Thiazolidinediones Or Sulphonylureas and Cardiovascular Accidents.Intervention Trial (TOSCA IT)

Cardiovascular (CV) disease is the most common cause of morbidity and mortality among diabetic patients.The UK Prospective Diabetes Study (UKPDS)clearly showed that tight glycemic control significantly decreases diabetes-related events. Therefore, achievement of HbA1c < 7% is a major goal in the treatment of type 2 diabetes since this parameter still represents the best predictor of micro and macrovascular complications. However, it is often difficult to maintain this goal because of the progressive deterioration of pancreatic beta-cell function and insulin sensitivity. The progressive nature of type 2 diabetes generally requires a stepwise therapeutic approach starting with lifestyle intervention (diet and increased physical activity). After lifestyle changes have failed, drug therapy is initiated and progressively intensified through the combination of different classes of hypoglycemic agents. Since insulin resistance is recognized as a major factor in the pathogenesis of type 2 diabetes and associated CV risk factors, drugs that improve insulin sensitivity are advised as initial pharmacological therapy. Currently, metformin is recommended as the first-line drug for patients with type 2 diabetes. Great uncertainty exists regarding the best therapeutic option in diabetic patients inadequately controlled with metformin, due to the lack of randomized controlled trials that have directly compared the efficacy of different combination regimens in achieving glycemic goals. The paucity of sound clinical evidence in this area is highlighted in a Consensus Statement from the American Diabetes Association and the European Association for the Study of Diabetes, that suggested either a sulphonylurea (SU) or a thiazolidinediones (TZD) as additional medication to metformin, in the absence of data demonstrating the superiority of one combination over the other. Addition of a SU or a TZD to concomitant metformin results in a substantial improvement in glucose control with a mean reduction in HbA1c of 1-1.5%. However, a direct evaluation of these treatment strategies, within a large clinical trial, would be most appropriate, given their different mechanisms of action, side effects and costs. Indeed, the only direct comparison between a SU and a TZD added to metformin has been obtained in a short-term study (24 weeks), which showed a greater reduction in HbA1c with the association metformin-glibenclamide (- 1.5%) compared with the association metformin-rosiglitazone (-1.1%, <0.001). More patients receiving metformin-glibenclamide reached HbA1c <7.0 % than did those receiving metformin-rosiglitazone (60% vs. 47%). In addition, a more favorable lipid profile was present in patients treated with metformin-glibenclamide, although this finding has not been confirmed in other studies. However, the short duration of the study precluded evaluation of CV endpoints.

This lack of information prevents an evidence-based choice between these two treatment options.

In this respect, several studies are underway to investigate the potential of different glucose-lowering therapies on CV outcomes. With regard to SU, retrospective cohort studies have documented a higher risk of adverse outcomes in patients treated with a SU in monotherapy or in combination with metformin as compared to metformin alone. In the UKPDS, diabetic patients treated with a combination of glibenclamide plus metformin showed an increase in CV mortality compared with other groups of treatment. However, these data derive from a post-hoc analysis and, therefore, should be interpreted with caution. In contrast, the ADOPT study has shown a lower proportion of CV events in the glibenclamide group compared with rosiglitazone and metformin. These conflicting reports underline the need to evaluate the rate of CV events induced by the combination metformin-SU in comparison with other treatment strategies. Regarding TZDs, a recent prospective study (PROACTIVE Study) has compared the efficacy of adding pioglitazone (from 15 to 45 mg/die) or placebo to the standard hypoglycemia therapy in the secondary prevention of CV events in type 2 diabetic patients. Pioglitazone treatment did not significantly affect the primary endpoint, which included all-cause mortality, non-fatal MI, stroke, acute coronary syndrome, endovascular or surgical intervention in the coronary or leg arteries and leg amputation, but significantly reduced the number of patients who experienced the secondary endpoint (all-cause mortality, nonfatal MI and stroke). However, patients treated with pioglitazone had 115 more episodes of heart failure than the placebo group. In addition, weight gain was 4 kg greater with pioglitazone than with placebo, and much greater than that expected on the basis of the improved glycemic control. These conflicting reports support the need for large-scale clinical trials that compare the efficacy of the association of metformin plus a SU vs. metformin plus a TZD on CV outcomes. For these reasons, the Italian Diabetes Society has organized a randomized controlled clinical trial on long term cardiovascular effects of these two therapeutic strategies. The aim of this study is to ascertain whether in patients with type 2 diabetes poorly controlled with metformin in monotherapy, the addition of pioglitazone reduces the rate of cardiovascular events as compared with the addition of a SU. Glucose control, major CV risk factors, safety, tolerability and economic costs with the two therapeutic regimens will also be compared.

