RESUMEN
BACKGROUND: Next-generation sequencing has had a significant impact on genetic disease diagnosis, but the interpretation of the vast amount of genomic data it generates can be challenging. To address this, the American College of Medical Genetics and Genomics and the Association for Molecular Pathology have established guidelines for standardized variant interpretation. In this manuscript, we present the updated Hospital Israelita Albert Einstein Standards for Constitutional Sequence Variants Classification, incorporating modifications from leading genetics societies and the ClinGen initiative. RESULTS: First, we standardized the scientific publications, documents, and other reliable sources for this document to ensure an evidence-based approach. Next, we defined the databases that would provide variant information for the classification process, established the terminology for molecular findings, set standards for disease-gene associations, and determined the nomenclature for classification criteria. Subsequently, we defined the general rules for variant classification and the Bayesian statistical reasoning principles to enhance this process. We also defined bioinformatics standards for automated classification. Our workgroup adhered to gene-specific rules and workflows curated by the ClinGen Variant Curation Expert Panels whenever available. Additionally, a distinct set of specifications for criteria modulation was created for cancer genes, recognizing their unique characteristics. CONCLUSIONS: The development of an internal consensus and standards for constitutional sequence variant classification, specifically adapted to the Brazilian population, further contributes to the continuous refinement of variant classification practices. The aim of these efforts from the workgroup is to enhance the reliability and uniformity of variant classification.
Asunto(s)
Pruebas Genéticas , Variación Genética , Humanos , Estados Unidos , Mutación , Reproducibilidad de los Resultados , Teorema de Bayes , Genoma HumanoRESUMEN
BACKGROUND: Acute myeloid leukemia (AML) is a myeloid neoplasm associated with a high morbidity and mortality. The diagnosis, risk stratification and therapy selection in AML have changed substantially in the last decade with the progressive incorporation of clinically relevant molecular markers. METHODS: In this work, our aim was to describe a real-world genomic profiling experience in AML and to demonstrate the impact of the European Leukemia Net 2022 update on risk stratification in AML. RESULTS AND DISCUSSION: One hundred and forty-one patients were evaluated with an amplicon-based multi-gene next-generation sequencing (NGS) panel. The most commonly mutated genes were FLT3, DNMT3A, RUNX1, IDH2, NPM1, ASXL1, SRSF2, NRAS, TP53 and TET2. Detection of FLT3 ITD with NGS had a sensitivity of 96.3% when compared to capillary electrophoresis. According to ELN 2017, 26.6%, 20.1%, and 53.3% of patients were classified as having a good, moderate, or unfavorable risk. When ELN 2022 was used, 15.6%, 27.8%, and 56.6% of patients were classified as favorable, moderate, or unfavorable risk, respectively. When ELN 2022 was compared to ELN 2017, thirteen patients (14.4%) exhibited a different risk classification, with a significant decrease in the number of favorable risk patients, what has immediate clinical impact. CONCLUSIONS: In conclusion, we have described a real-world genomic profiling experience in AML and the impact of the 2022 ELN update on risk stratification.