Yu Chen
Xiaolu Wang1,2- 1Department of General Practice, Hangzhou Third People’s Hospital, Hangzhou, Zhejiang,, China
- 2Department of General Practice, Hangzhou Third Hospital Affiliated to Zhejiang Chinese Medical University, Hangzhou, China
Immune checkpoint inhibitor–associated diabetes mellitus (ICI-DM) is a rare but life-threatening endocrine immune-related adverse event characterized by abrupt insulin deficiency and a high incidence of diabetic ketoacidosis (DKA). Unlike classical type 1 diabetes, ICI-DM often develops after only a few treatment cycles, shows a fulminant phenotype with disproportionally modest HbA1c elevation, and is typically irreversible and glucocorticoid-refractory, necessitating permanent insulin therapy. Mechanistically, PD-1/PD-L1 blockade disrupts pancreatic immune tolerance and permits autoreactive CD8+ T-cell–mediated β-cell destruction, with risk amplified by susceptible HLA haplotypes and pre-existing islet autoantibodies (most commonly GAD antibodies). Additional contributors include pancreatic inflammation, cytokine-driven immune activation, incretin axis perturbations (GLP-1/GIP), and β-cell dedifferentiation programs. Clinically, timely recognition requires proactive glucose monitoring, early ketone assessment, and evaluation of C-peptide and pancreatic enzymes, particularly because DKA can occur rapidly after symptom onset. This review synthesizes current evidence on ICI-DM epidemiology, pathogenesis, clinical presentation, and management, and discusses emerging preventive and disease-modifying strategies, including mesenchymal stromal cells and B-cell–targeted approaches, to optimize outcomes in patients receiving immunotherapy.
1 Introduction
Immune checkpoint inhibitors (ICIs) have reshaped modern oncology by restoring antitumor immunity through blockade of key inhibitory pathways, most notably programmed death-1 (PD-1), programmed death-ligand 1 (PD-L1), and cytotoxic T-lymphocyte-associated antigen-4 (CTLA-4) (1). To date, the US Food and Drug Administration has approved 14 ICI agents, encompassing anti-PD-1 antibodies (pembrolizumab, nivolumab, sintilimab, toripalimab, camrelizumab, tislelizumab), anti-PD-L1 antibodies (atezolizumab, durvalumab, avelumab, envafolimab, sugemalimab), and the anti-CTLA-4 antibody ipilimumab (2). By design, these therapies amplify cytotoxic T-cell activity and broaden immune effector function, but this immune disinhibition also generates a distinctive spectrum of immune-related adverse events (irAEs) that differs fundamentally from toxicities observed with cytotoxic chemotherapy or molecularly targeted agents (3).
Among irAEs, endocrine toxicities are increasingly recognized as clinically relevant complications as ICI use expands across tumor types and treatment lines (3). Reported endocrine dysfunction affects approximately 30% of treated individuals and most commonly involves the pituitary, thyroid, pancreas, and adrenal cortex (4–6). Although pancreatic endocrine injury is uncommon, immune checkpoint inhibitor–associated diabetes mellitus (ICI-DM) is clinically disproportionate to its incidence because it often presents abruptly and may progress to fulminant diabetic ketoacidosis (DKA), a life-threatening emergency (7–10). Accordingly, timely recognition, mechanistic understanding, and standardized management are essential for optimizing outcomes. In this review, we synthesize current evidence on the epidemiology, clinical phenotype, pathogenesis, and management of ICI-DM, with the goal of providing a practical framework for surveillance and multidisciplinary care in routine oncology practice.
2 Epidemiology of ICI−DM
The 2022 American Diabetes Association classification recognizes immune checkpoint inhibitor–induced diabetes as an immune-mediated form of type 1 diabetes mellitus (T1DM) (11). Clinically, this phenotype closely resembles, and can be difficult to distinguish from, fulminant T1DM (12, 13). Across published cohorts, the reported incidence of ICI-DM ranges from 0.2% to 1.4%; however, its clinical impact is disproportionate to its frequency because more than half of patients present with DKA at diagnosis (8–10). Importantly, β-cell autoantibody positivity and combination regimens that co-blockade PD-1/PD-L1 and CTLA-4 are consistently associated with a higher likelihood of DKA, indicating that host autoimmunity and treatment intensity jointly shape disease severity (14, 15).
The risk of ICI-T1DM varies by therapeutic class, with the highest frequency observed under PD-1/PD-L1 plus CTLA-4 dual therapy, followed by PD-1 monotherapy, PD-L1 monotherapy, and CTLA-4 monotherapy (15). A plausible explanation is a “double-hit” mechanism: CTLA-4 blockade broadly expands and activates T cells, whereas PD-1 pathway blockade removes peripheral tolerance checkpoints that otherwise restrain β-cell–directed autoreactivity, thereby facilitating organ-specific immune injury (16). In contrast, CTLA-4 monotherapy rarely precipitates diabetes, potentially reflecting both narrower clinical utilization and the constitutive expression of CTLA-4 in pituitary tissue, which may bias toxicity toward hypophysitis rather than pancreatic autoimmunity (17). Consistently, agent-level estimates report T1DM incidences of 3.5% with nivolumab, 2.3% with pembrolizumab, 0.7% with atezolizumab, and 0.3% with durvalumab, whereas ipilimumab alone is only infrequently implicated (18, 19).
Epidemiologically, ICI-DM is most commonly reported in melanoma and non–small-cell lung cancer, likely reflecting both disease burden and patterns of ICI utilization (8). Demographic trends across case series indicate a male predominance (20–24). Across cohorts, median onset is approximately 7–17 weeks, typically after about four treatment cycles (25, 26). More broadly, the interval from ICI initiation to overt hyperglycaemia spans 1 week to 12 months, underscoring the need for ongoing vigilance throughout—and potentially beyond—the treatment course (27).
