Trial Review

The safety and scientific validity of this study is the responsibility of the study sponsor and investigators. Listing a study does not mean it has been endorsed by the ANZCTR. Before participating in a study, talk to your health care provider and refer to this information for consumers
Trial registered on ANZCTR


Registration number
ACTRN12620001360909
Ethics application status
Approved
Date submitted
25/09/2020
Date registered
17/12/2020
Date last updated
10/05/2023
Date data sharing statement initially provided
17/12/2020
Type of registration
Prospectively registered

Titles & IDs
Public title
PREdiction and Diagnosis using Imaging and Clinical biomarkers Trial in Traumatic Brain Injury
Scientific title
Diagnosis and Prognosis of Traumatic Brain Injury Outcome using Magnetic Resonance Imaging
Secondary ID [1] 302352 0
None
Universal Trial Number (UTN)
U1111-1258-5007
Trial acronym
PREDICT-TBI
Linked study record

Health condition
Health condition(s) or problem(s) studied:
Traumatic Brain Injury 319132 0
Condition category
Condition code
Injuries and Accidents 317095 317095 0 0
Other injuries and accidents
Neurological 317096 317096 0 0
Other neurological disorders

Intervention/exposure
Study type
Observational
Patient registry
False
Target follow-up duration
Target follow-up type
Description of intervention(s) / exposure
Participants will undergo blood tests for specific biomarker analyses and Magnetic Resonance scans.
Blood samples (25ml) will be taken at the following times by a phlebotomy trained nurse or phlebotomist:
Within 36 hours of injury
Between 48 and 72 hours of injury
Day 4 post injury
Day 7 post injury
Day 14 post injury
3 month post injury (+/- 14 days)
MRI scans will be done up to 3 times by trained radiographers and/or technicians with medical cover always provided, at the following timepoints:
Following discharge from ICU
3 months post injury (+/- 14 days)
6 months post injury (+/- 21 days)
MRIs will take approximately one hour.
All study procedures will cease six months post injury.
Study outcome questionnaires will be administered 3 and 6 months post injury, either face to face or over the phone by a research coordinator.
These questionnaires are PHQ-9, PHQ-15, BRS, BRISC, GAD-7, PROMIS, GOS-E, PCL-5
Intervention code [1] 318642 0
Diagnosis / Prognosis
Comparator / control treatment
No control group
Control group
Uncontrolled

Outcomes
Primary outcome [1] 325180 0
The combined ability of MRI scans, blood biomarkers (Circulating cell-free DNA (ccfDNA) UCHL1 levels (GE/ml), , GFAP, NF-L, Aß40, Aß42, total tau and phospho-tau derivatives (pg/ml), Neural exosomes (particles/ml) and neural exosome biomarker profile to predict patient outcome
Timepoint [1] 325180 0
Timepoints are: Blood Samples: Within 36 hours of injury, between 48 - 72 hours post injury, Day 4 post injury, Day 7 post injury Day 14 post injury, 3 month post injury (+/- 14 days). MRI will be done: Post ICU discharge, 3 months post injury (+/- 14 days) (bloods will also be taken on this day - as above) and 6 months post injury (+/- 21 days). Outcome Measure Questionnaires will be done 3 months post injury (+/- 14 days) 6 months post injury (+/- 21 days).
Secondary outcome [1] 387102 0
The effectiveness of the use of MRI with advanced neuro-imaging such as the Diffusion Tension Imaging (DTI) and tractography to predict patient outcome.
Timepoint [1] 387102 0
6 months post injury
Secondary outcome [2] 389200 0
The effectiveness of the use of blood biomarkers will be reflected by a validated panel of neuro-inflammatory biomarkers that reflect blood-brain-barrier disruption.
Timepoint [2] 389200 0
6 months post injury
Secondary outcome [3] 389201 0
Patient Reported Outcome Measurement Information System - questionnaire that measures patient-reported outcomes (PROs), such as pain, fatigue, physical functioning, emotional distress, and social role participation that have a major impact on quality-of-life


