Leukemia Samples Research Articles

Characteristics and Outcomes of Therapy-Related Acute Myeloid Leukemia: Results of Retrospective Analysis of 116 Adult Patients

Background:Therapy related acute myeloid leukemia (t-AML) is a late complication following cytotoxic therapy for a primary neoplasm or a non-neoplastic disorder as autoimmune diseases and a highly fatal complication in patients treated for first primary malignancy. Therapy related AML are considered to have a worse prognosis and inferior outcome compared with de novo AML. The aim of this retrospective analysis is to describe clinical and biological characteristics and outcomes of these poor prognosis patients. Methods:This retrospective analysis include 116 adult patients in 2 institutions (Leon Berard cancer center and Lyon Sud university hospital) between January 2006 and June 2019 treated with chemotherapy and/or radiation for different previous malignancies and who subsequently developed AML. We analyzed demographic characteristics, parameters related to previous malignancies (type, delay until occurrence of AML and treatment), to AML (cytogenetic and mutational profile, treatment and outcome). Event for overall survival (OS) was death and patients were censored at the date of last contact if alive or at the date of allogeneic HSCT. Event for disease free survival were death and first relapse. Results:We analyzed retrospectively, 116 adult patients (57 male and 59 female) with median age of 67.5 [19 to 87] years diagnosed with t-AML according to 2016 revised WHO criteria and caused by exposure to chemotherapy and/or radiotherapy for previous solid tumors in 81 pts (70%) or hematologic malignancies in 35 pts (30%). Median time between diagnosis of previous neoplasm and AML was 5.4 [0.4-34.3] years. For the treatment of previous cancer, 48 pts (41%) received chemotherapy alone, 17 pts (15%) radiotherapy and 51 pts (44%) received both modalities. T-AML occurred after exposure to alkylating agents in 12 pts (12%), to agents targeting topoisomerase II in 7 pts (7%), to both agents in 65 pts (66%) and other treatment in 15 pts (15%). Cytogenetic profile was performed in 108 pts and was favorable in 4 pts (3.7%), intermediate in 46 pts (42.8%), unfavorable in 47 pts (43.5%) and assessment failed in 11 pts (10%). Eighty eight available leukemia samples were analyzed using molecular RT-PCR and, more recently, NGS profiling. Gene mutations concern 8 pts for FLT-3, 20 pts for Evi1, 1 pt for IDH1, 1 pt for IDH2 and 8 pts were TP53 mutated. Treatment of t-AML consisted in intensive chemotherapy (IC) in 59 pts (51%), hypomethylating agents (HMA) or low dose cytarabine in 29 pts (25%), best supportive care (BSC) in 28 pts (24%). Only 15 pts underwent allogeneic hematopoietic stem cell transplantation. Median overall survival (OS) [95% CI] was 7.75 months [5.55-12.32] and median progression free survival (PFS) [95% CI] was 6.14 months [5.22-9.07] in whole cohort. According to cytogenetic risk group and as expected, the median survival is significantly longer in favorable than in intermediate and unfavorable risk group (45.3 vs 15.2 vs 5.4 mths for OS with p = 0.005 and 45.3 vs 10.3 vs 5.3 mths for PFS with p = 0.007). Conclusion:This descriptive analysis confirm data of literature with poor prognosis and worse survival in unfavorable and intermediate risk groups of t-AML. However, pts with no comorbidities and candidate to IC had a significant better survival than those receiving HMA or BSC and should be considered for allogeneic transplantation Figure Disclosures No relevant conflicts of interest to declare.

Analysis of Genomic Landscape of Large Granular Lymphocyte Leukemia Reveals Etiologic Insights

Introduction: Large Granular Lymphocyte (LGL) leukemia is a rare lymphoproliferative disorder characterized by clonal expansion of either CD3+ cytotoxic T cells expressing T cell receptor (TCR) alpha beta or CD3- natural killer (NK) cells. A less frequent CD3+ T cell subtype expresses TCR gamma delta (GD). Here, we present the molecular landscape of known LGL types (T, NK, GD) from analysis of the largest patient cohort assembled to date. An integrative analysis of genomic datasets from all LGL subtypes is necessary to more precisely define the shared and unique etiology of this rare disorder. Methods: We collected paired saliva and PBMC samples and related clinical information from 116 LGL leukemia patients after informed consent. The assembled cohort consisted of 93 T-LGL, 11 NK-LGL, and 12 GD T-LGL patients. Genomic analyses were performed on the leukemic (PBMC) and germline (saliva) samples after whole exome sequencing (WES) and transcriptome sequencing (RNAseq, PBMC only). Results: Somatic mutations were detected in the previously described potential drivers STAT3 (n=56), TNFAIP3 (n=9), and PIK3R1 (n=4). We also identified somatic mutations in CDH8 (n=3) and CCL22 (n=4), which we postulate as putative drivers based on mutational clustering. CDH8 was mutated in all three LGL subtypes, but CCL22 somatic mutations were only observed in NK-LGL patients. We observed that STAT3 and CCL22 together account for 64% (7/11) of NK-LGL cases. STAT3 is the most recurrently mutated gene in LGL leukemia, yet concurrent molecular and clinical features are incompletely defined. Interestingly, patients with STAT3 mutations have a higher mutational burden (P=0.0006) compared to those with wild-type (WT) STAT3. This effect is independent of the age of the patient, which correlates with the mutational burden (R=0.26, P=0.0039) and agrees with the finding that the dominant mutational signatures in this cohort exhibit clock-like properties. We also observed that patients with STAT3 mutations are enriched (P=0.0273) for additional mutations in chromatin modifier enzymes such as KMT2D, TET2, DNMT3A, and SETD1B (Figure 1). We found that ~10% of the samples exhibit broad somatic copy-number aberrations, and a patient with somatic mutations in STAT3 and KMT2D displayed high-level microsatellite instability. STAT3 mutations were also significantly associated with increased expression of genes involved in apoptosis, complement activation, and interferon cytokine signaling compared to STAT3 WT (FDR < 0.05). Early-onset LGL patients with age 51 years or less (n=28), as defined by the bottom quartile of the cohort, displayed no differential enrichment of the somatic driver genes. Interestingly, the age of the patient was significantly associated with absolute neutrophil counts (ANC) (P = 0.0068), with younger patients exhibiting lower neutrophil counts, even after adjusting for the presence of STAT3 mutation, as it is associated with lower ANC. As neutropenia is a hallmark feature of LGL leukemia and often a trigger for initiating therapy, the association of young age with lower neutrophil counts and lower somatic mutational burden suggests other mechanisms may be involved. Focusing on the germline variants, we found that 17 of the patients (14.6%) had at least one pathogenic or likely pathogenic germline variant with known oncogenic association as annotated using CharGer. 5 patients had pathogenic mutations in known tumor suppressors including FANCC (n=1), BRCA1 (n=1), PALB2 (n=1), MUTYH (n=1), and SDHA (n=1), while 1 patient had pathogenic mutations in ALK, a known oncogene. Conclusions: We report on the genomic analyses done on whole exome and RNA-seq data from the largest cohort assembled for LGL leukemia to date. We show that the presence of STAT3 mutation is significantly associated with an increase in mutation burden and additional somatic mutations in chromatin modifiers, hinting at potential pathogenic mechanisms within STAT3 mutated patients. By combining LGL subtypes in our analysis, we were able to identify CDH8 as a putative driver that is present in T, NK, and GD subtypes. Additionally, we report CCL22 mutations specific to NK-LGL leukemia, however did not detect any subtype specific mutations in GD T-LGL. We found that about 15% of the patients carry at least one pathogenic germline variant with known oncogenic associations. These findings highlight emerging etiologic insights into this rare disorder. Disclosures Feith: Kymera Therapeutics: Membership on an entity's Board of Directors or advisory committees. Loughran:Keystone Nano: Membership on an entity's Board of Directors or advisory committees; Bioniz Therapeutics: Membership on an entity's Board of Directors or advisory committees; Kymera Therapeutics: Membership on an entity's Board of Directors or advisory committees; Dren Bio: Membership on an entity's Board of Directors or advisory committees.

