Critical Care Canada Forum Neuro Critical Care I · Critical Care Canada Forum Neuro Critical Care...
Transcript of Critical Care Canada Forum Neuro Critical Care I · Critical Care Canada Forum Neuro Critical Care...
How to use biomarkers of traumatic and ischemic brain injury
Jamie Hutchison
Professor of Pediatrics, University of Toronto
Critical Care Canada Forum Neuro Critical Care I
Nov 16th, 2011
Outline
• Definition of biomarkers
• Potential utility of biomarkers
• Steps necessary to bring biomarkers to the bedside
• Systematic review of the literature in Ped TBI
• Some specific studies - TBI
• Global cerebral ischemia and cardiac arrest
Traumatic Brain Injury (TBI)
• Most common cause of death and acquired disability in children and young adults
• Causes
• Falls
• MVA
• Bicycle
• Sports
• Assault
CIHI Analysis Head Injuries in Canada
Motor vehicle accident
(MVA)
Fall
Other
Inflicted injury
Sport
Causes of injuries (2000 – 2010 at the Hospital of Sick Children) in 97 children who had ICP monitoring
Definitions and Examples
• Analytes in biological samples
• Any measure that can predict a disease state or a response to a drug
• Examples – Physiological – Blood Pressure, Heart Rate
– Laboratory – Troponin
– Highly complex imaging modalities
– Multi-marker genomic/proteomic panels
– Proteins/lipids or metabolites in blood, CSF, saliva or urine
Patrick Kochanek – Head, Safar Resuscitation Research Centre, Pittsburgh
• Biomarkers have potential utility as diagnostic, prognostic, and therapeutic adjuncts in the setting of traumatic brain injury.
• They could be used to help determine which patients should receive which treatments.
• Two approaches are being used, namely, assessing markers of structural damage and quantifying mediators of the cellular, biochemical, or molecular cascades in secondary injury or repair.
Utility of biomarkers in disease
• Many biotechnology/pharmaceutical companies are interested in discovering and validating biomarkers as surrogate endpoints – Goal: Cost cutting
• $$$$ millions have been spent on failed drug trials in critical care
• Many companies hope that …If a new drug fails to improve the (surrogate) biomarker levels of disease then this drug should not be tested further and they should move on to the next drug
Problems with biomarker research
• Discovery must be done in one dataset and validation in another dataset – Often this sequence is not followed
• Sample sizes are often too low • Retrospective datasets are used • Sensitivity and specificity of tests vary from
manufacturer to manufacturer • Cross reactivity of the tests varies • Quality of assays varies across laboratories • Often there is a lack of knowledge of the biology ie. A
lack of understanding of how the biomarkers reflect the mechanisms of traumatic brain injury. Fingerprints of injury (eg. metabolomics) are difficult to interpret.
Rationale for studying biomarkers to for prognosis
• TBI can have devastating effects on the long-term function of children and adults.
• We remain unable to predict long-term function and neuropsychological outcome in the acute period (ie. In the ICU) post-injury.
• If we were able to accurately predict outcome early during hospitalization, we could stratify patients for rehabilitation interventions, to improve outcomes.
• End-of-life decisions are being made without accurate prognosis
History of biomarkers as tests of prognosis in TBI
• Single markers of structural damage to different brain cells (eg. S100B, neuron specific enolase) or small groups of molecules from the same family in serum and cerebrospinal fluid.
– Good correlation with outcome but not highly sensitive or specific
• Evolved to more complex approaches:
– Proteomics, lipidomics, metabolomics, multiplex immunoassays
Neuron
Oligodendrocyte
Astrocyte
Basement membrane
Leukocyte Glycoproteins
Endothelial cell
MOLECULAR FINGERPRINT
BLOOD
BRAIN CSF
The Molecular Fingerprint of the Neurovascular Unit
MOLECULAR FINGERPRINT
Protein and amino acid mediators of in injury Tsz-Yan (Milly) Lo
Time Brain Trauma
IEG
hsp
Cytokines, Adhesion molecules
Excitatory amino acids
Brain specific proteins
Apoptotic proteins
Systematic review: Biomarkers which are associated with outcome in children with severe TBI? Shibata ARO, LoT-Y, Hutchison JS, Guerguerian A-M
• Brain specific – Myelin Basic Protein (MBP), S100B, Glial Fibrillary Acid Protein (GFAP),
Nerve Growth Factor (NGF), neuron specific enolase (NSE)
• Inflammatory – L-selectin, Interleukin (IL)-1β, IL-6, IL-8 and IL-10
• Apoptotic – Bcl-2
• Other – Heme oxygenase 1 (HO-1), Quinolinic acid, fibrin degradation products
(FDP) and doublecortin (DCX).