3 Clinical manifestations of ICI-DM
Immune checkpoint blockade can precipitate immune-mediated pancreatic injury; thus, ICI-DM is an acute insulin-deficient syndrome that often resembles T1DM. Patients may be asymptomatic or develop polydipsia, polyuria, and weight loss. Hyperglycaemia typically occurs from several days to 12 months after the first infusion and frequently presents with ketosis or overt DKA, while a minority show isolated hyperglycaemia. Onset is abrupt, and about half of patients develop DKA requiring intensive-care management (28–30). Consistent with rapid β-cell loss, endogenous insulin secretion is often exhausted within three weeks of clinical onset (31), and DKA tends to develop roughly two weeks after initial hyperglycaemic symptoms are recognized (32).
Notably, β-cell failure in ICI-DM generally proceeds more rapidly. Once pancreatic autoimmunity is triggered, islet function deteriorates swiftly, producing a phenotype that resembles fulminant T1DM (FT1DM) (33). FT1DM is characterized by marked hyperglycaemia despite disproportionately low glycated haemoglobin, minimal or absent C-peptide, and a high propensity for early ketosis or DKA (34, 35). Notably, approximately 43% of ICI-DM cases meet diagnostic criteria for FT1DM, a markedly higher proportion than that reported in acute-onset T1DM cohorts (36, 37). Autoantibody-positive patients appear to be at higher risk of DKA and display a fulminant course, with precipitous collapse of β-cell reserve (14). β-cell destruction in ICI-DM is slower than in classical FT1DM but substantially faster than in typical acute T1DM, supporting the concept of an intermediate yet highly aggressive autoimmune trajectory (32). Therefore, early recognition is essential for oncologists, endocrinologists, and emergency physicians. Although no deaths have been definitively attributed to ICI-T1DM, fatal events may be under-recognized or misattributed to malignancy rather than therapy-related toxicity. In contrast to FT1DM, where islet autoantibodies are usually absent, approximately half of patients with ICI-T1DM test positive for at least one islet autoantibody, reinforcing an immune-mediated pathogenesis. Concomitant immune-related adverse events (irAEs) are also common, occurring in nearly 40% of cases, most frequently thyroid dysfunction, with hypophysitis and adrenal insufficiency also reported (14). Accordingly, systematic evaluation for coexisting endocrine and non-endocrine irAEs is critical to reduce missed or delayed diagnoses and to guide comprehensive supportive care.
Laboratory features can be diagnostically informative. HbA1c often underestimates the severity and tempo of hyperglycaemia, remaining normal or only modestly elevated at presentation despite marked glucose derangements (10, 13, 20). Exocrine pancreatic involvement has also been reported. A meta-analysis found that approximately half of ICI-treated recipients develop elevations in pancreatic enzymes, including asymptomatic hyperenzymemia in 3% and overt pancreatitis in 2% (31). However, whereas most patients with FT1DM exhibit increased lipase, amylase, or elastase-1, many individuals with ICI-T1DM do not, suggesting partial dissociation between endocrine failure and overt exocrine inflammation in this setting. Circulating glucagon levels are generally normal (10). Imaging findings are heterogeneous, patients may exhibit pancreatic atrophy (10), pancreatic enlargement, or features consistent with diffuse pancreatitis (38). Pancreatic enlargement may precede diabetes onset and can be followed by progressive pancreatic volume loss (8). Taken together, these observations underscore the need for vigilant metabolic surveillance during ICI therapy and for rapid initiation of insulin replacement and standardized DKA management protocols when hyperglycaemia is detected.
4 Pathophysiological mechanisms of ICI-DM
4.1 PD−1/PD−L1 signaling and ICI-DM
ICI-DM is strongly associated with disruption of the PD-1/PD-L1 axis. PD-1 is an inhibitory receptor expressed on T cells, B cells, activated monocytes, and dendritic cells, and it binds PD-L1, which is broadly expressed across leukocytes, lymphoid organs, and parenchymal tissues, including pancreatic islets. Notably, β cells from individuals with type 1 diabetes or with islet autoantibody positivity exhibit upregulated PD-L1 expression, consistent with an inducible tissue-protective checkpoint response (39). In non-obese diabetic (NOD) mice, anti-PD-1 therapy precipitates rapid-onset diabetes (40), and genetic ablation of PD-1 or PD-L1 markedly accelerates disease onset (41). Antigen-specific tolerance models similarly demonstrate that PD-1/PD-L1 blockade is sufficient to provoke autoimmune diabetes (42). Mechanistically, interferon-γ enhances β-cell PD-L1 expression in NOD islets (39), whereas loss of PD-1 on CD4+ T cells augment islet antigen–specific infiltration, underscoring the pathway’s role in restraining diabetogenic autoimmunity (43).
Physiologically, PD-1 signaling limits T-cell activation and proliferation, and tumors exploit this inhibitory checkpoint to evade immune elimination; CTLA-4, expressed on naïve and regulatory T cells, competes with the co-stimulatory receptor CD28 for CD80/86 binding, thereby attenuating early T-cell priming. Therapeutic antibodies targeting PD-1, PD-L1, or CTLA-4 restore T-cell effector function and promote tumor cytolysis, but the resulting immune hyperactivation also underpins immune-related adverse events (44). In the pancreas, activated T cells can destroy β cells and thereby abrogate insulin production. PD-L1 expression localizes to islets containing β cells and correlates with CD8+ T-cell infiltration; concordantly, whole-blood RNA profiling from newly diagnosed ICI-DM cases demonstrates PD-L1 upregulation (45). In NOD models, PD-1/PD-L1 blockade reliably triggers diabetes, whereas transgenic overexpression of either PD-1 or PD-L1 reduces incidence or delays onset (46–49). Importantly, these studies indicate preferential upregulation of PD-L1—but not CTLA-4 ligands CD80/86—on β cells, supporting a dominant role for the PD-1 pathway in islet immune tolerance (50). Human mechanistic evidence remains limited; however, in a 63-year-old man with well-controlled type 2 diabetes treated with nivolumab for renal cell carcinoma with pancreatic metastasis, histology from non-tumorous pancreas revealed dense T-cell infiltration dominated by CD8+ over CD4+ cells, with few B cells or macrophages, implicating CD8+ T cells as principal effectors in CIADM pathogenesis (51). Collectively, findings from animal models and the emerging human data support the concept that PD-1/PD-L1 blockade permits the survival and expansion of autoreactive T cells, culminating in β-cell eradication and the onset of ICI-T1DM (52, 53).