Biomarker-Driven Phase Ib Clinical Trial of OPB-111077 in Acute Myeloid Leukemia Increases Overall Response Rates

Background: Phase I clinical trial for oncology drugs have the highest clinical failure rate. Drug candidates may be discarded if they don´t show activity. Yet these drugs may have valuable activity in patient subgroups. Biomarkers attempt to identify these sensitive subpopulations, but are normally evaluated and validated in Phase II when good drugs may have been discarded. OPB-111077 is a novel, oral, low-molecular-weight compound that inhibits STAT3 and mitochondrial electron transport. A prior Phase I trial on 145 patients from all tumor types gave only 1 partial response (PR). A strong anti-proliferative activity was identified in Acute Myeloid Leukemia (AML) samples ex vivo. This new biomarker was used for patient selection in this trial. We present a novel design of a Phase Ib trial directly selecting patients with a potential biomarker, that can benefit the patient without increasing toxicity risks, in Relapsed/Refractory (RR) AML patients. Method: This is an open-label, phase Ib dose-escalation clinical trial to evaluate the safety profile, maximum tolerated dose (MTD) and preliminary efficacy of oral OPB-111077 in AML relapsed or refractory to chemotherapy patients. Patients > 18 years old with high risk AML, ECOG ≤2 and adequate organ functions were selected for biomarker screening. Biomarker consisted on a dose response curve for antiproliferative activity of agent in a bone marrow sample of patients incubated 72-96-120h (Figure), quantified as the Area Under the Curve (AUC). Patient samples whose ex vivo biomarker showed low sensitivity (70% worst) we excluded, since they were potentially predicted as resistant to the agent. The remaining 30% with higher sensitivity were included. OPB-111077 was administered orally on a once daily dose schedule in 28-day cycles until intolerable toxicity or disease progression. Two dose schemas were evaluated: 200 (DL1) and 250 mg/day (DL2). Dose limiting toxicities were any Grade ≥ 3 non-hematologic toxicity and any unexpected non-tolerable grade II that requires delay beyond 1 week until recovery and evaluated during the first 28 days of treatment. Adverse events were graded according to NCI-CTCAE vs 4.03 and response criteria was based on those given by Cheson et al. IC50 and area under the curve (AUC) determined by the Vivia ex vivo biomarker were correlated with clinical response. Results: We screened 47 RR AML patients, twelve patients were included and selected by personalized medicine test (5 in DL1 and 7 in DL2), median age 76 years, 92% men, 42% ECOG 0 were included in the study. AML patients were refractory in 41.7% and patients were in median of relapses 2 (range 1-6). No DLT was reported. Adverse events related to study treatment were reported in three patients, all G 1-2: in DL1 one patient had vomiting; in DL2 one patient had extrasystoles and other had anorexia, diarrhoea, epigastric discomfort, nausea and vomiting. Two patients died during study treatment due to disease progression and respiratory infection. Half of the patients discontinued study treatment due to disease progression and 25% due to adverse events (respiratory failure G5, respiratory infection G5 and extrasystoles G2). Over the 6 patients evaluable for efficacy, 3 of them achieved PR and 3 treatment failure as best response. Progression free survival and overall survival were 57 days (95% CI: 37 - 77) and 95 days (95% CI: 27 - 163) respectively. The biomarker AUC and IC50 values stratified patients consistent with their clinical response, although without enough sampling potency for statistical significance. Figure shows the relative position of patients included in the study in the context of the overall population of samples considered to build up the mixed effect population pharmacodynamic model used to analyze samples results and define inclusion criteria. In green those that correspond to patients who showed a positive response; in orange those who failed. Conclusions: OPB-111077 is generally well tolerated and has a manageable toxicity profile. The MTD was not reached. The 50% PR rate achieved by OPB-111077, albeit in a few patients, is substantially higher that the 0.7% (1/145) PR rate achieved in a prior phase I trial on all tumor types1. This innovative Phase Ib design selecting patients using a biomarker may enable to rescue drug failures in the future.

Etoposide, Cytarabine and Mitoxantrone- or Fludarabine, Cytarabine and Granulocyte Colony-Stimulating Factor-Based Intensive Reinduction Chemotherapy Is Recommended for Children with Relapsed Acute Myeloid Leukemia: The Results from the Japanese Pediatric Leukemia/Lymphoma Study Group (JPLSG) AML-05R Study

Background:The current event-free survival rates of children with acute myeloid leukemia (AML) approach 60%, with hematopoietic cell transplantation (HCT) as an effective second-line therapy and improvements in supportive care; however, the overall survival rates of children with relapsed AML still remains 70%, which is an unsatisfactory situation that urgently needs to be solved. Patients & Methods:In order to identify prognostic factors and improve their prognosis, we analyzed 111 patients with relapsed disease after treatment with the Japanese Pediatric Leukemia/Lymphoma Study Group (JPLSG) AML-05 protocol, who were registered in the retrospective JPLSG AML-05R study. We collected the following data: age at relapse, time from the diagnosis to relapse, site of relapse, FAB classification, chromosomal analysis results, reinduction chemotherapy regimen, rate of achieving a second complete remission (CR2) after initial reinduction therapy, detailed information on HCT, outcome and cause of death. Moreover, ninety-three leukemic samples obtained from 111 patients at the time of their diagnosis were available for a genetic analysis. Mutational analyses ofFLT3-ITD,KIT, N-andK-RAS, NPM1,WT1,andKMT2A-partial tandem duplication were performed. Screening ofNUP98-NSD1was also performed. The high or low expression ofMECOMandPRDM16were determined based on theABL1ratio. Results:The 5-year overall survival rate was 36.1%. t(8;21) was found in 27 patients and inv(16) was found in 5 patients.KMT2Arearrangement was found in 23 patients (24.7%).KITmutations were found in 17 patients (17.5%). The following mutations were also detected:FLT3-ITD (n=10),N-RAS(n=10),WT1(n=7),K-RAS(n=4),NUP98-NSD1(n=2),KMT2A-PTD (n=2), andNPM1(n=2). The high expression ofMECOMandPRDM16was found in 24 (25.8%) and 32 (34.4%) patients, respectively. A genetic analysis revealed the prognostic significance ofFLT3-ITD as a poor prognostic marker (p=0.04) and core binding factor-AML, t(8;21) and inv(16), as a good prognostic marker (p<0.01). The five-year overall survival rate in the AML-05 risk groups were as follows: high risk, 14.7%; intermediate risk, 32.3%; and low risk 61.7% (p<0.01). The major determinant of survival was duration from the diagnosis to relapse. The mean duration in the non-surviving group (10.1±4.1 months) was shorter than that in the surviving group (16.3±8.3 months) (p<0.01). Moreover, achieving CR2 prior to HCT was associated with a good prognosis (p<0.01). "Etoposide, cytarabine and mitoxantrone" (ECM)- or "fludarabine, cytarabine and granulocyte colony-stimulating factor" (FLAG)-based regimens were therefore recommended for reinduction therapy (p<0.01). Conclusions:Achieving CR2 prior to HCT by intensive reinduction chemotherapy, ECM, or FLAG, is important for improving the prognosis of patients with relapsed pediatric AML. Liposomal daunorubicin or 3 fractionated doses of gemtuzumab ozogamicin is an attractive option which is under consideration for addition to FLAG. Moreover, recent molecular targeted therapeutics, such as FLT3 inhibitors, may contribute to improving the prognosis of these patients. Larger prospective investigations are needed in order to establish individualized treatment strategies for patients with relapsed childhood AML. Disclosures No relevant conflicts of interest to declare.

