Janssen Defends Data Reanalysis In Risperdal Trial

Legal Intelligencer
Max Mitchell, The Legal Intelligencer

December 9, 2015    | 0 Comments

A Janssen Pharmaceuticals official and the statistician who recently conducted a hotly contested reanalysis of data purportedly linking Risperdal to gynecomastia were "hoagie buddies," according to the plaintiff's attorney in the ongoing Risperdal-related case.

To illustrate that point, Thomas R. Kline of Kline & Specter, who is representing plaintiff Timothy Stange in Stange v. Janssen Pharmaceuticals, read portions of emails between the statistician, Warren Bilker, and a Janssen official who had been tasked with helping to find a researcher to perform the reanalysis of data that first appeared in a 2003 medical article.

Kline used the email, and others, to refute arguments from the defense that Janssen played a limited role in the reanalysis, which has become a major point of contention. While Janssen has pointed to the newer study as confirmation of earlier findings suggesting no significant statistical relationship between Risperdal and gynecomastia—a condition causing excessive breast tissue in young males—the plaintiffs have argued that some results linking the condition to Risperdal were omitted from the earlier study to conceal the risks and manipulate the marketplace.

In the email, Bilker and the Janssen research vice president, Jesse Berlin, appeared to discuss meeting up for a dinner before going to a Philadelphia Phillies game. One of the places they talk about eating is at a sandwich restaurant by Citizens Bank Park, where the Phillies play.

"They're old hoagie buddies," Kline said. "Why, they're guys that are heading to the ball game. To the Phils."

The email, and others that appeared to discuss the vetting process between Janssen officials and Bilker in regards to his reanalysis, were read during trial Wednesday after a three-week break. The case had been put on hold so Janssen could provide more than 30,000 responsive documents, consisting of about 300,000 pages, related to the reanalysis.

The reanalysis at issue was of a 2003 medical article published in the Journal of Clinical Psychiatry, which has been referred to in the litigation as the Findling article.

However, Janssen development leader Dr. Danielle Coppola, who was on the stand Wednesday, testified the data was omitted from the initial analysis because it was not statistically significant, and the reanalysis, which included the omitted information, confirmed the results of the Findling report.

Coppola further testified Janssen had a limited role in the reanalysis, and only asked Bilker to perform the work after other statisticians were unable to do the project. She further noted that the authors of the Findling report had the ultimate say in who would perform the reanalysis.

According to Coppola, the reanalysis was done after questions about the validity of the findings were raised in litigation and in the media. She said the author of the Findling article initially reached out to Janssen about having a reanalysis performed, and Janssen had to be involved in the study because it is the only entity with access to all the Risperdal-related research data.

"I wanted to make sure the data was safe, and the label accurately reflected that," Coppola said.

When asked by defense attorney Melissa Graff of Drinker Biddle & Reath how the reanalysis compared to the original findings, which did not indicate a significant link between Risperdal and gynecomastia, Coppola said they were "entirely consistent."

Coppola said no information had been withheld from the authors of the Findling article. She also said that one of the original authors, who had threatened to take his name off the Findling article if there were any problems, was satisfied with the reanalysis, and the newer findings were also given to the U.S. Food and Drug Administration.

"We thought that, under the circumstances, it would be important to share," she said. "We all have an interest in the safety of the product."

Coppola also told the jurors that the table initially omitted did not deal with the frequency of patients experiencing elevated prolactin while taking Risperdal, but instead showed the proportion of elevated prolactin levels to all adverse reactions that Risperdal users experienced. She also noted that statistics mentioned on the label were based on clinical trial, while the omitted data came from reports of adverse effects by users.

When asked by Kline about the emails involving Bilker, one of which included Berlin telling Bilker that the authors of the Findling report "will want to be convinced that you are a legitimate scientist (a tough sell, I know)," Coppola said she had never seen the emails before, but knew that Bilker and Berlin knew each other. She said the Findling authors knew the two knew each other as well.

"They were buddies," Kline said, adding that the two appeared to be joking in the emails.

Kline went on to review the findings in the omitted tables regarding elevated levels of prolactin, and noted that the ratios of higher prolactin were significantly greater in children and adolescents than in adults.

The case, which is before Philadelphia Court of Common Pleas Judge Kenneth Powell, is expected to end before the end of the week.

Max Mitchell can be contacted at 215-557-2354 or mmitchell@alm.com. Follow him on Twitter @MMitchellTLI. •

Thank You Mr Mitchell and Legal Intelligencer