By Chris Lane
By Jeff Balke
By Aaron Reiss
By Angelica Leicht
By Dianna Wray
By Aaron Reiss
By Camilo Smith
By Craig Malisow
This is high-tech, last-ditch, no-other-options stuff. The company's pacemaker has been shown to succeed where medication and brain surgery fail, ending years of torture for people battling their own bodies. More than 30,000 children and adults have received an implant.
But Cyberonics is a big company that wants to be a giant. Its problem is that it has a single product, and treatment-resistant epilepsy is a limited market. Cyberonics estimates its potential U.S. population at 425,000.
The real money is in mood disorders, which is why the company spent $50 million in the last seven years studying the device's efficacy on treatment-resistant depression.
Cyberonics estimates there are 4.4 million Americans living with depression that's resistant to drugs and electroshock therapy. After years of battling the Food and Drug Administration, Cyberonics in February received the agency's conditional approval to enter its pacemaker in the depression market. Its stock soared. CEO Robert "Skip" Cummins told shareholders that not only would really depressed people no longer have to kill themselves, but the company was poised to dominate the $60 billion TRD market.
Next up, according to the patents Cyberonics has already secured, would be to use the same device to treat eating disorders, Alzheimer's, obesity, migraines, endocrine disorders, congestive heart failure, comas and the catch-all "pain."
But before it can get to depression, Cyberonics has to meet certain FDA conditions. And Cyberonics has an interesting history with the FDA.
In 2001, the agency issued a warning letter to the company for failing to properly report 60 deaths its pacemaker could have caused or contributed to. The company also failed to properly report 102 patient infections. Cyberonics had withheld the data from the FDA until two months after a routine inspection.
Cyberonics resolved those issues but received another warning letter last December, which the company must rectify for full approval.
Cyberonics is optimistic the FDA will give it the key to the depression market. In a press release, the company described the federal violations outlined in the warning letter as "outstanding bioresearch monitoring issues," which is the company's way of telling shareholders it must do the following:
1. Investigate and explain why thousands of patients had to be reimplanted with a second pacemaker;
2. Investigate and explain the causes of 81 deaths reported between 2001 and 2003; and
3. Investigate and explain the cause of each reported adverse effect, including instances where the device dislodged and migrated throughout the body; cases of vocal cord paralysis; and increased seizures.
Cummins is confident the company will hit the depression market by the end of May.
Sam Goldsmith is only five, but he's been to hell and back more times than anyone can count. At three months, after a course of routine immunization shots, he developed a fever that never really let go. Seizures rack his brain, knocking him to the floor. Whenever this happens, Dave and Monika Goldsmith divide duties: One holds their writhing child and the other grabs his E.T. backpack -- a pharmacy-in-a-bag -- and sets a timer.
If the seizure doesn't quit after five minutes, they shoot Sam with five milligrams of Valium. After ten minutes, they step up to Verset and call 911. If he's still seizing after 20 minutes, he's in the ICU, climbing a pharmacological ladder: Ativan. More Verset. Fosphenytoin. The highest rung is the phenobarbital coma, and it's lights out, Sam.
But then, there's also the magnet.
The magnet can stop a seizure cold. It triggers the pacemaker in Sam's chest, which zaps his vagus nerve, which leads to the brain. In medical terms, there is no explanation for the "mechanism of action," which means that no one -- not neurologists, not Cyberonics -- knows exactly why the device works. All that's understood is that vagus-nerve stimulation can reduce and even eliminate seizures in otherwise treatment-resistant epileptics.
After exhausting dozens of medications, the Goldsmiths wound up in the office of renowned neurologist Angus Wilfong. Wilfong, a medical director of the epilepsy program at Texas Children's Hospital, is an assistant professor of pediatrics and neurology at Baylor College of Medicine. Wilfong also founded a vagal-therapy program at the Children's Hospital of Wisconsin. Wilfong told the Goldsmiths about his patients' positive experiences with vagus-nerve stimulation, and he referred them to pediatric neurosurgeon John Laurent.
In 1987, Laurent co-founded Texas Children's Hospital's Brachial Plexus Program, the largest of its kind in the world. Laurent's team operated on babies who, at birth, tore or stretched the bundle of nerves between the neck and shoulders. Because of the size of their patients and the delicacy of the damaged nerves, brachial-plexus surgery is an especially difficult procedure.
