Relapsed Acute Lymphoblastic Leukemia Patients Research Articles

Machine learning-based evaluation of prognostic factors for mortality and relapse in patients with acute lymphoblastic leukemia: a comparative simulation study

BackgroundPredicting mortality and relapse in children with acute lymphoblastic leukemia (ALL) is crucial for effective treatment and follow-up management. ALL is a common and deadly childhood cancer that often relapses after remission. In this study, we aimed to apply and evaluate machine learning-based models for predicting mortality and relapse in pediatric ALL patients.MethodsThis retrospective cohort study was conducted on 161 children aged less than 16 years with ALL. Survival status (dead/alive) and patient experience of relapse (yes/no) were considered as the outcome variables. Ten machine learning (ML) algorithms were used to predict mortality and relapse. The performance of the algorithms was evaluated by cross-validation and reported as mean sensitivity, specificity, accuracy and area under the curve (AUC). Finally, prognostic factors were identified based on the best algorithms.ResultsThe mean accuracy of the ML algorithms for prediction of patient mortality ranged from 64 to 74% and for prediction of relapse, it varied from 64 to 84% on test data sets. The mean AUC of the ML algorithms for mortality and relapse was above 64%. The most important prognostic factors for predicting both mortality and relapse were identified as age at diagnosis, hemoglobin and platelets. In addition, significant prognostic factors for predicting mortality included clinical side effects such as splenomegaly, hepatomegaly and lymphadenopathy.ConclusionsOur results showed that artificial neural networks and bagging algorithms outperformed other algorithms in predicting mortality, while boosting and random forest algorithms excelled in predicting relapse in ALL patients across all criteria. These results offer significant clinical insights into the prognostic factors for children with ALL, which can inform treatment decisions and improve patient outcomes.

Mutational mechanisms in multiply relapsed pediatric acute lymphoblastic leukemia

Pediatric acute lymphoblastic leukemia (ALL) is marked by low mutational load at initial diagnosis, which increases at relapse. To determine which processes are active in (relapsed) ALL and how they behave during disease progression before and after therapy, we performed whole genome sequencing on 97 tumor samples of 29 multiply relapsed ALL patients. Mutational load increased upon relapse in 28 patients and upon every subsequent relapse in 22 patients. In addition to two clock-like mutational processes, we identified UV-like damage, APOBEC activity, reactive oxygen species, thiopurine-associated damage and an unknown therapy component as drivers of mutagenesis. Mutational processes often affected patients over longer time periods, but could also occur in isolated events, suggesting the requirement of additional triggers. Thiopurine exposure was the most prominent source of new mutations in relapse, affecting over half of the studied patients in first and/or later relapse and causing potential relapse-driving mutations in multiple patients. Our data demonstrate that multiple mutational processes frequently act in parallel as prominent secondary drivers with dynamic activity during ALL development and progression.

Abstract 5662: PDCD1 and CD48 significantly impact cancer recurrence and survival in pediatric acute lymphoblastic leukemia

Abstract Introduction: Immune checkpoints (ICPs) play a critical role in modulating the immune response, maintaining self-tolerance, and preventing excessive immune system activation. Dysregulation of immune checkpoints has been implicated in cancer progression and treatment resistance. In this study, we aimed to screen for dysregulated immune checkpoint genes that might be associated with cancer recurrence in pediatric acute lymphoblastic leukemia (ALL) following a course of cancer treatment. We hypothesized that certain ICPs would be significantly dysregulated in relapsed ALL patients, potentially providing insights into the underlying mechanisms driving therapy resistance and disease recurrence. Methods: 51 ICP genomic data (gene expression, copy number alterations (CNAs), and mutation) of133 patients with relapse information [relapse-free (n=29) and relapsed (n=104)] extracted from the Therapeutically Applicable Research to Generate Effective Treatments (TARGET) were examined in the study. Patient’s t-test (p-value < 0.05) was used to identify significant genetic factors between relapse-free and relapsed patients. Free and overall survival analyses were performed using log-rank test (p-value < 0.05) based on patient gene expression profiles. Results: Among the 51 genes examined, PDCD1, PVR, CD70, and CD48 were significantly upregulated in relapsed patients compared with relapse-free patients (P < 0.05). No significant CNAs nor mutations were detected in these four genes. Patients with high expression profiles of PDCD1 and CD48 at diagnosis were 1.71 and 1.47 more likely to relapse compared to patients with low expression profiles, respectively (P < 0.05). Furthermore, the upregulation of PDCD1 (P= 0.0004, HR= 1.972) and CD48 (P= 0.0344, HR= 1.533) demonstrated significantly increased mortality rates in patients following treatment, as indicated by overall survival analyses. Conclusion: The upregulation of PDCD1 and CD48 was significantly associated with relapse and survival of ALL patients. Promisingly, these two ICPs can potentially be used as prognostic biomarkers for relapse risk and rationale for alternative therapeutic strategies. Our study suggests that targeting PDCD1 and CD48 might be beneficial in enhancing the treatment efficiency and thereby reducing the recurrence risk. The study was funded by the Deputyship for Research and Innovation, Ministry of Education, Saudi Arabia (DRI-KSU-1273). Citation Format: Shatha Fawaz Alotaibi, Lames Abuhadi, Noura Alolyani, Moureq R. Alotaibi, Ali R. Alhoshani, Homood As Sobeai. PDCD1 and CD48 significantly impact cancer recurrence and survival in pediatric acute lymphoblastic leukemia [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 5662.

