Machine learning identifies the independent role of dysplasia in the prediction of response to chemotherapy in AML

The independent prognostic impact of specific dysplastic features in acute myeloid leukemia (AML) remains controversial and may vary between genomic subtypes. We apply a machine learning framework to dissect the relative contribution of centrally reviewed dysplastic features and oncogenetics in 190 patients with de novo AML treated in ALFA clinical trials. One hundred and thirty-five (71%) patients achieved complete response after the first induction course (CR). Dysgranulopoiesis, dyserythropoiesis and dysmegakaryopoiesis were assessable in 84%, 83% and 63% patients, respectively. Multi-lineage dysplasia was present in 27% of assessable patients. Micromegakaryocytes (q = 0.01), hypolobulated megakaryocytes (q = 0.08) and hyposegmented granulocytes (q = 0.08) were associated with higher ELN-2017 risk. Using a supervised learning algorithm, the relative importance of morphological variables (34%) for the prediction of CR was higher than demographic (5%), clinical (2%), cytogenetic (25%), molecular (29%), and treatment (5%) variables. Though dysplasias had limited predictive impact on survival, a multivariate logistic regression identified the presence of hypolobulated megakaryocytes (p = 0.014) and micromegakaryocytes (p = 0.035) as predicting lower CR rates, independently of monosomy 7 (p = 0.013), TP53 (p = 0.004), and NPM1 mutations (p = 0.025). Assessment of these specific dysmegakarypoiesis traits, for which we identify a transcriptomic signature, may thus guide treatment allocation in AML.

Clonal hematopoiesis of indeterminate potential and cardiovascular events in systemic lupus erythematosus (HEMATOPLUS study)

OBJECTIVE

The detection of somatic mutations in genes of myeloid cells in asymptomatic patients - defining clonal hematopoiesis of indeterminate potential (CHIP) - predisposes to cardiovascular events (CVE) in the general population. We aimed to determine whether CHIP was associated with CVE in SLE patients.

METHODS

The study is an ancillary study of the randomized, double-blind, placebo-controlled, multicenter trial PLUS study conducted from June 2007 through August 2010 at 37 centers in France involving 573 SLE patients. The search for somatic mutations by high-throughput sequencing of 53 genes involved in clonal hematopoiesis was performed on genomic DNA collected at PLUS inclusion. The CHIP prevalence was assessed in SLE and in a retrospective cohort of 479 patients free of hematological malignancy. The primary outcome was the incident CVE in SLE.

RESULTS

Screening for CHIP was performed in 438 SLE patients (38 [29-47] years, 91·8% female). Overall, 63 somatic mutations were identified in 47 patients defining a CHIP prevalence of 10·7% in SLE. Most SLE patients (78·7%) carried a single mutation. Most variants (62·5%) were located in the DNMT3A gene. CHIP was associated with age, age at SLE diagnosis and a lower frequency of antiphospholipid antibodies. CHIP occurred more than 20-years earlier (p < 0·00001) in SLE than in controls. The detection of CHIP at inclusion was not associated with the occurrence of CVE during follow up (HR = 0·42 (0·06 - 3·21), p = 0·406).

CONCLUSION

The prevalence of CHIP is high in SLE with respect to age but was not associated with incident CVE.

CLINICAL TRIAL REGISTRATION

ClinicalTrials.gov, https://clinicaltrials.gov, NCT05146414.

2021 Update on MRD in acute myeloid leukemia: a consensus document from the European LeukemiaNet MRD Working Party

Measurable residual disease (MRD) is an important biomarker in acute myeloid leukemia (AML) that is used for prognostic, predictive, monitoring, and efficacy-response assessments. The European LeukemiaNet (ELN) MRD Working Party evaluated standardization and harmonization of MRD in an ongoing manner and has updated the 2018 ELN MRD recommendations based on significant developments in the field. New and revised recommendations were established during in-person and online meetings, and a 2-stage Delphi poll was conducted to optimize consensus. All recommendations are graded by levels of evidence and agreement. Major changes include technical specifications for next-generation sequencing-based MRD testing and integrative assessments of MRD irrespective of technology. Other topics include use of MRD as a prognostic and surrogate end point for drug testing; selection of the technique, material, and appropriate time points for MRD assessment; and clinical implications of MRD assessment. In addition to technical recommendations for flow- and molecular-MRD analysis, we provide MRD thresholds and define MRD response, and detail how MRD results should be reported and combined if several techniques are used. MRD assessment in AML is complex and clinically relevant, and standardized approaches to application, interpretation, technical conduct, and reporting are of critical importance.

