Findings may help predict risk for immune checkpoint inhibitor-induced diabetes
Click Here to Manage Email Alerts
Clinical and genetic risk factors can help identify individuals at risk for immune checkpoint inhibitor-induced diabetes, according to results of a cohort study.
Researchers at an academic integrated health care system assessed data from 14,328 adults (45.9% women; median age, 66 years; range, 8-106) treated with immune checkpoint inhibitors (ICIs) between July 2010 and January 2022.
Sixty-four (0.45%) developed ICI-induced diabetes, equating to an incidence of 124.8 per 100,000 person-years.
Investigators identified preexisting type 2 diabetes (OR = 5.91; 95% CI, 3.34-10.45) and treatment with combination ICI (OR = 2.57; 95% CI, 1.44-4.59) as significant clinical risk factors of ICI-induced diabetes.
A logistic regression model revealed that an increase of 1 standard deviation in type 1 diabetes polygenic score resulted in an OR of 2.6 (95% CI, 1.7-4) for ICI-induced diabetes.
“ICI-induced diabetes is a serious adverse event because it is associated with life-threatening complications and it has a significant impact on quality of life,” Michelle Rengarajan, MD, PhD, instructor in medicine at Harvard Medical School and endocrinologist at Massachusetts General Hospital, told Healio. “Being better able to predict who is at risk and more closely monitor these patients, or treat them earlier, is a very important step.”
Healio spoke with Rengarajan about the findings, the benefits of identifying patients at risk for ICI-induced diabetes and the need for larger studies on this topic.
Healio: Prior to your study, what had been known about ICI-induced diabetes?
Rengarajan: There has been a reasonable amount of work done on this. ICI diabetes resembles type 1 diabetes — it’s a presumed loss of insulin-producing beta cells. It can be associated with some markers of beta cell autoimmunity that we see in type 1 diabetes and is a serious complication of ICI therapies. In the earliest days of ICI therapies, these patients often were presenting to the hospital with diabetic ketoacidosis, which was often the way people were diagnosed with ICI-induced diabetes.
Like those with type 1 diabetes, these patients are on insulin for the rest of their lives. They take multiple injections a day and need insulin anytime they have a meal. This is a diagnosis that requires a lot from the patient.
Healio: What motivated you to conduct this study?
Rengarajan: We wanted to be able to better understand and predict risk for ICI-induced diabetes. Additionally, as checkpoint inhibitors are being used [in more treating settings] and more broadly across tumors, there often are cases where decision-making has become more individualized. If someone was at especially high risk for type 1 diabetes in an adjuvant or neoadjuvant setting, this might influence decisions about whether or how checkpoint inhibitors should be administered in an effort to preserve quality of life.
Healio: How did you conduct the study?
Rengarajan: We examined a database of about 14,000 patients who were treated with checkpoint inhibitors in our hospital system, and used a number of different automatic filters to decrease to a smaller cohort of patients we thought were likely to have ICI-induced diabetes. Then we manually screened these approximately 200 patients. That helped us define the incidence and risk for development of ICI-induced diabetes, which we were able to confirm is quite rare. Then we examined particular features of those patients to see if we were able to predict risk factors for ICI-induced diabetes and could identify phenotypic subtypes of it.
Healio: What did you find?
Rengarajan: A key clinical risk factor was preexisting insulin resistance. This was seen in patients with prediabetes or well-controlled type 2 diabetes — these patients were at much higher risk for developing a new insulin requirement with ICI-induced diabetes.
Another risk factor we found, which has been true across a number of immune-related adverse events, was treatment with combination therapy — both a CTLA-4 inhibitor and a PD-1 or PD-L1 inhibitor. Taking out these critical T-cell checkpoints can lead to a substantial increase in autoimmune side effects. Genetically, we found that individuals with a very high polygenic risk score for type 1 diabetes were at higher risk for developing ICI diabetes.
This raises two interesting points. First, having a genetic predictor suggests that we might be better able to identify patients ahead of time who are at high risk for ICI diabetes. Second, these patients who were at high risk for type 1 diabetes required ICI treatment to develop diabetes. This raises an interesting question about why these patients didn’t get spontaneous type 1 diabetes before, and whether checkpoint inhibitors are somehow accelerating that process.
Healio: What are your next steps in research?
Rengarajan: To really move this work forward, we would need much larger cohorts of patients who could be evaluated at a single center. I would like to see more multicenter efforts, and ideally would like to work with other groups that would enable us to evaluate larger cohorts of patients.
Healio: Is there anything else you’d like to mention?
Rengarajan: We think that many of the features I am describing are probably not unique to ICI diabetes. We’re eager and curious to see if some of what we’re finding — like these genetic risk factors — would be able to potentially predict other immune-related adverse events, as well.
Reference:
For more information:
Michelle Rengarajan, MD, PhD, can be reached at mrengarajan@mgh.harvard.edu.
Collapse
Healio Interviews
Click Here to Manage Email Alerts