Single cell analysis exposes intratumor heterogeneity and suggests that FLT3-ITD is a late event in leukemogenesis

Accepted Manuscript Single cell analysis exposes intra-tumor heterogeneity and suggests that FLT3-ITD is a late event in leukemogenesis Roni Shouval, MD Liran I. Shlush, MD, PhD Shlomit Yehudai-Resheff, PhD Shahnaz Ali, PhD Neta Pery, MSc Ehud Shapiro, PhD Maty Tzukerman, PhD Jacob M. Rowe, MD Tsila Zuckerman, MD PII:

S0301-472X(14)00057-5

DOI:

10.1016/j.exphem.2014.01.010

Reference:

EXPHEM 3100

To appear in:

Experimental Hematology

Received Date: 19 August 2013 Revised Date:

5 January 2014

Accepted Date: 27 January 2014

Please cite this article as: Shouval R, Shlush LI, Yehudai-Resheff S, Ali S, Pery N, Shapiro E, Tzukerman M, Rowe JM, Zuckerman T, Single cell analysis exposes intra-tumor heterogeneity and suggests that FLT3-ITD is a late event in leukemogenesis, Experimental Hematology (2014), doi: 10.1016/j.exphem.2014.01.010. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Single cell analysis exposes intra-tumor heterogeneity and suggests that FLT3-ITD is a late event in leukemogenesis Roni Shouval, MD1*, Liran I. Shlush, MD, PhD1*, Shlomit Yehudai-Resheff, PhD4, Shahnaz Ali, PhD 1, Zuckerman, MD1,4 * Equal contribution of both authors 1

RI PT

Neta Pery, MSc1, Ehud Shapiro, PhD2, Maty Tzukerman, PhD1, Jacob M. Rowe, MD1,3, Tsila

Bruce Rappaport Faculty of Medicine and Research Institute, Technion – Israel Institute of

Technology, Haifa, Israel

Department of Biological Chemistry, Weizmann Institute of Science, Rehovot, Israel

3

Department of Hematology, Shaare Zedek Medical Center, Jerusalem, Israel

4

Department of Hematology & Bone Marrow Transplantation, Rambam Healthcare Campus, Haifa,

M AN U

SC

2

Israel

Running title: Single cell analysis of leukemic blasts Category: Molecular genetics

Corresponding Author: Tsila Zuckerman, MD

TE D

Text word count: 2,840

Department of Hematology and Bone Marrow Transplantation

P.O. Box 9602

Israel

AC C

Haifa 31096

EP

Rambam Health Care Campus

Tel: 972 4 8542541

Fax: 972 4 8542343

e-mail: [email protected]

1

ACCEPTED MANUSCRIPT Abstract FMS-like tyrosine kinase 3 receptor- internal tandem duplication (FLT3-ITD) commonly occurs in acute myeloid leukemia (AML) and is considered rare in acute lymphocytic leukemia (ALL). Acute leukemia has poor prognosis, mainly due to relapse. Standard FLT3-ITD diagnostic techniques are based on genomic PCR and have recently incorporated GeneScan to identify variations of the FLT3

RI PT

gene. As this is an average-based assay utilized in a heterogeneous leukemic cell population, we hypothesized that cells of acute leukemia, considered FLT3-ITD-negative by standard methods, could possess a fraction of FLT3-ITD-positive cells. The present study employed single cell mutation analysis to evaluate the FLT3-ITD status in newly diagnosed AML (n=5) and ALL (n=3) patients. A total of 541 single leukemic cells and 36 mononuclear cells from healthy volunteers were analyzed.

SC

7/8 patients considered FLT3-ITD-negative according to bulk DNA analysis, appeared to possess a small fraction of FLT3-ITD-positive cells based on single cell analysis (SCA). Moreover, this approach

M AN U

revealed the heterogeneity of the tumor as evident by different FLT3-ITD mutations present in the same patient. The presence of a minor clone carrying FLT3-ITD in almost all patients tested, provides an evidence that this lesion is a common late event in leukemogenesis. Additionally, 3 relapsed patients demonstrated loss of heterozygosity (LOH) of the normal allele, affecting 25-100% of the cells found to be FLT3-ITD-positive. Though further clinical testing is warranted, these findings may have

Abstract word count: 240

TE D

implications on the prognostic significance of FLT3-ITD and the use of targeted therapy.

