Somatic Mutations in T Cells in Patients with Hematological Diseases

Somatic mutations accumulate in cells during aging, and the mutation rate and spectrum vary between tissues. Mutation accumulation can be associated with inflammation: for example, myeloid driver gene mutations in hematopoietic stem cells (clonal hematopoiesis, CH), has been linked with atherosclerosis. Somatic mutations can also accumulate in T cells, where they are connected with immune dysregulation in large granular lymphocytic leukemia, characterized by autoimmunity and STAT3 mutations in T cells. However, the mutation spectrum in T cells in other hematological disorders is not yet fully discovered. We hypothesized that T cell lineage specific somatic mutations are associated with T cell clonality in various diseases involving the hematopoietic system.

We systematically characterized the spectrum of somatic mutations in CD4+ and CD8+ T cells in altogether 207 samples from 95 patients with chronic graft-versus-host disease (cGVHD, number of patients: 40), myelodysplastic syndrome (MDS, n = 15), immune thrombocytopenia (n = 2), aplastic anemia (AA, n = 21, previously published), and immunodeficiency (n = 17, previously published). Age at sampling was similar between disease cohorts. DNA was extracted from bead separated CD4+ and CD8+ cells from peripheral blood or bone marrow samples. We applied a custom-made targeted NGS panel, consisting of 2533 immune-related genes. Variant calling was performed using Mutect2. In variant filtering, we used 42 healthy T cell and skin samples as panel of normals. To explore the association between somatic mutations and clonal T cell expansions, we performed TCRb deep sequencing from 148 matched patient samples.

With mean sequencing depth of 382x, we identified 888 non-synonymous somatic variants in total. 784 (86%) of the variants were only found in either CD4+ or CD8+ fraction. 52 variants were shared between individual patient's CD4+ and CD8+ T cells, suggesting that they were acquired prior to T cell lineage differentiation. On the average, we observed 3.8 fraction-specific variants per sample. Mean variant allele frequency (VAF) was 4.5% and 12.7% for fraction-specific and shared variants, respectively. Variant types included missense (91%), nonsense (5.1%), splice-site (1.7%), frameshift (1.3%), and non-stop (0.2%) mutations. The most common mutational signatures were age-associated Signature 1 and defective DNA mismatch repair associated Signature 15.

Immunodeficiency patients had lower number of fraction-specific variants compared to AA, MDS and cGVHD patients (p = 0.027). In all cohorts, CD8+ T cells had acquired more variants than CD4+ T cells (p = 0.00012; mean [median] for CD8: 4.4 [4] and CD4: 3.2 [2]). The highest fraction-specific VAF in each sample was associated with the largest T cell clone size in CD8+ but not in CD4+ samples (R = 0.59, p = 3.1e-09 and p = ns, respectively). In CD8+ T cells, age was associated with number of fraction-specific variants (R = 0.38, p = 1.1e-4) but not with the highest CD8-specific VAF (p = ns).

Among the variants shared between individual patients' CD4+ and CD8+ T cells, we found mutations in DNMT3A (n = 3) and TET2 (n = 2) and ASXL1 (n = 1), the most recurrently mutated CH genes. In addition to shared mutations, in 12% of patients had fraction-specific myeloid driver gene mutations (e.g. DNMT3A, TET2, KRAS, NRAS) in CD4+ or CD8+ T cells. To identify pathways enriched for putative driver mutations, we performed analyses of fraction-specific T cell variants with OncodriveFM. The most significantly mutated gene ontology pathways included T cell co-stimulation, leukocyte migration, Ras protein signal transduction, and positive regulation of MAP kinase activity.

In conclusion, T cell fraction-specific somatic mutations are common in patients with various hematological diseases, and they are enriched in pathways important for T cell function. CD8+ T cells accumulate more mutations than CD4+ T cells. In CD8+ T cells, the number of somatic mutations was associated with age, reflecting stochastic accumulation of mutations. Based on mutational signatures, somatic mutations in T cells are mainly result of endogenous mutational processes. In addition, mutational clone size CD8+ T cells was significantly associated with the largest TCR clone, suggesting that mutations have accumulated in expanded CD8+ T cell clones. Further studies are warranted to determine the functional effects of somatic mutations in T cells.

Mustjoki:Novartis: Honoraria, Research Funding; Pfizer: Honoraria, Research Funding; Bristol-Myers Squibb: Honoraria, Research Funding.