Dados do Resumo

Título

Nanopore RNA sequencing as an accurate and feasible tool for childhood acute lymphoid leukemia classification

Introdução

Childhood acute lymphoid leukemia (ALL) has achieved overall survival (OS) rates of over 80%. However, late and imprecise diagnoses in low and middle-income countries (LMICs) lead to significantly lower OS, as robust diagnosis requires advanced, yet costly techniques to classify genomic subtypes. Nanopore sequencing was shown to be accurate and cost-effective for ALL classification by researchers from St. Jude Children’s Research Hospital (SJCRH) and the University of North Carolina (UNC).

Objetivo

In partnership with SJCRH and UNC, this project aims to validate the accuracy of ALL classification through Nanopore sequencing in a pediatric Brazilian cohort, with further correlation between transcriptome data and clinical-pathological information.

Métodos

This retrospective multicenter cohort study was approved by the IRB (2746/2024). Clinicopathologic data will be collected from 118 pediatric patients diagnosed with ALL from 2001 to 2023. Furthermore, to perform the RNA sequencing, total RNA from buffy coat or bone marrow samples will be extracted from TRIzol and quantified by Nanodrop and Qubit. The libraries will be prepared with a PCR-cDNA Barcoding kit, and sequencing will be conducted using MIN106D flow cells with the MinION Mk1b device. The basecalling and demultiplexing will be performed on MinKNOW and Dorado (v0.4.1), and alignment with ENSEMBL (v110) on Minimap2. A machine learning algorithm, standardized by researchers from SJCRH and UNC, will classify the samples with a gene expression profile input on a PLS-DA model with pairwise cross-validation and an SVM classifier. Additional data analysis, such as gene differential expression, will be performed on R studio.

Resultados

As previously reported, this algorithm achieved accurate classification with a prediction probability higher than 0,8 in 96,2% of the samples among acute leukemia lineages, and 94,1% among B-ALL genomic subtypes. Therefore, this level of accuracy is also expected in this cohort, as data from larger cohorts will enhance algorithm performance. Also, There may be evidence of some differential gene expression profile among our Brazilian cohort that may be significantly associated with clinical pathological data, disease progression, and outcomes.

Conclusões

Nanopore has been reported as a novel cost-effective approach for molecular analysis and evidence has been increasing on its potential as a feasible tool for childhood cancer classification, especially in acute leukemias. Once this hypothesis is corroborated in this cohort, this project may improve the standard of care and advance the availability of more precise early diagnosis in LMICs, potentially increasing OS rates.

Financiador do resumo

Barretos Cancer Hospital Educational and Research Program, CAPES, FAPESP, St. Jude Children's Research Hospital, University of North Carolina

Palavras Chave

Childhood Acute Lymphoid Leukemia; RNA Sequencing; Nanopore Sequencing