Disparities in acute myeloid leukemia treatments and outcomes.

PURPOSE OF REVIEW: This review aims to summarize different contributors to survival disparities in acute myeloid leukemia (AML) patients. The focus is set on African-American (hereafter referred to as Black) patients, with separate consideration of self-reported race and ancestry. It aims to also highlight the interconnectivity of the different features that impact on despair survival. RECENT FINDINGS: The main themes in the literature covered in this article include the impact of social deprivation, clinical trial enrollment and biobanking, structural racism and ancestry-associated differences in genetic features on survival outcomes. SUMMARY: An increasing number of studies have not only shown persistent survival disparities between Black and non-Hispanic White AML patients, but uncovered a multitude of contributors that have additive adverse effects on patient outcomes. In addition to potentially modifiable features, such as socioeconomic factors and trial enrollment odds that require urgent interventions, there is emerging data on differences in disease biology with respect to genetic ancestry, including frequencies of known AML-driver mutations and their associated prognostic impact.

DNA methylation epitypes highlight underlying developmental and disease pathways in acute myeloid leukemia.

Acute myeloid leukemia (AML) is a molecularly complex disease characterized by heterogeneous tumor genetic profiles and involving numerous pathogenic mechanisms and pathways. Integration of molecular data types across multiple patient cohorts may advance current genetic approaches for improved subclassification and understanding of the biology of the disease. Here, we analyzed genome-wide DNA methylation in 649 AML patients using Illumina arrays and identified a configuration of 13 subtypes (termed "epitypes") using unbiased clustering. Integration of genetic data revealed that most epitypes were associated with a certain recurrent mutation (or combination) in a majority of patients, yet other epitypes were largely independent. Epitypes showed developmental blockage at discrete stages of myeloid differentiation, revealing epitypes that retain arrested hematopoietic stem-cell-like phenotypes. Detailed analyses of DNA methylation patterns identified unique patterns of aberrant hyper- and hypomethylation among epitypes, with variable involvement of transcription factors influencing promoter, enhancer, and repressed regions. Patients in epitypes with stem-cell-like methylation features showed inferior overall survival along with up-regulated stem cell gene expression signatures. We further identified a DNA methylation signature involving STAT motifs associated with FLT3-ITD mutations. Finally, DNA methylation signatures were stable at relapse for the large majority of patients, and rare epitype switching accompanied loss of the dominant epitype mutations and reversion to stem-cell-like methylation patterns. These results show that DNA methylation-based classification integrates important molecular features of AML to reveal the diverse pathogenic and biological aspects of the disease.

High-dimensional genomic feature selection with the ordered stereotype logit model.

For many high-dimensional genomic and epigenomic datasets, the outcome of interest is ordinal. While these ordinal outcomes are often thought of as the observed cutpoints of some latent continuous variable, some ordinal outcomes are truly discrete and are comprised of the subjective combination of several factors. The nonlinear stereotype logistic model, which does not assume proportional odds, was developed for these 'assessed' ordinal variables. It has previously been extended to the frequentist high-dimensional feature selection setting, but the Bayesian framework provides some distinct advantages in terms of simultaneous uncertainty quantification and variable selection. Here, we review the stereotype model and Bayesian variable selection methods and demonstrate how to combine them to select genomic features associated with discrete ordinal outcomes. We compared the Bayesian and frequentist methods in terms of variable selection performance. We additionally applied the Bayesian stereotype method to an acute myeloid leukemia RNA-sequencing dataset to further demonstrate its variable selection abilities by identifying features associated with the European LeukemiaNet prognostic risk score.

Long-term survival can be achieved in a significant fraction of older patients with core binding factor acute myeloid leukemia treated with intensive chemotherapy.

