Predicting cfDNA Shedding Patterns for Biopsy Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods lack the ability to accurately predict when a liquid biopsy versus a tissue biopsy should be performed for cancer patients, especially in cases with multiple or inaccessible lesions, leading to invasive procedures and compromised detection accuracy.
Innovation Solution
A method and system for predicting cell-free DNA (cfDNA) shedding using a trained model based on lesion and cfDNA datasets, which clusters new samples to determine the shedding pattern and decide on the appropriate biopsy type.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If tissue biopsy is performed to obtain accurate lesion molecular profile, then measurement precision is improved, but ease of operation deteriorates due to high invasiveness
Solution Approach 1:
The patent introduces cfDNA as an intermediary substance that carries lesion molecular information from the tissue to liquid form. By analyzing cfDNA shed from lesions into the bloodstream, the system obtains accurate molecular profiles without direct tissue contact, thus maintaining measurement precision while eliminating the invasiveness of tissue biopsy
Solution Approach 2:
The patent replaces the mechanical tissue biopsy procedure with a liquid biopsy approach. Instead of physically extracting tissue samples through invasive procedures, the system collects liquid samples (blood, urine, stool) containing cfDNA that can be analyzed to obtain the same molecular profile information, thereby substituting a minimally invasive procedure for a highly invasive one
2Ease of operation
If liquid biopsy is used to reduce invasiveness, then ease of operation is improved, but measurement precision deteriorates due to variable shedding patterns
Solution Approach 1:
The patent performs preliminary actions by collecting multiple cfDNA samples over time and performing tissue biopsies at predetermined intervals. This longitudinal sampling approach allows the system to establish baseline shedding patterns and update the machine learning model with fresh data, ensuring that the cfDNA measurements remain accurate and reflective of current lesion characteristics
Solution Approach 2:
The patent implements feedback mechanisms where cfDNA shedding patterns are continuously monitored and fed back into the machine learning model. The model uses this feedback to refine its predictions and determine when tissue biopsies are necessary to update the reference data. This closed-loop system ensures that measurement precision is maintained by adjusting the sampling strategy based on observed shedding variations
3Measurement precision
If frequent tissue biopsies are performed to monitor treatment response, then measurement precision is improved, but loss of time increases due to procedural complexity
Solution Approach 1:
The patent applies partial action by performing tissue biopsies only when necessary rather than at every monitoring interval. The machine learning model determines the optimal timing for tissue biopsies based on cfDNA shedding patterns, performing them only when the accumulated data indicates a need for reference updates. This reduces the total number of invasive procedures while maintaining measurement precision through strategic sampling
Data Source
AI summary
A method is provided for training a predicting cfDNA shedding model using a plurality of lesion and cfDNA datasets. A new cfDNA shedding sample and the plurality of lesion and cfDNA datasets are clustered to predict a shedding pattern. A diagnostic type is determined for a subsequent cfDNA shedding sample based on the predicted shedding pattern.