Targeted-Capture Sequencing Approach to V(D)J Rearrangement, Immunoglobulin (Ig) Translocation and Mutation Detection in Multiple Myeloma (MM): Potential Applications in Minimal Residual Disease (MRD) Analysis and Comparison to Multiparameter Flow Cytometry (MFC) in the Myeloma Canada Research Network (MCRN) 001 Study with Intravenous Busulfan and Melphalan (BuMel) Conditioning Followed By Lenalidomide Maintenance in Newly Diagnosed Multiple Myeloma (NDMM)

Background

Advances in MM treatment have resulted in increased rates of deeper responses, however most patients (pts) relapse, reflecting persistent disease. More sensitive techniques are required to accurately evaluate depth of response and guide therapeutic decisions.

Ig V(D)J gene rearrangements occur early in B-cell development, such that plasma cell (PC) dyscrasias are typified by a unique molecular V(D)J signature. MM is additionally characterized by translocations into Ig loci and recurrent mutations, which are tumor specific and can be exploited for MRD assessment.

Current techniques for MRD assessment include MFC, which involves the simultaneous detection of multiple (≥8) abnormal PC surface antigens and NGS, predominantly with LymphoSIGHT™. IMWG 2016 guidelines incorporate MRD assessment to a sensitivity of 1/10⁵ nucleated cells by either technique. We have developed an alternative target-capture based NGS approach for detection of mutations, Ig gene (IGH, IGK and IGL) translocations and Ig loci V(D)J rearrangements, which can also be utilized for MRD analysis.

We validated this assay on 68 human myeloma cell lines (HMCLs) and established a reliable limit of translocation detection of at least 1/10³ with the potential to reach 1/10⁵ with additional depth of coverage. This assay was then applied to both cfDNA and CD138+ sorted bone marrow (BM) samples from selected pts on the MCRN001 trial and compared to 8-color MFC (sensitivity 1/10⁴ cells) and standard MM response assessment.

Methods

DNA libraries were prepared with KAPA HyperPrep® for Illumina TruSeq in conjunction with NEXTflex-96 DNA Barcodes (Bio Scientific) for HMCL or custom molecular adapters and index primers for molecularly barcoded libraries for MCRN patient samples. Capture probes (xGen® Lockdown®, Integrated DNA Technologies) were designed to pull down Ig V, J and C genes from IGH, IGL and IGK, known Ig translocation hotspots and 38 recurrently mutated genes (2638 probes, 316.5kb). All captured libraries were sequenced on Illumina NextSeq500.

The MCRN-001 trial enrolled 77 NDMM pts who received bortezomib-based induction and BuMel conditioning (Busulfan 3.2mg/kg², Melphalan 140mg/m²) followed by ASCT on day 0 and lenalidomide maintenance (10mg/d) starting on day 100 (ASH 2016, abstract 4632). For MRD analysis, BM was collected for 8-color MFC and plasma samples for cfDNA at day 100, every 3 months for the first year and every 6 months thereafter. BM samples were collected for genomic studies at day 100, 6 and 12 months after ASCT and incorporated into MRD analysis.

Results

Targeted capture sequencing detected t(4;14), t(11;14), and t(14;16) described in 18/18, 11/13 and 10/11 HMCLs respectively (Table 1). Candidate V(D)J rearrangements in HMCLs were confirmed by whole genome sequencing in 3 cases with additional validation underway. Dilution of HMCL into normal DNA established a reliable limit of detection of at least 1/10³, with the potential to reach 1/10⁵ with additional sequencing.

All 77 pts enrolled onto the MCRN-001 study were evaluable for MRD by MFC. 35/77 (45%) pts achieved a VGPR or better after induction, increasing to 60/69 (87%) at 12 months. MRD negativity by MFC was 16/77 (21%) and 27/69 (39%) respectively.

11 pre-induction (PI) samples, 15 follow up BM samples from 9 pts and 47 cfDNA samples from 20 pts at various timepoints have been collected. Of 11 PI samples, 8 pts had 13 follow up BM samples available for MRD analysis, one of which failed QC. cfDNA analysis is ongoing and will be presented.

NGS translocation analysis detected t(11;14) in 3 follow up patient samples, all of which were MRD positive by MFC and had t(11;14) by cytoFISH at diagnosis. In one case of MRD positivity by MFC, t(11;14) was not detected (Pt#2 month 12). V(D)J gene rearrangement analysis detected MRD in 5/13 follow up samples, and candidate mutations were identified in 4 BM samples, all of which were MRD positive by MFC. In total, 8/13 analyzed samples were MRD positive by NGS by ≥ 1 analytic method, with complete agreement to MFC in 14/14 cases (Table 2).

Conclusions

Targeted-capture NGS is a feasible, efficient technique for analysis of VDJ gene rearrangement, translocations and targeted mutation detection in MM and can be utilized for MRD assessment. In this small series, MRD assessments by NGS analysis correlates closely 8-color MFC. Further follow up is required to assess the impact of MRD status by these measurements on long term outcomes.

View largeDownload slide

View largeDownload slide

Close modal