End-induction Minimal Residual Disease Research Articles

Prevalence, clinical and haematological profile and outcome of BCR-ABL1-like acute lymphoblastic leukaemia in Indian children: A prospective observational study

BackgroundBCR-ABL1-like acute lymphoblastic leukaemia (ALL) represents a perplexing subclass of ALL mainly due to genomic heterogeneity and non-availability of standardized diagnostic tests. The data on prevalence, clinical–haematological profile and outcome are limited, more so from low-middle income countries. MethodsThis prospective observational study enrolled children<14 years with B-ALL. Recurrent genetic/chromosomal aberrations were excluded. BCR-ABL1-like aberrations were analysed using Polymerized Chain Reaction (PCR) or Next-Generation Sequencing (NGS). Two groups with/without BCR-ABL1-like ALL (Mut+/Mut-) were compared. ResultsOut of 214 eligible children; 75 with “B-other ALL” were analysed for BCR-ABL1-like aberrations. Their prevalence was 34/214 (15.8%) in B-ALL and 34/75 (45.3%) in “B-other” ALL (PCR:33.3% and NGS:53.3%). Majority of the aberrations were JAK2E16 and ILR7e5/ILR7e6 with PCR and RAS pathway and PAX5 fusion with NGS. Baseline demographic, clinical and laboratory parameters including aberrant flowcytometry were comparable in Mut+ and Mut-groups. Children with day-8 absolute blast count (ABC) were higher in Mut+ group. High end-induction minimal residual disease (MRD) was comparable in two groups. Number of NCI-standard risk (SR) at diagnosis with high D-8 ABC or high MRD was higher in Mut+ group. Similar results were found when aberrations were analysed by NGS alone. In all 56 children are alive, 19 had an event (relapse/death). The outcomes were comparable in two groups with median follow-up of 1075 days (IQR: 660–1527); when analysis was based on combined as well as NGS-based methodology alone. ConclusionsWe report 15.8% prevalence of BCR-ABL1-like ALL in children by PCR or NGS. High D-8 ABC was associated with BCR-ABL1-like ALL with no impact on outcomes.

Outcome of Infants Treated on Total Therapy for Infants with Acute Lymphoblastic Leukemia I: Results from a Non-Randomized Multi-Center Study

Infants with KMT2A rearranged (KMT2Ar) acute lymphoblastic leukemia (ALL) have a poor prognosis with an event-free-survival (EFS) of 33.6-36.9% on recent Children's Oncology Group and Interfant trials. In preclinical studies, we identified bortezomib and vorinostat as novel active agents in primary KMT2Ar infant ALL specimens that have not been included in infant regimens to date. Total therapy for infants with acute lymphoblastic leukemia I (TINI) evaluated this novel combination in a multi-institutional pilot trial for infants with de novo ALL. In this study, bortezomib and vorinostat were incorporated into an ALL chemotherapy backbone containing dexamethasone, mitoxantrone, and pegasparaginase during induction and reinduction chemotherapy cycles. A run-in dose escalation evaluated 3 dose levels of vorinostat which was then followed by a dose expansion phase. The primary endpoint of the study was tolerability; secondary endpoints included minimal residual disease (MRD), 3-year EFS and overall survival (OS). Fifty patients enrolled between 2017-2021. Median follow up was 3.4 years (range 1.6-6.8 years). The six patients in the dose escalation phase had no dose-limiting toxicities. Forty-four patients received vorinostat at the third dose level (DL3) of 180 mg/m 2/dose. 68% of patients treated at DL3 had CNS involvement at diagnosis. 68% of DL3 patients carried a KMT2Ar. The most frequent KMT2A partner genes were AFF1 (40%), MLLT1 (20%), and MLLT3 (13%). The most frequent grade 3-4 non-hematologic adverse events during induction include neutropenic fever (45%), hypertension (43%), infection (40%), anorexia (32%), diarrhea (18%) and perineal skin ulceration (16%). There were 3 episodes of sepsis and 1 induction death due to parainfluenza pneumonia. Fewer events and no deaths or proven sepsis occurred during reinduction, with only grade 3 hypertriglyceridemia exceeding 10% of patients affected (18%). 79% of all patients and 69% of KMT2Ar patients became MRD negative by flow cytometry on day 22 of induction following our investigational block. Following induction intensification with cyclophosphamide, cytarabine, and mercaptopurine, 77% and 66% of all patients and KMT2Ar patients respectively were MRD negative at the time of count recovery. Eight of the forty-four patients, all KMT2Ar, underwent stem cell transplant in first remission due to persistent MRD positivity at the end of consolidation. The 3-year EFS and OS probabilities (SE) of all forty-four DL3 patients were 56.5% (8.0) and 70.5% (7.3), respectively. The 3-year cumulative incidence of relapse (CIR) with death considered a competing event was 29.3% (6.9) and the cumulative incidence of death (CID) as a first event was 8.9% (4.5). Patients lacking a KMT2Ar had superior outcomes with 85.7% (9.4) EFS and 92.9% (6.9) OS. The EFS and OS of KMT2Ar patients were 44.8% (9.5) and 62% (9.1) respectively. CIR and CID for KMT2Ar patients was 43.5% (9.6) and 7.2% (5) respectively. KMT2Ar patients ≥90 days at diagnosis had superior outcomes compared to their younger counterparts with 59.1% (10.5) EFS and 77.3% (8.9) OS. End of induction MRD was predictive of an event in this subgroup, with 85.6% (9.4) EFS for MRD negative patients and 14.3% (13.2) EFS for MRD positive patients. In contrast, KMT2Ar patients &amp;lt; 90 days of age at diagnosis had poor outcomes irrespective of MRD with 12.5% (11.7) EFS and 25% (15.3) OS. Notably KMT2Ar patients &amp;lt; 90 days who were MRD negative at the end of induction had a median time to relapse of 600 days (range 282-969 days), a late time point for this malignancy that typically relapses within a year from initial diagnosis. This suggests the regimen effectively reduced tumor burden leading to an extended duration of remission but failed to sufficiently eradicate disease to prevent relapse. We conclude that the incorporation of bortezomib and vorinostat for treatment of infants with ALL was well tolerated. MRD responses in KMT2Ar patients were robust. Although the study was not powered to evaluate outcomes, EFS and OS in KMT2Ar patients suggest a clinical signal worth pursuing given the low proportion of patients transplanted in first remission. The successor study, TINI 2 (NCT05848687), will build upon the bortezomib/vorinostat backbone by incorporating 2 cycles of blinatumomab and the menin inhibitor ziftomenib in combination with chemotherapy during reinduction.

