EnteroMedics Inc., based in Roseville, Minnesota, said Monday that the Food and Drug Administration has approved a new clinical trial for its Maestro anti-obesity device that uses electricity to suppress hunger.

The news represents a major victory for EnteroMedics whose previous study ended in failure. Last October, initial results from the EMPOWER studies of more than 400 patients showed people who used the Maestro device lost just as much weight as patients implanted with the dummy device.

In a previous interview, CEO Mark Knudson said wires from the sham device connected to the nerve released low-intensity energy whenever the company performed a safety check on the device. As a result, the Maestro device delivered 10,000 times more electricity to the therapy group but did not perform any better than the control group’s dummy device.

Since January, the company has been negotiating with the FDA on a study through the agency’s Investigational Device Exemption program.

The new 12 month study, dubbed ReCharge, will enroll 234 morbidly obese people who will randomly be implanted with Maestro or a sham device. This time, the dummy device will have no wires connected to the nerve. Patients will also use an updated Maestro device in which the battery pack will be worn around the belt instead of an implant.

The company hopes patients with Maestro will lose 20 percent more weight than people implanted with the dummy device. The trial will begin the second half of 2011.

Mark Knudson, CEO, EnteroMedics Inc.

Someday, in the not-so-distant future, EnteroMedics Inc. officials might look back at late 2009 and laugh. Or cry.

That ultimately depends on whether the Food and Drug Administration ultimately approves or rejects the company’s groundbreaking Maestro device. Approval means enormous financial success, the only implantable neurostimulation device in the United States designed to treat obesity. Rejection means the end of EnteroMedics, not to mention millions of investment dollars and years of painstaking research down the tubes.

But win or lose, October 2009 will certainly be a key date in the company’s history. That’s when EnteroMedics received the results of its key EMPOWER trial. The news was bad: patients receiving the VBLOC therapy lost no more weight than the patients implanted with a dummy device.

“Everyone was completely stunned when the results came out,” CEO Mark Knudson said.

That could have spelled doom for the company as investors dumped the stock. But now armed with fresh capital and encouraging talks with the Food and Drug Administration, EnteroMedics hopes to conduct a follow up study next year, with an eye on FDA approval in 2012.

That, of course, is much later time line the company had originally envisioned. But given its precarious position, 2012 is looking awfully good.

In some ways, EnteroMedics is the perfect case study of how the promise of a breakthrough technology ran smack into the grim financial and regulatory headaches confronting medical device start-ups today.

Several companies have been exploring the idea of using using electricity to control feelings of hunger but EnteroMedics is far ahead of the pack. Its Maestro device shoots currents down the Vagus nerve, interrupting signals between the brain and stomach. Not only could Maestro combat obesity, but the company argued the device could also treat diabetes and hypertension.

But being first in line also has its downsides. First of all, the company must design a randomized, blinded clinical trial that compares the technology being tested to a control group either through an alternative treatment or, if none exists, a placebo. But unlike pharma trials in which patients are given sugar pills, medical device companies must implant something in the patient to act as the placebo. Some experts even argue that implanting a device in a patient with no therapeutic value is unethical.

“It’s very difficult,” Knudson said. “It’s the biggest issue facing medical device companies in Minnesota. With devices, there’s no such thing as a sugar pill.”

Mark Kroll, a medical device engineer and a former top executive at St. Jude Medical Inc., says it’s also tough to enroll the right patients. People who are willing to participate in such a trial are probably “highly motivated” to lose weight and thus more susceptible to believe the dummy device is working, he said, a phenomenon known as the placebo effect. In other words, a patient who is told they have a device to help them lose weight might actually lose the weight.

From 2001 to 2006, the percentage of new pharmaceutical products cut from development after Phase II clinical trials, when drugs are first tested against placebo, rose by 20 percent, according to Wired magazine. The failure rate in more extensive Phase III trials increased by 11 percent, mainly due to surprisingly poor showings against placebo.

There’s also mounting evidence of “an enhanced placebo effect” in medical device trials, especially when “under conditions of heightened expectations,” according to a 2000 article published in the Journal of Clinical Epidemiology. The article noted strong placebo effects involving sham devices have frustrated researchers trying to prove the clinical value of acupuncture and devices that uses electricity to treat pain.

“The evidence for the existence of an enhanced placebo effect for devices and procedures is intriguing but by no means conclusive,” the article concludes. “Most studies on the subject were conducted before the era of obligatory informed consent and suffer from methodological flaws that cast doubt on their validity. Yet, the phenomenon cannot be ignored because its existence has important implications for the interpretation of many influential randomized control trials.”

For EnteroMedics, everything seemed fine until last October when the company dropped the bombshell: Initial results from the EMPOWER studies of more than 400 patients showed people who used the Maestro device just as much weight as patients implanted with the dummy device.

The sham device did contain wires connected to the nerve. Knudson believes the wires released low intensity energy whenever the company performed a safety check on the device. Still, the results were shocking: the Maestro device delivered 10,000 times more electricity to the therapy group but did not perform any better than the control group’s dummy device.

“When you’re the first people to plow the field, you find the boulders,” Knudson said. ‘When you do something new, you always find something unexpected.”

Knudson also noted the study proved Maestro worked. It also took much longer for the control group to achieve the same weight loss as the therapy group.

Investors weren’t comforted. EnteroMedics stock plummeted from $5.40 a share to less than $1 a share in just a few days. The company now trades around 60 cents a share.

“It’s painful when your stock gets hit like that,” Knudson said. “It makes it difficult to move forward.”

But the company had one distinct advantage: timing.

EnteroMedics went public in 2007, well before the economy plunged into the deepest recession since the Great Depression. The idea of investors eagerly buying shares of  a company with nothing yet to sell would be all but impossible in today’s risk averse economy.