Timepoint [3] 389201 0
6 months post injury
Secondary outcome [4] 389842 0
Generalised Anxiety Disorder 7 item Scale - a questionnaire for diagnosing common anxiety disorders
Timepoint [4] 389842 0
6 months post injury
Secondary outcome [5] 389843 0
Post Traumatic Stress Disorder Checklist - A 20-item self report scale to assess symptoms of PTSD
Timepoint [5] 389843 0
6 months post injury
Secondary outcome [6] 389844 0
Patient Health Assessment 9 questionnaire - to monitor the severity of depression and response to treatment
Timepoint [6] 389844 0
6 months post injury
Secondary outcome [7] 389845 0
Patient Health Assessment 15 questionnaire - to monitor the severity of depression and response to treatment
Timepoint [7] 389845 0
6 months post injury
Secondary outcome [8] 389846 0
Brief Resilience Scale - an assessment to assess resilience. It consists of six statements for individuals to agree or disagree with. When completed it generates a resilience score of between 6 and 30
Timepoint [8] 389846 0
6 months post injury
Secondary outcome [9] 389847 0
Brief Risk Resilience Index for Screening - questionnaire which assesses processes of emotion regulation, including risk for experiencing negative emotional states (negativity bias) and coping responses (emotional resilience, social skills).
Timepoint [9] 389847 0
6 months post injury
Secondary outcome [10] 389850 0
Glasgow Outcome Scale - Extended (GOS-E) - this is a global scale for functional outcome that rates patient status into one of five categories
Timepoint [10] 389850 0
6 months post injury

Eligibility
Key inclusion criteria
1. Age greater than or equal to 18 years old
2. Diagnosed with moderate or severe TBI*, with or without other injuries.
*TBI Definition - A TBI will be defined based on the post-resuscitation Glasgow Coma Scale (GCS) in the pre-hospital period and the first 24 hours of ICU admission, if admitted to ICU. Moderate and Severe TBI are defined by the following categories of GCS if in the opinion of the treating team GCS is not solely due to intoxication, sedation, or extracranial injury
• Moderate TBI is defined as a GCS score of 9-12
• Severe TBI is defined as a GCS score of less than or equal to 8
Minimum age
18 Years
Maximum age
No limit
Sex
Both males and females
Can healthy volunteers participate?
No
Key exclusion criteria
1. Previous major stroke
2. Pregnancy, or may be pregnant
3. In the opinion of the investigator the participant would be unlikely to be able to comply with study procedures and follow up (e.g. lives overseas).
4. Presence of underlying disease with a life expectancy of less than 6 months
5. Known contraindication to MRI that will prevent study procedures
6. Patients who have suffered a devastating TBI with either progression towards brain death at the time of assessment or where the treating medical team are not committed to ongoing full supportive care
7. Traumatic brain injury occurred more than 72 hours ago

Study design
Purpose
Natural history
Duration
Longitudinal
Selection
Defined population
Timing
Prospective
Statistical methods / analysis
Deep learning models for each of the MR (Magnetic Resonance), CT (Computerized Tomography), MR + CT, MR + CT + Biomarkers data will be constructed and trained for classifying and predicting the clinical outcome measures such as GOS-E (Glasgow Extended Scale - Extended). To ensure numbers are sufficient for the models, data augmentation will be conducted to generate a balanced training set, which is well accepted in the medical image analysis literature. The features, measures and regions within the data found to be most influential in the training and prediction of the clinical outcomes will be visualised, analysed and compared to clinical workflows to determine which measures and data is most beneficial for (Traumatic Brain Injury) TBI prognostication. Additionally, we will investigate the use of transfer learning, whereby an initial model is trained from a separate dataset and is then fine-tuned on the desired dataset. Specifically, MR images from the open Human Connectome Project (HCP) or the OASIS Study datasets could be used to boost the training numbers of the deep learning models, effectively increasing the number of images to build the generative models across the three timepoints. The state-of-the-art HPC facilities at UQ’s Research Computing Centre (RCC) would be employed to undertake the analysis. A standard n-fold cross validation will be used to split the imaging data into training and validation sets, while compared to standard regression modelling.

Recruitment
Recruitment status
Recruiting
Date of first participant enrolment
Anticipated
1/02/2021
Actual
17/07/2021
Date of last participant enrolment
Anticipated
1/12/2023
Actual
Date of last data collection
Anticipated
31/05/2024
Actual
Sample size
Target
300
Accrual to date
123
Final
Recruitment in Australia
Recruitment state(s)
NSW,NT,QLD,VIC
Recruitment hospital [1] 17582 0
Royal Brisbane & Womens Hospital - Herston
Recruitment hospital [2] 17583 0
Royal Darwin Hospital - Tiwi
Recruitment hospital [3] 17584 0
Princess Alexandra Hospital - Woolloongabba
Recruitment hospital [4] 17585 0
Liverpool Hospital - Liverpool
Recruitment hospital [5] 17586 0
Gold Coast University Hospital - Southport
Recruitment hospital [6] 17587 0
The Townsville Hospital - Douglas
Recruitment hospital [7] 21539 0
The Alfred - Melbourne
Recruitment postcode(s) [1] 31325 0
4029 - Herston
Recruitment postcode(s) [2] 31326 0
0810 - Tiwi
Recruitment postcode(s) [3] 31327 0
4102 - Woolloongabba
Recruitment postcode(s) [4] 31328 0
2170 - Liverpool
Recruitment postcode(s) [5] 31329 0
4215 - Southport
Recruitment postcode(s) [6] 31330 0
4814 - Douglas
Recruitment postcode(s) [7] 36446 0
3004 - Melbourne

