Tracing the origins of relapse in acute myeloid leukaemia to stem cells

Abstract

In acute myeloid leukaemia, long-term survival is poor as most patients relapse despite achieving remission1. Historically, the failure of therapy has been thought to be due to mutations that produce drug resistance, possibly arising as a consequence of the mutagenic properties of chemotherapy drugs2. However, other lines of evidence have pointed to the pre-existence of drug-resistant cells3. For example, deep sequencing of paired diagnosis and relapse acute myeloid leukaemia samples has provided direct evidence that relapse in some cases is generated from minor genetic subclones present at diagnosis that survive chemotherapy3,4,5, suggesting that resistant cells are generated by evolutionary processes before treatment3 and are selected by therapy6,7,8. Nevertheless, the mechanisms of therapy failure and capacity for leukaemic regeneration remain obscure, as sequence analysis alone does not provide insight into the cell types that are fated to drive relapse. Although leukaemia stem cells9,10 have been linked to relapse owing to their dormancy and self-renewal properties11,12,13, and leukaemia stem cell gene expression signatures are highly predictive of therapy failure14,15, experimental studies have been primarily correlative7 and a role for leukaemia stem cells in acute myeloid leukaemia relapse has not been directly proved. Here, through combined genetic and functional analysis of purified subpopulations and xenografts from paired diagnosis/relapse samples, we identify therapy-resistant cells already present at diagnosis and two major patterns of relapse. In some cases, relapse originated from rare leukaemia stem cells with a haematopoietic stem/progenitor cell phenotype, while in other instances relapse developed from larger subclones of immunophenotypically committed leukaemia cells that retained strong stemness transcriptional signatures. The identification of distinct patterns of relapse should lead to improved methods for disease management and monitoring in acute myeloid leukaemia. Moreover, the shared functional and transcriptional stemness properties that underlie both cellular origins of relapse emphasize the importance of developing new therapeutic approaches that target stemness to prevent relapse.

Access options

Rent or Buy article

Get time limited or full article access on ReadCube.

from$8.99

All prices are NET prices.

Figure 1: Mutational analysis reveals extensive pre-leukaemic genetic evolution and major clonal shifts at relapse.
Figure 2: AML relapse can originate from rare cells with a primitive functional and cellular phenotype.
Figure 3: AML relapse can originate from cells with a committed phenotype.
Figure 4: Distinct gene expression patterns are associated with relapse origin.

References

  1. 1

    Sarkozy, C. et al. Outcome of older patients with acute myeloid leukemia in first relapse. Am. J. Hematol. 88, 758–764 (2013)

    CAS  Article  Google Scholar 

  2. 2

    Goldie, J. H. & Coldman, A. J. The genetic origin of drug resistance in neoplasms: implications for systemic therapy. Cancer Res. 44, 3643–3653 (1984)

    CAS  PubMed  PubMed Central  Google Scholar 

  3. 3

    Ding, L. et al. Clonal evolution in relapsed acute myeloid leukaemia revealed by whole-genome sequencing. Nature 481, 506–510 (2012)

    CAS  ADS  Article  Google Scholar 

  4. 4

    Krönke, J. et al. Clonal evolution in relapsed NPM1-mutated acute myeloid leukemia. Blood 122, 100–108 (2013)

    Article  Google Scholar 

  5. 5

    Garg, M. et al. Profiling of somatic mutations in acute myeloid leukemia with FLT3-ITD at diagnosis and relapse. Blood 126, 2491–2501 (2015)

    CAS  Article  Google Scholar 

  6. 6

    Parkin, B. et al. Clonal evolution and devolution after chemotherapy in adult acute myelogenous leukemia. Blood 121, 369–377 (2013)

    CAS  Article  Google Scholar 

  7. 7

    Bachas, C. et al. The role of minor subpopulations within the leukemic blast compartment of AML patients at initial diagnosis in the development of relapse. Leukemia 26, 1313–1320 (2012)

    CAS  Article  Google Scholar 

  8. 8

    Greaves, M. Evolutionary determinants of cancer. Cancer Discov. 5, 806–820 (2015)

    CAS  Article  Google Scholar 

  9. 9

    Lapidot, T. et al. A cell initiating human acute myeloid leukaemia after transplantation into SCID mice. Nature 367, 645–648 (1994)

    CAS  ADS  Article  Google Scholar 

  10. 10

    Hope, K. J., Jin, L. & Dick, J. E. Acute myeloid leukemia originates from a hierarchy of leukemic stem cell classes that differ in self-renewal capacity. Nat. Immunol. 5, 738–743 (2004)