Single-Cell Transcriptomics Points to a Role of Stemness and Bone Marrow Niche Interactions in the Poor Response to Induction Therapy in T-Cell Acute Lymphoblastic Leukemia

T-cell acute lymphoblastic leukemia (T-ALL) has a poor prognosis with a 15-20% relapse rate. The induction therapy response is the strongest predictor of relapse and shows great variation between patients. However, whether intra-patient leukemic cell heterogeneity plays a role for the induction therapy response has not yet been investigated. We studied 13 patients with T-ALL (11 pediatric and 2 young adult, median follow-up 37 months). Of these, three were good responders (end of induction (EOI) minimal residual disease (MRD) undetectable/non-quantifiable by flowcytometry and PCR), while ten had a poor response (EOI MRD&amp;gt;0.01%, (range 0.05-97%). Of the ten poor responders, three experienced relapse and one died from resistant disease. Bone marrow samples from diagnosis (n=13) and paired samples from relapse (n=3) or MRD (n=4, MRD &amp;gt;5%) were subjected to combined single-cell RNA sequencing, single-cell surface marker profiling by antibody-derived tag (ADT)-sequencing, and single-cell TCR sequencing, using the 10X Genomics platform. A published healthy bone marrow dataset was used for cell annotation and clear distinction was observed between healthy cell subsets and leukemic blasts. TCR sequencing data revealed a hyperexpanded clone in six patients, with identical clonal sequences at diagnosis and relapse(n=2)/MRD(n=1). Surface marker profiling using a panel of 134 surface antigens showed good correspondence between leukemia-associated immunophenotypes found by ADT-sequencing and by flow cytometry for 16 common clinically used markers. To examine factors associated with a poor therapy response, we compared average gene expression of leukemic cells in poor and good responders at diagnosis using differential gene expression (DGE) and gene set enrichment analysis (GSEA). At diagnosis, poor responders showed upregulation of genes involved in extracellular matrix (ECM) degradation/organization, including the adhesion molecules ITGA5 and PECAM1, and MMP11, while showing downregulation of cell cycle genes. Accordingly, the fraction of cells predicted to be in G0/G1 phase at diagnosis of poor responders positively correlated with EOI MRD (r=0.52, p=0.07, Spearman's rank correlation). To investigate if this high fraction of quiescent cells in poor responders was associated with a more stem-like cell state, we examined the expression of a published hematopoietic stem cell (HSC) signature (Forsberg et al, 2010) that we previously selected for investigation in BCP-ALL (Modvig et al, Molecular Oncology, 2022). Indeed, patients with a poor response had a higher percentage of leukemic blasts with an HSC signature module score above zero (p=0.028, Kolmogorov-Smirnov, distribution in Figure 1), suggestive of stemness as a contributing factor to therapy resistance. To define molecular changes in leukemic blasts during therapy, we compared the leukemic blast gene expression profiles in matched diagnostic and MRD samples using DGE and GSEA. This revealed that the leukemic blasts within the first month of therapy downregulated cell cycle genes, while upregulating chemokine receptors, such as CXCR4. To identify specific genes associated with high MRD, we combined genes upregulated in poor responders (n=314, p&amp;lt;0.01, DGE) with genes associated with EOI MRD levels (n=98, r&amp;gt;0.8 and p&amp;lt;0.001, rank regression). This resulted in a 30-gene signature, comprising several genes previously associated with cancer stem cells, such as MICAL3, ELF4, and NLRP3, and genes encoding proteins involved in cancer drug resistance. These included sialyl transferases, acid ceramidase ( ASAH1) and zyxin ( ZYX). ZYX silencing was previously shown to downregulate the anti-apoptotic protein BCL-2 in leukemia. Accordingly, the percentage of cells with detectable BCL2 expression at diagnosis correlated with EOI MRD (r=0.58, p=0.04), supporting a role of BCL-2 inhibitor therapy in T-ALL. In conclusion, our data suggest that therapy-resistant cells are more stem-like and quiescent at diagnosis, display upregulation of genes involved in bone marrow niche ECM organization and adhesion, and genes related to drug resistance. Likewise, residual leukemic blast following therapy show less proliferation and CXCR4 upregulation. Thus, single-cell transcriptomics provides important molecular insights into therapy-resistant leukemic cells and could comprise a valuable prognostic tool in T-ALL.

Delayed Intensification Including Venetoclax and Bortezomib Prolongs Survival in Very High Risk Acute Lymphoblastic Leukaemia

Modern multi-drug risk-stratified response-adapted chemotherapy protocols lead to &amp;gt;85% cure rates in childhood acute lymphoblastic leukaemia (ALL). Between 2013-2022, 707 children (0-18 years) with acute lymphoblastic leukaemia (ALL) treated systematically on the ICiCLe-ALL-14 risk-adapted protocol achieved a 3-year event-free survival (EFS) of 73% (95% confidence interval, CI 69-77). A very high risk (VHR) subgroup (N = 107) was identified with poor outcome (3-year EFS, 38% [95% CI 25-51]). This included patients with induction failure, persistent minimal residual disease (MRD) and/or select high risk cytogenetic subtypes ( BCR::ABL1 other ABL-class fusion ALL, TCF3::HLF, low hypodiploidy, IKZF1plus with high end-induction MRD, KMT2A-rearranged infant ALL). Co-culture based high-throughput drug response profiling (DRP) was performed to identify alternate chemotherapeutic drugs for these VHR patients. Primary ALL blasts extracted by density centrifugation, were seeded on immortalized human bone marrow stromal cells and treated with a 14-drug panel: prednisolone; vincristine (VCR), daunorubicin; asparaginase (ASNase), bortezomib (BZM), venetoclax (VEN), panobinostat (PNB), mitoxantrone, cyclophosphamide, cytarabine, selinexor, dasatinib, 6-thioguanine and idarubicin. Seventy-four patients (25 VHR, 37 high-risk, 7 intermediate-risk, and 5 standard-risk,) were analysed. Venetoclax, BZM and PNB were identified as active agents (IC 50 &amp;lt;100nM for VEN and PNB, &amp;lt;10nM for BZM) in 71%, 42% and 75% of VHR samples, respectively. Based on drug-sensitivity patterns of VHR ALL, a pilot study was initiated to prospectively evaluate the addition of VEN and BZM in the frontline delayed intensification (DI) phase of VHR ALL patients (N=24) after informed consent (Figure 1). Response was assessed before and after VHR-DI in 46% (11/24) patients. Good response, i.e. decrease in MRD ≥ 1 log after VHR-DI, was noted in 82% (9/11) patients. Landmark analysis comparing survival of patients (from start of DI) treated on standard DI versus VHR-DI showed significant improvement in 2-year EFS from 54% (95% CI 40-66) to 76% (95% CI 51-89) ( p=0.037, log-rank test). Overall, VHR-DI was tolerated well. No significant difference was noted in grade 4 and 5 toxicities (Common Terminology Criteria for Adverse Events v4.0) between standard DI and VHR-DI ( p= 0.053 &amp; p=0.988 respectively, chi-square test). Fever-neutropenia and sepsis were the most frequent toxicities with VHR-DI. To identify potential mechanisms of resistance in poor-responders, transcriptome analysis was performed on diagnostic blast cells from VHR patients with good (N=5) or poor (N=5) response to VHR-DI and compared with highly sensitive SR patients (N=5). Upregulation of oxidative phosphorylation and ubiquitin-mediated proteolysis were found in good responders while non-responders showed upregulation of amino acid metabolic and RNA splicing pathways. In conclusion, drug response profiling identifies alternative combination therapies for VHR ALL. Our pilot study provides proof-of-concept of effectiveness and tolerability of inclusion of VEN and BZM in the post-induction treatment of patients with poor-response disease.

Outcomes in Children, Adolescents, and Young Adults With Down Syndrome and ALL: A Report From the Children's Oncology Group.

Patients with Down syndrome (DS) and B-ALL experience increased rates of relapse, toxicity, and death. We report results for patients with DS B-ALL enrolled on Children's Oncology Group trials between 2003 and 2019. We analyzed data for DS (n = 743) and non-DS (n = 20,067) patients age 1-30 years on four B-ALL standard-risk (SR) and high-risk trials. Patients with DS exhibited more frequent minimal residual disease (MRD) ≥0.01% at end induction (30.8% v 21.5%; P < .001). This difference persisted at end consolidation only in National Cancer Institute (NCI) high-risk patients (34.0% v 11.7%; P < .0001). Five-year event-free survival (EFS) and overall survival (OS) were significantly poorer for DS versus non-DS patients overall (EFS, 79.2% ± 1.6% v 87.5% ± 0.3%; P < .0001; OS, 86.8% ± 1.4% v 93.6% ± 0.2%; P < .0001), and within NCI SR and high-risk subgroups. Multivariable Cox regression analysis of the DS cohort for risk factors associated with inferior EFS identified age >10 years, white blood count >50 × 103/μL, and end-induction MRD ≥0.01%, but not cytogenetics or CRLF2 overexpression. Patients with DS demonstrated higher 5-year cumulative incidence of relapse (11.5% ± 1.2% v 9.1% ± 0.2%; P = .0008), death in remission (4.9% ± 0.8% v 1.7% ± 0.1%; P < .0001), and induction death (3.4% v 0.8%; P < .0001). Mucositis, infections, and hyperglycemia were significantly more frequent in all patients with DS, while seizures were more frequent in patients with DS on high-risk trials (4.1% v 1.8%; P = .005). Patients with DS-ALL exhibit an increased rate of relapse and particularly of treatment-related mortality. Novel, less-toxic therapeutic strategies are needed to improve outcomes.

