 |
|
 |
 |
 |
 |
 |
|
|
 |
|||||||||||||||||||||||||||||||||||||
 |
||||||||||||||||||||||||||||||||||||||||
| Home > Evidence of the Month > Archive | |
 |
|
Education |
| January 2006 Evidence of the Month |
 |
The PROactive Study
Alistair Emslie-Smith
Tayside Diabetes Network
Dundee, Scotland
Comment on:
Background
The results of the Prospective pioglitAzone Clinical Trial In macroVascular Events (PROactive) study were reported at the annual meeting of the European Association for the Study of Diabetes (EASD) in Athens in September 2005 and subsequently published in The Lancet in October 2005.
This was an international, multicentre, prospective, double-blind, placebo-controlled trial to assess whether the addition of pioglitazone to an existing diabetes regimen would prevent macrovascular events in people with Type 2 diabetes and evidence of pre-existing macrovascular disease (secondary prevention) and to assess the safety and tolerability of such treatment.
Methods
People with Type 2 diabetes (5238) and existing arterial damage (over 50% had ischaemic heart disease) were recruited. Participants were allocated to pioglitazone (force titrated to 45 mg/day) or placebo to be taken in addition to their glucose-lowering drugs and other medications. Glucose-lowering therapy of all types including insulin was to be adjusted towards best control (according to International Diabetes Federation [IDF] Europe guidelines). Thirty percent were taking insulin at baseline. The primary endpoint of the study was the composite of all-cause mortality, non-fatal myocardial infarction (MI) (including silent MI), stroke, acute coronary syndrome, endovascular or surgical intervention in the coronary or leg arteries, and amputation. Average time of follow-up was 34.5 months.
Key Results
- Pioglitazone was well tolerated, with fluid retention being the major problem.
- Glycaemic control (mean HbA1c , –0.5%), lipid levels (triglycerides, –13%; HDL cholesterol, +9%), and systolic blood pressure (–3 mmHg) were improved in the pioglitazone group.
- Addition of pioglitazone did not convincingly reduce the composite risk of the primary endpoint events (mean change, –10%, P=0.095).
- The investigators then analysed the data for a predefined ‘principal secondary endpoint’ (an unusual concept), the composite of all-cause mortality, non-fatal MI, and stroke. They found a risk reduction among participants in the pioglitazone group of 16%
(P=0.027). This is an absolute risk reduction of ~2% after 3 years (number needed to treat to prevent one event in 3 years = 50).
- All-cause mortality alone was not significantly decreased (study not powered for this).
- Many sub-analyses participants on statins did not appear to gain at all from pioglitazone therapy.
- The subjects treated with pioglitazone had twice as many episodes of heart failure (unadjudicated adverse event reporting) and four times as many episodes of oedema as the placebo subjects. The incidence of heart failure was twice as great as the reduction in incidence of macrovascular events.
- Weight gain was 4 kg greater in the pioglitazone group than in the placebo group, despite more insulin use in the latter.
- The bladder cancer rate in absolute terms was higher in the pioglitazone than in the placebo group (14 vs 6 people, respectively; most events occurred in the first year); breast cancer was lower (2 vs 11 people, respectively).
Clinical Relevance
- The study results are difficult to interpret with confidence and need careful thought.
- One approach to the main efficacy outcome is to observe that the direction of advantage for both the primary and principal secondary endpoint is consistent with the prior probability of advantage for pioglitazone based on results from other glucose-lowering drugs and pioglitazone's known effects on cardiovascular risk factors. Under this approach, the findings of 10% and 16% risk reductions are now our best estimates of the efficacy of the drug to guide clinical practice.
- There has, however, been much criticism in the literature about the validity of positive conclusions drawn from secondary endpoint data in the light of a ‘negative’ primary endpoint. Note this approach denies any previous knowledge of pioglitazone or its effects on surrogate outcomes such as glucose and triglycerides. It is then suggested that, at best, such data should be treated as exploratory or hypothesis-generating rather than as having hard clinical relevance.
- Are the improvements seen in blood glucose, lipid, and blood pressure control consistent with the possible risk reduction found? From the United Kingdom Prospective Diabetes Study (UKPDS) and other trials this seems likely. The observation is important as it means that other known effects of pioglitazone (such as reduction in inflammatory markers or insulin insensitivity) are having no added advantage.
- The lack of advantage in statin takers has received much coverage but should probably be ignored; it was one of many such analyses and thus statistically suspect. A further trial is needed to address this point.
- The fluid retention and heart failure issues are not adequately addressed by this study, except to note that excess heart failure deaths were very low compared with the numerical reduction in other cardiovascular deaths. As fluid retention and heart failure are not otherwise usually considered as critical health events as MI or stroke, it would appear that the problem is not a major safety concern (in people without prior heart failure).
- The relative risk of heart failure was not analysed in patients on insulin and pioglitazone, despite over 30% of participants being on both drugs (concomitant use of insulin and glitazones is contraindicated in Europe due to concerns about risk of heart failure).
- The finding that fewer people on pioglitazone progressed to insulin therapy is not of great significance; this would be expected of any glucose-lowering drug.
- There have been prior concerns about bladder tumours with peroxisome proliferator-activated receptor (PPAR) α-γ agonists - the group of drugs to which pioglitazone belongs. This is for their potential growth-promoting effects, not genotoxicity. This makes the findings of PROactive in this area concerning; they are under review by the regulators and manufacturers.
- The positive breast cancer findings were more of a surprise but consistent with some prior speculations. This finding needs further trials.
- Caution is needed in generalising the findings to people with diabetes without extant cardiovascular disease, to non-Caucasian populations, or to other drugs in the same classes.
The consistency of the findings with previous results for glucose-lowering drugs in the UKPDS is comforting. It is disappointing that the putative advantages in cardiovascular risk protection of this group of drugs have not been demonstrated in the trial. In these circumstances pioglitazone should be regarded as offering no clear advantage over other glucose-lowering drugs and generally inferior to metformin.
Other trials using drugs in this class will report in 2006, 2007, and 2009, notably, A Diabetes Outcome Progression Trial (ADOPT), Diabetes Reduction Approaches With Ramipril and Rosiglitazone Medications (DREAM), and Rosiglitazone Evaluated for Cardiac Outcomes and Regulation of Glycaemia in Diabetes (RECORD). Hopefully, these together with PROactive and other studies will further improve our understanding of the role of thiazolidinediones in clinical practice.
Further Reading
Yki-Järvinen H. The PROactive study: some answers, many questions. Lancet. 2005;366:1241-1242.
This Website Feature is funded by an unrestricted educational grant from sanofi-aventis and Pfizer Inc.
|
 |
|
Print this page |
|
||
 |
Add to Favourites |
|
|