Acute Myeloid Leukemia is mainly a disease of the elderly: however, the knowledge on the outcomes of treatment in core binding factor AML (CBFAML) in older population, is limited. We retrospectively collected data on 229 patients with CBF- AML followed long-term in the last two decades. A 5-year overall survival (OS) of 44.2% (95%CI, 39.9-47.5) and a 5-year event - free survival (EFS) of 32.9% (95%CI, 25.5-40.1) was observed. In a subgroup of >70-year patients who completed intensive therapy (induction + >3 courses of consolidation including autologous stem cell transplant: 10 patients) the median EFS was 11.8 months (95%CI, 9.4 - 15.2) and OS was 40.0% (95%CI, 36.4 - 44.1) at 5yr. In univariate analysis, age >70 (hazard ratio (HR) 1.78, [95%CI, 1.15 - 2.54], p=.008), failure to achieve remission following induction (HR, 8.96 [95%CI, 5.5 - 13.8], p=<.0001), no consolidation therapy (HR, 0.75 [95%CI, 0.47 - 1.84], p=.04) and less than 3 cycles of consolidation (HR, 1.48 [95%CI, 0.75 - 3.2], p=.0004), predicted poorer EFS. Our study shows that intensive therapy, in selected older CBF-AML patients, leads to longer survival. Achieving a CR seems to be the most important first step and at least 3 cycles of consolidation, an important second one. The analysis suggests that these patients should not be excluded from studies with intensive therapies.

Immunosuppression and outcomes in adult patients with de novo acute myeloid leukemia with normal karyotypes.

Acute myeloid leukemia (AML) patients rarely have long first remissions (LFRs; >5 y) after standard-of-care chemotherapy, unless classified as favorable risk at presentation. Identification of the mechanisms responsible for long vs. more typical, standard remissions may help to define prognostic determinants for chemotherapy responses. Using exome sequencing, RNA-sequencing, and functional immunologic studies, we characterized 28 normal karyotype (NK)-AML patients with >5 y first remissions after chemotherapy (LFRs) and compared them to a well-matched group of 31 NK-AML patients who relapsed within 2 y (standard first remissions [SFRs]). Our combined analyses indicated that genetic-risk profiling at presentation (as defined by European LeukemiaNet [ELN] 2017 criteria) was not sufficient to explain the outcomes of many SFR cases. Single-cell RNA-sequencing studies of 15 AML samples showed that SFR AML cells differentially expressed many genes associated with immune suppression. The bone marrow of SFR cases had significantly fewer CD4+ Th1 cells; these T cells expressed an exhaustion signature and were resistant to activation by T cell receptor stimulation in the presence of autologous AML cells. T cell activation could be restored by removing the AML cells or blocking the inhibitory major histocompatibility complex class II receptor, LAG3. Most LFR cases did not display these features, suggesting that their AML cells were not as immunosuppressive. These findings were confirmed and extended in an independent set of 50 AML cases representing all ELN 2017 risk groups. AML cell-mediated suppression of CD4+ T cell activation at presentation is strongly associated with unfavorable outcomes in AML patients treated with standard chemotherapy.

Mutational landscape and clinical outcome of patients with de novo acute myeloid leukemia and rearrangements involving 11q23/KMT2A.

Balanced rearrangements involving the KMT2A gene, located at 11q23, are among the most frequent chromosome aberrations in acute myeloid leukemia (AML). Because of numerous fusion partners, the mutational landscape and prognostic impact of specific 11q23/KMT2A rearrangements are not fully understood. We analyzed clinical features of 172 adults with AML and recurrent 11q23/KMT2A rearrangements, 141 of whom had outcome data available. We compared outcomes of these patients with outcomes of 1,097 patients without an 11q23/KMT2A rearrangement categorized according to the 2017 European LeukemiaNet (ELN) classification. Using targeted next-generation sequencing, we investigated the mutational status of 81 leukemia/cancer-associated genes in 96 patients with 11q23/KMT2A rearrangements with material for molecular studies available. Patients with 11q23/KMT2A rearrangements had a low number of additional gene mutations (median, 1; range 0 to 6), which involved the RAS pathway (KRAS, NRAS, and PTPN11) in 32% of patients. KRAS mutations occurred more often in patients with t(6;11)(q27;q23)/KMT2A-AFDN compared with patients with the other 11q23/KMT2A subsets. Specific gene mutations were too infrequent in patients with specific 11q23/KMT2A rearrangements to assess their associations with outcomes. We demonstrate that younger (age <60 y) patients with t(9;11)(p22;q23)/KMT2A-MLLT3 had better outcomes than patients with other 11q23/KMT2A rearrangements and those without 11q23/KMT2A rearrangements classified in the 2017 ELN intermediate-risk group. Conversely, outcomes of older patients (age ≥60 y) with t(9;11)(p22;q23) were poor and comparable to those of the ELN adverse-risk group patients. Our study shows that patients with an 11q23/KMT2A rearrangement have distinct mutational patterns and outcomes depending on the fusion partner.

