The multifaceted nature of diabetes mellitus induced by checkpoint inhibitors.

Immune checkpoint inhibitors (CPI) are increasingly being used in oncology, and many autoimmune side effects have been described. Diabetes mellitus (DM) has been reported in approximately 1% of subjects treated with programmed cell death-1 and programmed death ligand 1 (PD-1/PD-L1) inhibitors, alone or in association with CTLA-4 inhibitors. In the present mini-review, we aimed to describe different clinical pictures and pathophysiology associated with these forms of diabetes. Data on CPI-related DM was gathered from the largest case series in the literature and from our centre dedicated to immunotherapy complications (ImmuCare-Hospices Civils de Lyon). Most cases are acute autoimmune insulin-dependent diabetes which are similar to fulminant diabetes (extremely acute onset with concomitant near-normal HbA1c levels). Other cases, however, have a phenotype close to type 2 diabetes or appear as a decompensation of previously known type 2 diabetes. The occurrence of diabetes can also be a complication of autoimmune pancreatitis induced by CPI use. Finally, two cases of diabetes in a context of autoimmune lipoatrophy have recently been described. Regarding the wide variety of CPI-induced diabetes, the discovery of a glucose disorder under CPI should motivate specialised care for aetiological diagnosis and appropriate treatment.

Diagnosis, prediction, and natural course of HIV-1 protease-inhibitor-associated lipodystrophy, hyperlipidaemia, and diabetes mellitus: a cohort study.

BACKGROUND: The prevalence and severity of lipodystrophy syndrome with long-term therapy for HIV-1 infection that includes a protease inhibitor is unknown. We studied the natural course of the syndrome to develop diagnostic criteria and identifying markers that predict its severity. METHODS: We assessed 113 patients who were receiving HIV-1 protease inhibitors (mean 21 months) and 45 HIV-1-infected patients (28 with follow-up) never treated with a protease inhibitor. Lipodystrophy was assessed by questionnaire (including patients' rating of severity), physical examination, and dual-energy x-ray absorptiometry. Body composition and fasting lipid and glycaemic variables were compared with data obtained 8 months previously. Oral glucose tolerance was investigated. FINDINGS: There was 98% concordance between patients' reports of the presence or absence of lipodystrophy (reported by 83% of protease-inhibitor recipients and 4% of treatment-naïve patients; p=0.0001) and physical examination. Patients' ratings of lipodystrophy were significantly associated with declining total body fat (p=0.02). Lower body fat was independently associated with longer duration of protease-inhibitor therapy and lower bodyweight before therapy, and more severe lipodystrophy was associated with higher previous (p < 0.03) and current (p < or = 0.01) triglyceride and C-peptide concentrations, and less peripheral and greater central fat (p=0.005 and 0.09, respectively). Body fat declined a mean 1.2 kg over 8 months in protease-inhibitor recipients (p=0.05). The prevalence of hyperlipidaemia remained stable over time (74% of treated patients vs 28% of naïve patients; p=0.0001). Impaired glucose tolerance occurred in 16% of protease-inhibitor recipients and diabetes mellitus in 7%; in all but three patients these abnormalities were detected on 2 h post-glucose load values. INTERPRETATION: Diagnosis and rating severity of lipodystrophy is aided by the combination of physical examination, patient's rating, and measurement of body fat, fasting triglycerides, and C-peptide. Weight before therapy, fasting triglyceride, and C-peptide concentrations early in therapy, and therapy duration seem to predict lipodystrophy severity. Lipodystrophy was common and progressive after almost 2 years of protease inhibitor therapy, but was not usually severe. Hyperlipidaemia and impaired glucose tolerance were also common.

Limited joint mobility and lipodystrophy in children and adolescents with insulin-dependent diabetes mellitus.

One hundred forty-eight children, aged 11.9 +/- 3.7 years, who had insulin-dependent diabetes mellitus (IDDM) for 4.5 +/- 3.7 years and glycosylated Hb values (HbA1) of 10.5% +/- 4.5%, were examined for limited joint mobility (LJM) and lipodystrophy. In all diabetics, human biosynthetic insulin was used. Six hundred forty-eight pupils, aged 11.8 +/- 2.5 years, served as controls. LJM was found in 28.4% of the diabetics and in 7.5% of the controls (p < 0.001). The presence and severity of LJM was positively correlated with the duration of diabetes and negatively with height. There was no correlation between the presence of LJM and sex, chronological age, age of diabetes onset, HbA1 values, or retinopathy. Lipodystrophy at insulin injection sites was found in 37.1% of the diabetics. Hypertrophic lesions predominated. Our findings and those of the literature lead to the following conclusions: the prevalence of LJM in IDDM subjects of comparable age using Rosenbloom's criteria is almost uniform in all published studies (28%); the correlation, however, of prevalence and severity to different parameters presents discrepancies which cannot be interpreted at present. Human biosynthetic insulin does not protect from lipodystrophy and alterations of insulin injections still remain the only currently available preventive and therapeutic measure.