External Validation of the Newly Developed BETA-2 Scoring System for Pancreatic Islet Graft Function Assessment

Background: BETA-2 score using a single fasting blood sample was developed to estimate beta-cell function after islet transplantation (ITx) and was validated internally by a high ITx volume center (Edmonton). The goal was to validate BETA-2 externally, in our center.

Methods: Areas under receiver operating characteristic curves (AUROCs) were obtained to see if beta score or BETA-2 would better detect insulin independence and glucose intolerance.

Results: We analyzed values from 48 mixed meal tolerance tests (MMTTs) in 4 ITx recipients with a long-term follow-up to 140 months (LT group) and from 54 MMTTs in 13 short-term group patients (ST group). AUROC for no need for insulin support was 0.776 (95% confidence interval [CI] 0.539-1, P = .02) and 0.922 (95% CI 0.848-0.996, P < .001) for beta score and 0.79 (95% CI 0.596-0.983, P = .003) and 0.941 (95% CI 0.86-1, P < .001) for BETA-2, in LT and ST groups, respectively, and did not differ significantly. In LT group BETA-2 score ≥ 13.03 predicted no need for insulin supplementation with sensitivity of 98%, specificity of 50%, positive predictive value (PPV) of 93%, and negative predictive value (NPV) of 75%. In ST group the optimal cutoff was ≥13.63 with sensitivity of 92% and specificity, PPV, and NPV 82% to 95%. For the detection of glucose intolerance BETA-2 cutoffs were <19.43 in LT group and <17.23 in ST group with sensitivity > 76% and specificity, PPV, and NPV > 80% in both groups.

Conclusion: BETA-2 score was successfully validated externally and is a practical tool allowing for frequent and reliable assessments of islet graft function based on a single fasting blood sample.