    CAS  Article  Google Scholar 

  11. 11

    Ishikawa, F. et al. Chemotherapy-resistant human AML stem cells home to and engraft within the bone-marrow endosteal region. Nat. Biotechnol. 25, 1315–1321 (2007)

    CAS  Article  Google Scholar 

  12. 12

    Costello, R. T. et al. Human acute myeloid leukemia CD34+/CD38 progenitor cells have decreased sensitivity to chemotherapy and Fas-induced apoptosis, reduced immunogenicity, and impaired dendritic cell transformation capacities. Cancer Res. 60, 4403–4411 (2000)

    CAS  PubMed  PubMed Central  Google Scholar 

  13. 13

    Terpstra, W. et al. Fluorouracil selectively spares acute myeloid leukemia cells with long-term growth abilities in immunodeficient mice and in culture. Blood 88, 1944–1950 (1996)

    CAS  Article  Google Scholar 

  14. 14

    Eppert, K. et al. Stem cell gene expression programs influence clinical outcome in human leukemia. Nat. Med. 17, 1086–1093 (2011)

    CAS  Article  Google Scholar 

  15. 15

    Ng, S. W. et al. A 17-gene stemness score for rapid determination of risk in acute leukaemia. Nature 540, 433–437 (2016)

    CAS  ADS  Article  Google Scholar 

  16. 16

    Shlush, L. I. et al. Identification of pre-leukaemic haematopoietic stem cells in acute leukaemia. Nature 506, 328–333 (2014)

    CAS  ADS  Article  Google Scholar 

  17. 17

    Wunderlich, M. et al. AML xenograft efficiency is significantly improved in NOD/SCID-IL2RG mice constitutively expressing human SCF, GM-CSF and IL-3. Leukemia 24, 1785–1788 (2010)

    CAS  Article  Google Scholar 

  18. 18

    Corces-Zimmerman, M. R., Hong, W. J., Weissman, I. L., Medeiros, B. C. & Majeti, R. Preleukemic mutations in human acute myeloid leukemia affect epigenetic regulators and persist in remission. Proc. Natl Acad. Sci. USA 111, 2548–2553 (2014)

    CAS  ADS  Article  Google Scholar 

  19. 19

    The Cancer Genome Atlas Research Network. Genomic and epigenomic landscapes of adult de novo acute myeloid leukemia. N. Engl. J. Med. 368, 2059–2074 (2013)

  20. 20

    Andersson, A. K. et al. The landscape of somatic mutations in infant MLL-rearranged acute lymphoblastic leukemias. Nat. Genet. 47, 330–337 (2015)

    CAS  Article  Google Scholar 

  21. 21

    Klco, J. M. et al. Functional heterogeneity of genetically defined subclones in acute myeloid leukemia. Cancer Cell 25, 379–392 (2014)

    CAS  Article  Google Scholar 

  22. 22

    Goardon, N. et al. Coexistence of LMPP-like and GMP-like leukemia stem cells in acute myeloid leukemia. Cancer Cell 19, 138–152 (2011)

    CAS  Article  Google Scholar 

  23. 23

    Fong, C. Y. et al. BET inhibitor resistance emerges from leukaemia stem cells. Nature 525, 538–542 (2015)

    CAS  ADS  Article  Google Scholar 

  24. 24

    Subramanian, A. et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc. Natl Acad. Sci. USA 102, 15545–15550 (2005)

    CAS  ADS  Article  Google Scholar 

  25. 25

    Qiao, W. et al. PERT: a method for expression deconvolution of human blood samples from varied microenvironmental and developmental conditions. PLoS Comput. Biol. 8, e1002838 (2012)

    CAS  Article  Google Scholar 

  26. 26

    Kreso, A. & Dick, J. E. Evolution of the cancer stem cell model. Cell Stem Cell 14, 275–291 (2014)

    CAS  Article  Google Scholar 

  27. 27

    Notta, F., Doulatov, S. & Dick, J. E. Engraftment of human hematopoietic stem cells is more efficient in female NOD/SCID/IL-2Rgc-null recipients. Blood 115, 3704–3707 (2010)

    CAS  Article  Google Scholar 

  28. 28

    Mazurier, F., Doedens, M., Gan, O. I. & Dick, J. E. Rapid myeloerythroid repopulation after intrafemoral transplantation of NOD-SCID mice reveals a new class of human stem cells. Nat. Med. 9, 959–963 (2003)