Design: multicentre, randomized, open-label, comparative, parallel-group trial of 48 months duration (PROBE: Prospective Randomized Open Blinded End-Point).

RECRUITMENT WAS TERMINATED IN JANUARY 20147, WITH 3040 PATIENTS ENROLLED. Screening and follow-up took place at several Clinical Units which will work in collaboration with a network of Diabetes Outpatients Clinics all over Italy. All study procedures will be performed by trained study personnel according to a standardized protocol.

All eligible patients meeting the inclusion criteria, free from any condition listed in the exclusion criteria and giving the informed consent, have been randomized to one of two arms:

• Metformin (2 gr/die) + sulfonylurea
• Metformin (2 gr/die) + pioglitazone In the group randomized to metformin plus a sulfonylurea the choice between glibenclamide, gliclazide and glimepiride will be left to the study physician following local practices.

Participating centres were masked to the randomization sequences which were generated at the Epidemiology Unit of the "Consorzio Mario Negri Sud" (Dr Nicolucci). Patients entering the trial were allocated to the treatment group upon a telephone call from the clinic to the Epidemiology Unit after verification of all eligibility/exclusion criteria. The randomization sequences were stratified by clinic, gender and previous cardiovascular events.

Patients are seen for clinical follow up after 1, 3 and 6 months from the randomization and then every 6 months unless clinical conditions requiring more frequent medical consultations develop (according to GCP). Telephonic contacts are established with all patients in order to take information on adverse events,on glycaemic control and to modify drugs dosage.

The metformin dose will remain constant throughout the study; the initial dose of added drug will be 5 mg glibenclamide or 30 mg gliclazide or 2 mg glimepiride for the group randomized to metformin plus SU and 15 mg pioglitazone in the group randomized to metformin plus pioglitazone. If glucose control is unsatisfactory (fasting glucose >120mg/dl or post prandial glucose >160 mg/dl in more than 50% of home readings over a 8 weeks period), study medications will be uptitrated to the maximal effective daily dose (i.e. 15 mg glibenclamide, 120 mg gliclazide, 6 mg glimepiride, 45 mg pioglitazone). If a HbA1c >8.0% is observed at any follow up visit, patients will receive reinforcement of lifestyle education, compliance to study protocol will be assessed and a glycated hemoglobin measurement will be repeated after three months. A confirmed value >8.0% despite adherence to maximal doses of study medications for the previous three months, will lead to addiction of a single injection of insulin Glargine bed-time. Insulin titration will be performed on the basis of fasting capillary glycemia according to a pre-defined algorithm. If a HbA1c >8.0% is observed at two consecutive follow-up visits 3 months apart, one or more injections of analog insulin will be added. Insulin will be titrated down, according to a pre-specified algorithm, if frequent hypoglycemic episodes occur.

Use of concomitant medications (antihypertensive, lipid-lowering and antiplatelet agents) will be permitted throughout the study, according to Good Clinical Practice Guidelines.

Patients will stop the study medications but will remain in the study if any of the following conditions develops:

• alanine aminotransferase increases 2.5 or more times above baseline on two consecutive occasions (one month apart)
• signs and symptoms suggestive of heart failure evaluated according to the American Heart Association and the American Diabetes Association consensus on glitazones and heart failure (19).
• bladder cancer (in that case the study drug will be stopped only in the pioglitazone arm)
• medical conditions contraindicating oral hypoglycemic treatment and requiring stable insulin treatment.

The sample size is calculated based on the assumption that the study results would be relevant for the public health point of view in case add-on treatment with pioglitazone proves more effective than add-on sulfonylureas for the management of hyperglycemia and the prevention of the chronic complications of diabetes.

The sample size is thus estimated to detect with a 80% power of reduction in the risk of events of 20% (HR=0.80; metformin + Pioglitazone vs metformin + SU), with p<0.05 (one sided log rank test). Assuming an estimated occurrence rate of the primary end point of 3.5% per annum (24-26) and a 5% loss at follow-up, a total of 3371 patients will be enrolled with the event driven analysis at 498 end point events.

