By Kim Coghill

Washington Editor

BETHESDA, Md. ¿ Amylin Pharmaceutical Inc.¿s lead drug candidate for diabetes patients hit another brick wall Thursday when the Endocrinologic and Metabolic Drugs Advisory Committee voted against recommending the product for approval.

The candidate is Symlin (pramlintide), a synthetic analogue of the human hormone amylin. Amylin wants to market it as an adjunctive therapy to insulin for treating people with Type I and Type II diabetes.

The committee voted 8 to 1 against the use of Symlin for Type I patients, and 6 to 3 against the use for Type II patients.

Following the vote, Joe Cook, Amylin¿s chairman and CEO, held a press conference and told reporters, ¿I still see [Symlin] as a business opportunity and we¿ve got to find a way to make it work. My take on this is, the jury¿s still out.¿

The new drug application for Symlin is the subject of 13 years of research, 660,000 pages of material, six trials ¿ some that failed ¿ and thousands of patients, including almost 3,500 who participated in the most recent Phase III trials, one for Type I patients and the other for Type II.

The FDA will decide later on whether to approve the drug.

Patients using Symlin would inject it three times a day, in conjunction with, but in a separate syringe as, insulin.

According to Amylin, Symlin works with insulin to control blood sugar, and as an added bonus, it helps reduce weight gain, particularly in Type II patients.

But all is not well with Symlin. Dragos Roman, a researcher with the FDA¿s Division of Metabolic and Endocrine Drug Products, said in the trials there were some problems with adverse events, namely that use of the product with insulin was associated with a fourfold increase in auto-related accidents in Type I diabetes patients.

Roman¿s comments echoed the FDA¿s briefing document criticizing Symlin and the trials saying that random inspections of 44 patient records disclosed one hypoglycemic event that did not appear in the database and one car accident that did not appear in the database. The document then questions whether the safety database is reliable.

The FDA also said there was a higher incidence of severe hypoglycemia in patients on Symlin than in the control group.

Other adverse events associated with use were nausea and anorexia.

The trials included 1,717 Type I patients (538 on placebo) and 1,693 Type II patients (420 on placebo).

¿It seems this drug does have efficacy, but it is rather a small amount of efficacy and I think we could get the same effect by increasing the insulin,¿ said Deborah Grady, committee member and professor of medicine and epidemiology at the University of California at San Francisco. ¿What worries me about hypoglycemia in this drug is the nausea and anorexia. I know we can train the physicians, but what about when the patients get out on their own, will we see more hypoglycemia? I worry that hypoglycemia will become more prevalent.¿

The committee voted 8 to 1 that the trials proved efficacy, but also voted 8 to 1 saying the product doesn¿t appear to be safe.

¿They¿re saying we¿ve shown efficacy,¿ Cook said. ¿The question is, how are we going to initiate therapy? We¿ll talk with the FDA and do another trial. We¿ve got plenty of people out there who will participate. We¿ll enroll them and do another trial.¿

Orville Kolterman, Amylin¿s senior vice president of clinical affairs, didn¿t seem to think the hypoglycemia levels would be as severe outside of trials because insulin levels could be better controlled and monitored. He said the trials were blinded so patients didn¿t know whether they were taking a placebo and may or may not have adjusted insulin.

Several committee members said Symlin is attractive because it gives Type I patients another option for medication and it opens the possibility of weight loss as opposed to weight gain.

Drug¿s Been on A Long, Bumpy Road

An early sign of trouble for the drug surfaced in 1997 when Amylin and then-partner Johnson & Johnson, of New Brunswick, N.J., determined that an initial Phase III trial for Type II diabetes failed to show lower glucose on an intent-to-treat basis after 12 months. (See BioWorld Today, Aug. 19, 1997.)

Seven months down the road, and with $175 million invested, J&J pulled out of the deal, leaving Amylin to continue on its own. (See BioWorld Today, March 3, 1998.)

The loss was severe, forcing the San Diego-based company to cut its staff by 75 percent, from about 190 to 45 employees. Today that company employs 220 people.

But the bad news wasn¿t over yet.

In late 1998, Amylin¿s stock dropped 77 percent, closing at 69 cents, down $2.25, when two European/Canadian Phase III studies failed to show statistical significance. Symlin failed to achieve statistical significance at a 90 microgram dosage, but did achieve statistical significance at the 60 microgram dose. The company reported that it was more difficult to identify the right dose for Type II patients because they tend to be slightly more obese. (See BioWorld Today, Oct. 22, 1998.)

Despite the bad news, San Diego-based Amylin dug its heels in and kept going. A year later, the company raised $18.5 million by selling 3.7 million shares of stock at $5 per share to a select group of institutional and private investors. (See BioWorld Today, Oct. 8, 1999.)

A year later the outlook seemed much better.

In November 1999, Amylin released data from a one-year pivotal trial in Type I diabetes showing a statistically significant reduction in the primary glucose control.

The study showed when Symlin was added to a patient¿s insulin regimen, the primary glucose control endpoint (glycated hemoglobin; HbA1c) was significantly improved at six months, compared to those who took only insulin. For the group taking Symlin three times a day, HbA1c was reduced by 0.3 percent at six months, compared to those receiving insulin alone (p=0.008). (See BioWorld Today, Nov. 10, 1999.)

HbA1c is the accepted clinical standard for evaluating long-term glucose control. The primary endpoint was change in HbA1c from baseline to six months for the Symlin groups compared with placebo groups.

In February 2000 the company raised $100 million through the sale of 8.3 million shares at $12 each to help pay for the cost of further developing and launching Symlin. And a couple of months ago, the company raised another $35 million in a private placement that included 3.5 million newly issued shares at $10 per share. (See BioWorld Today, Feb. 23, 2000, and May 15, 2001.)

According to the company, marketing applications for Symlin were submitted in Europe in May and have been accepted for filing.

Amylin¿s second diabetes drug candidate, AC2993, is in Phase II evaluation for the Type II indication and AC2993 LAR, a long-acting formulation, is in Phase I development with results expected in the third quarter.

The company¿s third product, AC3056, is in Phase I evaluation as a potential treatment for metabolic disorders relating to cardiovascular disease.

Amylin¿s stock (NASDAQ:AMLN) was not traded Thursday. It was held at $7.95.