RNA-SEQ IMMUNOPROFILING OF PERIPHERAL BLOOD
Aspects of the disclosure relate to methods, systems, and computer-readable storage media, that are useful for characterizing subjects having cancer. The disclosure is based, in part, on methods for immunoprofiling a cancer subject and the subject's prognosis and/or likelihood of responding to an immunotherapy based upon analysis of leukocyte populations in the peripheral blood of the subject.
1 . A method for determining a leukocyte immunoprofile type of a subject having, suspected of having, or at risk of having cancer, the method comprising:
using at least one computer hardware processor to perform:
obtaining RNA expression data for peripheral blood mononuclear cells (PBMC) isolated from a biological sample obtained from the subject;
processing the RNA expression data to determine cell composition percentages for at least 20 cell types listed in Table 3;
generating a leukocyte signature for the subject using the determined cell composition percentages for the at least 20 cell types, the leukocyte signature comprising the cell composition percentages for the at least 20 cell types; and
identifying, using the leukocyte signature and from among a plurality of leukocyte
immunoprofile types, a leukocyte immunoprofile type for the subject.
2 . The method of claim 1 , wherein the RNA expression data comprises bulk RNA expression data.
3 . The method of claim 1 , wherein processing the RNA expression data comprises applying a cell deconvolution technique comprising one or more machine learning models to obtain the cell composition percentages.
4 . The method of claim 1 , wherein the cell composition percentages comprise cell composition percentages for: Naive CD4+ Tregs, Naive CD4+ T cells, Naive CD8+ T cells, Naive B cells, Effector memory CD8+ T cells, Classical monocytes, and Non-classical monocytes.
5 . The method of claim 1 , wherein the cell composition percentages comprise cell composition percentages for: Naive CD4+ T cells, Naive CD8+ T cells, Naive B cells, Non-switched Memory IgM B cells, Class-switched memory B cells, Central memory, CD4+ T cells, CD4+ Tregs, Transitional memory CD4+ T cells, Central memory CD4+ T cells, All memory CD4+ T cells, CD4+ T cells, CD4+ T cells, Eosinophils, Basophils, Plasmacytoid dendritic cells, Dendritic cells, PD1high CD8+ T cells, Transitional memory CD8+ T cells, Cytotoxic NK cells, Regulatory NK cells, All memory CD8+ T cells, CD8+ T cells, CD8+ TEMRA, Effector memory CD8+ T cells, Neutrophils, Granulocytes, Classical monocytes, and Non-classical monocytes.
6 . The method of claim 1 , wherein the leukocyte signature comprises cell composition percentages for: Naive CD4+ Tregs, Naive CD4+ T cells, Naive CD8+ T cells, Naive B cells, Effector memory CD8+ T cells, Classical monocytes, and Non-classical monocytes.
7 . The method of claim 1 , wherein the leukocyte signature comprises cell composition percentages for: Naive CD4+ T cells, Naive CD8+ T cells, Naive B cells, Non-switched Memory IgM B cells, Class-switched memory B cells, Central memory, CD4+ T cells, CD4+ Tregs, Transitional memory CD4+ T cells, Central memory CD4+ T cells, All memory CD4+ T cells, CD4+ T cells, CD4+ T cells, Eosinophils, Basophils, Plasmacytoid dendritic cells, Dendritic cells, PD1high CD8+ T cells, Transitional memory CD8+ T cells, Cytotoxic NK cells, Regulatory NK cells, All memory CD8+ T cells, CD8+ T cells, CD8+ TEMRA, Effector memory CD8+ T cells, Neutrophils, Granulocytes, Classical monocytes, and Non-classical monocytes.
8 . The method of claim 1 , wherein the plurality of leukocyte immunoprofile types is associated with a respective plurality of leukocyte immunoprofile types,
wherein identifying, using the leukocyte signature and from among a plurality of leukocyte immunoprofile types, the leukocyte immunoprofile type for the subject comprises:
associating the leukocyte signature of the subject with a particular one of the plurality of leukocyte immunoprofile types, and
identifying the leukocyte immunoprofile type for the subject as the leukocyte immunoprofile type corresponding to the particular one of the plurality of leukocyte immunoprofile types to which the leukocyte signature of the subject is associated.
9 . The method of claim 8 , wherein associating the leukocyte signature of the subject with a particular one of the plurality of leukocyte immunoprofile types comprises:
processing the leukocyte signature with a trained classifier to obtain an output indicative of the particular one of the plurality of leukocyte immunoprofile types.
10 . The method of claim 9 , wherein the trained classifier comprises a trained neural network classifier, optionally, a tabular prior-data fitted network transformer (TabPFN) classifier.
11 . The method of claim 8 , wherein the associating the leukocyte signature of the subject with a particular one of the plurality of leukocyte immunoprofile types comprises:
determining, for each particular one of the plurality of leukocyte immunoprofile types, a score indicating whether the leukocyte signature of the subject is associated with that particular leukocyte immunoprofile type,
wherein determining the score for a particular leukocyte immunoprofile type comprises applying a linear regression model associated with the particular leukocyte immunoprofile type, to the cell composition percentages in the leukocyte signature.
