Reconstitution of donor hematopoiesis after allogeneic hematopoietic stem cell transplantation (HSCT) forms the basis for effective graft-versus-leukemia responses, but mixed chimerism is not an infrequent outcome. How the donor and host hematopoietic system interact under conditions of mixed chimerism remains incompletely understood. Multi-modal single cell sequencing platforms are increasingly available and provide information regarding cell identities and interactions at high resolution. However, the analysis of post-transplant immune reconstitution requires consistent distinguishing of donor- and recipient-derived cells, which for sparse single cell sequencing data until now has remained a challenge.
Recently, mitochondrial DNA (mtDNA) mutations have been recognized for their potential as personal genetic barcodes that can be detected with mitochondrial single cell assay for transposase-accessible chromatin with sequencing (mtscATAC-seq). We hypothesized that individual-specific mtDNA mutations could provide a sensitive and robust approach for distinguishing donor- from recipient-derived cells, and therefore tested this approach on bone marrow (BM) samples from patients with relapsed acute myeloid leukemia (AML) post-HSCT. We employed ATAC with select cell surface antigen profiling by sequencing (ASAP-seq), which enables the detection of mtDNA mutations within distinct surface marker-defined cell populations alongside chromatin accessibility. We selected serial samples collected from the ETCTN 10026 study, which tested combined decitabine (days 1-5, every 4 weeks, start cycle 0) and ipilimumab (day 1, every 4 weeks, start cycle 1) in relapsed AML post-HSCT. We focused on 13 samples (study entry, on treatment and disease progression) from 3 patients: AML1012 (HSCT from a matched related donor [MRD]), and AML1010 and AML1026 (matched unrelated donor [MUD]-HSCT).
In total, we obtained 33,943 ASAP-seq profiles, including 3,283 single T cells. While clustering using single cell chromatin profiles alone only allowed identification of either CD4 + or CD8 + T cells, integration with surface marker expression enabled more detailed annotations of 8 T cell subpopulations and NK cells. Further, phenotypically distinct subpopulations such as CD57 + CD4 + and CD8 + T cells shared highly similar chromatin profiles, and 11.1% of CD4 + and 33.7% of CD8 + T cells would have been mislabeled based on clustering of chromatin profiles alone. Thus, ASAP-seq identified T cell subsets with markedly improved accuracy and resolution than scATAC-seq alone.
Upon evaluation of mtDNA mutations to discriminate donor- and recipient-derived single T cells, we found that this was unreliable for MRD-HSCT (AML1012), but highly robust in the setting of MUD-HSCT (AML1010, AML1026), consistent with maternal inheritance of mitochondrial genomes. For the latter two patients, we identified 48 donor- and 26 recipient-specific mtDNA mutations, all with high heteroplasmy (range 82 - 99%). Presence of donor- and recipient-derived mtDNA mutations was mutually exclusive, and recipient-specific mtDNA mutations were also detectable in AML cells. Clinical bulk and mtDNA mutation-based single T cell chimerisms were highly correlated (r = 0.97).
AML1010 had sustained complete T cell chimerism (>97%) during study treatment. In AML1026, the mtDNA mutation-based T cell chimerism rose from 55% to 71% after 1 cycle of decitabine and then remained stable until disease progression 3 months later. This was associated with increased percentage of donor-derived CD4 + T cells (45% [study entry] vs. 71% [after 1 cycle of decitabine], p < 0.01), while donor-derived CD8 + T cells remained unchanged at 76%. Across all studied timepoints in AML1026, donor versus recipient skewing was also highest in CD4 + T cell subsets, with fewer naïve (20% vs. 31%, p < 0.01) but more donor-derived CD57 + CD4 + T cells (13% vs. 3%, p < 0.01).
We demonstrate that mtDNA mutations can discriminate between donor- and recipient-derived single cells, enabling detection and in-depth characterization of chimeric immune cell dynamics after MUD HSCT. This approach will allow to systematically dissect conditions of mixed chimerism in the post-transplant setting with larger studies.
DeAngelo: Abbvie: Research Funding; Takeda: Consultancy; Servier: Consultancy; Pfizer: Consultancy; Novartis: Consultancy, Research Funding; Jazz: Consultancy; Incyte: Consultancy; Forty-Seven: Consultancy; Autolus: Consultancy; Amgen: Consultancy; Agios: Consultancy; Blueprint: Research Funding; Glycomimetrics: Research Funding. Neuberg: Madrigal Pharmaceuticals: Other: Stock ownership; Pharmacyclics: Research Funding. Sankaran: Cellarity: Consultancy; Forma: Consultancy; Novartis: Consultancy; Branch Biosciences: Consultancy; Ensoma: Consultancy. Soiffer: Jasper: Consultancy; Jazz Pharmaceuticals, USA: Consultancy; Precision Biosciences, USA: Consultancy; Juno Therapeutics, USA: Other: Data Safety Monitoring Board; Kiadis, Netherlands: Membership on an entity's Board of Directors or advisory committees; Rheos Therapeutics, USA: Consultancy; Gilead, USA: Other: Career Development Award Committee; NMPD - Be the Match, USA: Membership on an entity's Board of Directors or advisory committees; Takeda: Consultancy. Garcia: Genentech: Research Funding; Prelude: Research Funding; Pfizer: Research Funding; AstraZeneca: Research Funding; Takeda: Consultancy, Membership on an entity's Board of Directors or advisory committees; Astellas: Consultancy, Membership on an entity's Board of Directors or advisory committees; AbbVie: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding. Wu: Pharmacyclics: Research Funding; BioNTech: Current equity holder in publicly-traded company.
ipilimumab to modulate anti-leukemia immunity in the post-transplant and transplant-naive context
This icon denotes a clinically relevant abstract
This feature is available to Subscribers OnlySign In or Create an Account Close Modal