T-cellacutelymphoblasticleukemia (T-ALL) is an aggressive hematopoietic malignancy that is characterized by infiltration of the bone marrow with immature lymphoblasts that express T-cell surface markers (1). T-ALL accounts for 10%-15% of pediatric and 25% of adult acutelymphoblasticleukemia (ALL). T-ALL patients suffer from huge tumor burden, mediastinal enlargement and also have high risk of central nervous system (CNS) infiltration (1). These aggressive features in T-ALL make the treatment difficult and often result in poor prognosis. Thus, the cure rate of T-ALL had been lower than 10% for a long time due to high rate of relapse (1). Nowadays, various treatment approaches including aggressive-intensified combinational chemotherapy, radiotherapy, and stem cell transplants have improved the outcome of T-ALL treatment, with a 5-year event-free rate up to 75% in children and more than 50% in adults (1). However, the treatment of T-ALL remains a problem in the clinic due to resistance and relapse (1). Although the detailed mechanisms that contribute to poor outcome of the disease remain elusive, activating mutations in Notch1 were found in more than 50% of T-ALL cases, highlighting Notch1 as a key player in T-ALL (1, 2).
The abnormal chromosomal number in ALL defines distinct subtypes with different response to treatment. High hyperdiploid is a subtype defined based on cytogenetic nomenclature as chromosomal count between 47 and 57; the definition criteria are universally accepted. High hyperdiploid is one of the common childhood malignancies comprising 30% of all pediatric B cell–precursor ALL. Molecularly, high hyperdiploid ALL is characterized by massive aneuploidy (abnormal number of chromosomes), authenticating a nonrandom gain of chromosomes. For example, some or all of +X, +4, +6, +10, +14, +17, +18, and +21 and other trisomies have been reported. However, the pathogenetic phenomenon of chromosomal gains remains poorly understood, but it generally is believed that gene dosage effects are of significance (Chilton et al., 2014). Genetic abnormalities like driver fusion gene is not observed in the vast majority of high hyperdiploid ALL cases. However, there is a possibility that there is yet unidentified primary aberrations present due to the low resolution of most genetic screening techniques. Previously such concealed events have been reported in aneuploid tumors, for example, the identification of structural dysregulation resulting in rearrangements of cytokine receptor-like factor 2 (CRLF2) in a large number of ALL patients with Down syndrome and microdeletions leading to the transmembrane protease, serin 2 (TMPRSS2)/v-its erythroblastosis virus E26 oncogene homolog (ERG) hybrid gene in prostate cancer (Mullighan et al., 2009). Profiling of a fusion gene in high hyperdiploid ALL would be of prima facie clinical importance, which may perhaps simplify the diagnostic procedures and hence provide novel treatment options. Clinical features of high hyperdiploid ALL was associated with a relatively low WBC count and a B-cell precursor immunophenotype. The prognosis of five-year overall survival rates (OS) is close to 90%.
The switch from pro-proliferative to pro-differentiative pre-BCR signaling is succeeded by the suppression of SLC gene expression, the disappearance of pre-BCRs from the cell surface and the upregulation of the somatic recombination machinery to commence IgLC gene rearrangement at the two IgLC loci, κ and λ (Nemazee, 2017; Melchers, 2015). Morphologically this is reflected by the transition from large-cycling to small resting pre-B cells. IgLC gene recombination starts at the κ and proceeds to the λ gene loci until it yields an in-frame V to D gene segment rearrangement resulting in an IgLC capable of pairing with the preformed IgHC to assemble a mature BCR. Once expressed on the cell surface the BCRs antigen specificity determines the future fate of the developing B-cell. Due to the stochastic nature of IgHC and IgLC rearrangements a considerable amount of newly formed BCRs cross-react with self-antigens (Nemazee David, 2008). Binding of these autoreactive BCRs to their cognate autoantigen in the bone marrow typically elicits a strong receptor response followed by rapid receptor internalization, which signals the self-reactive B-cell to reinitiate somatic recombination at its IgLC gene loci. This process of receptor editing continues until it yields a non-self- reactive BCR specificity or until the death of the affected B-cells via apoptosis. Given the high frequency of autoreactive BCRs in the initial pool of BCR specificities (Wardemann et al, 2003) receptor editing contributes significantly to the final composition of the B-cell repertoire. In keeping with the fate of autoreactive B-cell progenitors that fail receptor editing, pre-B-cells lacking BCRs of any specificity are also removed from the pool of developing B-cells. Ultimately, the survival and positive selection of immature B-cells at the immature B-cell checkpoint requires a minimum signaling activity from the BCR, below the threshold for autoreactivity. This type of low- level BCR activation has been referred to as tonic BCR signaling. The nature of these signals, particularly as to whether they emerge as a consequence of the interaction of the BCR with low-affinity self-antigens or as a result of the random association and activation of BCR molecules on the cell surface or a combination of both remains to be determined (Monroe, 2006; Jianying Yang & Michael Reth, 2010).
