Lilly Zyprexa data didn’t match: NY Times

For at least a year, Eli Lilly provided information to doctors about the blood sugar risks of Zyprexa that did not match the data the company circulated internally when it first reviewed its clinical trial results, The New York Times reported on its Web site today, citing company documents. The original results, the Times report said, showed that patients on Zyprexa, Lilly’s pill for schizophrenia, were 3.5 times as likely to experience high blood sugar levels as those taking a placebo, according to a February 2000 memo sent to top Lilly scientists. But the results that Lilly eventually provided to doctors until at least late 2001 were very different, The Times reported. Those results indicated that patients taking Zyprexa were only slightly more likely to suffer high blood sugar as those taking a placebo, or an inactive pill, the report said. In a statement issued in response to today’s Times report, Lilly said the newspaper’s selection of data points was taken out of context and is “misleading.” Lilly also called the way the newspaper obtained the company’s internal document “illegal.” According to the Lilly statement, as part of the company’s ongoing safety monitoring in 1999, Lilly undertook an examination of its Zyprexa clinical trial database to help the firm further understand whether a relationship between Zyprexa and diabetes existed. The database at the time included, more than 75 clinical trials involving more than 5,000 patients, Lilly said “We were committed to proactively submitting the results of this analysis to the FDA for Zyprexa's label -- regardless of outcome,” Lilly said. As part of Lilly's final, standard quality check of the data, scientists discovered there were errors involving which patients were included or excluded in the analysis. The corrected analysis -- showing 3.1 percent of Zyprexa patients developed possible hyperglycemia or diabetes, vs. 2.5 percent of placebo patients -- was then submitted to the FDA for a label change, Lilly said.