SC I ENCE TRANS LAT IONAL MED I C I N E | R E S EARCH ART I C L E

CANCER

1Laboratory for Human Disease Models, RIKEN Center for Integrative MedicalSciences, Yokohama, Kanagawa 230-0045, Japan. 2Laboratory for IntegrativeGenomics, RIKEN Center for Integrative Medical Sciences, Yokohama, Kanagawa230-0045, Japan. 3Department of Medical Oncology, Dana-Farber Cancer Institute,Boston, MA 02215, USA. 4Department of Hematology, Toranomon Hospital, Tokyo105-8470, Japan. 5RIKEN Program for Drug Discovery and Medical TechnologyPlatforms, Yokohama, Kanagawa 230-0045, Japan. 6The Jackson Laboratory, BarHarbor, ME 04609, USA. 7Kazusa DNA Research Institute, Kisarazu, Chiba 292-0818, Japan.*Corresponding author. Email: fumihiko.ishikawa@riken.jp

Saito et al., Sci. Transl. Med. 9, eaao1214 (2017) 25 October 2017

Copyright © 2017

The Authors, some

rights reserved;

exclusive licensee

American Association

for the Advancement

of Science. No claim

to original U.S.

Government Works

Dow

nloaded from

Overcoming mutational complexity in acute myeloidleukemia by inhibition of critical pathwaysYoriko Saito,1 Yoshiki Mochizuki,2 Ikuko Ogahara,1 Takashi Watanabe,2 Leah Hogdal,3

Shinsuke Takagi,4 Kaori Sato,1 Akiko Kaneko,1 Hiroshi Kajita,1 Naoyuki Uchida,4

Takehiro Fukami,5 Leonard D. Shultz,6 Shuichi Taniguchi,4 Osamu Ohara,2,7

Anthony G. Letai,3 Fumihiko Ishikawa1*

Numerous variant alleles are associated with human acute myeloid leukemia (AML). However, the same variantsare also found in individuals with no hematological disease, making their functional relevance obscure. ThroughNOD.Cg-PrkdcscidIl2rgtmlWjl/Sz (NSG) xenotransplantation, we functionally identified preleukemic and leukemicstem cell populations present in FMS-like tyrosine kinase 3 internal tandem duplication–positive (FLT3-ITD)+

AML patient samples. By single-cell DNA sequencing, we identified clonal structures and linked mutations within vivo fates, distinguishing mutations permissive of nonmalignant multilineage hematopoiesis from leukemogenicmutations. Although multiple somatic mutations coexisted at the single-cell level, inhibition of the mutation stronglyassociatedwith preleukemic to leukemic stem cell transition eliminated AML in vivo. Moreover, concurrent inhibitionof BCL-2 (B cell lymphoma 2) uncovered a critical dependence of resistant AML cells on antiapoptotic pathways.Co-inhibition of pathways critical for oncogenesis and survival may be an effective strategy that overcomes geneticdiversity in human malignancies. This approach incorporating single-cell genomics with the NSG patient-derivedxenograft model may serve as a broadly applicable resource for precision target identification and drug discovery.

http:

by guest on May 11, 2021

//stm.sciencem

ag.org/

INTRODUCTIONAcute myeloid leukemia (AML) is a biologically and clinically hetero-geneous entity. Recent studies using deep DNA and RNA sequencinghave demonstrated intrapatient heterogeneity (1–4).Mutations in genes,such as NPM1, TET2, WT1, IDH1/2, and DNMT3A, are commonlyfound in AML (5–8), and next-generation DNA sequencing (NGS) sug-gests that certain mutations occur earlier than others based on variantallele frequencies (9–11). Preleukemic stem cells carrying these early so-matic mutations may contribute to leukemogenesis and disease relapse(9, 10). On the other hand, large-scale population-based sequencingstudies have revealed that hematopoietic cells in 5 to 18.4% of elderlysubjects with nonmalignant conditions, such as diabetes mellitus andcardiovascular diseases, harbored somatic mutations in genes includingASXL1, DNMT3A, and TET2 (12–14). In these subjects, the mutationswere associated with a 0.5 to 1% annual rate of progression to hemato-logicalmalignancies. This raised some fundamental questions regardingleukemogenesis and treatment strategies. First, which among AML-associated recurrentmutations contribute to leukemogenesis? Second,which of the many mutations and pathways must be targeted forgreatest clinical efficacy? To address these questions, we examinedmutational profiles of phenotypically and functionally defined humanAML cell populations to link mutations with in vivo fates. We thenused a functional single-cell genomic approach to identify critical tar-gets, allowing in vivo elimination of human AML cells with multiplecoexisting mutations.

RESULTSIn vivo fates of human AML cells are linked with distinctmutational profiles through NSG xenotransplantationWe obtained bone marrow (BM) or peripheral blood (PB) samplesfrom 27 patients with FMS-like tyrosine kinase 3 internal tandemduplication–positive (FLT3-ITD+) AML (table S1).Most of the patientshad poor prognostic factors, such as complex chromosomal abnormal-ities in addition to FLT3-ITDmutation, and/or had a known aggressivedisease (for example, primary resistance or relapse after multiple stemcell transplantations). Because AML-associated hematopoiesis consistsof both normal and malignant cells, we profiled patterns of recurrentmutations in patient-derived cell populations purified according to cellsurface phenotype that defines hematopoietic stem cells (HSCs), mul-tipotent progenitor cells, multilymphoid progenitors, and maturelymphoid and myeloid cells (15). To link these mutations with in vivofates, we transplanted the cell populations in newborn NSG mouse re-cipients (Fig. 1). If human lymphoid andmyeloid subsets were engraftedin NSG recipients (multilineage human hematopoietic repopulation),then the transplanted subpopulation contained normal HSCs and/orpreleukemic stem cells. If NSG recipients developed leukemia withuncontrolled proliferation of myeloid blasts and without lymphoiddifferentiation, then the transplanted subpopulation contained leukemia-initiating cells (LICs). As expected, frequencies of hematopoietic sub-populations and their population-levelmutational profiles varied amongpatients, and frequencies ofmutated alleles varied among subpopulationswithin individual patients (representative data from six patients in Fig. 2and fig. S1; sequence information in table S2). Upon transplantation,subpopulations with similar surface phenotypes isolated from differentpatients showed distinct behaviors in vivo. For instance, the patient 21–derivedCD34+CD38−CD90−CD45RA− cell population initiatedAML inNSGmice and therefore containedLICs (Fig. 2A). In contrast, in patients20, 23, and 24, theCD34+CD38−CD90−CD45RA− cell population recon-stituted multilineage human hematopoiesis in NSG mice and thereforecontained multilineage-engrafting HSCs or preleukemic stem cells,

1 of 12

SC I ENCE TRANS LAT IONAL MED I C I N E | R E S EARCH ART I C L E

by guest on May 11, 2021

http://stm.sciencem

ag.org/D

ownloaded from

whereas LICs were present in the CD34+CD38−CD90−CD45RA+ cellpopulation (Fig. 2B and fig. S1, AandB). In patient 13, theCD34+CD38−

population reconstituted multilineage human hematopoiesis, and theCD34−CD33+ population contained LICs, whereas patient 1–derivedCD34+CD38− cells initiated AML in vivo (Fig. 2C and fig. S1C). Theseobservations are consistent with recent reports showing variable cellsurface phenotype of LICs (16).Wenext examinedmutational profilesin these subpopulations with defined in vivo fates. In patient 20, thesame set of mutations (FLT3-ITD,DNMT3A, andWT1) was present inCD34+CD38−CD90−CD45RA− and CD34+CD38−CD90−CD45RA+

subpopulations, but these subpopulations showed distinct in vivo fates(Fig. 2B). This functional difference may be due to uneven distributionof mutations identified in bulk AML cell populations in single-cellclones, resulting in disparate combinations of mutations and divergentin vivo fates. Therefore, we performed single-cell DNA sequencing offunctionally defined preleukemic stem cell– and LIC-containing sub-populations alongwith humanmultilineage hematopoietic cells and leu-kemia cells they generate in vivo to define clonal structures and identifymutation(s) associated with leukemia-initiating versus multilineage-engrafting function.

