Beyond the Expected: Life-Threatening Immune-Related Adverse Events from Checkpoint Blockade — A Case Series
Punam Rabha, MD, DrNB*, Ashutosh Jain, MD, DM 1, Jerryes Pious Wisely 1, MD 2,
Narendra Bhardwaj, MBBS 3, Vinod Raina MD, MRCP, FRCP 4
*Correspondence to: Punam Rabha, Department of Medical Oncology, Fortis Memorial Research Institute, Sector 44, Gurugram, Haryana 122002, India.
ORCID: 0000-0001-5441-651X
Copyright
© 2026 Punam Rabha is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Received: 06 October 2026
Published: 01 November 2026
DOI: https://doi.org/10.5281/zenodo.23910554
Abstract
Background: Immune checkpoint inhibitors (ICIs) have transformed the management of multiple solid tumors, but their use is accompanied by immune-related adverse events (irAEs) that can affect virtually any organ system. Although most irAEs are low-grade, a minority are severe or life-threatening and require prompt recognition. We describe three patients who developed distinct, uncommon, high-severity irAEs during ICI therapy.
Methods: We report a retrospective case series of three patients treated with programmed death-1 (PD-1) or PD-1/PD-ligand 1 (PD-L1)–directed therapy at a tertiary oncology referral center. Clinical, laboratory, and imaging data were extracted from the medical record and are presented descriptively.
Results: Case 1, a 54-year-old man with resected clear-cell renal-cell carcinoma receiving adjuvant pembrolizumab, developed ICI-induced hypophysitis after 13 cycles, presenting with severe hyponatremia and biochemical evidence of central hypothyroidism, hypogonadotropic hypogonadism, and secondary adrenal insufficiency; pituitary MRI was normal. Hormone replacement was initiated and the patient remained disease-free at 1-year follow-up. Case 2, a 74-year-old woman on maintenance pembrolizumab after trimodality therapy for esophageal adenosquamous carcinoma, presented with fulminant insulin-deficient diabetes and diabetic ketoacidosis (DKA); C-peptide was nearly undetectable. She required prolonged subcutaneous insulin and permanent discontinuation of immunotherapy. Case 3, a 68-year-old woman receiving dostarlimab with chemotherapy for metastatic high-grade serous endometrial carcinoma, developed an extensive maculopapular eruption after the second dose. Skin biopsy showed full-thickness epidermal necrosis; high-dose systemic corticosteroids produced gradual resolution over 2–3 months, and dostarlimab was permanently discontinued.
Conclusions: ICI therapy can produce rare, severe, and organ-specific irAEs—including hypophysitis, fulminant diabetes with DKA, and extensive cutaneous eruptions—that may occur after variable latency and can mimic other acute medical conditions. Awareness of these presentations, coupled with a low threshold for endocrine and dermatologic evaluation, allows earlier diagnosis and may reduce irAE-associated morbidity.
Keywords: immune checkpoint inhibitors; immune-related adverse events; hypophysitis; fulminant diabetes; diabetic ketoacidosis; cutaneous toxicity; pembrolizumab; dostarlimab
Introduction
Immune checkpoint inhibitors (ICIs) targeting programmed death-1 (PD-1), programmed death-ligand 1 (PD-L1), and cytotoxic T-lymphocyte–associated protein 4 (CTLA-4) have become standard therapy across a broad range of malignancies. By releasing physiologic brakes on T-cell activation, these agents restore antitumor immunity but can also break peripheral tolerance to self-antigens, producing immune-related adverse events (irAEs).
irAEs are broadly classified as infusion reactions or as immune-mediated toxicities affecting specific organs, with the skin, gastrointestinal tract, endocrine glands, liver, and lungs most commonly involved. Less frequent presentations—including hypophysitis, fulminant insulin-dependent diabetes, and severe cutaneous reactions—are individually rare but carry substantial morbidity and, without timely recognition, potential mortality. Population-level data indicate that up to 80% of patients receiving ICI therapy experience some form of irAE, that approximately 35% require systemic corticosteroids, and that approximately 20% discontinue treatment because of toxicity.¹ Biomarker research has focused predominantly on predicting efficacy (for example, through PD-L1 expression, tumor mutational burden, and mismatch-repair status),²,³ and comparatively little is established about which patients are at risk for autoimmune toxicity or when it will occur; susceptibility appears to relate more to host immune characteristics than to tumor biology.
