Highlights
Analysis of 13,966 patients across 158 National Cancer Institute (NCI) trials highlights that immune checkpoint inhibitor-induced diabetes (ICI-D) has an overall cumulative incidence of 0.52%.
The risk of ICI-D is significantly higher with combination immunotherapy (OR 2.68) compared to PD-1/PD-L1 monotherapy, whereas concurrent chemotherapy appears to reduce the risk (OR 0.38).
Despite its low incidence, ICI-D is clinically aggressive: 90% of affected patients required hospitalization at diagnosis, and 43% necessitated intensive care unit (ICU) admission.
Extreme hyperglycemia serves as a critical clinical marker, helping clinicians differentiate ICI-D from other more common causes of hyperglycemia in oncology patients.
Background and Clinical Context
The introduction of programmed cell death 1 (PD-1) and programmed cell death ligand 1 (PD-L1) inhibitors has revolutionized the management of diverse malignancies. However, this therapeutic leap comes with the cost of immune-related adverse events (IRAEs). While dermatologic, gastrointestinal, and pulmonary toxicities are well-documented, endocrine toxicities—specifically immune checkpoint inhibitor-induced diabetes (ICI-D)—remain a challenge due to their sudden onset and life-long impact.
Unlike Type 2 diabetes, which typically develops over years of insulin resistance, ICI-D represents a rapid, T-cell-mediated destruction of pancreatic beta cells. This results in an absolute insulin deficiency that can manifest as fulminant diabetic ketoacidosis (DKA). Given the expanding use of these agents in both metastatic and adjuvant settings, understanding the precise incidence and risk factors for ICI-D is paramount for patient safety and clinical vigilance.
Study Design and Methodology
To characterize the landscape of ICI-D, researchers conducted a comprehensive review of the NCI Cancer Therapy Evaluation Program (CTEP) database. The study analyzed data from 158 clinical trials conducted between June 2015 and December 2022, encompassing nearly 14,000 patients. These trials included diverse regimens involving PD-1 or PD-L1 agents administered in both academic and community medical centers.
The investigators retrieved adverse event (AE) reports related to diabetes, hyperglycemia, and acidosis. Clinical information was manually curated to distinguish true ICI-D from transient steroid-induced hyperglycemia or pre-existing Type 2 diabetes. The primary outcomes included the cumulative incidence of ICI-D, the clinical characteristics at presentation, and the identification of treatment-related risk factors using logistic regression analysis.
Key Findings: Incidence and Treatment Regimen Influence
The study found that the overall cumulative incidence of ICI-D was 0.52 per 100 treated patients. While this confirms the condition is rare, the incidence fluctuated significantly based on the therapeutic backbone used.
The Impact of Combination Therapy
Patients receiving combination immunotherapy (e.g., dual checkpoint inhibition) faced a substantially higher risk. The incidence in the combination group was 0.94% compared to 0.37% for those on PD-1/PD-L1 monotherapy. The logistic regression revealed an odds ratio (OR) of 2.68 (95% CI, 1.69-4.24), suggesting that intensified immune activation significantly increases the likelihood of autoimmune beta-cell destruction.
The Protective Effect of Chemotherapy
One of the more intriguing findings was the lower incidence of ICI-D in patients receiving concurrent chemotherapy. The incidence was 0.26% with chemotherapy versus 0.65% without (OR 0.38; 95% CI, 0.21-0.71; P = .002). This suggests that the immunosuppressive effects of cytotoxic chemotherapy may modulate the aggressive T-cell response directed at the pancreas, potentially providing a protective buffer against this specific IRAE.
The Clinical Burden of ICI-D
The morbidity associated with ICI-D is disproportionately high relative to its frequency. The study noted that 90% of patients diagnosed with ICI-D required hospitalization. Even more concerning, 43% of these patients required ICU-level care, often due to severe metabolic derangements such as DKA. This high healthcare utilization underscores the need for oncology teams to have rapid protocols for identifying and managing metabolic emergencies.
Diagnostic Differentiation
Differentiating ICI-D from other forms of hyperglycemia is essential for appropriate management. Patients on cancer therapy often experience hyperglycemia due to high-dose corticosteroids or stress. However, the study found that the degree of hyperglycemia is a powerful differentiator. Patients with ICI-D presented with significantly higher glucose levels than those with other etiologies. Clinicians should maintain a high index of suspicion for ICI-D when glucose levels rise precipitously, especially if accompanied by low C-peptide levels or positive autoantibodies, although the latter are not always present in ICI-D.
Expert Commentary and Mechanistic Insights
From a mechanistic standpoint, ICI-D is thought to occur when the inhibition of the PD-1/PD-L1 pathway removes the “brakes” from autoreactive T-cells that target pancreatic islets. The observation that chemotherapy reduces this risk aligns with the hypothesis that generalized lymphodepletion or the anti-inflammatory effects of certain chemo-regimens can dampen this specific autoimmune cascade.
Limitations of the study include the reliance on adverse event reporting, which may lead to under-recognition of milder cases. However, the use of NCI CTEP data provides a robust, large-scale view that is often missing from single-center retrospective cohorts. Experts suggest that as immunotherapy moves into the adjuvant setting for earlier-stage cancers, the long-term management of ICI-D will become an increasingly important part of survivorship care.
Conclusion and Summary
The NCI trial conglomerate analysis confirms that ICI-D is a rare but life-threatening complication of modern immunotherapy. The risk is nearly three times higher with combination immunotherapy and significantly lower when chemotherapy is part of the regimen. Given that 90% of cases lead to hospitalization, early detection through routine glucose monitoring is vital. For the clinician, a sudden, severe spike in blood glucose in a patient receiving PD-1/PD-L1 inhibitors should be treated as an endocrine emergency until proven otherwise.
References
Quandt ZE, Finnigan S, Hill V, et al. Immune Checkpoint Inhibitor-Induced Diabetes Across National Cancer Institute Trials That Included PD-1 or PD-L1 Agents. JAMA Oncol. 2025; Published online December 26, 2024. doi:10.1001/jamaoncol.2025.5594.