Impact of T-cell Receptor Status on Mutational Landscape and Outcome in T-ALL.

Impact of T-cell Receptor Status on Mutational Landscape and Outcome in T-ALL.

Impact of minimal residual disease on relapse in childhood acute lymphoblastic leukemia: Lessons learnt from a tertiary cancer center in India

Prognostic predictive value of end of induction minimal residual disease (EOI-MRD) is well established in acute lymphoblastic leukemia (ALL). We evaluated the factors likely to affect EOI-MRD positivity (>0.01%) by flow cytometry and relapse in different BFM-95 (Berlin–Frankfurt–Munich) risk groups among children and adolescents. In this retrospective study, data of 223 newly diagnosed patients with ALL was analyzed. Association between demographic and pretreatment characteristics with EOI-MRD was assessed. Risk factors for relapse were analyzed using univariate and multivariate Cox regression. Proportion of the SR (standard risk), MR (moderate risk), and HR (high risk) patients was 18.8%, 60.9%, 20.3%, respectively. Positive EOI-MRD among these risk groups was observed in 11.9%, 18.3%, and 55.5% patients respectively (p value <.01%). MRD positivity was more likely to be associated with older age (>10 years) and BFM-HR patients (p value .0008 and <.0001). Thirty-four (15.2%) patients relapsed in the whole cohort. On univariate analysis, statistically significant factors for RFS (relapse-free survival) included hyperleukocytosis, high-risk cytogenetics, NCI (National Cancer Institute) high risk, poor day-8 prednisolone response, BFM–HR and positive EOI-MRD status. Of all these only EOI-MRD retained its impact by multivariate analysis. Positive EOI-MRD significantly predicted relapse in BFM-MR with 5-year RFS of 88.0% and 68.4% (p value .02). Five-year RFS of EOI-MRD negative and positive groups were 86.4% and 65.5%, respectively (p value .004). EOI-MRD is a powerful tool to predict relapse in children and adolescent with ALL especially in BFM-MR. Application of MRD in HR patients needs to be redefined in conjunction with other variables.

Correlation of Minimal Residual Disease (MRD) at the End of Induction (EOI) and Event Free Survival (EFS) in T-Cell Lymphoblastic Lymphoma (T-LL), a Report from the Children's Oncology Group (COG) Trial AALL1231

Background: Defining prognostic variables that correlate to EFS in T-LL remains a clinical challenge. Traditional variables such as stage or radiologic response to therapy have failed to correlate with EFS in recent trials. AALL1231 was a COG phase 3 clinical trial for newly diagnosed with T Acute Lymphoblastic leukemia (T-ALL) or T-LL patients that randomized children and young adults (age 1-30 years) to a modified augmented BFM (aBFM) backbone with no bortezomib (Arm A) or with bortezomib (Arm B) during induction and delayed intensification (DI) (1.3mg/m2 x 4 doses per block). T-LL patients were stratified as standard (SR), intermediate (IR), or very high risk (VHR), based upon minimal detectable disease (MDD) in the bone marrow at diagnosis, prior steroid pre-treatment and radiographic response at the end of induction. Bone marrow (BM) samples to assess MRD by flow cytometry at the EOI were an optional submission for T-LL participants and those collected were analyzed to assess its correlation to EFS. Results: AALL1231 accrued 209 T-LL patients from 2014 until closure in 2017. At the end of induction, 43.6% were in radiologic remission, 55.4% had a partial response and 1% had stable disease or no response. There were 86 (41%) patients for whom EOI samples for MRD assessment were submitted. Patients with MRD <0.1% (n=75) at EOI had a superior EFS versus those with MRD >0.1% (n= 11), (89.0 + 4.4% verses 63.6 + 17.2%, p= 0.025). Overall survival did not significantly differ between the two groups (88.9 + 4.4% versus 72.7 + 15.5% p= 0.15). As previously reported, Arm B demonstrated a superior outcome for both EFS and OS for the entire cohort. Within risk groups, IR and SR patients had similar EFS outcomes (Arm A 73.9 + 7.5% versus 80.4 + 6.7, Arm B 87.2 + 5.8 % versus 90.5 + 4.8%) with only two patients the VHR, both in Arm B. Cox regression for EFS using Arm A as a reference demonstrated that MRD EOI ≥0.1% was associated with a greater risk of inferior outcome (Hazard Ratio, HR = 3.73 (1.12-12.40, p = 0.032), (Table 1) which was independent of treatment arm assignment. OS failed to reach statistical significance for patients MRD EOI ≥0.1%, HR = 2.714 (0.72-10.44, p=0.14). Furthermore, Cox regression failed to demonstrate a superior EFS with decreasing MDD at diagnosis (>5% versus 1-5%, versus <1%), HR=2.67 (0.336-21.145, p=0.141, HR= 0.830 (0.255-2.699, HR= 0.57 (0.289-1.073, p= 0.080)(Table 2). Conclusions: In this pediatric T-LL trial, BM EOI MRD by flow cytometry was associated with improved EFS in in both arms (+/- bortezomib) in both univariate and multivariate analyses. In contrast, race, age, gender, risk group, MDD, stage and radiologic response to therapy were not prognostic. Consideration to incorporate MRD at EOI into the design of future trials will establish its value in defining risk groups for subsequent therapeutic trials. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal

Abstract 3633: Disparities in relapse among a large multi-ethnic population of children diagnosed with acute lymphoblastic leukemia (ALL): A report from the Reducing Ethnic Disparities in Acute Leukemia (REDIAL) Consortium