Functional genomic approach combining patient-derivedxenograft model and single-cell DNA sequencingdistinguishes leukemogenic from permissive mutationsWe examined DNMT3A, TET2, NPM1, and WT1 mutations andFLT3-ITD among single cells isolated from multilineage-engraftingcell– and LIC-containing patient-derived populations and their in vivoprogeny (Fig. 3A). Patient-derived multilineage-engrafting pre-leukemic stem cells showed mutational heterogeneity at the single-celllevel, carrying combinations of multiple mutations (Fig. 3, B and C).DNMT3A mutation was identified both in multilineage-engraftingCD34+CD38−CD90−CD45RA− preleukemic cells (patients 20 and 24)

Saito et al., Sci. Transl. Med. 9, eaao1214 (2017) 25 October 2017

and AML-initiating CD34+CD38−CD90−CD45RA− cells (patient 21) atthe single-cell level (Fig. 3B). In vivo–generated single B cells from patient20– and patient 24–derived preleukemic stem cells harbored DNMT3Amutation (andWT1mutation in the case of patient 20), whereas therewere no FLT3-mutated B cell clones. There were no FLT3-ITD muta-tions in patient 24–derived preleukemic stem cells and in vivo–generatedsingle B cells. Although there were FLT3-ITD–mutated single cells in thepatient 20–derived preleukemic stem cell population, those FLT3-ITD–mutated subclones did not contribute to normal lymphopoiesis. Thesefindings indicate thatDNMT3A andWT1mutations are permissive andcan coexist in a single cell without hindering human multilineage differ-entiation. In contrast, FLT3-ITD was identified in substantial proportionsof patient 21–, patient 20–, and patient 24–derived LICs and engraftedAML cells at the single-cell level. Likewise, CD34+CD38− cells derivedfrom patients 1 and 13 exhibited distinct in vivo fates: At the single-cell level, the latter carried wild-type (WT) FLT3, and the formerharbored FLT3-ITD (Fig. 3C). Note that, although FLT3-ITD+ singlecells were a minority among the LIC population in patient 13, everyengrafted AML cell harbored FLT3-ITD mutation. In addition, FLT3-ITD+ single cells were enriched among LIC-containing CD34−CD33+

population at the time of relapse in patient 13. These findings suggestthat acquisition of the FLT3-ITD mutation acts as a critical trigger forleukemia initiation, working in cooperation with accumulated muta-tions in DNMT3A, TET2, NPM1, and/or WT1.

Kinase inhibition effectively targets human AML withmutational diversityTherapeutic efficacy of targeting such a mutation among multiple co-existing mutations is an important question for clinical translation.We addressed this by in vivo inhibition of the FLT3 pathway in anNSG patient-derived xenograft (PDX) model. To serve as a realisticplatform for in vivo therapeutic testing, a PDX model must reflect the

A

B

Fig. 1. In vivo fates of patient-derived AML cells defined by mutational profile. (A) Somatic mutation profiles were identified in patient cell subpopulationsdefined by surface phenotype based on developmental hierarchy of human hematopoiesis. (B) The in vivo fate of each subpopulation was determined throughtransplantation into newborn NSG mice. If repopulation by multilineage hematopoiesis occurred, then the transplanted subpopulation contained hematopoietic orpreleukemic stem cells; if AML engraftment occurred, then the transplanted subpopulation contained LICs.

2 of 12

SC I ENCE TRANS LAT IONAL MED I C I N E | R E S EARCH ART I C L E

by guest on May 11, 2021

http://stm.sciencem

ag.org/D

ownloaded from

-

A

B

C

Fig. 2. Mutational profiles and in vivo engraftment of patient-derived subpopulations. Three additional patients are shown in fig. S1. In each patient sample, humanCD45+CD3+CD19− T cells and human CD45+CD3−CD19+ B cells were identified. Within CD3−CD19− non-T, non-B cells, subpopulations were identified on the basis of CD34,CD38, CD90, and CD45RA surface expression. These populations underwent polymerase chain reaction (PCR) for FLT3-ITDmutation and DNA sequencing (DNA-seq) for the othergenes indicated. Variant allele frequencies are shownas heatmaps. In patients 21 (A) and 20 (B), the CD34+CD38−CD90−CD45RA− andCD34+CD38−CD90−CD45RA+ subpopulationswere identified. The in vivo fates of CD34+CD38−CD90−CD45RA− subpopulations differed between patients 20 and 21, showing engraftment with multilineage human hema-topoiesis in patient 20 (indicated by green rectangles) but initiation of AML in patient 21 (indicated by red rectangles). In patient 13 (C), the CD34+CD38− subpopulation showedmultilineage repopulation, whereas the CD34−CD33+ subpopulation with additional FLT3-ITD and NPM1 mutations initiated AML. AML-engrafted recipients showed no B cellengraftment (indicated by gray dashed outlines on flow cytometry plots). Detailed information on variants found in each patient is shown in table S2.

Saito et al., Sci. Transl. Med. 9, eaao1214 (2017) 25 October 2017 3 of 12

SC I ENCE TRANS LAT IONAL MED I C I N E | R E S EARCH ART I C L E

by guest on May 11, 2021

http://stm.sciencem

ag.org/D

ownloaded from

In vivo– In vivo–A

B

C

Fig. 3. Mutations contributing to distinct in vivo cell fates identified by single-cell functional genomics. (A) Patient-derived subpopulations with defined in vivofates and their in vivo progeny underwent single-cell mutation profiling. Through this strategy, mutations present in patient-derived preleukemic stem cells (pre-LSCs)and LIC clones were tracked and linked to in vivo fates. (B and C) Using samples from five patients, patient-derived multilineage-engrafting and LIC-containing pop-ulation and engrafted B cells and AML cells were subjected to single-cell DNA sequencing for variants detected in each indicated gene in each patient. FLT3-ITDsequences with highly variable repeated sequence patterns were detected by single-cell PCR. In (B) and (C), each column of rectangles represents an individual cell.The presence or absence of mutations in each gene is shown by colors of rectangles, as indicated.

Saito et al., Sci. Transl. Med. 9, eaao1214 (2017) 25 October 2017 4 of 12

SC I ENCE TRANS LAT IONAL MED I C I N E | R E S EARCH ART I C L E

mutational diversity of human AML cells. At the single-cell level, wedetected AML cells with various patterns of mutations in engrafted re-cipients. In addition, engrafted AML cells retained the mutations pres-ent in patient-derived LIC-containing cells from 12 patients examined,indicating that engrafted AML cells reflect mutational diversity ofpatient-derived AML cells (Fig. 4). Therefore, we went on to examinethe effect of FLT3 pathway inhibition in human AML cells with mul-tiple coexistingmutations by using RK-20449, a pyrrolo-pyrimidine de-rivative inhibitor of Src family kinase HCK and FLT3 that we hadpreviously identified (17). By treating 56 NSGmice that were engraftedwith FLT3-ITD+AML from 19 patients, we found significant responsesto the single agent RK-20449 in vivo in all 19 cases (data and associatedP values are shown in table S3). For five patients, RK-20449 completelyeliminated AML cells in the BM, spleen, and PB of all recipient micetreated, despite the presence of mutations not directly targeted by

Saito et al., Sci. Transl. Med. 9, eaao1214 (2017) 25 October 2017

RK-20449 (patient 1: DNMT3A, NPM1, and TET2; patient 2: IDH1;patient 16: FLT3D835Hpointmutation andNRAS) (Fig. 5A and tableS2). For 11 additional patients, RK-20449 treatment alone resulted incomplete responses in the spleen inmost of the recipientmice tested (Fig.5B). Although statistically significant (P < 0.05) treatment effects wereobserved in groupwise comparisons, residual RK-20449–resistantAML cells were present in the BM of at least one treated mouse for14 of 19 patients (Fig. 5, B and C).

BCL-2 inhibition enhances mitochondrial priming toapoptosis induced by kinase inhibitionResistance to RK-20449 may be mediated through pre-existing ornewly acquired somatic mutations (18). However, enrichment or newacquisition of AML-associated somatic mutations was not identified infive AML cases after in vivo RK-20449 treatment (fig. S2). In addition,

by guest on May 11, 2021

http://stm.sciencem

ag.org/D

ownloaded from

theWT FLT3 genewas not identified in re-sistant cells, indicating that emergence ofFLT3WT cells is not a substantial mecha-nism for resistance. This is consistent withmutational profiles of patient samples seri-ally obtained at primarypresentation and atrelapse (fig. S3). In six cases examined, nei-ther emergence of FLT3WT cells nor sub-stantial increase in frequencies of somaticmutations was detected, with the exceptionof FLT3 D835H mutation–positive cells inpatient 16 emerging at the time of relapse.In vivo RK-20449 treatment resulted intranscriptional up-regulation of S100A8(associated with drug resistance in leuke-mia),HSPA5 (promotes cell survival underendoplasmic reticulum stress and sup-presses ferroptosis), and IFI6 (negativelyregulates apoptosis) (fig. S4) (19–23).There-fore, we functionally assessed dependenceonantiapoptoticmechanisms inRK-20449–resistant human AML cells by dynamicBCL-2 (B cell lymphoma2) homology do-main 3 (BH3) profiling (24). Some humanmalignancies are dependent on specificantiapoptotic proteins for survival and aretherefore sensitive to the small-moleculeantagonists of those proteins (25–29). Dy-namic BH3 profiling determines how“primed” cells are to apoptotic cell deathand how changing conditions (such as ex-posure to drugs) affect baseline primingby quantifyingmitochondrial cytochromec release in response to BH3-only peptidesthat activate proapoptotic effectors BAXand BAK. Despite patient-to-patient vari-ability, RK-20449 treatment lowered thehalf maximal inhibitory concentration(IC50) of BIM peptide for mitochondrialcytochrome c release, indicating enhancedproapoptotic signaling in FLT3-ITD+ hu-man AML cells (Fig. 5D). This was con-sistent with our finding that RK-20449alone completely eliminated AML cells

Fig. 4. Profiles of AML-associated mutations in nine genes in patient- and recipient-derived AML cells from12 AML cases. Variant allele frequencies of indicated genes are represented as heat maps. Patient, LIC-containingpopulation from the patient; 1′, human AML cells from primary recipients; 2′, human AML cells from secondary re-cipients. All patient-derived and recipient-derived leukemia populations were positive for FLT3-ITD by PCR. Non-ITD FLT3mutations were identified by sequencing. Information on variants is shown in table S2.

