Well, thanks, OP, that's wonderful insight. And I love the resources that you have at Dana-Farber. Now Lee, what is your take on this? What do you see in your patients? - Yeah, so I agree with OP actually. I think the diabetic educator is really play a key role. I have quite a few patients. One part is, you know, patients want to focus on the treatment for their cancer because they say that's something to take their life away. Some of them, you know, even this determine, I don't care about my sugar and I just want to focus on my cancer treatment. So I think the important part is a little patient aware about, you know, the management is important to, not only to improve the glycemic control, it is also important to improve the tumor response and the cancer outcome. So I typically, this kind of patient, I will connect with our next practitioner who specialized through this diabetes and to discuss about, for this type one, discuss about insulin pump and also continuous glucose monitoring. I think eventually when they get used to this, they're happy about this, about management. - Thanks both OP and Le. You know, I think there's so many support mechanisms which is great that you have that type of support for your patients. And hopefully this will allow others to learn from your program as well. Le, do you know, and do we as a field know why patients get autoimmune type one diabetes from these immune checkpoint inhibitors? Is it through the TESO activation that we're listening or what's the mechanism there? - Yeah, so that's a good question. We do not have a lot of data to help us to identify, you know, what is the mechanism on the line? I think number one, it's similar, but clearly distinguished from classic type one diabetes because even look at antibody positive rate, I mentioned earlier, in classic autoimmune type one diabetes, the positive antibody rate is more than 90% but in this kind of population, checkpoint inhibitor induced type one diabetes is only 50%. And I could unset all these diabetes in checkpoint inhibitors, you know, most of them are kind of, this is a unique feature. In terms of a mechanism, you know, it has a strong link to anti-PD-1 and this study show beta cell express PD-L1. So when you have anti-PD-1 to block PD-L1 and you might increase the toxicity. So, but, you know, such a low incidence rate is only 0.5%. It's not explain everything that's from my view. And also the other part is, if you look at pancreatitis and autoimmune diabetes, we did a study looking for that actually, they're about 20% overlap, but it's not everyone with type one diabetes are going to develop pancreatitis or versus a patient develop pancreatitis, autoimmune, I mean, like a checkpoint inhibitor induced the pancreatitis, they do not develop on type one diabetes. So the mechanism I can tell you is not, we don't know yet. So we need to do more to find it out. - So Lee, that gets me to asking one more question. I think that was a great insight into the mechanisms and what we know and what we don't know about type one diabetes, but do you see any therapies for conventional type one diabetes emerging from the experience that we've learned from these drug-induced acutely and abruptly developing type one diabetes patients? - Yeah, so I would say it's not really clear. I guess initially people talk about use high dose steroids as an anti-inflammatory treatment to help to, you know, to inhibit inflammation and hopefully to reverse this process. But, you know, steroids itself cause hyperglycemia and I don't see any data in the literature to say it's working. And they are, people use anti-TNF used for other toxicity like colitis or, you know, emulitis, but I do not see any literature data myself to say very useful this type one diabetes. I don't see them, but that's a potential to look for. But on the other hand, you always have to keep in mind emul checkpoint inhibitor is to induce auto-immunity and increase immune response against the cancer. If you use something to inhibit your immune system and what about outcome of tumor response? You might also inhibit the tumor response. So we have to weigh this balance. We're going to take a tumor response and improve the side effect. Thanks, Lad, that's really helpful. OP, when Afrin mentioned an uncle endocrine program that you direct at Dana-Farber, I was curious to learn about it. Can you share insights about your program? Yeah, absolutely. This program came to be in 2015 when I went from being a fellow to going on faculty and was sort of where I had my passion based on my experiences during fellowship. So we basically see patients who are either in the midst of their cancer treatment or who are many, many years after their cancer treatment, so long-term survivors with any endocrine complaint that they might have. And it consists of me as the endocrinologist and we're housed in the adult survivorship program at Dana-Farber, so we also have all the services that are available within that program. We end up working very closely with the clinicians at Brigham for a variety of reasons. One of them is that we have many very subspecialized programs at the Brigham. For example, Lee is very involved in the neuroendocrine program and we have amazing neurosurgeons. So for patients that are looking for neurosurgery for a pituitary lesion, for example, I would send them to see Lee and his colleagues rather than see them myself. So it's a very collaborative program between Dana-Farber and the Brigham and has really served the patients well, but I think also the oncologists, it's sort of become a, I think the oncology and chronology program has become a resource for the oncologists to reach out to when they have endocrine issues or concerns or questions relating