Key words: FLT3-internal tandem duplication, acute myeloid leukemia, acute lymphocytic leukemia,

AC C

EP

loss of heterozygosity, single cell mutation analysis, genomic PCR

2

ACCEPTED MANUSCRIPT Introduction Mutations in the FMS-like tyrosine kinase 3 receptor (FLT3) gene are one of the most common molecular abnormalities in de novo AML (prevalence 15-35%)[1, 2]. The FLT3-ITD is an in-frame mutation involving the juxtamembrane region of the FLT3 protein causing constitutive activation of the receptor’s tyrosine kinase activity, which promotes proliferation and survival of hematopoietic

RI PT

progenitor cells through multiple signaling pathways[2]. The presence of FLT3-ITD mutation in AML is known to be associated with poor prognosis[3]; hence, this abnormality is taken into account both in molecular risk stratification of normal cytogenetic AML and in consideration of treatment options[4]. FLT3 inhibitors are at advanced stages of clinical evaluation and have demonstrated potential benefit when combined with standard chemotherapy in the management of AML patients harboring the FLT3-

SC

ITD mutation[2]. In contrast to AML, FLT3-ITD mutation is rarely observed in ALL[1, 5, 6]. Based on the comparison of paired samples obtained at diagnosis and relapse, it has been suggested that

M AN U

FLT3-ITD is a late event in leukemogenesis related to its instability[7, 8]. Other studies have suggested FLT3-ITD to be a driver mutation in AML[9].

Currently, genomic PCR performed on peripheral blood (PB) or bone marrow (BM) blast cells is the standard method for determining the FLT3-ITD mutational status[10]. However, in a heterogeneous process such as leukemia, averaged measurement of cell populations may miss a cryptic population substructure among leukemic clones (i.e., a small subpopulation positive for the mutation which might be concealed by the mean negative status of the general population)[11-14]. Given all the above, we

TE D

investigated whether patients with newly diagnosed/relapsed acute leukemias, judged to be FLT3-ITD negative by standard genomic PCR, may possess FLT3-ITD positive cells, which evade detection by current diagnostic methods. Furthermore, the current study also assessed the efficacy of SCA in

Patients and methods

AC C

Patient samples

EP

revealing intra-clonal diversity and loss of heterozygosity.

Single cell analysis (SCA) was employed to study the FLT3-ITD mutational status in five adult AML patients, three adult ALL patients and in four healthy controls. All participants provided signed informed consent, and the study was approved by the Institutional Review Board (IRB) of the Rambam Health Care Campus (Approval # 028009). For five of the eight study patients, samples were available both from diagnosis and relapse. Cell preparation Blast cells were identified using specific leukemia-associated immunophenotype (LAIP) markers (Table 1). Bulk DNA was extracted from mononuclear cells isolated from BM or PB at

3

ACCEPTED MANUSCRIPT diagnosis and relapse, using density gradient centrifugation with Ficoll lymphocyte separation medium (MP Biomedicals). Single cells were isolated by re-suspending the bulk of cells in PBS supplemented with 2% FBS and 1 mM EDTA pH 8.0, until a dilution of about 1 cell/0.5 ul was obtained. Then, 0.5 ul drops of this dilution were placed at the center of multiple wells of a 96-well (flat bottom) plate.

RI PT

Microscope observation was used to identify wells with exactly one cell.

DNA isolation using whole genome amplification (WGA)

WGA was performed using the IllustraGenomiPhi V2 DNA Amplification kit (GE Healthcare Life Sciences) according to the manufacturer’s instructions. Briefly, single cells were transferred to PCR tubes (0.2 ml volume) using 3 µl sample buffer from the GenomiPhi V2 DNA Amplification kit. In the

SC

optimized protocol, 1.5 µl cell lyses solution (600 mM KOH, 10 mM EDTA, 100 mM dithiothreitol (DTT)) was added to each single cell. Cell lysis was carried out for 10 min at 30°C, followed by the

M AN U

addition of 1.5 µl neutralizing solution (4 vol 1 MTris-HCl, pH 8.0, added to 1 vol 3MHCl). Furthermore, 4 µl sample buffer, 9 µl reaction buffer, and 1 µl enzyme mixture supplied with the GenomiPhi V2 DNA Amplification kit were added to complete the reaction. The amplification was then carried out at 300C for 4 hours followed by heat inactivation at 650C for 10 minutes.

FLT3-ITD mutation detection

FLT3-ITD PCR mutation analysis was performed on the WGA DNA products according to the former

TE D

specifications (Repli-g mini kit, Qiagen #150025, 1500100) and primers (Table 2)[12]. Similarly, FLT3ITD status was determined for bulk DNA extracted from mononuclear cells using the automated iPrepTM apparatus (invitrogen), with the iPrep™ Card: gDNA Blood #10012. The reaction preparation was done following manufacturer's instructions. The PCR products were separated on gel

EP

electrophoresis (Figure 2). MV-4 11 cell line was used as a positive control.