CNS Prophylaxis in Adult ALL Patients after Stem Cell Transplantation: An EBMT Survey on General Practice

Introduction Allogeneic hematopoietic stem cell transplantation (HSCT) is an effective treatment for acute lymphoblastic leukemia (ALL) resulting in long-term remission. Despite advances in therapy, disease progression remains the major cause of mortality following HSCT accounting for 20-50% of all deaths. The central nervous system (CNS) is the most common extra-medullary site of disease progression after transplant in ALL. The established practice of CNS prophylaxis for ALL has largely been modeled in the pediatric population, given the higher incidence of ALL in this age group compared with adults. Although the use of CNS prophylaxis as part of the upfront treatment for ALL has led to significant decreases in CNS relapse and improved outcomes overall, the routine use of post-HSCT prophylactic CNS therapy as a strategy to prevent CNS relapse after transplant in adult patients is still controversial. Studies that utilized post-HSCT CNS prophylaxis have reported disparate results and there is no generalized consensus regarding the use of post-transplant CNS prophylaxis to prevent relapse. Methods In order to assess the current practice of post-HSCT prophylactic CNS therapy in preventing CNS relapse in ALL patients, we carried out a survey among EBMT centres to describe intrathecal (i.t.) prophylaxis or irradiation in HSCT for ALL in both patients with or without CNS manifestations at any time of disease (first or second complete remission [CR1, CR2] and relapsed/refractory [R/R] ALL). EBMT centres were asked whether they usually give i.t. prophylaxis or cranial irradiation before and after HSCT, and which drugs they use. Results Of 410 invited centres, 55 (13%) replied to the survey. Tables 1 shows the number of centres which use i.t. prophylaxis or cranial irradiation. Most of the 55 centres used both pre-and post i.t. prophylaxis and triple prophylaxis (steroids, cytarabine [AraC] and methotrexate [MTX]) was the most frequently adopted. As expected in the pre-transplant phase, almost 70% of centres use i.t. chemotherapy irrespective of the phase of the disease or of the use of total body irradiation (TBI) as conditioning regimen. In the post-transplant phase, the rate of i.t prophylaxis varies according to disease phase, TBI use, and history of CNS involvement pre HSCT. In patients w/o history of CNS involvement in CR1, i.t. prophylaxis is given in 23.6% and 21.5% (with or w/o TBI use, respectively). Notably in CR2 pts, i.t. prophylaxis is given in 20% of cases if the conditioning regimen includes TBI, and in 74.5% of cases if TBI is not used; 32.7% and 46.9% R/R pts (with or without TBI use, respectively) are given i.t. prophylaxis. In the setting of patients with history of CNS involvement in CR1, i.t. prophylaxis is given in and 30.9% and 38.2% (with or without TBI use, respectively); in CR2 pts, i.t. prophylaxis is given in 38.9% of cases if the conditioning regimen does not include TBI, and in 29.1% of cases if TBI is used; 38.9% and 45.3% R/R pts (with or without TBI use, respectively) are given i.t. prophylaxis. Pre-transplant cranial irradiation is planned in less than 10% of centres in CR1 and CR2 patients without CNS involvement, irrespective of the use of TBI in the conditioning regimen, while it is used in 22.4% and 25% of patients with R/R disease with or without TBI use for transplant, respectively. In the setting of patients with CNS involvement, among those in CR1 and CR2, approximately 30% receive pre-transplant cranial irradiation irrespective of TBI use, while 32.7% and 34.6% of patients receive cranial irradiation (with or without TBI use, respectively). Its use after transplant is uncommon irrespective of disease status at treatment. Conclusions CR2 pts (without CNS involvement) seem to be at higher risk of CNS relapse if TBI is not included in the conditioning regimen. R/R pts are always considered at higher risk for CNS relapse if TBI is not included in the conditioning regimen. Cranial irradiation before treatment is used in almost 1 out of 3 pts with CNS involvement despite the use of TBI as the conditioning regimen, and 1 out 4 R/R pts without CNS involvement at transplant. The rate of cranial irradiation after transplant is extremely low, irrespective of diseases status, previous CNS involvement or TBI use. In conclusion, the use of i.t. prophylaxis after HSCT in patients with or without CNS involvement is variable among European transplant centres, also in the era of novel therapies for patients with ALL.

In Silico Analysis of miR-181a and miR-155 and Their Target Genes in the Pathogenesis of Acute Lymphoblastic Leukaemia

Acute lymphoblastic leukaemia (ALL) represents 80% of acute leukaemia in children and 20% in adults. In most cases, the cause remains unknown; however, genetic predisposition is found in most patients. Clinically, a patient is diagnosed with ALL when peripheral blood smear or bone marrow aspirate is overpopulated with lymphoblasts. Although the 5-year overall survival rate exceeds 90% using available treatments, the life-threatening effects still affect some patients [1]. Therefore, further improvement in clinical outcomes and reduction of adverse effects require novel therapeutic approaches such as microRNAs (miRNAs) biomarkers. In addition, miRNAs have demonstrated distinct expression profiles in classifying the ALL subtypes and involved in various cancers [2].
 
