AUTHORS
Jessica M. Kent, N/A, Yuchen Jin2, Binod Dhakal, MD, MS3, Oren Pasvolsky, MD4, Mahmoud R. Gaballa, MD5, Hector Garcia Pleitez, MD6, Sarvari V. Yellapragada, MD7, Krina Patel Jr, MD, MSc8, Muzaffar Qazilbash, MD9, Anita D'Souza, MD10, Aniko Szabo, PhD11, Carolina D Schinke, MD12, Meera Mohan, MD, MS13
AFFILIATIONS
1 Medical College of Wisconsin, Milwaukee, Wisconsin, United States
2 Medical College of Wisconsin, Milwaukee, Wisconsin, United States
3 Medical College of Wisconsin, Milwaukee, Wisconsin, United States
4 UT MD Anderson Cancer, Houston, Texas, United States
6 UT MD Anderson Cancer, Houston, Texas, United States
7 UT MD Anderson Cancer, Houston, Texas, United States
8 UT MD Anderson Cancer, Houston, Texas, United States
9 UT MD Anderson Cancer, Houston, Texas, United States
10 Medical College of Wisconsin, Milwaukee, Wisconsin, United States
11 Medical College of Wisconsin, Milwaukee, Wisconsin, United States
12 University of Arkansas for Medical Sciences, Little Rock, Arkansas, United States
13 UT MD Anderson Cancer, Houston, Texas, United States
ABSTRACT
Background: Bone marrow (BM) minimal residual disease (MRD) testing is an established endpoint in multiple myeloma (MM) clinical trials to measure disease response. The International Myeloma Working Group (IMWG) recognizes MRD negativity at a sensitivity of at least 10-5. Current MRD assays achieve sensitivities ranging from 10-4 to 10-6, with deeper levels of MRD negativity associated with better outcomes. MRD assessment using Next Generation Sequencing (NGS) is increasingly used in clinical practice. In this study, we assessed the feasibility of detecting a trackable baseline clonotype identification ID and the depth of sensitivity achievable by NGS-based MRD testing in real-world settings.
Methods: We identified MM patients who had a BM identification (ID) sample and MRD testing by NGS (ClonoSEQ) performed at 2 academic centers: Medical College of Wisconsin and University of Arkansas for Medical Sciences. Patients were included if they had BM MRD testing at any point during their disease course, regardless of treatment or disease status. Collected data included baseline ID sequences, number of tracked and detected ID sequences in subsequent tests, tumor cells detected per million counts, and number of cells assessed. The presence of any residual sequence was considered MRD positive (MRDpos), and its absence was MRD negative (MRDneg).
Results: A total of 647 unique patients were included. A baseline trackable baseline ID was identified in 90% of patients. Among these patients, majority (57%) had 2-3 trackable V(D)J sequences at baseline; 12% had 1, 18% had 4-5, and 2% had >5 sequences. A total of 1,411 MRD tests were performed among 595 patients. Overall, 686 (49%) of BM samples were MRDpos and 725 (51%) were MRDneg. The median number of trackable V(D)J sequences decreased from 3 at baseline to 2 during subsequent MRD testing. Among the BM MRDpos samples, 32% had 1-10 tumor cells per million, corresponding to a sensitivity range of 10-5 to 10-6. Additionally, 40% of samples achieved a sensitivity greater than 10-5 and 28% achieved a sensitivity greater than 10-6. Notably, in 28%, residual disease was detectable only at a sensitivity of 10-6, which would have been classified as MRDneg by less sensitive assays. Of the 725 (51%) BM MRDneg samples, the median number of cells analyzed was 2.8 million (range: 0.04–6.29 million). Of these, 604 (83%) of samples assessed at least 1 million cells, achieving a sensitivity of 10-6.
Conclusions: Identification of a baseline trackable ID was successful in the majority of MM patients. With a substantial proportion of MRD tests achieving a sensitivity of 10-6 and over a quarter of positive samples having residual disease detectable only at a 10-6 sensitivity, these findings support the feasibility of achieving deep MRD sensitivities in real-world clinical practice and support future research in the optimal sensitivity thresholds for MRD assessment beyond 10-5 that may further improve patient outcomes.
DISCLOSURES
Jessica M. Kent, N/A: No relevant conflicts of interest to declare.