5 of 12

SC I ENCE TRANS LAT IONAL MED I C I N E | R E S EARCH ART I C L E

Saito et al., Sci. Transl. Med. 9, eaao1214 (2017) 25 October 2017 6 of 12

by guest on May 11, 2021

http://stm.sciencem

ag.org/D

ownloaded from

s

A

B

C

D E F

Fig. 5. Induction of apoptosis via en-hanced BCL-2 dependence in FLT3-ITD+

AML cells with diverse coexisting somaticmutations by in vivo kinase inhibition.Overall, treatment with RK-20449 resultedin significant reduction of AML cells in theBM, spleen, and PB of recipients (P < 1 ×10−19 for each; data are tabulated in tableS3). To document patient-to-patient varia-bility, RK-20449 responses were classifiedas follows: complete response (A), if all re-cipients treated showed residual BM humanCD45+ chimerism of <5%; good response(B), if the case did not meet the criterionfor complete responder but all recipientsshowed <50% residual BM human CD45+;partial response (C), if at least one of the re-cipients showed >50% residual BM humanCD45+. For each response group, PB timecourse of human AML chimerism (leftmostpanels) for RK-20449–treated recipientsand final BM (middle panels) and spleen(rightmost panels) human AML chimerismof RK-20449–treated and untreated recipi-ents are shown. Pretreatment PB humanAML cell chimerism is shown at week 0.The numbers of recipients for each patient/each treatment group and pre-/posttreatmentAML chimerism are shown in table S3. Inall response groups, BM, spleen, and PBchimerism was significantly reduced withRK-20449 treatment. For the BM and spleen,final chimerism for recipients in each treat-ment group was compared. For the PB,pre- and posttreatment chimerism for RK-20449–treated recipients was compared.P < 5.8 × 10−5 by two-tailed t test for allcomparisons. In each scatter graph, dottedlines are drawn at 0, 5, and 50%. (D) Dy-namics of apoptotic response to BIM pep-tide in the presence of RK-20449 wasmeasured for six AML cases. Bars repre-sent the BIM IC50 of cytochrome c lossin RK-20449–treated human CD45+ cellsas percentages of IC50 in cells treated withdimethyl sulfoxide (DMSO) alone. Enhance-ment of apoptotic response to BIM byRK-20449 showed substantial patient-to-patient variability. (E) Dynamics ofapoptotic response to BH3-only peptidesBAD, HRK, and NOXA in the presence ofRK-20449 or DMSO alone was measuredfor seven AML cases. Bars represent thepercentage reduction of IC50 in the pres-ence of RK-20449 compared with DMSOalone. RK-20449 enhanced apoptotic re-sponse to BAD and HRK peptides withsubstantial patient-to-patient variability,whereas apoptotic response to NOXA pep-tide was not substantially enhanced by RK-

20449. (F) Apoptotic response to a BCL-2 inhibitor ABT-199 was enhanced by RK-20449 in seven AML cases. Bars show increased cytochrome c loss in human AML celltreated with ABT-199 and RK-20449 compared with those treated with DMSO alone.

SC I ENCE TRANS LAT IONAL MED I C I N E | R E S EARCH ART I C L E

by guest on May 11, 2021

http://stm.sciencem

ag.org/D

ownloaded from

A

B

C

Fig. 6. Eradication of FLT3-ITD+ AML cells in vivo through combined inhibition of kinase and antiapoptotic pathways. (A and B) Mice engrafted with AML derivedfrom 12 patients associated with good or partial responses to RK-20449 alone received four different treatments (no treatment, ABT-199 alone, RK-20449 alone, and com-bination). The numbers of recipients for each patient/each treatment group and pre-/posttreatment AML chimerism are shown in table S4. (A) PB human CD45+ chimerism isshown over time. Recipients were phlebotomized weekly, and pretreatment PB human CD45+ AML chimerism is shown at time 0. Mean PB human CD45+ chimerism for eachpatient/each treatment group and statistics comparing the treatment groups are shown in table S5. (B) Final BM and spleen human AML chimerism is shown for miceengrafted with AML derived from 12 patients. Nine cases showed complete responses, and three cases showed good responses to combination treatment. Each circlerepresents an AML-engrafted recipient. Mean BM/spleen human CD45+ chimerism for each patient/each treatment group and statistics comparing the treatment groupsare shown in table S6. In each scatter graph, dotted lines are drawn at 0, 5, and 50%. (C) Residual human AML initiation capacity of human CD45+ cells after in vivo treatmentwas assessed by serial transplantation for four treatment groups. To compare the amount of residual LICs in recipients after in vivo treatment, each secondary recipient wastransplanted with human CD45+ cells sorted from 2.5% (by cell number) of viable BM cells remaining in treated recipients. The mean and SEM for human CD45+ AML cellchimerism in the BM of secondary recipients are shown. Each circle represents a secondary recipient.

Saito et al., Sci. Transl. Med. 9, eaao1214 (2017) 25 October 2017 7 of 12

SC I ENCE TRANS LAT IONAL MED I C I N E | R E S EARCH ART I C L E

by http://stm

.sciencemag.org/

Dow

nloaded from

with DNMT3A, TET2, IDH1, and/orWT1mutations. In addition, expo-sure to RK-20449 facilitated mitochondrial cytochrome c release in re-sponse to BAD and HRK peptides, indicating enhanced mitochondrialsensitivity toBCL-2 andBCL-XL inhibition (Fig. 5E). In addition, dynamicBH3profiling showed that a selectiveBCL-2 inhibitor,ABT-199, enhancedRK-20449–induced apoptosis in FLT3-ITD+ human AML cells (Fig.5F). This suggested that co-inhibition of antiapoptotic signal, specifical-ly BCL-2, might result in a more robust eradication of human AMLwith diverse coexisting mutations.

Combined inhibition of kinase and antiapoptosispathways eliminates AML with genetic complexity andclonal diversityTo examine whether combining RK-20449 andABT-199 eliminates re-sistant AML cells, we chose cases that were not completely eradicated invivo by RK-20449 alone. Inmost of the cases, the BCL-2 inhibitor ABT-199 alone resulted in transient and limited responses. In contrast, in all12 cases, combination treatment significantly reducedhumanAMLchi-merism in the PB, BM, and spleen (Fig. 6, A and B, and tables S4 toS6; P < 0.05 in all comparisons). Combination treatment completelyeliminated human AML cells in vivo without targeting coexistingmutations in 9 of 12 cases (Fig. 6B; cases 3, 5 to 10, 13, and 14). Wecompared residual leukemia-initiating capacity by transplanting a pre-determined fraction of viable human CD45+ AML cells from treatedmice into secondary NSG recipients (Fig. 6C). We found that BMtreated with RK-20449 alone or ABT-199 alone contained enoughviable LICs to initiate AML in every secondary recipient transplanted,whereas AML cells remaining after combination treatment engrafted inonly 1 of 23 mice, indicating that combination treatment moreeffectively reduced the frequency of LICs in vivo (Fig. 6C). On the otherhand, combination treatment did not have significant effects on humanleucocytes, T cells, B cells, and myeloid cells in human cord blood(CB) HSC–engrafted NSG recipients (table S7).

guest on May 11, 2021

DISCUSSIONMultilineage-engrafting preleukemic HSCs carrying somatic mutationsare thought to be poised for leukemogenesis and may act as a reservoirfor leukemic progression and relapse (9, 10). To prevent such events,early-occurring somaticmutations or foundermutations with high var-iant allele frequencies have been considered as critical therapeutic tar-gets (30). However, some of these somatic mutations were found inindividuals with advanced age with no apparent hematological disease,associated with a relatively low rate of progression to hematologicalmalignancies (fig. S5, top left) (31). Here, we demonstrated that thecontribution of these somatic mutations to normal hematopoieticdifferentiation or to leukemogenesis could inform therapeutic targetselection in AML. By integrating population- and single-cell–levelgenomics and in vivo functional assessment in PDXs, we found thatrelatively late acquisition of FLT3-ITD on the background of permis-sive earlier-occurring mutations in DNMT3A, TET2, NPM1, andWT1, alone or in combination, triggered in vivo leukemogenesis(fig. S5, top right). Moreover, acquisition of FLT3-ITD triggered leu-kemogenesis along a broad spectrum of hematopoietic differentiation.Through single-cell sequencing, we found substantial mutational heter-ogeneity in patient-derived preleukemic cells and NSG-engrafted hu-man lymphoid cells, whereas patient-derived LICs were less mutationallyheterogeneous at the single-cell level, with substantial enrichment ofFLT3-ITD+ single cells. This is consistent with the hypothesis that late

Saito et al., Sci. Transl. Med. 9, eaao1214 (2017) 25 October 2017

acquisition of FLT3-ITDmutation results in selective clonal expansion.Single-cell RNA sequencing may help fully characterize clonalstructures of preleukemic and leukemic stem cells.