Genome-wide association study identifies susceptibility loci for acute myeloid leukemia

Acute myeloid leukemia (AML) is a hematological malignancy with an undefined heritable risk. Here we perform a meta-analysis of three genome-wide association studies, with replication in a fourth study, incorporating a total of 4018 AML cases and 10488 controls. We identify a genome-wide significant risk locus for AML at 11q13.2 (rs4930561; P = 2.15 × 10-8; KMT5B). We also identify a genome-wide significant risk locus for the cytogenetically normal AML sub-group (N = 1287) at 6p21.32 (rs3916765; P = 1.51 × 10-10; HLA). Our results inform on AML etiology and identify putative functional genes operating in histone methylation (KMT5B) and immune function (HLA).

Standardisation of pathogenicity classification for somatic alterations in solid tumours and haematologic malignancies

BACKGROUND

The difficulty in interpreting somatic alterations is correlated with the increase in sequencing panel size. To correctly guide the clinical management of patients with cancer, there needs to be accurate classification of pathogenicity followed by actionability assessment. Here, we describe a specific detailed workflow for the classification of the pathogenicity of somatic variants in cancer into five categories: benign, likely benign, unknown significance, likely pathogenic and pathogenic.

METHODS

Classification is obtained by combining a set of eight relevant criteria in favour of either a pathogenic or a benign effect (pathogenic stand-alone, pathogenic very strong, pathogenic strong, pathogenic moderate, pathogenic supporting, benign supporting, benign strong and benign stand-alone).

RESULTS

Our guide is concordant with the ACMG/AMP 2015 guidelines for germline variants. Interpretation of somatic variants requires considering specific criteria, such as the disease and therapeutic context, co-occurring genomic events in the tumour when available and the use of cancer-specific variant databases. In addition, the gene role in tumorigenesis (oncogene or tumour suppressor gene) also needs to be taken into consideration.

CONCLUSION

Our classification could contribute to homogenize best practices on somatic variant pathogenicity interpretation and improve interpretation consistency both within and between laboratories.

Genetic identification of patients with AML older than 60 years achieving long-term survival with intensive chemotherapy

To design a simple and reproducible classifier predicting the overall survival (OS) of patients with acute myeloid leukemia (AML) ≥60 years of age treated with 7 + 3, we sequenced 37 genes in 471 patients from the ALFA1200 (Acute Leukemia French Association) study (median age, 68 years). Mutation patterns and OS differed between the 84 patients with poor-risk cytogenetics and the 387 patients with good (n = 13), intermediate (n = 339), or unmeasured (n = 35) cytogenetic risk. TP53 (hazards ratio [HR], 2.49; P = .0003) and KRAS (HR, 3.60; P = .001) mutations independently worsened the OS of patients with poor-risk cytogenetics. In those without poor-risk cytogenetics, NPM1 (HR, 0.57; P = .0004), FLT3 internal tandem duplications with low (HR, 1.85; P = .0005) or high (HR, 3.51; P < 10-4) allelic ratio, DNMT3A (HR, 1.86; P < 10-4), NRAS (HR, 1.54; P = .019), and ASXL1 (HR, 1.89; P = .0003) mutations independently predicted OS. Combining cytogenetic risk and mutations in these 7 genes, 39.1% of patients could be assigned to a "go-go" tier with a 2-year OS of 66.1%, 7.6% to the "no-go" group (2-year OS 2.8%), and 3.3% of to the "slow-go" group (2-year OS of 39.1%; P < 10-5). Across 3 independent validation cohorts, 31.2% to 37.7% and 11.2% to 13.5% of patients were assigned to the go-go and the no-go tiers, respectively, with significant differences in OS between tiers in all 3 trial cohorts (HDF [Hauts-de-France], n = 141, P = .003; and SAL [Study Alliance Leukemia], n = 46; AMLSG [AML Study Group], n = 223, both P < 10-5). The ALFA decision tool is a simple, robust, and discriminant prognostic model for AML patients ≥60 years of age treated with intensive chemotherapy. This model can instruct the design of trials comparing the 7 + 3 standard of care with less intensive regimens.

