Fantastic Voyage

Once doctors had begun the early trials of the VNS device, they started to notice something else. Not only was the device cutting down on seizures but the patients "all perked up after getting the implant," says Dr. B.J. Wilder, a University of Florida neurologist, now retired, who did many of the earliest implants. "They were in a better mood."

That led to research into whether the VNS device might help patients who suffered from prolonged bouts of depression that didn't ease no matter what drugs they tried. Brain imaging showed that vagus-nerve stimulation was activating parts of the brain that influence mood, and there were also changes in the levels of certain feel-good neurotransmitters. The discovery certainly stimulated some good feelings at Cyberonics. The company figures the depression market is about 10 times the size of the epilepsy market.

One day in January 2002, Cyberonics executives gathered in CEO Cummins's corner office to await some of the early results from the first big study in depression. The study involved 235 patients. Half had received eight weeks of vagus-nerve stimulation on top of their usual regimen of pharmaceuticals, and the other half had served as a control group. They'd gotten the implant and were taking their drugs--but the implant had not been turned on during those first eight weeks.

As the fax machine began to spit out data, "you could've heard a pin drop," recalls Alan Totah, Cyberonics' vice president of regulatory affairs. "We said, 'Holy mackerel, what happened?' " The data didn't show any statistically significant difference between patients in the control group and those who'd actually been receiving vagus-nerve stimulation.

Everyone had a theory as to why the results were so bad. The electrical stimulation hadn't been high enough. Eight weeks of stimulation wasn't long enough to have a real impact. The patients in this study were more depressed than those in an earlier pilot study, which had shown better results.

Following the bad news, three Cyberonics executives who were involved with that study left the company. Richard Rudolph, Cyberonics' chief medical officer, argued to start another study from scratch. But that would've been time consuming and expensive, and Cummins didn't feel it was necessary.

Fortunately, a change in the clinical-trials strategy produced better data a year later. One out of every six patients who had gotten the VNS device found that their depression had vanished, or that they had only a few symptoms. Just one out of 18 patients in a control group saw that kind of improvement.

In October 2003, Cyberonics shipped the requisite 10 copies of its application to the FDA to try to gain approval for treating depression. Each copy of the application filled 87 three-ring binders. The agency had previously granted Cyberonics "expedited review" status, something it does for devices that may represent a breakthrough.

It's the first week of January 2004. Cyberonics has arranged a special conference call with the Wall Street analysts who follow the company's stock. Seated at a conference table in Houston with eight other executives, Cummins is wearing dark-blue pin-striped pants, a light-blue striped shirt, and a bright yellow tie. His head is shaved, and he's wearing rimless glasses. He looks as if he were dressed by a Hollywood costume designer for the role of someone who runs a bond-trading desk.

The call is a blend of good news and bad news. The company is lowering its earnings estimates for the last two quarters of its 2004 fiscal year--bad news that is saved for the very end of the call. The good news seems based largely on guesswork and wishful thinking. FDA approval for depression is expected to come relatively easily, since the device is the same as the one that has already been implanted in 22,000 epilepsy patients, and it will be made in the same FDA-inspected faci-lity at Cyberonics' Houston headquarters. And the FDA's review of Cyberonics' filing documents seems to be going smoothly.

When the FDA is ready to summon Cyberonics to a panel meeting, it will send a letter. As of January, the company hadn't received it. Still, on the conference call, Totah, the head of regulatory affairs, makes what amounts to an educated guess that the company will be invited to present to an FDA panel in April. Totah says if the company gets that chance--the process sounds a bit like going to see the Wizard--Cyberonics expects the FDA will make its decision by July.

Those are big suppositions for a company with Cyberonics' history. And on the basis of those suppositions, Cummins announces to Wall Street, the company will spend $5 million hiring 140 sales and marketing staffers over the next six months and training the company's existing sales staff to sell the VNS device to psychiatrists. Cummins tells the analysts that 4.4 million people in the United States suffer from deep depression that doesn't respond to existing treatments, and he predicts that Cyberonics can sell $2.8 billion of the VNS devices to that population in the next six years.