CHICAGO—During the 2017 annual Federation of Clinical Immunology Societies (FOCIS) meeting, a session focused on precision immunology and its advances. Precision immunology describes the identification of host, immune system and tumor factors that can be used to select an immunotherapy approach. Thus, the first step in precision immunology is to identify soluble factors, immune cell characteristics and genome immune signatures that may correlate with clinical outcomes. The work in precision immunology is just beginning, but several approaches and several biomarkers are coming to the forefront.
You Might Also Like
Explore This IssueOctober 2017
Also By This Author
Advances in precision immunology will ultimately require precision controls. “To the degree that we have standards for immunology, they are from a pretty homogeneous population,” explained Kelly Zalocusky, PhD, a postdoctoral fellow from the University of California, San Francisco, during the session.
To counter that, she and her colleagues created the 10,000 Immunomes Project, a reference of diverse immune measurements from general population controls entered into the National Institute of Allergy and Infectious Diseases (NIAID) ImmPort database. The inaugural version of the reference contains 10,344 unique subjects and more than 42,000 samples (51% female and 49% male). Data types include secreted proteins (Luminex), flow cytometry, mass cytometry via Cytometry by Time of Flight, gene expression, clinical lab tests, hemagglutination inhibition assay titers, human leukocyte antigen type and others. The researchers filtered through the subjects and studies in the ImmPort database and extracted the responses from healthy subjects, thereby creating custom control cohorts.
Specifically, the 10,000 Immunomes Project contains high-throughput analysis of secreted proteins in 2,776 distinct subjects, with up to 63 proteins per subject. The investigators adapted an Empirical Bayes algorithm to correct the significant batch effects in the gene expression arrays. Before correcting for the batch effect, they found that protein measurements clustered by study accession. However, the clustering disappeared after batch correction.
The investigators performed a meta-analysis of Luminex data captured and identified known and novel associations between secreted protein levels, age, sex and ethnicity. Their analysis confirmed that leptin levels are higher in female than in male subjects. The researchers also documented numerous novel findings, including elevated serum CXCL5 levels in African-American subjects relative to other races.
The research by Dr. Zalocusky and colleagues also includes a meta-analysis of cytometry data from 1,415 distinct subjects, with up to 24 identified cell subsets per subject.1 The investigators developed an algorithm to automatically find cell subset percentages and name them according to the Human Immune Monitoring Center ontology. Their analysis revealed associations between cell subset percentages, age, sex and ethnicity. Example: They found that age had a strong effect on many cell subsets, including a decrease in naive CD8+ T cells with age and an increase in central memory CD4+ T cells with age.