Target capture next-generation sequencing (NGS) for use in molecular-based research of myeloid measurable residual disease (MRD)

Want to download this as a PDF? Download now

Introduction

Improved methods of disease status monitoring with detection of low-level variants are essential for early therapeutic interventions and to potentially prevent disease recurrence.

At present, two different types of methods are used for detection of measurable residual disease (MRD): immunophenotypic based on multiparameter flow cytometry (MFC), and molecular methods which include real-time quantitative polymerase chain reaction (RQ-PCR), digital droplet PCR (ddPCR) or next-generation sequencing (NGS). Each of these methods differ in their applicability, specificity and sensitivity to detect MRD.

NGS provides a solution for evaluation of multiple genes in a single assay. Together with significant reduction in sequencing cost and improved accuracy NGS can now be used in monitoring.

In this study we aimed to evaluate OGT’s SureSeq™ Myeloid MRD Complete NGS Workflow Solution V2 for suitability with deep sequencing and rare variant detection required for investigation of MRD monitoring.

Methods

Workflow

Libraries were generated using OGT’s Universal NGS Workflow Solution V2 (Figure 1). The workflow is ideally suited for low frequency variant detection through incorporation of Unique Dual Indexes (UDIs) and Unique Molecular Identifiers (UMIs).

Figure 1: DNA to sequencer in 1.5 days with minimal handling time.

Panel

The SureSeq Myeloid MRD Panel has been designed in collaboration with leading cancer experts to incorporate 13 key genes relevant to AML research. Utilizing OGT’s intelligent panel design capabilities, the SureSeq Myeloid MRD Panel in combination with OGT’s Universal NGS Workflow Solution V2 offers accurate detection of a range of variants: single nucleotide variants (SNVs), insertion-deletions (indels) and internal tandem duplications (ITDs) in FLT3 down to 0.05% Variant Allele Frequency (VAF).

Table 1: The SureSeq Myeloid MRD Panel targets SNVs, indels and FLT3- ITDs in 46 hotspot exons across 13 genes relevant to AML research.

Sequencing

Sequencing was conducted using 2 x 150 bp reads on an Illumina NextSeq 550 High output^® V2 300.

Samples

The workflow and panel capabilities were demonstrated by using 36 research samples orthogonally validated and shown to carry variants within panel targeted regions. Twenty-nine samples had at least one variant and seven had no variants. In addition, the workflow and the panel were tested using Myeloid Reference DNA Standard (Horizon Discovery), with 6 SNVs, 2 indels and a 300 bp FLT3-ITD. DNA samples were diluted to create a range of frequencies down to 0.05% variant allele frequency (VAF).

Bioinformatic analysis

Sequencing data analysis was performed using OGT’s proprietary cloud-based Interpret NGS Analysis Software which includes a specific MRD hotspot analysis pipeline. This pipeline has been fine-tuned to the SureSeq Myeloid MRD Complete NGS Workflow Solution V2 to achieve optimal sensitivity and specificity by adding extra modules to OGT’s standard somatic pipeline. These additions include: improved base error correction (UMI processing), UMI-based QC metrics, global and local error models for SNV calling aimed at reducing false-positives, improvements to our ITD detection algorithm, and sample specific monitoring plots in the user interface.

Results I

Highly uniform UMI coverage across targeted regions

High depth and uniformity of coverage was achieved for all targeted genes and genomic regions allowing for confident detection of low frequency gene specific variants.

Uniform coverage is essential to enable reliable variant detection across all targets. High uniformity is demonstrated (Figure 2) – including difficult to target: NPM1 ex 11.

Figure 2: IGV plot showing coverage profile of target regions in the Myeloid MRD panel.

Results II

Detection of SNVs and indels in the critical regions

Data presented here are from 29 research samples that were processed using the SureSeq Myeloid MRD Complete NGS Workflow Solution V2 in combination with OGT’s Interpret NGS Analysis Software.

Table 2 lists the range of SNVs detected in the 29 research samples. These include SNVs in key genes CSF3R, FLT3, IDH1, IDH2, JAK2, KIT, RUNX1, SF3B1, TP53 that range from 0.85 - 0.039% VAF (Table 2).

Table 2: SNV detection: Data generated using the SureSeq Myeloid MRD Panel in combination with the OGT’s Universal NGS Workflow Solution V2 and OGT’s Interpret NGS Analysis Software. Rank and Pval relate to local and global background error.

Table 3 lists the range of Indels detected in the 29 research samples. These include 6 NPM1 type A insertion and other genes TP53, CEBPA, RUNX1 (Table 3).

Table 3: Indel detection: Data generated using the SureSeq Myeloid MRD Panel in combination with the OGT’s Universal NGS Workflow Solution V2 and OGT’s Interpret NGS Analysis Software.

Results III

Detection of SNVs, indels and FLT3-ITD using Myeloid Reference DNA Standard

SureSeq Myeloid MRD Complete NGS Workflow Solution V2 in combination with OGT’s Interpret NGS Analysis Software confidently detected all anticipated variants within the Myeloid Reference standard including an NPM1 insertion and a 300 bp FLT3-ITD.

Figure 3: Example of a SNV detection FLT3 c.2503 G>T (p.Asp835Tyr).

Table 4: Detected SNVs, indels and ITD in Myeloid Reference DNA Standard (Horizon Discovery).

*P-value and Rank metrics allow for improved variant filtering at sites with high background, as only variants with a P-value < 0.05 and a Rank > 95% will be considered.

Results IV

Detection of a FLT3 internal tandem duplications (FLT3-ITDs)

FLT3-ITDs are challenging to target, and subsequently detect due to their inherent repeat content and length (up to 300 bp). The unique detection algorithms incorporated into Interpret NGS Analysis Software enable accurate detection and quantification of FLT3-ITDs, including multiple and large ITDs (Figure 2).

Figure 4a: Panel A shows the output from OGT’s Interpret NGS Analysis Software for a low-frequency 57bp FLT3-ITD at 0.78%

Figure 4b: Panel B shows a subset of the supporting clipped reads.

Table 5: FLT3-ITD detection: Data generated using the SureSeq Myeloid MRD Panel in combination with the OGT’s Universal NGS Workflow Solution V2 and OGT’s Interpret NGS Analysis Software.

Conclusions

• High depth and uniformity of coverage was achieved for all targeted genes and genomic regions allowing for confident detection of low frequency variants
• We have demonstrated that the SureSeq Myeloid MRD Panel in combination with the OGT’s Universal NGS Workflow Solution V2 and OGT’s Interpret NGS Analysis Software can be reliably used to detect AML relevant gene specific variants SNVs, indels and FLT3-ITDs as low as 0.05% VAF
• Our approach provides researchers with the capability to use capture-based NGS technology to simultaneously characterize AML gene variants in MRD monitoring.