Clinical and molecular relevance of genetic variants in the non-coding transcriptome of patients with cytogenetically normal acute myeloid leukemia.

Expression levels of long non-coding RNA (lncRNA) have been shown to associate with clinical outcome of patients with cytogenetically normal acute myeloid leukemia (CN-AML). However, the frequency and clinical significance of genetic variants in the nucleotide sequences of lncRNA in AML patients is unknown. Herein, we analyzed total RNA sequencing data of 377 younger adults (aged <60 years) with CN-AML, who were comprehensively characterized with regard to clinical outcome. We used available genomic databases and stringent filters to annotate genetic variants unequivocally located in the non-coding transcriptome of AML patients. We detected 981 variants, which are recurrently present in lncRNA that are expressed in leukemic blasts. Among these variants, we identified a cytosine-to-thymidine variant in the lncRNA RP5-1074L1.4 and a cytosine-to-thymidine variant in the lncRNA SNHG15, which independently associated with longer survival of CN-AML patients. The presence of the SNHG15 cytosine-to-thymidine variant was also found to associate with better outcome in an independent dataset of CN-AML patients, despite differences in treatment protocols and RNA sequencing techniques. In order to gain biological insights, we cloned and overexpressed both wild-type and variant versions of the SNHG15 lncRNA. In keeping with its negative prognostic impact, overexpression of the wild-type SNHG15 associated with higher proliferation rate of leukemic blasts when compared with the cytosine-to-thymidine variant. We conclude that recurrent genetic variants of lncRNA that are expressed in the leukemic blasts of CN-AML patients have prognostic and potential biological significance.

Controlled variable selection in Weibull mixture cure models for high-dimensional data.

Medical breakthroughs in recent years have led to cures for many diseases. The mixture cure model (MCM) is a type of survival model that is often used when a cured fraction exists. Many have sought to identify genomic features associated with a time-to-event outcome which requires variable selection strategies for high-dimensional spaces. Unfortunately, currently few variable selection methods exist for MCMs especially when there are more predictors than samples. This study develops high-dimensional penalized Weibull MCMs, which allow for identification of prognostic factors associated with both cure status and/or survival. We demonstrated how such models may be estimated using two different iterative algorithms. The model-X knockoffs method was combined with these algorithms to control the false discovery rate (FDR) in variable selection. Through extensive simulation studies, our penalized MCMs have been shown to outperform alternative methods on multiple metrics and achieve high statistical power with FDR being controlled. In an acute myeloid leukemia (AML) application with gene expression data, our proposed approach identified 14 genes associated with potential cure and 12 genes with time-to-relapse, which may help inform treatment decisions for AML patients.

Discovery of clinically relevant fusions in pediatric cancer.