Abstract Introduction: While end-induction minimal residual disease (MRD) is the strongest prognostic factor for relapsed ALL, approximately half of all relapses occur in children who are MRD negative. Latino ethnicity is also a risk factor for relapse. To further explore these associations, we conducted an interim analysis of risk factors for relapse in a large multi-ethnic population of children diagnosed with ALL. Methods: The REDIAL Consortium includes patients diagnosed with ALL at six major pediatric cancer centers in the Southwestern U.S. The study period was 2004 to 2018, and we included individuals who were 1-23 years of age when diagnosed with ALL. Time to relapse was defined as time from ALL diagnosis to the initial relapse event, with individuals censored at date of death, last follow-up, or bone marrow transplant. Demographic and clinical factors evaluated included race/ethnicity (Latino, non-Latino Black, non-Latino White, non-Latino other), sex, age at diagnosis (1-5, 6-10, 11-15, &amp;gt;15 years), ALL immunophenotype (B-cell, T-cell), National Cancer Institute (NCI) risk group, central nervous system involvement, enrollment on a Children’s Oncology Group clinical trial, end-induction disease failure, and end-induction bone marrow flow cytometric MRD. Cox proportional hazards models were used to calculate adjusted hazards ratios (HRs) and 95% confidence intervals (CIs). Analyses were further stratified based on end-induction MRD status (positive ≥0.01%, negative &amp;lt;0.01%). Results: Overall, there were 1,710 ALL patients with a median age at diagnosis of 5 years (interquartile range: 3-11 years). The majority of patients were Latino (60.1%) and male (56.9%). Of the 379 MRD-positive patients, 74 (19.5%) relapsed, compared to 138 of 1,233 (11.2%) MRD-negative patients (p&amp;lt;0.0001). In adjusted models, factors associated with a greater risk of relapse included MRD positivity (HR=1.72, 95% CI: 1.26-2.36), older age at diagnosis (&amp;gt;15 vs. 1-5 years, HR=1.98, 95% CI: 1.19-3.29), and NCI high-risk group (HR=1.74, 95% CI: 1.20-2.52), while patients enrolled on a clinical trial were less likely to relapse (HR=0.76, 95% CI: 0.57-0.99). Among MRD-positive patients, Latinos were less likely to relapse (HR=0.60, 95% CI: 0.33-0.99) compared to non-Latino Whites, whereas Latinos who were MRD negative were more likely to relapse (HR=1.68, 95% CI: 1.09-2.59). Conclusion: In a large contemporary multi-ethnic cohort of &amp;gt;1,700 children with ALL, we observed significant disparities in relapse by MRD status, age at diagnosis, NCI risk group, clinical trial enrollment, and race/ethnicity. Notably, nearly 65% of relapse events occurred in MRD-negative patients. Further analyses are ongoing in REDIAL to evaluate the impact of other factors including cytogenetics and novel biomarkers of relapse. Citation Format: Pagna Sok, Austin L. Brown, Olga A. Taylor, M. Brooke Bernhardt, Juan C. Bernini, Rodrigo A. Erana, Timothy Griffin, Kenneth Heym, Van T. Huynh, Laura Klesse, Kathleen Ludwig, Sandi L. Pruitt, Karen R. Rabin, Michael E. Scheurer, Philip J. Lupo. Disparities in relapse among a large multi-ethnic population of children diagnosed with acute lymphoblastic leukemia (ALL): A report from the Reducing Ethnic Disparities in Acute Leukemia (REDIAL) Consortium [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 3633.

Abstract 3671: Association between residence in a Hispanic enclave and end-induction minimal residual disease among children with acute lymphoblastic leukemia in Texas

Abstract Background: Acute lymphoblastic leukemia (ALL) remains a leading cause of death in children and adolescents. We have demonstrated that Hispanic enclaves - neighborhoods with high proportions of Spanish-speaking residents, recent immigrants, and ethnic-specific businesses - are associated with inferior overall survival in children with ALL. However, associations of enclaves with other outcomes remain poorly understood. In the present study, our objective was to determine whether residence in an enclave was associated with end-induction minimal residual disease (MRD), a strong predictor of ALL mortality. Methods: This was a retrospective study of N=142 children aged 0-18 years, treated at Texas Children’s Hospital (Houston, TX) from 2007-2018. MRD was defined as ≥0.01% leukemic blasts in bone marrow at day 29 of therapy. We used the 2010 census tract geography and a modified version of the Hispanic enclave index to identify enclaves. We assigned an enclave index score to each census tract, computed quartiles based on the statewide distribution, and mapped children to quartiles based on address at diagnosis (Q1: least ethnically distinct neighborhoods; Q4: most ethnically distinct neighborhoods). Data on sex, self-reported race/ethnicity, National Cancer Institute (NCI) risk group, primary language, and ALL cytogenetics were abstracted from electronic health records. Cytogenetics were considered either unfavorable (BCR-ABL1 fusion, KMT2A rearrangement, hypodiploidy, or intrachromosomal amplification of chromosome 21) or favorable (ETV6-RUNX1 fusion, double trisomies of chromosomes 4 and 10)/neutral (neither favorable nor unfavorable). We estimated the odds ratio (OR) and 95% confidence interval (95% CI) of MRD according to enclave index score using logistic regression. Results: Neither individual ethnicity nor primary language varied by MRD status, but a greater proportion of MRD-positive children lived in the most ethnically distinct neighborhoods (53% vs. 36%, p=0.04). Enclave index score was associated with MRD after adjusting for sex, NCI risk group, and cytogenetics (OR 1.57 per quartile increment, 95% CI 1.09-2.33). We also observed non-significantly increased odds of MRD among children in Q3 (OR 1.53, 95% CI 0.41-6.55) and Q4 (OR 2.84, 0.83-11.58) relative to Q1. Conclusions: We found that residence in a Hispanic enclave was associated with MRD in children with ALL. Neither ethnicity nor primary language differed by MRD status, suggesting that these did not confound the association with enclave index score. Neighborhood factors may influence early treatment outcomes in ALL, and children living in Hispanic enclaves may constitute a high-risk population. Citation Format: Joshua P. Muniz, John P. Woodhouse, Amy E. Hughes, Sandi L. Pruitt, Karen R. Rabin, Michael E. Scheurer, Philip J. Lupo, Jeremy M. Schraw. Association between residence in a Hispanic enclave and end-induction minimal residual disease among children with acute lymphoblastic leukemia in Texas [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 3671.

Effects of age, obesity, and body surface area on asparaginase-associated toxicities during acute lymphoblastic leukemia induction therapy: A report from the Children’s Oncology Group.

7000 Background: Asparaginase is integral to pediatric-inspired regimens (PIR) to treat acute lymphoblastic leukemia (ALL) in adolescents and young adults (AYA). However, asparaginase-associated toxicities (AAT) often preclude delivery of planned therapy. Older age, obesity and/or large body surface area (BSA) have been associated with higher risk of AAT in PIR, but data are conflicting, and the impact of dose modification based on these factors is unknown. Methods: We examined induction toxicity data from patients ages 1-30 years enrolled in the frontline Children’s Oncology Group (COG) trials for high-risk B-ALL (AALL0232, 2004-2011) and T-ALL (AALL0434, 2007-2014). During Induction, patients received pegaspargase (2,500 IU/m2 without prescribed dose-capping) plus daunorubicin, vincristine, and prednisone or dexamethasone. AAT were defined as CTCAE v4 hyperbilirubinemia (Grade ≥3), elevated alanine aminotransferase (ALT) (Grade ≥4), thrombosis (any), or pancreatitis (any, included consolidation phase). Obesity was classified using population norms as body mass index (BMI) ≥30 (or ≥95th percentile for age/sex). BSA was analyzed continuously and dichotomized at 1.5 m2 (equivalent to pegaspargase 3,750 IU, the threshold for permissible dose-capping in PIR). The association of AAT with end-Induction minimal residual disease (MRD) ≥0.01% was assessed. Results: Among 4,925 patients, 25% were ≥15 years, 39% had BSA &gt;1.5m2, and 18% had obesity. Multivariable logistic analyses inclusive of BMI and BSA together found increased risk for any AAT in age groups ≥10 years (10-15y, odds ratio (OR) 2.0, 15-20y OR 2.2, ≥21 OR 3.3, p=0.002). Only patients with both obesity and high BSA (&gt;1.5m2) were at additional risk (OR 3.3, p&lt;0.0001). Similarly, risks for hyperbilirubinemia, ALT elevations, and thrombosis were increased in patients with both high BSA and obesity (OR 3.5, 95% confidence interval [CI] 2.2-5.7), OR 3.3, 95%CI 1.7-6.6, and OR 3.1 95%CI 1.5-6.5, respectively), but not in those with high BSA without obesity. The risk of hyperbilirubinemia was greater with increasing obesity (p&lt;0.0001) and was also higher in all age groups ≥10 years (OR 6.3-7.9, p&lt;0.0001). Age was not associated with thrombosis or ALT elevation; risk for pancreatitis was associated with Hispanic ethnicity, but not with age, BMI, or BSA. AAT were not associated with pooled trial MRD ≥0.01%. Conclusions: We report here the largest dataset of AAT in children and AYAs receiving ALL Induction therapy without routinely prescribed dose-capping of pegaspargase. Risk for AAT was increased in patients &gt;10 years and in those with obesity, but not high BSA alone. Dose capping may not be necessary for children and AYAs with high BSA without obesity. Prospective studies of AAT pharmacogenomics and modifiable risk factors will support safer dosing in PIR. Clinical trial information: NCT00075725, NCT00408005.

