Systems and methods for identifying cancer treatments from normalized biomarker scores
Inventors
Bagaev, Alexander • Frenkel, Feliks • Ataullakhanov, Ravshan
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-11705220-B2
Publication Date
2023-07-18
Expiration Date
Abstract
Techniques for generating therapy biomarker scores and visualizing same. The techniques include determining, using a patient's sequence data and distributions of biomarker values across one or more reference populations, a first set of normalized scores for a first set of biomarkers associated with a first therapy, and a second set of normalized scores for a second set of biomarkers associated with a second therapy, generating a graphical user interface (GUI) including a first portion associated with the first therapy and having at least one visual characteristic determined based on a normalized score of the respective biomarker in the first set of normalized scores; and a second portion associated with a second therapy and having at least one visual characteristic determined based on a normalized score of the respective biomarker in the second set of normalized scores; and displaying the generated GUI.
Core Innovation
The invention relates to using sequencing data about at least one biological sample of a subject together with biomarker information stored in at least one database. The biomarker information indicates a distribution of values for each biomarker in a reference subset of a plurality of biomarkers, where each biomarker in the reference subset is associated with at least one candidate therapy associated with a clinical trial. The reference subset includes biomarkers associated with different genes, including a first biomarker and a second biomarker with respective first and second distributions of values.
Using the sequencing data, the invention determines un-normalized biomarker scores for at least some biomarkers in the reference subset, including a first un-normalized biomarker score for the first biomarker and a second un-normalized biomarker score for the second biomarker. It then determines normalized biomarker scores for the at least some biomarkers by adjusting the un-normalized biomarker scores to a common scale, with the first and second un-normalized biomarker scores adjusted using the first and second distributions of values, respectively.
The invention identifies the subject as a member of one or more cohorts using the normalized biomarker scores for the subject. Each cohort is associated with a positive or negative outcome of the at least one candidate therapy, and the method outputs an indication of the one or more cohorts in which the subject is identified as a member. The disclosed computing system performs the same steps.
Claims Coverage
The provided claim set contains three independent claims, with four main inventive features shared across them: obtaining sequencing data; retrieving biomarker value distributions linked to candidate therapies in clinical trials; computing un-normalized and normalized biomarker scores by adjusting to a common scale; and assigning the subject to one or more therapy-outcome cohorts and outputting the cohort indication.
Sequencing data with database biomarker distributions linked to clinical trial therapies
Obtaining sequencing data about at least one biological sample of a subject; accessing, in at least one database, biomarker information indicating a distribution of values for each biomarker in at least a reference subset of a plurality of biomarkers, each biomarker in the reference subset being associated with at least one candidate therapy associated with a clinical trial, including a first biomarker associated with a first gene and a second biomarker associated with a second gene different from the first gene, and the biomarker information including a first distribution of values for the first biomarker and a second distribution of values for the second biomarker.
Un-normalized biomarker scoring from sequencing data
Determining, using the sequencing data, un-normalized biomarker scores for at least some biomarkers in the reference subset of the plurality of biomarkers, including a first un-normalized biomarker score for the first biomarker and a second un-normalized biomarker score for the second biomarker.
Normalization to a common scale using reference distributions
Determining normalized biomarker scores for the at least some biomarkers at least in part by adjusting the un-normalized biomarker scores to a common scale, where adjusting comprises adjusting the first and second un-normalized biomarker scores to the common scale using the first and second distributions of values respectively.
Cohort assignment based on therapy positive/negative outcomes
Identifying the subject as a member of one or more cohorts using the normalized biomarker scores for the subject, where each of the one or more cohorts is associated with a positive or negative outcome of the at least one candidate therapy, and outputting an indication of the one or more cohorts in which the subject is identified as a member.
All independent claims share the core workflow: sequencing-data acquisition, retrieving biomarker value distributions tied to candidate therapies in clinical trials, calculating un-normalized and normalized biomarker scores by adjusting to a common scale using the stored distributions, and assigning the subject to one or more cohorts associated with positive or negative therapy outcomes with an output indicating the cohort memberships.
Stated Advantages
Not explicitly described in patent.
Documented Applications
Not explicitly described in patent.
Interested in licensing this patent?