Trial Review

Did you know?

The ANZCTR now automatically displays published trial results and simplifies the addition of trial documents such as unpublished protocols and statistical analysis plans.

These enhancements will offer a more comprehensive view of trials, regardless of whether their results are positive, negative, or inconclusive.

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
ACTRN12624001085561p
Ethics application status
Submitted, not yet approved
Date submitted
6/08/2024
Date registered
9/09/2024
Date last updated
9/09/2024
Date data sharing statement initially provided
9/09/2024
Type of registration
Prospectively registered

Titles & IDs
Public title
Pretrained Resnet for surgical Outcome prediction using PET Hypometabolism and Excised Tissue (PROPHET)
Scientific title
Pretrained Resnet for surgical Outcome prediction using PET Hypometabolism and Excised Tissue (PROPHET) in participants planned for resective epilepsy surgery: Prospective Validation Study
Secondary ID [1] 312694 0
None
Universal Trial Number (UTN)
Trial acronym
Linked study record

Health condition
Health condition(s) or problem(s) studied:
Drug resistant epilepsy 334696 0
Condition category
Condition code
Neurological 331263 331263 0 0
Epilepsy

Intervention/exposure
Study type
Interventional
Description of intervention(s) / exposure
This is a prospective cohort study of patients who are undergoing resective epilepsy surgery for drug resistant epilepsy, designed to prospectively validate the clinical tool PROPHET, which predicts outcome following epilepsy surgery. PROPHET is a deep-learning based tool that utilises the preoperative T1-weighted MRI, preoperative 18F-FDG-PET scan, and a mask of the resection region as inputs to develop a predicted classification of the outcome following epilepsy surgery (Engel 1 vs Engel 2-4). PROPHET was developed by our team using retrospective data, and in model development, the mask of the resection region was generated from the difference map between the preoperative and postoperative T1-weighted MRIs, and therefore represented the actual resection region. However, to be clinically useful prior to epilepsy surgery, this tool will use a mask of the intended resection region in place of the actual resection region.

For all participants in this study, the intended resection region will be annotated manually prior to epilepsy surgery using brain imaging software such as ITK-SNAP or BrainLab. A prediction of the surgical outcome will be subsequently generated using the annotated intended resection region, preoperative MRI and preoperative PET (Model A). The predicted outcome generated by PROPHET will not be re-identified with study participants. Therefore, it will not be disclosed to the treating epileptologist or neurosurgeon prior to, or following, epilepsy surgery. The predicted outcome generated by PROPHET will not impact the clinical decision making about whether the patient will proceed to epilepsy surgery or the nature operation to be performed. The PROPHET predicted outcome will also not be shared with the participant.

We will subsequently observe the actual surgical outcome following epilepsy surgery up to 12 months, as measured by the Engel Surgical Outcome Scale. We will also collect the postoperative MRI in the subset of subjects for whom a postoperative MRI was acquired for clinical purposes.

The actual resection region will be generated for all patients who have a postoperative MRI, by generating a difference map between the preoperative and postoperative MRI (using the same method as in PROPHET development). Model B will be a repeat prediction using the preoperative MRI, preoperative PET and actual resection region.

For both models (model A using the intended resection region, and model B the actual resection region), the predicted Engel outcome and the actual Engel outcomes will be compared using area under the receiver operator characteristic (AUC) to validate the model.
Intervention code [1] 329223 0
Diagnosis / Prognosis
Comparator / control treatment
Model B, which uses the actual resection region will be the comparator group. Model B uses inputs (preop MRI, preop FDG-PET and actual resection region) that were used in model development and internal/ external validation.
Control group
Active

Outcomes
Primary outcome [1] 339034 0
PROPHET validity
Timepoint [1] 339034 0
12 months postoperatively
Secondary outcome [1] 438315 0
PROPHET validity
Timepoint [1] 438315 0
12 months postoperatively
Secondary outcome [2] 438316 0
Similarity between the intended resection region and the actual resection region
Timepoint [2] 438316 0
12 months postoperatively
Secondary outcome [3] 438317 0
Change in quality of life
Timepoint [3] 438317 0
Measured at 0 months and 12 months postoperatively
Secondary outcome [4] 438318 0
Change in depression
Timepoint [4] 438318 0
Measured at 0 months and 12 months postoperatively
Secondary outcome [5] 438319 0
Change in anxiety
Timepoint [5] 438319 0
Measured at 0 months and 12 months postoperatively

Eligibility
Key inclusion criteria
1. The participant can provide informed consent on their own behalf.
2. The participant is planned for resective epilepsy surgery (either temporal or extratemporal surgery).
3. A preoperative volumetric T1-weighted brain MRI is available.
4. A preoperative brain 18F-FDG-PET scan is available.
5. The participant is fluent in English.
6. The participant is eligible for Medicare
Minimum age
18 Years
Maximum age
No limit
Sex
Both males and females
Can healthy volunteers participate?
No
Key exclusion criteria
1. Age less than or equal to 17 years old
2. Prior resective epilepsy surgery
3. Non-resective epilepsy surgeries, such as laser interstitial thermal therapy (LITT), vagus nerve stimulator (VNS) implantation, deep brain stimulator (DBS) implantation

Study design
Purpose of the study
Diagnosis
Allocation to intervention
Non-randomised trial
Procedure for enrolling a subject and allocating the treatment (allocation concealment procedures)
Methods used to generate the sequence in which subjects will be randomised (sequence generation)
Masking / blinding
Open (masking not used)
Who is / are masked / blinded?