The Goldsmiths researched VNS, poring over literature for a better understanding of the risks and benefits involved. The most they could find was that, because the vagus nerve is so close to the vocal cords, some patients experienced hoarseness or a tickling in their throats. They decided the minimal risk was worth offering Sam a chance at a seizure-free life. In 2002, Laurent implanted Sam's pacemaker.
With Wilfong and Laurent, the Goldsmiths placed Sam among the finest care in the country. But after about a year of continued seizures, they got a call from Wilfong's office.
We may have a problem.
X-rays showed that Laurent had inverted the wires connecting to Sam's vagus nerve, just as he had in at least 74 other children, according to a class-action lawsuit filed in Harris County District Court in 2003. (The Goldsmiths are not part of the suit.)
But the news wasn't entirely bad. According to the Goldsmiths, Wilfong said that the device was still working properly and wouldn't need to be replaced.
By the end of 2003, it looked like Sam had fallen into the category of patients who were neither harmed nor helped by the therapy. Wilfong suggested brain surgery -- a procedure that has helped many treatment-resistant epileptics like Sam. So in January 2004, surgeons at Texas Children's spent ten hours carving out a chunk of the boy's temporal lobe.
The surgery was a disappointment. Sam continued to seize and, a few months after the operation, wound up in Memorial Hermann Hospital during a grand-mal. According to the Goldsmiths, the doctors turned off Sam's pacemaker in order to conduct an MRI. With a subsequent X-ray, doctors saw the device was inverted, and they refused to turn it back on. They did not know how inverted wires were affecting Sam, the Goldsmiths say.
"That raised a whole lot of red flags for me and Monika," Dave Goldsmith says.
The device remained off for two months, and the Goldsmiths saw no difference in their son. It was clear: The $15,000 bundle of titanium and wires in Sam's body was not worth a cent.
The Goldsmiths returned to Wilfong and demanded he refer them to a surgeon who could implant a new pacemaker correctly. After a successful replacement, the Goldsmiths noticed a moderate reduction in Sam's seizures.
Laurent died in his River Oaks home in November 2004. The class-action suit's remaining defendants are Wilfong and Cyberonics. Wilfong declined to comment for this story.
Cyberonics spokesperson Liza Heapes says Laurent received the company's standard "surgical training kit," which includes an instructional book and video and a dummy pacemaker for practice. Using this training kit, surgeons have implanted devices in 30,000 adults and children around the world. Thousands have had to be removed and/or replaced, and the company has not told the FDA why.
In the Goldsmiths' case, the explanation is clear: A renowned neurologist referred them to a gifted surgeon who, with manufacturer training, botched what the company calls a "straightforward" procedure.
"I think what putting the first VNS in incorrectly took from us," Dave Goldsmith says, "was a whole year and a half or almost two years of it not doing as much as it could've done for us, at the very least. And possibly harming him in some way."
First of all, if you're talking to Cyberonics, don't call it a pacemaker. That word rarely appears in company literature.
They prefer Pulse Generator. The Pulse Generator provides the trademarked Vagus Nerve Stimulation Therapy, named after the nerve whose Latin name means "wandering." The vagus nerve wanders from the brain stem through the neck, thorax and abdomen.
About the size and shape of a silver dollar, the pacemaker is implanted below the left collarbone, and two wires are attached to the vagus nerve. Cyberonics' patient's manual lists a total of nine side effects, including skin irritation, blood clotting, paralysis of nearby nerves and muscles, and changes in heart rate. The physician's manual lists 49 side effects, including device migration, insomnia, gastric ulcers and formation of fibrous tissue. And, unlike the patient's manual, the physician's manual explains that the electrical leads can in fact paralyze the nerve they're supposed to stimulate.
Cyberonics describes implantation as a simple procedure taking between 30 and 90 minutes. In most patients, the pacemaker stimulates the left vagus nerve for 30 seconds, stops for five minutes, then turns back on for another 30 seconds. Patients who feel a seizure coming on, or are already in the midst of one, can use a magnet to automatically activate the device. The FDA in 1997 approved it for people more than 12 years old who suffered from treatment-resistant epilepsy. These were the unfortunate patients who had exhausted every pill, diet and surgical alternative. Although the FDA does not approve it in younger patients, off-label use -- implanting the device in kids under 12 -- is legal and common. However, Cyberonics' policy lists any adverse event in off-label use as "user error." Meaning that, technically, if Sam's device were to spontaneously combust, it would be the Goldsmiths' fault.
