Short-term treatment with insulin can induce remission of type 2 diabetes (T2DM) in some patients but not others. We thus sought to identify baseline predictors of diabetes remission in response to short-term insulin-based therapy. In this study, adult patients with T2DM of <7 years duration were randomized to 8-weeks treatment with (i) insulin glargine, (ii) glargine + thrice-daily lispro, or (iii) glargine + twice-daily exenatide, followed by 12-weeks washout that enabled assessment of remission (defined as A1c<6.5% after ≥3 months without glucose-lowering therapy). At baseline, 8-weeks, and washout, beta-cell function was assessed with 4 measures: Insulin Secretion-Sensitivity Index-2 (ISSI-2), insulinogenic index/HOMA-IR, ΔCpeptide0-120/Δglucose0-120×Matsuda, and ΔISR0-120/Δgluc0-120×Matsuda. Diabetes remission was achieved in 31 of 90 participants (34.4%). Compared to their peers, those who went on to remission had lower A1c (p<0.001) and better beta-cell function at baseline (all 4 measures p≤0.01). The non-remission and remission groups did not otherwise differ in baseline insulin sensitivity/resistance (Matsuda, HOMA-IR), BMI, duration of diabetes, pre-trial diabetes medications, or allocated insulin-based therapy during the trial. On logistic regression analyses, each baseline measure of beta-cell function emerged as a significant predictor of remission (log ISSI-2: adjusted OR 4.41 (95%CI: 1.71-11.34); log insulinogenic index/HOMA-IR: 2.21 (1.26-3.89); log ΔCpeptide0-120/Δglucose0-120×Matsuda: 1.62 (1.00-2.64); log ΔISR0-120/Δgluc0-120×Matsuda: 1.87 (1.09-3.23). Similarly, higher baseline ISSI-2 tertile predicted longer time to glycemic relapse after cessation of the insulin-based therapy (log-rank P=0.029). In conclusion, beta-cell function is the dominant baseline determinant of the likelihood of achieving remission of diabetes in response to short-term insulin-based therapy.


R.Retnakaran: Other Relationship; Sanofi, Eli Lilly and Company, Research Support; Boehringer Ingelheim/Mount Sinai Hospital, Novo Nordisk. J.Pu: None. A.Emery: None. C.K.Kramer: Research Support; Boehringer Ingelheim Inc. B.Zinman: Consultant; Abbott Diabetes, Eli Lilly and Company, Novo Nordisk A/S, Novo Nordisk Canada Inc., Boehringer Ingelheim Inc., Merck & Co., Inc.


Canadian Institutes of Health Research (MOP136938)

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at