Identification of Clonal Igh Gene Rearrangements By High-Throughput Sequencing of Cell Free DNA in Multiple Myeloma Patients

Background:

The achievement of complete remission is an important therapeutic goal and prognostic factor for overall survival in patients with B-cell malignancies. Molecular monitoring of disease with PCR-based strategies is used to assess the depth of treatment response, detect minimal residual disease (MRD), and identify patients at increased risk of relapse. Immunoglobulin heavy chain (IGH) gene rearrangements are used as molecular marker in approximately 80% of lymphoma and multiple myeloma (MM) patients since they represent lineage-specific markers and the third complementarity determining region  (CDR3) is unique to each clone.  We and others have demonstrated that next generation sequencing (NGS) technologies provide the opportunity to identify and quantify clonotypes with consensus primers combining the benefits of high sensitivity and universal applicability thus overcoming ASO-PCR and RQ-PCR limitations. Very recently, NGS has been applied to characterize the tumor-specific VDJ recombination of the immunoglobulin genes in the serum of patients with diffuse large B-cell lymphoma and used as a novel non-invasive strategy to predict clinical disease recurrence after first-line treatment (Roschewski et al., Lancet Oncology 2015).

The present study was designed to assess whether the Ion Torrent Personal Genome Machine (IT-PGM)-based sequencing and analysis we established (Gimondi et al., ASH 2014), could be used to detect circulating tumor DNA encoding the clonal immunoglobulin gene sequence in the plasma of patients with Multiple Myeloma.

Methods:

Genomic DNA (gDNA) was extracted from CD138⁺ plasma cells immunomagnetically selected from the bone marrow blood of six MM patients and amplified using seven different family-specific IgH-V primers and a consensus JH primer (Voena et al., Leukemia 1997). IgH clonality was assessed by Sanger sequencing in order to define the patient specific DNA rearrangement. Plasma samples of the six MM patients were collected and cell free DNA (cfDNA) extracted (Qiagen). The total amount of cfDNA was estimated by fluorometric measurement  (median 105 ng, range 37-171 ng). For library construction and NGS, paired samples of gDNA (500ng) and cfDNA (at least 37ng)  were amplified as previously described (Gimondi et al., ASH 2014). PCR products were evaluated for quality and length by high-sensitivity dsDNA chips (Agilent). PCR amplicons were barcoded, pooled and sequenced on the Ion Torrent PGM. NGS data were analyzed by using the IMGT-High V-Quest web server tool and the statistical software R.

Results:

Rearranged IGHV-D-J loci from cfDNA of each MM patient could be amplified using 7 different IGHV family primers and a consensus JH primer, despite the limited abundance of DNA recovered from plasma samples. PCR products were sequenced on an IT-PGM 316 chip, yielding at least 110K reads per sample (mean 280K reads) with an average coverage of 130x (at least 50x). Clonal IgH sequences were quantified as a fraction of the complete IGHV-D-J rearranged reads. The clonality of our samples was assessed by determining the percentage of reads identical to the most abundant CDR3 sequence in each sample. The number of clonal sequences corresponding to the highest expressed IGH clonotype in gDNA was consistent with those identified in cfDNA samples (range 74%-82%  and range 65%-73% respectively). Furthermore, the clonotypes identified by high-throughput sequencing of gDNA and cfDNA samples demonstrated a 100% sequence identity with the patient-specific IGHV-D-J clonal rearrangement identified by Sanger sequencing.

Conclusions:

We demonstrate that next generation sequencing of cfDNA from the plasma of Multiple Myeloma patients is feasible, accurate, and sensitive in identifying the tumor-derived VDJ recombination of the immunoglobulin genes without prior knowledge of the tumor clonotype and might represent an alternative when bone marrow biopsies are unavailable. Moreover, given the patchy bone infiltration of malignant plasma cells, cfDNA analysis is a non-invasive approach that might give a more precise quantification of disease burden. In addition, NGS analysis of the IGHV-D-J rearranged sequences provides a deep and detailed characterization of the patient immune repertoire, thus possibly allowing clonal evolution evaluations and monitoring of MRD in follow-up samples.