Malignant transformation is frequently associated with disease-specific epigenetic alterations, but the underlying mechanisms and pathophysiological consequences remain poorly understood. Here, we used global comparative DNA methylation profiling at CG-rich regions of 27 acute myeloid leukemia (AML) samples to select a subset of aberrantly methylated CG-rich regions (~400 regions, ~15,000 CpGs) for quantitative DNA methylation profiling in a large cohort of AML patients (n = 196) using MALDI-TOF analysis of bisulfite-treated DNA. Meta-analysis separated a subgroup of CG-rich regions showing highly correlated DNA methylation changes that were marked by histone H3 lysine 27 trimethylation in normal hematopoietic progenitor cells. While the group of non-polycomb group (PcG) target regions displayed methylation patterns that correlated well with molecular and cytogenetic markers, PcG target regions displayed a much weaker association with genetic features. However, the degree of methylation gain across the latter panel showed significant correlation with active DNMT3A levels and with overall survival. Our study suggests that both epigenetic as well as genetic aberrations underlay AML-related changes in DNA methylation at CG-rich regions and that the former may provide a marker to improve classification and prognostication of adult AML patients.

Profiling of aberrant DNA methylation in acute myeloid leukemia reveals subclasses of CG-rich regions with epigenetic or genetic association.

BACKGROUND
Machine learning (ML) has been applied to an increasing number of predictive problems in laboratory medicine, and published work to date suggests that it has tremendous potential for clinical applications. However, a number of groups have noted the potential pitfalls associated with this work, particularly if certain details of the development and validation pipelines are not carefully controlled.


METHODS
To address these pitfalls and other specific challenges when applying machine learning in a laboratory medicine setting, a working group of the International Federation for Clinical Chemistry and Laboratory Medicine was convened to provide a guidance document for this domain.


RESULTS
This manuscript represents consensus recommendations for best practices from that committee, with the goal of improving the quality of developed and published ML models designed for use in clinical laboratories.


CONCLUSIONS
The committee believes that implementation of these best practices will improve the quality and reproducibility of machine learning utilized in laboratory medicine.


SUMMARY
We have provided our consensus assessment of a number of important practices that are required to ensure that valid, reproducible machine learning (ML) models can be applied to address operational and diagnostic questions in the clinical laboratory. These practices span all phases of model development, from problem formulation through predictive implementation. Although it is not possible to exhaustively discuss every potential pitfall in ML workflows, we believe that our current guidelines capture best practices for avoiding the most common and potentially dangerous errors in this important emerging field.

Machine Learning in Laboratory Medicine: Recommendations of the IFCC Working Group.

BACKGROUND. Variable response to chemotherapy in acute myeloid leukemia (AML) represents a major treatment challenge. Clinical and genetic features incompletely predict outcome. The value of clinical epigenetic assays for risk classification has not been extensively explored. We assess the prognostic implications of a clinical assay for multilocus DNA methylation on adult patients with de novo AML.

METHODS. We performed multilocus DNA methylation assessment using xMELP on samples and calculated a methylation statistic (M-score) for 166 patients from UPENN with de novo AML who received induction chemotherapy. The association of M-score with complete remission (CR) and overall survival (OS) was evaluated. The optimal M-score cut-point for identifying groups with differing survival was used to define a binary M-score classifier. This classifier was validated in an independent cohort of 383 patients from the Eastern Cooperative Oncology Group Trial 1900 (E1900; {"type":"clinical-trial","attrs":{"text":"NCT00049517","term_id":"NCT00049517"}}NCT00049517).

RESULTS. A higher mean M-score was associated with death and failure to achieve CR. Multivariable analysis confirmed that a higher M-score was associated with death (P = 0.011) and failure to achieve CR (P = 0.034). Median survival was 26.6 months versus 10.6 months for low and high M-score groups. The ability of the M-score to perform as a classifier was confirmed in patients ≤ 60 years with intermediate cytogenetics and patients who achieved CR, as well as in the E1900 validation cohort.

CONCLUSION. The M-score represents a valid binary prognostic classifier for patients with de novo AML. The xMELP assay and associated M-score can be used for prognosis and should be further investigated for clinical decision making in AML patients.

/pdf/a-clinical-measure-of-dna-methylation-predicts-outcome-in-de-530s64wfjw.pdf

A clinical measure of DNA methylation predicts outcome in de novo acute myeloid leukemia

