Sunday, 09 January 2022 13:46

Reproductive crisis in cancer research Featured

Eight years ago, the Reproducibility Project: Cancer Biology (RP: CB) was started. The goal was to replicate the results of top cancer laboratories. Looking at high-ranking research articles from 2010 and 2012, the team has now come to a sobering conclusion: The reproducibility of the work is less than 50%.

The project involved a collaboration between the Center of Open Science, Charlottesville, Virginia, and Science Exchange, based in Palo Alto, California. The goal was to investigate the reproducibility of preclinical research in cancer biology.  For this purpose, 193 experiments from 53 high-impact papers had to be replicated.

In the end, the original list of 53 papers with 193 key experiments was reduced to only 23 papers with 50 experiments - i.e., less than 50% could be reproduced. None of the 193 experiments in the original work were described in enough detail to allow the team to design protocols for replicating the experiments; they had to ask the original authors for clarifications. Many original papers were missing important descriptive and inferential statistics. In 32% of the experiments, the authors were not helpful at all or had not responded to questions at all. In 67% of peer-reviewed protocols, changes were needed to complete the research. Only 41% of these changes could be implemented.

Since 2017, the results from each individual investigation have been published, mostly as individual papers in eLife. In total, the experimental work cost $1.5 million.

We cannot calculate at this point how many animals had to uselessly lose their lives for this.

Timothy M Errington, T. M., Denis, A., Perfito, N., Iorns, E. & Nosek, B. A. (2021). Reproducibility in Cancer Biology: challenges for assessing replicability in preclinical cancer biology. eLife 2021;10:e67995 doi: 10.7554/eLife.67995.
Kaiser, J. (2021). More than half of high-impact cancer lab studies could not be replicated in controversial analysis. Science, Vol 374, Issue 6573. doi: 10.1126/science.acx9770,