AC C

FLT3-ITD detection using GeneScan device FLT3-ITD identification in AML patients was performed by PCR with specific oligos for exons 11 and 12 of the FLT3 gene with DNA extracted from a bulk. 5 pmol/µl of each oligo (for oligos sequence see Table 2 below), dNTPs mix 5mM each, 0.1 µl Supertherm DNA polymerase (start PCR master mixThermo) and 20 ng of DNA were used in a total reaction volume of 25µl. PCR program included 2.5 minutes on 940C followed by 30 cycles of 30 seconds at 940C, 1 minute of 570C, and 2 minutes at 720C, terminating by 45 seconds at 600C. PCR products were then prepared for Genetic Analyzer reaction by addition of 9.5 µl formamide and 0.5 µl Red DNA size standard. If a single pick of 328 base pairs (BP) was obtained, the sample was determined as wild type (WT); however, if an additional

4

ACCEPTED MANUSCRIPT pick of a longer PCR product was obtained, the sample was determined as positive for FLT3-IDT[3, 15].

High Resolution Melting reaction PCR and melting analyses of the FLT3-ITD mutation were performed on the 36/72 well capacity

RI PT

Rotor-Gene Q 2plex real-time PCR cycler (QIAGEN, Hilden, Germany). Each PCR mix contained the following: 20 ng of human genomic DNA sample, 5 nM of each of the FLT3-ITD forward and reverses primers[16], 5µl Thermo-Start PCR Master Mix 2X (Thermo Fisher Scientific Inc., MA, USA) and 1µl LCGreen Plus 10X in TE (Idaho Technology Inc., UT, USA) in a 10 µl reaction volume. The following conditions were used: for PCR initial activation step, 15min at 950C; for amplification, 30 sec at 950C,

SC

30 sec at 600C, and 45 sec at 720C for 50 cycles and a single cycle of 30 sec at 950C, 2 min at 500C. Immediately after completing these steps, melting analysis of the PCR product was performed in the

M AN U

same device. The melting signal was acquired from 70 °C to 90 °C at a ramp rate of 0.1 °C/sec. Melting data were visualized and analyzed using RotorGene Q series 1.7 software. Melting curves of examined samples were normalized and the difference temperature graphs were compared against wild-type control samples (Figure 3). Study controls

Several controls were used in this study. T-cells from each patient were analyzed in parallel to the leukemic cells. The T-cells were chosen since they were not part of the leukemic clone and hence

TE D

reflect intra-patient control. T-cells were FACS sorted using CD3 antibody, the gate selection was strict to ensure high T-cell purity[17]. The blast plot was determined by CD45+ with APC-Cy7 on the X axis and the level of cell granulation on the Y axis.

Mononuclear cells from PB of 4 healthy volunteers were also examined throughout the study to

EP

evaluate the specificity of the methods.

In addition, we analyzed the ability to detect FLT3-ITD using SCA and compared it to the conventional

Results

AC C

GeneScan used for bulk DNA.

A total of 553 cells were analyzed, out of which 79 cells were excluded due to low DNA quality, unsuccessful whole genome amplification or inability to verify positive PCR results by high resolution melting (HRM). In addition, 38 cells from healthy volunteers were analyzed in parallel. The results obtained for 474 of the patient cells are presented in Table 3. Six out of the seven patients considered FLT3-ITD negative by standard diagnostic PCR (patients #2-8) (Tables 3 and 4), hence clinically negative, were found to possess a small fraction of single cells positive for this mutation (Figure 1). FLT3-ITD+ cells were present in all ALL patients (#6-8) (Table 3), even though this mutation is rarely

5

ACCEPTED MANUSCRIPT observed in ALL[1, 6]. None of the cells obtained from healthy volunteers possessed a FLT3-ITD mutation.

Comparison between SCA and GeneScan methods The above patient samples were additionally analyzed for the presence of the FLT3-ITD in bulk

RI PT

DNA using conventional PCR and GeneScan device. In 3 out of 13 examined samples there was concordance between the results obtained by Single Cell Analysis and Genescan detection of the lesion in bulk cells (table 3): patient #1 was positive by both methods, while patients #5 and # 7 were negative by both methods at diagnosis. In 7 of 13 samples discordance was observed between the two methods either at diagnosis or relapse, with GeneScan being positive in all the above cases while

SC

the bulk DNA was negative. T-cells from each patient were examined in parallel both at diagnosis and relapse as intra-patient control. In none of these T-cells FLT3-ITD was detected. Additionally, cells

Comparison between diagnosis and relapse

M AN U

from healthy volunteers were found to be negative by both methods.