In a recent clinical study, midostaurin, a small-molecule inhibitor ofFLT3, improved both 5-year event-free and overall survivals in FLT3-ITD+ AML patients when combined with standard chemotherapy,making it the first U.S. Food and Drug Administration–approved tar-geted agent in AML (32–34). Our findings are consistent with this clin-ical finding: Targeting FLT3-ITD alone resulted in effective eliminationof AML in vivo despite coexisting earlier-occurring mutations (fig. S5,bottom left and right). Furthermore, dynamic BH3 profiling identifiedan additional target, BCL-2, in kinase inhibitor–resistantAMLcells. Co-inhibition of BCL-2 resulted in complete eradication of human AMLcells resistant to kinase inhibition. Note that this approach was effectivein cases with poor prognoses and highly aggressive clinical courses,multiple coexisting mutations, and/or complex chromosomal abnor-malities. With the advent of high-throughput sequencing technologiesand the genomic information gained from each patient, precisionmed-icine is becoming more and more feasible. By functionally connectinggenomic information with in vivo fate and behavior of patient-derivedcells at the level of single cells through a PDX model, we could identifytherapeutic targets with improved precision to promote more effectivedrug discovery in genetically complex human malignancies.

MATERIALS AND METHODSStudy designThe overall objective of this study was to explore strategies for eliminat-ing FLT3-ITD+ human AML by (i) tracing profiles of known somaticmutations associated with AML and other malignancies in single-cellclones of patient-derived hematopoietic cells to human hematopoieticcells engrafting in NSG xenotransplantation recipients to identify mu-tations associated with leukemia initiation, (ii) examining whether theantiapoptosis pathway can be a therapeutic target independent ofleukemogenic somatic mutations, and (iii) testing whether targetingthese pathways by small-molecule inhibitors would result in efficientelimination of human AML in vivo. To do so, we isolated various sub-populations of patient-derived hematopoietic cells by fluorescence-activated cell sorting (FACS) and performedNGS of bulk and single-cellgenomic DNA for known malignancy-associated somatic mutations inparallel with NSG xenotransplantation. By performingNGS of bulk andsingle-cell DNA in human hematopoietic cells that were engrafted inNSG recipients, we traced and identified somatic mutations affiliatedwith in vivo leukemia-initiating AML cell clones and preleukemic cellclones that reconstituted nonmalignant human hematopoiesis in vivo.To determine the degree to which human AML cells resistant to kinaseinhibition were dependent on the antiapoptotic pathway, we performeddynamic BH3 profiling. To test the efficacy of kinase inhibition and in-hibition of the antiapoptotic pathway, we treated human FLT3-ITD+

AML-engrafted NSG recipients with a dual kinase inhibitor, RK-20449, and a BCL-2 inhibitor, ABT-199, and assessed human AML cellchimerism by flow cytometry in the PB weekly during treatment and inthe BM, spleen, and PB at the time the animal was sacrificed.

We did not predetermine the numbers of mice in each treatmentgroup. To ensure that each comparison was sufficiently powered, weperformed power calculation for each comparison (either two-tailedt test or paired two-tailed t test) using StatMate (GraphPad). We foundthat the comparisons deemed statistically significant (P < 0.05) werepowered at 85 to 99% in detecting the differences observed using the

8 of 12

SC I ENCE TRANS LAT IONAL MED I C I N E | R E S EARCH ART I C L E

by guest on May 11, 2021

http://stm.sciencem

ag.org/D

ownloaded from

statistical test used. PB sampling was carried out before treatment andevery week after start of treatment. Recipients were checked daily andsacrificed when they showed signs of progressive disease such as ruffledfur andweakness. No randomization and blindingwere performed, andthere were no exclusions.

For in vivo treatment experiments, recipientswith similar extent ofPBhuman AML engraftment chimerism were chosen as a set for indicatedtreatments. Pretreatment humanCD45 chimerismand statistical data areshown in tables S3 to S5. There were no significant differences found inpretreatment human AML engraftment between treatment groups.

Because it is logistically difficult and undesirable from the standpointof animal health and comfort to obtain replicate samples from highlyimmunosuppressed NSG recipients, PB was sampled only once everyweek for all recipients in all experiments. Human AML chimerism ob-tained at the time of sacrifice was also evaluated once in the BM, spleen,and PB of the recipients. Overall experimental replication for in vivotreatment studies was ensured by the numbers of patient samples testedand the numbers of recipients treated for each group. For each patientsample, treatment experiments were performed as a set of untreatedcontrol, treatment A, treatment B, and/or treatment C, with A, B,and C being ABT-199 alone, RK-20449 alone, or combination, respec-tively. Therefore, there were at least three experimental replicates foreach patient sample.

Ethical statementsWritten informed consent was obtained from all patients. The studywas performed with authorization from the Institutional Review Boardfor Human Research at RIKEN and Toranomon Hospital, in accord-ance with the ethical standards of responsible committees on humanexperimentation at each institution. CB samples were purchased fromLonza. All experiments using NSG mice (35, 36) were performed withauthorization from and according to guidelines established by the Insti-tutional Animal Committees at RIKEN and the Jackson Laboratory.

MiceMicewere bred andmaintainedunder defined flora at the animal facilityofRIKENandat the JacksonLaboratory. Both female andmaleNSGmicereceived transplants at 2 days of age. Treatment studies were conductedwhen sufficient engraftment was observed at about 6 weeks of age.

Flow cytometryThe following monoclonal antibodies (mAbs) were used for flow cy-tometry: mouse anti-human CD19 (catalog nos. 562653, 555412, and341093), CD3 (catalog nos. 563800, 562426, and 555341), CD33 (cata-lognos. 562854 and 555450), CD34 (catalog no. 348791), CD38 (catalogno. 340439 and 555459), CD4 (catalog no. 555348), and CD45 (catalognos. 563204, 641399, and 563204); and rat anti-mouse Ter119 (catalog no.557915) and CD45 (catalog nos. 563891 and 563410) (BD Biosciences).Analyses were performedwith FACSAria III and FACSCanto II (BDBio-sciences). Toobtain cells for xenogeneic transplantation, BV786 (BrilliantViolet 786)–conjugated anti-CD3, BV605-conjugated anti-CD19, BV421-conjugated anti-CD33, PE-Cy7 (phycoerythrin–cyanin 7)–conjugatedanti-CD34, allophycocyanin-conjugated anti-CD38, fluoresceinisothiocyanate–conjugated anti-CD90, andPE-conjugated anti-CD45RAmAbs were used. For single-cell sorting, a 100-mm nozzle was used.

TransplantationNSGnewbornmice received 1.5-gray total body irradiation, followed byintravenous injection of purified human cells. For primary transplanta-

Saito et al., Sci. Transl. Med. 9, eaao1214 (2017) 25 October 2017

tion shown in Figs. 2 to 4, cells of the indicated phenotype were sortedfrom BM or PB cells obtained from the patients, and 5 × 102 to 2 × 105

cells were transplanted per recipient, depending on the frequency of thecell population. For in vivo treatment experiments, the knownLIC pop-ulation from each patient was transplanted to create human AML-engrafted recipients. Donor cells were purified according to their cellsurface phenotype using mAbs against human CD34, CD38, CD90,CD45RA, CD3, CD19, and CD33. The extent of engraftment of humancells in the NSG recipients was assessed by retro-orbital phlebotomyand flow cytometry.

Genome analysisDNAwas extracted from human cells purified from patient samples orrecipient organs using the DNeasy Blood and Tissue kit (Qiagen). PCRdetection of FLT3-ITD was performed using the TaKaRa PCR FLT3/ITD Mutation Detection set (Takara Bio).

The bulk DNA sequences were determined by NGS. After shearingwith a Covaris S220 (Covaris), the fragmented genomic DNA (10 ng)was converted to an NGS sequencing library with a KAPAHyper Prepkit (Kapa Biosystems) according to the protocol provided by the sup-plier. Targeted sequencing of AML-related genes was carried out by ahybridization capture method with xGen AML Cancer Panel v1.0(Integrated DNA Technologies) according to the protocol providedby the supplier. The hybridization-captured DNA library was subjectedto NGS in a paired-end read mode (200 cycles) with an Illumina HiSeq1500 (Illumina). The obtained DNA sequences were mapped onto hu-man genome sequence (hg19) using Burrows-Wheeler Aligner (BWA)(v0.7.12; http://bio-bwa.sourceforge.net/) and then realigned with aRealigner Target Creator in the Genome Analysis Toolkit (v1.6-13;https://software.broadinstitute.org/gatk/index.php). After treatmentwith Fix Mate Information in Picard (v1.119; http://broadinstitute.github.io/picard/) and Quality Score Covariate and Table Recalibrationin the Genome Analysis Toolkit (v1.6-13), variants were detected withVarScan (v2.3.6; read depth, >10; http://varscan.sourceforge.net/somatic-calling.html).

