Therapeutic approaches for treating MM have advanced to include novel drugs, particularly immunotherapies. The combined use of proteasome inhibitors, thalidomide analogs, and CD38-targeting monoclonal antibodies currently represents the mainstay of modern myeloma therapy. New monoclonal antibodies, T-cell activators, and cell therapy are also in the process of entering the clinics. Although a definite cure for MM is still lacking, the introduction of new drugs with different mechanisms and improved treatment approaches has significantly improved the survival of MM patients [27]. MRD has a strong predictive value in various disease states and treatment conditions [28, 29]. It can identify the likelihood of relapse and enable early intervention. Evaluation of MRD rates is also used as an endpoint to accelerate drug testing and approval in many trials [30, 31].

Although many methods, such as NGF and multi-parametric flow cytometry, can be used to detect MRD, there is no standard method. At present, NGF is the most common method used for detecting MRD in clinics. MRD detection based on NGF is fast, efficient, and economical; however, it requires complex visualization and professional data analysis. Furthermore, false-negative MRD detection can occur in some patients due to immunophenotypic changes post-treatment [32].

The use of NGS for detecting MRD has increasingly been implemented in clinical practice. A study has compared NGS and NGF in detecting MRD and concluded a strong correlation between the two approaches [33].

MM patients in China tend to be younger than those in Europe or the United States. Furthermore, there is a high demand for effective management of MM in this population. However, the detection of MRD in Chinese MM patients has primarily relied on flow cytometry, which is gradually becoming insufficient to meet patient needs. Although international studies have previously reported on MRD monitoring via second-generation sequencing, there is limited research on the Chinese population.

To the best of our knowledge, this study is the first to compare NGS with NGF in China. To monitor MRD by using NGS, the clonal IGH rearrangement at the time of diagnosis must be known. In this study, both NGS and CE were used to detect the clonal IGH rearrangements in 60 newly diagnosed Chinese MM patients, and the consistency between the two methods was 98.3%. The overall detection rate of the IGH-FR1/FR2/FR3 combination was 70.0% via NGS. Additionally, unique clonal IGH rearrangements were observed in 42 patients. Therefore, NGS could detect clonal rearrangements in most of the newly diagnosed MM patients. Such detection can serve as a molecular biomarker at the time of diagnosis, enabling MRD monitoring during clinical treatment. To evaluate the feasibility of NGS in follow-up MRD monitoring of MM patients, we analyzed the limit and repeatability of NGS in detecting MRD. According to the IMWG guidelines, MM patients are considered MRD-negative if there are no clonal plasma cells in the bone marrow, with a minimum sensitivity of 1 in 105 nucleated cells via the NGS method [14]. The results of the study presented here confirmed that NGS has good sensitivity in MRD detection, and demonstrated a linear curve ranging from 10–6 to 10–1, with a correlation coefficient of 0.985. Using this method, it is possible to detect one tumor plasma cell in 1,000,000 nucleated cells, indicating a limit of detection of 10–6. Thus, NGS exhibits high sensitivity in MRD detection in MM patients. In addition, this approach showed good repeatability in MRD detection in these patients. In samples with different tumor loads, the MRD levels were estimated at 10–2, 10–3, and 10–4 via NGS, and the intra- and inter-assay variation was relatively low.

Currently, the major approaches recommended for MRD assessment in MM patients at home and abroad are the multi-parameter NGF and NGS technologies. There are relatively many reports on multi-parameter NGF in MRD detection in MM patients [34, 35], whereas the applicability of NGS has seldom been reported in China yet. In this study, 43 samples from 36 patients were evaluated at follow-up by using both NGS and NGF. Our results revealed a consistency rate of 79.1% between the two methods, showing that both methods have high consistency. Interestingly, out of the cases analyzed, 9 showed inconsistent MRD results, with MRD levels being detectable via NGS but undetectable via NGF. It is worth noting that none of the samples identified as MRD-positive via NGF were found to be negative via NGS. Discrepancies between NGF and NGS in detecting MRD can be attributed to differences in sensitivity and detection principles. NGF, which relies on antibodies targeting cell surface proteins, and NGS, which identifies genetic mutations, focus on distinct biological markers [36, 37]. This divergence in methodological focus can lead to scenarios where MRD is detectable by one technique but remains undetected by the other, reflecting the distinct detection capabilities inherent to each method [38].

MM patients undergoing CAR-T therapy targeting MM surface antigens, such as CD138 and CD229, may experience blocking of these antigen-binding sites for several months [39], This necessitates adjustments in the use of NGF for MRD detection. Interestingly, after ASTC, 2 patients tested MRD-negative via NGF. However, the NGS method revealed MRD levels of 2.83 × 10–5 and 1.10 × 10–4 in these patients. Notably, both patients demonstrated a very good partial response according to the evaluation of treatment effectiveness following treatment with VRd. Previous studies have also reported [40] that after induction treatment or transplantation, MRD that turns negative indicates a better clinical prognosis.

Retrospective studies have shown that making treatment decisions based on MRD results (including stopping, intensifying, or changing the treatment) can improve progression-free survival in comparison with patients whose treatment remains unmodified after MRD assessment [31, 32, 41]. The prognostic value of MRD, as determined by NGS, offers a robust basis for informed treatment adjustments, encompassing de-escalation, intensification, or modification strategies to halt disease progression and improve outcomes [29, 41]. Furthermore, by uncovering the genetic and immunologic drivers of MRD, NGS facilitates the development of targeted therapies, advancing personalized medicine in MM. This transformative approach not only promises improved therapeutic efficacy and patient well-being but also significantly shifts the MM management paradigm [41]. Although NGS presents a higher per-sample costs, its superior sensitivity in detecting MRD at very low levels offers potential long-term cost savings. Early and accurate MRD detection can guide more effective treatment adjustments, potentially reducing the overall treatment costs by avoiding unnecessary therapies and hospitalizations.

Emerging technologies, especially the integration of artificial intelligence (AI) and machine learning (ML) are set to enhance MRD detection in MM by processing complex datasets more efficiently, automating the identification of novel MRD markers, and enabling personalized treatment plans through predictive modeling [42, 43]. Concurrently, the discovery of new biomarkers such as extracellular matrix proteins, angiogenic factors, p53-related protein kinase, circulating tumor cells, and microRNAs is redefining MM diagnosis and treatment [44,45,46,47,48]. The future of MM management is geared towards integrating these technologies and biomarkers into a personalized, predictive, and patient-centered care framework.