Novel Alkylating Agent Melflufen Displays Potent Efficacy in Plasma Cell Leukemia Samples and Other High-Risk Subtypes of Multiple Myeloma

Introduction: Despite the progress that has been made for standard risk multiple myeloma (MM), subsets of patients with the most advanced and aggressive plasma cell dyscrasias still suffer comparatively poor outcomes. One example is plasma cell leukemia (PCL), which carries a median overall survival of under two years. For patients with PCL, response to front line therapy occurs but is often short-lived, ultimately giving way to aggressive multi-drug resistant disease and patient mortality. Thus, there is a need for the development of new strategies that improve the prognoses for these patients. Melflufen (melphalan flufenamide) is a first-in-class peptide-drug conjugate that is currently in late-phase clinical trials for multiple myeloma. This highly lipophilic agent is preferentially retained in malignant plasma cells (MPCs), where overexpressed aminopeptidases lead to trapping of the alkylator melphalan. We evaluated the anti-myeloma effects of melflufen on patient samples treated ex vivo, and found pronounced sensitivity to melflufen in most samples, with particularly potent efficacy in PCL samples. Methods: Bone marrow aspirate or peripheral blood samples were obtained from patients with plasma cell disorders after IRB approval and informed consent. Ex vivo efficacy of melflufen and melphalan were compared using our Myeloma Drug Sensitivity Testing (My-DST) platform that optimizes viability and tests the malignant cells in the context of the normal cells from their microenvironment (Walker et al, Blood Advances, 2020). In brief, mononuclear cells from patients with plasma cell dyscrasia, including MM and PCL, were isolated and cultured in triplicate wells with titrations of melphalan, melflufen or untreated controls for 48 hours. Post-treatment survival was measured by high-throughput flow cytometry with antibodies for CD138, CD38, CD45 and CD19, and a live/dead dye to discriminate viable MPCs from normal bone marrow cells. EC50 values were determined from these titrations using nonlinear regression curve fits. When the EC50 for melflufen was established in My-DST, a single dose concentration of 10 nM was used to screen patient samples and distinguish relative sensitivity or resistance. Results: Using the My-DST approach with 48 hour drug treatments, melflufen significantly decreased the viable MPC populations, whereas melphalan had little effect (Fig 1A). Concurrent titrations revealed significantly higher MPC sensitivity to melfufen (mean melphalan EC50 = not reached, mean melflufen EC50 = 22.9 nM) (Fig 1B). By comparison to another alkylator, cyclophosphamide's active metabolite has an EC50 of 3.75 µM in this assay. Response to melflufen was accentuated in 2/3 PCL samples tested (HTB-1802.1, HTB-1389.1), with the EC50 < 1nM (Fig 1B). Melflufen demonstrated toxicity in CD45 positive white blood cells, which is consistent with neutropenia observed in clinical trials (data not shown). In single dose screening studies in additional MM patient samples, 4/8 (50%) showed >20% decrease in viable MPCs after incubation with melflufen at 10 nM (Fig 1C). Overall, using those parameters for ex vivo "response" to meflufen, 3/3 patients with PCL responded, 5/6 patients with del(17p) responded, and 3/3 patients with c-MYC translocations responded (Fig 1C, italics). In addition, 3/5 samples from patients that were clinically daratumumab-refractory displayed sensitivity to melflufen. Of five samples from patients with prior exposure to alkylators, four were sensitive to melflufen. Conclusion: Overall, these data support that the peptide-drug-conjugate melflufen shows a broad efficacy across samples from patients with plasma cell disorders. Patients facing poor prognoses, including those with PCL, high-risk cytogenetics and daratumumab-refractory disease, have a great need for new treatments. Thus, the encouraging ex vivo results with melflufen in samples from these aggressive subsets support further clinical exploration. In particular, our preliminary data suggest that plasma cell leukemia patients may be exquisitely sensitive to melflufen. To follow-up these findings, we will expand the number of samples tested from PCL and other forms of high-risk MM samples. Ultimately, if the trend for accentuated sensitivity in plasma cell leukemia holds, a clinical approach for melflufen in these patients may improve outcomes for this group. Figure 1 Disclosures Lockerbie: Oncopeptides AB: Current Employment. Flanagan:Oncopeptides AB: Current Employment. Lehmann:Oncopeptides AB: Current Employment. Forsberg:Celgene: Speakers Bureau; Genentech, Inc., Sanofi, Karyopharm, Abbvie: Research Funding. Mark:Takeda: Consultancy; Kayopharm: Consultancy; Bristol-Myers Squibb: Research Funding; Janssen: Research Funding; Celgene: Consultancy; Amgen: Consultancy; Sanofi: Consultancy; Janssen: Consultancy. Sherbenou:Oncopeptides Inc.: Research Funding.

Preliminary Results from a Phase 1 Study of Cfi-400495, a PLK4 Inhibitor, in Patients with Acute Myeloid Leukemia and High Risk MDS

