Method and system for predicting response to immune checkpoint inhibitor based on a density of immune cells
Inventors
Yoo, Donggeun • OCK, Chanyoung • Paeng, Kyunghyun
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12400735-B2
Publication Date
2025-08-26
Expiration Date
Abstract
The present disclosure relates to a method, performed by at least one computing device, for predicting a response to an immune checkpoint inhibitor. The method includes receiving a first pathology slide image, detecting one or more target items in the first pathology slide image, determining at least one of an immune phenotype of at least some regions in the first pathology slide image or information associated with the immune phenotype based on the detection result for the one or more target items, and generating a prediction result as to whether or not a patient associated with the first pathology slide image responds to the immune checkpoint inhibitor, based on the immune phenotype of the at least some regions in the first pathology slide image or the information associated with the immune phenotype.
Core Innovation
The invention uses an information processing system that inputs an H&E stained pathology slide image into an artificial neural network model. The artificial neural network model detects target items including immune cells, tumor cells, a cancer area, and a stroma area, with immune cells and tumor cells detected in units of cells and the cancer area and stroma area detected in units of areas. This slide-based detection is used to generate a prediction result for treatment response.
For each H&E stained pathology slide image, the system determines at least one region of interest (ROI) of a predetermined size and calculates a density of immune cells in the cancer area in the ROI and/or a density of the immune cells in the stroma area in the ROI. Based on the calculated density, the system determines an immune phenotype of the ROI by comparing the immune-cell density with a threshold density. The system also calculates an inflamed score for the patient based on a ratio of ROIs having an immune phenotype of a specific class.
The system classifies the patient associated with the H&E stained pathology slide image as a responder or a non-responder to the treatment based on the inflamed score, and outputs the prediction result. In additional embodiments, the invention outputs an immune phenotype map describing immune phenotypes for each ROI and can provide a report including the H&E stained pathology slide image, the treatment-response prediction result, a graph of immune phenotype ratios, and immune cell density information.
Claims Coverage
Two independent claims are identified. Each includes the same core inventive workflow of ANN-based detection on H&E slides, ROI-based immune-cell density calculation, threshold-based immune phenotype determination, inflamed score computation from ROI phenotype ratios, and responder/non-responder classification.
Ann-based detection of immune and tissue targets on H&E slides
Detect target items including immune cells, tumor cells, a cancer area and a stroma area in an H&E stained pathology slide image using an artificial neural network model, by inputting the H&E stained pathology slide image into the artificial neural network model and detecting the immune cells and the tumor cells as target items in units of cells and detecting the cancer area and the stroma area as target items in units of areas.
ROI immune-cell density calculation
Determine at least one region of interest (ROI) of a predetermined size in the H&E stained pathology slide image, and calculate, for each of the at least one ROI, at least one of a density of the immune cells in the cancer area in the ROI or a density of the immune cells in the stroma area in the ROI.
Threshold-based immune phenotype and inflamed score
Generate a prediction result as to whether or not a patient associated with the H&E stained pathology slide image responds to a treatment, by determining an immune phenotype of the at least one ROI by comparing at least one of the density of the immune cells in the cancer area in the ROI or the density of the immune cells in the stroma area in the ROI with a threshold density, calculating an inflamed score for the patient based on a ratio of ROIs having an immune phenotype of a specific class in the at least one ROI, and classifying the patient associated with the H&E stained pathology slide image as a responder or a non-responder to the treatment based on the inflamed score.
Across both independent claims, the coverage centers on ANN-based detection of immune and tissue targets in H&E stained pathology slide images, followed by ROI immune-cell density calculation, threshold-based immune phenotype determination, inflamed score computation from ratios of ROI phenotypes, and responder/non-responder classification based on the inflamed score.
Stated Advantages
Not explicitly described in patent.
Documented Applications
Not explicitly described in patent.
Interested in licensing this patent?