Table 1 summarizes representative irAEs by organ system and specialty. We describe three patients treated at our center who illustrate uncommon, high-acuity irAE phenotypes: ICI-induced hypophysitis, fulminant diabetic ketoacidosis, and extensive cutaneous toxicity. Our objective is to characterize the clinical, biochemical, and radiologic features of these events to assist clinicians in earlier recognition and management.
Methods
Study Design and Patients
We conducted a retrospective, descriptive case series of three adult patients with solid-organ malignancies who were treated with PD-1– or PD-L1–directed immune checkpoint inhibitor therapy, with or without concurrent cytotoxic chemotherapy, at a single tertiary oncology referral center. Patients were selected because each developed a severe, organ-specific immune-related adverse event during the course of treatment. Clinical history, tumor histology and staging, treatment regimen, and the temporal relationship between ICI exposure and toxicity onset were recorded for each patient.
Clinical and Laboratory Evaluation
For each case, relevant laboratory investigations, cross-sectional or magnetic resonance imaging, and, where applicable, autoimmune or endocrine panels were obtained as part of routine clinical care to characterize the suspected irAE and to exclude alternative etiologies, including infection, tumor progression, and non-immune metabolic derangement.
Treatment
Immune checkpoint inhibitor regimens, concomitant systemic therapies, and supportive interventions administered for management of the acute irAE—including corticosteroids, hormone replacement, insulin therapy, and critical care support, as applicable—were documented from the medical record.
Response Assessment and Endpoints
The primary endpoints of this descriptive series were the clinical, biochemical, and radiologic characterization of each irAE and its short-term clinical course. Because of the case-series design, no formal efficacy or survival endpoints were assessed.
Statistical Analysis
Given the descriptive nature of this three-patient case series, no formal statistical hypothesis testing was performed. Findings are reported descriptively, with individual values and reference ranges, and are contextualized against incidence and outcome data from the published literature.
Ethical Considerations
This case series was conducted in accordance with the Declaration of Helsinki. As a retrospective report of three patients, formal institutional ethics committee review was not required, consistent with common practice for case-based publications of this size. Written informed consent for publication of clinical details and clinical photographs was obtained from all patients or their legal representatives.
Case
Case 1: ICI-Induced Hypophysitis after Adjuvant Pembrolizumab
A 54-year-old man with hypertension underwent left radical nephrectomy with para-aortic lymph-node dissection for clear-cell renal-cell carcinoma (grade 4, pT2aN0, 10 cm, with a rhabdoid component). He subsequently received adjuvant pembrolizumab (200 mg every 3 weeks); positron-emission tomography–computed tomography (PET-CT) after 9 cycles (December 27, 2024) showed a complete metabolic response. After the 13th cycle, he presented with headache, nausea, non-projectile vomiting, and fatigue. Laboratory evaluation revealed severe hyponatremia (sodium 109 mmol/L; reference range 136–145) with a potassium of 4.83 mmol/L and calcium of 9.1 mg/dL; blood counts and renal and hepatic parameters were otherwise unremarkable (hemoglobin 15.4 g/dL; total leukocyte count 3.51×10?/L; platelets 241×10?/L; absolute neutrophil count 2.99×10?/L; albumin 4.2 g/dL; creatinine 1.37 mg/dL). Endocrine testing showed thyrotropin of 0.386 μIU/mL with free triiodothyronine of 2.37 pg/mL and free thyroxine of 0.84 ng/dL; total and free testosterone were low at 84.90 ng/dL and 2.40 pg/mL, respectively, with luteinizing hormone of 4.98 mIU/mL and follicle-stimulating hormone of 8.57 mIU/mL. Adrenocorticotropic hormone (ACTH) was low at 2.4 pg/mL (reference range 7.4–64.3), with an inappropriately low cortisol of 1.39 μg/dL (reference range 5.0–25.0). Urine sodium was less than 7 mmol/L, urine osmolality was 84 mOsmol/kg, and serum osmolality was 251 mOsmol/kg. Contrast-enhanced MRI of the brain (March 18, 2025) showed a small T2/FLAIR-hyperintense focus in the left pons consistent with a chronic ischemic focus; minimal bilateral periventricular white-matter hyperintensities consistent with chronic small-vessel ischemic disease; age-appropriate diffuse cerebral atrophy; and FLAIR/DIR hyperintensity involving the intraconal and extraconal left optic nerve. Dedicated pituitary imaging showed a normal-appearing pituitary gland and infundibulum, without hypothalamic abnormality. The combination of central hypothyroidism, hypogonadotropic hypogonadism, and secondary adrenal insufficiency, in the absence of a structural pituitary lesion, supported a diagnosis of ICI-induced hypophysitis. Further immunotherapy was discontinued and hormone replacement therapy, including steroids, was started. The patient remained disease-free at 1-year follow-up imaging.