Introduction: CFI-400945 is a first-in-class, potent, selective, orally active inhibitor of Polo-like kinase 4 (PLK4) (Ki=0.26nM), a master regulator of centriole duplication, necessary for genomic integrity (Mason et al. Cancer Cell 2014; 26:163-76). CFI-400945 has activity in leukemia cell lines and primary leukemia samples including those with complex karyotype, inversion 3 and monosomy 7 (Minden. personal communications). This suggests that CFI-400945 may provide an effective treatment of patients with AML. The objectives of this phase 1 trial was to establish the safety, tolerability, and recommend phase II dose (RP2D) of CFI-400945 in patients with acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS). Methods: Patients with relapsed/refractory AML or MDS and patients with untreated AML who refused induction chemotherapy or who are not candidates for intensive chemotherapy were eligible. Dose escalation followed a standard 3+3 design with a starting dose of 64 mg orally once daily. Plasma levels of CFI-400945 free base were measured on Days 1, 2, & 29 of Cycle 1 and Day 15 on all subsequent cycles. Peripheral blood and/or bone marrow were obtained at baseline, Day 8 of Cycle 1 and Day 1 of each subsequent cycle prior to dosing for pharmacodynamic monitoring. Safety assessments using the NCI CTCAE version 4.03 were performed. Results: From May 2018 to June 2019, nine patients have been enrolled on study across three pre-defined dose levels (64 mg [n=3], 96 mg [n=4], and 128 mg [n=2]). Three patients had untreated AML, five patients had relapsed/refractory AML and one patient had myelodysplastic syndrome/myeloproliferative disorder (MDS/MPN). Patient characteristics at diagnosis are outlined in Table 1. Six (67%) patients had baseline high throughput sequencing; the most frequent mutations were TP53 (33%), TET2 (33%), KRAS (33%) and DNMT3A (33%). A total of 20 cycles were administered with a median of 1 cycle (range, 0 to 7 cycles). The most common non-hematological drug related toxicities of any grade, which occurred in over 20%, were diarrhea (44%), headache (44%), colitis (33%), vomiting (33%), bilirubin increase (22%), dizziness (22%), fatigue (22%), and nausea (22%). One patient on the 96 mg dose level was not evaluable for DLT and hence, replaced. Both patients treated at the 128 mg/day dose level developed DLTs, consisting of grade 3 colitis and grade 5 sepsis and colitis. Pharmacokinetic profile indicated low interpatient variability between patients. Maximum exposure did not correlate with toxicity Six patients were evaluable for disease response. Two (33%) achieved complete remission (CR), 3 pts (50%) had stable disease (with one patient having a 78% reduction in marrow blast count). The patient with MDS/MPN who did not complete 1 cycle of therapy progressed to AML (Figure 1). Both patients who obtained a CR had an early response within 2 cycles. One CR has been durable for 218 days with no measurable residual disease (MRD) by flow cytometry. The additional patient, who obtained a CR with incomplete platelets recovery, with subsequent best response of CR, had a sustained response for 91 days before relapse was confirmed by bone marrow examination (Figure 1). Conclusion: Single agent CFI-400945 has activity in patients with poor risk AML. The RP2D in this population is 96 mg once daily. Dose expansion is occurring at the RP2D level. A phase 2 study with CFI-400945 single agent or in combination study with azacitidine or decitabine is planned. Disclosures Leber: Novartis: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Abbvie: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Alexion: Honoraria, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Lundbeck: Honoraria, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Takeda/Palladin: Honoraria, Membership on an entity's Board of Directors or advisory committees; Treadwell: Honoraria, Membership on an entity's Board of Directors or advisory committees; Amgen: Honoraria, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; BMS/Celgene: Honoraria, Membership on an entity's Board of Directors or advisory committees; Pfizer: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Otsuka Pharmaceutical: Honoraria, Membership on an entity's Board of Directors or advisory committees; Janssen: Honoraria, Membership on an entity's Board of Directors or advisory committees. Bray:Treadwell Therapeutics: Current Employment; TIO Discovery: Current Employment. Gupta:Pfizer: Consultancy; Novartis: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Sierra Oncology: Consultancy, Membership on an entity's Board of Directors or advisory committees; Bristol MyersSquibb: Honoraria, Membership on an entity's Board of Directors or advisory committees; Incyte: Honoraria, Research Funding. Maze:Novartis: Honoraria; Pfizer: Consultancy; Takeda: Research Funding. McNamara:Novartis: Honoraria. Schimmer:Jazz: Honoraria; Otsuka: Honoraria; Medivir AB: Research Funding; AbbVie Pharmaceuticals: Other: owns stock ; Takeda: Honoraria, Research Funding; Novartis: Honoraria.

Proteomics in Pediatric Acute Myeloid and T-Cell Lymphoblastic Leukemia: Shared Individual Protein Expression Patterns Co-Cluster into Overall Distinct Combinations

INTRODUCTION: Pediatric acute lymphoblastic leukemia (ALL) and acute myeloid leukemia (AML) are heterogeneous diseases mediated by changes in protein expression. As most chemotherapeutic agents target proteins, and because overall survival of pediatric AML is far inferior to both pre-B and T-ALL, we aimed to compare the proteomic landscape of pediatric T-ALL and AML, with the goal of determining common AML-T-ALL pathways that are potentially targetable with novel agents. METHODS: Reverse phase protein arrays (RPPA) analysis was used to measure protein expression in 858 acute leukemia samples (358 T-ALL and 500 AML, 723 pediatric (< 18 yrs.), 135 adults (≥18 yrs.)) and 61 normal CD34+ samples using 270 validated antibodies. Expression levels were normalized against CD34+ cells. Proteins were allocated into 30 functionally related subgroups (Protein Functional Group (PFG)). A progeny clustering algorithm was applied to each PFG to search for strong correlations between proteins and to identify an optimal number of Protein Clusters (PC). Block clustering identified PC that recurrently co-occurred together (Protein Constellation (CON)) and patients that expressed similar combination of CON were defined as Protein Signature (SIG). Proteins that were differentially expressed were identified using the Student's t-test or ANOVA, with a Bonferroni adjusted p-value (0.05/ 270 = 0.00019)). RESULTS: Of the 270 analyzed proteins, 131 proteins (49%) were differentially expressed between T-ALL and AML; 60 were higher in T-ALL, 71 in AML. Similar to our previous analysis in adult AML and ALL, cell cycle regulators (CDKN1A, CDKN1B) and 2 of the 5 histone marks (H3K36Me3 & H3K4Me3) were higher expressed in T-ALL compared to AML. Heat shock proteins (HSP90AA1_B1, HSPA1A_L, HSPB1 and HSPB1-pSer82) were higher in AML as well as translation proteins EIF2S1, EIF4E and EIF4EBP1 and ribosomal proteins RPS6-pSer235_236 and RPS6KB1, while expression of the translation inhibitory proteins EIF2S1-pSer51 and EIF2AK2-pThr451 was lower in AML compared to T-ALL. Next, cluster analysis in the context of 30 PFG resulted in 133 PC. The majority (n=102) of PC were expressed in both diseases, 30 PC (22.6%) were AML-specific, and only one PC was specific to T-ALL (characterized by high CDKN1A, CDKN1B and CCND1, but low WEE1, CCNB1 and RB1-pSer). Co-clustering of the 133 PC identified 14 CON that formed 17 SIG. Three CON (5, 9, 10) were associated with AML, 2 with T-ALL (2, 13) and 8 CON were observed in both diseases. In contrast, 15 of SIG were associated with either T-ALL or AML, and two SIG (9, 10) included a mixture of both diseases (P < 0.001, annotation bar Figure 1 "Disease")). SIG were associated with gender (P < 0.001), but not with CNS-status and ethnicity (Hispanic vs. non-Hispanic). No age-specific (kids vs. adults) signatures were observed. For each SIG and CON, proteins significantly higher or lower expressed compared to the normal CD34+ cells were identified. CONCLUSIONS: This study provides support for our previous hypothesis that pediatric T-ALL and AML can be characterized by recurrent protein expression patterns. While most PC and CON were found in both diseases, SIG (i.e. combinations of protein expression patterns) were specific to either T-ALL or AML. We found similar results when comparing B-ALL to AML in adults. Shared CON indicate that there are common protein expression patterns between pediatric T-ALL and AML. Proteins or pathways with similar utilization (e.g. CON3, 5) in both diseases may allow for information on clinical utility from one disease to be applicable to the other. Those with differential utilization are likely to be uninformative with respect to clinical utility in the other disease. Figure. "MetaGalaxy" analysis for pediatric AML and T-ALL. Each row represents one protein clusters (n = 133), each column represents one patient (n = 858). Blue indicates membership for that particular protein cluster. Annotation bar shows strong correlation with disease (yellow = T-ALL, blue = AML). No associations were seen for age (blue = adult, pink = pediatric) or Ethnicity (blue = Hispanic, yellow = non-Hispanic). Figure 1 Disclosures No relevant conflicts of interest to declare.