to their patients, also relating to their research. So I've helped a lot of the oncologists in various endocrine aspects of their studies, which has been very satisfying. - Well, that's amazing to hear, OP. Now, in your practice, what is the most common diagnosis that you see? - Yeah, so when I first started, there actually had been a survivorship program for endocrinology for long-term survivors of hematologic cancers. And that was primarily Hodgkins. So a big chunk of my patients have hypothyroidism or thyroid nodules due to mental radiation for their Hodgkins disease. But that was sort of a legacy population that I took over. But nowadays, I would say that the majority of my patients are the patients with osteoporosis related to the use of aromatase inhibitor. And that's in part because that's who I... That's sort of who I... My colleague in the survivorship program is a breast oncologist. So she has been very good at promoting the program in the breast oncology group. There was a while where a period where the most common diagnosis was immune checkpoint inhibitor induced thyroiditis. But between the oncologists just seeing a lot more patients and gaining more familiarity with the management of this condition themselves. And also we developed a clinical algorithm for the management of thyroid function abnormalities in patients receiving immune checkpoint inhibitors. Most of those patients never see me now so that the number there has really dropped. I usually only see the more complicated cases of thyroiditis at this point. - That sounds amazing, the infrastructure that you've put into place. How does the referral process work and how do oncologists like myself reach you for a consultation for their patients? - We certainly have a formal referral in our electronic health record, which goes to my scheduler. But typically I've reviewed these referrals to make sure that they are more appropriate for me as opposed to one of the specialty programs that I alluded to at the Brigham. But probably the majority of referrals start off as an email from the oncologist to me asking for some advice on a patient. And it allows me to give some short-term advice that the oncologist can start so they can start to send some labs or get a bone density scan and sort of get the ball rolling. And then I can schedule the patient into my clinic at the next available appointment. - Well, that's an amazing infrastructure OP. And that was some great insight into your very unique program. And I wanna thank you on air for guiding me in building something similar at Duke to serve our patients here locally. I'm excited to work with our patients and oncologists here and basically derive what I can derive from your experience that would be so valuable in shaping ours. Thanks so much, OP. - You're welcome, I'm glad to be of help. - OP, any parting thoughts you would like to share with our audience? - Well, I guess I would just say that the management of any medically complex patient is going to be a teamwork experience. And that is true for the patient with cancer also. And more and more, I think it's becoming clear that the endocrinologist should be part of that team as we start to see more endocrine side effects of some of the cancer therapies and not to mention incidental endocrine findings on imaging or lab tests that are sent in these patients. So I think the management of the oncology patient should involve a multidisciplinary approach and don't forget about the endocrinologist. - Oh yeah, definitely don't forget the endocrinologist, right? (laughs) So Ali, any parting thoughts for our audience from your side? - Yeah, topical with OP, it's a multi-discipline management. We need to work with oncologists, endocrinologists. Actually, recently we have a project of work with oncologists on glycemic control. So initially actually it was a little issue rise from oncology infusion center. Every patient when they're going to have the infusion they do BMP, chem seven. And sometimes patient with diabetes, they can have a sugar, like a broad sugar, like a more than 300. And then what should they do? So oncologists scratch their head and they want us to help. And we kind of work with them and we cannot generate the algorithm for this acute, with so-called acute hyperglycemia or current just infusions. And the patient can't wait for the infusion, they go home, everything fine, they go home on the same day. So what should they do about this hyperglycemia? So we generate the algorithm. In the meantime, we also work with oncologists to address, you mentioned earlier, the patient diabetes management with co-existent malignancy. We try to improve the quality of life by providing better guidance and better device to help them with the beta-glycemic control. - Well, that's amazing. I think the bottom line that I'm hearing from both UNOP is that it takes a village to manage patients and co-manage patients. And my favorite specialists to work with and colleagues to work with are our own colleges. So I think we all make a great team. And it sounds like at Brigham and Dana, you all make a great team as well. So thank you, OP and Lee for joining me and Tian today for our podcast checkpoint now. I am especially delighted since today I'm not the only endocrinologist here and it was very encouraging to hear from fellow endocrinologists who share the same interests. Thanks again for joining. - Thank you for having us. - Thank you, I enjoyed it. - Awesome. Thanks again, La and OP for our audience. Please remember to tune in again in two weeks. You can reach us at checkpoint now [email protected] and please remember to follow us on Twitter at Checkpoint Now MD. (upbeat music) (upbeat music) (upbeat music) (upbeat music) (upbeat music)