The purpose of this study is to evaluate alpha- and beta-cell function during combination treatment with saxagliptin in addition to dapagliflozin and metformin compared to placebo in addition to dapagliflozin and metformin in subjects with T2DM on stable metformin background therapy.

List of Abbreviations AE Adverse event ALT Alanine aminotransferase ANOVA Analysis of variance APTT Activated partial thromboplastin time AST Aspartate aminotransferase AUCins Area under the serum insulin concentration time curve BMI Body mass index CRF Case report form CTA Clinical trial application ECG Electrocardiogram FSFV First subject first visit GCP Good clinical practice GIR Glucose infusion rate GIRmax Maximum glucose infusion rate HbA1C N-(1-deoxy)-fructosyl-haemoglobin HBsAg Hepatitis B surface antigen HIV Human immunodeficiency virus ICH International conference on harmonisation IEC Independent ethics committee INR International normalised ratio IRB Institutional review board IU International unit i.v. Intravenous(ly) LSFV Last subject first visit LSLV Last subject last visit MedDRA Medical Dictionary of Regulatory Activities OAD Oral antidiabetic drug q.d. daily SAE Serious adverse event SAP Statistical analysis plan s.c. Subcutaneous(ly) SMPG self-measured plasma glucose SOP Standard operating procedure WHO-DDE World Health Organization Drug Dictionary Enhanced

Under physiological conditions blood glucose is kept within a narrow range by complex interactions of several signalling pathways. In this context, fine-tuning of alpha- and beta cell activity is fundamental to avoid excessive metabolic excursions. The pathogenesis of type 2 diabetes mellitus (T2DM) is driven by insulin resistance, followed by an increasing imbalance between alpha and beta cell activity which is characterised by relative insulin deficiency and increased glucagon secretion especially in the postprandial state.

Individual arrangement of complementary antidiabetic drugs in the escalation of pharmacological intervention in T2DM is a major challenge. There is substantial need to provide a scientific rationale for the best effective combination of antidiabetic drugs with regard to their potency to address different aspects in the pathophysiology of T2DM. Therefore, studies evaluating potential synergistic and/or complementary effects of pharmacological interventions in T2DM deem imperative.

This study aims to evaluate the effect of the DPP-IV inhibitor saxagliptin (as compared to placebo) in addition to the SGLT-2 inhibitor dapagliflozin on insulin resistance, pancreatic alpha and beta cell function.

Triple therapy with metformin, SGLT-2 inhibitors, and DPP-IV inhibitors was shown to be efficacious and safe in several studies and has been approved for the treatment of T2DM by the European Medical Agency. Therefore, subjects with T2DM on a stable metformin background therapy will be enrolled in this trial.

This is a single center, phase IV, prospective, placebo-controlled, exploratory, proof-of-mechanism study.

This is a two-step treatment trial with a first open label and a second randomised, double-blinded, placebo controlled treatment phase. In treatment phase 1 (TP1, 30±4 days), 26 subjects will receive dapagliflozin 10 mg once daily as add on to their pre-existing metformin monotherapy. Thereafter, in treatment phase 2 (TP2, 30±4 days), subjects will be randomised to additional saxagliptin 5 mg or corresponding placebo treatment in a 1:1 ratio. Trial subjects will subsequently undergo both treatment phases, and endpoint assessments will be performed at the beginning and end of each treatment phase on Visits 2, 3, and 4Assignment of treatment allocation will take place at the investigational site. The total study duration for a subject will be between 64 to 84 days.

The use of a combined euglycaemic-hyperinsulinaemic and hyperglycaemic glucose clamp protocol is chosen in order to assess parameters of insulin resistance as well as alpha- and beta-cell function under highly standardised conditions within a small study population.

Trained members of staff will perform all administrations of the IMP at the clinical site. Drug accountability will be performed from Visit 2 to Visit 4 based on subject diaries and returned study medication.

