Experts put FREEDOM findings in perspective

Despite significant advances in the field of revascularization in recent years, including the introduction of drug-eluting stents, data from the highly anticipated FREEDOM trial document that CABG is the optimal treatment strategy for the intermediate term among patients with diabetes and multivessel CAD.

These results generated a lot of attention after the presentation by Valentin Fuster, MD, PhD, physician in chief at Mount Sinai Medical Center and director of Mount Sinai Heart Institute in New York, at the American Heart Association Scientific Sessions 2012 and simultaneous publication in The New England Journal of Medicine.

“The FREEDOM trial was the culmination of almost 7 years of work aimed at addressing the question of how to treat patients with diabetes in multivessel disease,” Spencer B. King III, MD,Cardiology Today Editorial Board member and president of Saint Joseph’s Heart and Vascular Institute in Atlanta, said in an interview.

“The question has been answered many times for balloon angioplasty and earlier stenting techniques, but in the current era, the question remained: Would DES make PCI an appropriate choice compared with CABG for patients with diabetes and multivessel disease?” said King, who was a member of the trial steering committee.

Interviewed experts highlighted other questions raised by the FREEDOM trial, including the need for further examination of different approaches to PCI and how patients fare after long-term follow-up, which is being planned by the FREEDOM investigators.

Results from the clinical trial

FREEDOM was designed against a specific background, according to Fuster.

“Of patients who undergo catheterization and coronary intervention, about 25% have diabetes and multivessel CAD,” he said in an interview. “This is a large population, and there is some suggestion that, in this group of patients, bypass surgery is superior to PCI. We had to find an answer to this.”

Fuster and colleagues designed FREEDOM as a 5-year superiority trial. The investigators randomly assigned 1,900 patients (mean age, 63 years; 29% women) with diabetes and multivessel CAD to CABG or PCI with a DES. Eighty-three percent of patients had three-vessel disease. Patients were followed for a minimum of 2 years; median follow-up was 3.8 years. The study was conducted at 140 centers worldwide between 2005 and 2010.

Results showed that the primary outcome — first occurrence of all-cause mortality, nonfatal MI or nonfatal stroke — occurred more frequently in those who underwent PCI vs. CABG. Five-year rates were 26.6% in the PCI group vs. 18.7% in the CABG group (P=.005). MI rates also significantly favored CABG over PCI (6% vs. 13.9%; P<.0001), as did all-cause mortality (16.3% vs. 10.9%; P<.001). Stroke, however, occurred more frequently in the CABG group (5.2% vs. 2.4%; P=.03).

Importantly, Fuster said, these results were consistent across all prespecified subgroups, including across complexity of disease. Event rates were higher in patients assigned PCI as compared with patients assigned CABG for those with SYNTAX scores of <22 (23.2% vs. 17.2%), 23 to 32 (27.2% vs. 17.7%) and >33 (30.6% vs. 22.8%).

“In patients with diabetes and multivessel coronary disease, CABG was of significant benefit as compared with PCI,” Fuster said during a press conference at the AHA Scientific Sessions 2012. “CABG surgery is the preferred method of revascularization for patients with diabetes and multivessel disease.”

Cost-effectiveness analysis revealed differences

“This prospective economic evaluation was carried out alongside the FREEDOM trial to provide additional insight into the relative value of CABG vs. PCI in the drug-eluting stent era from the perspective of the US health care system,” Elizabeth A. Magnuson, ScD, of Saint Luke’s Mid American Heart Institute in Kansas City, Mo., said during a press conference at AHA Scientific Sessions 2012.

Due to the prolonged recovery period and higher complication rates, the initial cost of CABG was $8,622 higher than PCI (P<.001). During 5 years of follow-up, high rates of resource use in the PCI arm, largely due to a greater number of repeat revascularization procedures and CV hospitalizations, significantly increased costs associated with the procedure. After accounting for these added costs, CABG costs remained $3,600 higher, according to Magnuson.