Single-cell variation analysis was carried out for single cells sorted ona BD FACSAria into 96-well plates. Single-cell whole-genomeamplification byMultiple Annealing and Looping-Based AmplificationCycles (MALBAC) (37), associated with a substantially low allelicdropout rate at ~1% and a false-positive rate of ~4 × 10−5, was used(37–39). After single-cell genome amplificationwith aMALBAC SingleCellWGAkit (YikonGenomics), target regions of genes of interestwerePCR-amplified with primers including well indexes by PCR. The first-round gene-specific PCRwas conducted using gene-specific primers by25-cycle PCRwithGflexDNAPolymerase (Takara Bio) in a single-plexmode. The second-round PCR was performed to link Illumina anchorsequences at both sides of the first-round PCR products. PCR con-ditions were as in the first-round PCR, except that PCR primers werereplaced with those for attaching Illumina anchoring sequences. Be-cause of low PCR efficiency, the first-round PCRs forWT1 and CEBPAwere carried out as follows: ForWT1, PCRwas performed over 40 cyclesusing Taq DNA polymerase (Qiagen) under a three-step thermalcycling of 94°C for 30 s, 55°C for 30 s, and 72°C for 30 s; for CEPBA,PCR was performed over 36 cycles using Taq DNA polymerase andQ-solution (Qiagen) under a two-step thermal cycling of 95°C for 1minand 68°C for 6 min. Primer sequences are described in table S8. ThePCR products were sequenced in a paired-end read mode (300 cycles)on an Illumina MiSeq. The obtained DNA sequences were mapped ontothe human genome sequence (hg19) with BWA (v0.7.12), and paired-end

9 of 12

SC I ENCE TRANS LAT IONAL MED I C I N E | R E S EARCH ART I C L E

by guest on May 11, 2021

http://stm.sciencem

ag.org/D

ownloaded from

readsweremergedwith SAMtools (v1.0; readdepth, >100; http://samtools.sourceforge.net/).Variant detection and frequency calculationwere carriedoutwithmpileup in SAMtools.Here, variation genotypewas assigned onan assumption that the sequencing error rate is lower than 2% and anapparent variation with allele frequency lower than 2% was regarded asa WT. Three independent amplification and sequencing rounds wereperformed before calling a locus WT. No germline tissue was availablefor evaluation of somatic status ofmutations.CEBPA,DNMT3A, FLT3,IDH1,TET2, andWT1 variants were included or excluded according togene-specific characteristics, as described by Lindsley et al. (40).

For RNA sequencing (RNA-seq) analysis, total RNA was extractedfromFACS-purified cells treatedwithTRIzol (ThermoFisher Scientific).RNA-seq libraries were prepared using the SureSelect Strand-SpecificRNA Library Preparation kit (Agilent Technologies) according to themanufacturer’s protocol andwere sequenced by aHiSeq 1500 (Illumina).The sequence reads were mapped to the human genome (hg19) usingTopHat2 software (v2.0.8). Cufflinks (v2.1.1) was run with the samereference annotation with TopHat2 to generate FPKM (fragments perkilobase of transcript per million mapped reads) values. Statistical eval-uation of gene expression change was performed using the edgeRalgorithm with read counts on exons determined using the R program.

In vivo treatmentIn vivo treatment experiments were performed with AML-engraftedNSG recipients using RK-20449 (17) and ABT-199 (41, 42). The recip-ients were treated with RK-20449 (30 mg/kg) intraperitoneally twice aday, ABT-199 (70 mg/kg) orally once a day, or both RK-20449 andABT-199. The mice were euthanized when they became moribund orafter 4 to 6 weeks of treatment, and human AML chimerism in the BM,spleen, and PB was determined using flow cytometry. In secondarytransplantation experiments, eachmouse received 7-aminoactinomycinD(−) viable humanCD45+ cells from 2.5% of total BM that remained inAML-engrafted recipients at the time of sacrifice to simulate relapseoccurring from residual viableAMLcells. All treated recipients and theirpre- and posttreatment engraftment data are tabulated in tables S3 andS4. No sample size pre-estimation was performed. To ensure that eachcomparison was sufficiently powered, we performed power calculationfor each comparison (either two-tailed t test or paired two-tailed t test)using StatMate (GraphPad). Each comparison deemed significant (P <0.05) was powered at 85 to 99% for detecting the differences observed.

Dynamic BH3 profilingCells were harvested from the BM of recipients engrafted with AMLderived from patients with FLT3-ITD+ AML, and BH3 profiling wasperformed using the plate-based assay previously described (43, 44).For dynamic BH3 profiling, 2 × 106 harvested cells were exposed to500 nM RK-20449 in 2 ml of hematopoietic growth medium supple-mented with stem cell factor (50 ng/ml), FLT3 ligand (50 ng/ml), andthrombopoietin (50 ng/ml) for 4 hours at 37°C in humidified atmo-sphere containing 5% CO2. After surface labeling with anti-humanCD45 and dead cell exclusion with Zombie NIR (BioLegend), the cellswere permeabilized and exposed to BH3 peptides (0.39 mMBIM, 80 mMBAD, 80 mMHRK, or 80 mMNOXA) or 1 mMABT-199, and retainedintracellular cytochrome c wasmeasured by flow cytometry using anti–cytochrome c antibody.

Statistical analysisFor in vivo treatment experiments, numerical data are presented asmeans ± SEM. The differences were examined with two-tailed t test

Saito et al., Sci. Transl. Med. 9, eaao1214 (2017) 25 October 2017

(GraphPad Prism, GraphPad). Statisticalmethods for genomic analysesare included in the “Genome analysis” section.

For in vivo treatment experiments, pre- and posttreatment datawereobtained from the PB of eachmouse. To detect differences between pre-and posttreatment data obtained from the same mouse, paired two-tailed t test (pairing pre- and posttreatment values of each mouse)was used. Because pretreatment data from the BM/spleen of the miceare not available, differences in the BM and spleen of the mice wereevaluated by unpaired two-tailed t test. We designed the experimentsuch that there were three to six engrafted recipients per treatmentgroup fromagiven patient. Therefore, in comparisons restricted tomiceengrafted with AML from a particular patient, n was insufficient (threeto six in each group) formeaningful tests of normality or variation. In allsuch comparisons, performing nonparametric tests did not yield con-tradicting results on the significance of the differences detected, and thecomparisons were sufficiently powered, although n was small, becausethe sizes of the differences detectedwere sufficiently large and the valuesobservedwithin each groupwere sufficiently tightly distributed (SDwassufficiently low). For comparisons of treatment groups including allmice (across patient samples), we chose to use the t test because para-metric tests on non-Gaussian data are robust as long as sample sizes aresufficient. In these cases, using nonparametric tests also did not yieldcontradicting results.

Because n was relatively small (<20) for groups restricted to a par-ticular patient sample, independent data points were plotted in all rele-vant figures, and all data are tabulated in tables S1 to S8. In comparisonsof whole treatment groups (containing experiments using samples frommultiple patients), normality testing using two different tests (D’Agostino-Pearson omnibus test and Shapiro-Wilk test) yielded variable results.However, because n is relatively large, detected differences are relativelylarge, and deviations among data points within each group are relativelysmall; parametric test (two-tailed t test) should be robust. Performingnonparametric tests on these data sets did not contradict the results ofthe parametric two-tailed t test.

In comparisons between groups restricted to individual patients,nwasinsufficient to obtain ameaningful estimate on the variance. For aggregatedata ofmice engraftedwithAML frommultiple patients, nwas sufficient-ly large; however, in such comparisons, we did not expect variances to besimilar because patients have highly heterogeneous disease characteristicsand biology. As expected, the F test used to compare variances showedsimilar variances between some groups but not in others. In comparisonsbetween groups restricted to individual patients, nwas insufficient to ob-tain ameaningful estimate on the variance.Wedidnot expect variances tobe similar for aggregate data acrossmultiple patient samples and compar-isons made on mice engrafted with AML from multiple patients withhighly heterogeneous disease characteristics and biology.

SUPPLEMENTARY MATERIALSwww.sciencetranslationalmedicine.org/cgi/content/full/9/413/eaao1214/DC1Fig. S1. Identification of hematopoietic subpopulations in patient samples by surfacephenotype and in vivo function in three additional patients.Fig. S2. Mutational profiles of AML cells with and without in vivo RK-20449 treatment.Fig. S3. Mutational profiles of AML patient samples obtained serially during clinical course.Fig. S4. Altered transcription profiles of human AML cells with in vivo RK-20449 treatment.Fig. S5. Overcoming genetically complex AML by targeting leukemia-initiating mutation andantiapoptotic BCL-2 pathway.Table S1. Patient characteristics (provided as an Excel file).Table S2. Information of identified variants (provided as an Excel file).Table S3. Human AML chimerism in the PB, BM, and spleen of AML-engrafted recipientstreated with RK-20449 alone (provided as an Excel file).