patients signed informed consent following institutional guidelines and in accordance with the Declaration of Helsinki. Patients received induction chemotherapy with highdose cytarabine containing regimens (n5 132) or all trans retinoic acid (ATRA) and arsenic trioxide (ATO) for acute promyelocytic leukemia (APL) (n5 17) (Supporting Information Table II). The xMELP assay was performed on each sample and M-score was determined as previously described [4,5]. Standard molecular (FLT3-ITD and NPM1) and cytogenetic annotation was performed as per clinical care. Cytogenetic risk was classified according to Medical Research Council criteria [7]. In this cohort, 81% of patients achieved CR following induction and 57% were alive at 2 years (Supporting Information Table II). Median follow-up was 29.2 months (range 0.1–112.2 months). Mean and median M-score for the cohort were 91.8 (95% confidence interval [CI] 86.1–97.5) and 88.7 (range, 26.7–183.9), respectively, similar to the UPENN and ECOG cohorts. Consistent with previous results, mean M-score was lower in patients with core binding factor (CBF) favorable-risk leukemia (58.2, 95% CI 50.9–65.6) compared to patients with intermediate and adverse cytogenetics (100.1, 95% CI 92.4–107.9 and 118.7, 95% CI 107.3–130.0, respectively; P< 0.00001), and was not associated with sex, presenting WBC, or FLT3-ITD/NPM1 mutation profile (Supporting Information Table I). Mean M-score for patients with APL was also lower than patients with intermediate or unfavorable risk cytogenetics at 82.3 (95% CI 69.9–94.7), although not as low as the CBF subgroup. Notably, mean M-score was higher in patients >60 years compared to younger patients (106.7, 95% CI 96.0–117.4 vs. 85.4, 95% CI 78.9 vs. 91.8, Supporting Information Table I), and in patients with secondary AML compared to those with de novo AML (104.9, 95% CI 91.7–118.2 vs. 87.6, 95% CI 81.5–93.8, P5 0.01). We found that mean M-score for patients surviving at 2 years was significantly lower than for deceased patients (80.1; 95% CI 72.6–87.7 vs. 109.2; 95% CI 101.8–116.6, P< 0.00001, Figure 1A) and patients achieving CR had a lower mean M-score compared to those with primary refractory disease (85.3; 95% CI 79.2–91.4 vs. 118.7; 95% CI 108.4– 129.1, P< 0.00001, Figure 1B). A univariate Cox survival analysis demonstrated that a 10-unit increase in M-score was associated with a 10% increase in the hazard of death (HR 1.1; 95% CI 1.1–1.2; P< 0.0001, Supporting Information Table III) and a 40% increase in the odds of failing to achieve CR (OR 1.4; 95% CI 1.2–1.6; P< 0.0001). These associations are more robust than observed with the original cohorts. We segregated the AML cohort into subgroups based on a previously defined binary M-score cut-point (M-score5 86) and found a significant difference in survival for low and high M-score groups determined by this classifier (log-rank P< 0.00001). High Mscore was associated with an increased hazard of death (HR 2.8; 95% CI 1.7–4.5, P< 0.0001, Supporting Information Table III) with 2-year OS of 80% (95% CI 70–90%) compared to 39% (95% CI 28–50%) for low and high M-score groups, respectively (Supporting Information Table IV). CR rates for low and high M-score groups were 97% (95% CI 93–100%) and 67% (95% CI 57–77%), respectively. Subgroup analysis showed that for both de novo AML (n5 113) and secondary AML (n5 36), the M-score classifier significantly distinguishes prognostic categories (log-rank P5 0.001 and P5 0.02, respectively). Additionally, the M-score classifier is prognostic for OS in younger patients with intermediate cytogenetics (n5 45, log-rank P5 0.04) and for patients who achieve CR (n5 120, log-rank P5 0.004). We performed multivariable analysis of M-score and “standard” clinicopathologic prognostic factors (age, diagnostic WBC, cytogenetics, de novo versus secondary, and FLT3-ITD status). Since the UTMDACC cohort had a relatively large number of patients with APL or CBF AML with favorable OS, we restricted the analysis to younger (age <60) patients with intermediate or unfavorable risk leukemia. In this prognostically diverse subgroup, the M-score is marginally associated with OS on multivariable analysis (P5 0.054, Supporting Information Table V). We have previously demonstrated that multi-locus methylation assessment can be used to determine AML prognosis at diagnosis, and that it has several characteristics advantageous to clinical implementation including universal applicability, rapid turnaround, and ease of interpretation. The reproducibility of the xMELP assay and M-score for prognosis in this separate cohort of patients treated with intensified induction therapy strategies confirms the broad applicability of this assay. A novel finding in this study is that M-score can identify a subgroup of patients with secondary AML who experience a 2-yr OS >75% (Supporting Information Table IV). An important finding confirmed by this study is that M-score retains prognostic significance in AML patients achieving CR with induction chemotherapy, suggesting value for guiding individualization of postremission therapy in responding patients. Patients with high M-scores could be considered for referral for stem cell transplantation or investigational approaches for maintaining remission. Further prospective studies that incorporate M-score into current prognostic schemes are warranted, as well as studies to compare value to technically difficult flow-based minimal-residual disease assays. Extending this analysis to a larger number of patients is of great interest as these results would have immediate implications for prognostication and design of therapeutic regimens. COURTNEY D. DINARDO, MARLISE R. LUSKIN, MARTIN CARROLL, CATHERINE SMITH, JENNA HARRISON, SHERRY PIERCE, STEVE KORNBLAU, MARINA KONOPLEVA, TAPAN KADIA, HAGOP KANTARJIAN, GERALD B. WERTHEIM,* AND STEPHEN R. MASTER* Department of Leukemia, University of Texas MD Anderson Cancer Center, Houston, Texas; Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts; Perelman School of Medicine at the University of Pennsylvania, Philadelphia, Pennsylvania; Department of Pathology & Laboratory Medicine, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, Pennsylvania; Department of Pathology & Laboratory Medicine, Weill Cornell Medical College, New York, New York Additional Supporting Information may be found in the online version of this article. Courtney D. DiNardo, Marlise R. Luskin, Gerald B. Wertheim, and Stephen R. Master contributed equally to this manuscript. Conflict of interest: M.R. Luskin, M. Carroll, S.R. Master, and G.B. Wertheim have the provisional patent application 62/040,821. *Correspondence to: Stephen R. Master, Department of Pathology and Laboratory Medicine, Weill Cornell Medical College. 525 E. 68th St. Suite F-715, Box 79, New York, NY 10065. E-mail: srm9012@med.cornell.edu or Gerald Wertheim, Department of Pathology and Laboratory Medicine, Children’s Hospital of Philadelphia, 5140 Floor Main Building, 324 S. 34th St. Philadelphia, PA 19104. E-mail: wertheimg@email.chop.edu Received for publication: 18 September 2016; Revised: 24 October 2016; Accepted: 27 October 2016 Published online: 2 November 2016 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/ajh.24597