federal states, representing a total base population of 33 million people, were included (Table 1). Patients age 15 or older with a primary diagnosis of ALL (ICD-10 code C91.0) in 1997–2006 and with mortality follow up through December, 2006 were included. Cancer topography, morphology, and behavior were originally coded in accordance with the International Classification of Disease for Oncology (ICD-O)-3 guidelines and later converted into ICD-10 using ‘IARCcrgTools’ . Patients with both B-cell and T-cellacutelymphoblasticleukemia are covered under this diagnostic code in ICD-10. For some registries, data were available starting from later years only. Cases both with and without preceding cancers were included. Because there were data quality issues for patients age 70+ in some of the German registries, only data for patients age 15–69 were included. In order to compare population level survival for ALL in Germany with survival in the United States (US), data from the Surveillance, Epidemiology, and End Results (SEER13) database were analyzed . The same inclusion criteria as for patients from the German cancer registries were applied for the same time period. The SEER13 database includes data from 13 regional cancer centers in the US, covering a population of about 39 million people. Centers are chosen for inclusion based on their high quality and epidemiologically interesting population groups. The SEER population is considered to be similar to the general US population with respect to most sociodemographic characteristics , although it may be more affluent than average and may have slightly higher than average survival for some cancers .
LCH is a rare, clinically heterogeneous neoplasm of immature dendritic cells (its manifestations range from isolated bone lesions to systemic disease with involvement of two or more visceral organs) that is most frequent in children (Jaffe et al., 2001). Development of LCH following T- ALL is extremely rare. Few brief reports have stated a relationship between LCH and systemic Juvenile Xantogranuloma (JXG) following T-ALL (Rodig et al., 2007; Perez-Becker et al., 2010). It is suggested that LCH/JXG are clonally related to T-ALL and present persistent expression of constitutively active NOTCH1. Perez-Becker et al. reported the case of a 5-year- old female that developed an aggressive JXG only 5 months after the diagnosis of T-cellacutelymphoblasticleukemia. They reported identical bi-allelic T-cell receptor-γ (TCR-γ) rearrangement in both neoplasms. Rodig and coworkers describe the case of a 3;7 year old patient who eighteen months after initial diagnosis developed an aggressive Langerhans cell Histiocytosis. Both neoplasms harbored not only the same (TCR-γ) rearrangement, but also identical, synergistic activating NOTCH1 mutations, affecting exons 27 and 34. Moreover, the analysis of 24 cases of LCH and Rosai-Dorfman disease in patients without prior history of T- ALL revealed no mutations. These reports highlight the unique nature of the mutations found in this work, emphasizing its exceptional character.
T-ALL cells have shown to directly interact with CXCL12-producing MSCs in a dynamic BM microenvironment, where leukemic cells were highly nonmotile and strongly associated with stromal cells [ 22 ]. Hawkins et al. investigated the migratory property of xenotransplanted primary human Tcellacutelymphoblasticleukemia (T-ALL) in NOD/SCID/γ mice using time-lapse intravital microscopy [ 23 ]. In contrast with our results, they demonstrated the significant enhanced migration velocity of chemoresistant cells with cell division events immediately after the third dose of VDL, suggesting that faster migration and a lack of long-lasting interactions with the BM microenvironment are conserved characteristics of chemoresistance in T-ALL [ 23 ]. However, this observation was made after a 2 day treatment with VDL. This is a critical difference, as it is possible that at this earlier time point, cytotoxic events are still occurring and not all cells are fully chemoresistant. In other words, a mixed cell population of viable drug-resistant cells and dying cells may have been studied instead of a predominantly drug-resistant cell population. By contrast, in our model, 7 days of chemotherapy were chosen, the timepoint where viability stops decreasing and a state of surviving, predominantly drug-resistant cells is achieved. This may be one explanation for the difference to our results. Of note, it was also shown using intravital imaging that AML cells are migratory, and in contrast with T-ALL, chemoresistant AML cells become less motile [ 24 ]. Further studies are warranted to distinguish motility differences in different subtypes of leukemia and different drug treatments to further elucidate the effects on motility.