Minimal residual disease comparison between Ig/TCR PCR versus NGS assays in children with Philadelphia chromosome-positive acute lymphoblastic leukemia: A report from the COG AALL1631 study.

10023 Background: Minimal residual disease (MRD) assessment by immunoglobulin/T-cell receptor (Ig/TCR) polymerase chain reaction (PCR) is currently being used in the international pediatric Philadelphia chromosome-positive acute lymphoblastic leukemia (Ph+ALL) trial EsPhALL2017/AALL1631 for risk stratification. MRD concordance has previously been demonstrated between Ig/TCR PCR and flow cytometry in Ph+ALL. We sought to assess concordance of MRD assessment between conventional Ig/TCR PCR and next-generation sequencing (NGS) assays. Methods: MRD was assessed in all pts on AALL1631 by Ig/TCR PCR at end-induction IB; those with MRD &lt;5x10-4 were classified as standard-risk (SR) and randomized to treatment with imatinib and one of two chemotherapy regimens without hematopoietic stem cell transplant (HSCT), whereas pts with end-induction 1B MRD ≥ 5x10-4 were considered high-risk (HR) and assigned to HSCT after consolidation chemotherapy. Residual diagnostic and end-induction IB samples from consenting pts were assessed for NGS MRD by the clonoSEQ assay (Adaptive Biotechnologies) in blinded fashion and subsequently compared to Ig/TCR MRD to determine concordance as related to MRD-based HSCT recommendations ( ie, MRD ≥ 5x10-4 consistent with HR group assignment). MRD values were calculated using the kappa statistic for agreement above chance. Results: Sixty-seven pts had matched samples available for MRD assessment at end-induction 1B by both Ig/TCR PCR and NGS (Table). NGS MRD was evaluable for all 67 pts and stratified as 62 SR (&lt;5x10-4) and 5 HR (≥5x10-4). In contrast, Ig/TCR PCR results were inevaluable for 3 pts (unsatisfactory sample quality) and indeterminate (positive, but not quantifiable) in 4 pts. Of the remaining 60 pts, 55 met SR and 5 HR criteria using Ig/TCR PCR. There was only 1 discordant case between the two methods for MRD-based HSCT recommendation among these 60 pts with a kappa statistic for agreement above chance of 0.88. Conclusions: NGS and Ig/TCR PCR assays were highly concordant in MRD assessment for risk stratification at a threshold of 5x10-4 in pediatric pts with Ph+ALL enrolled on AALL1631. Of note, the NGS assay yielded MRD results amenable for risk stratification in 100% pts compared to 89.6% for the Ig/TCR PCR methodology. These data support the use of NGS MRD testing for risk stratification in pediatric Ph+ALL.[Table: see text]

Ocular abnormalities at diagnosis and after the completion of treatment in children and adolescents with newly diagnosed acute lymphoblastic leukemia.

Ocular abnormalities (OA) in pediatric patients with acute lymphoblastic leukemia (ALL) are common findings both at diagnosis and later in follow-up. The frequency, predictors, and prognostic impact of OA in the context of recent ALL protocols are not well characterized. Single-center retrospective analysis of the medical records of 224 patients with ALL enrolled on Dana-Farber Cancer Institute (DFCI) ALL Consortium Protocol 05-001. Overall, 217 (98%) patients had at least one ophthalmic exam. Retinal hemorrhages were the most frequent abnormalities at diagnosis (11%) and cataracts at later time points (13%). OA at diagnosis were associated with age ≥10years and with the severity of anemia and thrombocytopenia; they were also univariately associated with lower 5-year event-free survival (EFS) (high risk [HR]=3.09 [95% CI: 1.38-6.94]; p=.006), but not in a disease-free survival (DFS) model adjusted for end-induction minimal residual disease (p=.82). The cumulative incidence of cataract was 13.1% ± 2.8% at 43months from diagnosis; its development was associated with high presenting white blood cell count (≥50,000/μl) (p=.010), male sex (p=.036), higher risk group (p=.025), and cranial radiation (p=.004). Cataract was associated with decreased visual acuity. OA at diagnosis, present in 12% of patients, were associated with older age, anemia, and thrombocytopenia and did not carry a significant prognostic impact. Cataracts were detected in over 10% of patients and were associated with decreased visual acuity, thus supporting routine screening after completion of therapy, especially for those treated with high-risk protocols.

Racial, Ethnic, and Socioeconomic Factors Result in Disparities in Outcome Among Children with Acute Lymphoblastic Leukemia Not Fully Attenuated By Disease Prognosticators: A Children's Oncology Group (COG) Study

▪Introduction: Health disparities are major issue for racial, ethnic, and socioeconomically disadvantaged groups. Though outcomes in childhood acute lymphoblastic leukemia (ALL) have steadily improved, identifying persistent disparities is critical. Prior studies evaluating ALL outcomes by race or ethnicity have noted narrowing disparities or that residual disparities are secondary to differences in leukemia biology or socioeconomic status (SES). We aimed to identify persistent inequities by race/ethnicity and SES in childhood ALL in the largest cohort ever assembled for this purpose.Methods: We identified a cohort of newly-diagnosed patients with ALL, age 0-30.99 years who were enrolled on COG trials between 2004-2019. Race/ethnicity was categorized as non-Hispanic white vs. Hispanic vs. non-Hispanic Black vs. non-Hispanic Asian vs. Non-Hispanic other. SES was proxied by insurance status: United States (US) Medicaid (public health insurance for low-income individuals) vs. US other (predominantly private insurance) vs. non-US patients (mainly jurisdictions with universal health insurance). Event-free and overall survival (EFS, OS) were compared across race/ethnicity and SES. The relative contribution of disease prognosticators (age, sex, white blood cell count, lineage, central nervous system status, cytogenetics, end Induction minimal residual disease) was examined with Cox proportional hazard multivariable models of different combinations of the three constructs of interest (race/ethnicity, SES, disease prognosticators) and examining hazard ratio (HR) attenuation between models.Results: The study cohort included 24,979 children, adolescents, and young adults with ALL. Non-Hispanic White patients were 13,872 (65.6%) of the cohort, followed by 4,354 (20.6%) Hispanic patients and 1,517 (7.2%) non-Hispanic Black patients. Those insured with US Medicaid were 6,944 (27.8%). Five-year EFS (Table 1) was 87.4%±0.3% among non-Hispanic White patients vs. 82.8%±0.6% [HR 1.37, 95 th confidence interval (95CI) 1.26-1.49; p<0.0001] among Hispanic patients and 81.9%±1.2% (HR 1.45, 95CI 1.28-1.56; p<0.0001) among non-Hispanic Black patients. Outcomes for non-Hispanic Asian patients were similar to those of non-Hispanic White patients. US patients on Medicaid had inferior 5-year EFS as compared to other US patients (83.2%±0.5% vs. 86.3%±0.3%, HR 1.21, 95CI 1.12-1.30; p<0.0001) while non-US patients had the best outcomes (5-year EFS 89.0%±0.7%, HR 0.78, 95CI 0.71-0.88; p<0.0001). There was substantial imbalance in traditional disease prognosticators (e.g. T-cell lineage) across both race/ethnicity and SES, and of race/ethnicity by SES. For example, T-lineage ALL accounted for 17.6%, 9.4%, and 6.6% of Non-Hispanic Black, Non-Hispanic White, and Hispanic patients respectively (p<0.0001).Table 2 shows the multivariable models and illustrates different patterns of HR adjustment among specific racial/ethnic and SES groups. Inferior EFS among Hispanic patients was substantially attenuated by the addition of disease prognosticators (HR decreased from 1.37 to 1.17) and further (but not fully) attenuated by the subsequent addition of SES (HR 1.11). In contrast, the increased risk among non-Hispanic Black children was minimally attenuated by both the addition of disease prognosticators and subsequent addition of SES (HR 1.45 to 1.38 to 1.32). Similarly, while the superior EFS of non-US insured patients was substantially attenuated by the addition of race/ethnicity and disease prognosticators (HR 0.79 to 0.94), increased risk among US Medicaid patients was minimally attenuated by the addition of race/ethnicity or disease prognosticators (HR 1.21 to 1.16). OS disparities followed similar patterns but were consistently worse than in EFS, particularly among patients grouped as non-Hispanic other.Conclusions: Substantial disparities in survival outcomes persist by race/ethnicity and SES in the modern era. Our findings suggest that reasons for these disparities vary between specific disadvantaged groups. Additional work is required to identify specific drivers of survival disparities that may be mitigated by targeted interventions. [Display omitted] DisclosuresGupta: Jazz Pharmaceuticals: Consultancy, Membership on an entity's Board of Directors or advisory committees. Teachey: NeoImmune Tech: Research Funding; Sobi: Consultancy; BEAM Therapeutics: Consultancy, Research Funding; Janssen: Consultancy. Zweidler-McKay: ImmunoGen: Current Employment. Loh: MediSix therapeutics: Membership on an entity's Board of Directors or advisory committees.