With FDA approval, Cyberonics grew from an office in a League City strip mall to a marble-floored HQ in Clear Lake. According to the company's 2004 annual report, Cyberonics' fiscal 2004 net income was $6.8 million, with worldwide net sales at $110.7 million. About 350 of the company's 550 employees live in Houston.
Although the company always aimed for the treatment-resistant-epilepsy market, doctors grew curious when some patients reported elevated moods.
The company launched clinical trials for depression in 1998, and before the trials were over, Cummins issued press releases touting the therapy's success. By 2001, Europe and Canada had approved the device for treatment of depression. However, the company knew that widespread use would be impossible without FDA approval. Thus began the story of Cyberonics v. FDA, with Skip Cummins emerging as the company's most outspoken officer.
Cummins was a general partner in a venture capital partnership when he joined the fledgling Cyberonics in 1988. Seven years later, he was appointed president and CEO. (Cummins was not outspoken with the Houston Press. Shortly after agreeing to an interview and a tour of the facility, Cummins shocked Wall Street by selling $14 million worth of Cyberonics stock in mid-February. He then declined the Press interview.)
A 1999 Houston Chronicle story on Cyberonics described Cummins as a six-foot-two, 215-pound intense outdoorsman who held Dartmouth College's javelin record and spent his free time helicopter-skiing and snowboarding.
When the FDA's advisory panel recommended approval of VNS for depression in June 2004, Cyberonics' stock more than doubled. Cummins thanked the FDA and the panel in a press release.
But in August, when the FDA rejected the panel's recommendation -- a rarity in the agency's history -- Cummins was less cordial.
"We are shocked and bewildered," Cummins stated in a press release.
The FDA reported that the company relied on nonrandomized clinical studies with a potential bias. The agency also expressed concern over instances of study patients whose depression deepened, and the fact that there was no statistical difference between a group of patients who had the device turned on and a control group whose devices were off.
Richard Malone, a psychiatrist and professor of psychiatry at Drexel College of Medicine in Philadelphia, was one of two advisory panel members to vote against recommending the treatment for depression.
"For the VNS, there was no evidence that the real treatment was better than the fake treatment," Malone says. "So then you have a treatment that doesn't have evidence of working but does have evidence of side effects."
John Rush, a distinguished psychiatrist with the University of Texas-Southwestern Medical College, designed the clinical depression study. He told the panel in June that, at the time of the study, conducting a yearlong controlled study would have compromised patient safety.
"We didn't know if they would get better [or if] they would get worse," Rush told the panel, according to transcripts. "We didn't know how many would kill themselves. We had no idea. No one has ever reported this So we were wrestling with a -- really a totally new territory, a terribly difficult illness with a very high risk of disability [or] death. You saw a hospitalization, we had a patient suicide who was a physician and so on."
But Malone disagrees.
"They claimed that they couldn't do more studies because it wouldn't be ethical," he says. "But I just replied that it's in the same way not an ethical thing to recommend treatment."
Although the FDA initially rejected the panel's approval recommendation, it reversed its decision in February after the company submitted supplements to the original study.
With their depression therapy deemed "approvable," Cyberonics' stock jumped 42 percent the day after the announcement. Cummins told shareholders the company would enter the depression market in May, and he technically didn't have to tell them anything at all.
The FDA does not disclose approvable letters, deferring instead to the manufacturer. A company can choose not to reveal the terms of approvable letters to the public and to shareholders, meaning the disclosure of potential public health concerns is up to corporations, whose sole legal purpose is to generate profit.
Shortly after the approvable letter, Cummins sold stock for the first time, puzzling some shareholders. However, he explained to investors that the 350,000 shares he sold was less than a quarter of his holdings.
"I did what I had to do to reduce my family's exposure and risks, knowing that the 1.15 million options I still hold will be more than enough when this company's enormous potential and value are recognized in the next two years," Cummins explained in a letter.
The FDA warning letter demanding reasons for unexplained deaths and thousands of explanted devices never had any effect on stock. As long as a product is on the market, and as long as no one talks about warning letters, a company will continue making money.
No one likes talking about warning letters, and no one likes talking about unproven therapy.
FDA medical device director Dan Schultze declined to be interviewed, referring questions to a spokesperson who stopped answering them after a while.
Cyberonics board member Stanley Appel, chairman of the neurology department at Baylor College of Medicine, also refused comment.
Spokespeople for the Houston and national chapters of the National Alliance for the Mentally Ill and the Depression and Bipolar Support Alliance, as well as the Mental Health Association of Greater Houston, said they did not offer opinions on therapy.
