Fulminant type 1 diabetes mellitus induced by tislelizumab in a patient with lung adenocarcinoma: a case report

Abstract

Background

Although immune checkpoint inhibitor-related endocrine adverse reactions are common in clinical practice, fulminant type 1 diabetes induced by tislelizumab remains rare.

Case presentation

A 68-year-old Asian female patient was diagnosed with stage IV BRAF V600E-mutant lung adenocarcinoma via pleural biopsy in August 2020 and initially treated with dabrafenib and trametinib. In March 2023, she started treatment with tislelizumab owing to a new metastasis in the left lower lung. On 20 May 2024, the patient was admitted to the hospital with a 3-day history of the sudden onset of dry mouth, polydipsia, and polyuria, accompanied by mild nausea and vomiting. Laboratory examination results showed hyperglycemia, ketosis, and mild metabolic acidosis, but HbA1c was mildly elevated at 7%. The C-peptide release test indicated severe insulin deficiency. She was diagnosed with fulminant type 1 diabetes, treated with intravenous insulin and fluid resuscitation, and discharged. A total of 6 months after discontinuing tislelizumab, her pancreatic function had not recovered.

Conclusion

Tislelizumab can induce fulminant type 1 diabetes via immune-mediated mechanisms. Proactive monitoring of fasting glucose, HbA1c, and islet function is essential for patients with cancer receiving tislelizumab therapy.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13256-026-05837-y.

Introduction

As a novel class of antitumor drugs, immune checkpoint inhibitors (ICIs) have significantly improved survival rates and patient prognosis. They now serve as first-line therapeutic agents for various solid tumors, including melanoma, non-small cell lung cancer, hepatocellular carcinoma, and esophageal cancer [1]. Common immune checkpoints include cytotoxic T lymphocyte-associated antigen 4 (CTLA-4) and programmed death-1/programmed death-ligand 1 (PD-1/PD-L1) inhibitors. These drugs can enhance the activation of the host’s T lymphocytes, thereby enabling them to recognize and kill tumor cells more effectively. Unlike conventional chemotherapy and targeted therapies, endocrine system disruption is among the most common immune-related adverse events (irAEs) linked to immune checkpoint inhibitors. Studies have demonstrated that approximately 7–40% of patients treated with immune checkpoint inhibitors develop endocrine-related irAEs. The most prevalent conditions are thyroid dysfunction and hypophysitis. Less common injuries include immunotherapy-associated type 1 diabetes mellitus, adrenal insufficiency, and hypoparathyroidism [2].

In this study, we present clinical manifestations, laboratory findings, and therapeutic approaches for a patient who developed fulminant type 1 diabetes (FT1DM) after receiving tislelizumab treatment for a lung malignant tumor, aiming to elucidate the  underlying mechanisms and propose a management framework for clinicians encountering similar cases.

Case description

A 68-year-old Asian female patient was diagnosed with stage IV BRAF V600E-mutant lung adenocarcinoma in August 2020 through enhanced chest computed tomography (CT) and pleural biopsy and initially received therapy with dabrafenib plus trametinib (Fig. 1). In March 2023, the patient started tislelizumab treatment owing to a chest CT showing a new metastasis in the left lower lung (Fig. 2). During immunotherapy, serial fasting glucose measurements remained within normal range, fluctuating between 4.62 and 5.85 mmol/L.

Fig. 1.

Enhanced chest computed tomography showing a solid lesion in the right lung (arrow)

Fig. 2.

Enhanced chest computed tomography showing a new solid lesion in the lower left lung (arrow)

In April 2024, this patient was diagnosed with hypoadrenocorticism due to nausea, vomiting, and electrolyte disturbances, and the treatment with tislelizumab was suspended. The patient started on oral prednisone acetate 5 mg daily. On 20 May 2024, the patient was admitted to the Department of Endocrinology on an emergency basis with a 3-day history of the sudden onset of dry mouth, polydipsia, and polyuria, accompanied by mild nausea and vomiting. A timeline figure of the patient’s disease progression is shown in Fig. 3. She had a history of hypothyroidism, was on long-term treatment with levothyroxine tablets 50 μg daily, and denied a history of diabetes mellitus. Physical examination showed a temperature of 36.1 °C, pulse rate of 84 beats per minute, respiration rate of 21 breaths per minute, and a blood pressure of 129/69 mmHg. The patient’s body mass index (BMI) was 17.48 kg/m².