 Previously, miR-181a and miR-155 have been discussed, either acting as oncogenic or tumour suppressive miRNAs in carcinogenesis. The study aims to investigate the role of miR-181a and miR-155 and their predicted target genes in the pathogenesis of ALL. Hence, in silico analysis was conducted using miRDB 6.0 software to identify the predicted target genes and GeneCodis 4.0 software to identify signalling pathways regulated by miR-181a and miR-155. As presented in Table 1, miR-181a was found to be downregulated whereas miR-155 was significantly upregulated in ALL groups as compared to control groups [3].
 Analysis of target expression was conducted to identify predicted target genes in ALL cell line, DND-41. In this study, the predicted target genes are regulated by miR-181a and miR-155 (with target score ≥ 80 and target expression ≥ 20) revealed 57 genes and 29 genes, respectively. The most significant genes regulated by miR-181a and miR- 155 were DUSP6 and H3F3A, respectively. DUSP6 is a regulator for hyperactivation of the extracellular signal- regulated kinase (ERK) in relapsed pre-B ALL patients. Thus, this reaction caused lesions in the RAS pathway and oncogenic tyrosine kinases (e.g., BCR-ABL1).
 
 Additionally, a complete transformation of pre-B ALL clones was done by BCR-ABL1 and correlated with the activity of other MAPKs such as MAPK14 and MAP2K1. Therefore, ERK was identified as a negative feedback regulator of the RAS-ERK pathway [4].
 
 H3F3A was reported to be involved in the pathogenesis of paediatric leukaemia caused by mutated genes that encoded proteins during epigenetic regulation. The mutation in H3F3A led to the inability of this residue to undergo periodic regulatory post-translational modifications and blocked trimethylation of all H3 during T-ALL pathogenesis. Specifically, the recurrent T-ALL patients found the lysine-to-arginine mutations at H3K27 and H3K36. Hence, the reduced expression of H3 proteins was associated with unfavourable prognosis [5].
 
 The functional enrichment analysis of KEGG pathway was conducted by using GeneCodis 4.0 software. The most significant pathway targeted by miR-181a and miR-155 was p53 signalling pathway and MAPK pathway, respectively.
 
 SESN3 was mainly responsive to p53 protein by inducing its expression upon oxidative stress and strictly associated with the antioxidant response during p53 signalling pathway. Hence, SESN3 could be a potential therapeutic drug for anticancer therapy, such as decreased mTORC1 mechanism in leukaemia patients [6]. Besides that, FOS was disrupted the MAPK pathway components, specifically MEK1 and MEK2 which could increase sensitivity to chemotherapy in relapsed ALL patients. These components were involved in ERK phosphorylation that critically drives chemoresistance during MAPK/ERK signalling pathway [7].
 
 In conclusion, differentially expressed miRNAs have promising roles and can be used as effective biomarkers for diagnostic, prognostic, predictive and therapeutic effects in ALL patients.

A comprehensive in silico analysis of pathogenic nsSNPs in the NT5C2 gene involved in relapsed ALL

Background: About 10-20% of children suffering from acute lymphoblastic leukemia (ALL), experience a relapse, which is a major cause of their death. Purine nucleotide analogs are frequently prescribed to maintain the treatment of ALL. Cytosolic 5´-nucleotidase (NT5C2) catalyzes the 5´ dephosphorylation of purine analogs. Gain-of-function mutations in the NT5C2 gene result in resistance to the treatment with purine analogs and subsequently in the relapse of the disease.
 Materials and Methods: In this descriptive study, bioinformatics tools were used to assess the effect of single nucleotide polymorphisms (SNPs) in the NT5C2 gene on the function and structure of the protein. So, 352 missense variants were retrieved from the NCBI database and analyzed by SIFT, PROVEAN, PMut, PANTHER, PolyPhen2, SNPs & Go, and PhD-SNP servers. Then, structural evaluations were performed using HOPE, NetSurp-2.0, and PyMOL. Moreover, stability and evolutionary preservation were assessed by I-Mutant2.0 and ConSurf, respectively.
 Results: As many as 31 nsSNPs were predicted to be affecting the protein function and stability. Also, the native residues were found to be evolutionarily preserved. The structural evaluation demonstrated that a change of hydrophobicity, flexibility, size, charge, or surface accessibility due to 24 nsSNPs would lead to the change of noncovalent interactions and then the conformation of the protein.
 Conclusion: Identification of biomarkers is significant in the prediction of relapses in ALL children. In this study, bioinformatics tools served to identify 24 high-risk deleterious nsSNPs in the NT5C2 gene. These mutations can be used to predict resistance to chemotherapy and relapse in ALL patients.

The expense of sending cerebrospinal fluid for analysis on all lumbar punctures in pediatric acute lymphoblastic leukemia patients.

Central nervous system (CNS) relapse in pediatric acute lymphoblastic leukemia (ALL) patients is uncommon. The cerebrospinal fluid (CSF) of patients with ALL is routinely sampled at each intrathecal chemotherapy treatment to screen for CNS relapse. The analysis of CSF is both time consuming and resource intensive and must be completed approximately 20 times per patient throughout treatment. Our objective was to examine the expense of routine screening on all CSF samples for CNS relapse in ALL patients, and to identify if CNS relapse can be detected clinically. We identified all patients diagnosed with ALL at the Children's Hospital of Eastern Ontario (CHEO) between January 2001 and June 2021. We collected the total number of CSF samples in these patients and the number of CSF samples positive for CNS relapse. An in-depth chart review on the patients who relapsed in the CNS was completed to identify symptoms at relapse. Over the study period, 351 patients were diagnosed with ALL and underwent a total of 6515 lumbar punctures (LPs), each of which examined the CSF. The cost of CSF sample analysis is $14.32 (Canadian dollars [CDN]); thus, the total cost for the study sample was $93,294.80 (CDN). There were 14 CNS relapses and although symptoms including headache, vomiting, and fatigue were common, two patients were asymptomatic at relapse. Given the marginal cost of routine CSF screening and the lack of specific and sensitive symptoms for CNS relapse, we conclude that the routine practice of sending all CSF samples for analysis of CNS relapse in ALL patients is relatively inexpensive and beneficial.

