Single Tube Molecular Profiling of ALL Samples Reveals Integrated Molecular Signature and Highlights Gene Regulatory Networks

Background:

Over 20 acute lymphoblastic leukemia (ALL) subtypes are defined by molecular signature, which is historically mainly dependent on genomic rearrangements (translocations and gene fusions). Advancements in NGS strategies allow high-resolution profiling, taking in to account also gene expression patterns, single nucleotide variants (SNVs) and insertions and deletions (INDELs).

Previous studies showed that molecular signature correlates to prognosis and treatment compatibility, making accrue molecular profiling a major interest point for both clinical and research communities, helping identify novel therapeutic targets and designing clinical management approaches.

Many current NGS approaches rely on panels-based assays, which screen DNA or RNA to provide one piece of the puzzle, resulting in higher cost and processing time and requiring larger initial inputs. We report a single tube NGS panel that uses total nucleic acid as input for simultaneous screening of DNA and RNA, allowing for simultaneous profiling of gene expression, fusions and variants, establishing a work frame for comprehensive molecular profiling within a single assay.

Methods:

This study used 102 ALL clinical samples, 85 bone marrow (BM) and 17 peripheral blood (PB). Libraries were prepared using a custom Qiagen Multimodal NGS panel targeting 302 DNA targets and 229 RNA fusions. Enriched amplicon libraries were sequenced with unique dual indices on Illumina NovaSeq 6000. Data were analyzed on our in-house bioinformatic pipeline. Expression values were normalized to GUSB. Previous in-house studies were used to establish baseline cut-offs for upregulation of 31 genes known to be associated with ALL. Variants enrichment for genes was done by normalizing the number of variants to the size of the gene and genes with elevated variant loads were identified, to highlight molecular pathways within our cohort.

Results:

Since expression can be modulated by both genomic variants and rearrangements, we performed a principal component analysis (PCA) on our RNA expression data to identify the clustering effect within the samples that could be driven by either or both of these genetic factors. The PCA identified several distinct clusters, with PC1 accounting for 39.3% of the variance, and PC2 accounting for 15.3%. The biological source (BM vs. PB) did not have an impact on the clusters, allowing us to analyze both sample types together. We then investigated whether the predetermined 31 over-expressed genes could be driving the clusters. Three clear groups were observed based on over-expression quantification. One group with low overexpression levels (0-1 genes), the second with medium levels (8-11), and one with high-level of over expression (>16 genes). These clusters also corresponded to T-ALL (high variance)-vs B-ALL (low to medium variance) classification, and showed clear segregation of samples with over-expression of EPOR (high variance cluster) from the rest of the samples.

RNA expression data also showed a high correlation between expression of PAX5, RUNX1 and TCF3; all are known to interact as part of a gene regulatory network, highlighting its importance to ALL development.

Finally, we investigated SNVs' role within the different clusters. While we could not identify any single variant/gene to be especially enriched with any of the clusters, we observed that KMT2D has an enrichment in variants within our cohort, underscoring that epigenetics is an important component of the disease landscape.

Conclusions:

We demonstrate the use of a single tube multimodal NGS assay for comprehensive genomic profiling. Our analysis shows that this powerful and cost-effective tool can help identify molecular signatures, classify ALL sub-types, identify regulatory networks and molecular mechanisms, which will help not only properly diagnose but also identify and develop therapeutic targets clinical management strategies.

Zelinger:NeoGenomics Laboratories, Inc.: Current Employment. Koo:NeoGenomics Laboratories, Inc.: Current Employment. Alarcon:NeoGenomics Laboratories, Inc.: Current Employment. Ko:NeoGenomics Laboratories, Inc.: Current Employment. Thomas:NeoGenomics Laboratories, Inc.: Current Employment. Jung:NeoGenomics Laboratories, Inc.: Current Employment. Wang:NeoGenomics Laboratories, Inc.: Current Employment. Ye:NeoGenomics Laboratories, Inc.: Current Employment. Kulkarni:NeoGenomics Laboratories, Inc.: Current Employment.