Case Presentation: A 53-year-old male with no history of diabetes was admitted to the hospital due to complaints of severe fatigue, dizziness, and abrupt weight loss. He was found to have a random blood glucose level of 536 and a bicarbonate level of 15. On admission, he had a venous pH of 7.30 and a beta-hydroxybutyrate level of 5.4. He was managed for diabetic ketoacidosis with IV insulin. Workup revealed a low C-peptide level (0.47), indicative of severe endogenous insulin deficiency. His hemoglobin A1C level was 7.8. Islet cell antibody screen was negative. History revealed that he was being managed for metastatic adenocarcinoma of the lung with nivolumab after eventually progressing through first line therapy of carboplatin, pemetrexed, and bevacizumab followed by pemetrexed and bevacizumab maintenance. He had been taking a fixed dose of 240 mg once every two weeks, for approximately one and a half years without any side effects and good disease control. Two months before presentation, his dose had been increased to a fixed dose 480 mg once every four weeks consistent with the March 2018 FDA label update. He was diagnosed with new-onset Type 1 Diabetes Mellitus (T1DM) secondary to immune therapy and has maintained on insulin since.

Discussion: Immune checkpoint inhibitors (ICIs) including PD1 inhibitors such as nivolumab have rapidly become a household name in the world of oncology. Immune-related adverse events (irAE) are a known complication of ICIs. Although severe grade 3-5 irAEs are rare (frequency of typically <1% in phase III trials), they often present as an abrupt onset of autoimmune disorders. Now with ICIs being FDA approved for a wide range of malignancies and being used in a rapidly increasing number of patients, a rise in the incidence of these complications is inevitable. Our patient is the first presented case of an immune-mediated endocrinopathy triggered by escalating the dose of nivolumab.
Nivolumab has recently been FDA-approved for a new, higher (480 mg flat every 4 weeks) dose. Literature shows reported cases of new-onset autoimmune and fulminant diabetes in patients treated with nivolumab. The published experience has generally shown this toxicity to be irreversible and not responsive to either immunosuppression or drug withdrawal. Pre-treatment HLA genotyping has been proposed to identify patients at risk for fulminant diabetes, however, there are no known markers to help identify patients prone to autoimmune diabetes. This makes surveillance for autoimmune diabetes a challenge and patients usually present with DKA. Based on prior data, the median time of T1DM onset, after nivolumab initiation is 2.5 to 4.4 months. Our patient is unique as he tolerated the 240 mg nivolumab dose for one and a half year but developed autoimmune diabetes within two months of starting the higher dose. This presentation has not been reported in the literature before.

Conclusions: Our patient’s unique presentation opens the question as to whether the higher dose may be associated with an increased risk of irAE even in patients previously tolerating every two-week therapy. We advocate for closer monitoring in patients starting treatment with high dose nivolumab, or those transitioning to the higher dose. This should help identify life-threatening complications earlier. Further research is needed to patients at risk of these irAEs.