Effects of NRAS Mutations on Leukemogenesis and Targeting of Children With Acute Lymphoblastic Leukemia.

Through the advancements in recent decades, childhood acute lymphoblastic leukemia (ALL) is gradually becoming a highly curable disease. However, the truth is there remaining relapse in ∼15% of ALL cases with dismal outcomes. RAS mutations, in particular NRAS mutations, were predominant mutations affecting relapse susceptibility. KRAS mutations targeting has been successfully exploited, while NRAS mutation targeting remains to be explored due to its complicated and compensatory mechanisms. Using targeted sequencing, we profiled RAS mutations in 333 primary and 18 relapsed ALL patients and examined their impact on ALL leukemogenesis, therapeutic potential, and treatment outcome. Cumulative analysis showed that RAS mutations were associated with a higher relapse incidence in children with ALL. In vitro cellular assays revealed that about one-third of the NRAS mutations significantly transformed Ba/F3 cells as measured by IL3-independent growth. Meanwhile, we applied a high-throughput drug screening method to characterize variable mutation-related candidate targeted agents and uncovered that leukemogenic-NRAS mutations might respond to MEK, autophagy, Akt, EGFR signaling, Polo−like Kinase, Src signaling, and TGF−β receptor inhibition depending on the mutation profile.

Small Molecule Inhibition of PDE6D-RAS Interaction Suppresses the Growth of Acute Lymphoblastic Leukemia Cell Lines

Small Molecule Inhibition of PDE6D-RAS Interaction Suppresses the Growth of Acute Lymphoblastic Leukemia Cell Lines

Economic Burden of Hematopoietic Stem Cell Transplantation (HSCT) in Patients with Acute Lymphoblastic Leukemia (ALL) in the US

IntroductionAcute lymphoblastic leukemia (ALL) is a rare disease that affects lymphoblasts derived from lymphoid stem cells and due to the acute nature of this cancer, it is rapidly fatal without treatment (National Cancer Institute, 2016). The average 5-year overall survival (OS) rate was 35% for adult ALL and over 90% of the relapsed or refractory patients eventually die with a median OS of 3-6 months (Kantarjian HM et al. 2010, Bassan and Hoelzer 2011, Forman and Rowe 2013). Allogeneic hematopoietic stem cell transplantation (HSCT) is a potentially curative option for patients with B-cell ALL. Previous studies find that HSCT is associated with high costs and healthcare resource use. However, there is no study examining the allogeneic HSCT costs specifically for the adult ALL patients.ObjectiveThe aim of this study is to quantify the economic burden of allogeneic HSCT for adult B-cell ALL patients and relapsed adult B-cell ALL patients in the US.MethodsThis was a retrospective cohort study using Truven Health MarketScan® Commercial and Medicare Supplemental Databases. Patients with ≥1 inpatient claims or ≥2 non-diagnostic outpatient claims at least 30 days apart for any ALL diagnosis (ICD-9-CM 204.00-204.02, ICD-10-CM C91.00-C91.02) during 4/1/2009 - 10/31/2016 were extracted. All patients were required to have ≥1 hospitalization with diagnosis-related group in 834-839 on or after the first ALL diagnosis, have ≥1 allogeneic HSCT hospitalization (index date is the admission date for the initial HSCT hospitalization), are ≥18 years old on index date, and have ≥6 months of continuous enrollment prior to index date. Patients with medications for T-Cell ALL during the study period were excluded. Relapsed ALL patients were required to have ≥1 inpatient claims or ≥2 non-diagnostic outpatient claims at least 30 days apart for relapsed ALL diagnosis (ICD9: 204.02, ICD10: C91.02). Patients were followed until the earliest of inpatient death, end of continuous enrollment, or end of the study period (10/31/2016). Outcomes included all-cause and HSCT-related costs and total hospital length of stay (LOS) observed during the index hospitalization, and during 100, 180, or 365 days post-index periods (for the subset of patients with at least 100, 180, or 365 days of continuous enrollment after index date or an inpatient death during that follow-up). The costs for the HSCT-related complications such as graft versus host disease (both acute and chronic), hepatic veno-occlusive disease (VOD), graft failure, serious infections, bronchiolitis obliterans, secondary malignancies, and inpatient death were included, as well as the costs for transplant conditioning regimens during the 14 days before HSCT. Costs were sum of the plan paid amount and out-of-pocket amount in 2016 dollars. All descriptive analyses were conducted for patients with ALL (including both frontline and relapsed ALL) and relapsed ALL, separately.ResultsA total of 692 ALL patients (mean age: 45.6; male: 56.1%; mean Charlson comorbidity index: 3.2) were identified. Among all the ALL patients, 123 had relapsed ALL. Among ALL patients, mean (SD; median) costs were $275,642 ($316,109; $212,742) and mean (SD) LOS was 31.1 (20.9) days during the initial HSCT hospitalization (Table 1). During the first year post index, mean (SD; median) all-cause and HSCT-related costs were $543,146 ($416,148; $455,245) and $394,069 ($378,893; $301,977), respectively, with all-cause and HSCT-related mean (SD) LOS of 52.5 (41.5) and 49.2 (40.3) days. Among relapsed ALL patients, mean (SD; median) costs were $319,643 ($310,479; $255,545) during the initial HSCT hospitalization (Table 2). Mean (SD; median) all-cause and HSCT-related costs were $680,503 ($651,399; $556,244) and $535,245 ($577,152; $395,031) during the first year post index. Mean (SD) LOS was 33.9 (18.0) days during the initial HSCT hospitalization and 61.5 (46.4) and 59.1 (45.8) days during the first year post index for all-cause and HSCT-related hospital stays, respectively. The initial HSCT hospitalization accounts for about 47% and 51% of the total costs in the first year after HSCT for ALL and relapsed ALL patients, respectively.ConclusionsAllogeneic HSCT in ALL is associated with substantial economic burden, especially among relapsed ALL patients. During the first year post HSCT, about half of the total costs occurred during the initial HSCT hospitalization. [Display omitted] DisclosuresZhang:Amgen Inc.: Employment, Equity Ownership. Cong:Amgen Inc.: Employment, Equity Ownership.