Analysis of Single Cell Transcriptomics in Paired Pediatric T-ALL Samples Collected at Diagnosis and Following End of Induction Therapy Reveals an MRD-Associated Stem Cell Signature

IntroductionDespite recent improvement in outcomes for de novo disease, pediatric T-cell acute lymphoblastic leukemia (T-ALL) remains challenging to treat at relapse. Investigation into genomic markers of treatment response and therapy resistance offers an opportunity to further enhance outcomes for these patients. We previously identified a T-ALL blast-associated gene signature at diagnosis (Dx) and characterized the immune microenvironment in Dx T-ALL marrow samples using single cell transcriptome analysis (Bhasin et al. Blood 2020(ASH)). This approach allowed us to generate a granular expression map of both the T-ALL landscape and the Dx bone marrow (BM) immune microenvironment. Here we expand this work by evaluating samples collected from the same patients Dx and End of Induction (EOI) BM samples from pediatric T-ALL patients. The use of paired samples provides insight into treatment-induced changes in the microenvironment. Further, the inclusion of both minimal residual disease (MRD) positive and MRD negative samples allowed us to compare differences between these groups.MethodsUsing the 10X genomics platform, we profiled the single cell transcriptome of ~18,000 BM and immune microenvironment cells from viably frozen samples collected from T-ALL patients at Dx or EOI. Five paired Dx and EOI samples and one EOI sample from a patient with relapsed T-ALL were evaluated, for a total of 11 samples. Three paired samples were MRD positive at EOI and two were MRD negative; the relapsed sample was MRD negative. Cell clustering was performed using the Seurat package and differential expression analysis was performed using R/Bioconductor packages (Hao et al. Cell 2021). Cell communication analysis was conducted using the CellChat R tool (v 1.0.0) to infer cell-cell communication within the EOI MRD positive and MRD negative subsets and compare their communication networks (Jin et al. Nature Comm 2021).ResultsUsing our previously described blast-associated gene signature (Bhasin et al. ASH 2020) we were able to identify residual blast populations at EOI in MRD-positive samples. Comparative analysis of gene profiles at Dx and EOI showed significant changes in the microenvironment cell populations with highest increase in erythroid cell populations after induction therapy. The gene expression profiles were significantly different for immune cells at Dx and EOI and the relapsed sample had greater similarity to the Dx samples indicating a persistent immunosuppressive environment. Clustering analysis of the EOI samples (3 MRD positive and 2 MRD negative) demonstrated the presence of patient specific blast cells in MRD positive samples that retained patient-specific transcriptomeheterogeneity at EOI (Fig.1A). Analysis of communication networks between different cell types based on receptor and ligand expression levels between different cell types identified a CD34 + cluster of stem cells that had different interactions with other immune populations in the MRD positive and negative subsets. Differential expression analysis between the MRD positive and MRD negative cells in this CD34 + stem cell cluster identified higher expression of myeloid associated genes such as CEBPB, CEBPD, AZU1 in the MRD negative group relative to the MRD positive cells, which showed higher expression of B-cell related genes such as IGHM, VPREB1, CD79A/ B along with upregulation of P13K signaling in B-lymphocytes, B-cell receptor signaling and autophagy pathways. Analysis of upstream regulators based on the differential gene signature between the MRD positive and MRD negative group demonstrated upregulation of MYC and TCF3 activity and inhibition of TGFB1, CSF3 and CEBPA in MRD positive compared to MRD negative samples (Fig.1B).Conclusions: Leukemic blasts exhibit patient-specific gene expression signatures that are present at EOI in MRD positive samples. Exploration of the impact of minimal residual disease at EOI revealed differential gene expression patterns in stem cells from MRD positive samples, characterized by activation of B cell related signaling pathways and regulators such as MYC and TCF3. In contrast, a more myeloid-like expression signature was observed in stem cells from MRD negative samples. These findings open the avenues for exploration of therapeutic targets of T-ALL progression. [Display omitted] DisclosuresDeRyckere: Meryx: Other: Equity ownership. Graham: Meryx: Membership on an entity's Board of Directors or advisory committees, Other: Equity ownership.

Blinatumomab Associated Seizure Risk in Patients with Down Syndrome and B-Lymphoblastic Leukemia: An Interim Report from Children's Oncology Group (COG) Study AALL1731