BACKGROUND: Pediatric cancers typically have a distinct genomic landscape when compared to adult cancers and frequently carry somatic gene fusion events that alter gene expression and drive tumorigenesis. Sensitive and specific detection of gene fusions through the analysis of next-generation-based RNA sequencing (RNA-Seq) data is computationally challenging and may be confounded by low tumor cellularity or underlying genomic complexity. Furthermore, numerous computational tools are available to identify fusions from supporting RNA-Seq reads, yet each algorithm demonstrates unique variability in sensitivity and precision, and no clearly superior approach currently exists. To overcome these challenges, we have developed an ensemble fusion calling approach to increase the accuracy of identifying fusions. RESULTS: Our Ensemble Fusion (EnFusion) approach utilizes seven fusion calling algorithms: Arriba, CICERO, FusionMap, FusionCatcher, JAFFA, MapSplice, and STAR-Fusion, which are packaged as a fully automated pipeline using Docker and Amazon Web Services (AWS) serverless technology. This method uses paired end RNA-Seq sequence reads as input, and the output from each algorithm is examined to identify fusions detected by a consensus of at least three algorithms. These consensus fusion results are filtered by comparison to an internal database to remove likely artifactual fusions occurring at high frequencies in our internal cohort, while a "known fusion list" prevents failure to report known pathogenic events. We have employed the EnFusion pipeline on RNA-Seq data from 229 patients with pediatric cancer or blood disorders studied under an IRB-approved protocol. The samples consist of 138 central nervous system tumors, 73 solid tumors, and 18 hematologic malignancies or disorders. The combination of an ensemble fusion-calling pipeline and a knowledge-based filtering strategy identified 67 clinically relevant fusions among our cohort (diagnostic yield of 29.3%), including RBPMS-MET, BCAN-NTRK1, and TRIM22-BRAF fusions. Following clinical confirmation and reporting in the patient's medical record, both known and novel fusions provided medically meaningful information. CONCLUSIONS: The EnFusion pipeline offers a streamlined approach to discover fusions in cancer, at higher levels of sensitivity and accuracy than single algorithm methods. Furthermore, this method accurately identifies driver fusions in pediatric cancer, providing clinical impact by contributing evidence to diagnosis and, when appropriate, indicating targeted therapies.

Mutations associated with a 17-gene leukemia stem cell score and the score's prognostic relevance in the context of the European LeukemiaNet classification of acute myeloid leukemia.

Leukemia stem cells (LSC) are more resistant to standard chemotherapy and their persistence during remission can cause relapse, which is still one of the major clinical challenges in the treatment of acute myeloid leukemia (AML). A better understanding of the mutational patterns and the prognostic impact of molecular markers associated with stemness could lead to better clinical management and improve patients' outcomes. We applied a previously described 17-gene expression score comprising genes differently expressed between LSC and leukemic bulk blasts, for 934 adult patients with de novo AML, and studied associations of the 17-gene LSC score with clinical data and mutation status of 81 genes recurrently mutated in cancer and leukemia. We found that patients with a high 17-gene score were older and had more mutations. The 17-gene score was found to have a prognostic impact in both younger (aged <60 years) and older (aged ≥60 years) patients with AML. We also analyzed the 17-gene LSC score in the context of the 2017 European LeukemiaNet genetic-risk classification and found that for younger patients the score refined the classification, and identified patients currently classified in the European LeukemiaNet Favorable-risk category who had a worse outcome.

Identification of NRAS isoform 2 overexpression as a mechanism facilitating BRAF inhibitor resistance in malignant melanoma.

Activating mutations in BRAF are found in 50% of melanomas and although treatment with BRAF inhibitors (BRAFi) is effective, resistance often develops. We now show that recently discovered NRAS isoform 2 is up-regulated in the setting of BRAF inhibitor resistance in melanoma, in both cell lines and patient tumor tissues. When isoform 2 was overexpressed in BRAF mutant melanoma cell lines, melanoma cell proliferation and in vivo tumor growth were significantly increased in the presence of BRAFi treatment. shRNA-mediated knockdown of isoform 2 in BRAFi resistant cells restored sensitivity to BRAFi compared with controls. Signaling analysis indicated decreased mitogen-activated protein kinase (MAPK) pathway signaling and increased phosphoinositol-3-kinase (PI3K) pathway signaling in isoform 2 overexpressing cells compared with isoform 1 overexpressing cells. Immunoprecipitation of isoform 2 validated a binding affinity of this isoform to both PI3K and BRAF/RAF1. The addition of an AKT inhibitor to BRAFi treatment resulted in a partial restoration of BRAFi sensitivity in cells expressing high levels of isoform 2. NRAS isoform 2 may contribute to resistance to BRAFi by facilitating PI3K pathway activation.

