Previous data from our clinical trial has demonstrated that the combination of dapagliflozin plus pioglitazone cause robust decrease in the plasma glucose concentration without significant increase in plasma ketone concentration in subjects with T1DM receiving multiple insulin injections or insulin pump. Although patients receiving insulin therapy with pump have participated in the study, none has insulin delivered in automated pump (780 pump) which automatically adjusts the rate of insulin infusion based upon the measured plasma glucose concentration. Although automated insulin pumps have been introduced to practice only last year, they are gaining popularity in the care of T1DM patients worldwide. The aim of this amendment is to demonstrate the efficacy (decrease in HbA1c) and safety (no significant increase in plasma ketone concentration) of combination of dapagliflozin plus pioglitazone in T1DM patients receiving insulin therapy with automated insulin pump.

Insulin Pumps Participants The interim analysis of our previous study has demonstrated that the combination of dapagliflozin plus pioglitazone cause robust decrease in the plasma glucose concentration without significant increase in plasma ketone concentration in subjects with T1DM receiving multiple insulin injections or insulin pump. Although patients receiving insulin therapy with pump have participated in the study, none has insulin delivered in automated pump (780 pump) which automatically adjusts the rate of insulin infusion based upon the measured plasma glucose concentration. Although automated insulin pumps have been introduced to practice only last year, they are gaining popularity in the care of T1DM patients worldwide. The aim of this amendment is to demonstrate the efficacy (decrease in HbA1c) and safety (no significant increase in plasma ketone concentration) of combination of dapagliflozin plus pioglitazone in T1DM patients receiving insulin therapy with automated insulin pump.

1. The primary objective of the study is to examine the decrease in HbA1c caused by pioglitazone plus dapagliflozin versus placebo in T1DM patients receiving automated insulin delivery (780 pump).
2. Secondary objectives of the study are to examine the effect of pioglitazone plus dapagliflozin on plasma ketone concentration, time in range, body weight and rate of hypoglycemia.

Research Design:

Patients will be screened as follows:

1. Screening: Patients will be screened with medical history and physical examination. Blood will be drawn for CBC, Chem 20, lipid profile, TSH, fasting insulin, C-Peptide, ketones, pregnancy test and urine analysis. All measurements will be performed in the morning after 10-12 hours overnight fast.
2. Run in Period (week 1-4) Following determination of eligibility, subjects will receive Fasting Measurement of: plasma glucose, insulin, C-peptide, glucagon, GLP-1, GIP, HbA1c, FFA, ketones, lactate, ProBNP, CBC, blood chemistries, lipid profile, erythropoietin, creatinine, uric acid, Na, K, Cl, HCO3, PH, Ca, urinary creatinine to microalbumin ratio and urinary creatinine to glucose ratio. EGFR will be calculated using the National Kidney Foundation (NKF) and the American Society of Nephrology (ASN) calculator CKD-EPI using Creatinine Equation while adjusting for body surface to estimate GFR. This is centralized for all centers and done by the team in Texas diabetes Institute. A copy of the eGFR calculation and results will be included in the patients file. After completing the fasting measurements, subjects will start a 4 week run in period during which they will be instructed to continue their daily insulin regimen unchanged unless instructed otherwise by the study team. They also will be instructed to continue their measurement of daily glucose levels at least 4 times each day (before each meal and at bed time) to determine the baseline rate of hypoglycemia. Subjects will be instructed to continue with this schedule of home measurement of plasma glucose levels until the end of the study (week 32), and to bring the results of the measurement at each follow-up visit.

During the run in period, patient compliance will be evaluated by examining their diary of insulin dose injection, and blood sugar levels. The daily insulin dose in the diary will be contrasted with the measured fasting plasma insulin concentration. It should be noted that, the vast majority of patients will be recruited by the investigators from T1DM patients who are under their care for long period of time and their compliance with T1DM treatment is well known to investigators (see page 18 below).