    CAS  Article  Google Scholar 

  29. 29

    Hu, Y. & Smyth, G. K. ELDA: extreme limiting dilution analysis for comparing depleted and enriched populations in stem cell and other assays. J. Immunol. Methods 347, 70–78 (2009)

    CAS  Article  Google Scholar 

  30. 30

    Bandelt, H. J., Forster, P. & Röhl, A. Median-joining networks for inferring intraspecific phylogenies. Mol. Biol. Evol. 16, 37–48 (1999)

    CAS  Article  Google Scholar 

  31. 31

    Novershtern, N. et al. Densely interconnected transcriptional circuits control cell states in human hematopoiesis. Cell 144, 296–309 (2011)

    CAS  Article  Google Scholar 

  32. 32

    Gamazon, E. R. et al. Comprehensive genetic analysis of cytarabine sensitivity in a cell-based model identifies polymorphisms associated with outcome in AML patients. Blood 121, 4366–4376 (2013)

    CAS  Article  Google Scholar 

Download references

Acknowledgements

This work was supported by grants from the Ontario Institute for Cancer Research with funds from the province of Ontario, the Cancer Stem Cell Consortium with funding from the Government of Canada through Genome Canada and the Ontario Genomics Institute (OGI-047), and the Canadian Institutes of Health Research (CSC-105367), the Canadian Cancer Society, the Terry Fox Foundation, a Canada Research Chair to J.E.D. L.I.S. was funded by the Benjamin Pearl fellowship from the McEwen Centre for Regenerative Medicine and an American Society of Hematology Scholar Award. This research was funded in part by the Ontario Ministry of Health and Long Term Care, whose views are not expressed here. We thank all the members of the Dick laboratory for comments and P. Zandstra for supporting this work.

Author information

Affiliations

Authors

Contributions

L.I.S. and A.M. developed the study and workflow, and performed all the xenografting, cell purification, and genotyping analysis. M.D.M. provided clinical annotated samples from the Princess Margaret Cancer Centre AML biobank. L.H., J.D.M., and T.J.H. performed sequencing. A.T.-G., J.J.F.M., A.R.-B., I.J.-Z., and R.M. performed genotyping. J.L.M. and M.D. supported xenografting, S.A. helped with identification of genetic variants and PDVs. S.W.K.N., V.V., C.X., and G.B. provided gene expression, and statistical, bioinformatic, and perturbation (PERT) analysis. L.I.S. and A.M. wrote the paper. L.I.S., A.M., J.C.Y.W., M.D.M., and J.E.D. revised the paper. J.E.D. provided funding and study supervision.

Corresponding author

Correspondence to John E. Dick.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Additional information

Reviewer Information Nature thanks M. Greaves, R. Levine and the other anonymous reviewer(s) for their contribution to the peer review of this work.

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data figures and tables

Extended Data Figure 1 Experimental design and estimated order of mutation acquisition for patients in PM AML cohort.

a, Experimental design. Patient samples obtained at diagnosis and relapse were sorted into multiple mature and progenitor populations by flow cytometry. In parallel, samples were injected into NSG or NSG-SGM3 mice, and human myeloid and T cells were sorted from the xenografts. WGS was performed on blasts and T cells, and ddPCR probes were designed for selected variants. These were then analysed by ddPCR in all the cell populations sorted from patient samples and xenografts for calculation of VAF. b, Estimated order of mutation acquisition for patients in PM AML cohort. Estimated order of acquisition of pre-leukaemic (grey) and leukaemic (black) PDVs, based on median joining network analysis.

Extended Data Figure 2 Analysis of relapse origin and clonal relationships for patients 1 and 6.

a, VAFs of PDVs as detected by ddPCR in various cell populations sorted directly from the diagnosis (Dx), relapse (Rel), or remission (Rem) samples of patient 1. No relapse variants were detected by ddPCR in this patient. White squares indicate populations in which no variants were detected. Phylogenetic tree showing clonal relationships for patient 1, based on analysis of VAFs of all pre-leukaemic and leukaemic mutations. The genetic distance between any two respective symbols on the phylogenetic tree was estimated using a neighbour joining method based on the degree of genetic relatedness (Nei’s genetic distance). Each coloured box contains highly related cell populations and xenografts inferred to represent a clone (green, pre-leukaemic; blue, diagnostic; red, relapse; blended blue and red, equal representation of diagnostic and relapse populations). b, VAFs of relapse variants as detected by ddPCR in various cell populations sorted directly from the diagnostic or relapse samples of patient 6, or in CD33+ myeloid cells isolated from xenografts generated from the diagnostic sample of the same patient. Abbreviations and display criteria are the same as in a. Phylogenetic tree showing clonal relationships for patient 6. The display criteria are the same as for a.