MicroRNA-3151 inactivates TP53 in BRAF-mutated human malignancies.

The B-Raf proto-oncogene serine/threonine kinase (BRAF) gene is the most frequently mutated gene in malignant melanoma (MM) and papillary thyroid cancer (PTC) and is causally involved in malignant cell transformation. Mutated BRAF is associated with an aggressive disease phenotype, thus making it a top candidate for targeted treatment strategies in MM and PTC. We show that BRAF mutations in both MM and PTC drive increased expression of oncomiR-3151, which is coactivated by the SP1/NF-κB complex. Knockdown of microRNA-3151 (miR-3151) with short hairpin RNAs reduces cell proliferation and increases apoptosis of MM and PTC cells. Using a targeted RNA sequencing approach, we mechanistically determined that miR-3151 directly targets TP53 and other members of the TP53 pathway. Reducing miR-3151's abundance increases TP53's mRNA and protein expression and favors its nuclear localization. Consequently, knockdown of miR-3151 also leads to caspase-3-dependent apoptosis. Simultaneous inhibition of aberrantly activated BRAF and knockdown of miR-3151 potentiates the effects of sole BRAF inhibition with the BRAF inhibitor vemurafenib and may provide a novel targeted therapeutic approach in BRAF-mutated MM and PTC patients. In conclusion, we identify miR-3151 as a previously unidentified player in MM and PTC pathogenesis, which is driven by BRAF-dependent and BRAF-independent mechanisms. Characterization of TP53 as a downstream effector of miR-3151 provides evidence for a causal link between BRAF mutations and TP53 inactivation.

Prognostic and biologic significance of long non-coding RNA profiling in younger adults with cytogenetically normal acute myeloid leukemia.

Long non-coding ribonucleic acids (RNAs) are a novel class of RNA molecules, which are increasingly recognized as important molecular players in solid and hematologic malignancies. Herein we investigated whether long non-coding RNA expression is associated with clinical and molecular features, as well as outcome of younger adults (aged <60 years) with de novo cytogenetically normal acute myeloid leukemia. Whole transcriptome profiling was performed in a training (n=263) and a validation set (n=114). Using the training set, we identified 24 long non-coding RNAs associated with event-free survival. Linear combination of the weighted expression values of these transcripts yielded a prognostic score. In the validation set, patients with high scores had shorter disease-free (P<0.001), overall (P=0.002) and event-free survival (P<0.001) than patients with low scores. In multivariable analyses, long non-coding RNA score status was an independent prognostic marker for disease-free (P=0.01) and event-free survival (P=0.002), and showed a trend for overall survival (P=0.06). Among multiple molecular alterations tested, which are prognostic in cytogenetically normal acute myeloid leukemia, only double CEBPA mutations, NPM1 mutations and FLT3-ITD associated with distinct long non-coding RNA signatures. Correlation of the long non-coding RNA scores with messenger RNA and microRNA expression identified enrichment of genes involved in lymphocyte/leukocyte activation, inflammation and apoptosis in patients with high scores. We conclude that long non-coding RNA profiling provides meaningful prognostic information in younger adults with cytogenetically normal acute myeloid leukemia. In addition, expression of prognostic long non-coding RNAs associates with oncogenic molecular pathways in this disease. clinicaltrials.gov Identifier: 00048958 (CALGB-8461), 00899223 (CALGB-9665), and 00900224 (CALGB-20202).

NRAS isoforms differentially affect downstream pathways, cell growth, and cell transformation.