Fig. 3.

A timeline figure of the patient’s disease progression

Arterial blood gas results were as follows: pH was 7.37 (7.35–7.45), base excess was −8.4 mmol/L (−2–3 mmol/L), total CO₂ was 18.9 mmol/L (22–29 mmol/L), and actual bicarbonate was 17.9 mmol/L (18–23 mmol/L). Blood glucose was 33.88 mmol/L. Electrolyte levels were as follows: sodium at 133.7 mmol/L, chloride at 94.4 mmol/L, and potassium at 4.19 mmol/L. Glycosylated hemoglobin A1c (HbA1c) was mildly elevated at 7%. The concentration of β-hydroxybutyric acid was 2.2 mmol/L. Urine glucosuria and ketonuria were 2+, respectively. Renal function tests showed a blood urea nitrogen (BUN) level of 12.36 mmol/L and a creatinine level of 90.9 μmol/L. The blood lipase was 146 U/L, and the blood amylase was 74 U/L. Cortisol was 0.33 μg/dL, and adrenocorticotropic hormone was 1 pg/mL. Further evaluation of autoimmune diabetes mellitus antibodies suggested that the glutamic acid decarboxylase antibody (GADA) level was 45.64 IU/mL, while both the anti-insulin antibody and anti-islet cell antibody tests were negative. Other laboratory tests showed no significant abnormality (Table 1). In conclusion, the patient was diagnosed with diabetic ketosis and immediately administered insulin intravenously to reduce blood glucose levels, along with fluid resuscitation. Concurrently, levothyroxine was initiated at a dose of 75 μg once daily.

Table 1.

Laboratory examination results of the patient

Blood tests		Reference interval
White blood cells (WBC) (109/L)	8.32	3.5–9.5
Red blood cells (RBC) (1012/L)	3.99	3.8–5.1
Hemoglobin (g/L)	119	115–150
Platelets (109/L)	182	125–350
Creatinine (μmol/L)	90.9	57–111
Blood urea nitrogen (BUN) (mmol/L)	12.36	1.79–7.14
Uric acid (UA) (μmol/L)	489.9	150–416
Aspartate aminotransferase (AST) (U/L)	21.1	15–40
Alanine aminotransferase (ALT) (U/L)	7.8	9–50
Total bilirubin (μmol/L)	13.2	0–26
Direct bilirubin (μmol/L)	2.6	0–6.8
Indirect bilirubin (μmol/L)	10.6	0–15
K+ (mmol/L)	4.19	3.5–5.3
Na+ (mmol/L)	133.7	137–147
Cl− (mmol/L)	94.4	99–110
Ca2+ (mmol/L)	2.47	2.11–2.52
Mg2+ (mmol/L)	0.85	0.75–1.02
C-reactive protein (mg/L)	0.88	0–6
Procalcitonin (ng/mL)	0.1	0–0.5
Total T3 (TT3) (ng/mL)	0.51	0.60–1.55
Total T4 (TT4) (ng/mL)	9.52	5.42–12.74
Free Triiodothyronine (FT3) (ng/mL)	2.19	2.30–4.80
Free Thyroxine (FT4) (ng/mL)	1.47	0.62–1.24
Thyroid stimulating hormone (TSH) (uIU/mL)	0.99	0.56–5.91
Cortisol (μg/dL)	0.33	4.26–24.85
Adrenocorticotropic hormone (ACTH) (pg/mL)	1	7.2–63.4
Amylase (U/L)	74	35–135
Lipase (U/L)	146	0–60
PH	7.37	7.35–7.45
PCO2 (mmHg)	31	35–48
Base excess (BE) (mmol/L)	−8.4	−2–3
Total CO2 (mmHg)	18.9	22–29
HCO3− (mmol/L)	17.9	18–23
Carcinoembryonic antigen (CEA) (ng/mL)	2.79	0–5
β-hydroxybutyric acid (mmol/L)	2.2	< 0.5
Urine sugar (mmol/L)	2+	Neg
Urinary ketone (mmol/L)	2+	Neg
Glucose (mmol/L)	33.88	3.89–6.11
HbA1c (%)	7	< 6.5
GADA (IU/mL)	45.64	0–10
Anti-insulin antibody (RU/mL)	0.04	0–20
Anti-islet cell antibody (RU/mL)	0.56	0–20