Case 2: Fulminant Insulin-Deficient Diabetes with Ketoacidosis after Pembrolizumab
A 74-year-old woman with mid-esophageal carcinoma (cT3N+M0) received 6 cycles of neoadjuvant CAPOX plus pembrolizumab, followed by Ivor Lewis esophagectomy; pathology showed adenosquamous carcinoma (ypT2N0) with PD-L1 expression of 90%. She continued maintenance pembrolizumab and subsequently presented to the emergency department with a 1-day history of dyspnea and altered sensorium. On initial assessment she was hypotensive (blood pressure 80/60 mm Hg), tachycardic (pulse 120/min), tachypneic (respiratory rate 36/min), and hypoxemic (oxygen saturation 88%), with a random blood glucose of 502 mg/dL and severe ketonuria. Other blood parametres are listed in table 2 and autoimmune diabetes panel in table 3. She was resuscitated with intravenous fluids and vasopressor support and received empirical broad-spectrum antibiotics and corticosteroids, with subsequent transfer to the intensive care unit for persistent hypotension and suspected pembrolizumab-induced diabetic ketoacidosis (DKA). Non-contrast computed tomography of the head showed age-related atrophic changes with chronic small-vessel ischemic white-matter changes, without acute intracranial pathology. She was treated with an intravenous insulin infusion and supportive care, with de-escalation of antibiotics and vasopressors as renal function and hemodynamics improved, and resumption of oral intake. Subsequently the patient was converted to subcutaneous short-acting insulin with adequate glycemic control. Short-acting insulin according to a sliding scale was continued for 4 months after discharge, and immunotherapy was permanently discontinued. These findings were consistent with fulminant, ICI-induced, insulin-dependent (type 1–like) diabetes mellitus presenting as DKA.
Case 3: Extensive Cutaneous Toxicity after Dostarlimab
A 68-year-old woman with metastatic high-grade serous carcinoma of the endometrium with peritoneal deposits (mismatch-repair–deficient; estrogen receptor positive 80%; progesterone receptor positive 30%; HER2 score 0; CA-125 49.7 U/mL) started weekly nab-paclitaxel and carboplatin with dostarlimab every 3 weeks. After the second dose of dostarlimab, she developed an erythematous eruption involving the palms and soles that progressively extended to the back and chest (Figure 1). Laboratory evaluation at the time of the eruption showed mild anemia (hemoglobin 9.2 g/dL), a leukocyte count of 5.99×10³/μL with 80% neutrophils, and a platelet count of 140×10³/μL. Renal function was preserved. Liver-associated enzymes were mildly elevated (aspartate aminotransferase 60 U/L; alanine aminotransferase 72 U/L; gamma-glutamyl transferase 80 U/L; alkaline phosphatase 134 U/L), without other evidence of hepatic irAE. Skin biopsy from the lesion showed full-thickness epidermal necrosis, subepidermal blistering, and minimal inflammatory infiltrate (keratinocyte apoptosis). Direct immunofluorescence testing was negative, which ruled out autoimmune blistering disease. High-dose systemic corticosteroids were initiated and then tapered gradually according to the response of the eruptions. Topical steroid cream was also applied. Systemic corticosteroids were gradually tapered on discharge, and over 2–3 months the skin eruptions responded. Dostarlimab was permanently discontinued. The CTCAE grade of the eruption was not formally assigned in the source record; the requirement for systemic corticosteroids and the extent of involvement are consistent with at least grade 3 toxicity.