Neuroblastoma rat sarcoma (RAS) viral oncogene homolog (NRAS), a small GTPase, is one of the most thoroughly studied oncogenes that controls cell growth, differentiation, and survival by facilitating signal transduction. Here, we identify four novel naturally occurring NRAS isoforms (isoforms 2-5) in addition to the canonical isoform (isoform 1). Expression analyses performed on a panel of several different human malignancies and matching normal tissue revealed distinct isoform expression patterns. Two of the novel isoforms were found in the nucleus and cytoplasm, whereas the others were exclusively cytoplasmic. The isoforms varied in their binding affinities to known downstream targets and differentially regulated the RAS signaling pathway. Strikingly, forced expression of isoform 5, which encodes only a 20-aa peptide, led to increased cell proliferation and to transformation by activation of known NRAS targets. These discoveries open new avenues in the study of NRAS.

Expression and prognostic impact of lncRNAs in acute myeloid leukemia.

Long noncoding RNAs (lncRNAs) are transcripts longer than 200 nucleotides, located within the intergenic stretches or overlapping antisense transcripts of protein coding genes. LncRNAs are involved in numerous biological roles including imprinting, epigenetic regulation, apoptosis, and cell cycle. To determine whether lncRNAs are associated with clinical features and recurrent mutations in older patients (aged ≥60 y) with cytogenetically normal (CN) acute myeloid leukemia (AML), we evaluated lncRNA expression in 148 untreated older CN-AML cases using a custom microarray platform. An independent set of 71 untreated older patients with CN-AML was used to validate the outcome scores using RNA sequencing. Distinctive lncRNA profiles were found associated with selected mutations, such as internal tandem duplications in the FLT3 gene (FLT3-ITD) and mutations in the NPM1, CEBPA, IDH2, ASXL1, and RUNX1 genes. Using the lncRNAs most associated with event-free survival in a training cohort of 148 older patients with CN-AML, we derived a lncRNA score composed of 48 lncRNAs. Patients with an unfavorable compared with favorable lncRNA score had a lower complete response (CR) rate [P < 0.001, odds ratio = 0.14, 54% vs. 89%], shorter disease-free survival (DFS) [P < 0.001, hazard ratio (HR) = 2.88] and overall survival (OS) (P < 0.001, HR = 2.95). The validation set analyses confirmed these results (CR, P = 0.03; DFS, P = 0.009; OS, P = 0.009). Multivariable analyses for CR, DFS, and OS identified the lncRNA score as an independent marker for outcome. In conclusion, lncRNA expression in AML is closely associated with recurrent mutations. A small subset of lncRNAs is correlated strongly with treatment response and survival.

MonoSeq Variant Caller Reveals Novel Mononucleotide Run Indel Mutations in Tumors with Defective DNA Mismatch Repair.

Next-generation sequencing has revolutionized cancer genetics, but accurately detecting mutations in repetitive DNA sequences, especially mononucleotide runs, remains a challenge. This is a particular concern for tumors with defective mismatch repair (MMR) that accumulate strand-slippage mutations. We developed MonoSeq to improve indel mutation detection in mononucleotide runs, and used MonoSeq to investigate strand-slippage mutations in endometrial cancers, a tumor type that has frequent loss of MMR. We performed extensive Sanger sequencing to validate both clonal and subclonal MonoSeq mutation calls. Eighty-one regions containing mononucleotide runs were sequenced in 540 primary endometrial cancers (223 with defective MMR). Our analyses revealed that the overall mutation rate in MMR-deficient tumors was 20-30-fold higher than in MMR-normal tumors. MonoSeq analysis identified several previously unreported mutations, including a novel hotspot in an A7 run in the terminal exon of ARID5B.The ARID5B indel mutations were seen in both MMR-deficient and MMR-normal tumors, suggesting biologic selection. The analysis of tumor mRNAs revealed the presence of mutant transcripts that could result in translation of neopeptides. Improved detection of mononucleotide run strand-slippage mutations has clear implications for comprehensive mutation detection in tumors with defective MMR. Indel frameshift mutations and the resultant antigenic peptides could help guide immunotherapy strategies.