Evidence based translational medicine

1
Evidence based translational medicine
Clinical trial

• Systematic review and meta-analysis
• how powerful is the treatment?
• what is the quality of evidence?
• what is the range of evidence?
• is there evidence of a publication bias?
• What are the conditions of maximum efficacy?

2
Cooling for stroke

• Cooling seems to work in patients who have brain
  injury due to cardiac arrest
• Theres lots of stories about individual patients
  who should have extensive brain damage but dont
• Many labs use cooling as a positive control in
  their animal studies
• Preliminary evidence from clinical trials in
  stroke is encouraging

3
How powerful is the treatment in animals?
101 publications 222 experiments 3256 animals
43.5 protection (40.1-47.0)
4
Evidence based clinical trial designHypothermia
for acute ischaemic stroke
Criterion Animal data EuroHYP-1
How powerful is the treatment? gt40 improvement in outcome Powered to detect 7 improvement in outcome
What is the quality of evidence? Efficacy maintained in high quality studies Randomised, blinded outcome assessment, intensely monitored
Is there evidence of a publication bias? Yes, but gt35 improvement in adjusted outcome Registered
What is the range of evidence? Good duration of cooling, delay to treatment, intensity, hypertension, reperfusion Patients gt18yo with moderate to severe stroke treated within 6 hrs
What are the conditions of maximum efficacy? Temperature dependent otherwise robust across dimensions Target 34-35C
5
(No Transcript)
6
Evidence based translational medicine
Clinical trial

• Multi Centre Animal Studies
• confirm efficacy
• robust and monitored conduct of experiments
• transparent analysis and reporting
• deliberate heterogeneity

• Systematic review and meta-analysis
• how powerful is the treatment?
• what is the quality of evidence?
• what is the range of evidence?
• is there evidence of a publication bias?
• What are the conditions of maximum efficacy?

7
Data More are better
Cumulative meta-analysis of the efficacy of lytic
treatments (eg tPA) in thrombotic animal models
of stroke
NXY-059
Hypothermia
8
Clinical trials and in vivo studies
9
MultiPARTMulticentre Preclinical Animal Research
Team
10
What weve found out so far

• Animal studies which do not report simple
  measures to avoid bias give larger estimates of
  how good drugs are
• Most animal studies do not report simple measures
  to reduce bias
• Publication and selective outcome reporting
  biases are important and prevalent
• You cannot assume rigour, even in Journals of
  impact
• You can only find these things out by studying
  large numbers of studies

11
What were looking into now

• Can we advise scientists on more efficient
  research design?
• Can we use this approach better to understand
  animal models of mental illness?
• Can we automate some of the techniques required?
• Can we help publishers improve quality?
• 3000 publications are added to PubMed each day
  can we build tools to provide up-to-date research
  summaries?
• Can we use this approach to design better
  clinical trials?

12
Deaths from Stroke, ScotlandAge 70 to 79
13
Abraham Maslow To a man with a hammer, every
problem looks like a nail

• George Santayana
• Those who cannot remember the past are condemned
  to repeat it

• Karl Marx
• Hegel said somewhere that history tends to
  repeat itself. He forgot to add that the first
  time is tragedy, the second is farce

14
(No Transcript)
15
Acknowledgements