Figure 1. Cutaneous immune-related adverse event in Case 3 following dostarlimab. Clinical photographs obtained after the second dose of dostarlimab show an erythematous, maculopapular eruption. Panel A: dorsal and palmar surfaces of the hands. Panel B: trunk and back, showing widespread, partially confluent erythema. Panel C: dorsal feet and soles. Panel D: additional view of the trunk demonstrating the extent of cutaneous involvement. The distribution and progression—beginning acrally (palms and soles) and extending centripetally to the trunk—are consistent with an ICI-associated maculopapular eruption. Skin biopsy demonstrated full-thickness epidermal necrosis and subepidermal blistering with minimal inflammatory infiltrate; direct immunofluorescence was negative. (Abbreviation: ICI, immune checkpoint inhibitor.)
Results
Case 1: ICI-Induced Hypophysitis after Adjuvant Pembrolizumab
A 54-year-old man with hypertension underwent left radical nephrectomy with para-aortic lymph-node dissection for clear-cell renal-cell carcinoma (grade 4, pT2aN0, 10 cm, with a rhabdoid component). He subsequently received adjuvant pembrolizumab (200 mg every 3 weeks); positron-emission tomography–computed tomography (PET-CT) after 9 cycles (December 27, 2024) showed a complete metabolic response. After the 13th cycle, he presented with headache, nausea, non-projectile vomiting, and fatigue.
Laboratory evaluation revealed severe hyponatremia (sodium 109 mmol/L; reference range 136–145) with a potassium of 4.83 mmol/L and calcium of 9.1 mg/dL; blood counts and renal and hepatic parameters were otherwise unremarkable (hemoglobin 15.4 g/dL; total leukocyte count 3.51×10?/L; platelets 241×10?/L; absolute neutrophil count 2.99×10?/L; albumin 4.2 g/dL; creatinine 1.37 mg/dL). Endocrine testing showed thyrotropin of 0.386 μIU/mL with free triiodothyronine of 2.37 pg/mL and free thyroxine of 0.84 ng/dL; total and free testosterone were low at 84.90 ng/dL and 2.40 pg/mL, respectively, with luteinizing hormone of 4.98 mIU/mL and follicle-stimulating hormone of 8.57 mIU/mL. Adrenocorticotropic hormone (ACTH) was low at 2.4 pg/mL (reference range 7.4–64.3), with an inappropriately low cortisol of 1.39 μg/dL (reference range 5.0–25.0). Urine sodium was less than 7 mmol/L, urine osmolality was 84 mOsmol/kg, and serum osmolality was 251 mOsmol/kg.
Contrast-enhanced MRI of the brain (March 18, 2025) showed a small T2/FLAIR-hyperintense focus in the left pons consistent with a chronic ischemic focus; minimal bilateral periventricular white-matter hyperintensities consistent with chronic small-vessel ischemic disease; age-appropriate diffuse cerebral atrophy; and FLAIR/DIR hyperintensity involving the intraconal and extraconal left optic nerve. Dedicated pituitary imaging showed a normal-appearing pituitary gland and infundibulum, without hypothalamic abnormality.
The combination of central hypothyroidism, hypogonadotropic hypogonadism, and secondary adrenal insufficiency, in the absence of a structural pituitary lesion, supported a diagnosis of ICI-induced hypophysitis. Further immunotherapy was discontinued and hormone replacement therapy, including steroids, was started. The patient remained disease-free at 1-year follow-up imaging.
