Diagnostic utility of the insulin-to-C-peptide molar ratio in differentiating insulin autoimmune syndrome and exogenous insulin antibody syndrome from insulinoma.

OBJECTIVE

Although an insulin-to-C-peptide molar ratio greater than 1 has traditionally been considered suggestive of insulin autoimmune syndrome (IAS), its diagnostic value remains controversial, and the optimal cutoff is unclear. This study aimed to evaluate the diagnostic performance of the insulin-to-C-peptide molar ratio for distinguishing IAS and exogenous insulin antibody syndrome (EIAS) from insulinoma and to identify the optimal diagnostic cutoff.

METHODS

We retrospectively collected clinical and biochemical data from 19 patients with IAS, 33 with EIAS, and 70 control patients with pathologically confirmed insulinoma. Receiver-operating-characteristic (ROC) curve analysis was performed separately for IAS versus insulinoma and EIAS versus insulinoma to assess the diagnostic performance of the insulin-to-C-peptide molar ratio in the fasting state and during hypoglycemic episodes.

RESULTS

The insulin-to-C-peptide molar ratio was significantly higher in IAS and EIAS than in insulinoma at both fasting and during hypoglycemic episodes. During hypoglycemic episodes, the area under the ROC curve (AUC) was 0.970 for IAS versus insulinoma and 0.944 for EIAS versus insulinoma, with optimal cutoffs of 0.382 and 0.552, respectively. In the fasting state, the corresponding AUCs were 0.916 and 0.961, with optimal cutoffs of 0.309 and 0.386. Compared with the conventionally used cutoff of 1, the optimal cutoffs substantially improved sensitivity while maintaining high specificity, particularly for IAS. Insulin concentration alone also showed good diagnostic performance for distinguishing IAS from insulinoma, with AUCs of 0.985 and 0.943 and optimal cutoffs of 83.1 and 58.86 μIU/mL during hypoglycemic episodes and in the fasting state, respectively, but was less informative for EIAS, with AUCs of 0.853 and 0.729 and optimal cutoffs of 67.85 and 71.94 μIU/mL, respectively.

CONCLUSION

The insulin-to-C-peptide molar ratio, measured either during hypoglycemic episodes or in the fasting state, showed good diagnostic performance for distinguishing IAS and EIAS from insulinoma. The conventionally used cutoff of 1 appears to be too high, and lower cutoffs may provide better sensitivity while preserving specificity. Validation in larger cohorts is needed.