Long-term outcome of imatinib 400 mg compared to imatinib 600 mg or imatinib 400 mg daily in combination with cytarabine or pegylated interferon alpha 2a for chronic myeloid leukaemia: results from the French SPIRIT phase III randomised trial

The STI571 prospective randomised trial (SPIRIT) French trial is a four-arm study comparing imatinib (IM) 400 mg versus IM 600 mg, IM 400 mg + cytarabine (AraC), and IM 400 mg + pegylated interferon alpha2a (PegIFN-α2a) for the front-line treatment of chronic-phase chronic myeloid leukaemia (CML). Long-term analyses included overall and progression-free survival, molecular responses to treatment, and severe adverse events. Starting in 2003, the trial included 787 evaluable patients. The median overall follow-up of the patients was 13.5 years (range 3 months to 16.7 years). Based on intention-to-treat analyses, at 15 years, overall and progression-free survival were similar across arms: 85%, 83%, 80%, and 82% and 84%, 87%, 79%, and 79% for the IM 400 mg (N = 223), IM 600 mg (N = 171), IM 400 mg + AraC (N = 172), and IM 400 mg + PegIFN-α2a (N = 221) arms, respectively. The rate of major molecular response at 12 months and deep molecular response (MR4) over time were significantly higher with the combination IM 400 mg + PegIFN-α2a than with IM 400 mg: p = 0.0001 and p = 0.0035, respectively. Progression to advanced phases and secondary malignancies were the most frequent causes of death. Toxicity was the main reason for stopping AraC or PegIFN-α2a treatment.

Minimal residual disease monitoring in acute myeloid leukemia with non-A/B/D-NPM1 mutations by digital polymerase chain reaction: feasibility and clinical use

Not available.

Optical genome mapping, a promising alternative to gold standard cytogenetic approaches in a series of acute lymphoblastic leukemias

Acute lymphoblastic leukemias (ALL) are characterized by a large number of cytogenetic abnormalities of clinical interest that require the use of several complementary techniques. Optical genome mapping (OGM) is based on analysis of ultra-high molecular weight DNA molecules that provides a high-resolution genome-wide analysis highlighting copy number and structural anomalies, including balanced translocations. We compared OGM to standard techniques (karyotyping, fluorescent in situ hybridization, single nucleotide polymorphism-array and reverse transcription multiplex ligation-dependent probe amplification) in 10 selected B or T-ALL. Eighty abnormalities were found using standard techniques of which 72 (90%) were correctly detected using OGM. Eight discrepancies were identified, while 12 additional anomalies were found by OGM. Among the discrepancies, four were detected in raw data but not retained because of filtering issues. However, four were truly missed, either because of a low variant allele frequency or because of a low coverage of some regions. Of the additional anomalies revealed by OGM, seven were confirmed by another technique, some of which are recurrent in ALL such as LMO2-TRA and MYC-TRB fusions. Despite false positive anomalies due to background noise and a case of inter-sample contamination secondarily identified, the OGM technology was relatively simple to use with little practice. Thus, OGM represents a promising alternative to cytogenetic techniques currently performed for ALL characterization. It enables a time and cost effective analysis allowing identification of complex cytogenetic events, including those currently inaccessible to standard techniques.

Targeting RUNX1 in acute myeloid leukemia: preclinical innovations and therapeutic implications

Introduction: RUNX1 is an essential transcription factor for normal and malignant hematopoiesis. RUNX1 forms a heterodimeric complex with CBFB. Germline mutations and somatic alterations (i.e. translocations, mutations and abnormal expression) are frequently associated with acute myeloid leukemia (AML) with RUNX1 mutations conferring unfavorable prognosis. Therefore, RUNX1 constitutes a potential innovative and interesting therapeutic target. In this review, we discuss recent therapeutic advances of RUNX1 targeting in AML.Areas covered: Firstly, we cover the clinical basis for RUNX1 targeting. We have subdivided recent therapeutic approaches either by common biochemical pathways or by similar pharmacological targets. Genome editing of RUNX1 induces anti-leukemic effects; however, off-target events prohibit clinical use. Several molecules inhibit the interaction between RUNX1/CBFB and control AML development and progression. BET protein antagonists target RUNX1 (i.e. specific BET inhibitors, BRD4 shRNRA, proteolysis targeting chimeras (PROTAC) or expression-mimickers). All these molecules improve survival in mutant RUNX1 AML preclinical models.Expert opinion: Some of these novel molecules have shown encouraging anti-leukemic potency at the preclinical stage. A better understanding of RUNX1 function in AML development and progression and its key downstream pathways, may result in more precise and more efficient RUNX1 targeting therapies.