Case 2: Fulminant Insulin-Deficient Diabetes with Ketoacidosis after Pembrolizumab
A 74-year-old woman with mid-esophageal carcinoma (cT3N+M0) received 6 cycles of neoadjuvant CAPOX plus pembrolizumab, followed by Ivor Lewis esophagectomy; pathology showed adenosquamous carcinoma (ypT2N0) with PD-L1 expression of 90%. She continued maintenance pembrolizumab and subsequently presented to the emergency department with a 1-day history of dyspnea and altered sensorium. On initial assessment she was hypotensive (blood pressure 80/60 mm Hg), tachycardic (pulse 120/min), tachypneic (respiratory rate 36/min), and hypoxemic (oxygen saturation 88%), with a random blood glucose of 502 mg/dL and severe ketonuria.
She was resuscitated with intravenous fluids and vasopressor support and received empirical broad-spectrum antibiotics and corticosteroids, with subsequent transfer to the intensive care unit for persistent hypotension and suspected pembrolizumab-induced diabetic ketoacidosis (DKA). Non-contrast computed tomography of the head showed age-related atrophic changes with chronic small-vessel ischemic white-matter changes, without acute intracranial pathology.
She was treated with an intravenous insulin infusion and supportive care, with de-escalation of antibiotics and vasopressors as renal function and hemodynamics improved, and resumption of oral intake. Subsequently the patient was converted to subcutaneous short-acting insulin with adequate glycemic control. Short-acting insulin according to a sliding scale was continued for 4 months after discharge, and immunotherapy was permanently discontinued. These findings were consistent with fulminant, ICI-induced, insulin-dependent (type 1–like) diabetes mellitus presenting as DKA.
Case 3: Extensive Cutaneous Toxicity after Dostarlimab
A 68-year-old woman with metastatic high-grade serous carcinoma of the endometrium with peritoneal deposits (mismatch-repair–deficient; estrogen receptor positive 80%; progesterone receptor positive 30%; HER2 score 0; CA-125 49.7 U/mL) started weekly nab-paclitaxel and carboplatin with dostarlimab every 3 weeks. After the second dose of dostarlimab, she developed an erythematous eruption involving the palms and soles that progressively extended to the back and chest (Figure 1).
Laboratory evaluation at the time of the eruption showed mild anemia (hemoglobin 9.2 g/dL), a leukocyte count of 5.99×10³/μL with 80% neutrophils, and a platelet count of 140×10³/μL. Renal function was preserved. Liver-associated enzymes were mildly elevated (aspartate aminotransferase 60 U/L; alanine aminotransferase 72 U/L; gamma-glutamyl transferase 80 U/L; alkaline phosphatase 134 U/L), without other evidence of hepatic irAE. Skin biopsy from the lesion showed full-thickness epidermal necrosis, subepidermal blistering, and minimal inflammatory infiltrate (keratinocyte apoptosis). Direct immunofluorescence testing was negative, which ruled out autoimmune blistering disease. High-dose systemic corticosteroids were initiated and then tapered gradually according to the response of the eruptions. Topical steroid cream was also applied. Systemic corticosteroids were gradually tapered on discharge, and over 2–3 months the skin eruptions responded. Dostarlimab was permanently discontinued. The CTCAE grade of the eruption was not formally assigned in the source record; the requirement for systemic corticosteroids and the extent of involvement are consistent with at least grade 3 toxicity.
Discussion
This case series illustrates three distinct, uncommon, and potentially severe immune-related adverse events associated with immune checkpoint inhibitor therapy: hypophysitis, fulminant diabetic ketoacidosis, and extensive cutaneous toxicity. Together, these cases underscore that irAEs can present at any point during treatment, can mimic unrelated acute illnesses, and require a specific index of suspicion for diagnosis.
ICI-induced hypophysitis occurs most frequently with combined anti-PD-1/anti-CTLA-4 therapy (9–10%), less often with anti-CTLA-4 monotherapy (2–6%), and rarely with anti-PD-1 monotherapy (approximately 1%).? Hypophysitis related to anti-CTLA-4 antibodies typically manifests within the first 3–4 months of therapy, whereas anti-PD-1–associated cases, as in our patient, tend to occur later (median onset approximately 6 months). The pathogenic mechanism of anti-PD-1–induced hypophysitis remains poorly defined, and, as in our patient, pituitary imaging is frequently normal despite biochemical panhypopituitarism—a feature that can delay diagnosis if endocrine testing is not specifically pursued.