Frequency and Prognostic Significance of Recurrent Gene Mutations in Pediatric B-ALL: Report from the DFCI ALL Consortium

The application of next-generation sequencing (NGS) approaches to leukemia markedly expanded our understanding of the molecular landscape of pediatric acute lymphoblastic leukemia (ALL), but NGS integration into clinical care and therapeutic decision making is still limited. The prognostic impact of discovered mutations in uniformly treated patients with newly diagnosed B-ALL is not well characterized, and there is no consensus regarding the clinical significance of many of these findings. We investigated the frequency of mutations affecting common molecular pathways in pediatric B-ALL that are potentially druggable with targeted therapies, to define the prognostic role of these mutations, as measured by end-induction minimal residual disease (MRD) and event-free survival (EFS). We analyzed 159 patients (median age 6 years, range 1-18) with newly diagnosed Ph-negative B-ALL treated between 2005-2015 according to the Dana-Farber Cancer Institute (DFCI) ALL Consortium Protocols 05-001 and 11-001. Diagnostic leukemia samples were sequenced using a validated clinical NGS panel. We focused on mutations affecting the following pathways: Ras (NRAS, KRAS, PTPN11, NF1 and BRAF), cell cycle regulation (CCND3, CDKN2A/B and RB1), PI3K signaling (PTEN, PIK3CA, PIK3R1, PIK3C2B, MTOR, TSC1 and TSC2), Polycomb repressive complex 2 (PRC2) (EED, EZH2 and SUZ12), and JAK/STAT signaling (CRLF2, JAK2, JAK3, MPL, SH2B3 and SOCS1). Additionally, we analyzed mutations affecting CREBBP, FLT3, PAX5, SETD2 and TP53. Mutations occurring in the ExAC database at a frequency greater than 0.01% were excluded. A Fisher exact test and Wilcoxon rank sum test were used for categorical and continuous variables. High end-induction MRD, defined as > 10-3, was assessed by an Ig-TCR PCR assay. EFS was estimated using Kaplan and Meier method and tested between groups using a log-rank test. Induction death/failure, relapse or death were considered as events. Overall, 108 of the 159 patients (68%) carried at least one mutation in the studied genes. Most common mutations were in the Ras pathway (47%), CREBBP (9%), JAK pathway (8%), FLT3 (8%), PI3K pathway (5%), PAX5 (4%), SETD2 (4%), TP53 (4%), cell cycle regulation (3%) and PRC2 complex (3%). We investigated the distribution of mutations among common cytogenetic groups: ETV6-RUNX1, high hyperdiploidy (HHD) (51-65 chromosomes), hypodiploidy (<45 chromosomes), KMT2A-rearranged, TCF3-PBX1, and intra-amplification chromosome 21 (iAMP21). There was a strong association between TP53 and hypodiploidy (43% vs 2%; p=0.001). Ras pathway and FLT3 mutations were enriched in HHD (67% vs 39%; p=0.004) and (17% vs 4%; p=0.016), respectively. FLT3 mutations were mutually exclusive with ETV6-RUNX1 (12% vs. 0%, p=0.038), and Ras pathway mutations were rare in this subgroup (p=0.018). PAX5 mutations were enriched in children >10 years of age (p=0.035). There were no associations with sex. Ras pathway mutations were associated with high end-induction MRD (68% vs 42%; p=0.045), and this association was stronger for clonal (variant allele frequency (VAF) >25%) mutations (n=33) (63% vs 19%; p=0.0002). Focusing on HHD, CREBBP mutations frequently co-occurred with clonal Ras pathway mutations (83% vs 17%; p=0.032). Among HHD patients with evaluable MRD (n=33), 7 patients had high MRD, and 3 of these had CREBBP mutations (p=0.052). The analyzed cohort was enriched for higher risk ALL disease (Table 1). The 5-year EFS was 80%±3% among these 159 patients. Overall, no EFS difference was observed based on Ras pathway (p=0.35). Among HHD ALL, the presence of CREBBP or clonal Ras pathway mutations was significantly associated with inferior 5-yr EFS (50%±20% vs 97%±3%; p=0.0006, Figure 1) and (70%±13% vs 100%; p=0.007, Figure 2), respectively. In conclusion, our findings provide insight into the prognostic significance of the most common mutations in pediatric HHD B-ALL in a uniformly treated cohort of patients as part of the DFCI ALL Consortium. The presence of CREBBP and clonal Ras pathway mutations may be associated with upfront chemotherapy resistance as demonstrated by high end-induction MRD. Further analysis of Ras pathway mutations segregated by VAF is warranted. Future trials may integrate these findings into risk stratification of HHD ALL. With prospective continued clinical use of NGS assays, we will further clarify the role of mutations and their contribution to disease outcomes in B-ALL. Disclosures Mar: Blueprint Medicines Corporation: Current Employment, Current equity holder in publicly-traded company. Stegmaier:Novartis: Research Funding; Auron Therapeutics: Consultancy. Silverman:Takeda: Other: advisory board; Servier: Other: advisory board; Syndax: Other: advisory board.

Comparison of Deep Whole Exome Versus Targeted Gene Sequencing for Assessment of Persistent Molecular Disease in Acute Myeloid Leukemia Samples

Background There is growing evidence that genomics-based assessment of persistent molecular disease (PMD) may be a useful risk stratification tool for patients with acute myeloid leukemia (AML). However, there is no consensus about the optimal approach for detection and monitoring of leukemia-associated mutations (LAMs) for PMD testing. One approach is whole exome sequencing (WES), which provides a comprehensive assessment of the clonal architecture by identifying LAMs across the exome at diagnosis, which can then be measured in follow-up samples. Another approach is to use highly sensitive targeted gene sequencing (TGS) to detect persistent LAMs even at very low levels, although detection is limited to genes interrogated by the panel. Although both of approaches have been shown to predict outcome in retrospective studies, there are substantial differences in the number of LAMs identified and the limit of detection, and a head-to-head comparison has not yet been reported. Here, we use a defined cohort enrolled in a prospective clinical trial to compare PMD results from deep WES to error-corrected TGS. Methods Cohort: Patients were age 18-60 with de novo AML classified as intermediate risk by European Leukemia Net criteria, who achieved a morphologic remission after undergoing standard induction therapy. All patients were enrolled in a prospective clinical trial (NCT02756962). Sequencing: Deep WES was performed using DNA from normal tissue (buccal swab or skin) and pre- and post-induction (~day 30) bone marrow (BM) samples to achieve an average coverage depth of ~600x. LAMs identified via paired tumor/normal analysis of the pre-induction sample were queried in the post-induction sample for assessment of PMD. PMD testing by TGS was performed using error-corrected sequencing of 40 genes recurrently mutated AML genes to an average error-corrected coverage depth of ~4500x. Definitions: Based on previously published work, we used two separate variant allele frequency (VAF) cutoffs to define PMD. For WES, PMD+ was defined as ≥1 LAM with a VAF >2.5% (Klco JAMA 2015). For TGS, PMD+ was defined as having ≥1 LAM with a VAF >0.5% (Duncavage NEJM 2018). LAMs were sub-classified per Table 1. TGS mutations not identified as LAMs by WES at diagnosis were excluded from the primary analysis. Results 31 patients were studied. LAMs are summarized in Table 1. 20 patients (65%) were PMD+ by WES after induction, and 22 patients (71%) were PMD+ by TGS. The concordance between WES and TGS was 81% (25/31) (Table 2). Two patients were PMD+ by WES only, due to the persistence of LAMs in the exome space, but not in the targeted panel. Four patients were PMD+ by TGS only, due to the presence of recurrent mutations at VAFs below the detection limit of WES (range: 0.59 - 1.90%). Two patients were PMD+ by both assays, but due to different mutations. All other patients who were PMD+ by both assays had at least one overlapping mutation. Analysis of the mutations that persisted after therapy in both assays showed that 26% of patients (8/31) were PMD+ by TGS because of mutations in DNMT3A, TET2, or ASXL1 (i.e., "DTA" mutations). All of these patients were also PMD+ by WES, with 7 of the 8 patients having ≥1 additional non-DTA mutation (median: 3, range: 1 - 7). In an exploratory analysis, 22 additional mutations in 13 patients were identified by TGS that were not detected by WES on the diagnostic sample. This included two patients who were PMD-, but who had new mutations in TET2 and DNMT3A, respectively, likely representing selection for ancestral clones that were unrelated to the AML founding clone. Conclusion Concordance between WES and TGS-based PMD assessment was high. Discordant results were generally driven by non-recurrent mutations detected by WES, and low-level mutations detected by the high coverage, error-corrected TGS. Although isolated DTA mutations were common on TGS, WES analysis showed additional LAMs accompanied these variants in most cases, indicating the persistence of an ancestral leukemic clone that may provide useful prognostic information. We also observed new, low-level mutations that emerged after therapy in 42% of patients, some of which were not part of the leukemic clone identified at diagnosis. This indicates that use of highly sensitive PMD approaches may be challenging without pre-induction mutation testing, which is required to understand the relevance of markers of persistent molecular disease. Disclosures Jacoby: AbbVie: Research Funding; Takeda: Consultancy; Jazz Pharmaceuticals: Research Funding.