4.2 Susceptibility genes/HLA polymorphisms and ICI-DM
The contribution of heritable factors to conventional T1DM is well established. HLA genotyping in selected T1DM cohorts has consistently shown that diabetes-predisposing HLA configurations are markedly over-represented compared with the general population (26). Approximately 90% of patients carry the DR3-DQ2 or DR4-D08 haplotypes (54, 55), with DR3-D02 and DR4-D08 single haplotypes comprising the principal risk alleles (56). In contrast, DR4-D04 and DR9-D09 haplotypes have been associated with fulminant diabetes in East-Asian populations (57). These observations provide a rationale for considering pre-treatment assessment of high-risk HLA alleles in patients scheduled to receive ICIs, with the goal of estimating the likelihood of ICI-T1DM and informing surveillance intensity.
Beyond classical HLA risk, emerging genomic data suggest that additional susceptibility loci may modulate ICI-DM pathogenesis. Comprehensive genomic analyses have identified 15 genes with elevated expression in ICI-DM (58). The most frequently reported coding alteration is a missense variant in NLRC5; additional missense variants in DNAJB11, PXN, and XRCC3 (58). Variants in NLRC5 and CEMIP2 appear particularly enriched in CIADM (58). Mechanistically, NLRC5 acts as a key transactivator of HLA class I genes, and its dysregulation may increase islet immunogenicity and thereby amplify autoimmune responses in predisposed individuals (59). This concept is biologically plausible given that the highly polymorphic HLA region on chromosome 6 is strongly linked to autoimmunity, and that class I and class II alleles are essential for CD8+ and CD4+ T-cell activation, respectively (60). The association of class II–restricted alleles with organ-specific autoimmunity further emphasizes the tissue-selective nature of HLA-mediated risk (60). Several studies have examined HLA haplotypes specifically in ICI-DM. In a retrospective series of 200 patients, HLA typing was available for 78 individuals; among these, 51% carried an HLA-DR4 haplotype, reported as a higher prevalence than that observed in Caucasian classical T1DM, whereas 14% carried HLA-DR3 and 10% possessed putatively protective haplotypes (61).
4.3 Islet autoantibodies and ICI-DM
T1DM is initiated by T cell–mediated destruction of pancreatic β cells, culminating in progressive insulin deficiency (62). In parallel, B cells contribute to disease initiation and risk stratification through the production of islet-directed autoantibodies (63). Approximately 95% of patients with classical T1DM test positive for at least one pancreatic autoantibody (63), and the presence of two or more distinct autoantibodies is associated with an estimated 50% risk of developing clinical T1DM (64). In the context of ICI-T1DM, pancreatic autoantibody positivity is observed in a substantial subset of patients and is widely regarded as a supportive biomarker of immune-mediated β-cell injury. Clotman et al. (26) reported that, among 39 patients with ICI-T1DM, 22 (56%) were positive for pancreatic autoantibodies, and all seropositive cases had glutamic acid decarboxylase autoantibodies (GADA). Additional autoantibodies included protein tyrosine phosphatase IA-2A, islet cell antibodies (ICA), insulin autoantibodies (IAA), and ZnT8A (26). Across cohorts, approximately 40% of patients with ICI-DM harbor classic T1DM-associated autoantibodies, with GADA being the most frequently detected specificity (65). Clinically, autoantibody positivity has been associated with a more rapid disease onset (66). For example, GADA-positive patients have been reported to develop ICI-DM after an average of three treatment cycles, whereas GADA-negative patients develop disease after approximately six cycles on average (67). Moreover, compared with patients lacking T1DM-related autoantibodies, seropositive individuals show a higher incidence of diabetic ketoacidosis (68).
4.4 T−cell subsets and ICI-DM
T cells play a crucial role in autoimmune diseases; however, their involvement in ICI-DM remains poorly understood. Research by Lozano et al. (69) demonstrated that the enrichment of activated memory CD4+ T cells in peripheral blood prior to ICI treatment, along with TCR diversity, is associated with the occurrence of irAEs. Existing autoimmune diseases or preclinical autoimmune states may serve as indicators for the potential development of severe irAEs. Previous studies have suggested a link between CD8+ T cells and latent autoimmune diabetes in adults (LADA). Furthermore, autopsy studies of ICI-DM patients have revealed marked infiltration of T lymphocytes in the pancreatic islets, with CD8+ T cells being more prominent than CD24+ B cells (70), further suggesting a potential association between CD8+ T cells and the development of ICI-DM. Additionally, other T helper subsets, such as Th17 and Th2, may also contribute to the pathogenesis of ICI-DM (71).
4.5 Glucagon-like peptide-1/gastric inhibitory polypeptide and CIADM
GIP and GLP-1 are intestinal peptides that enhance insulin secretion during meals and play a critical role in glucose homeostasis. Changes in their secretion have been linked to the development of abnormal glucose metabolism. Studies have shown that alterations in gut microbiota diversity or imbalances in key microbial species can affect enteroendocrine cell function, consequently influencing blood glucose levels in patients (72). A study focusing on the intestinal endocrine system in CIADM patients treated with ICIs demonstrated reduced GLP-1 levels in these patients. Additionally, GIP concentrations were found to be lower both during fasting and in oral glucose tolerance tests (OGTT) (73). These findings suggest that ICIs may influence gut microbiota, thereby altering enteroendocrine cell function and leading to decreased GLP-1 and GIP levels, which may contribute to the development of CIADM.
