Epigenetic Leukocyte Counts from Dried Blood Spots of Pediatric Patients

Extract

To the Editor:

We have recently described quantification of leukocytes from dried blood spots of adult patients by analysis of site-specific DNA methylation levels (1). In this follow-up study, we investigated if the method is also applicable for children. Leukocyte counts in pediatric patients are important for diagnostics of immunodeficiency and inflammation and to track leukocyte counts during immunosuppressive therapy. Typically, automated cell counters combined with flow cytometry form the basis of conventional techniques for subtyping of leukocytes. However, this method requires fresh blood samples, the required volume often exceeds what can be harvested from capillary blood, instrumentation is not available at many general practitioners, and standardization of these measurements is not trivial (2). Thus, there are certain limitations with conventional white blood counts that could be solved by innovative approaches—such as epigenetic blood counts.

Epigenetic changes control and reflect hematopoietic differentiation. Each leukocyte subtype has a characteristic DNA methylation pattern (1, 3). Based on such DNA methylation patterns, deconvolution of the composition of leukocytes is therefore possible, and this is even possible with targeted analysis at individual CG dinucleotides (CpGs) for each cell type tested (3). In contrast to conventional counts, epigenetic measurements are also feasible with dried blood spots, enabling long-term storage and easy shipment for centralized analysis. Furthermore, only small amounts of blood are required that can also be obtained from capillary sampling (i.e., finger- or heel-pricks) (1), which may be particularly useful for pediatric patients. However, DNA methylation at many CpGs changes significantly from birth to late adolescence (4). Therefore, epigenetic algorithms may have difficulty in estimating the cell composition in blood samples from newborns and younger children (5). We recently established optimized assays to determine DNA methylation with digital droplet PCR at 7 cell-type specific CpG sites: cg22381196 for granulocytes, cg23054181 for lymphocytes, cg04468741 for monocytes, cg05074138 for CD4 T cells, cg04329870 for CD8 T cells, cg02212339 for B cells, and cg05355684 for natural killer cells (1). These assays were performed in a cohort of adult patients (n = 310; age range: 18 to 82 years). For epigenetic cell counts, methylation data from 50 adult patients was used to train single linear regression models, which were then validated in the rest of the cohort, revealing high correlations with conventional counts (r = 0.95, r = 0.97, r = 0.82, r = 0.84, r = 0.94, r = 0.96, and r = 0.72, respectively) (1).