Comprehensively analyze the expression and prognostic role for ten-eleven translocations (TETs) in acute myeloid leukemia.

BackgroundThe ten-eleven translocation (TET) family oxidize 5-methylcytosines (5mCs) and promote the locus-specific reversal of DNA. The role of TETs in acute myeloid leukemia (AML) is mostly unknown.MethodsTETs mRNA expression levels were analyzed via Gene Expression Profiling Interactive Analysis (GEPIA). The association TETs expression levels and methylation with prognosis by UALCAN GenomicScape, and METHsurv. We analyzed TETs’ aberration types, located mutations, and structures via cBioPortal. GeneMANIA performed the functional network. Gene ontology (GO) enrichment was analyzed via LinkedOmics. MiWalK identified miRNAs, miTarbase, and TargetScan. Transcription factor (TF) targets were analyzed via ChEA3. GSCAlite analyzed the role of these defined genes in cancer pathways and potential drug targets. Finally, we selected AML patients in our department to investigate the mutated types of TETs.ResultsTETs expression level results showed TET1 (P=0.003) and TET2 (P=0.004) overexpressed in Haferlach leukemia samples, TET3 (P=4.04e-8) downregulation in Andersson leukemia samples. TET2 and TET3 overexpression but TET1 downregulation in the GEPIA database. Overexpression of TET2 leads to positive outcomes (P=0.0091). The upregulation of TET2 led to poor survival for CN-AML patients, but downregulation of TET3 indicated a satisfactory prognosis. The hypermethylation of TETs like cg24705708 (P=0.036), cg05976228 (P=0.022), cg19127638 (P=0.022), cg15254238 (P=0.025), cg07669489 (P=0.037) indicate poor outcomes. Overexpression of GALNS (P=0.024) as an adverse biomarker, downregulation of E2F5 (P=0.037), MAP7 (P=0.019), and NRIP1 (P=0.0013) indicated good prognosis. Regulatory network analysis indicated TETs’ functions, including covalent chromatin modification, histone modification, DNA methylation, or demethylation. Enrichment functions involving. TETs participate in several cancer pathways, including DNA repair response and receptor tyrosine kinase (RTK) signaling pathway. TETs are sensitive to belinostat, ceranib-2, docetaxel, tivantinib, and vincristine.ConclusionPresent study showed that TETs have different expressions in AML, and the expression levels of TETs lead to different outcomes of AML. The TETs cancer pathway analysis will also provide potential therapy methods for AML patients with TETs aberrations.

Detection of Pathogenic Isoforms of IKZF1 in Leukemic Cell Lines and Acute Lymphoblastic Leukemia Samples: Identification of a Novel Truncated IKZF1 Transcript in SUP-B15.

Simple SummaryAbnormal RNA splicing plays a fundamental role in leukemogenesis in acute lymphoblastic leukemia (ALL). Many cases of high-risk B-cell ALL cases, including BCR-ABL1+ and BCR-ABL1-like ALL, share a common molecular mechanism of aberrant fusion transcripts involving tyrosine kinase genes combined with dysregulation of the transcription factor and lymphocyte differentiation factor IKZF1. Dysfunction of IKZF1 in ALL is caused by mutation and gene deletion but also alternative splicing resulting in exon skipping with production of aberrant IKZF1 proteins. We report here an assay to detect aberrantly spliced isoforms of IKZF1 in ALL to assist in diagnosis, outcome prediction, and therapy selection in ALL and the identification of a novel altered IKZF1 product in a model ALL cell line.Leukemia-associated alternative splicing of IKZF1 can result in proteins with loss of one to four copies of its N-terminal zinc finger domains (N-ZnF). The best characterized pathogenic splice isoforms, Ik-6 and Ik-8, have been commonly found in BCR-ABL1+ acute lymphoblastic leukemia (ALL) and a subset of BCR-ABL1-like ALL. Infantile and childhood ALL that express these pathogenic IKZF1 isoforms have shown inferior clinical outcomes and can be resistant to tyrosine kinase inhibitors. Using ALL cell lines, we designed and validated a method to detect abnormal IKZF1 transcripts. In the SUP-B15 leukemia cell line, we noted novel IKZF1 transcripts that include both an Ik-6 splice and a transcript with a 14 base pair insertion at the C-terminus. There was also increased IKZF2 protein in SUP-B15 as compared to other ALL lines. Expression of Ik-6 could be suppressed by treatment with the pro-apoptotic type II histone deacetylase inhibitor givinostat. In 17 adult ALL samples, we noted the Ik-6 isoforms in 6 of 15 BCR-ABL1−, and 1 of 2 BCR-ABL1+ cases, with Ik-8 also expressed in one case. Cases with Ik-6 expression showed inferior survival as well as older age at presentation, lower expression of CD10 and more commonly a diploid karyotype.

A non-invasive tool for early detection of acute leukemia in children using a paper-based optoelectronic nose based on an array of metallic nanoparticles

Volatile organic compounds (VOCs) in blood samples can be used as useful biomarkers to diagnose various human diseases. This study describes the potential of a paper-based sensor array for detecting leukemia using blood VOCs. Blood samples were collected from 59 new leukemia cases and 47 healthy cases as a control group. Each blood sample was divided into two parts; one for a laboratory test and the other was used in our study. Samples were mixed with heparin and then transferred to a sterile container, and a sensor was stacked on its cap. This sensor array contains 16 nanoparticles deposited on a sheet of hydrophobic paper in a 4 × 4 array format. Containers were stored in an oven at 60 °C for 4.5 h. Then, the image of sensors was recorded by a scanner and compared to the image before exposing the blood vapor. The sensor responses were subjected to different multivariate statistical methods to develop models that discriminate between control and leukemia samples. The interaction of nanoparticles with the volatile metabolome of blood caused aggregation and consequently changing in the color of nanoparticles. The color changes resulted in a specific pattern for blood samples with leukemia, which is different from those obtained from healthy specimens. The discrimination analysis was approved by pattern recognition methods such as principal component analysis with 97% accuracy. Among 59 patients, the mean age was 6.02 ± 4.55 years (range 1–16 y). The mean total response was 652.83 ± 117.02. The rock curve showed an accuracy of 96% for classifying patients from the control group. The logistic regression model showed that 93.6% of healthy and 93.2% of patients were classified correctly by using this method. These statistics agree with the classification results obtained by principal component analysis. For every 5000-unit increase in platelet count, the chance of leukemia decreased by 9%. Additionally, the chance of being categorized as a patient decreased by 10% for every 20-unit increase in total response. The electronic nose using VOC’s of blood is a non-invasive and inexpensive tool for detecting new cases of leukemia with high sensitivity and specificity. Platelet count is an essential para-clinical parameter determining the total response of the sensors. Follow up studies with a larger sample size are warranted to elucidate its clinical applicability.