67 by a p53-mediated mechanism . The tumor suppressor gene tp53 can upregulate BTG2 and downregulate cyclin D1, thus leading to a cell cycle arrest at G1/S-phase via pRB-pathway [52,84,100]. Further, BTG2 can induce downregulation of cyclin E and thus cause a cell cycle arrest in the absence of p53, Rb and cyclin D1 . However, our RNA seq data has not revealed any significant changes in any cyclin molecule expression. Thus, the underlying mechanism of BTG2 upregulation after MEIS2-knockdown in acutelymphoblasticleukemia has still to be investigated. BTG2 induction appears to be not cell-cycle-dependent, but rather after a stimulus such as differentiation or DNA damage . BTG2 is induced by cAMP and the promotor of BTG2 is found to be enriched with cyclic AMP response elements (CRE) , a noteworthy observation considering that following MEIS2 knockdown, RNAseq analysis showed a significant upregulation of CREB5 (cAMP response element binding protein 5), a transcription factor which belongs to the CRE (cAMP response element)-binding protein family . As the name already suggests CREB proteins bind to cAMP-response-elements (CRE) within target genes. CREB5, which is highly homologous with CRE-BP1, binds to CRE as a homodimer or a heterodimer with CRE-BP1 and exerts functions as a CRE- dependent trans-activator . Once CREB proteins are activated, the transcription of different genes involved in cell growth and survival as well others such as proto- oncogenes or cell cycle genes (cyclin A1 and cyclin D2) is induced [2,28,141,163]. Interestingly, CREB showed a two-three fold overexpression in primary AML cells and CREB5 overexpression was found in a subgroup of ALL with RCSD1-ABL1- fusion protein [18,134]. In the study by Shankar et al.  knockdown of CREB by RNAi led to suppression of cell proliferation and survival. Surprisingly, we observed a reduced cell proliferation and a cell cycle arrest although the expression of CREB5 increased several-fold after MEIS2-knockdown. CREB exerts its functions through many target genes. Impey et al. could identify BTG2, a cAMP regulated gene, as a CREB target .
One of the hyperdiploidy cluster, i.e. cluster ‘HD4’ comprised only three cases characterized by the highest number of chromosome gains in this study (median 11) and a pattern that was distinct to the other two high hyperdiploid clusters HD2 and HD3 (Fig 2A and 2B). Most of the classical chromosomes were involved, however, gains of chromosomes 4 and 17 were absent or less frequent and, instead, non-classical gains of chromosomes 5, 11, 19, and 22 were present in all three cases (Fig 2B). Furthermore, all patients harbored a TP53 mutation (Fig 2C). Due to the unusual chromosome pattern and the presence of TP53 mutations we suspected that these relapses may have descended from a hypodiploid leukemia. Subsequent CytoScan HD array analysis of these cases showed copy-neutral LOH in nearly all disomic chromosomes, thus confirming the hypodiploid origin. As no hypodiploid founder clone could be detected in any of the three cases by DI or FISH (Table 1B), this cluster was termed
The successful clinical application of new therapeutic approaches asks for the evaluation in reliable preclinical models. Non-obese diabetic/ severe combined im- muno-deficiency (NOD/ SCID) mice are characterized by the absence of functional T- and B-cells, reduced NK-cell activity and a defective complement system [13,121]. Leukemia cells are isolated from diagnostic samples from patients with BCP-ALL and are transplanted onto NOD/SCID mice by intravenous injection. This model enables the engraftment of human leukemia cells in the recipient mice which can be confirmed by a high proportion of human leukemia cells in the peripheral blood or organ bone marrow and spleen using flow cytometry. Importantly, leuke- mia cells engrafted in mice maintain the original phenotypic and genotypic charac- teristics of the primary BCP-ALL [7,13,79]. Our group has identified different en- graftment phenotypes of patient-derived ALL samples. A short period of less than 10 weeks after transplantation until manifestation of the disease in the recipient animals (time to leukemia short, TTL short ) was found to be associated with poor