4.6 Pancreatic inflammation/β−cell dedifferentiation and ICI-DM
Many CIADM patients exhibit elevated serum amylase and/or lipase at, or preceding, diabetes onset, suggesting pancreatitis and subsequent pancreatic atrophy (74, 75). Biopsy specimens show peri−islet inflammation with interferon−γ and tumor−necrosis−factor−α expression. In PD−L1–blocked diabetic mice, transcriptomic analyses highlight enrichment of interferon−γ response genes, alongside β−cell dedifferentiation characterized by reduced Ins1/2, Chga, Mafa and Nkx6−1, and increased Gcg, Sst and Cxcl9 expression (76). Thus, cellular and humoral immunity, genetic polymorphisms, pancreatic inflammation and β−cell dedifferentiation collectively orchestrate CIADM pathogenesis. Continuous monitoring of islet autoantibodies and lipase may therefore improve early detection and severity assessment.
5 Therapeutic strategies of ICI-DM
ICI-DM often progresses rapidly once it develops. After a definitive diagnosis, immediate initiation of exogenous insulin is strongly recommended, and oncologists should proactively counsel patients regarding the risk of ICI-DM and the high likelihood of permanent insulin dependence. When diabetic ketoacidosis occurs, urgent implementation of standard critical-care protocols—including aggressive intravenous fluid resuscitation, careful electrolyte correction, and continuous intravenous insulin infusion—is essential. Glucocorticoids are the first-line treatment for most immune-related adverse events; accordingly, multiple groups have attempted steroid therapy in ICI-DM (77–79). However, these interventions have been consistently ineffective. In contrast to many other irAEs, ICI-DM is generally irreversible and steroid-refractory (6); therefore, long-term glycaemic control typically relies on sustained insulin replacement (80). Beyond supportive management, emerging disease-modifying strategies are being explored. Mesenchymal stromal cells (MSCs), the most extensively studied cellular therapy platform, secrete bioactive mediators that promote tissue repair, limit fibrosis, and modulate immune responses. In male NOD mice, autologous MSC administration significantly suppressed PD-L1–associated diabetes onset (81). Quantitative histology demonstrated a larger residual β-cell area in MSC-treated animals, with pancreatic insulin content 1.9-fold higher than in untreated controls. MSC therapy also reduced islet-infiltrating T cells by 45% and CXCL9-positive macrophages by 67%, supporting a protective effect through attenuation of immune cell recruitment and tissue infiltration (81). In parallel, rituximab—a chimeric monoclonal antibody targeting the B-cell surface antigen CD20—has demonstrated the ability to delay T1DM progression in clinical trials (82) (Table 1). Figure 1.
Table 1. Pathogenetic mechanisms and emerging therapeutic strategies in ICI-DM.
Figure 1. Immune checkpoint inhibitor-associated diabetes mellitus.
6 Conclusion
Immune checkpoint inhibitor–associated diabetes mellitus (ICI-DM) is an uncommon yet potentially fatal immune-related adverse event that is increasingly encountered as ICIs are deployed across tumor types. In contrast to classical type 1 diabetes, ICI-DM typically presents with an abrupt transition from euglycaemia to profound insulin deficiency, frequently culminating in diabetic ketoacidosis and requiring intensive acute care. Current evidence supports a dominant role for PD-1/PD-L1 blockade in dismantling pancreatic immune tolerance and unleashing autoreactive, β-cell–directed cytotoxic T-cell responses, with disease kinetics shaped by host susceptibility factors such as high-risk HLA haplotypes, pre-existing islet autoantibodies, and permissive inflammatory programs within the pancreas. The emerging picture also implicates contributory mechanisms, including exocrine pancreatic inflammation, cytokine-driven immune activation, incretin axis perturbation, and β-cell dedifferentiation, that may collectively accelerate islet failure and account for the fulminant clinical phenotype.
From a clinical standpoint, early recognition and standardized management are paramount. Given the high incidence of ketosis/DKA and the frequent dissociation between marked hyperglycaemia and modest HbA1c elevations, routine glucose surveillance during ICI therapy, rapid assessment of ketones and C-peptide at symptom onset, and evaluation for concurrent endocrine irAEs are essential. Importantly, ICI-DM is characteristically refractory to glucocorticoids and is irreversible in most cases, making lifelong insulin replacement the mainstay of treatment and patient education a critical component of care. Looking forward, risk-adapted monitoring strategies incorporating HLA typing, islet autoantibodies, and pancreatic enzyme dynamics may enable earlier detection and mitigation. Therapeutically, adjunctive approaches aimed at preserving β-cell mass or tempering immune infiltration, such as mesenchymal stromal cells or B-cell–directed therapies, are biologically plausible and warrant prospective evaluation. Ultimately, improving outcomes will require multidisciplinary pathways that integrate oncology and endocrinology, alongside mechanistic studies to define actionable biomarkers and prevention-oriented interventions.
Author contributions
YC: Writing – original draft, Writing – review & editing. XW: Writing – original draft. SD: Writing – original draft.
Funding
The author(s) declared that financial support was received for this work and/or its publication. This study was supported by the Science and Technology Special Fund for the Development and Support of the Biomedical and Health Industry in Hangzhou (2024WJC130).