Intervention assignment
Single group
Other design features
Phase
Not Applicable
Type of endpoint/s
Statistical methods / analysis
The estimated sample size for this study is 50 participants. This number has been estimated to compare the expected model accuracy (area under the curve [AUC] = 0.8) to the null hypothesis (AUC = 0.5), in which the model does not perform better than random chance. In this calculation, power was set at 0.8, type 1 error at 0.05, and the expected proportion of positive cases (Engel class I) at 0.7. The estimated sample size will be re-evaluated in the interim feasibility analysis.

Obtaining a postoperative MRI is standard of care at our institution, at the 3 month mark. Therefore, we anticipate approximately 80% of the cohort (n=40) to be available for Model B.

For the analysis of model performance, two models will be assessed:
1. Model A: PROPHET prediction using the intended resection region
2. Model B: PROPHET prediction using the actual resection region

Data for each model will be classified as:
• True positive: model prediction “seizure free”; ground truth “seizure free”
• True negative: model prediction “not seizure free”; ground truth “not seizure free”
• False positive: model prediction “seizure free”; ground truth “not seizure free”
• False negative: model prediction “not seizure free”; ground truth “seizure free”

Using these classifications, we will calculate the AUC, kappa agreement between model prediction and outcome, sensitivity, specificity, positive predictive value, and negative predictive value.


Recruitment
Recruitment status
Not yet recruiting
Date of first participant enrolment
Anticipated
1/10/2024
Actual
Date of last participant enrolment
Anticipated
Actual
Date of last data collection
Anticipated
Actual
Sample size
Target
50
Accrual to date
Final
Recruitment in Australia
Recruitment state(s)
VIC
Recruitment hospital [1] 26904 0
The Alfred - Melbourne
Recruitment postcode(s) [1] 42966 0
3004 - Melbourne

Funding & Sponsors
Funding source category [1] 317125 0
University
Name [1] 317125 0
In kind support, Monash University
Country [1] 317125 0
Australia
Primary sponsor type
Hospital
Name
Alfred Hospital
Address
Country
Australia
Secondary sponsor category [1] 319387 0
None
Name [1] 319387 0
Address [1] 319387 0
Country [1] 319387 0

Ethics approval
Ethics application status
Submitted, not yet approved
Ethics committee name [1] 315878 0
Alfred Hospital Ethics Committee
Ethics committee address [1] 315878 0
https://www.alfredhealth.org.au/research/ethics-research-governance
Ethics committee country [1] 315878 0
Australia
Date submitted for ethics approval [1] 315878 0
31/07/2024
Approval date [1] 315878 0
Ethics approval number [1] 315878 0

Summary
Brief summary
Epilepsy surgery is the best treatment option for people living with epilepsy that cannot be controlled with medication alone, also known as drug resistant epilepsy. In people living with drug resistant epilepsy, epilepsy surgery offers a better chance of stopping seizures than medications alone, however, not all patients who have epilepsy surgery become seizure free. Many research studies that have tried to understand why some patients do not become seizure free after epilepsy surgery have focused on population data rather than looking at patients on an individual level. We have developed a tool that uses machine learning, a form of artificial intelligence, to make a personalised prediction of the outcome following epilepsy surgery. In its development, this tool used the brain imaging (magnetic resonance imaging [MRI] and positron emission tomography [PET] scans), as well as the actual surgical area, of patients who have already had epilepsy surgery. The aim of this study is to assess the accuracy of the tool when using the intended/ planned surgery in place of the actual resection region.
Trial website
Trial related presentations / publications
Public notes

Contacts
Principal investigator
Name 136070 0
Prof Terence O'Brien
Address 136070 0
School of Translational Medicine, Monash University, 99 Commercial Rd Melbourne VIC 3004
Country 136070 0
Australia
Phone 136070 0
+61 3 9903 0555
Fax 136070 0
Email 136070 0
[email protected]
Contact person for public queries
Name 136071 0
Dr Merran Courtney
Address 136071 0
School of Translational Medicine, Monash University, 99 Commercial Rd Melbourne VIC 3004
Country 136071 0
Australia
Phone 136071 0
+61 3 9903 0555
Fax 136071 0
Email 136071 0
[email protected]
Contact person for scientific queries
Name 136072 0
Dr Merran Courtney
Address 136072 0
School of Translational Medicine, Monash University, 99 Commercial Rd Melbourne VIC 3004
Country 136072 0
Australia
Phone 136072 0
+61 3 9903 0555
Fax 136072 0
Email 136072 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?

No Supporting Document Provided



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.