Plasma Cell-Free DNA Chromosomal Instability As Surrogate Biomarker of Treatment Response and Minimal Residual Diseases for Multiple Myeloma

Background: Minimal residual disease (MRD) is becoming standard diagnostic care for multiple myeloma. Here we investigate cell-free DNA chromosomal instability as minimal invasive biomarker for minimal residual disease monitoring.

Methods: 55 patients were recruited, including 47 newly diagnosed and 8 relapsed multiple myelomas. Plasma samples were collected before treatments, end of 2 cycle and 4 cycle of treatments. Treatment response was assessed by using IMWG criteria. Cell-free DNA was analyzed by illumine HiSeq X10, followed by chromosomal instability analyses by a customized bioinformatics workflow, ultrasensitive chromosomal aneuploidy detector (UCAD), and cfDNA CIN responding to treatment was summarized as cfDNA MRD INDEX.

Results: In this cohort study, 53 (96.3%) patients were found with treatment responses after 4 cycle of treatments, including 16 (29.1%) complete, 13 (23.6%) very good partial, 17 (30.9%) partial and 6 (10.9%) marginal responses. The other 2 (3.64%) recurrent MM experiences disease progression.

At baseline, plasma cfDNA chromosomal aberrations were found in 14/17(82.3%) multiple myeloma patients. Less patients were found with detectable chromosomal changes after treatments (58.8% after C2, 41.2% after C4, versus 82.3% before treatments, P<0.01). The treatment responses identified in plasma cfDNA were summarized as cfDNA MRD INDEX. Low cfDNA MRD INDEX predicts better treatment responses (Fisher exact test, Odds ratio=7.3, P=4.69e-05). And it predicts complete response with sensitivity 100% and specificity 83.3%, with cutoff value 0.044.

Furthermore, cfDNA CIN MRD index were found linearly correlated with percentage of malignant plasma cells from bone marrow aspiration as examined by flow cytometry assay (R-square=0.883, P<0.01). Cutoff -0.06 predicts 100% MRD negatives at specificity 100%.

Conclusions: Plasma cfDNA CIN might be used for monitoring multiple myeloma patients treatment responses, especially for predicting minimal residual diseases.