825P - Ultra-sensitive cfDNA analysis for minimally invasive measurable residual disease detection and profiling in multiple myeloma

Background

MRD assessment in bone marrow (BM-MRD) is a relevant prognostic factor in multiple myeloma (MM). However, MRD is an established test in MM clinical trials but not in routine practice. Less invasive methods to monitor MRD in peripheral blood (PB) could facilitate its implementation.

Methods

This study enrolled 243 patients monitored in PETHEMA/GEM clinical trials. In all, BM-MRD was processed using NGF. In 83 patients, PB-MRD was examined in cell-free DNA (cfDNA) using CloneSight, a highly sensitive (>10^-4) NGS method based on patient-specific multiplexed amplicon mini-panels covering somatic mutations identified at diagnosis. PB-MRD was then evaluated by BloodFlow, another high-sensitive method (10^-7) that combines immunomagnetic enrichment with NGF.

Results

CloneSight was performed in 83 patients, 9 were excluded due to lack of suitable somatic mutations for PB-MRD tracking. A total of 194 samples were studied, 12 (6%) were PB-MRD positive, from 9 of the 74 (12%) patients. In 166 samples, BM-MRD was simultaneously analyzed using NGF. The concordance between PB-MRD using CloneSight and BM-MRD by NGF was 67% (62% double negative and 5% double positive). The frequency of CloneSight-/NGF+ and CloneSight+/NGF- discordant assessments was 31% and 2% respectively. Of note, only 3 of the 74 (4%) MRD negative patients relapsed thus far. Upon analyzing PB-MRD with both methods described, another 4/74 patients were CloneSight-/BloodFlow+, of whom 1 progressed. Of note, the only case being CloneSight+/BloodFlow- relapsed. The negative predictive value (NPV) of PB-MRD using BloodFlow was about 80%. CloneSight also offered a high PPV of 100%. The landmark median PFS of patients with negative vs positive PB-MRD using CloneSight was NR vs 14 months, respectively (HR:11.7, P<.001).

Conclusions

CloneSight and BloodFlow are empowered to assess PB-MRD with high sensitivity based on the respective detection of cfDNA and circulating tumor cells (CTCs). Presence of mutations and CTCs was associated with a shorter PFS. The high PPV achieved indicates that identification of such events in the group of BM-MRD-positive patients could stratify those at immediate risk of relapse.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.