FDA's Circulatory System Devices advisory panel is encouraging TherOx to sponsor a new clinical trial of its SSO2 SuperSaturated Oxygen Therapy Downstream system to demonstrate the therapy improves clinical outcomes in heart attack patients.

At its March 18 meeting in Gaithersburg, Md., the panel voted 9-5 to recommend FDA not approve a PMA for the device. The majority were unconvinced that the targeted oxygen therapy improves clinical patient outcomes, even though trial data showed the therapy created a statistically significant reduction in the surrogate endpoint of myocardial infarction size.

The TherOx SSO2 is a catheter-based device designed to help cardiovascular interventionists reduce the size of an infarct and preserve heart muscle by delivering oxygen-enriched blood directly to targeted coronary arteries immediately following percutaneous coronary intervention such as stenting.

"Too many times in cardiovascular medicine we have gone on the wrong path to chase surrogate [end points] and ended up with less than good circumstances," said panelist Clyde Yancy, M.D., Baylor Heart and Vascular Institute in Dallas. "Clinical outcomes data continue to trump everything else."

"I feel for the sponsor, and I also respect their discipline," said Yancy, who voted against approval. Because the consequences can be so profound, "we do need an additional therapy" for acute myocardial infarction patients, "but we need a higher threshold of evidence."

AMIHOT II Data Fail To Persuade Panel

TherOx' PMA is based on the results of the Acute Myocardial Infarction with HyperOxymic Therapy (AMIHOT II) trial, which followed 222 patients treated with SS02 therapy and 79 controls for 30 days, or through hospital discharge.

The primary effectiveness endpoint was superiority of infarct size reduction over PCI with stenting alone at 14 days post procedure, as measured by single photon emission computed tomography (SPECT). The study showed a 6.5% median reduction in the size of a left ventricle infarct in the SSO2 arm versus the control group.

The company said it chose the endpoint for efficacy based on prior studies indicating that a 5% median infarct size reduction represents a "clinically meaningful" reduction in the amount of dead heart muscle following a heart attack. However, several panelists pointed out that although SS02 produced statistically significant changes in infarct size, the differences were small.

David Naftel, Ph.D., of the University of Alabama at Birmingham, said he voted in favor of approval because TherOx had worked well with the FDA and had met the study's endpoints, though he acknowledged that the trial results were not impressive. He added that he had hoped "it would just jump out that this was an incredible technology and that everybody benefited. I didn't see that."

The 30-day major adverse cardiac event rate was 5.4% in the SSO2 therapy group versus 3.8% in the control group, a statistically non-inferior difference.

Adverse events in the AMIHOT II trial included four patient deaths in the treatment group within 30 days versus none for the controls, a statistically insignificant difference.

Although the trial met its safety endpoints, several panelists suggested the negative safety trend influenced their opinion about the overall value of SSO2.

Valluvan Jeevanandam, M.D., Columbia University, suggested that he might have voted to approve SSO2 with the conditions that the device be indicated only for a small subset of heart attack patients when the procedure begins within three hours of symptom onset.

According to TherOx, SSO2 therapy takes about 90 minutes and should be performed within six hours of a heart attack.

But Jeevanandam said he was deterred by a "hint of safety issues" in the clinical data and the lack of robust clinical efficacy data.

Judah Weinberger, M.D., of Columbia University, told TherOx executives that he was not happy making his decision to recommend against approval for the SSO2 system, since the trial met its primary endpoint.

"I think that we look at it with eyes in 2009 that might be a little different than when you looked at it with eyes in 2005," when FDA and TherOx agreed on the study design. But without more evidence of a clinical benefit, SSO2 does not meet "the statutory burden of risk/benefit," Weinberger said.

The hypotheses for the AMIHOT II trial were derived from the earlier AMIHOT I clinical trial - which had failed to meet its endpoints - using a Bayesian statistical analysis that was formulated with help from FDA.

The agency has had draft guidance out on the use of Bayesian models in clinical trials since 2006 but has not yet finalized it (¹ Also see "Device Center Enthusiastic About Bayesian Trial Submissions" - Medtech Insight, 31 July, 2006., p. 7).

Yancy asked FDA to consider whether the TherOx PMA process might better inform the agency about the ongoing application of Bayesian analysis in clinical trial design, although he did not go so far as to say its use should necessarily be a disqualifier.