Whenever a deviation is noted by the Investigator, the Sponsor and/or monitor should be informed and the implications of the deviation must be reviewed and discussed. The deviation must be documented, explaining the reason for the deviation (root cause analysis), actions taken and the impact of the deviation on the subject(s) and/or the trial. The safety laboratory will perform a first check on their values for safety parameters and flag any values outside the reference range. The Investigator must evaluate all results outside the reference range. Before the Investigator starts the trial, the laboratory reference ranges must be available in the Investigator's Trial File. The results provided by the safety laboratory will be part of the trial database. Data Management is the responsibility of Profil Institut für Stoffwechselforschung GmbH, Neuss Germany.

The objectives of the monitoring procedures are to ensure that (i) the safety and rights of the trial subjects are respected, (ii) that accurate, valid and complete data are collected, and (iii) that the trial is conducted in accordance with the trial protocol, the principles of GCP and local legislation.

The monitor must be given direct access to the Trial Investigator File and source documents (original documents, data and records). Direct access includes permission to examine, analyse, verify and reproduce any record(s) and report(s) that are important to evaluation of the clinical trial.

Key tasks of the monitor include verifying the presence of informed consent, the adherence to the inclusion/exclusion criteria, the documentation of SAEs, and the recording of all safety and efficacy variables. The monitor will also confirm the completeness of patient records, the adherence to the protocol and the progress in subject enrolment.

Profil Institut für Stoffwechselforschung GmbH, Neuss, Germany, will be responsible for the statistical analysis. The statistical planning and conduct of analyses of the data from this study will follow the principles defined in relevant ICH guidelines and the Sponsor's biostatistical standard operating procedures (SOPs). All statistical analyses will be performed using SAS® (SAS Institute Inc., Cary, North Carolina, United States of America [USA]), version 9.3 or later.

For the lack of existing data for confirmatory sample size calculation, the study is designed as an exploratory proof-of-mechanism study with an experimental assessment set-up. All obtained results will be handled with equal emphasis in a strictly exploratory sense. The data obtained from the trial will serve for the purpose of thesis generation and the design of potential confirmatory studies.

Dapagliflozin as well as Saxagliptin have been approved for treatment of T2DM in adult male and female patients. Because it is moreover not expected that the effects of either drug on the endpoint measures will differ between male and female subjects with T2DM, a specific gender distribution is not planned for this study.

Based on a previous study in twelve patients with T2DM treatment with dapagliflozin, a group size of 12 patients in each group has a power of 80% to detect a difference at a significance level of 0,05 (two-tailed).

Before data are released for statistical analysis, a blinded review of all data will take place to identify protocol deviations that may potentially affect the results. This review will be performed without revealing which trial product the subjects are assigned to. The blinding of the trial products will be maintained for everyone involved in allocating subjects to the analysis sets until data are released for statistical analysis. Furthermore, outliers will be identified by data review according to ICH-E9, using a fake randomisation. In addition, protocol deviations, which may potentially affect the results, will be identified and it will be evaluated if subjects and/or data should be excluded from the analysis.

The subjects and observations excluded from analysis sets, and the reason for this, will be described in the clinical trial report.

All available data will be included in data listings and tabulations. In general no imputation of values for missing data will be performed.

In general, metric variables and derived parameters will be presented using descriptive summary statistics including arithmetic means, median, minimum, maximum and standard deviations. Categorical variables will be presented using descriptive summary statistics including number of patients and percentages. Percentages of patients will be based on non-missing values.

In addition, for all efficacy variables as well as the derived parameters the geometric means and coefficients of variation will be presented as well.

Descriptive summaries will also be presented. Unless otherwise stated, all formal tests will be conducted at the two-sided 5% level of significance. No alpha adjustment will be done. All analyses on primary and secondary efficacy endpoints will be interpreted in an exploratory manner.