10 of 12

SC I ENCE TRANS LAT IONAL MED I C I N E | R E S EARCH ART I C L E

Table S4. Human AML chimerism in the PB, BM, and spleen of AML-engrafted recipientstreated with ABT-199 alone, RK-20449 alone, or combination (provided as an Excel file).Table S5. Statistics comparing pre- and posttreatment PB human AML chimerism in recipientstreated with ABT-199 alone, RK-20449 alone, or combination (provided as an Excel file).Table S6. Statistics comparing BM and spleen human AML chimerism in recipients treated withABT-199 alone, RK-20449 alone, or combination (provided as an Excel file).Table S7. Effect of in vivo exposure to combined RK-20449 and ABT-199 on humanmultilineage hematopoiesis (provided as an Excel file).Table S8. PCR primers for targeted sequencing of MALBAC products (provided as an Excel file).Reference (45)

by guest on May 11, 2021

http://stm.sciencem

ag.org/D

ownloaded from

REFERENCES AND NOTES1. Cancer Genome Atlas Research Network, Comprehensive molecular profiling of lung

adenocarcinoma. Nature 511, 543–550 (2014).2. C. Kandoth, M. D. McLellan, F. Vandin, K. Ye, B. Niu, C. Lu, M. Xie, Q. Zhang, J. F. McMichael,

M. A. Wyczalkowski, M. D. M. Leiserson, C. A. Miller, J. S. Welch, M. J. Walter,M. C. Wendl, T. J. Ley, R. K. Wilson, B. J. Raphael, L. Ding, Mutational landscape andsignificance across 12 major cancer types. Nature 502, 333–339 (2013).

3. J. M. Klco, D. H. Spencer, C. A. Miller, M. Griffith, T. L. Lamprecht, M. O’Laughlin, C. Fronick,V. Magrini, R. T. Demeter, R. S. Fulton, W. C. Eades, D. C. Link, T. A. Graubert,M. J. Walter, E. R. Mardis, J. F. Dipersio, R. K. Wilson, T. J. Ley, Functionalheterogeneity of genetically defined subclones in acute myeloid leukemia.Cancer Cell 25, 379–392 (2014).

4. C. A. Miller, R. K. Wilson, T. J. Ley, Genomic landscapes and clonality of de novo AML.N. Engl. J. Med. 369, 1473 (2013).

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

6. F. Delhommeau, S. Dupont, V. Della Valle, C. James, S. Trannoy, A. Massé, O. Kosmider,J.-P. Le Couedic, F. Robert, A. Alberdi, Y. Lécluse, I. Plo, F. J. Dreyfus, C. Marzac,N. Casadevall, C. Lacombe, S. P. Romana, P. Dessen, J. Soulier, F. Viguié, M. Fontenay,W. Vainchenker, O. A. Bernard, Mutation in TET2 in myeloid cancers. N. Engl. J. Med. 360,2289–2301 (2009).

7. T. J. Ley, L. Ding, M. J. Walter, M. D. McLellan, T. Lamprecht, D. E. Larson, C. Kandoth,J. E. Payton, J. Baty, J. Welch, C. C. Harris, C. F. Lichti, R. R. Townsend, R. S. Fulton,D. J. Dooling, D. C. Koboldt, H. Schmidt, Q. Zhang, J. R. Osborne, L. Lin, M. O’Laughlin,J. F. McMichael, K. D. Delehaunty, S. D. McGrath, L. A. Fulton, V. J. Magrini, T. L. Vickery,J. Hundal, L. L. Cook, J. J. Conyers, G. W. Swift, J. P. Reed, P. A. Alldredge, T. Wylie,J. Walker, J. Kalicki, M. A. Watson, S. Heath, W. D. Shannon, N. Varghese, R. Nagarajan,P. Westervelt, M. H. Tomasson, D. C. Link, T. A. Graubert, J. F. DiPersio, E. R. Mardis,R. K. Wilson, DNMT3A mutations in acute myeloid leukemia. N. Engl. J. Med. 363,2424–2433 (2010).

8. P. S. Ward, J. Patel, D. R. Wise, O. Abdel-Wahab, B. D. Bennett, H. A. Coller, J. R. Cross,V. R. Fantin, C. V. Hedvat, A. E. Perl, J. D. Rabinowitz, M. Carroll, S. M. Su, K. A. Sharp,R. L. Levine, C. B. Thompson, The common feature of leukemia-associated IDH1and IDH2 mutations is a neomorphic enzyme activity converting a-ketoglutarate to2-hydroxyglutarate. Cancer Cell 17, 225–234 (2010).

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

10. L. I. Shlush, S. Zandi, A. Mitchell, W. C. Chen, J. M. Brandwein, V. Gupta, J. A. Kennedy,A. D. Schimmer, A. C. Schuh, K. W. Yee, J. L. McLeod, M. Doedens, J. J. F. Medeiros,R. Marke, H. J. Kim, K. Lee, J. D. McPherson, T. J. Hudson; HALT Pan-Leukemia Gene PanelConsortium, A. M. K. Brown, F. Yousif, Q. M. Trinh, L. D. Stein, M. D. Minden,J. C. Y. Wang, J. E. Dick, Identification of pre-leukaemic haematopoietic stem cellsin acute leukaemia. Nature 506, 328–333 (2014).

11. J. S. Welch, T. J. Ley, T. J. Ley, D. C. Link, C. A. Miller, D. E. Larson, D. C. Koboldt,L. D. Wartman, T. L. Lamprecht, F. Liu, J. Xia, C. Kandoth, R. S. Fulton, M. D. McLellan,D. J. Dooling, J. W. Wallis, K. Chen, C. C. Harris, H. K. Schmidt, J. M. Kalicki-Veizer, C. Lu,Q. Zhang, L. Lin, M. D. O’Laughlin, J. F. McMichael, K. D. Delehaunty, L. A. Fulton,V. J. Magrini, S. D. McGrath, R. T. Demeter, T. L. Vickery, J. Hundal, L. L. Cook, G. W. Swift,J. P. Reed, P. A. Alldredge, T. N. Wylie, J. R. Walker, M. A. Watson, S. E. Heath,W. D. Shannon, N. Varghese, R. Nagarajan, J. E. Payton, J. D. Baty, S. Kulkarni, J. M. Klco,M. H. Tomasson, P. Westervelt, M. J. Walter, T. A. Graubert, J. F. DiPersio, L. Ding,E. R. Mardis, R. K. Wilson, The origin and evolution of mutations in acute myeloidleukemia. Cell 150, 264–278 (2012).

12. G. Genovese, A. K. Kähler, R. E. Handsaker, J. Lindberg, S. A. Rose, S. F. Bakhoum,K. Chambert, E. Mick, B. M. Neale, M. Fromer, S. M. Purcell, O. Svantesson, M. Landén,M. Höglund, S. Lehmann, S. B. Gabriel, J. L. Moran, E. S. Lander, P. F. Sullivan, P. Sklar,H. Grönberg, C. M. Hultman, S. A. McCarroll, Clonal hematopoiesis and blood-cancer riskinferred from blood DNA sequence. N. Engl. J. Med. 371, 2477–2487 (2014).

Saito et al., Sci. Transl. Med. 9, eaao1214 (2017) 25 October 2017

13. S. Jaiswal, P. Fontanillas, J. Flannick, A. Manning, P. V. Grauman, B. G. Mar, R. C. Lindsley,C. H. Mermel, N. Burtt, A. Chavez, J. M. Higgins, V. Moltchanov, F. C. Kuo, M. J. Kluk,B. Henderson, L. Kinnunen, H. A. Koistinen, C. Ladenvall, G. Getz, A. Correa, B. F. Banahan,S. Gabriel, S. Kathiresan, H. M. Stringham, M. I. McCarthy, M. Boehnke, J. Tuomilehto,C. Haiman, L. Groop, G. Atzmon, J. G. Wilson, D. Neuberg, D. Altshuler, B. L. Ebert,Age-related clonal hematopoiesis associated with adverse outcomes. N. Engl. J. Med.371, 2488–2498 (2014).

14. M. Xie, C. Lu, J. Wang, M. D. McLellan, K. J. Johnson, M. C. Wendl, J. F. McMichael,H. K. Schmidt, V. Yellapantula, C. A. Miller, B. A. Ozenberger, J. S. Welch, D. C. Link,M. J. Walter, E. R. Mardis, J. F. Dipersio, F. Chen, R. K. Wilson, T. J. Ley, L. Ding, Age-relatedmutations associated with clonal hematopoietic expansion and malignancies.Nat. Med. 20, 1472–1478 (2014).

15. F. Notta, A. Zandi, N. Takayama, S. Dobson, O. I. Gan, G. Wilson, K. B. Kaufmann, J. McLeod,E. Laurenti, C. F. Dunant, J. D. McPherson, L. D. Stein, Y. Dror, J. E. Dick, Distinctroutes of lineage development reshape the human blood hierarchy across ontogeny.Science 351, aab2116 (2016).

