Endocrine Abstracts (2017) 51 OC7.1 | DOI: 10.1530/endoabs.51.OC7.1

JUMP: Maternal family history of diabetes and non-white ethnicity adversely affects beta cell response in young people with Type 2 Diabetes

Renuka Dias1,2, Kayleigh Aston2, gianluca tornese3, David Dunger4, Jerry Wales5, Julian Hamilton-Shield6, Anthony Barnett7 & Timothy Barrett2


1Birmingham Women and Children’s Hospital, Birmingham, UK; 2University of Birmingham, Birmingham, UK; 3University of Trieste, Trieste, Italy; 4Addenbrooke’s Hospital, Cambridge, UK; 5Lady Cilento Children’s Hospital, Brisbane, Australia; 6University of Bristol, Bristol, UK; 7Heart of England NHS Trust, Birmingham, UK.


Background: Type 2 diabetes (T2D) in children and adults continues to rise. The relative contribution of pancreatic beta cell failure is not established. We aimed to analyse beta cell function in a subgroup of the UK T2D (JUMP) cohort.

Methods: Participants were recruited at 58 hospital centres across England and Wales (n=204). Patients were 5–18 years, diagnosed with T2D (ADA criteria) and a BMI SDS>+1.0. At baseline, demographics, clinical and medical history, biochemistry and anthropometric data were collected (n=168). A subgroup of the cohort underwent standard short OGTT (n=39) with glucose, C-peptide and insulin levels taken at 0 and 30 minutes for calculation of HOMA-IR, HOMA-%S, HOMA-%B, Insulinogenic Index (IGI) and Disposition index (DI) using the Web-based HOMA calculator (https://www.dtu.ox.ac.uk/homacalculator). Modelling was applied to the linear regression analyses adjusting for demographics, growth or disease status.

Results: Previously baseline demographics of the JUMP cohort have been reported (BSPED 2015). There is a predominance of females (73.8%), mean age at diagnosis of 13.2 years and mean BMI SDS +2.85. The majority of CYP were non-white (57.8%). Subgroup analysis of the short OGTT data showed no significant difference versus the overall cohort (White population 60%, age at diagnosis 12.9 years, BMI SDS at diagnosis +2.9). Family history (most significantly in diabetes in mothers versus fathers, P<0.05) was strongly associated with reduced DI. This was consistent also with IGI and HOMA-%S. Longer duration of disease, non-white ethnicity, older age at diagnosis and post-pubertal status were consistently associated with a decreased DI. Greater BMI and taller stature may be associated with decreased DI but only achieved statistical significance in single models. Sex and medications were not associated with DI.

Conclusions: The JUMP cohort represents a UK national prospective cohort study of CYP with T2D. Non-white ethnicity, longer duration of disease have been reported in association with reduced DI (and inadequate beta cell compensation in response to reduced insulin sensitivity). Our novel finding is that a maternal FH of diabetes significantly increases the risk of a low DI. Further follow-up will establish whether this is a risk factor for earlier development of complications.