Immune Checkpoint Inhibitors Cause of Type 1 Diabetes?

This transcript has been edited for clarity. 

As many of us know, there is a therapy for cancer, known as immune checkpoint inhibitors, that in addition to treating cancer can also cause new-onset type 1 diabetes. I've always been interested in this because I don't really understand what's happening, and also because patients who develop this form of type 1 diabetes really have type 1 diabetes.

There's no honeymoon period. They're often in diabetic ketoacidosis (DKA), and they need to be treated from the get-go as people who are completely insulin deficient. There have been a number of recent case series discussing this type of diabetes, but there's still much that we don't know. 

First, immune checkpoint inhibitors are different from conventional cancer treatments. I'm not an expert on this, but I do know they function by releasing the brakes on the immune system, which then allows the immune system to specifically target and combat tumor cells, thus helping treat the cancer.

There are a number of different agents, and different agents cause different kinds of side effects. The endocrine side effects include a pituitary disruption, autoimmune thyroid disease, adrenal disease, type 1 diabetes, and hypoparathyroidism. Again, it depends on the type of immune therapy as to the most common endocrine disorder you will see.

Type 1 diabetes most commonly occurs with anti-PD-1 or anti–PD-L1 antibody monotherapy. Within a few months to a year or two after these therapies are initiated, patients will develop fulminant, new-onset type 1 diabetes. This often presents as DKA, probably in about half of the cases. Again, it's because they're so insulin deficient. 

Immune checkpoint inhibitor–induced type 1 diabetes occurs at varying rates, ranging from 0.2% up to 8% of people treated with these agents. I think what's important is to be aware that it can happen, and to monitor patients for signs and symptoms of DKA.

There's nothing we can do to predict it, so measuring autoantibodies in advance or obtaining a family history doesn't help. I think obtaining a baseline fasting glucose level and A1c can be useful to follow patients and also to rule out preexisting diabetes that wasn't otherwise diagnosed. We do think there are specific HLA alleles associated with an increased risk, but that's not commonly measured in our patients before they begin these therapies.

Within a few weeks to a few months to a year or two after starting an immune checkpoint inhibitor, patients may develop diabetes. Oddly, many people do not have positive islet autoantibodies who develop this kind of type 1 diabetes. They do have very low levels of insulin and C peptide, as though they have overwhelming beta-cell failure. 

They often also don't have a very high A1c level at baseline because they haven't had diabetes for a while. I think in many of these cases, it really is a very sudden-onset, very severe form of type 1 diabetes, and they haven't been exposed to hyperglycemia for a long period of time.

To make matters slightly more complicated, there are people on these agents who develop a somewhat slower-evolving form of diabetes that looks a little bit more like type 2 diabetes. I'm really talking about this more severe form of what seems to be sudden, new-onset type 1 diabetes that requires insulin therapy for life.

I try to make people's lives as simple as possible, particularly when they're dealing with cancer. I'll start people on a multiple daily insulin injection regimen, but often it will end up being easier for them to manage their diabetes if they're on an automated insulin delivery system.

I tend to put them on a system with a pod, like the Omnipod 5 system, just because that's simpler for many than dealing with some of the more complicated systems. I would review the pump systems with your patients to see if there's one they feel that they could live with and handle. I put everybody on continuous glucose monitoring, and I think it goes without saying that it makes diabetes management easier. 

I do treat this in the same way I treat type 1 diabetes in general, except it really isn't LADA, or latent autoimmune diabetes in the adult. It's adult-onset type 1, but adult-onset type 1 without much residual insulin secretion that needs to be treated very intensely from the beginning.

Again, quality of life is very important. I spend a large amount of time working with patients and trying to help them achieve the kind of control that's reasonable, as well as handling this new disease that they got on top of receiving cancer treatment. 

As one very bright 13-year-old asked me, “Don't you think people would rather have type 1 diabetes than cancer?” I said, “Yes, probably it's best to treat the cancer.” 

I think these are great drugs for cancer, but these are agents we need to be aware of in terms of their ability to cause diabetes. 

This has been Dr Anne Peters for Medscape. Thank you.