Panelists Offer Advice For A New Trial

During their discussions, some panel members acknowledged the logistical challenges a small company faces in conducting new trials.

"I do hope the sponsor at this point finds the means and continues on this as they have for 15 years," said panelist John Somberg, M.D., of Rush University Medical Center in Lake Bluff, Ill.

He advised TherOx to conduct a new clinical trial enrolling the sickest population with the greatest potential for adverse outcome potentials and compare the clinical benefits of SSO2 versus a control.

Yancy argued that the least burdensome approach to proving the value of SSO2 therapy would be a single-arm study using historical data as a control.

However, Michael Domanski, M.D., National Institutes of Health, argued a single-arm study would not be adequate. He recommended developing "a careful clinical endpoints trial to clear the air. I don't think a single-arm study is going to tell you anything."

Panel chairman Jeffrey Borer, M.D., Weill Cornell Medical College, New York, said that without a control, "the possibility of unintentional selection bias would be of great concern."

AMIHOT II primary investigator Gregg W. Stone, M.D., Columbia University Medical Center, argued that any unresolved questions about SSO2 would be better answered in a post-market study rather than another pre-market study that would delay getting the therapy on the market. He called the SSO2 system "an important device for the high-risk patients for whom there is no therapeutic alternative."

Stone said that the rates of MACE, death, myocardial infarction, target vessel revascularization and stroke "really were within statistical comparability within the confines of this trial" and that the rate of "safety events, even the mortalities, was lower than you might have expected in this group."

To show superiority for MACE, the company would have to study "thousands and thousands of patients," Stone said.

In addition to offering guidance for new clinical trials, the panel advised TherOx to develop a comprehensive user-training plan. Borer pointed out that training is especially important to preventing safety problems since this therapy is likely to be used only infrequently and physicians would not have a lot of practice with it.

The panelists also recommended that the company further examine the safety risks of stent occlusion and myocardial rupture with SSO2 therapy.

Sponsors of the device suggested that a larger, off-the-shelf catheter with a larger sheath may have played at least some role in early adverse events in the AMIHOT trials, and that the smaller catheter designed and subsequently approved for study with SSO2 likely led to better outcomes later.

Several panelists also suggested that TherOx collect new animal study data to better characterize the physiochemical mechanism of action of SSO2 therapy and the risk of oxygen toxicity in general.

Approval With Conditions Considered But Rejected

At certain points during the meeting, it appeared that the panel might be willing to recommend approval with the condition that more evidence be collected post-market. However, FDA representatives reminded the panel it had to vote strictly on the data at hand.

Three potential conditions were proposed in an earlier, failed vote, including limiting indications for SSO2 use to patients with a lesion to the left anterior descending artery, mandating a post-approval study and requiring the entire SSO2 procedure be completed within the cath lab rather than allowing a patient to be transported with an infusion catheter still in place.

However, when the motion to approve the device with conditions was raised, the majority of panelists voted against it.

TherOx Considers Its Options

Privately-held TherOx has been studying SSO2, its flagship product, since 1994 and started U.S. clinical trials of the system in 1999 (² Also see "TherOx Aqueous Oxygen System Offers Coronary, Cerebral Applications" - Medtech Insight, 8 February, 1999., p. 23).

The product received its CE mark in 2001.

Kevin Larkin, the firm's CEO, told "The Gray Sheet" that TherOx had already opened a dialogue with FDA following the panel's vote.

He believes the panel may not have properly appreciated the importance of infarct-size reduction in the trial. "The 6.5% reduction in infarct size is absolute ... that is a 25% reduction relative to the control group. If you take a 26% infarct size and you reduce it to under 20%, that is a 25% relative reduction. That is saving a quarter of the myocardium that was not saved in the control group," he said. "So maybe we didn't, as a company, do an adequate job of characterizing that 6.5% on other terms that the panel members could relate to."

"These are all things that we will explore with FDA," he said.

While TherOx would not rule out anything that helps bring the product to market, Larkin said the company would prefer to be able to do it without the need to partner with a larger device firm, but that the company "would not be bound and determined to do it alone."

TherOx filed a proposed initial public offering with the Securities and Exchange Commission last September but withdrew it last month amid weakness in the broader market.

- Monica Hogan (m.hogan@elsevier.com)

PMA Panel Asks TherOx For New Data On Cath-Lab-Based Oxygen Therapy

More from Archive

More from Medtech Insight