In fact, EnteroMedics along with Virtual Radiological Corp. of Minnetonka and AGA Medical Holdings Inc. of Plymouth, are not just the last three local medical device companies to go public but are the last three companies of any kind in Minnesota to go public since the end of 2007.

It’s hard to say what would have happened to EnteroMedics in the fall of 2009 if it was still a private company. Could the company raised more money from wary venture capitalists to conduct another trial? Knudson, for his part, is glad he didn’t have to find out.

“We had enough cash to stay solvent but not for another test because we didn’t think we needed it,” said Paul Klingenstein, a managing partner of Aberdare Ventures and an EnteroMedics board member. “It’s extremely difficult for a young company to gain the financial resources to take to a Phase III study. We didn’t fully understand that but now we do.”

In January, EnteroMedics managed to raise another $4.8 million from selling stock to institutional investors. The company also said the FDA encouraged it to submit an Investigational Device Exemption application to run another clinical trial that would validate the therapy results of the EMPOWER study.

EnteroMedics officials believe they can design the right study. The new trial will still use a dummy device but with no wires connected to the nerve. Patients will also use an updated Maestro device in which the battery pack will be worn around the belt instead of an implant. The company hopes the upgrade will boost patient compliance to 95 percent from just over 50 percent in the original study.

“I’m absolutely confident in the company,” Klingenstein, the board member, said. “But will I predict what the FDA is going to do? No way.”

EnteroMedics and Leptos are both developing a device that uses electricity to treat obesity. But that’s where the similarities end, EnteroMedics CEO Mark Knudson told me in a quick phone call.

“I see no connection,” he said.

In the story, I wrote EnteroMedics’ problems with gaining FDA approval for its VBLOC therapy spelled bad news for companies like Leptos, who are far behind EnteroMedics in clinical development. I theorized EnteroMedics’ FDA woes spooked investors in the competition.

If EnteroMedics fails to win regulatory approval, companies following it will no doubt face a tough time raising money, Knudson said. But Leptos’ demise probably had a more simple reason: Its technology didn’t work, he said.

Leptos CEO Hans Neisz declined to tell me why the company folded, except to say it had nothing to do with raising money.

I should note that Knudson, who was gracious and friendly, invited me to visit his company next month. I gladly accepted.

The company, which is developing a device that uses electricity to treat obesity and diabetes, Thursday said data from a key clinical trial showed its VBLOC therapy “remain encouraging, with signs of clinically meaningful weight loss and control of obesity.” Furthermore, patients using a more advanced version of the Maestro device for 14  hours a day lost weight and demonstrated significantly reduced levels of HbA1c (glycated hemoglobin) levels in the blood.

Sounds good. One major problem: EnteroMedics is a company dangling precariously off a cliff. Last fall, Enter0Medics said initial results from its Empower clinical study failed to meet its goals.

Huh? Didn’t I just say the company reported positive results this week?

Let’s back up for a second. EnteroMedics, founded in 2002, heads a class of promising start-ups pioneering the use of  neuromodulation to treat obesity. Leptos Biomedical Inc. in Fridley, IntraPace Inc. of Mountain View, Calif., and MetaCure USA Inc. in Orangeburg, N.Y., all seek to control hunger by shooting electrical impulses into the digestive system.

Medtronic Inc. also is researching this area. In 2005, the company formed a new unit, Medtronic Obesity Management, after purchasing Transneuronix Inc., which focused on developing a pacemaker-like device to stimulate the stomach.

But EnteroMedics is far ahead of the pack; its Maestro device shoots electricity into the vargus nerve, blocking signals from the brain to the stomach and pancreas that cause feelings of hunger, fullness and satisfaction. The Food and Drug Administration will decide on whether to grant pre-market approval sometime this quarter.

Everything seemed fine until last October when EnteroMedics dropped a bombshell: Initial results from the Empower studies of more than 400 patients showed the company did not hit its designed targets. People who used the Maestro device lost weight — but so did the people who didn’t.

In every clinical trial, there are two groups of patients: The treated group receive the actual therapy while a control group gets a placebo or standard treatment. The results are then compared. Ideally, treated patients should demonstrate a higher degree of efficacy than the control group.

That didn’t happen with Empower. The number of patients showing an excess weight loss (EWL) of greater than 25 percent were identical in the treated and control groups. Du-oh!  At 12 months, the patients who used the real device for more than 9 hours a day demonstrated an average EWL of 23.1 percent compared to 22.6 percent in the control group.

At the JP Morgan Healthcare conference in San Francisco Thursday, EnteroMedics CEO Mark Knudson said evidence showed the body in the control group somehow blocked electrical signals on the vargus nerve without the use of the device, an unexpectedly powerful placebo effect. He also noted the effect was gradual in the control group while treated patients experienced an immediate blocking effect.

The result doesn’t necessarily mean Maestro doesn’t work. It just means the control group did just as well. But whether that’s good enough for the FDA remains to be seen. And given the mood of the FDA these days, I wouldn’t hold my breath.

“We don’t know how the FDA is going to react” to the data, Knudson said.

If the FDA rejects Maestro, EnteroMedics is finished. Given the poor economy, finding investors willing to continue writing checks would be next to impossible. (EnteroMedics did say Thursday that it raised another $4.8 million.)

Just look at what happened last fall. After the company reported the disappointing Empower results, EnteroMedics stock  plummeted from $5.40 a share to less than $1 a share in just a few days. The company now trades around 70 cents a share. EnteroMedics also immediately reduced its workforce by 40 percent to 33 employees.

Hang on to your hats EnteroMedics investors. It’s going to be a bumpy ride.