In this retrospective study, we showed that the predictive factors of dipeptidyl peptidase-4 (DPP-4) inhibitors lowering the glycosylated hemoglobin (HbA1c) level after 12 months in patients with type 2 diabetes mellitus included a decrease in the HbA1c level after 3 months of treatment, high baseline HbA1c level, low baseline body mass index, and the absence of coronary artery disease, suggesting that the most suitable candidates for treatment with DPP-4 inhibitors are diabetics who are not obese and do not have coronary artery disease. In addition, the long-term efficacy of DPP-4 inhibitors can be predicted by decrements of HbA1c after 3 months of treatment.

First, in our study, the physicians evaluated patients from multiple aspects, including physical activity, cognitive function, and stability of cardiovascular disease, considering the risk of hypoglycemia, after which they set the individual treatment goals and prescribed DPP-4 inhibitors. A majority of patients had already been treated with other anti-diabetic drugs, including α-glucosidase inhibitors, sulfonylureas, biguanides, glinides, and thiazolidinediones. DPP-4 inhibitors are associated with a relatively low risk of hypoglycemia and are weight-neutral [23], whereas dose increments of the baseline drugs are associated with various side effects, especially for old patients (e.g., diarrhea for α-glucosidase inhibitors, weight gain for sulfonylureas, and thiazolidinediones). Therefore, for these patients, additional DPP-4 inhibitor administration instead of dose increments of the baseline drugs was chosen, expecting further glycemic effects with less side effects.

Second, there may be substantial synergistic or heterogeneous responses to combination therapy with other anti-diabetic drugs [4]. We reanalyzed the data to elucidate this issue. Stepwise multiple regression analysis showed that co-administration of glinides or sulfonylureas, which are insulin secretagogues, with DPP-4 inhibitors affected the magnitude of decrease in the HbA1c level after 12 months; however, co-administration of α-glucosidase inhibitors, biguanides, or thiazolidinediones did not lead to any statistically significant effects, indicating that combination therapy with stimulators of insulin secretion and DPP-4 inhibitors shows more synergistic effects on lowering HbA1c.

Third, a novel finding of this study was that the absence of coronary artery disease itself was one of the predictors of DPP-4 inhibitor efficacy. As several studies have shown, shorter history of diabetes may be a predictor of greater reduction in HbA1c after treatment with DPP-4 inhibitors [5]; however, in our retrospective study, we were unfortunately not able to ob tain reliable data concerning the duration of diabetes. Further prospective studies are needed to clarify the effects of DPP-4 inhibitors and the duration of diabetes.

Lastly, as Dr. Kim pointed out, identifying the predictive factors with reference to the baseline HbA1c could be interesting when considering the individual responses to DPP-4 inhibitors. The baseline HbA1c may be regulated by multiple factors, including lifestyle-related factors and the administered drugs. Due to the small sample size, we were not able to identify the predictive factors with reference to the baseline HbA1c, and further studies are necessary to elucidate such predictive factors.

We appreciate Dr. Kim for the interest in our study and for the valuable comments and suggestions.