After the patient was stabilized, she switched to a continuous subcutaneous insulin pump to manage hyperglycemia. The oral glucose tolerance test (OGTT) and C-peptide release test indicated that pancreatic islet β-cell function was failing. OGTT results were: 0 minutes: 4.08 mmol/L; 30 minutes: 7.90 mmol/L; 60 minutes: 12.31 mmol/L; 120 minutes: 16.91 mmol/L; 180 min: 22.89 mmol/L. C-peptide release test values were: 0 minutes: 0.15 ng/mL; 30 minutes: 0.12 ng/mL; 60 minutes: 0.07 ng/mL; 120 minutes: 0.09 ng/mL; 180 minutes: 0.08 ng/mL (Table 2).

Table 2.

OGTT and C-peptide release test of the patient

OGTT test (mmol/L)		Reference interval
0 hours	4.08	3.89–6.11
0.5 hours	7.9
1 hour	12.31
2 hours	16.91	0–7.8
3 hours	22.89
C-peptide release test (ng/mL)
0 hours	0.15	1.1–4.4
0.5 hours	0.12
1 hour	0.07
2 hours	0.09
3 hours	0.08

After discharge, the patient consistently adhered to an intensive insulin therapy regimen utilizing a three-short-one-long insulin protocol. At the 3-month follow-up, the patient’s GADA remained positive. By the 6-month follow-up, the patient’s GADA had converted to negative. However, fasting C-peptide levels remained suboptimal (0.09 ng/mL), demonstrating no significant improvement.

Discussions

Fulminant type 1 diabetes mellitus (FT1DM) is a new subtype of type 1 diabetes mellitus (T1DM) that is characterized by severe damage to the body’s β-cells over several days, with a sharp rise in blood glucose and a slightly increased glycated hemoglobin. Three key diagnostic criteria for FT1DM are the occurrence of diabetic ketosis or ketoacidosis within 7 days, a plasma glucose level of 16.0 mmol/L or higher along with an HbA1c below 8.7%, and a fasting serum C-peptide level less than 0.3 ng/mL, plus a post-meal C-peptide level less than 0.5 ng/mL [3–5]. In this paper, the patient denied a history of diabetes. She exhibited a rapid onset within 7 days, with typical hyperglycemia symptoms and moderate nausea and vomiting. Laboratory results showed hyperglycemia (glucose level of 33.88 mmol/L), ketosis, mild metabolic acidosis (actual bicarbonate was 17.9 mmol/L), and severely impaired β-cell function, indicating significant insulin deficiency, though HbA1c was only mildly elevated at 7%, all of which met the diagnostic criteria for FT1DM. Furthermore, this case featured elevated pancreatic enzymes (lipase), consistent with the typical pancreatic involvement seen in FT1DM.

Distinct from classic type 1 diabetes, FT1DM typically manifests in older individuals and is classified as type 1B diabetes owing to the frequent absence of islet autoantibodies. However, this case showed a positive GADA among islet autoantibodies. Previous studies have also found that islet-related autoantibodies can be positive in some patients with FT1DM, including GADA, zinc transporter-8 (ZnT8A), and the protein tyrosine phosphatase IA2 (IA-2A) [6, 7]. Studies have shown that the positive rate of GADA in Chinese patients with FT1DM is significantly higher than that in the Japanese population (23.3% versus 5.1%, P < 0.01) [8]. Two additional studies have also shown that the positive rate of GADA in Chinese patients with FT1DM can range from 30.7% to 34% [9, 10]. Collectively, these findings indicated that GADA negativity is not a prerequisite for FT1DM diagnosis, and GADA positivity rates vary across populations, potentially owing to genetic susceptibility. Moreover, compared with GADA-negative patients, GADA-positive patients were more likely to develop diabetic ketoacidosis after receiving PD-1/PD-L1 treatment.