The aim of this study was to assess the expression of protein tyrosine kinases (PTK) in acutelymphoblasticleukemia (ALL) patient samples at the protein level. Xenotransplanted primary ALL blasts and ALL cell lines were used as model system for the functional analysis of the role of PTK in ALL. The analysis revealed that Lyn, a member of the Src family kinase (SFK), was prominently expressed in a subgroup of ALL patient samples. To further investigate the biological consequence of elevated Lyn expression in ALL cells, Nalm6 and CALL3 cells were used as a model which recapitulated the high and low Lyn expression profile observed in patient specimens, respectively. Lyn is known to be associated with pre-BCR after receptor crosslinking. Analysis of the functional role of Lyn upon shRNA mediated Lyn repression and pre-BCR crosslinking showed that phosphorylation of downstream signaling proteins was strikingly reduced or delayed in Nalm6 cells. In addition, cell proliferation was substantially reduced in Nalm6 Lyn-knockdown cells. Conversely, an increase in the tyrosine phosphorylation was found in CALL3 Lyn-knockdown cells. Membrane microdomain, called lipid rafts, were shown to concentrate and regulate SFK 111 . However, data of the Lyn localization in the plasma membrane indicates that, whereas Lyn was exclusively present within defined lipid rafts in CALL3 cells, the protein was aberrantly localized all over the membrane in Nalm6 cells. The Lyn mislocalization was likely independent of lipid rafts and it could enable Lyn to interact with other proteins located outside rafts structures and promote its activation. Ultimately, preliminary data suggests that overexpression of Lyn is implicated in resistance to tyrosine kinase inhibitor (TKI) treatment.
Materials and Methods leukemia such as severely impaired general condition, lethargy, ruffled fur and impaired posture. Upon clear evidence for leukemia related morbidity, mice were killed and autopsy was performed. At autopsy cell suspensions from spleen and bone marrow were prepared. Spleen tissue was minced and passed through nylon cell sieves (70 µm) followed by ficoll gradient centrifugation for 20 minutes at 1300 rpm at room temperature. Bone marrow cells were isolated flushing the femoral cavity with HBSS. Cell suspension from spleen and bone marrow were stained with allophycocyanin (APC)-conjugated anti-human B-cell antigen (CD19), Percp conjugated human leukocyte common antigen (CD45) and phycoerythrin (PE)-conjugated anti-murine leukocyte common antigen (Ly-5) for 20 minutes at 4º C followed by washing with BSA washing solution. Blood staining was done according to the same protocol described in the previous section. Cells have been acquired by BD LSRII multiparameter flow cytometry. Time to leukemia (TTL) was determined for each patient sample transplanted as weeks from the date of transplant to the date of clinical manifestation of the disease.
Cancer can arise in nearly every tissue and affect every organ in the human body, which results in a wide spectrum of different tumor subtypes with independent underlying processes of tumor formation. Tumor development is characterized by accumulation of consecutive gene mutations (Nordling C. O. 1953) whereas Knudson postulated in 1971 that two independent mutation events, either of germline or somatic origin, are needed to convert healthy into carcinogenic cells (Knudson, JR 1971). These events either lead to gain of function products (proto- oncogenes) resulting in increased cell proliferation or in loss of function products of tumor suppressors which further promote dysregulation of cell cycle processes or cell death evading mechanisms (Croce 2008; Hanahan, Weinberg 2000; Sherr 2004). These mutations can be either inherited when they affect the reproductive cells or occur in somatic cells as an acquired mutation (Lichtenstein et al. 2000; Peto 2001). A great number of these acquired mutations occur spontaneously or were induced upon environmental factors like chemicals, radiation or intracellular ROS-species (Martincorena, Campbell 2015). Impaired DNA-repair mechanisms or mutations that remain undetected, subsequently have an impact on the DNA structure, resulting in either chromosomal changes or an altered gene sequence and lead to different protein formation with a gain or loss in protein function (Martincorena, Campbell 2015; Weinberg 2007).