Conflict of interest
The authors declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
References
1. Lao D, Lu C, Chen S, Liu F, Li W, Shangguan D, et al. Recent advances in immune checkpoint inhibitor-induced type 1 diabetes mellitus. Lnt Lmmu-Nopharmacel. (2023) 122:110414. doi: 10.1016/J.Intimp.2023.110414
2. Ribas A and Wolchok JD. Cancer immunotherapy using check-point blockade. Science. (2018) 359:1350–5. doi: 10.1126/science.aar4060
3. Zhu J, Lu0 M, Liang D, Gao S, Zheng Y, He Z, et al. Type 1 diabetes with immune-checkpoint inhibitors: A systematic analysis of clinical trials and A pharmacovigilance study of post-marketing data. Int Immu-Nopharmacol. (2022) 110:109053. doi: 10.1016/J.Intimp.2022109053
4. Sznol M, Postow MA, Davies MJ, Pavlick AC, Plimack ER, Shaheen M, et al. Endocrine-related adverse events associated with immune checkpoint blockade and expert insights on their management. Caneer Treat Rev. (2017) 58:70–6. doi: 10.1016/j.ctrv.2017.06.002
5. Fan Y, Geng Y, Shen L, and Zhang Z. Advances on immune-related adverse events associated with immune checkpoint inhibitors. Front Med. (2021) 15:33–42. doi: 10.1007/s11684-019-0735-3
6. Quandt Z and Young A. immune checkpoint inhibitor diabetes mellitus: A novel form of autoimmune diabetes. Clin Exp Immunol. (2020) 200:131–40. doi: 10.1111/cei.13424
7. Kanji S, Morin S, Agtarap K, Purkayastha D, Thabet P, Bosse D, et al. Adverse events associated with immune checkpoint inhibitors: overview of systematic reviews. Drugs. (2022) 82:793–809. doi: 10.1007/s40265-022-01707-1
8. De Filette J, Jj P, Decoster L, et al. Immune checkpoint inhibitors and type L diabetes mellitus: A case report and systematic review. Eur J Endocrinol. (2019) 181:363–74. doi: 10.1530/EJE-19-0291
9. Wright JJ, Salem JE, Johnson DB, Lebrun-Vignes B, Stamatouli A, Thomas JW, et al. Increased reporting of immune checkpoint inhibitor-associated diabetes. Diabetes Care. (2018) 41:E150–1. doi: 10.2337/Dc18-1465
10. Stamatouli AM, Ouandt Z, Perdigoto AL, Clark PL, Kluger H, Weiss SA, et al. Collateral damage: insulin-dependent diabetes induced with checkpoint inhibitors. Diabetes. (2018) 67:1471–80. doi: 10.2337/dbi18-0002
11. Eisayed NA, Aleppo G, Aroda YR, Bannuru RR, Brown FM, Bruemmer D, et al. 2. Classification and diagnosis of diabetes: standards of care in diabetes-2023. Diabetes Care. (2023) 46:S19–40. doi: 10.2337/dc23-S002
12. Barroso-Sousa R, Wt B, Carrido-Castro AC, Hodi FS, Min L, Krop IE, et al. Incidence of endocrine dysfunction following the use of different immune checkpoint inhibitor regimens: A systematic review and mela-analysis. JAMA Oncol. (2018) 4:173–82. doi: 10.1001/jamaoncol.2017.3064
13. De Filette J, Jansen Y, Schreuer M, and Lal L. Incidence of thyroid-related adverse events in melanoma patients treated with pembrolizumab. J Clin Endocrinol Metab. (2016) 101:4431 –4439. doi: 10.1210/jc.2016-2300
14. Akturk HK, Kahramangil D, Sarwal A, and Lal L. Immune checkpoint inhibitor-induced type 1 diabetes: A systematic review and meta-analysis. Diabetes Med. (2019) 36:1075–81. doi: 10.1111/dme.14050
15. Lu J, Zhou H, Zhang Y, Fang W, Yang Y, Huang Y, et al. Reporting of immune checkpoint inhibitor therapy-associated diabetes,2015-2019[J/0l. Diabetes Care. (2020) 43:E79–80. doi: 10.2337/De20-0459
16. Chen X, Affinati AH, Lee Y, Turcu AF, Henry NL, Schiopu E, et al. Immune checkpoint inhibitors and risk of type L diabetes. Diabetes Care. (2022) 45:1170–6. doi: 10.2337/dc21-2213
17. Zheng Z, Lu Y, Yang J, Tan C, Zhou L, Wang X, et al. Diabetes mellitus induced by immune checkpoint inhibitors. Diabetes Metab Res Rev. (2021) 37:E3366. doi: 10.1002/Dmir.3366
18. Yendmuris PS. Diabetes and immune checkpoint inhibitors-letter. Clin Caneer Res. (2023) 29:4016. doi: 10.1158/1078-0432.CCR-23-1742
19. Giezen TJ, Mantel-Teeuwisse AK, Meyboom RH, Straus SM, Leufkens HG, Egberts TC, et al. Mapping the safety profile of biologicals: A disproportionality analysis using the who adverse drug reaction database, vigibase. . Drug Saf. (.2010) 33:865–78. doi: 10.2165/11538330-000000000-00000
20. Tsang V, Mcgrath RT, Clifton-Blgh RJ, Scolyer RA, Jakrot V, Guminski ADs, et al. Check point inhibitor-associated autoimmune diabetes is distinct from typc1 diabetes. J Clin Endocrinol Metab. (2019) 104:5499–506. doi: 10.1210/jc.2019-00423
21. Kotwal A, Haddox C, Block M, and Kudva YC. Immune checkpoint inhibitors: an emerging cause of insulin-dependent diabetes. BMJ Open Diabetes Res Care. (2019) 7:E000591. doi: 10.1136/Bmjdre-2018-000591
22. Tan MH, Iyengar R, Mizokami-Stout K, Yentz S, MacEachern MP, Shen LY, et al. Spectrum of immune checkpoint inhibitors-induced endocrinopathies in cancer patients: A scoping review of case reports. J Clin Diabetes Endocrinol. (2019) 5:1. doi: 10.1186/s40842-018-0073-4
