SYSTEMS AND METHODS FOR IDENTIFYING CANCER TREATMENTS FROM NORMALIZED BIOMARKER SCORES
Techniques for determining predicted response of a subject to multiple therapies using the subject's sequencing data. The techniques include accessing biomarker information indicating a distribution of values for each biomarker in at least a reference subset of a plurality of biomarkers across a respective group of people, each of the plurality of biomarkers being associated with at least one therapy in a plurality of therapies; determining, using the sequencing data and the biomarker information, a normalized score for each biomarker in at least a subject subset of the plurality of biomarkers to obtain a set of normalized biomarker scores for the subject; and determining, using the set of normalized biomarker scores for the subject, therapy scores for the plurality of therapies, each of the therapy scores indicative of predicted response of the subject to administration of a respective therapy in the plurality of therapies.
1 . A system, comprising:
at least one computer hardware processor;
at least one database that stores biomarker information; and
at least one non-transitory computer-readable storage medium storing processor-executable instructions that, when executed by the at least one computer hardware processor, cause the at least one computer hardware processor to perform:
obtaining sequencing data about at least one biological sample of a subject;
accessing, in the 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 across a respective group of people, each of the plurality of biomarkers being associated with at least one therapy in a plurality of therapies;
determining, using the sequencing data and the biomarker information, a normalized score for each biomarker in at least a subject subset of the plurality of biomarkers to obtain a set of normalized biomarker scores for the subject, wherein the subject subset of the plurality of biomarkers is a subset of the reference subset of the plurality of biomarkers; and
determining, using the set of normalized biomarker scores for the subject, therapy scores for the plurality of therapies, each of the therapy scores indicative of predicted response of the subject to administration of a respective therapy in the plurality of therapies.
2 . The system of claim 1 , wherein the plurality of biomarkers includes a first biomarker, and wherein determining a normalized score for each biomarker in at least the subject subset of the plurality of biomarkers comprises:
determining a first normalized score for the first biomarker using the distribution of values for the first biomarker.
3 . The system of claim 2 , wherein determining the first normalized score comprises:
determining a first un-normalized score for the first biomarker using the sequencing data;
determining a first Z-score based on the first distribution of values for the first biomarker; and
determining the first normalized score for the first biomarker based on the first un-normalized score and the first Z-score.
4 . The system of claim 1 , wherein determining therapy scores for the plurality of therapies comprises determining a first therapy score for a first therapy in the plurality of therapies as a sum of two or more scores in the set of normalized biomarker scores for the subject.
5 . The system of claim 1 , wherein determining therapy scores for the plurality of therapies comprises determining a first therapy score for a first therapy in the plurality of therapies at least in part by:
determining weights for two or more scores in the set of normalized biomarker scores for the subject; and
determining the first therapy score as a weighted sum of the two or more scores, summands of the sum being weighted by the determined weights.
6 . The system of claim 5 , wherein determining the weights comprises determining the weights using a statistical model.
7 . The system of claim 6 , wherein determining the weights comprises determining the weights using a generalized linear model.
8 . The system of claim 6 , wherein determining the weights comprises determining the weights using a logistic regression model.
9 . The system of claim 1 , wherein the plurality of therapies comprises a first therapy and a second therapy different from the first therapy, and wherein determining therapy scores for the plurality of therapies comprises:
determining a first therapy score for the first therapy using a first subset of the set of normalized biomarker scores for the subject; and
determining a second therapy score for the second therapy using a second subset of the set of normalized biomarker scores for the subject,
wherein the second subset is different from the first subset.
10 . The system of claim 1 , wherein the processor-executable instructions further cause the at least one processor to perform:
providing the determined therapy scores to a user.
11 . The system of claim 1 , wherein the processor-executable instructions further cause the at least one processor to perform:
ranking the plurality of therapies based on the determined therapy scores.
12 . The system of claim 1 , wherein the processor-executable instructions further cause the at least one processor to perform:
recommending at least one of the plurality of therapies for the subject based on the determined therapy scores.
13 . The system of claim 12 , wherein recommending the at least one of the plurality of therapies comprises:
ranking the plurality of therapies based on the determined therapy scores; and
recommending at least a threshold number of top-ranked therapies for the subject.
14 . The system of claim 1 , wherein the plurality of therapies comprises at least two therapies selected from the group consisting of: an anti-PD1 therapy, an anti-CTLA4 therapy, an IL-2 therapy, an IFN alpha therapy, an anti-cancer vaccine therapy, an anti-angiogenic therapy, and an anti-CD20 therapy.
15 . The system of claim 1 , wherein the processor-executable instructions further cause the at least one processor to perform:
generating a graphical user interface (GUI) comprising:
a first portion associated with a first therapy in the plurality of therapies, the first portion including a first plurality of GUI elements, each of the first plurality of GUI elements being associated with a respective biomarker in the plurality of biomarkers 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 in the plurality of therapies, the second portion including a second plurality of GUI elements different from the first plurality of GUI elements, each of the second plurality of GUI elements being associated with a respective biomarker in the plurality of biomarkers 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.
16 . The system of claim 15 , wherein the at least one visual characteristic comprises color of a GUI element and/or size of the GUI element.
17 . The system of claim 15 , wherein the processor-executable instructions further cause the at least one processor to perform:
in response to receiving, via the GUI, a user selection of the first therapy, presenting, via the GUI, information about at least one biomarker with which at least one of the first plurality of GUI elements is associated.
18 . The system of claim 15 , wherein the first therapy is associated with a first therapy score and the second therapy is associated with a second therapy score, and wherein the first portion and the second portion are positioned, relative to one another in the GUI, based on relative magnitude of the first therapy score and the second therapy score.
19 . The system of 1 , wherein each of the plurality of biomarkers is selected from the group consisting of: a genetic biomarker, a cellular biomarker, a saccharide biomarker, a lipid biomarker, a heterocyclic biomarker, an elementary compound biomarker, an imaging biomarker, an anthropological biomarker, a personal habit biomarker, a disease-state biomarker, and an expression biomarker.
20 . The system of claim 1 , wherein each of the therapies are selected from immunotherapy and targeted therapy.
21 . At least one non-transitory computer-readable storage medium storing processor-executable instructions that, when executed by at least one computer hardware processor, cause the at least one computer hardware processor to perform:
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 across a respective group of people, each of the plurality of biomarkers being associated with at least one therapy in a plurality of therapies;
determining, using the sequencing data and the biomarker information, a normalized score for each biomarker in at least a subject subset of the plurality of biomarkers to obtain a set of normalized biomarker scores for the subject, wherein the subject subset of the plurality of biomarkers is a subset of the reference subset of the plurality of biomarkers; and
determining, using the set of normalized biomarker scores for the subject, therapy scores for the plurality of therapies, each of the therapy scores indicative of predicted response of the subject to administration of a respective therapy in the plurality of therapies.
22 . A method, comprising:
using at the at least one computer hardware processor to perform:
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 across a respective group of people, each of the plurality of biomarkers being associated with at least one therapy in a plurality of therapies;
determining, using the sequencing data and the biomarker information, a normalized score for each biomarker in at least a subject subset of the plurality of biomarkers to obtain a set of normalized biomarker scores for the subject, wherein the subject subset of the plurality of biomarkers is a subset of the reference subset of the plurality of biomarkers; and
determining, using the set of normalized biomarker scores for the subject, therapy scores for the plurality of therapies, each of the therapy scores indicative of predicted response of the subject to administration of a respective therapy in the plurality of therapies.