By Mary Welch

Oh what a difference two weeks can make.

Hemosol Inc. said results of its Phase III randomized, double-blind study of Hemolink, a human hemoglobin-based blood substitute, did achieve statistically significant results. Two weeks ago the Toronto-based company said that preliminary results showed the blood substitute didn't prove Hemolink's benefits. (See BioWorld Today, June 13, 2000, p. 1.)

"As a public company, we are required to make disclosures whether they are convenient for a company or not," said John Kennedy, Hemosol's president and CEO. "When we had the preliminary results we were right at the edge of the power calculator. We could have been wrong on either side. We could have said it didn't look like statistical significance was achieved - and be wrong. Or, we could have said it did, and then be proved wrong. We just needed two more weeks to be certain of what we had."

The trial had two main purposes. The first was to determine if Hemolink could be given safely in a large number of patients undergoing routine coronary artery bypass grafting (CABG) surgery. The second was to determine whether the transfusion of donor red blood cells (RBCs) could be effectively avoided or reduced in patients given Hemolink relative to patients in a control group. Those in the control group received penaspan, a standard volume expander, instead of Hemolink.

The Phase III trial involved 288 patients in Canada and the UK. In that trial, 27 percent of the control group received a transfusion of donor RBCs as opposed to only 17 percent of the Hemolink group. The 37 percent reduction relative to the control group is statistically significant (p<0.05).

"It worked," Kennedy said. "It actually worked better than we expected. It showed that Hemolink effectively reduced the number of patients who had to have donor RBCs. Of those Hemolink patients who did require donor RBC, they had to get less and get it later. From a risk standpoint and patient safety, that's important - not to mention economically important."

In fact, those Hemolink patients requiring a transfusion needed 20 percent fewer red blood cells than those in the control group. In addition, the median time to first donor RBC transfusion was 14 hours longer in those taking Hemolink than those in the control group. Finally, the rate of transfusion of donor platelets to control bleeding following surgery was 62 percent lower in the Hemolink group.

The company now plans to file for marketing approval in Canada next month. Approval, if granted, should take about a year, Kennedy said.

"We were just coming off our annual meeting in May when the preliminary results came in," Kennedy recalled. "I was suffering from a lack of humbleness and then had to say that our Canadian filing would be delayed. It turned out to be a delay from June to July. So there's a happy ending."

Hemosol plans to start filing for approval in select European countries in either September or October, Kennedy said.

The trial's results also mean that the company will apply this year for a listing on Nasdaq. Currently it is traded on the Toronto Stock Exchange. "There are lots of advantages to being listed on Nasdaq," Kennedy said. "Our compatriots say that when you decide to go down there and be listed that it's better to go after you've gotten your Phase III results. Stocks on Nasdaq can go up and down quickly, so it's better if you have solid Phase III data. You're dealing from a position of strength."

A larger U.S. trial is just getting started. Patients are being enrolled, but treatment should start in September. About 40 sites will participate. The trial should last about a year, with a new drug application filed with the FDA in the second half of 2001, Kennedy said.

Hemosol's stock (TSE:HML) closed Monday at C$16.90, up C50 cents. But the stock didn't rebound nearly as much as it fell two weeks ago, when it dropped 32 percent, or C$7.90, to close at C$17.