23. Baden MY, Imagawa A, Abiru N, and Lal L. Characteristics and clinical course of type L diabetes mellitus related to anti-programmed cell death-L therapy. Diabetol Int. (2019) 10:58–66. doi: 10.1007/s13340-018-0362-2
24. Singh V, Chu Y, Gupta W, and Zhao CW. A tale of immune-related ad-verse events with sequential trials of checkpoint inhibitors in A patient with metastatic renal cell carcinoma. Cureus. (2020) 12:E8395. doi: 10.7759/Cureus.8395
25. Galligan A, Xu W, Fourlanos S, Nankervis A, Chiang C, Mant AM, et al. Diabetes associated with immune checkpoint inhibition: presentation and management challenges. Diabetes Med. (2018) 2018-06-16):13762. doi: 10.1111/Dme.13762
26. Clotman K, Janssens K, Specenier P, Weets I, and De Block CEM. Programmed cell death-L inhibitor-induced Type L diabetes mellitus. J Clin Endocrinol Metab. (2018) 103:3144–54. doi: 10.1210/jc.2018-00728
27. Zhang R, Cai XL, Lu L, Han XY, and Ji LN. Type 1 diabetes induced by immune checkpoint inhibitors. China Med J (Engl). (2020) 133):2595–8. doi: 10.1097/CM9.0000000000000972
28. Dougan M and Petropaolo M. Time to dissect the autoimmune etiology of cancer antibody immunotherapy. J Clin Invest. (2020) 130:51–61. doi: 10.1172/JCI131194
29. Imagawa A, Hanafusa T, Miyagawa J, and Matsuzawa Y. A novel sub-Type of type 1 diabetes mellitus characterized by A rapid onset and an absence of diabetes-related antibodies, osaka iddm study group. N Engl J Med. (2000) 342:301–7. doi: 10.1056/NEJM200002033420501
30. Shiba M, Inaba H, Ariyasu H, Inagaki Y, Matsuno S, Iwakura H, et al. Fulminant type 1 diabetes mellitus accompanied by positive conversion of anti-insulin antibody after the administration of anti-ctla-4 antibody following the discontinuation of anti-pd-L antibody. Intern Med. (2018) 57:2029–34. doi: 10.2169/internalmedicine.9518-17
31. Zhang W, Chen J, Bi J, Ding N, Chen X, Wang Z, et al. Combined diabetic ketoacidosis and hyperosmolar hyperglycemic state in type 1 diabetes mellitus induced by immune checkpoint inhibitors: under recognized and under reported emergency in icis-dm. Front Endocrinol (Lausanne). (2022) 13:1084441. doi: 10.3389/Fendo.2022.1084441
32. Tachibana M and Imagawa A. Type 1 diabetes related to immune checkpoint inhibitors. Best Praet Res Clin Endoerinol Metab. (2022) 36:101657. doi: 10.1016/J.Beem.2022101657
33. Buzzetti R, Maddaloni E, Gaglla J, Leslie RD, Wong FS, Boehm BO, et al. Adult-onset auto-immune diabetes. Nat Rev Dis Primers. (2022) 8:63. doi: 10.1038/s41572-022-00390-6
34. Imagawa A, Hanafusa T, Awata T, Ikegami H, Uchigata Y, Osawa H, et al. Report of the committee of the Japan diabetes society on the research of fulminant and acute-onset type 1 diabetes mellitus: new diagnostic criteria of fulminant type 1 diabetes mellitus (2012). J Diabetes Lnvesting. (2012) 3:536–9. doi: 10.1111/jdi.12024
35. Imagawa A and Hanafusa T. Fulminant type 1 diabetes–an important subtype in east asia. Diabetes Metab Res Rev. (2011) 27:959–64. doi: 10.1002/dmrr.1236
36. Imagawa A, Hanafusa T, Uchigata Y, Kanatsuka A, Kawasaki E, Kobayashi T, et al. Fulminant type L diabetes; A nationwide survey in Japan. Diabetes Care. (2003) 26:2345–52. doi: 10.2337/diacare.26.8.2345
37. Cuy, Wang Y, Li P, Wei H, Chen L, Liu Q, et al. Fulminant type L diabetes in children: A multicenter study in China. J Diabetes Res. (2017) 2017:6924637. doi: 10.1155/2017/6924637
38. Chang LS, Barroso-Sousa R, Tolaney SM, Hodi FS, Kaiser UB, and Min L. Endocrine toxicity of cancer immunotherapy targeting immune checkpoints. Endocr Rev. (2019) 40:17–65. doi: 10.1210/er.2018-00006
39. Osum KC, Al B, Martinov T, Sahli NL, Mitchell JS, Tucker CG, et al. interferon-gamma drives programmed death-ligand L expression on islet β Cells to limit T cell function during autoimmune diabetes. Sci Rep. (2018) 897:8295. doi: 10.1038/S41598-018-26471-9
40. Ansari MJ, Ad S, Chitnis T, Smith RN, Yagita H, Akiba H, et al. The programmed death-1(Pd-1) pathway regulates autoimmune diabetes in nonobese diabetic (Nod)Mice. J Exp Med. (2003) 198:63–9. doi: 10.1084/Jem.20022125
41. Keir ME, Sc L, Guleria I, Latchman YE, Qipo A, Albacker LA, et al. Tissue expression of pd-L1 mediates peripheral T cell tolerance. J Exp Med. (2006) 203:883–95. doi: 10.1084/Jem.20051776
42. Liu Y, Zhang H, Zhou L, Li W, Yang L, Li W, et al. Immunotherapy-associated pancreatic adverse events: current understanding of their mechanism, diagnosis, and management. Front Oncol. (2021) 11:627612. doi: 10.3389/Fonc.2021.627612
43. Pauken KE, Mk J, Azuma M, and Fife BT. Pd-L, but not pd-L1, expressed by islet-reactive cd4*T cells suppresses infiltration of the pancreas during type 1 diabetes. Diabetes. (2013) 62:2859–69. doi: 10.2337/D612-1475