16. L. Quek, G. W. Otto, C. Garnett, L. Lhermitte, D. Karamitros, B. Stoilova, I.-J. Lau,J. Doondeea, B. Usukhbayar, A. Kennedy, M. Metzner, N. Goardon, A. Ivey, C. Allen, R. Gale,B. Davies, A. Sternberg, S. Killick, H. Hunter, P. Cahalin, A. Price, A. Carr, M. Griffiths,P. Virgo, S. Mackinnon, D. Grimwade, S. Freeman, N. Russell, C. Craddock, A. Mead,A. Peniket, C. Porcher, P. Vyas, Genetically distinct leukemic stem cells in humanCD34- acute myeloid leukemia are arrested at a hemopoietic precursor-like stage.J. Exp. Med. 213, 1513–1535 (2016).

17. Y. Saito, H. Yuki, M. Kuratani, Y. Hashizume, S. Takagi, T. Honma, A. Tanaka, M. Shirouzu,J. Mikuni, N. Handa, I. Ogahara, A. Sone, Y. Najima, Y. Tomabechi, M. Wakiyama,N. Uchida, M. Tomizawa-Murasawa, A. Kaneko, S. Tanaka, N. Suzuki, H. Kajita, Y. Aoki,O. Ohara, L. D. Shultz, T. Fukami, T. Goto, S. Taniguchi, S. Yokoyama, F. Ishikawa,A pyrrolo-pyrimidine derivative targets human primary AML stem cells in vivo.Sci. Transl. Med. 5, 181ra52 (2013).

18. L. I. Shlush, A. Mitchell, L. Heisler, S. Abelson, S. W. K. Ng, A. Trotman-Grant,J. J. F. Medeiros, A. Rao-Bhatia, I. Jaciw-Zurakowsky, R. Marke, J. L. McLeod, M. Doedens,G. Bader, V. Voisin, C. Xu, J. D. McPherson, T. J. Hudson, J. C. Y. Wang, M. D. Minden,J. E. Dick, Tracing the origins of relapse in acute myeloid leukaemia to stem cells.Nature 547, 104–108 (2017).

19. C. Hetz, The unfolded protein response: Controlling cell fate decisions under ER stressand beyond. Nat. Rev. Mol. Cell Biol. 13, 89–102 (2012).

20. J. A. P. Spijkers-Hagelstein, P. Schneider, E. Hulleman, J. de Boer, O. Williams, R. Pieters,R. W. Stam, Elevated S100A8/S100A9 expression causes glucocorticoid resistance in MLL-rearranged infant acute lymphoblastic leukemia. Leukemia 26, 1255–1265 (2012).

21. M. Yang, P. Zeng, R. Kang, Y. Yu, L. Yang, D. Tang, L. Cao, S100A8 contributes to drugresistance by promoting autophagy in leukemia cells. PLOS ONE 9, e97242 (2014).

22. X.-y. Yang, M.-y. Zhang, Q. Zhou, S.-y. Wu, Y. Zhao, W.-y. Gu, J. Pan, J.-n. Cen, Z.-x. Chen,W.-g. Guo, C.-s. Chen, W.-h. Yan, S.-y. Hu, High expression of S100A8 gene is associatedwith drug resistance to etoposide and poor prognosis in acute myeloid leukemiathrough influencing the apoptosis pathway. Onco Targets Ther. 9, 4887–4899 (2016).

23. S. Zhu, Q. Zhang, X. Sun, H. J. Zeh III, M. T. Lotze, R. Kang, D. Tang, HSPA5 regulatesferroptotic cell death in cancer cells. Cancer Res. 77, 2064–2077 (2017).

24. J. Montero, K. A. Sarosiek, J. D. DeAngelo, O. Maertens, J. Ryan, D. Ercan, H. Piao,N. S. Horowitz, R. S. Berkowitz, U. Matulonis, P. A. Jänne, P. C. Amrein, K. Cichowski,R. Drapkin, A. Letai, Drug-induced death signaling strategy rapidly predicts cancerresponse to chemotherapy. Cell 160, 977–989 (2015).

25. T. N. Chonghaile, J. E. Roderick, C. Glenfield, J. Ryan, S. E. Sallan, L. B. Silverman, M. L. Loh,S. P. Hunger, B. Wood, D. J. DeAngelo, R. Stone, M. Harris, A. Gutierrez, M. A. Kelliher,A. Letai, Maturation stage of T-cell acute lymphoblastic leukemia determines BCL-2versus BCL-XL dependence and sensitivity to ABT-199. Cancer Discov. 4, 1074–1087(2014).

26. V. Del Gaizo Moore, J. R. Brown, M. Certo, T. M. Love, C. D. Novina, A. Letai, Chroniclymphocytic leukemia requires BCL2 to sequester prodeath BIM, explaining sensitivity toBCL2 antagonist ABT-737. J. Clin. Invest. 117, 112–121 (2007).

27. G. L. Kelly, S. Grabow, S. P. Glaser, L. Fitzsimmons, B. J. Aubrey, T. Okamoto, L. J. Valente,M. Robati, L. Tai, W. Douglas Fairlie, E. F. Lee, M. S. Lindstrom, K. G. Wiman,D. C. S. Huang, P. Bouillet, M. Rowe, A. B. Rickinson, M. J. Herold, A. Strasser, Targetingof MCL-1 kills MYC-driven mouse and human lymphomas even when they bearmutations in p53. Genes Dev. 28, 58–70 (2014).

28. A. Letai, M. D. Sorcinelli, C. Beard, S. J. Korsmeyer, Antiapoptotic BCL-2 is required formaintenance of a model leukemia. Cancer Cell 6, 241–249 (2004).

29. T.-T. Vo, J. Ryan, R. Carrasco, D. Neuberg, D. J. Rossi, R. M. Stone, D. J. DeAngelo,M. G. Frattini, A. Letai, Relative mitochondrial priming of myeloblasts and normal HSCsdetermines chemotherapeutic success in AML. Cell 151, 344–355 (2012).

30. G. Brumatti, N. Lalaoui, A. H. Wei, J. Silke, ‘Did he who made the lamb make thee?’ Newdevelopments in treating the ‘fearful symmetry’ of acute myeloid leukemia.Trends Mol. Med. 23, 264–281 (2017).

11 of 12

SC I ENCE TRANS LAT IONAL MED I C I N E | R E S EARCH ART I C L E

by gueshttp://stm

.sciencemag.org/

Dow

nloaded from

31. D. P. Steensma, R. Bejar, S. Jaiswal, R. C. Lindsley, M. A. Sekeres, R. P. Hasserjian,B. L. Ebert, Clonal hematopoiesis of indeterminate potential and its distinction frommyelodysplastic syndromes. Blood 126, 9–16 (2015).

32. J. S. Garcia, R. M. Stone, The development of FLT3 inhibitors in acute myeloid leukemia.Hematol. Oncol. Clin. North Am. 31, 663–680 (2017).

33. D. A. Sallman, J. E. Lancet, What are the most promising new agents in acute myeloidleukemia? Curr. Opin. Hematol. 24, 99–107 (2017).

34. R. M. Stone, S. J. Mandrekar, B. L. Sanford, K. Laumann, S. Geyer, C. D. Bloomfield,C. Thiede, T. W. Prior, K. Döhner, G. Marcucci, F. Lo-Coco, R. B. Klisovic, A. Wei, J. Sierra,M. A. Sanz, J. M. Brandwein, T. de Witte, D. Niederwieser, F. R. Appelbaum, B. C. Medeiros,M. S. Tallman, J. Krauter, R. F. Schlenk, A. Ganser, H. Serve, G. Ehninger, S. Amadori,R. A. Larson, H. Döhner, Midostaurin plus chemotherapy for acute myeloid leukemia witha FLT3 mutation. N. Engl. J. Med. 377, 454–464 (2017).

35. F. Ishikawa, M. Yasukawa, B. Lyons, S. Yoshida, T. Miyamoto, G. Yoshimoto, T. Watanabe,K. Akashi, L. D. Shultz, M. Harada, Development of functional human blood andimmune systems in NOD/SCID/IL2 receptor g chainnull mice. Blood 106, 1565–1573(2005).

36. L. D. Shultz, B. L. Lyons, L. M. Burzenski, B. Gott, X. Chen, S. Chaleff, M. Kotb, S. D. Gillies,M. King, J. Mangada, D. L. Greiner, R. Handgretinger, Human lymphoid andmyeloid cell development in NOD/LtSz-scid IL2Rgnull mice engrafted withmobilized human hemopoietic stem cells. J. Immunol. 174, 6477–6489 (2005).

37. C. Zong, S. Lu, A. R. Chapman, X. S. Xie, Genome-wide detection of single-nucleotide andcopy-number variations of a single human cell. Science 338, 1622–1626 (2012).

38. D. Grün, A. van Oudenaarden, Design and analysis of single-cell sequencing experiments.Cell 163, 799–810 (2015).

39. M. L. Leung, Y. Wang, J. Waters, N. E. Navin, SNES: Single nucleus exome sequencing.Genome Biol. 16, 55 (2015).

40. R. C. Lindsley, W. Saber, B. G. Mar, R. Redd, T. Wang, M. D. Haagenson, P. V. Grauman,Z.-H. Hu, S. R. Spellman, S. J. Lee, M. R. Verneris, K. Hsu, K. Fleischhauer, C. Cutler,J. H. Antin, D. Neuberg, B. L. Ebert, Prognostic mutations in myelodysplastic syndromeafter stem-cell transplantation. N. Engl. J. Med. 376, 536–547 (2017).