A Clinical Endpoints Committee (CEC) composed of Cardiologists and Specialists in Internal Medicine with vascular expertise will review and adjudicate all potential end points, according to pre-specified criteria. This committee will be blind to the study medication assignment group and independent of the Steering Committee. Side effects (i.e. weight gain, peripheral edema, hypoglycemic episodes, etc.) will be monitored. All adverse events, whether or not attributed to study drugs, will be collected and recorded on the appropriate, ad hoc page(s) of the Case Report Form (CRF). The Data Safety Monitoring Board, composed of clinical trials expertise, will monitor events and side effects and will decide the study discontinuation in the case of a significant difference between the two arms in the rate of the primary endpoint or serious adverse events.

At baseline and at each follow-up visit (ever six months) the following parameters will be assessed:

• vital signs, anthropometry, check for any new signs/symptoms
• home glucose readings
• frequency of hypoglycemic episodes graded by severity.
• intercurrent illnesses/events, change(s) in use or dose of drugs
• history of hospitalization for myocardial infarction, stroke, major leg amputation (above ankle), acute coronary syndrome, heart failure, endovascular or surgical intervention on the coronary, leg or carotid arteries)
• angina or intermittent claudication At baseline and every months safety parameters (ALT, AST, macroematuria) and HbA1c will be evaluated.

At baseline and every 12 months serum lipids (total cholesterol, HDL cholesterol and triglycerides), microalbuminuria and C Reactive Protein (CRP) will be evaluated.

At baseline and every 12 months a standard resting ECG (Minnesota coding) will be performed.

All study investigations will be performed according to a standard protocol described in detail in the operating manual which will be produced by the Coordinating Center in collaboration with the Steering Committee. Prior to study start-up, the investigators will attend training and standardization sessions to reduce inter-observer variability; these sessions will be organized and supervised by the Coordinating Centre. Ad hoc CRFs will be developed to collect information which will be transmitted to the Epidemiology Unit.

The CRFs will be reviewed and data queries will be produced to correct missing or incorrect data on the CRFs. Periodically a Quality Team (appointed by the Steering Committee) will check the amount of CRFs per centre and produce recalls for centres not sending CRFs. An electronic data base will be created using a standardized procedure of data input.

The study conduct (i.e. timing, adherence to protocol etc..) at each participating centre will be monitored by regular visits of professional monitors (at the start-up, at the end of the study and twice a year). A Steering Committee will meet to review the study progress every 6 months, and an independent Data and Safety Monitoring Board (DSMB) will monitor safety outcomes throughout the course of the study. This Committee will assess at given intervals the safety data, the critical efficacy endpoints and will recommend whether to continue, modify, or stop the trial. Stringent statistical criteria will be set for early study termination in the event of a clear-cut difference between the treatment groups with respect to all-cause mortality. The DSMB should also recommend termination of the study for other serious safety reasons. The DSMB will meet independently of the Steering Committee.

The study will be conducted according to Good Clinical Practice (GCP)Guidelines. The protocol will be approved by Ethics Committees or Institutional Review Boards of each centre. Written informed consent will be obtained from all participants before beginning the study.

Several clinical centres, one epidemiology unit, and a central laboratory will participate in the study. The Coordinating Centre (Prof Riccardi) will be responsible for preparation of the study protocol together with the Steering Committee and for the preparation of the manual of operations, standardization and training sessions for the field investigators and for the monitoring of study conduct.

The Epidemiology unit (Dr Nicolucci) will provide a centralized randomization system for the patients allocation to study medications, will be responsible for data managing, creation of an appropriate electronic data base-planned according to the GCP- and for data analyses.

Statistical evaluation of end points will be performed under the responsibility of Prof Giuseppe Gallus (University of Milan) according to a protocol of analysis approved by the Steering Committee.

Adverse events monitoring and analysis will be provided by the Unit of Pharmacovigilance of the Federico II University of Naples (Prof. G. Di Renzo).

Pharmacoeconomics evaluation will be provided by the Interdepartmental Centre of Pharmacoeconomy of the Federico II University of Naples (Prof G.L. Mantovani).

Routine laboratory measurements including fasting glucose, haematology markers, liver enzymes, hepatitis markers, renal function tests, serum beta-human chorionic gonadotropin in premenopausal women will be performed at each centre. A central laboratory participating to an external quality control program will measure glycated hemoglobin (HPLC), standard lipid profile (total cholesterol, HDL cholesterol, triglycerides), macroematuria, microalbuminuria, CRP (ELISA; high sensitivity). Extra samples will be collected and stored (-70 °C) for future biochemical analyses. A genetic biobank will be prepared.