Home-measured blood glucose values will be reviewed at the end of week one of the run in period, subjects with evidence for hypoglycemia or who have FPG <120 mg/dl (measured with glucometer after overnight fast) will have their daily insulin dose adjusted according to the algorithm below (Basal insulin adjustment) to achieve a goal of a mean FPG >120 mg/dl as measured with glucometer in home-measured plasma glucose levels during week 4. Since the primary aim of the present study is to examine the efficacy of SGLT2 inhibitor with and without pioglitazone on the HbA1c, no adjustment in insulin dose will be done during the run in period or at the time of starting dapagliflozin in patients without evidence of hypoglycemia and FPG >120 mg/dl, in order not to affect the decrease in HbA1c caused by SGLT2 inhibitor and not to increase the risk of ketoacidosis. Changes in daily insulin dose will be allowed after starting dapagliflzoin therapy to avoid hypoglycemia (see algorithm below) and will be made based upon the home-measured plasma glucose levels and the CGM results, and the discretion of the investigator. Since previous studies (41) have demonstrated that <20% decrease in daily insulin dose minimizes the risk of ketoacidosis. The change in insulin dose at each follow-up visit will be limited to <10% of the daily dose. Plasma ketone concentration will be measured on the following day before making additional insulin dose adjustment if necessary. This plan of insulin dose adjustment minimizes the reduction in insulin dose to the minimum required to avoid hypoglycemia and minimizes the risk of ketoacidosis. Subjects with evidence of hypoglycemia will not be allowed to start dapagliflozin therapy before CGM confirms the elimination of hypoglycemic episodes.

Patients whose daily insulin dose required adjustment (because of hypoglycemia) will be seen weekly during the run in period, and the daily insulin dose will be adjusted to ensure a FPG concentration >120 mg/dl at 4 weeks without hypoglycemic events.

Eligible subjects will enter the run in period in which the pump parameters will be adjusted in patients experiencing hypoglycemic events in order to be eliminated.

During the run in period subjects will be asked to fill a questionnaire aims at examining the satisfaction from glucose management and the impact of combination therapy with pioglitazone plus dapagliflozin on the quality of glucose management (see questionnaire below). Subjects will be asked to fill the quality of life questionnaire at the las clinic follow-up visit.

Eligible subjects without evidence of hypoglycemia will be randomized into two groups to receive for 16 weeks in a double-blind fashion the following: Group 1 will be started on dapagliflozin 10 mg per day plus 30 mg pioglitazone. If well tolerated, the pioglitazone dose will be increased to 45 mg after 2 weeks. Subjects in Group 2 will receive matching placebo.

Subjects in both groups will be provided with ketone meters and will be asked to measure their plasma ketone concentration every morning. If the plasma ketone concentration exceeds 1.0 mM, subjects will be instructed to contact the study nurse coordinator.

Follow-up visits After starting therapy, subjects will be seen in the clinic (Clinic visit) every 4 weeks. Telephone calls (phone visits) also will be done by the study nurse every 2 weeks or as needed. During each follow-up visit (phone or clinic), brief medical history will be performed. Patients will be questioned about their medical history and their general health since last follow-up visit. They will be asked open ended questions e.g. "Have you had any health problems since the last visit". CGM data and ketone levels will be reviewed and recorded by the study nurse coordinator. If hypoglycemic events occur, changes to the pump parameters (e.g. carbs ratio, glucose target, and active insulin time) will be made to eliminate hypoglycemic events. Otherwise, no adjustment in pump parameters will be done during the study.

During the clinic visit, physical examination will be made, and body weight, blood pressure and pulse will be measured. Blood sample will be drawn for the measurement of FPG, ketone, glucagon, and FFA concentrations, and HbA1c. Subjects will be instructed to maintain telephone contact with the study nurse coordinator and inform him/her about any event of hypoglycemia, sudden increase in plasma glucose concentration, or an increase in plasma ketone concentration > 1.0 mM, or other health related issues should they occur.