• Amino acids – aspartate, glutamate and glycine
Biomarkers associated with age
• IL-10
• Caspase-1
• PARP
• Glutamate
• S-100b
• IL-6
• Trend: P-selectin, HO-1
Reference
Fraser D et al. PCCM 2010
Fraser D et al. PCCM 2010
AUC 0.98
S100B + L-selectin
S100B + IL-6
Sensitivity 100%
Specificity 96%
Lo TY et al. J Neurotrauma 2009
Multi-variable receiver operating characteristic curve for predicting Glasgow outcome scale at 6 months following
injury (N=28 children with TBI)
More complex approaches to biomarker discovery research
• Proteomics
• Lipidomics
• Metabolomics
403 416 417 504 523 832 ref **S100b 16. EMBP 14. FA7 1. OTC 5. VWF 2. ILRL1 3. TLR9 7. A1AT 4. TFPI2 <<GCS Scores>> 6. *A4 8. PON2 13. A1AG2 11. A2GL 9. SEPP1 12. FINC 19. *LCAP 21. MSLN 17. ITIH3 18. A2MG 22. MMP9 24. SGK1 25. VTNC 36. FA10 37. PON3 45. PGRP2 46. BORG5 30. FHR5 42. MMRN1 44. FCN3 31. CBPB2 33. PROC 53. RETBP 54. ZA2G 55. PLK2 58. CBG 65. A2AP 69. LRMP 64. F13A 71. ANGT 59. THRB 63. CETP 72. ZPI 60. MSMB 74. SGK3 62. GO45 77. RENAL 56. LRC25 57. CD72 66. CRIS1 78. TRFE 76. PAPPA 70. TLR7 47. PON1 81. GPX3 26. APOM 39. *ENOG 38. SAA4 48. EXL2 49. FUCO2 61. FHR2 67. AMBP 80. PROS 68. CD28 83. SGK2 82. BD01 79. *TAU 84. *SPTA2 40. APOL1 43. APOH 73. *MEF2B 75. ITIH4 10. SRF 20. SAMP 23. IL6RA 32. HRG 34. CLC4C 41. CO3 35. PLMN 15. CCL14 27. KLKB1 50. AACT 28. FETA 29. APOF 51. PLVAP 52. FHR1 86. IC1 87. FHR4 85. B2MG 88. EPO 89. PAP2 90. *NFH 92. FETUA 93. HPT 94. A1AG1 95. ALBU 91. A1BG
ref <<GCS Scores>> 1. OTC 5. VWF 2. ILRL1 3. TLR9 7. A1AT 4. TFPI2 6. *A4 8. PON2 13. A1AG2 11. A2GL 9. SEPP1 12. FINC 19. *LCAP 21. MSLN **S100b 16. EMBP 14. FA7 17. ITIH3 18. A2MG 22. MMP9 24. SGK1 25. VTNC 36. FA10 37. PON3 45. PGRP2 46. BORG5 30. FHR5 42. MMRN1 44. FCN3 31. CBPB2 33. PROC 53. RETBP 54. ZA2G 55. PLK2 58. CBG 65. A2AP 69. LRMP 64. F13A 71. ANGT 59. THRB 63. CETP 72. ZPI 60. MSMB 74. SGK3 62. GO45 77. RENAL 56. LRC25 57. CD72 66. CRIS1 78. TRFE 76. PAPPA 70. TLR7 47. PON1 81. GPX3 26. APOM 39. *ENOG 38. SAA4 48. EXL2 49. FUCO2 61. FHR2 67. AMBP 80. PROS 68. CD28 83. SGK2 82. BD01 79. *TAU 84. *SPTA2 40. APOL1 43. APOH 73. *MEF2B 75. ITIH4 10. SRF 20. SAMP 23. IL6RA 32. HRG 34. CLC4C 41. CO3 35. PLMN 15. CCL14 27. KLKB1 50. AACT 28. FETA 29. APOF 51. PLVAP 52. FHR1 86. IC1 87. FHR4 85. B2MG 88. EPO 89. PAP2 90. *NFH 92. FETUA 93. HPT 94. A1AG1 95. ALBU 91. A1BG
403 416 417 504 523 832
Hierarchical clustering of 95 differentially expressed proteins in comparison to admission GCS and S100B in children with TBI
GCS S100B
Haqqani, et al. J Neurotrauma 2007
Brain-selective proteins identified in sera (8 hours
post-TBI) of 6 paediatric patients
Protein name Tissue specificity