MiR-324-3p and miR-508-5p expression levels could serve as potential diagnostic and multidrug-resistant biomarkers in childhood acute lymphoblastic leukemia

Acute lymphoblastic leukemia (ALL) is one of the most frequent hematological malignancies in children, representing approximately 25 % of all pediatric cancers. Despite striking advances in ALL treatments, a small population of patients does not still respond to chemotherapy, raising the number of deaths in children. ABC transporters are one of the major causes of multidrug resistance (MDR) in cancers and overexpression of ABCA3 is directly associated with increased chemo-resistance in pediatric ALL. Here, we aimed to identify the microRNAs (miRNAs) which may regulate the expression of ABCA3 in childhood ALL. Bone marrow samples from a total of 50 ALLs and 59 controls were collected and after in silico and literature search, miR-324-3p and miR-508-5p were nominated from a list of putative miRNAs targeting ABCA3. Our qPCR analysis showed a low expression profile of selected miRNAs in pediatric ALL patients compared with non-cancer controls. Furthermore, we found that both miR-324-3p and miR-508-5p were significantly differentially expressed between patients with positive and negative minimal residual disease (MRD + vs MRD-) after one year of chemotherapy while only miR-508-5p was underexpressed in relapsed ALL patients. Additionally, a negative correlation was identified between the expression of these two miRNAs and ABCA3, supporting the regulatory effect of them on drug resistance through interacting with ABCA3. Overall, we suggested miR-324-3p and miR-508-5p as potential diagnostic and drug-resistant biomarkers in pediatric ALL. Moreover, our findings presented miR-508-5p to behave as a promising relapsed indicator in childhood ALL which can be applied in the development of novel therapeutic strategies.

Relapsed Childhood Acute Lymphoblastic Leukemia: A Single-Institution Experience.

BackgroundEven though the treatment outcomes of childhood acute lymphoblastic leukemia (ALL) have improved recently, relapse of the disease still remains a challenge in developing countries. This study aims to analyze the incidence of relapse and survival rates in childhood ALL.MethodsA retrospective study of 156 children with de novo ALL between 2012-2018 was conducted. Data on age, gender, relapse type, and relapse time were analyzed.ResultsA total of 26 (16.7%) patients experienced relapse, with a male-to-female ratio of 2.71:1. The relapse rate in the high-risk group was 1.6 times greater than that in the standard-risk group (61.5% vs. 38.5%). The median time from diagnosis to relapse was 29.3 months (38.5% in the early stage, 26.9% in the intermediate, and 34.6% in the late stage). The most common relapse site was bone marrow (38.5%), followed by the isolated central nervous system (CNS, 23.1%) and CNS plus bone marrow (23.1%); the least common site was testicle with or without bone marrow or CNS (15.2%). The median post-relapse survival time was 7.5 months.ConclusionModification of the protocol to use escalated methotrexate dose and providing new therapies such as stem cell transplantation can improve the overall survival rates in relapsed ALL patients.

Ras pathway mutation feature in the same individuals at diagnosis and relapse of childhood acute lymphoblastic leukemia.

Acute lymphoblastic leukemia (ALL) is the most common malignancy in children, while relapse and refractory ALL remains a leading cause of death in children. However, paired ALL samples of initial diagnosis and relapse subjected to next-generation sequencing (NGS) could construct clonal lineage changes, and help to explore the key issues in the evolutionary process of tumor clones. Therefore, we aim to analyze gene alterations during the initial diagnosis and relapse of ALL patients and to explore the underlying mechanism. Targeted exome sequencing technology was used to detect molecular characteristic of initial diagnosis and relapse of ALL in 12 pediatric patients. Clinical features, treatment response, prognostic factors and genetic features were analyzed. In our 12 paired samples, 75% of pre-B-cell acute lymphoblastic leukemia (B-ALL) patients had alterations in the Ras pathway (NRAS, KRAS, NF1, and EPOR), and Ras mutation are very common in patients with ALL relapse. TP53 mutations mainly existed in the primary clones and occurred at the initial diagnosis and relapse of ALL. Relapse-associated genes such as NT5C2 and CREBBP were observed in patients with ALL relapse; however, all patients included in this study had gene abnormalities in the Ras pathway, and NT5C2 and CREBBP genes may collaboratively promote ALL relapse. Among the 12 ALL patients, Ras pathway mutations are common in ALL relapse and may be associated with other recurrence-related genes alterations. The study with paired samples could improve the understanding of ALL relapse.

