March 21, 2012

expert reaction to the application of monkey research to human heart disease

A meta-analysis of research, published in PLoS ONE, into whether social status could predict heart disease in monkeys concluded that such research could not be applied to humans.

Sir Patrick Bateson FRS, Emeritus Professor of Ethology, University of Cambridge and Chair of the Bateson Review into primate research, said:

“Given the limitations imposed by small sample sizes, this meta-analysis is welcome and shows that, as always, scientists must be critical of all findings particularly when wide-ranging generalisations are drawn from them. Social hierarchy data are notoriously difficult to interpret both in animals and in humans and I am not surprised that the associations between hierarchy and CAD in monkeys were as messy as they have proved to be. It certainly doesn’t follow, however, that all non-human primate studies are worthless. To claim that would be another example of an unwarranted generalisation.”

Dr Mark Prescott, primatologist at the National Centre for the Replacement, Refinement and Reduction of Animals in Research (NC3Rs), said:

“The primary aim of the systematic review was to assess the strength of the evidence that social status/stress predicts ill health (coronary artery disease – CAD) in cynomolgus monkeys. The authors found that there does not appear to be a strong relationship between social status and CAD, and that this is sex-specific. There were too few data to draw from conclusions about social stress and CAD.

“The second aim of the review was to explore how the monkey evidence has been used in the debate about human inequalities in health. The review did not formally assess whether the monkey studies are predictive of human; nor did it formally assess publication bias. The authors looked at the % citations of the included monkey studies in human health inequalities literature (Table 3), and the ways in which the monkey data are being interpreted by commentators in that literature. The authors consider that commentators are being selective in their use of the available monkey data and that generalisation to human societies is not warranted.

“The methodological assessment of the included monkey studies (Table 2) gives further evidence of poor reporting in animal based research. In most cases it was not clear whether blinding was used, and a power or sample size calculation was not given. Similar findings were made in a large scale survey of reporting, experimental design and statistical analysis in biomedical research using laboratory animals, conducted on behalf of the NC3Rs and published in PLoS ONE in 2009. To improve standards of reporting and ensure that the data from animal experiments can be fully evaluated and utilised, including for systematic reviews of this kind, the NC3Rs has since developed the ARRIVE (Animal Research: Reporting In Vivo Experiments) guidelines. The guidelines were published in PLoS Biology in 2010 and to-date have been adopted by over 80 journals.*”

Professor Roger Lemon, Sobell Chair of Neurophysiology, UCL Institute of Neurology, said:

“This new PLoS ONE article, while clearly raising questions about this particular area of cardiovascular research, does not provide an argument against the benefits that NHP research brings to both fundamental knowledge and to human clinical advances.

1. I am surprised that this particular area was chosen for such a detailed analysis. It would not surprise me to discover that because macaque and human social structures are different that the monkey work might not have great relevance to human heart disease, but surely that doesn’t mean that these studies were not worthwhile in explaining the effects of social stress in monkeys on incidence of heart disease?

2. I am surprised that all the 14 papers covered seem to be the work of the same research group: in general scientific advances of every kind require validation by independent groups of researchers.

3. Most of the papers are quite old (all but one < 2000).

4. Most importantly, I would question the impact of the work, and therefore the premise that this work has influenced human clinical research unduly. The most cited paper has only 183 citations. Compare that, if you will, to the main monkey paper that paved the way for Deep Brain Stimulation treatment of Parkinson’s disease, with 1024 citations. (Bergman et al Science, 1990, so a slightly later paper than some of those covered in the PLoS review).”

‘The Monkey Puzzle: A Systematic Review of Studies of Stress, Social Hierarchies, and Heart Disease in Monkeys’ by Mark Petticrew and George Davey Smith, published in PLoS ONE on Wednesday 21 March 2012.
*Kilkenny C et al (2009) Survey of the quality of experimental design, statistical analysis and reporting of research using animals. PLoS ONE 4 (11): 1-11.