Comparison between samples obtained at diagnosis and relapse was performed in order to evaluate clonal evolution and sub-clonal heterogeneity. The FLT3-ITD status in single cells exhibited instability (inconsistency) between samples taken at diagnosis and relapse. While in 3 patients (#1,2,8, Table 3) there was a decrease in percentage of cells with FLT3-ITD mutation, in 2 other

TE D

patients (#6,7, Table 3) their percentage increased. Given the small sample size, statistical significance could not be achieved. Variability of the FLT3-ITD

EP

The study demonstrated high frequency of WT LOH, as 25-50% of FLT3-ITD+ cells obtained in patients # 1, 3 at diagnosis showed LOH of FLT3 WT allele (Figure 2) and one FLT3/ITD+ case with

Discussion

AC C

WT LOH was found in patient #8 at relapse (Table 3).

The current study using single cell analysis has demonstrated the presence of FLT3-ITD in a small sub-population of leukemic blasts obtained both at diagnosis and relapse in patients with AML and ALL who were mostly considered negative for the mutation by standard diagnostic methods. This method enabled characterization of intra-patient sub-clonal variability. Moreover, the present study has revealed instability of the FLT3-ITD clone with gain or loss of the mutation at relapse. Furthermore, this study disclosed the existence of LOH that was present in several patients both at diagnosis and relapse, and gave rise to 6 different allele combinations in a single patient. As recently

6

ACCEPTED MANUSCRIPT published by our group[7], lineage SCA and sequencing of the FLT3-ITD gene in patient #1 whose bulk DNA FLT3 genotype was WT/33/66, uncovered 6 possible allele combinations (WT/WT, 33/WT, 66/WT, —/33,—/66, 33/66).

Recently, acute leukemia has been shown to be a highly heterogeneous tumor incorporating

RI PT

different cell sub-populations already at diagnosis[12, 18-20]. This heterogeneity was demonstrated by various methods and can shed light on different mechanisms of relapse through which subpopulations acquire certain properties enabling their escape from chemotherapy[7, 12, 14, 21]. SCA promotes detection of heterogeneity/complexity of the tumor cell population at different stages of the disease [16]. Importantly, heterogeneity within the bulk tumor does not necessarily have to correlate with

SC

heterogeneity within the LSC population. Since it is assumed that only LSCs can sustain disease, determination of the FLT3-ITD status in LSCs might be more informative than analysis of the lesion in

M AN U

bulk tumor cells. However, analysis of the variation in the FLT3-ITD status of LSCs and their leukemogenic potential requires xenograft transplantation studies.

The finding that a small proportion of cells was positive for the FLT3-ITD mutation by SCA, while bulk DNA (both by standard PCR and GeneScan) was negative (representative example, Figure 2B), suggests a higher sensitivity of the SCA, which is capable of revealing cryptic sub-populations of leukemic cells.

Thiede et al.[3] as well as other investigators found that a high FLT3-ITD allelic burden[22], as

TE D

determined by the mutant to wild type (WT) ratio, conveys a poor prognosis for AML patients. This ratio is an indirect measure of the number of FLT3-ITD positive and negative cells. Using SCA, the present study has succeeded to directly evaluate the percentage of WT and mutant cells, and

mutational state[21].

EP

accurately determine the burden of LOH of the WT allele which is also considered a high-risk

AC C

As to the technical aspects of single cell sorting and genetic analysis, we used serial dilution with microscopic verification for single cell isolation. However, other methods, such as FACS sorting or microfluidics based techniques are faster, more accurate and less prone to subjective interpretation[23]. Detection of FLT3-ITD mutations using the GeneScan device is more sensitive than standard PCR but is still not optimal. In addition, qualitative and quantitative analysis of the allelic burden might be imprecise as it involves human interpretation[24]. SCA may provide a highly sensitive alternative tool for determining the existence of mutations and their burden. The detection of micropopulations positive for FLT3-ITD cells will necessitate reevaluation of the prognostic effect of the ITD mutation and its burden in AML. The existence of FLT3-ITD positive cells in clinically determined (bulk DNA) FLT3-ITD negative patients provides direct evidence that this mutation appears late in

7

ACCEPTED MANUSCRIPT leukemogenesis. An early driver event in leukemogenesis must be present in almost all leukemic cells, while later subclonal events will accumulate at lower frequency and with time can expand if they are given a strong selective advantage. The fact that at AML diagnosis low frequency of FLT3-ITD can be detected in the majority of patients suggests that this lesion can introduce a selective advantage to already established leukemia, however FLT3-ITD is not an obligatory step in leukemogenesis.