Epidemiology, clinical picture and long-term outcomes of FIP1L1-PDGFRA-positive myeloid neoplasm with eosinophilia: Data from 151 patients

FIP1L1-PDGFRA-positive myeloid neoplasm with eosinophilia (F/P+ MN-eo) is a rare disease: robust epidemiological data are lacking and reported issues are scarce, of low sample-size and limited follow-up. Imatinib mesylate (IM) is highly efficient but no predictive factor of relapse after discontinuation has yet been identified. One hundred and fifty-one patients with F/P+ MN-eo (143 males; mean age at diagnosis 49 years; mean annual incidence: 0.18 case per million population) were included in this retrospective nationwide study involving all French laboratories who perform the search of F/P fusion gene (study period: 2003-2019). The main organs involved included the spleen (44%), skin (32%), lungs (30%), heart (19%) and central nervous system (9%). Serum vitamin B12 and tryptase levels were elevated in 74/79 (94%) and 45/57 (79%) patients, respectively, and none of the 31 patients initially treated with corticosteroids achieved complete hematologic remission. All 148 (98%) IM-treated patients achieved complete hematologic and molecular (when tested, n = 84) responses. Forty-six patients eventually discontinued IM, among whom 20 (57%) relapsed. In multivariate analysis, time to IM initiation (continuous HR: 1,01 [0.99-1,03]; P = .05) and duration of IM treatment (continuous HR: 0,97 [0,95-0,99]; P = .004) were independent factors of relapse after discontinuation of IM. After a mean follow-up of 80 (56) months, the 1, 5- and 10-year overall survival rates in IM-treated patients were 99%, 95% and 84% respectively. In F/P+ MN-eo, prompt initiation of IM and longer treatment durations may prevent relapses after discontinuation of IM.

Plasmacytoid dendritic cells proliferation associated with acute myeloid leukemia: phenotype profile and mutation landscape

Neoplasms involving plasmacytoid Dendritic Cells (pDCs) include Blastic pDC Neoplasms (BPDCN) and other pDC proliferations, where pDCs are associated with myeloid malignancies: most frequently Chronic MyeloMonocytic Leukemia (CMML) but also Acute Myeloid Leukemia (AML), hereafter named pDC-AML. We aimed to determine the reactive or neoplastic origin of pDCs in pDC-AML, and their link with the CD34+ blasts, monocytes or conventional DCs (cDCs) associated in the same sample, by phenotypic and molecular analyses (targeted NGS, 70 genes). We compared 15 pDC-AML at diagnosis with 21 BPDCN and 11 normal pDCs from healthy donors. CD45low CD34+ blasts were found in all cases (10-80% of medullar cells), associated with pDCs (4-36%), monocytes in 14 cases (1-10%) and cDCs (2 cases, 4.8-19%). pDCs in pDC-AML harbor a clearly different phenotype from BPDCN: CD4+ CD56- in 100% of cases, most frequently CD303+, CD304+ and CD34+; lower expression of cTCL1 and CD123 with isolated lymphoid markers (CD22/CD7/CD5) in some cases, suggesting a pre-pDC stage. In all cases, pDCs, monocytes and cDC are neoplastic since they harbor the same mutations as CD34+ blasts. RUNX1 is the most commonly mutated gene: detected in all AML with minimal differentiation (M0-AML) but not in the other cases. Despite low number of cases, the systematic association between M0-AML, RUNX1 mutations and an excess of pDC is puzzling. Further evaluation in a larger cohort is required to confirm RUNX1 mutations in pDC-AML with minimal differentiation and to investigate whether it represents a proliferation of blasts with macrophage and DC progenitor potential.

Effects of azacitidine in 93 patients with IDH1/2 mutated acute myeloid leukemia/myelodysplastic syndromes: a French retrospective multicenter study

Isocitrate dehydrogenase 1 (IDH1) and 2 (IDH2) mutations in Myeloid Neoplams (MNs) exhibit DNA hypermethylation via 2-hydroxyglutarate (2HG) over-production. Clinical impact of azacitidine (AZA) remains inconsistent in IDH1/2-mutated MNs and the potential of serum 2HG as a suitable marker of response to AZA is unknown. To address these questions, we retrospectively analyzed 93 MNs patients (78 AML, 11 MDS, 4 CMML) with IDH1/2 mutations treated with AZA. After a median of 5 cycles of AZA, overall response rate was 28% (including 15% complete remission) and median OS was 12.3 months (significantly shorter in AML compared to MDS/CMML patients). In multivariate analysis of AML patients, DNMT3A mutation was associated with shorter OS while IDH1/2 mutation subtypes had no independent impact. No difference was observed in serum 2HG levels upon AZA treatment between responding and refractory patients suggesting that serum 2HG cannot be used as a surrogate marker of AZA response.