Abstract C005: NT5C2 small molecule inhibitor for the reversal of 6-MP resistance in acute lymphoblastic leukemia

Abstract Activating mutations in the cytosolic 5’-nucleotidase II (NT5C2) gene drive resistance to chemotherapy with 6-mercaptopurine (6-MP) in acute lymphoblastic leukemia (ALL) highlighting NT5C2 as a potential therapeutic target. In efforts to reverse 6-MP therapy resistance, we performed high throughput screens to identify small molecule inhibitors of NT5C2. Using a high throughput malachite-green based NT5C2 nucleotidase assay, we identified HTP_2 as a lead compound with low micromolar inhibitory activity against NT5C2 R367Q mutant recombinant protein. HTP_2 reverses 6-MP resistance in Nt5c2+/R367Qmutant mouse lymphoblasts and NT5C2 R29Q mutant expressing human cell lines. In addition, HTP_2 treatment results in increased sensitivity to 6-MP therapy in NT5C2 wild-type cells, highlighting the role of wild-type NT5C2 activity in the clearance of 6-MP. These results support a potential therapeutic use for NT5C2 inhibitors in potentiating the effects of 6-MP based chemotherapy in NT5C2 wild-type cells as well as NT5C2 mutant cells. Overall, this study presents a framework for the development of a NT5C2 inhibitor for the reversal of 6-MP resistance in relapsed ALL patients. Citation Format: Chelsea Dieck, Clara Reglero-Gomez, Arie Zask, Brent Stockwell, Adolfo Ferrando. NT5C2 small molecule inhibitor for the reversal of 6-MP resistance in acute lymphoblastic leukemia [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference on Molecular Targets and Cancer Therapeutics; 2019 Oct 26-30; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2019;18(12 Suppl):Abstract nr C005. doi:10.1158/1535-7163.TARG-19-C005

Circulating microRNAs as minimal residual disease biomarkers in childhood acute lymphoblastic leukemia

BackgroundTreatment stratification based on bone marrow minimal residual disease (MRD) at set time points has resulted in considerably improved survival in pediatric acute lymphoblastic leukemia (ALL). Treatment response is assessed using bone marrow samples. MicroRNAs (miRs) easily traffic among fluid spaces and are more stable than most other RNA classes. We examined the role of circulating miRs as putative less invasive MRD biomarkers.MethodsIn an exploratory experiment, expression of 46 preselected miRs was studied in platelet-free blood plasma samples of 15 de novo, 5 relapsed ALL patients and 10 controls by Custom TaqMan Array Advanced MicroRNA Card. Based on their high expression in ALL compared to controls, and on the reduction observed along the induction therapy, four miRs were selected for further analyses: miR-128-3p, -181a-5p, -181b-5p and 222-3p. Their expression was measured by qPCR at 4 time points in 27 de novo ALL patients treated in the ALL IC-BFM 2009 study.ResultsThe expression of all 4 miRs significantly decreased over the first week of therapy (miR-128-3p: log2 fold change − 2.86; adjusted p 3.6 × 10−7; miR-181b-5p: log2 fold change − 1.75; adjusted p 1.48 × 10−2; miR-181a-5p: log2 fold change -1.33; adjusted p 3.12 × 10−2; miR-222-3p: log2 fold change − 1.25; adjusted p 1.66 × 10−2). However, no significant further reduction in miR expression was found after the 8th day of therapy. Measured drop in expression of 2 miRs at day 8 strongly correlated with day 15 bone marrow flow cytometry MRD results (miR-128-3p: Pearson’s r = 0.88, adjusted p = 2.71 × 10−4; miR-222-3p: r = 0.81, adjusted p = 2.99 × 10−3).ConclusionIn conclusion, these circulating miRs might act as biomarkers of residual leukemia. MiR-128-3p and miR-222-3p in blood predict day 15 flow cytometry MRD results 7 days earlier. Although, their sensitivity falls behind that of bone marrow flow cytometry MRD at day 15.

Loss of Heterozygosity in the Short Tandem Repeat (STR) Loci Found in Tumor DNA of De Novo Diagnosed ALL Patients As a Factor Predicting Poor Outcome

