Cell-Free DNA Sequencing for Cancer Treatment Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current cancer treatment methods fail to effectively match specific cancers with effective drug treatments due to genetic and epigenetic differences among patients, and existing genomic analysis technologies are plagued by high error rates and biases, making it difficult to reliably detect de novo genomic alterations associated with cancer.
Innovation Solution
A system and method that uses a DNA sequencer to analyze genetic information at multiple time points, implement a machine learning algorithm to predict therapeutic responses, and detect genetic alterations in cell-free nucleic acids, thereby improving the accuracy of cancer diagnosis and treatment matching.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If current genomic analysis technologies are used to detect cancer genomic alterations, then the analysis can be performed, but the error rates and biases are orders of magnitude higher than required for reliable detection
Solution Approach 1:
The patent performs preliminary actions by sequencing the same genomic regions multiple times across different time points before making a detection call. This repeated measurement allows the system to distinguish true signals from noise through temporal consistency, thereby reducing error rates and improving reliability of genomic alteration detection.
Solution Approach 2:
The system uses feedback by comparing current sequencing results with prior time point data. This feedback mechanism allows the system to adjust and refine its detection calls, eliminating false positives and improving measurement precision through iterative verification against historical data.
2Adaptability or versatility
If cancer cells are constantly changing and mutating, then the disease state becomes a moving target, but this makes it more difficult to match cancers with effective drug treatments
Solution Approach 1:
The patent establishes a baseline genomic profile at multiple time points before treatment initiation and during treatment courses. This preliminary characterization creates a reference framework that allows the system to track and adapt to genomic changes in real-time, maintaining treatment matching capability despite cancer evolution.
Solution Approach 2:
The system transitions from static genomic analysis to dynamic, longitudinal monitoring. By continuously sequencing genomic regions at multiple time points, the system adapts to cancer evolution and can update treatment recommendations in response to changing genomic landscapes, thereby maintaining effectiveness despite loss of genomic stability.
3Measurement precision
If next-generation sequencing is used to detect genomic signals, then the technology can identify alterations, but the signals are so weak that detection is only possible in patients with terminally high tumor burden
Solution Approach 1:
The patent performs repeated preliminary sequencing at multiple time points, including early time points when tumor burden is low. By establishing baseline data early and accumulating information across time points, the system increases detection sensitivity without requiring high tumor burden, as the temporal pattern itself becomes the detection signal.
Solution Approach 2:
The system adds the temporal dimension to genomic analysis. Instead of relying solely on the strength of individual genomic signals (which requires high tumor burden), the system uses the pattern of signals across time as an additional dimension for detection, thereby improving sensitivity independent of tumor burden quantity.
Data Source
AI summary
Systems and methods are disclosed for generating a therapeutic response predict or detecting a disease, by: using a genetic analyzer to generate genetic information; receiving into computer memory a training dataset comprising, for each of a plurality of individuals having a disease, (1) genetic information from the individual generated at first time point and (2) treatment response of the individual to one or more therapeutic interventions determined at a second, later, time point; and implementing a machine learning algorithm using the dataset to generate at least one computer implemented classification algorithm, wherein the classification algorithm, based on genetic information from a subject, predicts therapeutic response of the subject to a therapeutic intervention.