Objective: Checkpoint inhibitor-associated autoimmune diabetes mellitus (CIADM), a variant of type 1 diabetes, is a rare immune-related adverse events (irAEs) caused by antibody-based immune checkpoint inhibitors. CIADM typically manifests as fulminant or acute-onset type 1 diabetes in the insulin-depleted state. However, we encountered a patient with slow-onset CIADM who initially presented with hyperglycemia without decreased insulin secretion after treatment with durvalumab (an anti-PD-L1 antibody).Patient: A 60-year-old man diagnosed with small-cell lung cancer on durvalumab combined with dexamethasone treatment experienced an increase in glycated hemoglobin (HbA1c) from 6.4% to 7.8% after three cycles.Results: Despite preserved endogenous insulin secretion (C-peptide, 2.47 ng/mL with a casual plasma glucose level of 287 mg/dL), basal insulin therapy was initiated considering CIADM. HbA1c levels remained stable (8.5–9.2%) for 3 months but increased to 13.4% at 18 weeks, indicative of CIADM. Declining endogenous insulin secretion resulted in ketosis; however, hyperglycemic crisis was prevented through basal insulin therapy.Conclusion: We emphasize that CIADM develops gradually and does not always occur in the course of fulminant or acute-onset type 1 diabetes; therefore, early initiation of insulin in the presence of hyperglycemia is crucial to prevent hyperglycemic crises.

Background: No consensus exists regarding which anthropometric measurements are related to bone mineral density (BMD), and this relationship may vary according to sex and age. A large Japanese cohort was analyzed to provide an understanding of the relationship between BMD and anthropometry while adjusting for known confounding factors. Methods: Our cohort included 10,827 participants who underwent multiple medical checkups including distal forearm BMD scans. Participants were stratified into four groups according to age (≥50 years or <50 years) and sex. The BMD values were adjusted for confounding factors, after which single and partial correlation analyses were performed. The prevalence of osteopenia was plotted for each weight index (weight or body mass index [BMI]) class. Results: Cross-sectional studies revealed that weight was more favorably correlated than BMI in the older group (R=0.278 and 0.212 in men and R=0.304 and 0.220 in women, respectively), whereas weight and BMI were weakly correlated in the younger age groups. The prevalence of osteopenia exhibited a negative linear relationship with weight among older women ≥50 years of age, and an accelerated increase was observed with decreasing weight in older men weighing <50 kg and younger women weighing <60 kg. When weight was replaced with BMI, the prevalence was low in most subgroups classified by weight. Conclusions Weight, rather than BMI, was the most important indicator of osteopenia but it might not be predictive of future bone loss.