Send This Page By E-mail
Print this page

Interview with Dr. Michael Davidson on the CHICAGO Trial

Michael H. Davidson, MD, FACC, FACP

On November 13 in Chicago, in a late-breaking clinical trials session of the 2006 American Heart Association Scientific Sessions, researchers presented results of the Carotid Intima-Media Thickness in Atherosclerosis Using Pioglitazone (CHICAGO) study. The study was simultaneously published online by JAMA (early-release article posted November 13, 2006).¹ In a multiethnic, multiracial population of patients with type 2 diabetes (N = 462), the CHICAGO study compared 2 oral antidiabetic agents, the thiazolidinedione pioglitazone and the sulfonylurea glimepiride, to determine whether the thiazolidinedione had a beneficial cardiovascular effect independent of its antiglycemic effect. The primary end point was change in carotid intima-media thickness as a surrogate for progression of atherosclerosis. In an interview with APOLLO, Dr. Michael H. Davidson, of Rush University Medical Center in Chicago, one of the CHICAGO investigators, discussed the results of the trial and their clinical significance.

Dr. Davidson, please explain, if you would, the association between diabetes and cardiovascular disease as we understand it in 2006?

Dr. Davidson: Diabetes is a state of very high risk for cardiovascular disease. In fact, the leading cause of death for people with diabetes is cardiovascular disease. The cardiovascular disease risk associated with metabolic syndrome is about two times higher than in the general population, and with progression to diabetes, the risk increases to about four times higher.

One clinical rule of thumb is that in estimating the cardiovascular risk of a patient with diabetes, we can take it to be equivalent to that of a person without diabetes with a low-density lipoprotein cholesterol (LDL-C) level approximately twice as high. That is consistent with clinical trial experience. Also, patients with diabetes in statin trials who receive the active treatment have a higher event rate than patients without diabetes who are given the placebo. This shows that patients with diabetes, even if they are taking a statin, have very high residual risk. Therefore, I believe this is a population that needs much more risk factor modification, in addition to statin therapy, to treat that residual risk.

We are going to discuss the thiazolidinedione (TZD) class of drugs. What do we know thus far about TZDs with regard to diabetes and cardiovascular disease risk?

Dr. Davidson: TZDs are insulin sensitizers and they have many so-called “pleiotropic effects” that are potentially beneficial. They're good at lowering blood glucose, maybe not as rapidly as a sulfonylurea, but over time they appear to provide good sustained glucose control. The cardiovascular benefit of TZDs derives from the fact that they may exert an antiatherosclerotic effect by reducing insulin levels. They also have a moderate positive effect on lipids, in that they lower triglyceride levels and raise high-density lipoprotein (HDL-C) levels. They tend to raise LDL-C levels, but they also tend to convert small LDL particles to less atherogenic large LDL particles. Therefore, the actual amount of circulating apolipoprotein B will be either decreased or unchanged, even if LDL-C levels themselves increase. In these ways they may modify the lipid pattern beneficially, making it potentially less atherogenic. In animal models they have some effect on blood pressure as well as apparent antiatherosclerotic effects.

What is the clinical trial experience to date regarding TZDs and atherosclerosis in humans?

Dr. Davidson: One important trial is PROactive (Prospective Pioglitazone Clinical Trial in Macrovascular Events),² a European study in which patients with diabetes and heart disease were randomized to pioglitazone or placebo in addition to usual therapy, both groups to be treated according to the 1999 International Diabetes Federation guidelines. In that study, glucose control was better, by about 0.5% of hemoglobin A_1c (HbA_1c), in the pioglitazone group, and there were small improvements in lipids and blood pressure. However, the difference in the predefined end point, which was a very large composite of cardiovascular end points, including some noncoronary vascular events such as amputation and leg revascularization, did not reach significance, although there was a reduction. Nonetheless, the traditional major adverse coronary events, such as myocardial infarction and coronary revascularization, were reduced by 16%, which was statistically significant compared with placebo, in 3 years, a relatively short period of time. However, there was an offsetting increased risk for heart failure, edema, and weight gain, which are the issues that we have to deal with when we consider the safety of TZDs. Overall, in my opinion, the PROactive trial had a positive cardiovascular outcome.

What we don't know so far is what beneficial effects TZDs may have that are independent of glucose lowering, the so-called pleiotropic effects. That question has been addressed to some extent by the CHICAGO (Carotid Intima-Media Thickness in Atherosclerosis Using Pioglitazone) study, in which progression of atherosclerosis was evaluated in patients achieving equivalent glucose control with either pioglitazone or a sulfonylurea in addition to metformin, and by the DREAM (Diabetes Reduction Assessment Using Ramipril and Rosiglitazone Medication) study using rosiglitazone,³ a different TZD, in patients with impaired fasting glucose or impaired glucose tolerance. The DREAM results showed a reduction in the onset of diabetes but not a cardiovascular outcome benefit.

Can you tell us about the CHICAGO study?