Hypoglycemia will be defined as blood sugar levels <70 mg/dl, or hypoglycemic symptoms that subside after carbohydrate ingestion and will be measured as time below range. Severe hypoglycemia will be defined as hypoglycemia requiring third party assistance.

subjects will be provided with pens of degludec as a back up for use in case of technical issue with the pump and will be advised to immediately inject an insulin dose equal to their basal insulin regimen in case of technical problem with the pump and contact one of the study team immediately to test the pump function.

Management of Plasma Ketone Concentrations Patients will be instructed at the time of starting therapy (dapagliflozin plus pioglitazone or placebo) to increase fluid consumption by 1 liter per day to compensate for urinary water loss (mean of 400 ml) caused by dapagliflozin. Patients also will be instructed about the signs and symptoms of diabetic keto acidosis (nausea, abdominal pain, rapid breathing etc.).

If at any time, plasma ketone concentration exceeds 1.0 mM, patients will be instructed to immediately contact the study nurse coordinator. Possible precipitating factor for ketosis will be examined (e.g., acute illness or stress). Insulin pump will be tested to rule out pump failure. Patients will be questioned about their diet and insulin injections to ensure that the increase in plasma ketone concentration is not due to non-adherence to therapy or due to pump or technical failure.

If precipitating factor for ketosis is identified. Subjects will be asked to hold therapy until it is resolved, after which time it will be restarted.

If a precipitating factor is ruled out and the increase in plasma ketone concentration >1.0 mM persists for 3 consecutive days, or the plasma ketone concentration continue to raise and exceed 1.5 mM, it will be considered drug induced ketosis. Subjects will be unblinded. If the ketosis occurred in subject receiving dapagliflozin plus pioglitazone, the study will be terminated. The insulin dose is escalated until the ketosis is resolved. However, if patient is receiving placebo, open label dapagliflozin plus pioglitazone (45 mg) will be started and continued for 16 weeks. Patients will continue with the regular visit schedule until the end of study.

If an increase in plasma ketone concentration occurs in subjects receiving open label dapagliflozin plus pioglitazone, the drug therapy will be discontinued, the insulin dose is escalated and the study is terminated.

Primary outcome of the Study: is the change from baseline to week 16 in the HbA1c.

Secondary outcomes include:

1. Change from baseline to week 16 in plasma ketone concentration
2. Change from baseline to week 16 in time in range, and time above range
3. Incidence of hypoglycemia measured as time below range
4. Change from baseline to week 16 in body weight

Inclusion Criteria:

- 1) Age >18 years 2) T1DM 3) Other than diabetes, subjects must be in good general health as determined by physical exam, medical history, Chem 20, CBC, TSH, urinalysis, and EKG.

4) Fasting C-peptide concentration <0.7 ng/ml 5) Poor glycemic control (HbA1c=7.0-11.0%) 6) Treatment with multiple daily insulin injections (basal plus prandial) or insulin pump 7) Total daily insulin dose ≥0.6 U/kg per day 8) Stable insulin dose (±4 units) in the preceding three months. 9) eGFR≥60 ml/min. 10) Weight stable over the preceding 3 months (± 3 pounds) and who do not participate in an excessively heavy exercise program

Exclusion Criteria:

• 1) Age >18 years 2) T1DM 3) Other than diabetes, subjects must be in good general health as determined by physical exam, medical history, Chem 20, CBC, TSH, urinalysis, and EKG.

  4) Fasting C-peptide concentration <0.7 ng/ml 5) Poor glycemic control (HbA1c=7.0-11.0%) 6) Treatment with multiple daily insulin injections (basal plus prandial) or insulin pump 7) Total daily insulin dose ≥0.6 U/kg per day 8) Stable insulin dose (±4 units) in the preceding three months. 9) eGFR≥60 ml/min. 10) Weight stable over the preceding 3 months (± 3 pounds) and who do not participate in an excessively heavy exercise program

Access will be allowed after communicating with PI and approval of steerng committee