The Folate Cycle Enzyme MTHFR Is a Critical Regulator of Cell Response to MYC-Targeting Therapies

Deciphering the impact of metabolic intervention on response to anticancer therapy may elucidate a path toward improved clinical responses. Here, we identify amino acid-related pathways connected to the folate cycle whose activation predicts sensitivity to MYC-targeting therapies in acute myeloid leukemia (AML). We establish that folate restriction and deficiency of the rate-limiting folate cycle enzyme MTHFR, which exhibits reduced-function polymorphisms in about 10% of Caucasians, induce resistance to MYC targeting by BET and CDK7 inhibitors in cell lines, primary patient samples, and syngeneic mouse models of AML. Furthermore, this effect is abrogated by supplementation with the MTHFR enzymatic product CH₃-THF. Mechanistically, folate cycle disturbance reduces H3K27/K9 histone methylation and activates a SPI1 transcriptional program counteracting the effect of BET inhibition. Our data provide a rationale for screening MTHFR polymorphisms and folate cycle status to nominate patients most likely to benefit from MYC-targeting therapies. SIGNIFICANCE: Although MYC-targeting therapies represent a promising strategy for cancer treatment, evidence of predictors of sensitivity to these agents is limited. We pinpoint that folate cycle disturbance and frequent polymorphisms associated with reduced MTHFR activity promote resistance to BET inhibitors. CH₃-THF supplementation thus represents a low-risk intervention to enhance their effects.See related commentary by Marando and Huntly, p. 1791.This article is highlighted in the In This Issue feature, p. 1775.

Biomarkers of Gemtuzumab Ozogamicin Response for Acute Myeloid Leukemia Treatment

Gemtuzumab ozogamicin (GO, Mylotarg®) consists of a humanized CD33-targeted antibody-drug conjugated to a calicheamicin derivative. Growing evidence of GO efficacy in acute myeloid leukemia (AML), demonstrated by improved outcomes in CD33-positive AML patients across phase I to III clinical trials, led to the Food and Drug Administration (FDA) approval on 1 September 2017 in CD33-positive AML patients aged 2 years and older. Discrepancies in GO recipients outcome have raised significant efforts to characterize biomarkers predictive of GO response and have refined the subset of patients that may strongly benefit from GO. Among them, CD33 expression levels, favorable cytogenetics (t(8;21), inv(16)/t(16;16), t(15;17)) and molecular alterations, such as NPM1, FLT3-internal tandem duplications and other signaling mutations, represent well-known candidates. Additionally, in depth analyses including minimal residual disease monitoring, stemness expression (LSC17 score), mutations or single nucleotide polymorphisms in GO pathway genes (CD33, ABCB1) and molecular-derived scores, such as the recently set up CD33_PGx6_Score, represent promising markers to enhance GO response prediction and improve patient management.

The complex genetic landscape of familial MDS and AML reveals pathogenic germline variants

The inclusion of familial myeloid malignancies as a separate disease entity in the revised WHO classification has renewed efforts to improve the recognition and management of this group of at risk individuals. Here we report a cohort of 86 acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS) families with 49 harboring germline variants in 16 previously defined loci (57%). Whole exome sequencing in a further 37 uncharacterized families (43%) allowed us to rationalize 65 new candidate loci, including genes mutated in rare hematological syndromes (ADA, GP6, IL17RA, PRF1 and SEC23B), reported in prior MDS/AML or inherited bone marrow failure series (DNAH9, NAPRT1 and SH2B3) or variants at novel loci (DHX34) that appear specific to inherited forms of myeloid malignancies. Altogether, our series of MDS/AML families offer novel insights into the etiology of myeloid malignancies and provide a framework to prioritize variants for inclusion into routine diagnostics and patient management.

Vitamin D Receptor Controls Cell Stemness in Acute Myeloid Leukemia and in Normal Bone Marrow

Vitamin D (VD) is a known differentiating agent, but the role of VD receptor (VDR) is still incompletely described in acute myeloid leukemia (AML), whose treatment is based mostly on antimitotic chemotherapy. Here, we present an unexpected role of VDR in normal hematopoiesis and in leukemogenesis. Limited VDR expression is associated with impaired myeloid progenitor differentiation and is a new prognostic factor in AML. In mice, the lack of Vdr results in increased numbers of hematopoietic and leukemia stem cells and quiescent hematopoietic stem cells. In addition, malignant transformation of Vdr-/- cells results in myeloid differentiation block and increases self-renewal. Vdr promoter is methylated in AML as in CD34+ cells, and demethylating agents induce VDR expression. Association of VDR agonists with hypomethylating agents promotes leukemia stem cell exhaustion and decreases tumor burden in AML mouse models. Thus, Vdr functions as a regulator of stem cell homeostasis and leukemic propagation.