To investigate the dapagliflozin-corrected treatment effect, the evaluation of the change from Visit 3 to Visit 4 in incremental AUCGluc270-390min / AUCIns270-390min will be performed using analysis of covariance (ANOVA) with change in incremental AUCGluc270-390min / AUCIns270-390min as response, treatment as fixed effect and baseline (Visit 2) value as covariate. From this model, least square means of change from Visit 3 to Visit 4 and 95% confidence intervals for these changes will be calculated. In addition, estimate of the placebo-corrected treatment effect (difference of least square mean of Visit 3 to Visit 4 changes between saxagliptin and placebo) and corresponding 95% confidence interval will be calculated.

If the data are homogeneous or not normally distributed and cannot be successfully transformed, the analysis will be performed by non-parametric technique using Wilcoxon Signed Rank Test for within treatment comparisons and Wilcoxon Rank-Sum Test for between treatment comparisons based on a two-sided alpha level of 0.05. In addition, the estimate of Hodges and Lehmann and the corresponding 95% non-parametric confidence interval will be shown.

The secondary efficacy endpoints derived from the hyperglycaemic clamp phase based on incremental AUCs will be calculated as described for the primary endpoint.

All AEs that will be entered into the CRF (all AEs since first dosing) will be coded. All AEs will be presented by individual listings and frequency tables broken down by body system and preferred terms as assigned by the Medical Dictionary for Regulatory Activity, as appropriate. Separate tables for serious adverse events, by severity, as well as by relationship to trial medication, will be provided.

Inclusion Criteria:

1. Diabetes mellitus type 2 for at least three months prior to Screening
2. HbA1c 7.0%-9.9%, both inclusive
3. Treatment with metformin (daily dose 1500 - 3000 mg)
4. Age 30-75 years, both inclusive
5. BMI 25-35 kg/m^2, both inclusive

Exclusion Criteria:

1. Use of any oral antidiabetic treatment except for metformin (i.e., sulphonylureas, DPP-IV inhibitors, thiazolidinediones, SGLT-2 inhibitors) within the last three months prior to Screening
2. Use of insulin or GLP-1 analogues within three months prior to Screening
3. Treatment with any other investigational drug within three months before screening
4. History of diabetes mellitus type 1
5. Acute infections within the last two weeks prior to Screening
6. Anamnestic history of hypersensitivity to the study drugs or to drugs with similar chemical structures
7. History of severe or multiple allergies
8. GFR (as calculated by the Cockroft-Gault equation) < 60 ml/min at Screening
9. State after kidney transplantation
10. Laboratory safety value(s) outside the reference range and deemed clinically relevant by the Investigator
11. Sexually active woman of childbearing age not practicing a highly effective method of birth control as defined as those which result in a low failure rate (i.e., less than 1% per year) when used consistently and correctly such as implants, injectables, combined oral contraceptives, hormonal IUDs, sexual abstinence or vasectomised partner
12. Pregnancy or breast feeding
13. Systolic blood pressure outside the range of 100-160 mmHg or diastolic blood pressure above 90 mmHg at Screening
14. Acute myocardial infarction or cerebral event (stroke/TIA) within six months prior to Screening
15. Uncontrolled unstable angina pectoris or history of pericarditis, myocarditis, endocarditis
16. Increased risk of thromboembolism, e.g. subjects with a history of deep leg vein thrombosis or family history of deep leg vein thrombosis, as judged by the Investigator
17. Hemodynamic relevant aortic stenosis, Aortic aneurysm
18. Repeated episodes of severe hypoglycaemia within six months prior to Screening
19. History of diabetic ketoacidosis, praecoma diabeticum, or diabetic coma
20. Recurrent urogenital infections
21. Haematuria
22. History of pancreatitis
23. Progressive fatal disease
24. Elective surgery planned during study participation
25. Acute or scheduled investigation with iodine containing radiopaque material
26. History of drug or alcohol abuse in the past two years
27. Donation of blood, major blood loss (>=500 ml), or major surgery within the last three months prior to Screening
28. Active hepatitis B, measured by positive tests of surface antigen HBsAg and/or active hepatitis C, measured by positive hepatitis C virus antibody tests (HCV) at Screening
29. Positive human immunodeficiency virus (HIV) antibodies or HIV 1 Ag at Screening