Patients who underwent CABG, however, had lower risk for death or MI compared with those in the PCI group. CABG was associated with 0.66 quality-adjusted life-years (QALY) gained and higher costs of approximately $5,400 per patient. Lifetime cost effectiveness of CABG was therefore $8,132 per QALY gained — significantly lower than the commonly used benchmark of $50,000 per QALY gained for considering a treatment to be cost effective, Magnuson said.

“For patients with diabetes and multivessel CAD, CABG provides not only better long-term clinical outcomes than PCI with DES, but also these benefits are achieved at an overall cost that represents an attractive use of societal health care resources. These findings provide additional support for existing guidelines that recommend CABG for diabetic patients with multivessel CAD,” she said.

The in-trial analysis reported by Magnuson was based on observed survival, health state utility and costs derived from observed health care resource use through 5 years in the FREEDOM trial and a lifetime analysis based on projections of survival, quality-adjusted survival and costs beyond 5 years.

Economic implications

In today’s fiscal climate and with the rising costs of medical care, evaluating the economic implications of different therapies is particularly relevant, said Mark Hlatky, MD, who served as a discussant on the cost-effectiveness analysis.

“Randomized trials are the best way to compare these therapies, and you can do a cost-effectiveness analysis alongside a randomized trial,” said Hlatky, who is professor of health research and policy and of cardiovascular medicine at the Stanford University School of Medicine and director of the Stanford-Kaiser Cardiovascular Outcomes Research Center.

Nevertheless, Hlatky said trial design must be considered because the outcomes of both the treatment and costs may differ from those observed in the study and therefore may not be generalizable in real-world situations. In particular, with trials involving surgery, upfront costs are easily documented during the trial period. The full benefits of the therapy, however, may not be observed within the follow-up period of a clinical trial.

“This is particularly so in FREEDOM because the survival curves for CABG and PCI didn’t separate until several years into follow-up. Consequently, the researchers appropriately performed a lifetime projection to account fairly for the effectiveness of CABG in reducing mortality,” Hlatky said. The results of which, he noted, were credible because they were based on the long-term consequences of the clinical outcomes observed during the trial period.

“With the observed differences in survival in the trial, coupled with the cost difference of about $3,600, CABG clearly provides good value for the money spent for patients who have diabetes and multivessel coronary disease,” he said.

Potential to affect clinical practice

Many experts in the field of cardiology said the results of the FREEDOM trial have the potential to significantly affect clinical practice.

“The initial finding of surgery being superior to angioplasty for diabetic patients with multivessel disease was hinted at in the NHLBI-sponsored EAST trial and confirmed in the BARI trial,” King said. “Now, with continuing improvement in surgery and PCI, the results seem remarkably the same after more than 20 years.”

King also said FREEDOM reinforces the need to carefully consider diabetes when evaluating treatment options.

“The results of this trial will reaffirm the importance of diabetes in decision-making regarding patients with multivessel disease. Some have come to believe that it was not the diabetes but rather the extent of disease in these patients that favored surgery. This trial provides strong evidence that the fact of diabetes remains important in decision-making, even in the era of very effective drug-eluting stenting,” King said.

Additionally, results underscore the importance of aggressive medical treatment and prevention before CABG or PCI, especially in terms of the progression of diabetes and its comorbidities, according to Fuster. Patients should not, he said, reach advanced stages of diabetes, hypertension and hypercholesterolemia.

“We have to put a lot of attention into how we prevent this late stage of disease from happening by taking care of obesity. About 70% of these patients were obese … so we have a lot to do to prevent this,” he said.

Further, Fuster said FREEDOM highlights the value of collaboration.

“We showed that interventionalists ... are able to work with surgeons, as is happening in percutaneous valve replacement,” he said. “This is a team effort.”

Currently, guidelines suggest CABG over PCI for patients with diabetes and multivessel disease, but the FREEDOM trial may influence stronger recommendations as well, according to George 
W. Vetrovec, MD, professor of medicine and chief of the cardiac cath lab at Virginia Commonwealth University. Nevertheless, the results should be evaluated in the context of the trial itself.

Alice K. Jacobs, MD, professor of medicine at Boston University School of Medicine, also discussed the need for further research.

“Whether, given the totality of evidence, the new data from FREEDOM will inform a Class I guideline recommendation in patients with triple-vessel disease and complex disease will no doubt be the subject of ongoing debate,” she said during her presentation as a discussant on the trial at AHA Scientific Sessions 2012.

Lingering questions

Despite data that largely support the use of CABG in patients with diabetes and multivessel disease, the FREEDOM results also raised some concerns.

For instance, Vetrovec said not all outcomes associated with CABG were positive.

“While surgery wins in terms of the primary endpoint, mortality endpoints for CABG are not a home run” he said. “Stroke risk was significantly higher in the CABG group. There’s a price for living longer, and that’s a 5% risk of having a stroke. Also, if you look at CV death, that trended toward surgery, but was not significant.”

Further, Vetrovec said the patient population studied may not necessarily be representative of all patients with diabetes and multivessel CAD. He cautioned against broad interpretation of the FREEDOM results and considering them applicable to all patients with diabetes and multivessel disease.

“FREEDOM represents only about 10% of the patients that the researchers screened as potential candidates. Then, there was even further reduction because of consent or other issues. So, not every diabetic is a FREEDOM patient. It’s about doing the best job of individualizing treatment for your patient,” he said.

In terms of future research, Jacobs emphasized the need for evaluating patient outcomes after more time has passed.

“It will be important to determine whether the relationship between the primary endpoints curves is maintained with longer-term follow-up as the saphenous vein bypass grafts begin to fail, and whether the continued evolution of new DES technology will diminish the advantage of CABG is unclear. But it does appear less likely if the mechanism of the effect of CABG is to protect the myocardium against new disease,” she said.

King also emphasized the need for longer-term follow-up, and noted that the duration of the FREEDOM trial may have favored CABG. “The time course of the current trial is optimal for surgical outcomes,” he said. “A longer follow-up of 7 to 10 years will be interesting, as this is a period when the concern about vein graft closure becomes more pertinent.”

Certain patients, regardless of having diabetes and multivessel disease, may experience better outcomes with PCI vs. CABG, King said.

“As with all trials, the results of FREEDOM express the average outcome, and even though there was no interaction based on SYNTAX scores, there remain patients with diabetes and multivessel disease who are best served by stenting based on their anatomic suitability and increased surgical risk,” he said. In addition, the issue of providing the internal mammary artery graft to the anterior descending artery and combining that with a hybrid approach to stenting other vessels has not been adequately investigated, especially in patients with diabetes.

Evaluating patients according to SYNTAX score could provide clinicians with more insight into the best treatment for those with diabetes and multivessel disease.

“Future research perhaps should concentrate on understanding which patients with very low SYNTAX scores and high-grade but discrete stenoses should still undergo stenting as the preferred intervention,” he said. “Many patients, of course, were not included in the trial, and the disclaimer about increased surgical risk should be part of any guideline statement.”

Overall, research into better management of CV risks and the consequences of diabetes is essential.

“No matter whether you look at bypass surgery or angioplasty, compared with nondiabetics, diabetics do worse. The research clearly needs to focus on figuring out what the incremental problem is with these patients and how to resolve it,” Vetrovec said.

Lingering questions aside, FREEDOM is an important trial, according to King. “This adequately powered National Heart, Lung and Blood Institute trial was pivotal and definitive for current practice,” he said.

“I hope the interventionalist community will look at these results as they are,” Fuster said.

Disclosure: Fuster, Hlatky, Jacobs, King and Vetrovec report no relevant financial disclosures. Magnuson reports receiving research grants from Abbott Vascular, AstraZeneca, Boston Scientific, Daiichi Sankyo, Edwards Lifesciences, Eli Lilly and Medtronic.