At present, although immune checkpoint inhibitors have become a powerful weapon for advanced malignant tumors and are widely used in clinical practice, the incidence of immune checkpoint inhibitor-related adverse events has risen annually, with thyroid dysfunction being the most common [11]. Compared with thyroid dysfunction, diabetes caused by immune checkpoint inhibitors is less frequent, occurring at a rate of 0.2–1.4% [12, 13].

As a humanized immunoglobulin G4 (IgG4) anti-PD-1 monoclonal antibody, tislelizumab was first approved on 26 December 2019 in China for the treatment of hematological malignancies and advanced solid tumors, including urothelial carcinoma, gastric and esophageal cancer, lung cancer, and hepatocellular carcinoma, among others [14]. Compared with other immune checkpoint inhibitors, reports of diabetes caused by tislelizumab are rare [2]. To date, only one case of tislelizumab-induced type 1 diabetic ketoacidosis has been documented, involving a patient with small cell lung cancer [15]. However, FT1DM caused by tislelizumab has not been reported in literature.

The pathogenesis of immune checkpoint inhibitor-related diabetes (ICI-DM) is complex, primarily involving immune factors and susceptibility genes [16]. PD-L1 is expressed not only in tumor cells but also in islet cells. Normally, the PD-1/PD-L1 pathway inhibits the activation of autoreactive T lymphocytes on islet cells, thereby protecting against the occurrence of diabetes. After blocking the PD-1/PD-L1 axis with PD-1/PD-L1 inhibitors, the inhibition of this activation is relieved, leading to an increase in the number or activity of T cells, which in turn results in islet cell infiltration and destruction [17–19]. An animal study demonstrated that 11-week-old non-obese diabetic (NOD) mice treated with anti-PD-1 inhibitors developed diabetes significantly earlier than controls. Histological analysis revealed markedly worsened insulitis and increased pancreatic CD8⁺ T cell proliferation [20]. Furthermore, several studies have shown that some human leukocyte antigen (HLA)-susceptible alleles may be associated with the pathogenesis of F1DM, such as HLA-DR4 and HLA-DR9 [16, 21, 22].

Since hyperglycemia induced by ICI is generally irreversible, and the patient’s pancreatic function is severely impaired, long-term insulin therapy is typically required for glycemic control. In subsequent disease management, if the patient’s glycemic levels stabilize and clinical circumstances warrant it, re-administration of ICIs may be reconsidered [23, 24].

Conclusion

Because the mechanism of ICI-DM mirrors antitumor mechanisms, it is highly challenging for patients with tumors who are receiving immunotherapy to completely prevent the onset of diabetes. Although the incidence of FT1DM is rare, its occurrence poses a life-threatening risk, necessitating early identification and treatment. In patients with cancer who are treated with ICIs, close monitoring of fasting blood glucose, HbA1c, fasting insulin, fasting C-peptide levels, and islet autoantibodies is essential. In addition, patients with appropriate medical and economic resources should undergo testing for HLA susceptibility genes.

Authors’ contributions

Yuanzhen Feng and Limin Han conceived and wrote the manuscript. Xiankang Cheng and Jiao Liu collected the clinical data, and Jun Qi revised the manuscript. All authors approved the final version of the manuscript.

Funding

This study was supported by the Science and Technology Planning Project of Changshou District (CSKJ2024028).

Data availability

All data of this patient are included in this report.

Declarations

Ethics approval and consent to participate

Ethics approval was acquired.

Consent for publication

Written informed consent was obtained from the patient for publication of this case report and any accompanying images. A copy of the written consent is available for review by the Editor-in-Chief of this journal.

Competing interests

The authors declare no competing interests.