In a prospective, randomized case-controlled study, the investigators hope to demonstrate a positive correlation of plasma levels of EPA and DHA as well as fish intake with the HbA1c-lowering effect of sitagliptin but not with the active comparator glimepiride.

<Background and Rationale>

DPP-4 inhibitors (DPP-4i) are becoming 1st line oral anti-diabetic drugs (OAD) in the management of type 2 diabetes in Japan; approximately 3-million patients, one third of those with diagnosis of diabetes, receive DPP-4i. The widespread use of DPP-4i in Japan is partly due to accumulating data from clinical trials demonstrating that DPP-4 inhibitors exert greater HbA1c-lowering effects in Japanese type 2 diabetes patients and also in non-obese Asian type 2 diabetes patients when compared to diabetes patients of other ethnicities (Ann Pharmacother 2012;46:1453-69; Diabetologia 2013;56:696-708).

We previously demonstrated that sources of diet such as fish can enhance the HbA1c-lowering effects of DPP-4i (J Diabet Investig 2012:3(5):464-467). In our retrospective studies, we also reported that DPP-4i treatments are more effective in type 2 diabetic patients with high plasma levels of EPA and DHA, polyunsaturated fatty acids that are rich in fish (J Diabet Investig 2012:3(5):464-467; J Diabet Investig 2012:3(6):498-502). In the course of analyzing the mechanism, we found that ingestion of fish prior to rice enhanced secretion of GLP-1 (Yabe D et al. Late breaking poster presentation in ADA 2014, manuscript in preparation). Further investigation of the ideal dietary content for combination with DPP-4i could be helpful in enhancing DPP-4i efficacy for more patients.

This prospective clinical trial studying food composition in addition to compliance of diet therapy in Japanese patients who received sitagliptin or active comparator glimepiride should provide new insights on cross-talk between diet and medicines

<Study Procedures>

1. Initial screening period

IC, screening assessment, eligibility check except for inclusion criteria (visit 1-4 week).

1. Hospital visit at fasting.

2. Patient's informed consent The investigator will hand over the written information to the patients and give an explanation of the contents of the study. The patients should be provided with ample time and opportunity to inquire about details of the study, and after confirmation that the patients fully understood the contents of the study, the informed consent form to participation in the study should be signed. (The informed consent form should describe the explanatory considerations specified in Ethical Guidelines for Clinical Studies, Part IV: Informed consent <detailed rule>) Prior to a patient's participation in the study, the written informed consent form should be signed or placed the name/seal, and personally dated by both the investigator and the patient. When CRC performed a supplemental explanation, he/she should also sign or place the name/seal, and date the informed consent form. The investigator will distribute a copy of the signed informed consent form and written information to the patient.

   After acquisition of informed consent to participation in the study until the end of the final observation, if information becomes available that may be relevant to the patient's willingness to continue participation in the study, the patient will be informed in a timely manner to confirm as to whether the patient is still willing to participate in the study. This communication should be recorded in writing. At this time, if the investigator considers that the written information should be revised, the written information should be promptly revised.

   The patients may withdraw from the study at any time. If a patient has withdrawn consent, the investigator should record the reasons for withdrawal in CRF.

3. Make an interim registration for a patient. A unique number (screening or baseline number) should be assigned for identification purposes. It should be noted that that under no circumstances should a patient be assigned more than one allocation number

4. Check if a patient satisfies required eligibility. Inquire about the background of a patient, past medical history, a history of surgery, and prior treatment drugs, and perform physical exam, vital signs (incl. ECG), hematology, blood biochemistry, pregnancy test (for women of childbearing potential), HbA1c, and fasting plasma glucose (FPG) as well as plasma EPA/DHA, FFQ and metabolomics.

5. Washout periods until Visit 2 are as shown below:

   • Patients who are not treated with an oral anti-diabetic agent: None

   • Patients who are on a single oral anti-diabetic agent: 4 weeks

     2. Baseline period

Physical and laboratory checkup, randomization and initiation of treatment (visit2, 0 week).

1. Hospital admission of patients at fasting.

2. In accordance with the study schedule, perform physical exam, vital signs (incl. ECG) and pregnancy test (for women of childbearing potential) should be determined. Concomitant drugs should be checked.

