Using tissue images and patient information to predict genetic changes in colorectal cancer
| ISRCTN | ISRCTN39695759 |
|---|---|
| DOI | https://doi.org/10.1186/ISRCTN39695759 |
- Submission date
- 17/09/2024
- Registration date
- 22/09/2024
- Last edited
- 18/09/2024
- Recruitment status
- No longer recruiting
- Overall study status
- Completed
- Condition category
- Cancer
Plain English summary of protocol
Background and study aims
Colorectal cancer (CRC) is a common type of cancer that can behave differently depending on genetic changes in the tumor. One of these genetic changes is called microsatellite instability (MSI), which can help doctors choose the best treatment and predict how the cancer will progress. However, current ways of predicting MSI are not always accurate for all patients. This study aims to develop a new method that combines information from tissue samples and patient medical records to better predict MSI in colorectal cancer.
Who can participate?
Adults who have been diagnosed with colorectal cancer confirmed through a tissue biopsy can participate in this study. Participants must be fully aware of their condition, have had tests done to check for certain genetic changes (such as MSI and mutations in RAS or BRAF genes), and be willing to sign a consent form. People who have other types of cancer, incomplete test results, or poor-quality tissue images will not be eligible to take part.
What does the study involve?
Participants in this study will not need to undergo any new medical tests. Instead, the study will use existing data, including tissue samples and test results from their previous treatments. Researchers will analyze the images of these tissue samples using advanced computer methods, combining the findings with patient medical records to develop a new prediction model.
What are the possible benefits and risks of participating?
There are no direct health benefits or risks for participants in this study, as no new treatments or procedures are involved. However, the findings from the study could help improve future treatment strategies for colorectal cancer by offering better tools to predict genetic changes in the cancer.
Where is the study run from?
Jinhua Science and Technology Bureau (China)
When is the study starting and how long is it expected to run for?
The study will begin in March 2024 and is expected to end in July 2024.
Who is funding the study?
Jinhua Science and Technology Bureau (China)
Who is the main contact?
Hangping Wei, applewhp@163.com
Contact information
Public, Scientific, Principal Investigator
No. 60 West Wuning Road
Dongyang
322100
China
 ORCID ID |
0000-0002-2904-7086 |
|---|---|
| Phone | +86 13735632532 |
| applewhp@163.com |
Study information
| Study design | Observational cross sectional study |
|---|---|
| Primary study design | Observational |
| Secondary study design | Cross sectional study |
| Study setting(s) | Hospital |
| Study type | Diagnostic |
| Participant information sheet | Not applicable (retrospective study) |
| Scientific title | Hybrid model for predicting microsatellite instability in colorectal cancer using hematoxylin & eosin-stained images and clinical features |
| Study acronym | MSI |
| Study objectives | Two deep learning methods (semi-supervised and fully-supervised) were used to extract features from pathological images. Subsequently, the pathomic signatures derived from these methods were integrated with clinical features to develop a hybrid model. The hybrid model was evaluated using an external validation cohort to calculate the area under the curve (AUC). |
| Ethics approval(s) |
Approved 11/03/2024, The Institutional Ethical Review Board of Dongyang Hospital, affiliated with Wenzhou Medical University (No. 60 West Wuning Road, Dongyang, 322100, China; +86057986859051; dongxiaofang2022@163.com), ref: 2024-YX-039 |
| Health condition(s) or problem(s) studied | Colorectal cancer |
| Intervention | This study included two patient cohorts: The Cancer Genome Atlas cohort (TCGA set, n = 559), divided into training and internal validation subsets in a 7:3 ratio, and the Dongyang CRC cohort (Dongyang set), n = 123, which served as an external testing cohort. Two deep learning methods (semi-supervised and fully-supervised) were used to extract features from pathological images. Subsequently, the pathomic signatures derived from these methods were integrated with clinical features to develop a hybrid model. The hybrid model was evaluated using an external validation cohort to calculate the area under the curve (AUC). |
| Intervention type | Other |
| Primary outcome measure | Tumor diagnosis and MSI status measured using patient records at a single time point |
| Secondary outcome measures | There are no secondary outcome measures |
| Overall study start date | 11/03/2024 |
| Completion date | 31/07/2024 |
Eligibility
| Participant type(s) | Patient |
|---|---|
| Age group | All |
| Sex | Both |
| Target number of participants | 120 |
| Total final enrolment | 123 |
| Key inclusion criteria | 1. Confirmed by pathological histology as a patient with colorectal cancer 2. Clear consciousness 3. Relevant tests have been conducted, including results of microsatellite and common gene mutation states (RAS, BRAF) 4. The patient voluntarily participates and signs an informed consent form |
| Key exclusion criteria | 1. Patients with non primary colon cancer or a history of other organ malignancies at the same time 2. Results without microsatellite and common gene mutation states (RAS, BRAF) 3. Other patients with incomplete clinical data 4. Those with unclear pathological images |
| Date of first enrolment | 12/03/2024 |
| Date of final enrolment | 01/07/2024 |
Locations
Countries of recruitment
- China
Study participating centre
Dongyang
322100
China
Sponsor information
Government
Jinhua Science and Technology Bureau
Jinhua
322100
China
| Phone | +86 57986859052 |
|---|---|
| dongxiaofang2022@163.com | |
| Website | - |
"ROR" |
https://ror.org/0347g4065 |
Funders
Funder type
Government
Government organisation / Local government
- Alternative name(s)
- Science and Technology Bureau of Jinhua City, Jinhua Municipal Science and Technology Bureau
- Location
- China
Results and Publications
| Intention to publish date | 31/07/2025 |
|---|---|
| Individual participant data (IPD) Intention to share | Yes |
| IPD sharing plan summary | Stored in non-publicly available repository, Available on request |
| Publication and dissemination plan | Plan to publish in peer-reviewed journals |
| IPD sharing plan | The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request. Hangping Wei, applewhp@163.com |
Editorial Notes
18/09/2024: Trial's existence confirmed by The Institutional Ethical Review Board of Dongyang Hospital, affiliated with Wenzhou Medical University.