INTRODUCTIONChildren with Down Syndrome (DS) and B-lymphoblastic leukemia (B-ALL) are at an increased risk of both relapse and treatment-related mortality, compared to those without DS. On COG study AALL1731 for de novo B-ALL, patients with DS and higher risk features (DS-High) are non-randomly treated with a regimen replacing intensive elements of conventional chemotherapy with three 28-day cycles of blinatumomab, with the combined goals of reducing toxicity and enhancing anti-leukemic efficacy. The DS-High group includes all NCI high risk (HR) patients; NCI standard risk (SR) patients with end-induction minimal residual disease positivity (>0.01%), unfavorable cytogenetics, CNS3 status, steroid pre-treatment, neutral cytogenetics with CNS2 status, or testicular disease. Neurotoxicity is a known risk of blinatumomab, with an incidence of 4% in block 1 and 1% in block 2 among pediatric patients with relapsed ALL (Brown et al, JAMA 2021). However, the specific risk in patients with DS has not been described to date. Here, we provide an early report of increased seizure incidence associated with blinatumomab in older DS-High patients enrolled on AALL1731 to date.METHODSWe reviewed seizure incidence among patients with DS enrolled on AALL1731 from June 2019 to June 2021 who had proceeded to receive blinatumomab. Blinatumomab was administered at a dose of 15 mcg/m 2/day, using dexamethasone pre-medication in cycle 1. Infusions were interrupted for seizures, with resumption at 5 mcg/m 2/d permitted following full resolution for grade 1-3 seizures.RESULTSAmong DS NCI HR patients, 8 of 47 (17%) had a seizure during blinatumomab infusion (Table 1). All 8 seizures occurred in patients over 10 years old. Six of the 8 seizures occurred in the first cycle of blinatumomab, most in the first 3 days of the infusion. Four had concomitant fever or cytokine release syndrome. Seizures were grade 2 (n=2) or grade 3 (n=6), and all resolved with full neurologic recovery. Of the 8 patients, 5 elected to resume blinatumomab; no further seizures occurred in these patients. There was no indication of increased seizure risk among NCI SR DS-High patients (1 seizure among 11 patients), or among DS or non-DS patients receiving blinatumomab on other study strata (0 of 7 DS SR-Avg; 1 of 146 non-DS SR-Avg; and 2 of 120 non-DS SR-High).CONCLUSIONSThe incidence of seizures associated with blinatumomab in DS-ALL patients older than 10 years appears higher than previously reported in children without DS. The majority of seizures occurred within the first 3 days, all fully resolved with no sequelae, and no patient who resumed blinatumomab infusion at a lower rate experienced further seizures. Seizure prophylaxis may be advisable in DS patients while receiving blinatumomab, particularly those >10 years of age. Further follow-up and a larger sample size are needed to confirm incidence and identify risk factors predisposing DS patients to neurologic toxicity. [Display omitted] DisclosuresLi: Novartis Canada: Membership on an entity's Board of Directors or advisory committees. Raetz: Pfizer: Research Funding; Celgene: Other: DSMB member. Loh: MediSix therapeutics: Membership on an entity's Board of Directors or advisory committees. Gupta: Jazz Pharmaceuticals: Consultancy, Membership on an entity's Board of Directors or advisory committees. Rau: Jazz Pharmaceuticals: Consultancy, Membership on an entity's Board of Directors or advisory committees, Other: Advisory Board; AbbVie Pharmaceuticals: Other: Spouse is employee and stock holder; Servier Pharmaceuticals: Consultancy. OffLabel Disclosure:This trial includes the use of blinatumomab in combination with chemotherapy for treatment of de novo B-lymphoblastic leukemia.

Abstract 2349: Comparison of the blood, bone marrow, and cerebrospinal fluid metabolomes in children with acute leukemia

Abstract Background: Children with acute leukemia (AL) are at risk for relapse, treatment toxicity, second neoplasms, and death. Understanding host and tumor response to therapy is necessary to achieve high survival rates and minimize adverse effects. Metabolomics, the characterization of small molecules in a biological system, may be useful for monitoring treatment response. In support of this, we previously identified metabolites in bone marrow (BM) plasma associated with end-induction minimal residual disease (MRD) and in cerebrospinal fluid (CSF) associated with fatigue. Because procedures for obtaining BM and CSF are invasive and subject patients to risk, we sought to determine whether the BM and CSF metabolomes of children undergoing therapy for AL were correlated with the blood metabolome. Methods: Blood plasma, BM, and CSF were collected at the end-induction from 11 patients with newly diagnosed AL treated at Texas Children's Hospital (Houston, TX). Global metabolomic profiling was performed by Metabolon (Durham, NC), using liquid chromatography-tandem mass spectrometry according to published methods. We assessed the number and type (e.g. lipid, amino acid) of metabolites in each sample. For each metabolite identified in ≥2 samples from ≥2 patients, we computed Spearman rank correlation coefficients to estimate the pairwise correlations between blood, BM, and CSF. Results: Patients were predominantly male (N=7) and Latino (N=9). Ten were diagnosed with B-cell acute lymphoblastic leukemia and one with mixed phenotype acute leukemia. Among 670 metabolites detected in ≥2 blood and BM plasma samples, 370 (58%) demonstrated moderate to strong correlations (Spearman's rho ≥0.5) and 317 (47%) were significant at p&amp;lt;0.05. Among 340 metabolites detected in ≥2 blood and CSF samples, 135 (40%) demonstrated moderate to strong correlations and 96 (28%) were significant at p&amp;lt;0.05. Among 318 metabolites detected in ≥2 BM plasma and CSF samples, 88 (28%) demonstrated moderate to strong correlations and 66 (21%) were significant at p&amp;lt;0.05. We observed generally moderate to strong correlations between BM and blood for 19 metabolites we previously reported were associated with MRD (median rho 0.51, interquartile range 0.24-0.69). We observed a strong positive correlation between CSF and blood for one of three metabolites we previously reported were associated with fatigue (dimethylglycine, rho=0.7, p=0.02), and a moderate inverse correlation for a second (gamma-glutamylglutamine, rho=-0.51, p=012). Conclusions: There is substantial correlation between the blood and BM plasma metabolomes at end-induction among children with ALL, including metabolites putatively associated with early treatment response. Correlations between CSF and blood or BM plasma are weaker, implying that the CSF metabolome is distinct. The blood plasma metabolome may approximate the BM but not CSF metabolome in children with ALL. Citation Format: Jeremy M. Schraw, J.P. Woodhouse, Melanie B. Bernhardt, Olga A. Taylor, Terzah M. Horton, Michael E. Scheurer, Mehmet F. Okcu, Karen R. Rabin, Philip J. Lupo, Austin L. Brown. Comparison of the blood, bone marrow, and cerebrospinal fluid metabolomes in children with acute leukemia [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 2349.

Genomic and Outcome Analyses of Philadelphia Chromosome-like (Ph-like) NCI Standard Risk B-Acute Lymphoblastic Leukemia (SR B-ALL) Patients Treated on Children's Oncology Group (COG) AALL0331