The mechanism underlying ICI-associated fulminant diabetes parallels that of hypophysitis: disinhibition of T-cell activity by PD-1/CTLA-4 blockade can unmask autoreactivity against insulin-producing pancreatic β-cells, precipitating rapid, often irreversible insulin deficiency. In a systematic review and meta-analysis of 38 randomized trials comprising 7,551 patients, the overall incidence of clinically significant endocrinopathy among ICI-treated patients was approximately 10%.? In a dedicated review of 42 published cases of ICI-associated type 1 diabetes, diabetic ketoacidosis was the presenting manifestation in the majority of patients (71%), and islet autoantibodies—predominantly glutamic acid decarboxylase (GAD) antibodies—were detectable in roughly half of cases; time to onset was shorter in autoantibody-positive patients (median approximately 5 weeks) than in autoantibody-negative patients (median approximately 9 weeks).? An earlier literature review and case analysis similarly found diabetic ketoacidosis to be the presenting feature in most reported cases.? These observations are consistent with our patient’s fulminant presentation with a near-absent C-peptide and a near-normal glycated hemoglobin, a combination that distinguishes acute ICI-induced β-cell failure from decompensated pre-existing type 2 diabetes.
Cutaneous irAEs are graded according to the Common Terminology Criteria for Adverse Events (CTCAE), ranging from asymptomatic or mild, localized rash without functional impairment (grade 1) to life-threatening presentations such as Stevens-Johnson syndrome or toxic epidermal necrolysis (grade 4) and fatal outcomes (grade 5). Management is graded accordingly: mild disease is managed with topical corticosteroids, emollients, and antihistamines without interrupting ICI therapy; more significant or persistent cutaneous involvement warrants oral corticosteroids and consideration of temporary treatment interruption; and severe (grade 3 or higher) disease requires same-day dermatology consultation, immunosuppressive-dose systemic corticosteroids, and, in refractory cases, adjunctive therapies such as intravenous immunoglobulin or calcineurin inhibitors. Proposed predictive biomarkers for dermatologic toxicity remain investigational and are not yet incorporated into routine practice.?
This series is limited by its small sample size (three patients), its descriptive, retrospective, single-center design, and the absence of standardized longitudinal follow-up data on irAE resolution and long-term hormonal or metabolic outcomes beyond the information available in the medical record.
These cases reinforce that severe irAEs, although individually rare, can affect any organ system, may occur at variable latency after ICI initiation, and can present with life-threatening acuity mimicking sepsis, stroke, or primary metabolic disease. Routine surveillance—including periodic endocrine panels and a low threshold for dermatologic and endocrinologic consultation—may facilitate earlier diagnosis. Further prospective, multicenter studies are needed to define reliable predictive biomarkers and to optimize surveillance strategies for high-severity irAEs.
Conclusion
Immune checkpoint inhibitors can cause rare but severe, organ-specific immune-related adverse events, including hypophysitis, fulminant diabetic ketoacidosis, and extensive cutaneous eruptions. Prompt recognition—supported by a high index of clinical suspicion, targeted endocrine and dermatologic evaluation, and timely initiation of hormone replacement, insulin therapy, or immunosuppression as appropriate—is essential to reduce morbidity and prevent life-threatening complications such as adrenal crisis or ketoacidosis.
Acknowledgments
The authors thank the clinical, nursing, and laboratory staff of the Department of Medical Oncology and Hematology/BMT, Fortis Memorial Research Institute, for their contribution to the care of the patients described in this report.
Author Contributions
Rabha P, Wisely J, Jain A, Bhardwaj N, and Raina V contributed to conception and design, clinical data acquisition, manuscript drafting, and critical revision. All authors reviewed and approved the final manuscript. [Please confirm and adjust individual contributions according to ICMJE criteria before submission.]
Conflicts of Interest / Disclosures
The authors declare no relevant conflicts of interest.
Data Availability Statement
De-identified clinical data supporting the findings of this case series are available from the corresponding author upon reasonable request, in accordance with institutional data-sharing policies.
References