activation and proliferation of the engaged T-cells and lysis of the tumor cells . It is already approved by the Food and Drug Administration (FDA) for ALL therapy . Blinatumomab was able to induce hematologic and molecular remissions in 69% of enrolled adult patients with relapsed or refractory B-cell precursor ALL , leading to long term survival and MRD negativity in 28% of patients . Even in pediatric patients Blinatumomab induced complete remissions in 6 out of 9 (67%) patients with a posttransplant B-precursor ALL relapse, a situation with no further standard of care therapy . Another trial in relapsed or refractory pediatric ALL showed 32% complete remission with 24% of patients reaching MRD negativity . Adverse reactions were severe cytokine release syndrome and neurological toxicities . For pediatric patients Blinatumomab is currently being tested in the setting of high risk first relapse B-precursor ALL in the IntReALL study group . Another intriguing option are chimeric antigen receptor T-cells (CAR-T cells) directed at CD19, expressed on precursor B-cell ALL cells . In this approach a patients’ own T-cells are genetically modified in vitro to express an artificial T-cell receptor, named chimeric antigen receptor (CAR). These receptors couple an anti- CD19 domain to intracellular signaling domains CD137 (a costimulatory receptor) and CD3-zeta (a signal-transduction component) of the T-cell receptor, in that way redirecting cytotoxic T-cells against ALL blasts [55, 127]. Initial studies treating adult and pediatric patients with relapsed or refractory precursor B-cell ALL report 100% (5 out of 5 patients)  and 90% (27 out of 30 patients)  complete remissions. Side effects included cytokine release syndrome and transient neurologic toxic effects, ranging from seizure to encephalopathy.
express these target-antigens. Finally, we tested the hypothesis that these drugs can interfere with leukemic engraftment of LSCs in NSG mice. Since our data showed that both, CD34 + /CD38 − and CD34 + / CD38 + ALL cells produce engraftment, we used Tcell-depleted MNCs (bulk of CD34 + cells) in these experiments. However, neither GO alone nor alemtuzumab alone were able to inhibit ALL engraftment in NSG mice, suggesting that some of the LSCs in these cell-fractions were resistant or did not express these targets. On the other hand, a combination of both drugs was found to substantially inhibit engraftment of ALL cells in NSG mice. These data suggest that CD33 and CD52 are variably expressed in sub- fractions of ALL LSCs, and that combined targeting is required to block LSC engraftment. An alternative explanation would be that ALL LSCs can recover rapidly from drug effects in vivo in mice unless multiple drugs are applied. We therefore transplanted the CD52 + ALL cell line BL-41 into NSG mice and asked whether continuous treatment with alemtuzumab is able to interfere with ALL expansion. Indeed, alemtuzumab i.p. injection three times a week led to a significant decrease in engraftment. However, again, no total depletion of ALL cells was achieved. Together, these data suggest that targeting through defined surface antigens may be a promising approach to attack LSCs, but these drugs have to be combined with each other or with other drugs to achieve optimal anti-neoplastic effects.
Figure 1. Unsupervised hierarchical clustering of gene expression data from bone marrow samples of 31 children with acutelymphoblasticleukemia. Cluster dendrogram based on 333 fil- tered probe sets (see methods for filter criteria). The upper main branch includes precursor B cell leukemias, whereas the lowest branch identifies two T-cell leukemias. The left column shows the patient number, cytogenetic characterization is listed in the middle column, risk group assignment is pre- sented in the right column. The dendro- gram did not change substantially when more (or even all) probe sets were used (results not shown). *normal indicates DNA index of 1 and no TEL-AML1, BCR- ABL, E2A-PBX1 or MLL-AF4 rearrange- ment.
To amplify leukemic blasts from acuteleukemia patients, peripheral blood or bone marrow aspirates were injected into 6 to 8 weeks old NSG mice via the tail vein (in 250 µl autoclaved and sterile filtered PBS). After injection, mice were treated with ciprofloxacin which was added to the drinking water in order to prevent infections. For expansion of PDX cells, freshly thawed PDX cells were injected into the tail vain of NSG mice (in 250 µl autoclaved and sterile filtered PBS). All animals were maintained under specific pathogen-free conditions in the research animal facility of the Helmholtz Zentrum München. Mice were sacrificed when blood measurement indicated leukemia disease or as soon as they showed any clinical signs of illness and human cells were isolated out of spleen or bone marrow subsequently. The spleen was homogenized through a 70 µm cell strainer and cells were purified using Ficoll gradient centrifugation (400 g, 30 min, rt, without rotor brake). After centrifugation, mononuclear cells could be harvested as a layer at the interphase. Cells were washed twice with PBS and once with patient medium (RPMI supplemented with 20% FCS, 1% pen/strep, 1% gentamycin and 2 mM glutamine) (400 g, 10 min, rt). After washing, cells were re-suspended in patient medium and stored at 37 °C for further use. Isolated bones were crushed in a porcelain mortar and suspended in PBS. The suspension was filtered through a 70 µm cell strainer and washed twice with PBS. Cells were re-suspended in patient medium and stored at 37 °C, respectively. Accuracy of sample identity was verified by repetitive finger printing using PCR of mitochondrial DNA (Hutter et al., 2004).
