Background: Allogeneic islet transplantation (IT) is a validated therapy in type 1 diabetes. The mechanisms underlying the decline of islet graft function with time are unclear. We evaluated the distinct relation between primary graft function (PGF) and 5-year IT outcomes.

Methods: This retrospective multi center cohort study enrolled all participants from the Collaborative Islet Transplant Registry, who received IT alone, or after kidney transplantation, between 01/19/1999, and 07/17/2020 with a calculable PGF (exposure of interest), measured 28 days after last islet infusion with a validated composite index of islet graft function (Beta2-score). Primary outcome was cumulative incidence of unsuccessful IT, defined as HbA1c ≥ 7.0% (53 mmol/mol) and/or severe hypoglycemia and/or fasting C-peptide < 0.2 ng/mL. Secondary outcomes were: 1) graft exhaustion (fasting C-peptide < 0.3 ng/mL), 2) inadequate glucose control (HbA1c ≥ 7.0% (53 mmol/mol) and/or severe hypoglycemia), and 3) requirement for exogenous insulin therapy (≥14 consecutive days). Relations between PGF and IT outcomes were explored with a competing risk analysis adjusted for all covariates suspected or known to impact outcomes. A predictive model based on PGF was built and internally validated by using bootstraps resampling method.

Results: In 39 centers, 1210 patients (712 (59.5%) females, mean age 47 years (SD 11) received a median of 10.8 thousand islet-equivalents per kg of bodyweight (IQR 7.4-13.5). Among them, 211 (17.6%) were islet after kidney recipient; 452 (37.4%) received a single islet infusion and 758 (62.6%) received multiple islet infusions. Mean PGF was 14.3 (SD 8.8). The 5-year cumulative incidence of unsuccessful IT was 70.7% (95%CI 67.2-73.9) (Figure 1), and was inversely and linearly related to PGF with adjusted subhazard ratio (sHR) of 0.77 (95% CI 0.72-0.82) per 5 units increase of Beta2-score (p<0.0001) (Figure 2). Secondary endpoints were similarly related to PGF. PGF predicted the 5-year cumulative incidences of unsuccessful IT, graft exhaustion, inadequate glucose control, and exogenous insulin therapy with a good accuracy with median C-statistic values of 0.70 (range 0.69-0.71); 0.76 (range 0.74-0.77); 0.65 (range 0.64-0.66); 0.72 (range 0.71-0.73), respectively. Finally, the prediction models developed from the present study were integrated into a software program allowing to display the prediction of cumulative incidences and median survival of the four IT study outcomes (https://lille-model.shinyapps.io/PGF-islet/).

Conclusion: This global multiple center study demonstrated a linear and independent relation between PGF and 5-year clinical outcomes of IT. The main study limitations are its retrospective design and the absence of analysis of complications.
Part of the results were presented at IPITA congress 2021, EPITA Congress 2023, The present study is accepted for publication in the Lancet Diabetes & Endocrinology