Funding & Sponsors
Funding source category [1] 306785 0
Government body
Name [1] 306785 0
Medical Research Future Fund (MRFF)
Country [1] 306785 0
Australia
Funding source category [2] 310606 0
Commercial sector/Industry
Name [2] 310606 0
Motor Accident Insurance Commission (MAIC)
Country [2] 310606 0
Australia
Primary sponsor type
University
Name
University of Queensland
Address
Queensland Brain Institute
St Lucia
Queensland 4072
Country
Australia
Secondary sponsor category [1] 307336 0
None
Name [1] 307336 0
Address [1] 307336 0
Country [1] 307336 0

Ethics approval
Ethics application status
Approved
Ethics committee name [1] 306944 0
Royal Brisbane & Womens Hospital Ethics Committee
Ethics committee address [1] 306944 0
Executive Suites
Lower Ground Floor
Dr James Mayne Building
Royal Brisbane & Women's Hospital
Brisbane
Queensland 4029
Ethics committee country [1] 306944 0
Australia
Date submitted for ethics approval [1] 306944 0
11/09/2020
Approval date [1] 306944 0
22/09/2020
Ethics approval number [1] 306944 0
HREC/2020/QRBW/66058

Summary
Brief summary
The objective of this study, in adults with moderate to severe TBI, is to use machine learning and determine and compare the univariate and multivariate associations between neuroimaging biomarkers, blood biomarkers, clinical data and neurological outcomes.

Specific aims are as follows:
1. To identify neuroimaging biomarkers of TBI, measured as structural and functional damage on brain MRI, on discharge from ICU/HDU (or if not admitted to ICU, once the participant is stable on the ward)and changes in structural damage and neuro-inflammation during recovery at 3- and 6-months after injury.
2. To determine the multivariate associations between neuroimaging biomarkers of TBI and neurological outcomes 3- and 6-months after injury, using a combination of deep learning and other machine learning methodologies.
3. To determine correlations between neuroimaging biomarkers, blood biomarkers (ccfDNA, exosomes, clinical data and neurological outcomes of TBI on discharge from ICU and at 3- and 6-months after injury.
4. To determine correlations between 3-month and 6-month changes in neuroimaging biomarkers, blood biomarkers, clinical data and neurological outcomes.
5. To determine the independent predictors of neurological outcomes at 3- and 6-months after injury using deep learning.
6. To assess aims 2-4 in the subgroup of patients who have MRI scans (during their ICU/HDU admission) as part of their clinical care
Trial website
Trial related presentations / publications
Public notes

Contacts
Principal investigator
Name 105534 0
A/Prof Fatima Nasrallah
Address 105534 0
Queensland Brain Institute
University of Queensland
St Lucia
Queensland 4072
Country 105534 0
Australia
Phone 105534 0
+61 7344 33004
Fax 105534 0
Email 105534 0
[email protected]
Contact person for public queries
Name 105535 0
Dr Tracey Evans
Address 105535 0
Queensland Brain Institute
The University of Queensland
St Lucia
Queensland 4072
Country 105535 0
Australia
Phone 105535 0
+61 7 3443 3822
Fax 105535 0
Email 105535 0
[email protected]
Contact person for scientific queries
Name 105536 0
A/Prof Fatima Nasrallah
Address 105536 0
Queensland Brain Institute
University of Queensland
St Lucia
Queensland 4072
Country 105536 0
Australia
Phone 105536 0
+61 7344 33004
Fax 105536 0
Email 105536 0
[email protected]

Data sharing statement
Will individual participant data (IPD) for this trial be available (including data dictionaries)?
No
No/undecided IPD sharing reason/comment


What supporting documents are/will be available?

Doc. No.TypeCitationLinkEmailOther DetailsAttachment
9223Ethical approval    380619-(Uploaded-22-09-2020-11-39-59)-Study-related document.pdf



Results publications and other study-related documents

Documents added manually
No documents have been uploaded by study researchers.

Documents added automatically
No additional documents have been identified.