Abstract
Background: Aplastic anemia (AA) and some subclasses of myelodysplastic syndromes (MDS) are immune-mediated bone marrow failure disorders. Previous studies indicate that cytotoxic T lymphocyte (CTL) clones play a role in the pathophysiology of these diseases by targeting hematopoietic stem cells. Current high throughput sequencing methods provide an unprecedented view to the characteristics of the T cell receptor (TCR) repertoire at the in-depth molecular level and opportunity to compare TCR repertoire in these diseases.
Methods: We investigated the TCR repertoire in patients with AA (n=12, mean age 58) and hypoplastic MDS (RA, RCMD) (n=20, mean age 61). Sorted CD8+ T cell fractions were analyzed with a multiplexed PCR assay that targets the variable CDR3 region of the rearranged TCRβ locus. Next-generation sequencing was performed and data was analyzed with the ImmunoSEQ, VDJtools (Shugay M et al. 2015) and VDJdb-standalone. The found rearrangements of T cell locus were compared to an age-matched control group (n=7, mean age 58) and previously published data from healthy adults (n=587), as well as to known immunodominant clones found in patients with large granular lymphocyte (LGL) leukemia. Lastly, VDJdb, a curated database of TCR sequences of known antigen specificity (https://vdjdb.cdr3.net) was used to investigate putative antigen-specific T cells.
Results: The overall T cell clonality of CD8+ T cells was significantly higher in patients with AA when compared to age matched controls (clonality index mean 0.39 vs. 0.28, p=0.03). However, in patients with hypoplastic MDS no significant difference in clonality was observed (0.30 vs. 0.28), although some patients had large expansions (clonality index >0.5). When all patients were analyzed as a group, age correlated both with the productive clonality (r=0.52, p=0.02) and the frequency of biggest productive rearrangement (r=0.53, p=0.01).
Several TCR rearrangements that were expressed in high frequency (>10% cumulative productive frequency) were shared among the AA and MDS patients, but were virtually absent in healthy individuals. Because LGL leukemia is closely connected to AA and hypoplastic MDS, and mono/oligoclonal LGL cells mediate autoimmune manifestations, we also compared our results to previously published TCR repertoire data of LGL leukemia patients. The immunodominant rearrangements found in AA and MDS patients (CTL rearrangements larger that 10%) were not shared with LGL leukemia cases. Similarly, the previously reported immunodominant rearrangements in LGL leukemia did not correspond to any major (>10%) clones in our patients. Only minor shared sequences were found which have also been discovered with a similar frequency in healthy controls.
Annotation of AA and MDS sequences to TCR database revealed that the reported antigen-specific sequences were not found in the immunodominant clones of our patients. The most frequent rearrangements with known antigen-specificity (up to 6.3% cumulative frequency) in our patients were specific for cytomegalovirus (CMV) and Epstein-Barr-virus. However, reported CMV-specific rearrangements were also found in a CMV-IgG negative patient, but with a low frequency (<10^-4 cumulative productive frequency).
From two patients we had both peripheral blood and bone marrow samples that were taken simultaneously. By comparing individual TCR sequences between 2 sample types we could observe that all T cell clones >0.1% were present in both samples with similar frequencies and there were only minor differences between the overall repertoires (r=0.97, r=0.96). We also compared T cell repertoires of patients before and after immunosuppressive treatment. A marked change in T cell repertoire was observed after successful treatment with cyclosporine A and anti-thymocyte globulin.
Conclusions: Next-generation TCRB sequencing of patients with AA and hypoplastic MDS confirmed that large cytotoxic T cell clones are found in these diseases. The molecular TCR structure of the clones was typical to AA and hypoplastic MDS and was not found in healthy individuals or patients with LGL leukemia. Furthermore, the rearrangements were not found among reported virus-specific sequences. Our results suggest that there is a common specific T cell target in AA and MDS. Deep molecular analysis of TCRs can further illuminate the mechanism of hematopoietic stem cell destruction in these diseases.
Mustjoki:Bristol-Myers Squibb: Honoraria, Research Funding; Ariad: Research Funding; Novartis: Honoraria, Research Funding; Pfizer: Honoraria, Research Funding.
Author notes
Asterisk with author names denotes non-ASH members.