Identification and Staging of B-Cell Acute Lymphoblastic Leukemia Using Quantitative Phase Imaging and Machine Learning.

Identification and classification of leukemia cells in a rapid and label-free fashion is clinically challenging and thus presents a prime arena for implementing new diagnostic tools. Quantitative phase imaging, which maps optical path length delays introduced by the specimen, has been demonstrated to discern cellular phenotypes based on differential morphological attributes. Rapid acquisition capability and the availability of label-free images with high information content have enabled researchers to use machine learning (ML) to reveal latent features. We developed a set of ML classifiers, including convolutional neural networks, to discern healthy B cells from lymphoblasts and classify stages of B cell acute lymphoblastic leukemia. Here, we show that the average dry mass and volume of normal B cells are lower than those of cancerous cells and that these morphologic parameters increase further alongside disease progression. We find that the relaxed training requirements of a ML approach are conducive to the classification of cell type, with minimal space, training time, and memory requirements. Our findings pave the way for a larger study on clinical samples of acute lymphoblastic leukemia, with the overarching goal of its broader use in hematopathology, where the prospect of objective diagnoses with minimal sample preparation remains highly desirable.

Standardizing a volume benchmark for cerebrospinal fluids for optimal diagnostic accuracy

Cerebral spinal fluid (CSF) cytomorphologic analysis remains the gold standard in the evaluation of malignant leptomeningeal involvement. However, collection of optimal volumes for adequate cytomorphologic evaluation is not standardized. Our study investigated optimal CSF volumes that result in a significant diagnostic result. A total of 4114 samples were retrospectively identified from 2014 to 2018, and 2557 samples had concurrent flow cytometry (FC) study. Each specimen was grouped as unsatisfactory, negative, atypical, or positive. Positive samples were grouped as either solid tumors, leukemia, or lymphoma by the type of malignancy detected. Demographic data as well as CSF source was recorded. Specimens with FC were separated by detection on cytology and/or FC. A t-test and ANOVA test were used to compare the average volumes for each group. Average volumes for negative, atypical, and positive samples are 7.48 mL (95% CI: 7.33, 7.63), 7.97 mL (95% CI: 7.37, 8.57), and 8.44 mL (95% CI: 7.46, 9.43), respectively. Average volumes for solid tumors, leukemia, and lymphoma positive samples are 12.0 mL (95% CI: 9.11, 14.89), 6.73 mL (95% CI: 5.94, 7.53), and 8.44 mL (95% CI: 6.78, 10.09). For cases with FC, the volumes are 10.11 mL (95% CI: 9.28, 10.96), 7.28 mL (95% CI: 6.87, 7.70), and 6.86 mL (95% CI: 6.25, 7.49) for positive cytology only, positive cytology/FC, and negative for both, respectively. Our results suggest that higher volumes produce better results for analysis. We recommend an optimal volume of 8.44 mL for cytologic work-up of malignancies. However, optimal volumes may differ based upon malignancy type and utilization of flow cytometry.

Ex vivo cultures and drug testing of primary acute myeloid leukemia samples: Current techniques and implications for experimental design and outcome.

New treatment options of acute myeloid leukemia (AML) are rapidly emerging. Pre-clinical models such as ex vivo cultures are extensively used towards the development of novel drugs and to study synergistic drug combinations, as well as to discover biomarkers for both drug response and anti-cancer drug resistance. Although these approaches empower efficient investigation of multiple drugs in a multitude of primary AML samples, their translational value and reproducibility are hampered by the lack of standardized methodologies and by culture system-specific behavior of AML cells and chemotherapeutic drugs. Moreover, distinct research questions require specific methods which rely on specific technical knowledge and skills. To address these aspects, we herein review commonly used culture techniques in light of diverse research questions. In addition, culture-dependent effects on drug resistance towards commonly used drugs in the treatment of AML are summarized including several pitfalls that may arise because of culture technique artifacts. The primary aim of the current review is to provide practical guidelines for ex vivo primary AML culture experimental design.

A deconvolution method and its application in analyzing the cellular fractions in acute myeloid leukemia samples

BackgroundThe identification of cell type-specific genes (markers) is an essential step for the deconvolution of the cellular fractions, primarily, from the gene expression data of a bulk sample. However, the genes with significant changes identified by pair-wise comparisons cannot indeed represent the specificity of gene expression across multiple conditions. In addition, the knowledge about the identification of gene expression markers across multiple conditions is still paucity.ResultsHerein, we developed a hybrid tool, LinDeconSeq, which consists of 1) identifying marker genes using specificity scoring and mutual linearity strategies across any number of cell types, and 2) predicting cellular fractions of bulk samples using weighted robust linear regression with the marker genes identified in the first stage. On multiple publicly available datasets, the marker genes identified by LinDeconSeq demonstrated better accuracy and reproducibility compared to MGFM and RNentropy. Among deconvolution methods, LinDeconSeq showed low average deviations (≤0.0958) and high average Pearson correlations (≥0.8792) between the predicted and actual fractions on the benchmark datasets. Importantly, the cellular fractions predicted by LinDeconSeq appear to be relevant in the diagnosis of acute myeloid leukemia (AML). The distinct cellular fractions in granulocyte-monocyte progenitor (GMP), lymphoid-primed multipotent progenitor (LMPP) and monocytes (MONO) were found to be closely associated with AML compared to the healthy samples. Moreover, the heterogeneity of cellular fractions in AML patients divided these patients into two subgroups, differing in both prognosis and mutation patterns. GMP fraction was the most pronounced between these two subgroups, particularly, in SubgroupA, which was strongly associated with the better AML prognosis and the younger population. Totally, the identification of marker genes by LinDeconSeq represents the improved feature for deconvolution. The data processing strategy with regard to the cellular fractions used in this study also showed potential for the diagnosis and prognosis of diseases.ConclusionsTaken together, we developed a freely-available and open-source tool LinDeconSeq (https://github.com/lihuamei/LinDeconSeq), which includes marker identification and deconvolution procedures. LinDeconSeq is comparable to other current methods in terms of accuracy when applied to benchmark datasets and has broad application in clinical outcome and disease-specific molecular mechanisms.