Introduction: Ph-like ALL is defined by a gene expression profile similar to Philadelphia chromosome-positive ALL but lacking the BCR - ABL1 fusion. Ph-like ALL is comprised of diverse genomic alterations activating kinase and cytokine receptor signaling, including chimeric fusions involving ABL1 and related genes (ABL2, CSF1R, PDGFRB ; ABL-class fusions), genomic rearrangements leading to CRLF2 over-expression, JAK2 gene fusions, and truncating rearrangements of EPOR . We have published several large studies describing the clinical features, treatment outcome, and genomic alterations in Ph-like ALL that focused on patients with NCI high risk (HR) ALL (age ≥10 years or WBC ≥50,000) or those with SR ALL (age 1-9.99 years and WBC <50,000) and elevated minimal residual disease (MRD) at the end of induction (EOI; Reshmi, BLOOD 2017 and Roberts, NEJM 2014). The purpose of the current study was to assess the frequency of genomic lesions in Ph-like SR ALL patients lacking HR features.Methods: We retrospectively analyzed 1023 unselected newly-diagnosed SR ALL patients enrolled on COG AALL0331 between 2006 and 2008, and compared characteristics of SR Ph-like cases to the HR patients described in our prior studies. All cases were screened using an 8-gene Taqman low-density array (LDA) PCR assay to identify the Ph-like GEP (Harvey, ASH 2013). Ph-like cases with elevated CRLF2 expression were tested for CRLF2 rearrangement (CRLF2 -R; P2RY8 - CRLF2 by Taqman PCR on the LDA card and IGH - CRLF2 by FISH). JAK mutations in CRLF2 -R cases were tested by Sanger sequencing. Ph-like cases without CRLF2 -R were tested for previously identified kinase fusions involving ABL1, ABL2, CSF1R, JAK2, NTRK3, and PDGFRB by RT-PCR. A subset of Ph-like cases without detected genomic alterations underwent RNA-sequencing using standard Illumina library preparation.Results: A total of 205 patients had LDA value ≥0.5, with 145 (14.2%) remaining after exclusion of BCR-ABL1 and ETV6-RUNX1 . Of these 145, 85 were CRLF2high (57.9%) as defined by CRLF2 expression levels on the LDA card. Among these 85, 42 (50%) had CRLF2 -R, compared to 80% in HR patients. P2RY8 - CRLF2 was seen in 36 cases, IGH - CRLF2 in 4 and 2 had CRLF2 -R with an unknown partner; 14 (32.6%) of these had a concomitant JAK mutation. Of the 60 Ph-like CRLF2low cases, five had a targetable kinase fusion identified by RT-PCR or RNA sequencing: two ABL - class fusions (ZMIZ1 - ABL1, EBF1-PDGFRB); one PAX5-JAK2 fusion; one ETV6-JAK2 fusion; and one ETV6 - NTRK3 fusion. The 145 Ph-like cases had inferior 7-year event-free survival (EFS; 83.1±3.5% vs. 90.6 ±0.1%; P=0.005) and non-significantly inferior 7-year overall survival (OS; 93.5±2.3% vs. 95.5 ±0.8%; P=0.184) compared to 869 non Ph-like cases. Ph-like ALL cases were more likely to have end induction MRD ≥ 0.01% (24.3% vs. 17.9%; P=0.049), and positive MRD was predictive of outcome for Ph-like (7-year EFS: 73.2±8.3%, vs. 86.6 ±3.7%; P=0.079; 7-year OS: 88.2 ±5.9% vs. 95.9±2.1%; P=0.162) and non Ph-like patients (7-year EFS: 79.7 ±3.7% vs. 93.3 ±1.0%; P <0.0001; 7-year OS: 90.2% SE 2.7 vs. 97.4±0.9%; P <0.0001). In multivariate analyses, Ph-like and MRD status EOI were significant predictors of EFS (p=0.016; <0.0001 respectively), while only MRD had a significant impact on OS.Conclusion: Ph-like ALL is less common in NCI SR than HR patients with significant differences in the frequency of various genomic alterations in Ph-like ALL cases based on NCI risk group. Our previous studies have shown that Ph-like ALL HR patients have a substantially increased risk of treatment failure and death as compared to non Ph-like HR patients (Mullighan, NEJM 2009 and Roberts, NEJM 2014). This cohort of SR Ph-like patients had a significantly inferior EFS and a trend to inferior OS, but the magnitude of effect was much smaller than that seen in HR Ph-like patients, with overall outcomes much better in SR as compared to HR Ph-like patients. These findings underscore the significant differences in Ph-like ALL frequency, genetic subtypes, and outcome based on age and initial WBC. [Display omitted] DisclosuresBorowitz:Beckman Coulter: Honoraria; Becton-Dickinson Biosciences: Research Funding; HTG Molecular: Honoraria. Zweidler-McKay:ImmunoGen: Employment. Mattano:HARP Pharma Consulting, LLC: Employment. Mullighan:Loxo Oncology: Research Funding; Amgen: Consultancy. Hunger:Amgen: Consultancy, Equity Ownership; Novartis: Consultancy; Erytech Pharmaceuticals: Consultancy; Jazz Pharmaceuticals: Honoraria.

Metabolomic Profiling Improves Prediction of Early Disease Response and Relapse in Pediatric Acute Lymphoblastic Leukemia

Introduction: Although outcomes for pediatric acute lymphoblastic leukemia (ALL) have improved dramatically, five-year survival remains dismal for those cases which relapse. Initial response to therapy, measured by minimal residual disease (MRD) at end-induction, is the most important predictor of outcome. However, approximately half of relapses occur in MRD-negative patients. We hypothesized that global metabolomic profiling of diagnostic bone marrow plasma could improve prediction of MRD positivity and relapse when combined with clinical features, while also providing insights into the biologic features of these chemotherapy-resistant groups.Methods: The study population (n=100) included children age 0-18 years with newly diagnosed ALL treated at Texas Children's Cancer Center. The cohort included 39 cases with MRD ≥ 0.01% at end-induction. Ninety-four children were followed for at least 3 years from diagnosis, of whom 23 relapsed. Diagnostic bone marrow plasma specimens were obtained with informed consent and subjected to metabolomic profiling via UPLC-MS/MS by Metabolon. Following compound identification and correction for batch effects, each compound was scaled such that the median concentration was one and missing values were imputed with the minimum detected concentration.Analysis was limited to 712 named compounds that were detected in at least two samples. We used Welch's two-sample t-test to compute p-values comparing mean concentrations in MRD positive vs. negative and relapsed vs. non-relapsed plasma and chi-squared test to assess differences in the frequency of detection. We then generated volcano plots to visualize the significance of each metabolite.Compounds demonstrating significance at the p < 0.001 level were then evaluated in logistic regression models for MRD and relapse, and the best predictors were obtained by backward elimination. Finally, we assessed the performance of models to predict MRD and relapse using (1) clinical features alone (immunophenotype, cytogenetics and National Cancer Institute risk category, plus end-induction MRD when predicting relapse); (2) metabolomic data alone; and (3) clinical features + metabolomic data. DeLong's test for correlated receiver operating characteristics curves was used to compare model performance.Results: In the assessment of individual metabolites, a higher concentration of pyruvate was associated with an increased odds ratio (OR) of MRD positivity (p = 0.005). Conversely, a significantly reduced OR for MRD was identified for arachidonylglycerol (p = 0.01). Reduced ORs for relapse were identified for increasing concentrations of two compounds involved in amino acid metabolism: picolinic acid and N-carboxyethylphenylalanine (p = 0.04 and 0.02, respectively), as well as for the vitamin E derivative gamma-CEHC (p = 0.03). Several glycerophospholipids demonstrated inverse associations with MRD positivity and relapse.Metabolomic profiling significantly improved prediction of both MRD and relapse as compared to clinical features alone (p = 0.001 and p = 0.03 respectively; Figure 1). Incorporating metabolomic data gave an absolute improvement in sensitivity of 8% for MRD and 19% for relapse.Conclusions: We demonstrate that global metabolomic profiling improves sensitivity and specificity when predicting risk of MRD positivity and relapse in ALL. Differences in pyruvate and lipid concentrations suggest metabolic changes associated with chemoresistance or disease recurrence which may be therapeutically targetable. If validated, these metabolomic biomarkers may improve clinical risk prediction and further our understanding of disease biology. [Display omitted] DisclosuresNo relevant conflicts of interest to declare.

Adverse prognostic impact of complex karyotype (≥3 cytogenetic alterations) in adult T-cell acute lymphoblastic leukemia (T-ALL)

The potential prognostic value of conventional karyotyping in adult T-cell acute lymphoblastic leukemia (T-ALL) remains an open question. We hypothesized that a modified cytogenetic classification, based on the number and type of cytogenetic abnormalities, would allow the identification of high-risk adult T-ALL patients. Complex karyotype defined by the presence of ≥3 cytogenetic alterations identified T-ALL patients with poor prognosis in this study. Karyotypes with ≥3 abnormalities accounted for 16 % (22/139) of all evaluable karyotypes, corresponding to the largest poor prognosis cytogenetic subgroup of T-ALL identified so far. Patients carrying karyotypes with ≥3 cytogenetic alterations showed a significantly inferior response to therapy, and a poor outcome in terms of event-free survival (EFS), overall survival (OS) and cumulative incidence of relapse (CIR), independently of other baseline characteristics and the end-induction minimal residual disease (MRD) level. Additional molecular analyses of patients carrying ≥3 cytogenetic alterations showed a unique molecular profile that could contribute to understand the underlying molecular mechanisms of resistance and to evaluate novel targeted therapies (e.g. IL7R directed) with potential impact on outcome of adult T-ALL patients.