44. Robert C. A decade of immune-checkpoint inhibitors in cancer therapy. Nat Commun. (2020) 11:3801. doi: 10.1038/s41467-020-17670-y
45. Wu L, Tsang VHM, Sasson SC, Menzies AM, Carlino MS, Brown DA, et al. Unravelling checkpoint inhibitor associated autoimmune diabetes: from bench to bedside [J/0l. Front Endocrinol (Lausanne). (2021) 12:764138. doi: 10.3389/Fendo.2021.764138
46. Ansari MJ, Salama AD, Chitnis T, Smith RN, Yagita H, Akiba H, et al. The programmed death-L(Pd-L) pathway regulates autoimmune diabetes in nonobese diabetes (Nod)Mice. J Exp Med. (2003) 198:63–9. doi: 10.1084/jem.20022125
47. Won TJ, Jung YJ, Kwon SJ, Lee YJ, Lee DI, Min H, et al. Forced expression of programmed death-L gene on T cell decreased the incidence of type 1diabetes. Areh Phar Res. (2010) 33:1825–33. doi: 10.1007/s12272-010-1115-3
48. Wang CJ, Chou FC, Chu CH, Wu JC, Lin SH, Chang DM, et al. Protective Role Of Programmed Death 1 Ligand 1 (Pd-L1) In Nonobese Diabete Mice: The Paradox In Transgene Models. Diabetes. (2008) 57:1861–9. doi: 10.2337/db07-1260
49. Mourad D, Azar NS, Eid AA, and Azar ST. Immune checkpoint inhibitor-induced diabetes mellitus: potential role of T cells in the underlying mechanism. Int J Mol Sei. (2021) 22:2093. doi: 10.3390/ijms22042093
50. Rui J, Deng S, Arazl A, Perdigoto AL, Liu Z, and Herold KC. β Cells that resist immunological attack develop during progression of autoimmune diabetes in nod mice. Cell Metab. (2017) 25:727–38. doi: 10.1016/j.cmet.2017.01.005
51. Yoneda S, Imagawa A, Hosokawa Y, Baden MY, Kimura T, Uno S, et al. T-lymphocyte infiltration to islets in the pancreas of A patient who developed type1 diabetes after administration of immune checkpoint inhibitors. Diabetes Care. (2019) 42:El16–8. doi: 10.2337/Dc18-2518
52. Lee HT, Lee SH, and Heo YS. Molecular interactions of antibody drugs targeting pd-L, pd-L1, and cta-4 in immuno-0ncology. Molecules. (2019) 24:1190. doi: 10.3390/molecules24061190
53. Pizarro C, García-Díaz DF, Codner E, Salas-Pérez F, Carrasco E, and Pérez-Bravo F. Pd-L1 gene polymorphisms and low serum level of pd-L1 protein are associated to type I diabetes in China. Diabetes Metab Res Rev. (2014) 30:761–6. doi: 10.1002/dmrr.2552
54. Nobleja. Immuno-genetics of type L diabetes: A comprehensive review. J Autoimmun. (2015) 64:101–12. doi: 10.1016/j.jaut.2015.07.014
55. Erlich H, Valdes AM, Noble J, Carlson JA, Varney M, Concannon P, et al. Hla dr-dq haplotypes and genotypes and type L diabetes risk: analysis of the type L diabetes genetics consortium families. Diabetes. (2008) 57:1084–92. doi: 10.2337/db07-1331
56. Steck AK and Rewers MJ. Genetics of type L diabetes. Clin Chem. (2011) 57:176–85. doi: 10.1373/clinchem.2010.148221
57. Imagawa A, Hanafusa T, Awata T, Ikegami H, Uchigata Y, Osawa H, et al. Report of the committee of the Japan diabetes society on the research of fulminant and acute-onset type 1 diabetes mellitus: new diagnostic criteria of fulminant type 1 diabetes mellitus (2012). J. Diabetes Lnvestig. (2012) 3:536–9. doi: 10.1111/jdi.12024
58. Caulfield J, Aizenbud L, Al P, Meffre E, Jilaveanu L, Michalek DA, et al. Germline genetic variants are associated with development of insulin-dependent diabetes in cancer patients treated with immune checkpoint inhibitors. J Immunother Cancer. (2023) 11:E006570. doi: 10.1136/Jitc-2022-006570
59. Szymczak F, Alvelos M, Marin-Caãas S, Castela Â, Demine S, Colli ML, et al. Transcription and splicing regulation by nlrc5 shape the interferon response in human pancreatic β Cells. Sci Adv. (2022) 8:5732. doi: 10.1126/SciAdv.Abn5732
60. Gough SC and Simmonds MJ. The hla region and autoimmune disease: associations and mechanisms of action. Cur Genomics. (2007) 8:453465. doi: 10.2174/138920207783591690
61. Lo Preiato V, Salvagni S, Ricci C, Ardizzoni A, Pagotto U, and Pelusi C. Diabetes mellitus induced by immune checkpoint inhibitors: type L diabetes variant or new clinical entity? Review of the literature. Rev Endocr Metabdisord. (2021) 22:337–49. doi: 10.1007/S11154-020-09618-W
62. Herold KC, Vignali DA, Cooke A, and Bluestone JA. Type 1 diabetes: translating mechanistic observations into effective clinical outcomes. Nat Rev Lmmunol. (2013) 13:243–56. doi: 10.1038/nri3422
63. Youssef N, Noureldein M, Daoud G, and Eid AA. Immune check-point inhibitors and diabetes: mechanisms and predictors. Diabetes Metab. (2021) 47:101193. doi: 10.1016/J.Diabet.2020.09.003
64. Ziegler AG, Rewers M, Simell O, Simell T, Lempainen J, Steck A, et al. Seroconversion to multiple islet auto-antibodies and risk of progression to diabetes in children. Jama. (2013) 309:2473–9. doi: 10.1001/jama.2013.6285