Yet Lydia Lewis, president of the Chicago-based DBSA -- a nonprofit that receives financial support from Cyberonics and pharmaceutical companies -- appeared before the advisory panel last June, advocating for more treatment options. Lewis chose her words carefully, lest she be accused of endorsing Cyberonics.
"I am not here to advocate for any particular therapy, including VNS, but rather for the critical need for new therapies," she told the panel, according to transcripts. "At DBSA, we know that new drugs and non-pharmacologic treatments are desperately needed. Far too many people are dying or living lives of quiet desperation, because they can't get sufficient relief from their symptoms of depression. The more treatments the FDA makes available, the more lives that will be saved."
Even outspoken FDA watchdog Public Citizen, a group that has weighed in extensively on the recent controversy over Vioxx, Celebrex and Bextra, said it has no staff members familiar with vagal therapy.
Fortunately, there are online forums where current or prospective Cyberonics patients can share information. Up until February 2004, Cyberonics had its own forum. But the company shut it down, explaining that "Increasingly the message board has been an instrument for abuse by certain individuals."
According to former forum members, the shutdown came on the heels of a patient posting a link to the FDA's database of product problems. The database lists 500 adverse events for Cyberonics between May and December 2004 and 1,414 for 2002. Data for 2003 is not available online. Events range from increased seizures to death, which is often described as "sudden unexplained death in epilepsy." In many cases, Cyberonics reported that surgeons refused to provide in-depth information so the company could find the cause of the problem. In one case of an inverted device, a neurologist "asked that he not be contacted regarding the event, as he has a limited staff and is subsequently not able to respond to mfr's request for additional info."
Cyberonics received a similar cold shoulder regarding a patient who died five days after implantation. In the case of one patient who developed a wheezing problem, the company could not investigate because whoever reported the problem forgot the patient's name. Some of the cases in which Cyberonics could figure out what went wrong are troubling as well, including the instance of a surgeon who implanted the device when it was already on. The device is not supposed to be activated until two weeks after surgery.
A week after Cyberonics removed its forum, former VNS patient Donna Baum launched what has become one of the Web's most popular alternatives. Baum, 50, had her pacemaker turned off after she experienced constant pain in her arm. She says the device reduced her seizures, and cut her postseizure recovery period from days to hours. But she also suffered breathlessness and nights where she'd wake up with vomit in her lungs. She says the device also eliminated her auras, which are signals some epileptics get of oncoming seizures.
Baum had her surgery before 2001, the year that Cyberonics placed its patient's manual online. Like many others on her forum, she didn't know a patient's manual existed until she was handed one after surgery. She says she wants her forum to be a place where current and prospective patients can get the kind of information she never got.
"I've never said the device is bad," she says from her home in Las Vegas. "I never said it needs to be pulled off the market. What I stated is 'research, research, research.' You need to know more than your doctor."
Naïveté and desperation play too big a role in prospective patients' decisions, she says.
"A lot of these people, they don't even think about it. They trust their doctor. You're raised to trust your doctor And some of them are looking for the Holy Grail. They've had horrible, devastating lives with epilepsy, and here you've got a rep saying, 'Hey, y'all, guess what, we've had patients that are seizure-free; we have patients that are completely off medication.' Now what does that imply to you, when it's said like that?"
But Alicia Kaminsky, whose seven-year-old son, Robby, has the device, says Baum's board is filled with nothing but people who are bitter because nothing -- not even VNS -- works for them.
Kaminsky, who lives in Florida, can't make it clearer: "Cyberonics has saved Robby's life."
One-year-old Robby was swimming with his older sister in the family pool when he experienced his first seizure. He went mute and sank to the bottom. That three-hour seizure landed him in the hospital for a week. Robby started his drug regimen at 18 months, and after several years Kaminsky didn't know how much longer she could take it.
"I cringe at feeding this child Valium before sending him off to kindergarten," she says.
Kaminsky, who also suffers from epilepsy, got in touch with a Cyberonics rep who referred her to a nearby neurologist. She says Cyberonics, which sets aside thousands of dollars a year for special-case patients who can't afford the operation or don't have insurance, paid for Robby's operation. She says the therapy has stopped at least 71 seizures since last September, allowing Robby to compete in his beloved motorcycle racing.
She says that prospective patients who research and locate the right doctor should have no trouble finding success with VNS. And she feels that even though Robby was younger than the approved age, the risk was worth it.