Dr. Davidson: The CHICAGO study was designed to examine the potential antiatherosclerotic effect of pioglitazone in patients with type 2 diabetes who had equal glucose control with either glimepiride, a sulfonylurea, or pioglitazone in addition to metformin therapy. Carotid intima-media thickness (CIMT), a well-known surrogate for atherosclerosis, was used to evaluate the outcomes. We wanted to test whether treatment with this TZD, pioglitazone, could result in less progression of atherosclerosis as measured by CIMT, compared with non-TZD drug treatment that achieved equally good glucose control.

Please tell us about the patients who were recruited in this trial.

Dr. Davidson: They were patients with type 2 diabetes from the metropolitan Chicago area, with an average age of about 60 years and duration of diabetes of around 90 months—so, a fairly long duration of diabetes. They could be receiving a broad variety of different therapies. Most of them were taking metformin or other antidiabetic drugs, but some were naive to treatment with an oral antidiabetic agent. Patients who had taken a TZD within the previous 12 weeks could not be enrolled. The average HbA_1C level was about 7.4, representing rather good control compared with other clinical trial populations, and the majority, about 60%, were taking statins. In addition, about 60% were taking an angiotensin-converting enzyme inhibitor or an angiotensin-receptor blocker. Therefore, it was a population with relatively well-controlled type 2 diabetes.

Therefore, a lower-risk population compared with some other trials, such as PROactive?

Dr. Davidson: Yes, PROactive enrolled patients with heart disease. In CHICAGO, the patients did not have heart disease.

Can you explain the study end points? What were you looking for in this trial?

Dr. Davidson: The primary end point was the CIMT, which was measured 2 centimeters from the bifurcation of the posterior wall, which is the point at which CIMT can be measured most accurately and with the least variability. All of the scanning was performed at a single center by the same sonographers using the same equipment, and the same blinded reader evaluated all of the scans.

We had other end points, including maximal CIMT and cardiovascular outcomes, as well. Although the trial was not statistically powered to assess clinical outcome, we did evaluate that as a secondary end point.

Has there been experience with CIMT as an end point in previous TZD trials?

Dr. Davidson: Yes. There have been at least 3 trials looking at CIMT.^4-6 There were 2 small trials using pioglitazone, with relatively short durations, which suggested that a benefit could be achieved.^5,6 However, a larger trial using troglitazone, which has been taken off the market, followed up 300 patients for 2 years and showed no effect.⁴ That study included a predominantly Hispanic population. The other 2 trials also had homogeneous populations, one Japanese⁵ and the other German,⁶ so it is important that CHICAGO truly represented the racially and ethnically diverse population of a large US city and reflected the diversity of the population of people with type 2 diabetes in the United States.

And what did the CHICAGO results demonstrate?

Dr. Davidson: We showed that there was a significant difference in mean CIMT progression between the pioglitazone-treated group and the glimepiride group. There was virtually no progression in the pioglitazone group over 18 months, maybe even a slight regression, whereas the glimepiride group showed a significant increase in CIMT over time. The mean difference at the end was –0.013 mm (P = .02). This was a significant difference achieved in a relatively short period of time, in a population of patients who all had well-controlled diabetes.

What was shown with regard to secondary end points?

Dr. Davidson: The maximal CIMT was reduced to a greater degree because it was thicker to begin with, and that difference was also significant. We also saw a difference in cardiovascular outcomes, although the study was not powered to evaluate that end point; there were 10 events in the glimepiride group compared with 4 in the pioglitazone group. As for adverse effects, there was a little more hypoglycemia in the glimepiride group. In the pioglitazone group, there was a little more edema and 1 hospitalization for heart failure among more than 200 patients, and 1 noncardiovascular death, due to pancreatic cancer, in an 80-year-old woman. So, on the whole, safety was comparable between the 2 arms, but cardiovascular benefit definitely showed a trend in favor of the pioglitazone arm.

How was HbA_1c affected?

Dr. Davidson: There was a slight reduction, of 0.2%, in HbA_1c at the end of the study in the pioglitazone group. The difference did not appear until 48 weeks. By then, the maximal CIMT was already significantly different. Therefore, we didn't achieve quite what we hoped with regard to equal glucose control, and the pioglitazone group had somewhat better control. In my opinion, however, because the difference in glycemic control appeared so late in the study, it cannot explain the CIMT differences. They really were due to other effects of the drug, independent of glycemic control.

And how was HDL-C affected?

Dr. Davidson: HDL-C rose by about 6 mg/dL in the pioglitazone group, almost a 15% increase. Triglycerides were reduced by 13%, with a treatment difference of 15% compared with the glimepiride group. LDL-C actually rose by 7%, but the increase was not significant, and blood pressure increased slightly and not significantly. The main lipid response was the HDL-C increase.

In the pioglitazone group, it seems that the change in mean CIMT was negative. Was there actually regression?

Dr. Davidson: I think it's more likely lack of progression. The mean value did fall below the baseline value at the end of the study, but the baseline and final values were very similar, basically just indicating lack of progression at 18 months.