AML-275: Pediatric Acute Lymphoblastic and Myeloid Leukemia Express Unique Combinations of Protein Expression Patterns

Context Pediatric acute lymphoblastic leukemia (ALL) and acute myeloid leukemia (AML) are heterogeneous diseases mediated by changes in protein expression. As most chemotherapeutic agents target proteins, and because overall survival of pediatric AML is far inferior to both pre-B and T-ALL, we aimed to compare the proteomic landscape of pediatric T-ALL and AML. Objective To compare protein expression in pediatric T-ALL and AML. DESIGN: Reverse phase protein arrays (RPPA) analysis was used to measure protein expression in 858 acute leukemia samples (358 T-ALL and 500 AML) and 61 normal CD34+ samples using 270 validated antibodies. Results Of the 270 analyzed proteins, 136 proteins (50%) were differentially expressed between T-ALL and AML; 61 were higher in T-ALL, 75 in AML (P Conclusions This study provides support for our previous hypothesis that pediatric T-ALL and AML can be characterized by recurrent protein expression patterns. While most PC and CON were found in both diseases, SIG were specific to either T-ALL or AML. We found similar results when comparing B-ALL to AML in adults. Shared CON indicate that there are common features between pediatric T-ALL and AML. Proteins or pathways with similar utilization in both diseases may allow for information on clinical utility from one disease to be applicable to the other. Those with differential utilization are likely to be uninformative with respect to clinical utility in the other disease. Grant Funding R01 CA164024 and R01 CA193776

Germline DNMT3A mutation in familial acute myeloid leukaemia

ABSTRACT Acute myeloid leukaemia (AML) is a heterogeneous myeloid malignancy characterized by recurrent clonal events, including mutations in epigenetically relevant genes such as DNMT3A, ASXL1, IDH1/2, and TET2. Next-generation sequencing analysis of a mother and son pair who both developed adult-onset diploid AML identified a novel germline missense mutation DNMT3A p.P709S. The p.P709S protein-altering variant resides in the highly conserved catalytic DNMT3A methyltransferase domain. Functional studies demonstrate that the p.P709S variant confers dominant negative effects when interacting with wildtype DNMT3A. LINE-1 pyrosequencing and reduced representation bisulphite sequencing (RBBS) analysis demonstrated global DNA hypomethylation in germline samples, not present in the leukaemic samples. Somatic acquisition of IDH2 p.R172K mutations, in concert with additional acquired clonal DNMT3A events in both patients at the time of AML diagnosis, confirms the important pathogenic interaction of epigenetically active genes, and implies a strong selection and regulation of methylation in leukaemogenesis. Improved characterization of germline mutations may enable us to better predict malignant clonal evolution, improving our ability to provide customized treatment or future preventative strategies.

Abstract 5910: Single-cell multiomic analysis of SNV, CNV, and protein expression

Abstract Recent advancements in precision medicine, while highly promising, presents a major technical challenge to researchers due to disease heterogeneity. The emergence of single-cell technologies has greatly refined the resolution in which sample diversity can be investigated, enhancing the efficiency of selecting appropriate molecular targets. Additionally, applying multiomic analysis on single cells would further improve the understanding of cell-to-cell heterogeneity by providing unique insights on cellular and genetic composition. Using a two-step droplet microfluidic technology, the Mission Bio Tapestri Platform enables multiplex-PCR based high-throughput targeted DNA sequencing in single cells to obtain single-nucleotide variation (SNV) and copy number variation (CNV) information. By leveraging this technology, a new workflow is developed to detect protein expression in addition to DNA genotype in the same single cells. In this approach, cells are labeled with a pool of oligonucleotide-conjugated antibodies prior to loading the cells into the Tapestri Instrument for targeted DNA analysis. Sequencing libraries are then prepared from both antibody oligonucleotides and the amplified DNA sequences, followed by identification of single-cell DNA genotypes and protein signatures from the sequencing readout. In a mixed population of four cell lines, single-cell SNV and CNV information from 127 targeted amplicons and the protein data from 10 antibodies independently classified the cells into appropriate clusters. This method has been successfully performed on clinical samples with myeloid malignancies. In an acute myeloid leukemia (AML) sample, combined single-cell SNV, CNV, and protein expression data illustrated the heterogeneity within the sample. The data clearly identified CD3+ T cells and CD19+ B cells without pathogenic SNVs and CNVs. CD34hiCD11blo and CD34loCD11bhi subpopulations were also identified within the cells carrying the same pathogenic SNVs and CNVs. We believe that this novel multiomic technology will enable new discoveries in the complex relationship between genotype and phenotype, leading to a better understanding of disease biology, and subsequently better design of diagnostics and therapies. Citation Format: Aik Ooi, Pedro Mendez, Dalia Dhingra, Nigel Beard, David Ruff. Single-cell multiomic analysis of SNV, CNV, and protein expression [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 5910.

Abstract 4544: Multifaceted mechanism of action of DuoHexaBody-CD37 involves both complement- and Fc gamma receptor-mediated cytotoxicity in pre-clinical B-cell lymphoma models

Abstract CD37 is a tetraspanin expressed on mature B cells where it orchestrates plasma membrane organization, receptor signaling, cell migration and adhesion. As CD37 is abundantly expressed on many mature B cell-derived malignancies, it represents an attractive target for new antibody therapies. DuoHexaBody®-CD37 is a novel biparatopic bispecific CD37 antibody with an Fc domain engineered to enhance antibody hexamerization upon binding to CD37 on the plasma membrane. DuoHexaBody-CD37 was shown to induce potent complement-dependent cytotoxicity (CDC) of malignant B-cell lines and primary chronic lymphocytic leukemia and non-Hodgkin lymphoma patient samples (Oostindie et al., Blood 2018 132:4170; van der Horst et al., Blood 2018 132:4179). Here, we demonstrate that the DuoHexaBody-CD37 mechanism of action also encompasses FcγR-mediated effector functions, including antibody-dependent cellular cytotoxicity (ADCC) and antibody-dependent cellular phagocytosis (ADCP). Daudi and Raji cells opsonized with DuoHexaBody-CD37 induced FcγRIIa and FcγRIIIa signaling in luciferase reporter assays. In agreement with efficient FcγR engagement, opsonization with DuoHexaBody-CD37 resulted in dose-dependent ADCC and ADCP by human healthy donor peripheral blood mononuclear cells (PBMCs) and monocyte-derived macrophages (MDMs), respectively. In a 4-hour Chromium-51 release assay, DuoHexaBody-CD37 induced ADCC in Daudi cells with PMBCs from 11 out of 12 tested donors (average EC50 of 9.87 [±9.98] ng/mL; maximum kill 20.4 [±9.2]%), and in Raji cells with 1 out of 3 donors. Image- and flow cytometry-based assays with MDMs illustrated that in presence of DuoHexaBody-CD37, Daudi cells were efficiently engulfed by effector cells from 3 different donors (average percentage phagocytic macrophages 29.7 ± 7.6%) resulting in target cell depletion (maximum depletion 75.9 [±18.0]%). Potent anti-tumor activity of DuoHexaBody-CD37 in vivo was reported in a screening approach using B-cell lymphoma patient-derived xenograft (PDX) models with single-mouse treatment groups. Two weekly doses of 5 mg/kg DuoHexaBody-CD37 resulted in strong tumor growth inhibition (tumor stasis or tumor regression) in 3/9 models compared to untreated tumors. Follow-up cohort PDX studies with eight mice per group confirmed potent, dose-dependent anti-tumor activity of DuoHexaBody-CD37 at doses as low as 1 mg/kg. In summary, DuoHexaBody-CD37 induces efficient tumor cell kill through CDC, ADCC and ADCP in vitro and shows potent anti-tumor activity in B-cell lymphoma PDX models in vivo. These data further strengthen the rationale for exploring the safety and efficacy of DuoHexaBody-CD37 in B-cell malignancy patients. Citation Format: Laurens P. Kil, Simone C. Oostindie, Kristin Strumane, Hilma J. van der Horst, Berris van Kessel, Marije B. Overdijk, Andreas Lingnau, Marcel Brandhorst, Jeroen van den Brakel, Margaret A. Lindorfer, Ronald P. Taylor, Martine E. Chamuleau, Tuna Mutis, A. Kate Sasser, Janine Schuurman, Paul W. Parren, Frank J. Beurskens, Esther C. Breij. Multifaceted mechanism of action of DuoHexaBody-CD37 involves both complement- and Fc gamma receptor-mediated cytotoxicity in pre-clinical B-cell lymphoma models [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 4544.