Rheumatoid arthritis (RA) is a chronic autoimmune, inflammatory disease, characterized
by synovitis in small- and medium-sized joints and, if not treated early and efficiently, joint
damage, and destruction. RA is a heterogeneous disease with a plethora of treatment
options. The pro-inflammatory cytokine tumor necrosis factor (TNF) plays a central role
in the pathogenesis of RA, and TNF inhibitors effectively repress inflammatory activity
in RA. Currently, treatment decisions are primarily based on empirics and economic
considerations. However, the considerable interpatient variability in response to treatment
is a challenge. Markers for a more exact patient classification and stratification are
lacking. The objective of this study was to identify markers in immune cell populations
that distinguish RA patients from healthy donors with an emphasis on TNF signaling.
We employed mass cytometry (CyTOF) with a panel of 13 phenotyping and 10
functional markers to explore signaling in unstimulated and TNF-stimulated peripheral
blood mononuclear cells from 20 newly diagnosed, untreated RA patients and 20
healthy donors. The resulting high-dimensional data were analyzed in three independent
analysis pipelines, characterized by differences in both data clean-up, identification of
cell subsets/clustering and statistical approaches. All three analysis pipelines identified
p-p38, IkBa, p-cJun, p-NFkB, and CD86 in cells of both the innate arm (myeloid dendritic
cells and classical monocytes) and the adaptive arm (memory CD4+ T cells) of the
immune system as markers for differentiation between RA patients and healthy donors.
Inclusion of the markers p-Akt and CD120b resulted in the correct classification of 18 of
20 RA patients and 17 of 20 healthy donors in regression modeling based on a combined
model of basal and TNF-induced signal. Expression patterns in a set of functionalmarkers
and specific immune cell subsets were distinct in RA patients compared to healthy
individuals. These signatures may support studies of disease pathogenesis, provide
candidate markers for response, and non-response to TNF inhibitor treatment, and aid
the identification of future therapeutic targets.