Overall, how important is that primary end point, clinically speaking?

Dr. Davidson: I think it's important to point out that this result cannot be compared with the results of other CIMT trials in an apples-with-apples comparison. CIMT was measured at the posterior wall, the thinnest part of the plaque, but the part with the least variability. We used an automated edge detector to get higher to the difference of measurements, which can dampen down an effect, but it gives us less variability of the measurement than manual measurements. What I think it important to point out is that there was a significant difference in plaque progression over time, which I consider to be an excellent surrogate determinant of cardiovascular risk and predictor of cardiovascular events. I think the trend in a positive direction in cardiovascular outcomes supports the conclusion that pioglitazone can provide cardiovascular benefit that is due to non–glucose-related effects.

Was this benefit seen across subgroups?

Dr. Davidson: Yes, we had set some prespecified subgroups initially: age, gender, HbA_1c above or below 7% at baseline, and, most important, statin use. In the PROactive trial, the statin-treated patients had no benefit, but in this trial, the statin users and nonusers both had equal benefit with pioglitazone, as did all the subgroups.

Have you drawn any conclusions about the mechanism of action by which pioglitazone exerts this antiatherosclerotic effect?

Dr. Davidson: We can't say for sure. Being a lipidologist, I am inclined to the view that the HDL-C increase is driving the benefit, but that is something worth looking at in further analyses.

You mentioned the safety profile. In CHICAGO, as compared with PROactive, there appeared to be less concern regarding heart failure. Can you comment on that?

Dr. Davidson: We had a lower-risk population, without heart disease, and a somewhat lower dose of pioglitazone—a maximum of 45 mg/d, with a mean dose slightly over 30 mg—as compared with PROactive. That there was only 1 case of heart failure among more than 200 patients treated with pioglitazone should be reassuring for clinicians who are considering this class of drugs for treating patients with diabetes.

In PROactive, there was also some slight concern about malignant neoplasms. Was there any concern of this kind in CHICAGO?

Dr. Davidson: No. We saw 1 case of pancreatic cancer in an 80-year-old woman. Since cancer is initiated several years before it becomes clinically evident, pioglitazone really cannot be implicated as a cause in the 1 case of cancer seen in this relatively short-term study.

Finally, Dr. Davidson, what do you think are the key clinical implications of the CHICAGO results?

Dr. Davidson: I would like to refer to the ADA (American Diabetes Association) guidelines, which say that we should start metformin as early as possible in the treatment of patients with diabetes but leave the decision about second-line treatment to clinical judgment. The options include a sulfonylurea, a TZD like pioglitazone, and even insulin. The CHICAGO study results, in my opinion, because of the cardiovascular benefits that we saw in the pioglitazone-treated group, tip the balance toward a TZD like pioglitazone as a second-line agent for patients with diabetes that is not adequately controlled with a single drug. Most patients with type 2 diabetes will require 2 agents to achieve glucose control. I think this study provides a rationale for selecting the TZD option for those patients.

Disclaimer

This editorial is intended for use by health care professionals for educational purposes only. The information conveyed and opinions expressed by the author in this editorial represent his own views, which do not necessarily reflect those of the APOLLO Steering Committee, The FCG Institute for Continuing Education, or AstraZeneca Pharmaceuticals LP. The posting of this editorial on cvspectrum.org does not imply endorsement either of the content or of the viewpoint expressed by the author.

References

1. Mazzone T, Meyer PM, Feinstein SB, et al. Effect of pioglitazone compared with glimepiride on carotid intima-media thickness in type 2 diabetes. JAMA. 2006;296:(doi10.1001/jama.296;21.jo60138).

2. Dormandy JA, Charbonnel B, Eckland DJ, et al. Secondary prevention of macrovascular events in patients with type 2 diabetes in the PROactive Study (PROspective pioglitAzone Clinical Trial In macroVascular Events): a randomised controlled trial. Lancet. 2005;366:1279-1289.

3. Gerstein HC, Yusuf S, Bosch J, et al. Effect of rosiglitazone on the frequency of diabetes in patients with impaired glucose tolerance or impaired fasting glucose: a randomised controlled trial. Lancet. 2006;368:1096-1105.

4. Hodis HN, Mack WJ, Zheng L, et al. Effect of peroxisome proliferator-activated receptor gamma agonist treatment on subclinical atherosclerosis in patients with insulin-requiring type 2 diabetes. Diabetes Care. 2006;29:1545-1553.

5. Minamikawa J, Tanaka S, Yamauchi M, Inoue D, Koshiyama H. Potent inhibitory effect of troglitazone on carotid arterial wall thickness in type 2 diabetes. J Clin Endocrinol Metab. 1998;83:1818-1820.

6. Langenfeld MR, Forst T, Hohberg C, et al. Pioglitazone decreases carotid intimamedia thickness independently of glycemic control in patients with type 2 diabetes mellitus: results from a controlled randomized study. Circulation. 2005;111:2525-2531.