RI PT

Furthermore, the existence of minor FLT3 ITD positive clones could have methodological and clinical implications. Studies evaluating FLT3 mutations as a parameter for risk stratification and management of leukemia usually utilized standard genomic PCR on bulk cells extracted from BM or PB. According to results of the present study, such an approach may not be sensitive enough. SCA is a laborious process that can be substituted by next-generation sequencing technologies which have the

SC

advantage of quantitative measurement of the fraction of FLT3-ITD positive cells on a large scale[23, 25]. A subsequent study with a larger number of single cells should be performed to further evaluate

M AN U

the sensitivity of this method, and compare it with the precise sensitivity of the deep sequencing technique. Given the importance of AML FLT3-ITD status in patients' molecular risk stratification and treatment, a small fraction of positive cells could be clinically important, and should be further evaluated in a larger sample set[26]. The significance of FLT3-ITD+ cells derived from adult ALL patients is unclear as it is a rare mutation and has not been shown to affect prognosis[1, 6]. Nevertheless, FLT3 over-expression may have a role in ALL leukemogenesis[5]. Our results also call into question the role of FLT3 inhibitors and the stage when they should

TE D

be employed. Utilizing these agents in a patient with a very low allelic burden may prevent the development of a poor-risk FLT3-ITD AML at relapse. Comparing the percentage of FLT3-ITD positive cells obtained from paired samples at diagnosis and relapse, revealed its instability, i.e., while in some cases the number of positive cells

EP

increased, it was found to decrease in others. This is consistent with earlier studies using standard diagnostic PCR [1, 27]. Whether this implies adaptation of leukemic cells to the environmental

AC C

pressure such as chemotherapy, which results in a gain of mutation associated with proliferative advantage (in cases where there was a increased percentage of positive cells at relapse) or operation of an unknown mechanism at relapse (in cases where there was a decreased percentage of positive cells at relapse) is yet to be determined. However, these data should still be taken with caution as more studies with a larger number of single cells should be performed. SCA has proved to be a useful tool in specifying the exact number of cells with LOH of the FLT3-WT allele, among FLT3-ITD+ cells. Concerns regarding a false allelic dropout, when using WGA on single cells, were raised in the past; however, Klein et al showed it to be a reliable method for LOH detection [24]. In the setting of a heterogeneous cell population, SCA may provide an alternative to high density

8

ACCEPTED MANUSCRIPT single nucleotide polymorphism (SNP) arrays for estimating LOH[28, 29]. The combination of the two methods can also be employed for result validation. LOH of the FLT3 WT allele is considered to be a recurrent non-random phenomenon in FLT3-ITD+ AMLs [28, 30]. While several mechanisms may underlie the acquired LOH, the one governing the FLT3-WT seems to be uniparental disomy [28-30]. We suggest that WT LOH may contribute to

RI PT

enhanced tumor survival in response to environmental pressures. FLT3WT/ITD cells dispensing of the WT allele will enjoy increased fitness, as demonstrated in knock-in mice [31]; on the other hand, dependence on FLT3-ITD signaling will occur, converting them susceptible to therapies, such as FLT3 inhibitors. The notion that WT FLT3 allele is practically a tumor suppressor has been supported in the study by Li et al, where FLT3−/ITD mice, that lost the WT allele, displayed a worse myeloproliferative

SC

neoplasm phenotype than the FLT3wt/ITD mice[32]. The development of such neoplasm in FLT3-ITD+ mice and the response of AML positive for the mutation, to FLT3 inhibitors [4, 33] suggest an active

M AN U

role of the FL3/ITD mutation in tumor evolution. Nevertheless, the high prevalence of a small fraction of FLT3-ITD+ single cells in leukemia patients, observed in our study, raises the question whether this mutation entails a selective advantage, or denotes a common "passenger" mutation in leukemia, representing a mutator phenotype[13].

The advent of next generation sequencing techniques and their potential clinical application could prove that a very low mutational burden is not an exclusive phenomenon for the FLT3-ITD mutation. To date, clinical molecular diagnostics have relied on average based methods (e.g., PCR on bulk

TE D

DNA). A shift in the diagnostic threshold for mutations such as FLT3-ITD would warrant elucidation of their pathophysiological role, prognostic meaning and stage for initiation of tyrosine kinase inhibitors.