Introduction: Loss of heterozygosity (LOH) was described for many malignancies including leukemia. Using chromosomal microarray analysis (CMA) we have shown preliminary (Risinskaya et al. HemaSphere, 2019, V.3, No S1, P.782) that LOH in STR loci in tumor cells with the normal karyotype represents the loss of extended chromosomal areas and could be associated with uniparental disomy of the entire chromosomal arm or the entire chromosome. Uniparental disomy results not only in the elimination of the wild type allele, but also may lead to the duplication of the mutant allele of certain candidate tumor gene(s). Retrospective analysis of archival samples have shown that in most cases STR loci absent at the relapse of disease were already lost at the onset of disease. We noted LOH at STR loci is especially frequent for relapsed ALL patients and could be used as prognostic factor for possible poor clinical outcome. The aim: To identify LOH in the blast cells of the patients with ALL at the onset of disease and to analyze therapy outcome relative to the patients without LOH. We expect LOH in the de novo ALL to be an unfavorable factor, and those patients possibly should be treated like high-risk group. Patients and Methods: This study includes a comparative analysis of the STR profiles of the DNA of tumor and normal cells from bone marrow samples on a cohort of 32 patients with de novo diagnosed Ph-negative ALL undergoing treatment according to the "RALL-2016" regimen at the National Research Center for Hematology (Moscow, Russia). Inclusion criteria: de novo diagnosed Ph-negative acute lymphoblastic leukemia (ALL) patients, 18-55 years old, intermediate risk group without MLL translocation t(4;11)(q21;q23), treatment by RALL-2016 protocol. Exclusion criteria: other diagnosis, adult patients older than 55 years, MLL translocation t(4;11)(q21;q23), pretreatment or treatment by other protocol. The presence of blast cells in bone marrow samples was confirmed morphologically. DNA was isolated from patient bone marrow samples taken at different disease stages. Control DNA samples were taken from blood of patients in complete remission and/or from buccal epithelium (buccal swab). STR-profiles were assessed by PCR with COrDIS Plus multiplex kit for amplification of 19 polymorphic STR-markers and amelogenin loci (Gordiz Ltd, Russia). The fragment analysis was performed on ABI3130 Genetic Analyzer. The data processing was accomplished using GeneMapper v.4-0 software. A multivariate survival analysis was used to assess the independent impact of the LOH as risk factor against the background of the abnormal tumor karyotype as standard factor for this cohort of patients. We have chosen Failure Free Survival (FFS) as primary endpoint for short-time follow-up and small patient sample size. Death from any reasons, relapse and second leukemia were chosen as failure events, survival time interval starts from begin of treatment. SAS V9/4 was used for statistical analysis of the data. Results: Of the 32 patients six were found with LOH in the certain STR loci (19%). Of these six, two were resistant to therapy and died from disease progression. One patient with LOH is currently in relapse and one patient with multiple LOH has a second leukemia (Fig.1a). Only two from six patients are "Failure Free" now, whereas only three patients from 26 in group without LOH died (two from infectious complications and one from GvHD after BMT). P value < 0,05 (Fig.1b). Conclusions: We have found statistically significant association of clinical failures with the LOH in STR loci measured at the onset of ALL. We assume that LOH is an unfavorable prognostic factor for de novo ALL patients and could be used for risk stratification and choice of adequate therapy implying allogeneic hematopoietic stem cells transplantation or modern innovative therapy approaches. Disclosures No relevant conflicts of interest to declare.

Pharmacological Inhibition of MAP2K7 Induces Apoptosis and Cell Cycle Arrest in T-Cell Acute Lymphoblastic Leukemia

Acute lymphoblastic leukemia (ALL) is the most common pediatric cancer and relapsed ALL is a leading cause of cancer-related death in children. While advances in frontline therapy have led to an 85% cure rate, relapsed ALL patients face a dismal prognosis necessitating identification of novel targets and development of alternative therapies. Amongst the two sub types of ALL, T-cell acute lymphoblastic leukemia (T-ALL) occurs less frequently but T-ALL patients have worse prognosis and higher rate of relapse. Our group discovered that the transcription factor KLF4 is heavily repressed via DNA methylation coinciding with aberrant activation of MAP2K7 and the MAP2K7 pathway (Shen et al. Leukemia, 2017). Additionally, we demonstrated, as a proof of principle, that pharmacological inhibition of JNK, the only downstream target of MAP2K7, possesses anti-leukemic properties against T-ALL but presented obstacles in terms of low potency and off-target effects. In order to overcome these limitations, we hypothesize that direct pharmacological inhibition of MAP2K7 improve specificity for targeted approach to T-ALL therapeutics. To examine this hypothesis, we explore the anti-leukemic effects of the MAP2K7 pathway inhibitor, 5Z-7-oxozeaenol, against T-ALL cells because it has been shown to covalently react with a unique cysteine-218 in the ATP binding pocket of MAP2K7. We found that T-ALL cell lines exhibit increased cytotoxic sensitivity to 5Z-7-oxozeaenol (IC50 ranging 0.2-1.1 µM) compared to a non-leukemic control (1.5 µM), and more potent than JNK inhibitor IC50 of 5µM. Additionally, 5Z-7-oxozeaenol reduces the amount of phospho-JNK, in a dose dependent manner, indicative of MAP2K7 pathway inhibition. We purified MAP2K7 protein to directly address specificity of kinase inhibition and found that 5Z-7-oxozeaenol inhibits enzymatic activity in vitro of MAP2K7. In addition, 5Z-7-oxozeaenol also inhibited TAK1, which is the MAP3K upstream of MAP2K7. In a panel of T-ALL cell lines, 5Z-7-oxozeaenol treatment induced apoptosis in MOLT3, Jurkat, and KOPTK1 T-ALL cell lines. Although P12-Ichikawa, RPMI-8402, DND-41, and ALL-SIL T-ALL cell lines underwent cell cycle arrest evidenced by a reduction in percentage of S/G2/M phase cells and increase in percentage of G0 phase cells, further increasing 5Z-7-oxozeaenol dosage proved sufficient for induction of apoptosis based on increase of caspase-3 and caspase-7 cleavage. Through reverse-phase protein array analysis we identified reduced expression of several cell cycle regulator proteins, including CDC25C, Cyclin B1, Cyclin D3, and CDC2 resulting from 5Z-7-oxozeaenol treatment. Conversely, we detected increased expression of cleaved caspase-3 and caspase-7, particularly in MOLT3 and Jurkat cell lines treated with MAP2K7 inhibitor. Immunoblot analysis revealed that 5Z-7-oxozeaenol inhibits the MAP2K7 signaling pathway and induces cell cycle arrest and apoptosis in T-ALL cells. Based on these findings, we demonstrated that 5Z-7-oxozeaenol induces cell cycle arrest and apoptosis in T-ALL cells through inhibition of the MAP2K7 pathway, suggesting that MAP2K7 represents a novel pharmacological target for the development of targeted therapy for high-risk patients with T-ALL. Disclosures No relevant conflicts of interest to declare.