activity. However, in that respect the NUP214/ABL1 fusion differs from the other ABL1 fusion, in that the coiled-coil motifs of NUP214 do not mediate the oligomerization of the NUP214/ABL1 protein, but mediate the interaction of NUP214/ABL1 with other nuclear pore proteins. This NUP214/ABL1 nucleo pore interaction is required to achieve the full NUP214/ABL1 transforming potential (De Keersmaecker et al., 2008). Neverthless, considering these two basic requirements for an ABL1 fusion partner, one would expect many genes in the cell to fulfil these two requirements. Thus, it is very surprising that there have only been 6 fusion partners of ABL1 identified till now. This indicates, that the fusion partners of ABL1 are required to have some other unknown important properties. It could be that they have to be able to activate some specific signalling pathways. For example, in addition to the coiled-coil domain, BCR contains a GRB2 binding site (Tyr177), which is required for the maximal transformation potential of BCR/ABL1 by activating the RAS and the GAB2/PI3K/AKT signalling pathways (Pendergast et al., 1993; Puil et al., 1994; Goga et al., 1995; He et al., 2002; Sattler et al., 2002). The ability of NUP214 to localize at the nuclear pore complex via its coiled-coil domains and its interaction with other nuclear pore proteins is a prerequisite to achieve the full transforming potential of NUP214/ABL1 (De Keersmaecker et al., 2008). There is also evidence that some of the fusion partners of ABL1 have tumor suppressor gene functions. This is known for example for the ETV6 gene and, as explained below, might also be the case for SHIP1.
tion analysis. Using cytometric bead array we could show that stimulated BCP-ALL cell lines secrete TNF-α and IL-12, hence creating an immunogenic milieu that pro- motes the maturation and priming of T cells. To learn more about the mechanisms of the synergistic eect of CpG B with IL-4 and CD40L, PCR of TLR9 mRNA (messenger ribonucleic acid) expression was performed. We observed an increase of TLR9 mRNA after stimulation with CD40L and IL-4, which may explain the additional stimulatory eect of CpG. On the other hand, upregulation of CD40 by CpG increased responsive- ness to soluble CD40L. The triple combination furthermore proved to be signicantly more eective in inducing anti-leukemic cytotoxic T lymphocytes (CTL). Importantly, such CTL expressed antileukemic cytotoxicity not only against treated but also against untreated BCP-ALL cells. We could show that this eect is specic by using third party BCP-ALL cells. With the long-term therapeutic aim to enable transfer of antileukemic CTL to patients, we tested our ndings in an autologous setting with BCP-ALL cells expanded in NOD/SICD mice and PBMC derived from the same patient at least 6 months after chemotherapy. We observed the same tendency, even though at a lower level.
pre-isolated RNA as input, but can be applied to several types of samples including cell lysates from sorted cells or crude lysates from frozen cell stocks. The isolation of RNA from samples is mostly performed using commercially available kits utilizing silica membrane columns. While these are suited to extract large amounts of high quality RNA from cells, they are time consuming and cost intensive. Second, due to utilizing early barcoding it is possible to integrate thousands of samples which is furthermore facilitated by the low costs of around 3 € per sample (excluding sequencing). This allows one to incorporate more samples into each experiment in order to increase the power to identify differential expression between groups and additionally makes the method highly suited to investigate large patient cohorts. Lastly, the demand for large amounts of input material for previous methods restricts their usage to characterize rare subtypes, yielding only a few hundred cells per biological replicate (Alpern et al., 2019). Being capable of generating high quality RNA-seq data from as little as 200 cells per sample, is a strong benefit of our approach, as due to the subclonal structure of AML, the further analysis of these subclones could promote a deeper understanding in treatment resistance and relapse emergence. However, the integration of those features also comes with some caveats. For example, due to the integration of UMIs the method enriches for the 3’ end of transcripts. Hence, the identification of transcript isoforms and mutations as well as allelic expression within samples based on the sequencing data is strongly limited. Both of these could be beneficial to understand the molecular pathways involved in leukemia (Batcha et al., 2019; Li et al., 2014; Petti et al., 2019).