In the current study, SCA revealed the existence of rare cells carrying the FLT3-ITD allele in patients

EP

considered negative for the mutation, which suggests that FLT3-ITD is a late event in

diagnosis.

AC C

leukemogenesis. Furthermore, this might explain why in some cases it is detected at relapse but not at

9

ACCEPTED MANUSCRIPT

Conflict of interest

AC C

EP

TE D

M AN U

SC

RI PT

The authors declare no conflict of interest.

10

ACCEPTED MANUSCRIPT Figure legends Figure 1. FLT3-ITD status in single leukemic cells Patients #2-8: Acute leukemias are considered to be FLT3-ITD negative by standard diagnostic PCR. Single cell analysis revealed a small fraction of FLT3-ITD positive cells in the majority of these patients.

Mixed Lineage Leukemia (MLL).

Figure 2. Single cell analysis of FLT3/ITD status in acute leukemias

RI PT

Diagnosis (D); Relapse (R); Acute Myeloid Leukemia (AML); Acute Lymphoblastic Leukemia (ALL);

SC

A. Example of PCR gel electrophoresis. Bulk DNA from patients #2 (Pt.2) and #3 (Pt.3) at diagnosis (D) is FLT3/ITD-. These patients carry FLT3/ITD+ single cells that are marked by (+), where (++)

M AN U

denotes FLT3/ITD+ with loss of heterozyosity (LOH) of the wild type FLT3 allele. B. A representative example of GeneScan plot of DNA obtained from patient # 3. Bulk DNA was used to amplify the FLT3 fragment. FLT3 products were then loaded to the GebeScan device and analyzed using gene mapper software. X axis represents length (bp) and Y axis represents frequency.

TE D

Figure 3. Representative example of HRM results for patient #1.

AC C

EP

Double pick denotes heterozygous state for FLT3-ITD/WT. Single pick denotes LOH.

11

ACCEPTED MANUSCRIPT

Table 1: Immunophenotyping

Immunophenotyping

1

CD13, CD15, CD33, CD34, CD38, HLA-DR, CD117

2

CD4, CD11B, CD33,CD38, DR, CD56,

3

CD13, CD15, CD33, CD38, DR, CD117

4

CD13, CD15, CD33, DR, CD117

5

CD11B, CD13, CD15, CD33, DR, CD117

6

CD2, CD4, CD38, HLA-DR

7

CD34, CD19, CD10, HLA-DR

8

CD15, CD19, CD22, CD38, HLA-DR

AC C

EP

TE D

M AN U

SC

RI PT

Patient #

12

ACCEPTED MANUSCRIPT Table 2. Oligos sequences Sequence

11F-FAM

5'-FAM-GCAATTTAGGTATGAAAGCCAGC -3'

12R

5’-CTTTCAGCATTTTGACGGCAAC -3’

AC C

EP

TE D

M AN U

SC

RI PT

Oligo name

13

ACCEPTED MANUSCRIPT

Table 3. Patient data summary

RI PT

Single cell analysis

Status Bulk DNA FLT3-ITD Status

Total FLT3-ITD+ cells LOH/FLT3-ITD+

1

AML

M1/M2

48XY +13+8,t(6:9)

D

Positive

28/84 (33.3%)

7/28 (25%)

1

AML

M1/M2

48XY +13+8,t(6:9)

R

N\A

1/31 (3%)

0

2

AML

M4/M5

Normal

D

Negative

1/19 (5.3%)

0

2

AML

M4/M5

Normal

R

N\A

0/59 (0%)

0

3

AML

Biphenotype

Normal

D

Negative

2/62 (3.2%)

1/2 (50%)

4

AML

M1/M2

Normal

D

Negative

1/50 (2%)

0

5

AML

M1/M2

48XY +2MARKERS D

Negative

0/21 (0%)

0

6

ALL

T-ALL

Normal

D

Negative

1/21 (4.8%)

0

6

ALL

T-ALL

Normal

R

Negative

2/25 (8%)

0

7

ALL

Pre-B ALL

Normal

D

Negative

0/17 (0%)

0

7

ALL

Pre-B ALL

Normal

R

Negative

1/15 (6.7%)

0

8

ALL

Pre-B ALL-MLL

4:11

D

Negative

4/29 (13.8%)

1/4 (25%)

8

ALL

Pre-B ALL-MLL

4:11

R

N\A

1/41(2.4%)

1/1 (100%)