3. Confirm that there is no abnormal finding, and randomize the patient [Randomization procedure] 1) The following requirements should be sent to secretariat of this study: Patient identification code Satisfying the eligibility criteria

   Assignments should be performed by dynamic allocation by the following factors:

   Gender Age BMI Plasma EPA

4. When the person in charge of registration assessed the entry of a patient is appropriate, a study drug to be administered will be designated.

5. Initiate a study drug at following doses (visit 2-4, 0-16 weeks):

Sitagliptin 50mg Glimepiride 0.5mg

3. Treatment period

Physical and laboratory checkup (visit3, 8 week; visit4, 16 week).

1. Hospital admission of patients at fasting.
2. Check the treatment compliance status, and the presence or absence of adverse event/hypoglycemic symptoms (for the glimepiride treatment group).
3. In accordance with the study schedule, perform physical exam, vital signs (incl. ECG) and pregnancy test (for women of childbearing potential) are determined. Concomitant drugs should be checked. HbA1c, and fasting plasma glucose (FPG) as well as plasma EPA/DHA, FFQ and metabolomics should be also checked.

3-1. Study configuration (parallel, crossover, etc.), Multicenter, randomized, prospective, open-label, 2 arm parallel, clinical trial.

(2 centers, see appendix) 3-2. Duration of the treatment period 16 weeks 3-3. Control group Glimepiride 3-4. Follow-up procedures A patient was to be withdrawn from the trial if

1. The patient withdrew consent, without the need to justify the decision.
2. The patient had to take any concomitant drugs interfering with the study medication.
3. The patient is no longer able to participate in the study for other medical reasons (e.g., surgery, AEs, or other diseases).
4. Occurrence of unacceptable hyperglycemia or hypoglycemia that may put patients at risk through continued participation.
5. The patient is not able to comply with study protocol (according to the discretion of the investigator or sub-investigator).

<Statistical Analysis and Sample Size Justification>

To evaluate the association of changes in HbA1c levels from baseline with plasma DHA/EPA, linear regression analyses are carried out at week16 in each treatment group. We construct multivariable logistic regression models to assess whether plasma DHA/EPA is independently associated with HbA1c reduction. We calculate Spearman correlation coefficients between changes in HbA1c and the following variables: compliance with diet therapy, estimated intakes of various food items including fish, and the clinical characteristics of the patients (age, sex, body mass index, duration of diabetes, baseline HbA1c) using IBM SPSS Statistics 22 (SAS Institute Inc., Cary, NC, USA).

Sample size:

In our retrospective analysis, estimated intakes of fish or DHA/EPA (n=72) as well as serum levels of DHA/EPA (n=20) are significantly correlated with HbA1c reduction by DPP-4 inhibitors in drug-naïve T2DM patients (J Diabet Investig 2012:3(5):464-7). In another retrospective analysis, serum levels of EPA are significantly correlated with HbA1c reduction by DPP-4 inhibitors in T2DM patients who received DPP-4 inhibitors as monotherapy (n=16) or add-on therapy (n=46) (J Diabet Investig 2012: 3(6):498-502).These studies support that the proposed study requires a sample size of 24 patients in each treatment group considering drop outs.

<Specific Drug Supply Requirements >

Both sitagliptin and glimepiride will be prescribed according to national health insurance. The patients who are eligible for this study will be randomized to treatment with sitagliptin or glimepiride at 0 week. Patients receiving an OAD at enrollment will be subjected to washout prior to randomization at 0 week. The dose for each treatment group is as follows:

Sitagliptin: Orally, 50mg/day (q.d) throughout the treatment period Glimepiride: Orally, 0.5mg/day (q.d) throughout the treatment period

<Adverse Experience Reporting>

An adverse event (AE) is any unfavourable and unintended sign (including an abnormal laboratory finding, for example), symptom, or disease temporally associated with the use of a medicinal product, whether or not considered related to the medicinal product. AEs will be reported to sponsor through the procedure similar to the marketed product.

Inclusion Criteria:

• Japanese T2DM patients, who are OAD naïve or receive 4-week wash out of one OAD (glinides, alpha-GI, metformin) before randomization.
• HbA1c:6.0-8.0 %
• BMI:18-30 kg/m2
• Age:20-75 years old
• CrCl: > 60mL/min

Exclusion Criteria:

• Patients on DPP-4 inhibitors or glimepiride
• Patients treated with pioglitazone
• Patients with moderate/severe renal impairments
• Patients with insulin or GLP-1 receptor agonists

16 weeks