National Editor

The slip-up by a scientific advisory board member during Tuesday's conference call with a group of investors prompted SuperGen Inc. to disclose the following day more details about interim data from its Phase III study with Dacogen (decitabine) in myelodysplastic syndrome.

Investors - some, anyway - liked what they saw. Dublin, Calif.-based SuperGen's stock (NASDAQ:SUPG) jumped $1.58 Wednesday, or 24.1 percent, closing at $8.15.

Because the board member inadvertently disclosed an interpretive analysis related to Kaplan-Meyer curves in the study, SuperGen went ahead and released them.

The company cautioned that the interim analysis, no matter how it's extrapolated, still represents less than half the available data. Complete, final analysis of the data is expected later this month.

Meanwhile, interrogators during a Wednesday afternoon conference call grilled company officials about Kaplan-Meyer, patient numbers, types of patients and their conditions - pretty much anything callers could think of, it seemed - during an event that had officials reiterating what they said in the press release earlier.

One dissatisfied investor remarked on the stock's favorable performance Wednesday, and said he wanted to "voice [his] displeasure" over the way the data situation was handled.

James Manuso, SuperGen's president and CEO, said during the conference call that the pre-planned interim analysis "occurred last summer in conjunction with the FDA" but it was "mutually decided" not to disclose details because "an earlier release may have jeopardized patient participation in the trial."

That same investor also grumbled about the $34 million financing disclosed by the company less than a week ago, in which about 4.9 million shares were privately placed at $7 each, a markdown from the previous day's closing bid. Without comment, a company official thanked him for calling. (See BioWorld Today, March 8, 2004.)

SuperGen officials could not be reached for comment.

Regarding Dacogen, SuperGen released an interim analysis of the Phase III trial study in February. At the time, the company said results showed those receiving the drug, which is designed to work by reversing DNA hypermethylation, had an increased time to acute myelogenous leukemia or death (median 105 days vs. 92 days, p=0.036), the study's primary endpoint.

The company said data from 92 patients had been collected, allowing for a final analysis, but said it would not begin doing that until all primary data points were verified.

After Tuesday's conference call, during which the Kaplan-Meyer information slipped out, SuperGen released on its website and in a press release the analysis conducted using the method, which is a commonly accepted way of computing survival benefit in trials.

The company pointed out that the analysis results offered in February were based on data from 19 high-risk patients, 18 classed as "intermediate 2" and 8 categorized as "intermediate 1," as defined by the International Prognostic Scoring System (IPSS) used to classify MDS patients. That makes 45 patients in all.

The Kaplan-Meyer curves, on the other hand, are based on data derived from all 170 patients at the time point of the 45th trial event. Though lacking a formal calculated median, the Kaplan-Meyer curves allow extrapolation from the 50 percent of patients alive without AML - which produces significantly different results than calculating the median of the first 45 patients, put forth in February.

Results from Kaplan-Meyer are not intended to be predictive, SuperGen warned, and might not be consistent with the final results on all 170 patients. The advisory board member based his analysis and conclusions on different data (the Kaplan-Meyer curves) than those used by the company (the median of the first 45 events) in preparing the median calculations disclosed about two months ago, so the extrapolations don't match.

Former analyst Peter Drake, who covered SuperGen before he retired from Prudential Vector Securities in Deerfield, Ill., accurately predicted about a year and a half ahead of time the company's difficulties with a deal for rubitecan, an oral topoisomerase I inhibitor for pancreatic cancer. The drug once was the subject of a potential $150 million deal with Abbott Laboratories, of Abbott Park, Ill., but the deal dissolved. (See BioWorld Today, March 6, 2002.)

SuperGen in January submitted to the FDA the final portion of its rolling new drug application for rubitecan, brand named Orathecin.

"Every company makes mistakes," Drake said Wednesday. "Every company runs into a wall. The definition of great companies are those that can, once hitting the wall, pick themselves up and find a way to get better from it."

He noted that SuperGen recently has "tried to raise money. It's when you're trying to raise money that you get into trouble."

Separately, SuperGen said Wednesday that note holders from the $21.3 million convertible debt transaction in June have approved the release of $10.6 million in funds previously classified as restricted cash and investments, so the firm now can use them for working capital.