AC C

EP

TE D

M AN U

SC

Pt. # Disease FAB Classification Cytogenetics

The table presents the leukemia type, subtype, FLT3-ITD status of bulk cells and single cells. The fraction of single cells with LOH of the WT allele amongst FLT3-ITD+ cells is displayed in the last column. FLT3-internal tandem duplication (FLT3-ITD); Loss of heterozygosity (LOH); Diagnosis (D); Relapse (R); Acute myeloid leukemia (AML); Acute lymphoblastic leukemia (ALL); Mixed lineage leukemia (MLL)

14

RI PT

ACCEPTED MANUSCRIPT

Table 4: Patient clinical data

Induction

Consolidation

Other mutations

1

37

X2

X1

Npm1CDF-

2

17000 (52% blasts) 53000

21

X1

X1

3

19000

34

X1

X3

4

8300

69

X2

X1

5 6 8

2300 100000 30000

63 30 49

X1 X1 X1

X1 X1 X1

FLT3NPM1CDFNPM1+ FLT3FLT3NPM1nd CDF-2 AML PHPH-

Transplant

Allo- SCT*

Allo- SCT

16

Auto-SCT**

18

No

9 5 3

No Allo- SCT Allo- SCT

EP

*Allo-SCT – Allogeneic stem cell transplantation;

Time from Dx to Rx*** (months) st 4 to 1 Rx nd and 6 to 2 Rx 3

SC

Age

M AN U

WBC

TE D

Patient No

*** Dx – diagnosis; Rx – relapse

AC C

** Auto-SCT – Autologous stem cell transplantation

15

ACCEPTED MANUSCRIPT References 1.

Kiyoi H, Yanada M, Ozekia K. Clinical significance of FLT3 in leukemia. Int J

Hematol 2005;82:85-92. 2.

Meshinchi S, Appelbaum FR. Structural and functional alterations of FLT3 in

acute myeloid leukemia. Clin Cancer Res 2009;15:4263-4269. 3.

Thiede C, Steudel C, Mohr B, et al. Analysis of FLT3-activating mutations in

RI PT

979 patients with acute myelogenous leukemia: association with FAB subtypes and identification of subgroups with poor prognosis. Blood 2002;99:4326-4335. 4.

Smith CC, Wang Q, Chin CS, et al. Validation of ITD mutations in FLT3 as a

therapeutic target in human acute myeloid leukaemia. Nature 2012;485:260-263.

Chang P, Kang M, Xiao A, et al. FLT3 mutation incidence and timing of origin

SC

5.

in a population case series of pediatric leukemia. BMC Cancer 2010;10:513. 6.

Xu F, Taki T, Yang HW, et al. Tandem duplication of the FLT3 gene is found

M AN U

in acute lymphoblastic leukaemia as well as acute myeloid leukaemia but not in myelodysplastic syndrome or juvenile chronic myelogenous leukaemia in children. Br J Haematol 1999;105:155-162. 7.

Shlush LI, Chapal-Ilani N, Adar R, et al. Cell lineage analysis of acute

leukemia relapse uncovers the role of replication-rate heterogeneity and microsatellite instability. Blood 2012;120:603-612.

Kottaridis PD, Gale RE, Langabeer SE, et al. Studies of FLT3 mutations in

TE D

8.

paired presentation and relapse samples from patients with acute myeloid leukemia: implications for the role of FLT3 mutations in leukemogenesis, minimal residual disease detection, and possible therapy with FLT3 inhibitors. Blood 2002;100:2393-

9.

EP

2398.

Testa U, Pelosi E. The Impact of FLT3 Mutations on the Development of

Acute Myeloid Leukemias. Leuk Res Treatment 2013;2013:275760. Kiyoi H, Naoe T. FLT3 mutations in acute myeloid leukemia. Methods Mol

AC C

10.

Med 2006;125:189-197. 11.

Altschuler SJ, Wu LF. Cellular heterogeneity: do differences make a

difference? Cell 2010;141:559-563. 12.

Ding L, Ley TJ, Larson DE, et al. Clonal evolution in relapsed acute myeloid

leukaemia revealed by whole-genome sequencing. Nature 2012;481:506-510. 13.

Welch JS, Ley TJ, Link DC, et al. The origin and evolution of mutations in

acute myeloid leukemia. Cell 2012;150:264-278. 14.

Walter MJ, Shen D, Ding L, et al. Clonal architecture of secondary acute

myeloid leukemia. N Engl J Med 2012;366:1090-1098.

16

ACCEPTED MANUSCRIPT 15.

