The aim of this project is to identify subgroups of patients with type 2 diabetes that respond well or poorly to particular drugs based on particular clinical characteristics such as their weight or kidney function, to enable better targeting of treatment for a particular individual.

This study will test 2 hypotheses of drug response supported by routine clinical and trial data. 600 patients with type 2 diabetes who have suboptimal glycaemic control on dual oral therapy will be recruited to a randomised double-blind crossover study of a DPP4 inhibitor, SGLT2 inhibitor and thiazolidinedione. Each patient will take each study drug in addition to their existing treatment for four months at a time. At the end of each treatment the patient's glucose control will be measured and information about their experience of the drug will be collected.

The study is a phase 4 randomised double-blind crossover study of a DPP4 inhibitor, SGLT2 inhibitor and thiazolidinedione as third line therapy in patients with Type 2 diabetes who have suboptimal glycaemic control on dual therapy with metformin and a sulphonylurea.

600 patients aged 30-80 who have been on stable doses of 2 classes of therapy (not including the trial IMPs or GLP1-agonist) for at least 3 months with HbA1c >58mmol/mol (7.5%) will receive three double-blinded third-line non-injectable therapies. On recruitment into the study participants will have underlying pathophysiology assessed in a mixed-meal tolerance test (MMTT) and samples will be collected for baseline analysis and storage for future biomarker analysis and discovery. Participants will then receive 16 weeks of each over-encapsulated blinded therapy in random order.

At the end of each treatment period, fasting blood will be taken to measure glycaemic response (HbA1c), fasting glucose and insulin concentrations trough drug levels and to confirm continued eligibility. Weight, blood pressure and. data about patient experience will also be collected including perceived side effects, preparedness to remain on therapy, psychological health and health related quality of life.

At the end of the study, patient treatment preference will be recorded after feeding back to the patient for each of the 3 therapies their HbA1c, weight change, frequency of hypoglycaemias, any patient reported side effects and the patient's verdict on each therapy will be recorded. Each participant will be asked which treatments they would take long term and the reason for their preference.

Inclusion Criteria:

• Clinical diagnosis of Type 2 diabetes
• Age ≥30 and ≤80
• Currently treated with two classes of oral glucose-lowering therapy (given either as separate or combined medications), that do not include a DPP4-inhibitor, a SGLT2-inhibitor or a thiazolidinedione.
• Diabetes duration ≥12months
• No change in diabetes treatment (new treatments or dose change) within previous 3 months
• HbA1c > 58mmol/mol (7.5%) and ≤110mmol/mol (12.2%) - confirmed at screening visit
• eGFR ≥ 60mls/min/1.73m² - confirmed at screening visit
• Able and willing to give informed consent

Exclusion Criteria:

• Changes in glucose-lowering therapy or dose within last 3 months

• HbA1c ≤ 58mmol/mol (7.5%) or >110mmol/mol (12.2%)

• eGFR <60mls/min/1.73m².

• Diabetes duration <12 months

• ALT >2.5 x upper limit of the assay normal range or known liver disease, specifically >30 μmol/L that is associated with other evidence of liver failure.

• Insulin treated within the last 12 months

• Limb ischaemia shown by absence of both pulses in one or both feet.

• Currently treated with corticosteroids

• Currently treated with rifampicin, gemfibrozil, phenytoin and carbamazepine

• Active infection (any infection requiring antibiotics at present)

• Foot ulcer requiring antibiotics within previous three months

• Recent (within 3 months) significant surgery or planned surgery (excluding minor procedures)

• Acute cardiovascular episode (angina, myocardial infarction, stroke, transient ischemic episode) occurring within the previous 3 months

• History of heart failure

• Current use of loop diuretic therapy (Furosemide or Bumetanide)

• History of bladder carcinoma

• Current/ongoing investigation for macroscopic haematuria

• History of Diabetic Ketoacidosis

• History of pancreatitis

• Pregnant, breastfeeding or planning a pregnancy over the study period

• Concurrent Participation on another Clinical Trial of an Investigational Medicinal Product, where the IMP is currently being taken, or without sufficient washout period* and without consultation with the CTIMP research team.

• Unable or unwilling to give informed consent

  • Sufficient washout period = five times the half-life of the IMP / potential IMP if involving a placebo / longest half-life if a trial includes more than one drug