65. Stamatouli AM, Quandt Z, Al P, Clark PL, Kluger H, Weiss SA, et al. Collateral damage: insulin-dependent diabetes induced with checkpoint inhibitors. Diabetes. (2018) 67:1471–80. doi: 10.2337/Dbi18-0002
66. Clotman K, Janssens K, Specenier P, Weets I, and De Block CEM. Programmed cell death-1 inhibitor-induced type L diabetes mellitus. J Clin Endocrinol Metab. (2018) 103:3144–54. doi: 10.1210/Jc.2018-00728
67. De Filette JMK, Jj P, Decoster L, Gao X, and Si J. Immune checkpoint inhibitors and type L diabetes mellitus: A case report and systematic review. Eur J Endocrinol. (2019) 181:363–74. doi: 10.1530/Eje-19-0291
68. Akturk HK, Kahramangil D, Sarwal A, Hoffecker L, Murad MH, and Michels AW. Immune checkpoint inhibitor-induced type L diabetes: A systematic review and meta-analysis. Diabetes Med. (2019) 36:1075–81. doi: 10.1111/Dme.14050
69. Lozano AX, Aa C, Nene A, Bacchiocchi A, Earland N, Vesely MD, et al. T cell characteristics associated with toxicity to immune checkpoint blockade in patients with melanoma. Nat Med. (2022) 28:353–62. doi: 10.1038/S41591-021-01623-Z
70. Yoneda S, Lmagawa A, Hosokawa Y, Baden MY, Kimura T, Uno S, et al. T-lymphocyte infiltration to islets in the pancreas of A patient who developed type L diabetes after administration of immune checkpoint inhibitors. Diabetes Care. (2019) 42:El16–118. doi: 10.2337/Dc18-2518
71. Kim KH, JY H, Cho J, Ku BM, Koh J, Koh JY, et al. Immune-related adverse events are clustered into distinct subtypes by ‘T-cell profiling before and early after anti-pd-L treatment. Oncoimmunology. (2020) 9:1722023. doi: 10.1080/2162402x.2020.1722023
72. Tomaro-Duchesneau C, Sl L, Roeth D, Sun L, Horrigan FT, Kalkum M, et al. Discovery of A bacterial peptide as A modulator of clp-L and metabolic disease. Sci Rep. (2020) 10:4922. doi: 10.1038/S41598-020-61112-0
73. Bastin M, Busieau P, Kuhn E, Rouault C, Taboureau O, Toulgoat A, et al. Incretin response in immune-checkpoint inhibitor-induced diabetes: an observational study. Diabetes Metab. (2021) 47:101212. doi: 10.1016/j.diabet.2020.11.004
74. Wu L, Tsang V, Am M, Sasson SC, Carlino MS, Brown DA, et al. Risk factors and characteristics of checkpoint inhibitor-associated autoimmune diabetes mellitus (Ciadm): A systematic review and delineation from type 1diabetes. Diabetes Care. (2023) 46:1292–9. doi: 10.2337/D022-2202
75. Byun DJ, Braunstein R, Flynn J, Zheng J, Lefkowitz RA, and Kanbour S. Immune checkpoint inhibitor-associated diabetes: A single-institution experience. Diabetes Care. (2020) 43:3106–9. doi: 10.2337/Dc20-0609
76. Perdigoto AL, Deng S, Du KC, Kuchroo M, Burkhardt DB, Tong A, et al. Immune cells and their inflammatory mediators modify β Cells and cause checkpoint inhibitor-induced diabetes. JCI Insight. (2022) 7:El56330. doi: 10.1172/Jci.Insight.156330
77. Okamoto M, Okamoto M, Gotoh K, Masaki T, Ozeki Y, Ando H, et al. Fulminant type? 1 diabetes mellitus with anti-programmed cell death-L therapy. J Diabetes Investig. (2016) 7:915–8. doi: 10.1111/jdi.12531
78. Teramoto Y, Nakamura Y, Asami Y, Imamura T, Takahira S, Nemoto M, et al. Case of type diabetes associated with less-dose nivolumab therapy in A melanoma patient. J Dematol. (2017) 44:605–6. doi: 10.1111/1346-8138.13486
79. Codwin JL, Jaggi S, Sirisena I, Sharda P, Rao AD, Mehra R, et al. Nivolumab-induced autoimmune diabetes mellitus presenting as diabetes ketoacidosis in A patient with metastatic lung cancer. J Immunother Cancer. (2017) 5:40. doi: 10.1186/s40425-017-0245-2
80. Dougan M and Pietropaolo M. Time to dissect the autoimmune etiology of cancer antibody immunotherapy. J Clin Lnvest. (2020) 130:51–61. doi: 10.1172/JCI131194
81. Gauci ML, Laly P, Vidal-Trecan T, Baroudjian B, Gottlieb J, Madjlessi-Ezra N, et al. Autoimmune diabetes induced by P-L inhibitor-retrospective analysis and pathogenesis: A case report and literature review. Cancer Mmunol Immunother. (2017) 66:1399–410. doi: 10.1007/s00262-017-2033-8
Keywords: autoimmunity, checkpoint inhibitor-associated diabetes mellitus, diabetic ketoacidosis, immune checkpoint inhibitors, PD-1, PD-L1, T cell, β-cell
Citation: Chen Y, Wang X and Duan S (2026) Immune checkpoint inhibitor-associated diabetes mellitus: mechanisms, clinical manifestations, and management strategies. Front. Endocrinol. 16:1679751. doi: 10.3389/fendo.2025.1679751
Received: 14 August 2025; Accepted: 18 December 2025; Revised: 17 December 2025;
Published: 13 January 2026.
Edited by:
Giulia Lanzolla, University of Cagliari, ItalyReviewed by:
Mohamed Essameldin Abdelgawad, Egypt Center for Research and Regenerative Medicine, EgyptCopyright © 2026 Chen, Wang and Duan. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Yu Chen, MTMxMTY3NDEyNjZAMTYzLmNvbQ==