41. R. Pan, L. J. Hogdal, J. M. Benito, D. Bucci, L. Han, G. Borthakur, J. Cortes, D. J. DeAngelo,L. Debose, H. Mu, H. Döhner, V. I. Gaidzik, I. Galinsky, L. S. Golfman, T. Haferlach,K. G. Harutyunyan, J. Hu, J. D. Leverson, G. Marcucci, M. Müschen, R. Newman, E. Park,P. P. Ruvolo, V. Ruvolo, J. Ryan, S. Schindela, P. Zweidler-McKay, R. M. Stone,H. Kantarjian, M. Andreeff, M. Konopleva, A. G. Letai, Selective BCL-2 inhibition byABT-199 causes on-target cell death in acute myeloid leukemia. Cancer Discov. 4,362–375 (2014).

42. A. J. Souers, J. D. Leverson, E. R. Boghaert, S. L. Ackler, N. D. Catron, J. Chen, B. D. Dayton,H. Ding, S. H. Enschede, W. J. Fairbrother, D. C. S. Huang, S. G. Hymowitz, S. Jin, S. L. Khaw,P. J. Kovar, L. T. Lam, J. Lee, H. L. Maecker, K. C. Marsh, K. D. Mason, M. J. Mitten,P. M. Nimmer, A. Oleksijew, C. H. Park, C.-M. Park, D. C. Phillips, A. W. Roberts, D. Sampath,J. F. Seymour, M. L. Smith, G. M. Sullivan, S. K. Tahir, C. Tse, M. D. Wendt, Y. Xiao,

Saito et al., Sci. Transl. Med. 9, eaao1214 (2017) 25 October 2017

J. C. Xue, H. Zhang, R. A. Humerickhouse, S. H. Rosenberg, S. W. Elmore, ABT-199, apotent and selective BCL-2 inhibitor, achieves antitumor activity while sparingplatelets. Nat. Med. 19, 202–208 (2013).

43. J. Ryan, A. Letai, BH3 profiling in whole cells by fluorimeter or FACS. Methods 61, 156–164(2013).

44. J. A. Ryan, J. K. Brunelle, A. Letai, Heightened mitochondrial priming is the basis forapoptotic hypersensitivity of CD4+ CD8+ thymocytes. Proc. Natl. Acad. Sci. U.S.A. 107,12895–12900 (2010).

45. K. Wang, M. Li, H. Hakonarson, ANNOVAR: Functional annotation of genetic variants fromhigh-throughput sequencing data. Nucleic Acids Res. 38, e164 (2010).

Acknowledgments: We would like to acknowledge and thank the patients and the clinicalcare team at the Toranomon Hospital, whose cooperation and effort made this studypossible. We thank R. C. Lindsley and J. S. Garcia for the critical review of the manuscript.Funding: This study was supported by the Basic Science and Platform TechnologyProgram for Innovative Biological Medicine (to F.I.) and the Project for Development ofInnovative Research on Cancer Therapeutics (to F.I.) from the Japanese Ministry ofEducation, Culture, Sports, Science, and Technology, the Japan Agency for MedicalResearch and Development (to F.I.), and the Maine Cancer Foundation and the NIH (grantsCA034196 and CA17983 to L.D.S.). Author contributions: F.I., Y.S., O.O., and A.G.L.conceptualized and designed the study. F.I., O.O., Y.M., and T.W. developed themethodology. I.O., L.H., K.S., A.K., and H.K. acquired the data. F.I., Y.S., O.O., A.G.L., Y.M.,and T.W. analyzed and interpreted the data. F.I., Y.S., O.O., A.G.L., and L.D.S. wrote,reviewed, and/or revised the manuscript. T.F. provided administrative, technical, andmaterial support. F.I. and Y.S. supervised the study. S. Takagi, N.U., and S. Taniguchiprovided the patient samples and clinical information. Competing interests: Y.S. and F.I.are inventors on patent application (62/394871) submitted by RIKEN that covers thecombinatory use of RK-20449 and BCL-2 inhibitors for AML. All other authors declare thatthey have no competing interests. Data and materials availability: Raw sequencedata reported in this paper are available at the National Bioscience Database CenterHuman database (Japan) (accession no. hum0116). NSG mice are available from theJackson Laboratory under a material transfer agreement.

Submitted 15 June 2017Accepted 25 September 2017Published 25 October 201710.1126/scitranslmed.aao1214

Citation: Y. Saito, Y. Mochizuki, I. Ogahara, T. Watanabe, L. Hogdal, S. Takagi, K. Sato,A. Kaneko, H. Kajita, N. Uchida, T. Fukami, L. D. Shultz, S. Taniguchi, O. Ohara, A. G. Letai,F. Ishikawa, Overcoming mutational complexity in acute myeloid leukemia by inhibition ofcritical pathways. Sci. Transl. Med. 9, eaao1214 (2017).

t on

12 of 12

May 11, 2021

pathwaysOvercoming mutational complexity in acute myeloid leukemia by inhibition of critical

Anthony G. Letai and Fumihiko IshikawaKaneko, Hiroshi Kajita, Naoyuki Uchida, Takehiro Fukami, Leonard D. Shultz, Shuichi Taniguchi, Osamu Ohara, Yoriko Saito, Yoshiki Mochizuki, Ikuko Ogahara, Takashi Watanabe, Leah Hogdal, Shinsuke Takagi, Kaori Sato, Akiko

DOI: 10.1126/scitranslmed.aao1214, eaao1214.9Sci Transl Med

target them.authors were able to identify mutations that drive the evolution of leukemia and figure out effective approaches totheir leukemogenic capacity in immunosuppressed mice. By combining this approach with genomic analysis, the

isolated subpopulations of leukemic cells with specific mutations and monitoredet al.address this issue, Saito can be difficult to know which of these mutations contribute to tumorigenesis and should therefore be targeted. To

A variety of mutations have been observed in cancer cells from patients with acute myeloid leukemia, but itThe right treatments for the right mutations

ARTICLE TOOLS http://stm.sciencemag.org/content/9/413/eaao1214

MATERIALSSUPPLEMENTARY http://stm.sciencemag.org/content/suppl/2017/10/23/9.413.eaao1214.DC1

CONTENTRELATED

http://stm.sciencemag.org/content/scitransmed/13/587/eabg1168.fullhttp://stm.sciencemag.org/content/scitransmed/12/538/eaax5104.fullhttp://science.sciencemag.org/content/sci/367/6477/586.fullhttp://stm.sciencemag.org/content/scitransmed/11/516/eaax2863.fullhttp://stm.sciencemag.org/content/scitransmed/11/508/eaaw8828.fullhttp://stm.sciencemag.org/content/scitransmed/10/457/eaar7939.fullhttp://stke.sciencemag.org/content/sigtrans/11/539/eaao5617.fullhttp://stke.sciencemag.org/content/sigtrans/11/526/eaan6776.fullhttp://stm.sciencemag.org/content/scitransmed/10/436/eaao3003.fullhttp://stm.sciencemag.org/content/scitransmed/10/431/eaam8460.fullhttp://stm.sciencemag.org/content/scitransmed/10/428/eaal1571.fullhttp://science.sciencemag.org/content/sci/358/6364/714.fullhttp://stm.sciencemag.org/content/scitransmed/9/374/eaaj2025.fullhttp://stm.sciencemag.org/content/scitransmed/8/368/368ra171.fullhttp://stm.sciencemag.org/content/scitransmed/8/359/359ra129.fullhttp://stm.sciencemag.org/content/scitransmed/8/357/357ra123.full

REFERENCES

http://stm.sciencemag.org/content/9/413/eaao1214#BIBLThis article cites 45 articles, 13 of which you can access for free

Terms of ServiceUse of this article is subject to the

registered trademark of AAAS. is aScience Translational MedicineScience, 1200 New York Avenue NW, Washington, DC 20005. The title

(ISSN 1946-6242) is published by the American Association for the Advancement ofScience Translational Medicine

of Science. No claim to original U.S. Government WorksCopyright © 2017 The Authors, some rights reserved; exclusive licensee American Association for the Advancement

by guest on May 11, 2021

http://stm.sciencem

ag.org/D

ownloaded from

PERMISSIONS http://www.sciencemag.org/help/reprints-and-permissions

Terms of ServiceUse of this article is subject to the

registered trademark of AAAS. is aScience Translational MedicineScience, 1200 New York Avenue NW, Washington, DC 20005. The title

(ISSN 1946-6242) is published by the American Association for the Advancement ofScience Translational Medicine

of Science. No claim to original U.S. Government WorksCopyright © 2017 The Authors, some rights reserved; exclusive licensee American Association for the Advancement

by guest on May 11, 2021

http://stm.sciencem

ag.org/D

ownloaded from