Digital cytometry of immune cell types from dried blood microsamples
Enumeration of circulating immune cells is an established clinical practice implemented across a plethora of indications. While flow cytometry is the gold standard for this task, cellular deconvolution algorithms provide a robust computational alternative using molecular data. These digital cytometry pipelines typically rely on millilitre-scale venous blood samples. Here, we evaluated the potential of capillary blood microsampling as a patient-centric and minimally invasive approach for the digital cytometry of major lymphocyte populations (T cells, B cells, and natural killer cells; TBNK). In a cohort of 200 participants, we compared RNA sequencing based deconvolution performance between matched PAXgene-stabilised venous blood and Mitra capillary blood microsamples. While venous transcriptomes enabled highly accurate TBNK estimation, patient-matched capillary microsamples exhibited a significant reduction in deconvolution accuracy. Computational adaptations, including the integration of capillary-specific single-cell reference profiles, yielded only limited improvements. Instead, our analyses revealed that the performance deficit was primarily driven by the decreased transcriptome complexity of the capillary microsample libraries. This study highlights the critical role of transcript diversity for cellular deconvolution and reveals the current pre-analytical challenges that must be overcome to harness transcriptomics based digital cytometry using dried blood microsamples.
Authors
- Timothy Rajakumar (ORCID: https://orcid.org/0000-0002-0334-4349)
- Quentin R. V. Ferry (ORCID: https://orcid.org/0000-0003-1843-8572)
- Bruno R. Steinkraus (ORCID: https://orcid.org/0000-0003-3757-1993)
- Vanessa Chong-Morrison (ORCID: https://orcid.org/0000-0002-2547-3694)
- T Yon Schrenck
- Marco Zeeck
- Keihan Ahmadi-Simab
Institutions
- Dermatologikum Hamburg (DE)
Publication Details
- Journal
- medRxiv
- Published
- 2026-11-16
- DOI
- https://doi.org/10.64898/2026.11.13.26368027
- Primary Topic
- Single-cell and spatial transcriptomics
- Type
- preprint