Non-parametric Heat Map Representation of Flow Cytometry Data: Identifying Cellular Changes Associated With Genetic Immunodeficiency Disorders

Abstract

Genetic primary immunodeficiency diseases are increasingly recognized, with pathogenic mutations changing the composition of circulating leukocyte subsets measured by flow cytometry (FCM). Discerning changes in multiple subpopulations is challenging, and subtle trends might be missed if traditional reference ranges derived from a control population are applied. We developed an algorithm where centiles were allocated using non-parametric comparison to controls, generating multiparameter heat maps to simultaneously represent all leukocyte subpopulations for inspection of trends within a cohort or segregation with a putative genetic mutation. To illustrate this method, we analyzed patients with Primary Antibody Deficiency (PAD) and kindreds harboring mutations in TNFRSF13B (encoding TACI), CTLA4, and CARD11. In PAD, loss of switched memory B cells (B-SM) was readily demonstrated, but as a continuous, not dichotomous, variable. Expansion of CXCR5+/CD45RA- CD4+ T cells (X5-Th cells) was a prominent feature in PAD, particularly in TACI mutants, and patients with expansion in CD21-lo B cells or transitional B cells were readily apparent. We observed differences between unaffected and affected TACI mutants (increased B cells and CD8+ T-effector memory cells, loss of B-SM cells and non-classical monocytes), cellular signatures that distinguished CTLA4 haploinsufficiency itself (expansion of plasmablasts, activated CD4+ T cells, regulatory T cells, and X5-Th cells) from its clinical expression (B-cell depletion), and those that were associated with CARD11 gain-of-function mutation (decreased CD8+ T effector memory cells, B cells, CD21-lo B cells, B-SM cells, and NK cells). Co-efficients of variation exceeded 30% for 36/54 FCM parameters, but by comparing inter-assay variation with disease-related variation, we ranked each parameter in terms of laboratory precision vs. disease variability, identifying X5-Th cells (and derivatives), naïve, activated, and central memory CD8+ T cells, transitional B cells, memory and SM-B cells, plasmablasts, activated CD4 cells, and total T cells as the 10 most useful cellular parameters. Applying these to cluster analysis of our PAD cohort, we could detect subgroups with the potential to reflect underlying genotypes. Heat mapping of normalized FCM data reveals cellular trends missed by standard reference ranges, identifies changes associating with a phenotype or genotype, and could inform hypotheses regarding pathogenesis of genetic immunodeficiency.

Keywords: CARD11; CTLA4; TACI; TNFSF13B; common variable immunodeficiency; flow cytometry; immunodeficiency.

Figure 1

Cellular subpopulations as measured by FCM. Schematic shows the 54 separate cell populations determined using the four antibody panels (see section Methods). Lines show the relationship to parent populations.

Figure 2

Heat mapping strategy for continuous green-black-red shading (A), or discontinuous shading (B). For continuous shading, colors vary continuously in shades of green (below control median), or red (above control median) with an intensity proportional to the difference from the median (0.5 centile). For discontinuous shading, colors vary stepwise as shown, based on the difference from the median, with the middle 50% of the control population in pale green, then progressively from light to dark blue (below control median), or yellow/orange/red/brown (above control median). Centiles for the control population are defined to vary from 1/(n+1) to 1–[1/(n+1)], depicted as “100%” in (B), where n is the size of the control population; such an approach allows identification of “extreme” or unprecedented values, namely those lying outside the bounds of the control population. Other percentages represent the proportion of the control population included in each indicated color stratum.

Figure 3

Heat mapping with continuous color shading (as depicted in Figure 2A) in PAD patients. Heat maps from individual patients (labeled a to v) are ordered from left to right based on increasing proportions of switched memory B cells (B-SM), with the cut-off of 0.4% of total lymphocytes shown. The expansion of CD21-lo B cells can be readily appreciated (labeled), as can the expansion in CXCR5+/CD45RA–CD4 T cells (X5-Th), and Tfh-effector cells and (as a percentage of CD4 T cells). Inset shows the comparisons between control values (n = 77) and PAD patients for X5-Th and Tfh-effector cells. Details of PAD patients presented in Supplementary Table 2.

Figure 4

Analysis of cellular parameters in CVID patients carrying TACI variants in comparison to TACI mutant control subjects without CVID. (A) Heat mapping with discontinuous shading for four unrelated CVID patients carrying the TACI A181E allele, one homozygous, along with two healthy individuals with A181E, or C104R alleles. Boxes highlight parameters for which cellular changes differ in those with CVID compared with TACI controls. (B) Scatter plots showing raw values for populations identified in (A), along with representative FCM contour plots of these critical parameters (C). Raw and centile data is presented in Supplementary Table 3.

Figure 5

Analysis of cellular parameters in a CTLA4 haploinsufficient kindred. (A) Heat mapping with discontinuous shading showing changes in cell populations for the CTLA4 haploinsufficient proband with CVID, in comparison with the clinically normal brother with the CTLA4 mutation, the unaffected proband's adult daughter without CTLA4 mutation and the internal control. Boxes highlight parameters for which cellular changes differ between the two CTLA4 mutants, the unaffected subjects, and between the CTLA4 mutants with or without clinical expression. (B) Scatter plots showing raw values for populations identified in (A), along with representative FCM contour plots of these critical parameters (C). Raw and centile data is presented in Supplementary Table 3.

Figure 6

Analysis of cellular parameters in a CARD11 mutant kindred. (A) Heat mapping with discontinuous shading showing changes in cell populations for two unrelated patients with dominant negative CARD11 mutations, along with their relative(s) without mutation. Boxes highlight cellular changes common to the two CARD11 mutants, but differing from the family members. (B) Scatter plots showing raw values for populations identified in (A), along with representative FCM contour plots of these critical parameters (C); in the B-cell contour plot, black boxes and numbers refer to total CD19+ cells, and red boxes and numbers refer to CD19+/CD27+ memory B cells. Raw and centile data is presented in Supplementary Table 3.

Figure 7

Comparison between CV percentages of 51 cellular parameters derived from the internal FCM control, and the same variation in the entire CPI cohort with widely varying immune conditions (see section Methods). The line shows the cut-off for the top 10 parameters with the highest ratio of disease-driven variation divided by internal variation, and these 10 parameters, and corresponding raw values appear in the table inset.

Figure 8

Cluster analysis of PAD patients based on centiles of the top 10 cellular parameters determined as outlined in the legend to Figure 7 (see text). Hereditary branches show relatedness in terms of cellular similarities, the major ones highlighted in boxes. Known genotypes are shown.