PF188 NGS COMBINED WITH ASO REAL‐TIME PCR BASED MRD DETECTION ASSAY FOR PEDIATRIC ALL PATIENTS

Background:Monitoring of minimal residual disease (MRD) has become routine clinical practice in frontline treatment of pediatric acute lymphoblastic leukemia (ALL) patients. MRD diagnostics has proven to be the strongest prognostic factor, allowing for risk group assignment into different treatment categories, ranging from significant treatment reduction to mild or strong intensification. MRD diagnostics is also guiding treatment decisions in relapsed ALL patients and patients undergoing stem cell transplantation.Aims:Our aim is to develop and apply a sensitive (≤ 10–4), universal, accurate, reliable and fast assay for MRD detection in ALL patients. To date, flow cytometry and polymerase chain reaction (PCR) analysis of rearranged immunoglobulin and T‐cell receptor genes (allele‐specific oligonucleotide [ASO]‐PCR) are claimed to meet these criteria, but classical flow cytometry does not reach a solid 10–4, whereas classical ASO‐PCR is time‐consuming and labor intensive.Methods:LymphoTrack® IGH + TRG Assay – MiSeq is applied on diagnosis sample using the Illumina MiSeq platform. This is used for the identification of clonal IGH VH‐JH and TRG V‐J rearrangements, DNA sequences and the distribution frequency of V and J segment utilization. The clone sequence is then compared against the germline sequences contained in the IMGT V‐QUEST databases for the identification of patient‐specific junctional regions following by the design of six different D‐N‐J specific ASOs primers for each clone according to UKALL protocol. Five point standard curve is used with triplicates at 5x10^‐5 and 1x10^‐5 for more reliable results at the critical lower level and six pooled negative controls for confirmation of the specificity of the assay.Results:Forty pediatric ALL patients have been tested at several time points. All informative patient sample dilutions reached 5x10^‐5, and most of the patients reached 1x10^‐5. The R2 of standard curves were within the expected range (≥0.98), and all slopes within range (3.1–3.9) with an average of 3.21 +/‐ 0.16. In most negative control samples no amplification was detected. In 8 assays amplification was detected in two negative control samples. In these cases the CT value was more than 3 CT from the Quantitative Range (QR).Our coupled NGS‐ASO‐PCR method sensitivity was compared to Flow cytometry and Fusion Transcript Real Time PCR methods. NGS‐ASO based MRD method has detected MRD populations in several samples were Flow cytometry or real time PCR failed. None of the patient tested were positive by other methods and was negative by ASO based MRD method.Summary/Conclusion:In the ALL field, MRD has become part of diagnostic patient care. Consequently, standardized MRD diagnostics should be available for assessment of treatment response in each individual ALL patient, to be used for personalized medicine such as accurate risk group assignment with risk‐adapted treatment. The coupled NGS‐ASO‐PCR method seams to fulfill a series of requirements for acceptance in the field, such as broad availability, easy implementation, applicability in the vast majority of patients, sufficient sensitivity (quantitative range of 5x10^‐5) and fast (short turn‐around time, particularly for follow‐up samples).

High sensitivity and clonal stability of the genomic fusion as single marker for response monitoring in ETV6-RUNX1-positive acute lymphoblastic leukemia.

Assessment of minimal residual disease (MRD) is an integral component for response monitoring and treatment stratification in acute lymphoblastic leukemia (ALL). We aimed to evaluate the genomic ETV6-RUNX1 fusion sites as a single marker for MRD quantification. In a representative, uniformly treated cohort of pediatric relapsed ALL patients (n=52), ETV6-RUNX1 fusion sites were compared to the current gold standard, immunoglobulin/T-cell receptor (Ig/TCR) gene rearrangements. Primer/probe sets designed to ETV6-RUNX1 fusions achieved significantly more frequent a sensitivity and a quantitative range of at least 10-4 compared to the gold standard with 100% and 73%versus 76% and 47%, respectively. The breakpoint sequence was identical at diagnosis and relapse in all tested cases. There was a high degree of concordance between quantitative MRD results assessed using ETV6-RUNX1 and the highest Ig/TCR marker (Spearman's 0.899, P<.01) with differences>½ log-step in only 6% of patients. A high proportion of ETV6-RUNX1-positive ALL relapses (40%) in our cohort showed a poor response to induction treatment at relapse, and therefore had an indication for hematopoietic stem cell transplantation, demonstrating the need of accurate identification of this subgroup. ETV6-RUNX1 fusion sites are highly sensitive and reliable MRD markers. Our data confirm that they are unaffected by clonal evolution and selection during front-line and second-line chemotherapy in contrast to Ig/TCR rearrangements, which require several markers per patient to compensate for the observed loss of target clones. In future studies, the genomic ETV6-RUNX1 fusion can be used as single MRD marker.

Digital-Droplet PCR, an Accurate Method for IG/TR PCR-MRD Stratification in Childhood Acute Lymphoblastic Leukemia

Digital-Droplet PCR, an Accurate Method for IG/TR PCR-MRD Stratification in Childhood Acute Lymphoblastic Leukemia