12 . The method of claim 8 , wherein the associating the leukocyte signature of the subject with a particular one of the plurality of leukocyte immunoprofile types comprises:
determining, for each particular one of the plurality of leukocyte immunoprofile types, a score indicating whether the leukocyte signature of the subject is associated with that particular leukocyte signature cluster,
wherein determining the score for a particular leukocyte signature cluster comprises applying a linear regression model associated with the particular cluster, to the cell composition percentages in the leukocyte signature.
13 . The method of claim 1 , further comprising generating the plurality of leukocyte immunoprofile types, the generating comprising:
obtaining multiple sets of RNA expression data from white blood cells (WBC) isolated from biological samples obtained from multiple respective subjects, each of the multiple sets of RNA expression data indicating cell composition percentages for at least 20 cell types listed in Table 3;
generating multiple leukocyte signatures from the multiple sets of RNA expression data, each of the multiple leukocyte signatures comprising cell composition percentages for at least 20 cell types listed in Table 3, the generating comprising, for each particular one of the multiple leukocyte signatures:
determining the leukocyte signature by determining the cell composition percentages using the RNA expression data in the particular set of RNA expression data for which the particular one leukocyte signature is being generated; and
clustering the multiple leukocyte signatures to obtain the plurality of leukocyte immunoprofile types.
14 . The method of claim 1 , further comprising:
updating the plurality of leukocyte immunoprofile types using the leukocyte signature of the subject, wherein the leukocyte signature of the subject is one of a threshold number of leukocyte signatures for a threshold number of subjects, wherein when the threshold number of leukocyte signatures is generated the leukocyte immunoprofile types are updated,
wherein the threshold number of leukocyte signatures is at least 50, at least 75, at least 100, at least 200, at least 500, at least 1000, or at least 5000 leukocyte signatures.
15 . The method of claim 14 , wherein the updating is performed using a clustering algorithm selected from the group consisting of a dense clustering algorithm, spectral clustering algorithm, k-means clustering algorithm, hierarchical clustering algorithm, and an agglomerative clustering algorithm.
16 . The method of claim 14 , further comprising:
determining a leukocyte immunoprofile type of a second subject, wherein the leukocyte immunoprofile type of the second subject is identified using the updated leukocyte immunoprofile types, wherein the identifying comprises:
determining a leukocyte signature of the second subject from RNA expression data from white blood cells isolated from a biological sample obtained from the second subject;
associating the leukocyte signature of the second subject with a particular one of the plurality of the updated leukocyte immunoprofile types; and
identifying the leukocyte immunoprofile type for the second subject as the leukocyte immunoprofile type corresponding to the particular one of the plurality of updated leukocyte immunoprofile types to which the leukocyte signature of the second subject is associated.
17 . The method of claim 16 , wherein the clustering is performed using a dense clustering algorithm, a spectral clustering algorithm, a k-means clustering algorithm, hierarchical clustering algorithm, and/or an agglomerative clustering algorithm.
18 . The method of claim 17 , wherein the clustering is performed using a spectral clustering algorithm.
19 . The method of claim 1 , wherein the plurality of leukocyte immunoprofile types comprises: a Naïve type (G1), a Primed type (G2), a Progressive type (G3), a Chronic type (G4), and a Suppressive type (G5).
20 . The method of claim 1 , further comprising identifying the subject as a candidate for treatment with an immunotherapy based upon the identifying the leukocyte immunoprofile type for the subject.
21 . The method of claim 1 , further comprising identifying the subject as a candidate for treatment with an immunotherapy when the subject is identified as having a Primed type.
22 . The method of claim 1 , further comprising administering a therapeutic agent to the subject based upon identification of the subject's leukocyte immunoprofile type.
23 . The method of claim 1 , further comprising administering an immunotherapy to the subject when the subject is identified as having a Primed type.
24 . The method of claim 1 , wherein the subject has head and neck squamous cell carcinoma (HNSCC).
25 . A system, comprising:
at least one computer hardware processor; and
at least one 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 a method for determining a leukocyte immunoprofile type of a subject having, suspected of having, or at risk of having cancer, the method comprising:
using at least one computer hardware processor to perform:
obtaining RNA expression data for peripheral blood mononuclear cells (PBMC) isolated from a biological sample obtained from the subject;
processing the RNA expression data to determine cell composition percentages for at least 20 cell types listed in Table 3;
generating a leukocyte signature for the subject using the determined cell composition percentages for the at least 20 cell types, the leukocyte signature comprising the cell composition percentages for the at least 20 cell types; and
identifying, using the leukocyte signature and from among a plurality of leukocyte
immunoprofile types, a leukocyte immunoprofile type for the subject.
26 . At least one 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 a method for determining a leukocyte immunoprofile type of a subject having, suspected of having, or at risk of having cancer, the method comprising:
using at least one computer hardware processor to perform:
obtaining RNA expression data for peripheral blood mononuclear cells (PBMC) isolated from a biological sample obtained from the subject;
processing the RNA expression data to determine cell composition percentages for at least 20 cell types listed in Table 3;
generating a leukocyte signature for the subject using the determined cell composition percentages for the at least 20 cell types, the leukocyte signature comprising the cell composition percentages for the at least 20 cell types; and
identifying, using the leukocyte signature and from among a plurality of leukocyte
immunoprofile types, a leukocyte immunoprofile type for the subject.