Search This Blog

Monday, October 16, 2006

Bad Apples at Bayer

Last week Bayer suspended two senior scientists over their handling of data related to its heart-surgery drug Trasylol. Back in February of this year the FDA started to reassess the safety of the drug after two separate studies suggested the drug was linked to heart attack, stroke and kidney problems.

The FDA sought advice from a panel of outside experts on Sept. 21 over how to address the potential safety concerns. The advisers ruled Trasylol was acceptable for preventing blood loss in certain patients undergoing heart bypass surgery.

Days later, Bayer said it had mistakenly withheld another study based on 67,000 hospital patient records that suggested the drug could increase the chances of death, serious kidney damage, congestive heart failure and stroke.

What did the FDA do wrong? They failed to discover the study of 67,000 patients. They hired experts! They had a high level meeting that involved life and death decision making. One can be sure that there were men with suits and ties, women in smart business outfits. PhDs, M.D.s were all around. The talk must have been highly articulate and spoken in tones of complete understanding. Yet they made the wrong decision. If Bayer had continued to hide the data, as the two senior scientists believed was okay to do, the committee would have never known about this.

The FDA needs to be aware of any test that is initiated. What were the results? Was the experiment cancelled? Why? Are the results trustworthy? In their defense the two fired scientists claimed that the study methods were suspect and the results were preliminary. This is no defense, as far as science is concerned. A test was initiated and the data suggested something. The FDA should be able to determine if the methods were good or bad.

In a previous job I had an electronic notebook. When I began an experiment I wrote it out in the materials and methods section. I tracked down my materials and logged lot numbers in the notebook. I put everything together. I laid out some assumptions that I was making so I could reassess them after the experiment. When I had finally documented every detail of the experiment, I went to the bench and ran the experiment. Every piece of data I got went into the notebook. If I accidently used a wrong buffer I noted it in the book. I was practicing science. If I made a mistake, I logged it into the book. If you fail to log a mistake, you are committing a sin in the world of science.

What the two scientists did was more than a sin. They knew that the conclusions from the study they hid was detrimental to a drug that was already in the spotlight concerning its safety. They did not provide the FDA with the information because their job is to protect the drug, for their employer, Bayer. Favorable data is highlighted and unfavorable data is hidden. It can be hidden via corporate confidentiality policy. It can simply not be brought up. The FDA is not operating in a manner that will uncover such deceptions. We have a system that is not working. In the Cargo Cult arena, the FDA is a blind person in charge of verifying the airplanes arrivals. The drug companies work on creating the sound and smells of airplanes landing. The blind FDA must rely on honesty. That is not science. The FDA is not a scientific organization. They hold meetings and talk but they do not practice the art of science. As such, we must now officially label the FDA as a charter member of the Cargo Cult Scientist.

1 comment:

Frans said...
This comment has been removed by a blog administrator.