Whole-transcriptome analysis in acute lymphoblastic leukemia: a report from the DFCI ALL Consortium Protocol 16-001

Abstract

The molecular hallmark of childhood ALL is characterized by recurrent, prognostic genetic alterations, many of which are cryptic by conventional cytogenetics. RNA-seq is a powerful next-generation sequencing technology with the ability to simultaneously identify cryptic gene rearrangements, sequence mutations and gene expression profiles in a single assay. We examined the feasibility and utility of incorporating RNA-seq into a prospective multi-center phase 3 clinical trial for children with newly-diagnosed ALL. Patients enrolled on the DFCI ALL Consortium Protocol 16-001 who consented to optional studies and had available material underwent RNA-seq. RNA-seq was performed in 173 ALL patients. RNA-seq identified at least one alteration in 157 (91%) patients. Fusion detection was 100% concordant with results obtained from conventional cytogenetic analyses. An additional 56 gene fusions were identified by RNA-seq, many of which confer prognostic or therapeutic significance. Gene expression profiling enabled further molecular classification into the following B-ALL subgroups: High hyperdiploid (n=36), ETV6-RUNX1/-like (n=31), TCF3-PBX1 (n=7), KMT2A-rearranged (n=5), iAMP21 (n=1), hypodiploid (n=1), BCR-ABL1/-like (n=16), DUX4-rearranged (n=11), PAX5 alterations (n=11), PAX5 P80R (n=1), ZNF384-rearranged (n=4), NUTM1-rearranged (n=1), MEF2D-rearranged (n=1) and Others (n=10). RNA-seq identified 141 nonsynonymous mutations in 93 (54%) patients; the most frequent were RAS-MAPK pathway mutations. Among 79 patients with both low-density array and RNA-seq data for the Ph-like gene signature prediction, results were concordant in 74 (94%) patients. In conclusion, RNA-seq identified several clinically-relevant genetic alterations not detected by conventional methods, supporting the integration of this technology in frontline pediatric ALL trials.

Conjoined Genes as Common Events in Childhood Acute Lymphoblastic Leukemia

“…NUP214::ABL1 was originally reported as a recurrent abnormality in T-ALL, accounting for 6% of adult and less than 2% of pediatric T-ALL, as recently reported by the Associazione Italiana di Onco-Ematologia Pediatrica [ 21 ]. Otherwise, only few cases have been reported in B-ALL [ 22 , 23 , 24 ]. Moreover, the fusion involving MAEA::CTBP1 (detected in one SR B-ALL case from our cohort), although with different breakpoints, has already been described in colon adenocarcinoma and acute myeloid leukemia [ 25 ].…”

Real-time molecular classification of leukemias

“…The training set used is described by Tran et al (3) and contains 1,134 ALL RNA-seq samples generated with diverse short-read sequencing protocols and platforms. These include both in-house (n=72) and public datasets (n=1,062) (5, 6) that are distributed amongst 17 well-documented ALL subtypes.…”

“…These include both in-house (n=72) and public datasets (n=1,062) (5, 6) that are distributed amongst 17 well-documented ALL subtypes. The data were corrected for batch effects using a Surrogate Variable Analysis (SVA) package (7) and the 500 most differentially expressed genes obtained from the DESeq2 likelihood ratio test (LRT) (8) , which were used to train a single layer neural network classifier in Tran et al (3) . As some samples in the test set might also be present in the training set for the in-house samples, we ensured that the latter were removed from the training set when spawning the neural network to avoid overfitting.…”

“…The advent of next-generation sequencing (NGS), specifically RNA-sequencing (RNA-seq), has expanded ALL taxonomy from 7 conventional to 19 molecular ALL subtypes, which are defined by gene alterations and/or distinct gene expression signatures; features that are either cryptic or not detected by conventional diagnostic pipelines (3) . Therefore, several groups are now incorporating RNA-seq into clinical workflows to refine the molecular classification and risk stratification of ALL (3) . However, NGS is currently hampered by a relatively long turnaround time for decision-making (typically 3 to 4 weeks), remains a costly endeavor and is only available in limited institutions.…”

Increasing Complexity of Molecular Landscapes in Human Hematopoietic Stem and Progenitor Cells during Development and Aging

“…Transcriptomic and next-generation sequencing technologies, together with cytogenetic analyses, have contributed to the 2016 and current updates in progress of the World Health Organization classification of hematological malignancies and serve to highlight the heterogeneity of pediatric acute leukemias [ 126 , 127 , 128 , 129 , 130 ]. High throughput sequencing assays have also been used to identify clinically relevant and novel fusion genes and mutations that are not detected by conventional cytogenetics [ 128 , 129 , 131 , 132 , 133 , 134 , 135 ]. More than 30 ALL subtypes have now been identified, with two-thirds being B ALL [ 130 ].…”