Extended Data Figure 3 Analysis of relapse origin and clonal relationships for patient 9.

VAFs of relapse variants as detected by ddPCR in various cell populations sorted directly from the diagnostic, remission, or relapse samples of patient 9, or in CD33+ myeloid cells isolated from xenografts generated from the diagnostic sample of the same patient. Abbreviations and display criteria are the same as in Extended Data Fig. 2a. VAFs of the only PDV detected by ddPCR in diagnostic blasts and in various cell populations sorted directly from the remission sample of patient 9. Phylogenetic tree showing clonal relationships for patient 9. The display criteria are the same as for Extended Data Fig. 2a.

Extended Data Figure 4 Analysis of relapse origin and clonal relationships for patient 2.

VAFs of relapse variants as detected by ddPCR in various cell populations sorted directly from the diagnostic, remission, or relapse samples of patient 2, or in CD33+ myeloid cells isolated from xenografts generated from the diagnostic sample of the same patient. Abbreviations and display criteria as in Extended Data Fig. 2a. VAFs of PDVs as detected by ddPCR in diagnostic blasts and in various cell populations sorted directly from the remission sample of patient 2. Phylogenetic tree showing clonal relationships for patient 2, based on analysis of VAFs of all relapse variants, pre-leukaemic, and leukaemic mutations. The display criteria are the same as for Extended Data Fig. 2a.

Extended Data Figure 5 Analysis of relapse origin and clonal relationships for patient 8.

VAFs of relapse variants as detected by ddPCR in various cell populations sorted directly from the diagnostic, remission, or relapse samples of patient 8, or in CD19+ B or CD33+ myeloid cells isolated from xenografts (DxGFM) generated from the diagnostic sample of the same patient. Abbreviations and display criteria as in Extended Data Fig. 2a. VAFs of PDVs as detected by ddPCR in diagnostic blasts and in various cell populations sorted directly from the remission sample of patient 8. Phylogenetic tree showing clonal relationships for patient 8. The display criteria are the same as for Extended Data Fig. 2a.

Extended Data Figure 6 Analysis of relapse origin and clonal relationships for patient 10.

VAFs of relapse variants as detected by ddPCR in various cell populations sorted directly from the diagnostic, remission, or relapse samples of patient 10, or in CD33+ myeloid cells isolated from xenografts generated from the diagnostic sample of the same patient. Abbreviations and display criteria are the same as in Extended Data Fig. 2a. VAFs of PDVs as detected by ddPCR in diagnostic blasts and in various cell populations sorted directly from the remission sample of patient 10. Phylogenetic tree showing clonal relationships for patient 10. The display criteria are the same as for Extended Data Fig. 2a.

Extended Data Figure 7 Analysis of relapse origin and clonal relationships for patients 4 and 11.

a, VAFs of relapse variants as detected by ddPCR in various cell populations sorted directly from the diagnostic, remission, or relapse samples of patient 4, or in CD33+ myeloid cells isolated from xenografts (DxM) generated from the diagnostic sample of the same patient. Abbreviations and display criteria are the same as in Extended Data Fig. 2a. Phylogenetic tree showing clonal relationships for patient 4. The display criteria are the same as for Extended Data Fig. 2a. b, VAFs of relapse variants as detected by ddPCR in various cell populations sorted directly from the diagnostic or relapse samples of patient 11. Abbreviations and display criteria are the same as in Extended Data Fig. 2a. Phylogenetic tree showing clonal relationships for patient 11. The display criteria are the same as for Extended Data Fig. 2a.

Extended Data Figure 8 Analysis of relapse origin and clonal relationships for patient 12.

VAFs of relapse variants as detected by ddPCR in various cell populations sorted directly from the diagnostic, remission, or relapse samples of patient 12, or in CD19+ B or CD33+ myeloid cells isolated from xenografts generated from the diagnostic sample of the same patient. Abbreviations and display criteria are the same as in Extended Data Fig. 2a. VAFs of the PDVs as detected by ddPCR in diagnostic blasts and in various cell populations sorted directly from the remission sample of patient 12. Phylogenetic tree showing clonal relationships for patient 12. The display criteria are the same as for Extended Data Fig. 2a.