"When you've got children so severely seizuring chances are they're going to go into cardiac arrest and die and you're left with no other option," she says. As for the warning letters, Kaminsky says she read neither and doesn't plan to.
"If a parent were to read those things, it would drive them crazy," she says. Then she puts it this way: "If a man were to read the warning labels on a bottle of Viagra, he'd never have sex again."
But Suzanne Parisian, who briefly served as the FDA's chief medical device officer in the mid-'90s, says the warning letters should be taken seriously.
"You have  people with epilepsy that had this implanted device that the FDA was never told that they died, and the company never did an investigation as to why they died, and the company's contending that the device is safe and effective," she says. "It's very unlikely that anybody will do any kind of an examination in these patients to find out why they're dying, and then you're asking them to expand the indication to depression patients."
Moreover, she says, the FDA is set up to put products on the market, not to play detective. Reviewers are dependent on manufacturer reports; officers base their decisions on paperwork -- they never see the actual devices. The agency also depends in large part on pharmaceutical and medical device companies' funding. Drug companies have paid steep "user fees" since 1992; medical device companies since 2002. In exchange, the FDA promises to expedite the decision-making process. It's supposed to buy efficiency, not approval, but Parisian calls it quid pro quo.
In the case of Cyberonics' approvable letter, "It's being done, No. 1, on the company's guarantee that they're going to be good boys because they're going to tell FDA everything," she says. "And yet we have a company with a history, that hasn't been telling FDA everything. So it doesn't really make sense, does it?"
"Very early on, there was a hint that there's something happening here that's very good," says James Wheless.
Fans of the short-lived series Houston Medical may remember Wheless as the tall, skinny guy with a beard. But in medical circles, Wheless is an authority on epilepsy and director of the Texas Comprehensive Epilepsy Program.
When Wheless talks about VNS, he's like a kid describing the world's coolest PlayStation game. In 2002, Wheless contributed to a series of articles on VNS therapy's first five years, published in a supplement to the journal Neurology. The non-peer-reviewed supplement, funded by Cyberonics, stated that VNS therapy had proved itself as an effective, mainstream treatment for treatment-resistant epilepsy.
Wheless's article states that while the therapy is not approved for patients under 12, the therapy is "particularly attractive for use in this population" because drug-related side effects are avoided. In 2002, more than 3,600 VNS patients were under 18; half were under 12.
Today, Wheless is excited about VNS therapy for depression, explaining that mood disorders are common in 40 percent of epileptic adults.
"Their current therapy is just so bad," he says. "Any help at all is a huge step up."
Wheless is not worried about Cyberonics' more recent warning letter, chalking it up to the FDA's "subtle concerns" over "relatively minor things."
Despite the FDA's lack of approval for children under 12, Wheless says the procedure is safe for young kids. Moreover, it can give them a better quality of life early on, he says.
"Why make someone suffer for 15 to 20 years with poorly controlled seizures?" he asks. Within the few first years of a child's life, doctors can tell if meds will control the epilepsy. And that's why most state Medicaid programs and major HMOs pay for procedures in kids under 12, he says. The $15,000 to $20,000 price tag might be a heavy initial cost, but by keeping patients out of ICUs and off expensive meds, the device is cost-effective.
Neurologist Gregory Holmes, president-elect of the American Epilepsy Society, believes in VNS efficacy for epilepsy. A professor at Dartmouth Medical School, Holmes says he's referred hundreds of patients for the treatment. He's also clear to state his endorsement is his personal opinion and not that of the epilepsy society.
"It's been very unusual in our hands, to have someone that becomes totally seizure-free with the device, but in the majority, it does reduce the seizure frequency or the intensity and sometimes seems to even improve the alertness and behavior of the patients," he says. "It's a good alternative; it's not a miracle, but it's good. I would be upset if it was pulled off the market for some reason. The harm of that would be quite, quite extensive."
In early March, Cummins told the Houston Chronicle he expected to field buyout offers from major pharmaceutical and medical device firms.
"I think any time somebody comes and recognizes the value of what has been created, that is very good for shareholders," he said. "There is always a right price for everything."
Meanwhile, the company is gearing up for TRD market domination, recruiting sales representatives and technicians across the country. If it does get full approval, its sales and technical force will get a taste of the $60 billion market. Now might be just the time to get a job at Cyberonics.
Those who sell the device must have a bachelor's degree and at least three years' experience.
Those who investigate product failure, assist in corrective and preventive action, and help develop test methods and quality inspection techniques must have a high school degree or the equivalent.