Identifying Pediatric A-beta plus Ketosis-Prone Diabetes (KPD)

KPD is characterized by diabetic ketoacidosis (DKA) in persons lacking the phenotype of autoimmune type 1 diabetes. The validated "Aβ" system classifies 4 KPD subgroups in adults, one of which (A-β+ KPD) is an atypical form that otherwise resembles type 2 diabetes (T2D) . Although recognized in children, better understanding of the prevalence and characteristics of pediatric A-β+ KPD is needed. The Rare and Atypical Diabetes Network (RADIANT) has modified the criteria for A-β+ KPD to: DKA within 6 months of diagnosis (dx) , not on SGLT2i, off insulin within 6 months of DKA with HbA1c <7% on oral agents alone, and negative islet autoantibodies. We compared clinical characteristics of pediatric T2D with and without DKA at dx, and in the former group estimated the prevalence of A-β+ KPD. A retrospective chart review identified 716 children with T2D over 3 years at Texas Children’s Hospital. Mean age at dx was 13.7 ± 2.4 yr; 63% females; 59% Hispanic, 29% African American (AA) , 9% non-Hispanic white, 3% others. Of them, 57 (8%) presented with DKA with negative antibodies. Children with DKA at dx, vs. those without it, were more often AA (45% vs. 28%, p=0.008) , male (60% vs. 35%, p=<0.001) and older (14.4 vs. 13.6 y/o, p=0.02) ; had lower random C-peptide (1.4 vs. 4.23 ng/mL, p<0.001) , higher glucose (375 vs. 240 mg/dL, p<0.001) , higher BMI-z score (2.38 vs. 2.25, p=0.02) and higher HbA1c (12.1 vs. 9.4%, p<0.001) at dx. In a multivariable model with age, sex, race, C-peptide, glucose, HbA1c and BMI z-score at dx, older age (p=0.004) , lower C-peptide (p<0.0001) , higher glucose (p=0.003) and BMI z-score (p=0.004) remained significantly associated with DKA. Eight children with DKA (14%) met all RADIANT criteria for A-β+ KPD. Most of the other children with DKA failed KPD criteria because they remained on insulin treatment for >6 months. Children with T2D and DKA at dx were more likely to be older, had lower C-peptide and higher glucose and BMI-z score at dx. Of them, 14% met all RADIANT A-β+ KPD criteria and are eligible for enrollment in RADIANT. Disclosure E.A.Kubota-mishra: None. M.Tosur: Advisory Panel; Provention Bio, Inc. Radiant study group: n/a. X.C.Huang: None. C.G.Minard: None. M.Astudillo: None. A.K.Refaey: None. G.Montes: None. S.Sisley: Speaker's Bureau; Rhythm Pharmaceuticals, Inc. A.Balasubramanyam: None. M.J.Redondo: Advisory Panel; Provention Bio, Inc.