Dive Brief:

• Members of an FDA advisory committee have found fault with the case made by Keystone Heart at a meeting to discuss its Class II medical device.  Keystone is owned by Venus Medtech, which calls itself the biggest transcatheter company in China.
• FDA convened the panel to assess if the clinical data support the substantial equivalence of Keystone's Triguard 3 cerebral embolic protection device to Boston Scientific's Sentinel. No vote was held on whether Keystone showed substantial equivalence but the tone of the debate points to challenges for the company.
• Feedback from some panelists was particularly negative, with one expert saying there is "compelling evidence that [Keystone Heart] not only didn't demonstrate efficacy, but there's a potential signal of harm."

Dive Insight:

Keystone is seeking 510(k) clearance to market Triguard 3 as a way to minimize the risk of cerebral damage caused by embolic debris during transcatheter aortic valve replacement. The company filed for 510(k) clearance, citing Boston Scientific's Sentinel as the predicate. 

FDA asked the experts to comment on whether the study data show Triguard 3 "satisfactorily meets" a secure positioning and stability special control needed to support substantial equivalence. The panelists spoke around the topic without directly answering whether the device meets the special control.

Richard Lange, the Texas Tech University Health Sciences Center physician who chaired the panel, summarized the views of other experts on the panel about positioning and vessel coverage. 

"Forty-five percent had incomplete coverage, 20% had no coverage and 10% [had] interference with a TAVR device, which is of concern," Lange said. "Sometimes it flipped, they had used another wire. The manufacturer has now made changes to it. Unfortunately, those changes have just been studied in a single-center study, which isn't a real-world experience, and it doesn't have the rigors of the current study."

Lange said "we just don't know" whether the changes made by Keystone will be enough to assure complete coverage. The inability to form firm conclusions based on the data was a theme that ran through the meeting, with other panelists noting the "dearth of information" and saying the "data is very poor."

Karen Furie, chief of neurology at Rhode Island Hospital, gave one of the strongest critiques of the data. Furie, like other panelists, focused on the overall population of patients in the trial, rather than the subpopulation Keystone zeroed in on to demonstrate efficacy. 

"Importantly, as we saw with the safety data, if you look at the NIH stroke scale worsening and new neurological impairment ... it's going in the wrong direction. The patients who got the device did worse clinically. To me, that's again more compelling evidence that they not only didn't demonstrate efficacy, but there's a potential signal of harm there," Furie said.

What that means for Keystone's prospects remains unclear. As industry representative Gary Jarvis, who works at Alfa Medical, said, the committee was convened to look at the substantial equivalence of a 510(k) but ended up discussing data mining and other topics that may be outside of the scope of FDA's decision about whether to clear Triguard 3.

"When we look at the trials themselves, both with the [Triguard 3] device and the Sentinel devices, they both met their pre-specified endpoints for safety. So we had a lot of discussion about safety, but those were already predefined and they were met. That in itself could potentially mean substantial equivalent just because they met the safety standard," Jarvis said. 

FDA will now review the feedback provided by the panelists before deciding on the next steps. The agency typically takes the advice of its advisory committees, though is not bound to do so.