Cell-Free DNA for Minimal Residual Disease Monitoring in Multiple Myeloma Patients

Background

Circulating nucleic acids, such as cell-free DNA (cf-DNA), are becoming a promising minimally-invasive diagnostic tool for cancer detection. Recent studies demonstrated that tumor-derived cf-DNA can be used to monitor tumor burden and response to treatment in patients (pts) with solid tumors as well as hematological malignancies (Dawson et al, 2013, Armand et al, 2013). In this study we investigated the clinical utility of cf-DNA in the monitoring of minimal residual disease (MRD) of pts with multiple myeloma (MM) carrying the tumor specific immunoglobulin (IGH) rearrangement.

Methods

Cf-DNA was extracted from 1 ml of serum sample from 13 MM patients enrolled in Italian CRD/MEL-200 and EMN-02 protocols. The total amount of cf-DNA was estimated by fluorometric measurement (median 560 ng, range 15-5158 ng) and the length of fragments was evaluated by high sensitivity dsDNA chips (Agilent). Patient specific clonal IGH rearrangement was identified at the time of diagnosis from bone marrow (BM) genomic DNA (gDNA) as previously reported (Ladetto et al, 2000). For each patient, MRD in BM and peripheral blood (PB) was estimated by real time quantitative PCR (qPCR) using ASO-specific primers and the quantification was based on serial 10-fold dilution standard curves from plasmid carrying the patient specific IGH rearrangement. The amount of IGH rearrangement in cf-DNA (cf-IGH) was estimated by qPCR and droplet digital PCR (ddPCR) (Bio-Rad) on diagnostic and follow up samples and was expressed as the amount of copies per 1 µg of total cf-DNA. qPCR and ddPCR results were interpreted according to the Euro-MRD guidelines (van der Velden et al, 2003).

Results

Overall, 54 cf-DNA samples from MM serum (13 diagnostic, 41 follow-up samples) were analyzed for the presence of patient specific IGH rearrangement. The most abundant fraction of cf-DNA was 180-220bp, than 350-400bp and 700-10000bp (in 100%, 85% and 68% of samples respectively), whereas longer fragments more often appeared in follow-up samples. By qPCR, cf-IGH at diagnosis were observed in 11/13 diagnostic samples. Only 3/13 pts were quantifiable (116, 85, 187 copies/1 µg of cfDNA) and 8/13 pts were positive but not quantifiable (PNQ) cf-IGH. By ddPCR, levels of cf-IGH at diagnosis were observed in 9/13 pts. 6/13 pts were quantifiable (246, 195, 96, 88, 184, 25 copies/1µg of cfDNA), and only 3/13 pts were PNQ. In follow-up samples, levels of cf-IGH were undetectable by qRT-PCR; however in 5 samples they were PNQ by ddPCR. Interestingly, in one available relapse sample, cf-IGH reappeared again to quantifiable level (61 copies by qRT-PCR and 190 copies by ddPCR). The levels of cf-IGH are quantifiable in samples with higher amount of tumor specific IGH rearrangements in BM or PB; however, no association was observed between cf-IGH level at diagnosis and disease burden estimated by the PCs infiltration in BM or the monoclonal immunoglobulin concentration in blood/urine.

Conclusions

These data show the potential utility of cf-IGH monitoring in MM pts. Although by qPCR, cf-IGH were detected in 11/13 pts, they were quantifiable only in 3/13 pts and ddPCR was more precise as it was able to quantify cf-IGH in 6/13 pts. Since cf-IGH copies were quantifiable only in diagnostic samples and in 1 available sample at the relapse, we conclude that higher amounts of serum are necessary to overcome the limitation of assay sensitivity. Potential advantages and predictive value, for monitoring tumor marker in a non-invasive manner, need to be further validated on larger cohort of samples using increased amount of cf-DNA. Work was supported by IGA grants NT12130, NT14575. This work is funded by a Black Swan Research Initiative grant by the International Myeloma Foundation "Dynamics of microRNA and cell-free DNA profiles during multiple myeloma progression“.