INTERNATIONAL MULTI-LABORATORY NEXT-GENERATION SEQUENCING FOR MRD ANALYSIS IN ALL. A PILOT STUDY BY THE EUROCLONALITY-NGS CONSORTIUM

Background:

Amplicon-based next generation sequencing (NGS) of immunoglobuline (IG) and T-cell receptor (TR) gene rearrangements to quantify minimal residual disease (MRD) in lymphoid neoplasms is already the focus of intense research, development, and application. This is mainly due to its potential to overcome limitations of the current molecular gold standard, real-time quantitative (RQ)-PCR, as it avoids laborious design and testing of patient specific assays and allows quantification of MRD with a more specific readout than RQ-PCR. However, standardization, quality control and validation of the technology in a multicentre, scientifically controlled and independent setting is highly warranted, but currently lacking.

Aims:

To investigate an optimized IG/TR based NGS approach (including quality controls and calibrators) for sensitive and quantitative MRD assessment in lymphoid neoplasms within the EuroClonality NGS Consortium.

Methods:

Within a European Multicentre trial a newly designed quality controlled one-step IGH NGS approach was tested in seven institutes (Kiel, Salamanca, Milano, Bristol, London, Prague, and Torino). Serial dilutions of diagnostic DNA and follow-up samples of thirty B-cell precursor ALLs with known complete IGH rearrangements as well as a set of positive and negative controls were analysed. All samples were spiked with pre-defined copy numbers of five reference IGH sequences. One-step PCR was used for library preparation with sets of FR2 primers harbouring sample specific barcodes and 500ng DNA per sample (75.000 copies), subsequently being sequenced with the Illumina V2 kit on the Illumina MiSeq NGS platform. Purpose-built bioinformatics methods were used to analyse data. MRD results of ALL follow-ups were compared to results of EuroMRD based RQ-PCR.

Results:

A total of 312 samples were sequenced within 24 deep sequencing runs producing 168 million reads. Regression analysis of reference IGH allele reads was used to calculate the read coverage per copy, and correct for different B-cell content per sample. Ratios of reads between pairs of reference IGH alleles were highly consistent in intra-and inter-centre analyses, independent of the total number of reads in the sample. This high consistency in the usage of the reference IGH alleles was confirmed through redundancy analysis based on normalized reads (adjusted p-value=0.24). When outliers were identified, indicating poor assay performance, those were excluded from further analysis. The IGH gene rearrangements of all 27 pooled positive B-cell line controls were identified in all centres with a high intra-and inter-centre consistency. NGS based MRD analysis in 116 ALL follow-up samples revealed MRD positivity in 69/116 samples, and RQ-PCR in 66/116 samples, with a good overall concordance between both approaches (R2=0.81). Discrepancies mainly affected samples with low MRD levels being consistent with Poisson sampling.

Summary and Conclusions:

The approach newly developed by the Euro- Clonality NGS Consortium shows great promise to allow for quality controlled NGS based MRD analysis in a multicentre setting with a high intra-and interlaboratory reproducibility. The critical next step, as part of the Consortium´s strategy, involves validation of this technology, prior to its implementation into routine practice, within multicentre clinical trials, in a scientifically controlled, independent, and multidisciplinary setting.

online 0