Meshinchi S, Alonzo TA, Stirewalt DL, et al. Clinical implications of FLT3

mutations in pediatric AML. Blood 2006;108:3654-3661. 16.

Abu-Duhier FM, Goodeve AC, Wilson GA, et al. FLT3 internal tandem

duplication mutations in adult acute myeloid leukaemia define a high-risk group. Br J Haematol 2000;111:190-195. 17.

Smith CA, Williams GT, Kingston R, et al. Antibodies to CD3/T-cell receptor

RI PT

complex induce death by apoptosis in immature T cells in thymic cultures. Nature 1989;337:181-184. 18.

Hope KJ, Jin L, Dick JE. Acute myeloid leukemia originates from a hierarchy

of leukemic stem cell classes that differ in self-renewal capacity. Nat Immunol

SC

2004;5:738-743. 19.

Greaves M. Darwin and evolutionary tales in leukemia. The Ham-Wasserman

Lecture. Hematology Am Soc Hematol Educ Program 2009:3-12.

Gerlinger M, Rowan AJ, Horswell S, et al. Intratumor heterogeneity and

M AN U

20.

branched evolution revealed by multiregion sequencing. N Engl J Med 2012;366:883892. 21.

Parkin B, Ouillette P, Li Y, et al. Clonal evolution and devolution after

chemotherapy in adult acute myelogenous leukemia. Blood 2013;121:369-377. 22.

Pratcorona M, Brunet S, Nomdedeu J, et al. Favorable outcome of patients

TE D

with acute myeloid leukemia harboring a low-allelic burden FLT3-ITD mutation and concomitant NPM1 mutation: relevance to post-remission therapy. Blood 2013;121:2734-2738. 23.

Walter MJ, Graubert TA, Dipersio JF, et al. Next-generation sequencing of

EP

cancer genomes: back to the future. Per Med 2009;6:653. 24.

Klein CA, Schmidt-Kittler O, Schardt JA, et al. Comparative genomic

hybridization, loss of heterozygosity, and DNA sequence analysis of single cells.

25.

AC C

Proc Natl Acad Sci U S A 1999;96:4494-4499. Ley TJ, Mardis ER, Ding L, et al. DNA sequencing of a cytogenetically normal

acute myeloid leukaemia genome. Nature 2008;456:66-72. 26.

Man CH, Fung TK, Ho C, et al. Sorafenib treatment of FLT3-ITD(+) acute

myeloid leukemia: favorable initial outcome and mechanisms of subsequent nonresponsiveness associated with the emergence of a D835 mutation. Blood 2012;119:5133-5143. 27.

Cloos J, Goemans BF, Hess CJ, et al. Stability and prognostic influence of

FLT3 mutations in paired initial and relapsed AML samples. Leukemia 2006;20:12171220.

17

ACCEPTED MANUSCRIPT 28.

Raghavan M, Smith LL, Lillington DM, et al. Segmental uniparental disomy is

a commonly acquired genetic event in relapsed acute myeloid leukemia. Blood 2008;112:814-821. 29.

Tuna M, Knuutila S, Mills GB. Uniparental disomy in cancer. Trends Mol Med

2009;15:120-128. 30.

Gupta M, Raghavan M, Gale RE, et al. Novel regions of acquired uniparental

RI PT

disomy discovered in acute myeloid leukemia. Genes Chromosomes Cancer 2008;47:729-739. 31.

Kharazi S, Mead AJ, Mansour A, et al. Impact of gene dosage, loss of wild-

type allele, and FLT3 ligand on Flt3-ITD-induced myeloproliferation. Blood

32.

SC

2011;118:3613-3621.

Li L, Piloto O, Nguyen HB, et al. Knock-in of an internal tandem duplication

Blood 2008;111:3849-3858. 33.

M AN U

mutation into murine FLT3 confers myeloproliferative disease in a mouse model.

Stone RM, Fischer T, Paquette R, et al. Phase IB study of the FLT3 kinase

inhibitor midostaurin with chemotherapy in younger newly diagnosed adult patients

AC C

EP

TE D

with acute myeloid leukemia. Leukemia 2012;26:2061-2068.

18

AC C EP

TE D

M AN U

SC

Leukemic cells

RI PT

ACCEPTED MANUSCRIPT

Figure 1

ACCEPTED MANUSCRIPT

Figure 2

300

320

M AN U

SC

RI PT

A

340

360

B 8000 6000

TE D

4000 2000

AC C

EP

0

ACCEPTED MANUSCRIPT

RI PT

Figure 3

SC

LOH

AC C

EP

TE D

M AN U

Heterozygous

H2O