Extended Data Figure 9 ROP and ROC correlate with TCGA gene expression clusters and demonstrate distinct molecular and clinical properties.

a, Gene set enrichment analysis of ROP and ROC clusters in the Princess Margaret Cancer Centre AML and TCGA relapsed AML cohorts. Diagnosis genes were ranked from top upregulated to top downregulated (ROC versus ROP cluster), and this rank list was compared using gene set enrichment analysis with the TCGA top 150 upregulated and top 150 downregulated genes (TCGA ROC-like versus TCGA ROP-like cluster). ROC genes correlate with TCGA ROC-like genes and ROP genes correlate with TCGA ROP-like genes. b, Comparison of genetic and clinical parameters between the TCGA ROC-like and TCGA ROP-like clusters identified distinct genomic and clinical properties associated with relapse origin.

Extended Data Figure 10 AML relapse is associated with increased stemness transcriptional programs and therapy resistance.

a, b, Proportion of bulk leukaemia populations that possess gene expression profiles of undifferentiated (a) or mature myeloid (b) cells, as determined by the perturbation (PERT) deconvolution analysis using gene expression data from known normal haematopoietic cell subsets. c, Leukaemia-initiating cell (LIC) frequency as determined by limiting dilution analysis. d, Gene-expression-based therapy resistance score. For all panels, data are shown for diagnostic (blue) and relapse (red) samples of patients in the PM cohort. Patients for whom the relapse origin (ROP or ROC) was functionally established are boxed.

Supplementary information

Supplementary Table 1

Clinical characteristics of the PM AML cohort. (XLSX 11 kb)

Supplementary Table 2

Immunophenotype of the diagnostic patient samples of the PM AML cohort. (XLSX 13 kb)

Supplementary Table 3

Protein damaging variants (PDVs) in the PM AML cohort. The reference allele (Ref) and alternative allele (Alt) were determined. Variant allele frequency (VAF) from whole genome sequencing of bulk cells from diagnosis (Dx), relapse (Rel) and Dx T cells (Tc) are provided. If the mutation appeared in a gene that was recurrently mutated in AML (more than 1% mutations in the TCGA AML dataset) the mutation is marked as recurrent (Y). If the variant was described in dbSNP, the major allele frequency (MAF) of the SNP from SNP142 is noted. (XLSX 47 kb)

Supplementary Table 4

Relapse variants (RVs) in the PM AML cohort. The reference allele (Ref) and alternative allele (Alt) were determined. Variant allele frequency (VAF) from whole genome sequencing of bulk cells from diagnosis (Dx), relapse (Rel) and Dx T cells (Tc) are provided. (XLSX 525 kb)

Supplementary Table 5

Somatic variants at Dx in the PM AML cohort. (XLSX 337 kb)

Supplementary Table 6

Frequency of nucleotide substitutions for RVs and somatic variants in the PM AML cohort. (XLSX 9 kb)

Supplementary Table 7

LSC frequency determined by limiting dilution assays in xenografts. The inverse of LSC frequency as well as upper and lower estimates are provided. Samples that generated multilineage grafts composed mainly of B cells with a minor CD33+ myeloid population are shaded. (XLSX 10 kb)

Supplementary Table 8

Protein damaging variants (PDVs) and relapse variants (RVs) validated by ddPCR in the PM AML cohort. For each patient, a matrix of sorted cell populations (rows) and variants (columns) is provided. VAFs were calculated from the ddPCR reaction. For patients #2, 8, and 15, NSG-SGM3 xenograft data is also provided. For patients # 8 and 10, diagnosis variants (present in diagnosis blasts and not in relapse blasts) are also provided. The presence of FLT3-ITD was determined by PCR, and the VAF recorded as 50% if positive and 0 if negative. (XLSX 87 kb)

Supplementary Table 9

Differential gene expression between 11 diagnosis and relapse AML samples by RNA sequencing. First worksheet (RC_Dx_Rel), read counts (RC) of RNA sequencing data. Second worksheet (RC_Diff_Exp) differential gene expression by RC. Third worksheet (RPKM_Diff_Exp), RNA sequencing data by RPKM and differential gene expression. Fourth and fifth worksheets, list of genes significantly differentially expressed between the ROP and ROC clusters of the PM AML cohort. Forth (ROc), top 40 genes that are more highly expressed in the ROc cluster. Fifth (ROp) 40 genes that are more highly expressed in the ROp cluster. (XLSX 10972 kb)

Supplementary Table 10

Summary table of results from the 11 AML samples including relapse origin, clinical, genetic and engraftment data. (XLSX 10 kb)

PowerPoint slides

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Shlush, L., Mitchell, A., Heisler, L. et al. Tracing the origins of relapse in acute myeloid leukaemia to stem cells. Nature 547, 104–108 (2017). https://doi.org/10.1038/nature22993

Download citation

Further reading

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Search

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing