Abiomed Presents Final PROTECT II Results for High-Risk PCI Patients

April 3, 2011—Abiomed, Inc. (Danvers, MA) announced that principal investigator William O'Neill, MD, presented the final results from the PROTECT II study at the American College of Cardiology's (ACC) 60^th annual scientific session in New Orleans.

The PROTECT II study was a prospective, multicenter, randomized trial designed to measure a composite of major adverse events (MAE) at 30 days, with 90-day follow-up in high-risk percutaneous coronary intervention (PCI) patients requiring hemodynamic support, comparing the company's Impella 2.5 to the intra-aortic balloon pump (IABP).

A company press release included data for a per-protocol (PP) population and noted the intent-to-treat (ITT) population. For purposes of the ACC presentation and future publications, both groups will be reported. The PP population includes patients who received treatment and met all the prespecified inclusion and exclusion criteria of the PROTECT II study. The PP population was prespecified, and patients were identified prospectively before the statistical analysis. The Impella product is a protocol-driven technology used for patients that meet the criteria of requiring hemodynamic support.

These reported data have been collected, monitored, and analyzed by a third-party academic research organization. The MAE were assessed at 30 and 90 days, Abiomed stated. The results included data from the entire study population and demonstrated a significant positive overall outcome for PP patients treated with Abiomed's Impella 2.5 at 90 days (P = .029).

According to Abiomed, the PROTECT II data demonstrated positive overall outcomes for the Impella arm over the IABP arm in the entire study cohort at 90 days, resulting in a 21% reduction in MAE over the IABP (P = .029). In the prespecified high-risk PCI without atherectomy subgroup (88% of study), Impella provided a significant benefit over the IABP at 30 days (P = .009) and at 90 days (P = .003) with a 29% reduction of MAE.

In the prespecified atherectomy group (12% of study), there was no overall statistical difference (P = .316) in MAE at 90 days. In this subgroup, the Impella arm demonstrated a significant increase in periprocedural creatine kinase-MB isoenzyme release (P = .03) and decreased repeat revascularization at 90 days (P = .006). Impella significantly reduced out-of-hospital MAE overall by 56% (P = .002) over the IABP arm.

Per the study conducted by an independent economic health organization (n = 227), the Impella reduction of MAE translated into overall lower hospital charges per patient, ranging from $12,000 (all patients) to $17,000 (survivors only) at 90 days. More economic analysis data will be presented in future health care forums highlighting reduced repeat revascularization, lower charges per readmission and reduced critical care length of stay.

Abiomed advised that the PROTECT II study was prematurely halted in December 2010 per the recommendation of the Data Safety Monitoring Board based on the 50% interim enrollment data, with the assumption that it would not meet its primary endpoint at 30 days.

According to the company, the treatment effect of Impella over IABP improved during the course of the trial, suggesting a learning curve. The superior hemodynamic support of Impella appears to have led to significant procedural differences between the two arms. The original study was powered at 80%, with the intended final population of 654 patients. The interim report included the 50% mark, and final analysis included an additional 19% of patients. The prespecified “roll-in subject” analysis demonstrated that, without the first patient in both arms, Impella was significantly better at 90 days (n = 307; P = .027).

The PROTECT II study PP results at 30 days compared the MAE rates for Impella-treated patients (n = 215) versus IABP-treated patients (n = 211). The overall MAE were 34.9% for patients treated with Impella versus 42.7% for patients treated with IABP (P = .1). For patients without atherectomy (88% of study), the rates were 29.5% for Impella patients versus 42.4% for IABP patients (P = .009). For patients with atherectomy (12% of study), the rates were 65.6% for Impella patients versus 45% for IABP patients (P = .143). The ITT results showed 30-day MAEs as similar (P = .312 for all patients, P = .057 without atherectomy, and P = .143 with atherectomy).

At 90 days, the overall PP results for Impella-treated patients (n = 213) versus IABP-treated patients (n = 210) showed MAE rates of 40.8% versus 51.4% (P = .029). For patients without atherectomy (88% of study), the event rates were 35.9% versus 51.1% (P = .003). For patients with atherectomy (12% of study), the rates were 68.8% versus 55% (P = .316). The ITT results showed 90-day overall MAE as similar (P = .087, for all patients, P = .015 without atherectomy, P = .316 with atherectomy).

“The PROTECT II study is a major advance that will help guide how interventional cardiologists treat the high-risk PCI patient population,” commented Dr. O'Neill. “At 90 days, there is a very significant advantage to the Impella-treated patients. As evidenced by this data, this superior level of hemodynamic support in the cath lab allowed for a more complete procedure, leading to a reduction in out-of-hospital MAE.”

Abiomed stated that the PROTECT II study is notable because of the complexity and advanced disease of the patient population receiving PCI requiring hemodynamic support (approximately 60% were turned down for surgery), the number of endpoints measured prospectively at 30 and 90 days post-hemodynamic-supported PCI, the analysis of use of atherectomy in low ejection fraction PCI patients, and that it measured IABP MAE.