https://onlinelibrary.wiley.com/doi/pdf/10.1002/ajh.24599

Cardiotoxicity risk with bortezomib versus lenalidomide for treatment of multiple myeloma: A propensity matched study of 1,790 patients

Whole-exome sequencing enhances prognostic classification of myeloid malignancies

BACKGROUND: Epigenetic dysregulation involving alterations in DNA methylation is a hallmark of various types of cancer, including acute myeloid leukemia (AML). Although specific cancer types and clinical aggressiveness of tumors can be determined by DNA methylation status, the assessment of DNA methylation at multiple loci is not routinely performed in the clinical laboratory.

METHODS: We recently described a novel microsphere-based assay for multiplex evaluation of DNA methylation. In the current study, we validated and used an improved assay [termed expedited microsphere Hpa II small fragment Enrichment by Ligation-mediated PCR (xMELP)] that can be performed with appropriate clinical turnaround time.

RESULTS: Using the xMELP assay in conjunction with a new 17-locus random forest classifier that has been trained using 344 AML samples, we were able to segregate an independent cohort of 70 primary AML patients into methylation-determined subgroups with significantly distinct mortality risk ( P = 0.009). We also evaluated precision, QC parameters, and preanalytic variables of the xMELP assay and determined the sensitivity of the random forest classifier score to failure at 1 or more loci.

CONCLUSIONS: Our results demonstrate that xMELP performance is suitable for implementation in the clinical laboratory and predicts AML outcome in an independent patient cohort.

Validation of DNA Methylation to Predict Outcome in Acute Myeloid Leukemia by Use of xMELP

Handle everyday research tasks with reliable, citation-backed results

Your personal Research Agent to handle research tasks with citation-backed results

Popular Tasks used by Researchers

How can I help with your research?

Meet SciSpace

Get more enhanced response by uploading the PDFs you want me to reference.

No relevant PDFs in your library

SciSpace is the AI research assistant for academics. Run systematic literature reviews on 280M+ papers, and write papers with cited sources. Trusted by 1M+ students, PhDs & researchers.

SciSpace | AI for Research

Analyze PDFs

Code & Manuscripts

Funding & Grants

Literature & Patents

Medical & Clinical Data

Systematic Review

Visualize & Present

Web & Data

Build a Google Scholar-like website for your research.

Build a website

Create charts and images for your research

Create a Chart

Write a paper for submission to a journal

Draft a manuscript

Patent Search

Design eye-catching scientific posters in minutes.

Scientific Poster Generation

Systematic Literature Review

One task is running at the moment. Your messages will be shown right after.

Drag and drop or click here to browse

Loved by <highlight>1 million+</highlight> researchers

Extract a list of specific topics and their sources from unstructured text

Topics

Compare and analyze relevant papers that matches with your search

Papers

Get insights from PDFs and bookmarked papers from your library

My library

Recent searches

Try searching for:

Catch AI-generated content in scholarly and non-scholarly content

{ai} Detector

Ai Writer

Get PDF Summaries, highlighted text explanations 

Chat with PDF

Effortlessly create in-text citations and bibliographies in APA and 2,500 other formats

Citation generator

Get explanations, summaries, and answers on academic papers

Ease up your research workflow with {scispace}'s cohort of exciting AI tools

Elevate your academic writing skills and convey your ideas the way you want

Paraphraser

Explore our range of reading and writing tools

Your file is being prepared and should be ready in a few minutes. If it's a large file, it might take a bit longer. You can close this window, and we'll email you the file when it's done.

You have reached a maximum limit of <strong>{limit}</strong> columns in the table. Remove at least <strong>1</strong> column to add or create another one.

Alison Rager

Author Tools

Chat about Author

Papers

Validation of DNA Methylation to Predict Outcome in Acute Myeloid Leukemia by Use of xMELP