Spectrin alpha chain, brain
Neuron-specific enolase (NSE)
Neurofilament triplet H protein (200 kDa neurofilament protein)
Amyloid beta A4 protein precursor (APP) (ABPP)
Prostaglandin-H2 D-isomerase precursor (PDG2 synthase)
Microtubule-associated protein tau (Neurofibrilary tangle protein)
Contactin 3 precursor (Brain derived immunoglobulin superfamily protein- BIG-1)
The alpha/gamma heterodimer and the gamma/gamma homodimer found in neurons
Highly enriched in the brain
Abundant in the brain and CNS where it is expressed in the blood-brain barrier and secreted into the CSF
Expressed in neurons. Isoform PNS-tau – in peripheral nervous system; others are expressed in the CNS
Expressed in the brain – frontal and occipital lobe, cerebellum and amygdala
Non-erythroid; cleaved by calpain and caspase-3; detected in CSF after TBI
Neuron-specific cytoskeletal protein
Haqqani, et al. J Neurotrauma 2007
Novel brain trauma biomarkers: Significant post-traumatic changes in serum levels of eicosanoids and leukotrienes
Lo T et al. PCCM 2011 A113
Objective: To determine if eicosanoids and leukotrienes are elevated in the serum of children with severe TBI compared to healthy controls
• Methods:
– Measured 22 lipids in serum using mass spectrometry
– Compared 6 severe TBI to 8 healthy controls
• Results:
– 8/22 (36%) lipids were elevated in the serum of children with TBI compared to controls
Novel post-traumatic outcome prediction using acute serum levels of amino and organic acids Lo T et al. PCCM 2011 A112
• Objective: Determine if serum amino and organic
acids at day 1 post-injury are associated with outcome at 6 months post-TBI in children
• Methods: – Measured 36 different amino and organic acids using mass
spectrometry
– Compared 4 children with bad outcome to 14 children with good outcome
• Results: – 6 amino acids ↓ in those with bad outcome
– 3 – methyl-histidine ROC AUC 1.0
Multimodal approach to prognosis
• Clinical parameters
– Eg. Severity of injury
• Serum and CSF biomarkers
• Novel MRI techniques
• Neuro electrophysiology
Sham 2VO
forebrain
hindbrain
no perfusion
MICe – John Sled
Perfusion CT
Systemic (serum) inflammatory response
Sham 24h 72h
Time post-ischemia
pg
/mg
of
tota
l pro
tein
*#
Sham 24h 72h
Time post-ischemia
ng
/mg
of
tota
l pro
tein
*#
*#
IL-1β expression in cerebral cortex
Sham Ischemia
2009
Thank-you
Acknowledgements
• BTBI study coordinator Judy VanHuyse
• Co-principal Investigator – Vicki Anderson
• Co-investigators
• Biomarkers and lab - Arsalan Haqqani, Danica Stanimirovic, Doug Fraser, Milly Lo, Ryan Salewski, Martin Post
• CCCTBG and CCCTG - Brent Winston, John Marshall, and Deborah Cook
• Funding – ONF/VNI