Several recent attempts to reproduce scientific studies have failed. About 100 scientists under the direction of former Amgen researcher C. Glenn Begley found that 47 of 53 “landmark” academic cancer studies could not be reproduced. A decade ago, cancer drug discoveries were supported by as many as 200 publications, and the number of such papers has fallen to as low as single digits. The issue may be due to several factors, including a lack of blinded studies, reproducibility requiring the same variables, a bias toward intended results and a penchant for career advancement. Similar issues occurred in 2011 at Bayer, when less than a quarter of the firm’s 47 cancer experiments could be validated. While at Merck, researcher George Robertson found holes in many studies on neurodegenerative diseases.

Source: Reuters