MRD2STOP 是一项实用性试验，评估多发性骨髓瘤 (MM) 中以可测量残留病 (MRD) 阴性为指导的维持治疗停止。符合条件的患者既往 MRD < 10-5，接受≥1 年的维持治疗，并前瞻性地通过正电子发射断层扫描、骨髓 (BM) 流式细胞术（检测限 [LoD] 10-5）确认患有不可检测的疾病，并且BM clonoSEQ (LoD 10-6)。通过 clonoSEQ 分析富含 CD138 细胞的 BM 抽吸物，以实现 MRD 10-7 灵敏度。我们评估了疾病复发的发生率和无进展生存期 (PFS)，按 10-7 状态分层。 47 名患者在中位 36 个月后停止维持治疗。 19% (9/47) 观察到基线 MRD ≥ 10-7。停药后的中位随访时间为 30 个月。 11 例患者出现疾病复发（MRD 10 ≥ -6），其中 5 例疾病进展。一名患者死于第二种癌症。基线 MRD < 10-7 的患者的估计 3 年疾病复发累积发生率为 20%，而 MRD ≥ 10-7 的患者为 75%（HR 7.8，95% CI 2.2-27.6，p= 0.001）。与 MRD < 10-7 相比，基线 MRD ≥ 10-7 与较差的 PFS 相关（HR 10.1，95% CI 1.6-62.3；3 年 PFS 49% vs 92%）。 MM 和 MRD < 10-6 的患者停止维持治疗导致疾病复发率较低。 MRD < 10-7 可能是较高的戒烟阈值，需要进一步验证。© 2024。作者。

MRD2STOP is a pragmatic trial evaluating maintenance therapy cessation guided by measurable residual disease (MRD) negativity in multiple myeloma (MM). Eligible patients had previous MRD < 10-5, received ≥1 year of maintenance, and were prospectively confirmed to have undetectable disease by positron emission tomography, bone marrow (BM) flow cytometry (limit of detection [LoD] 10-5), and BM clonoSEQ (LoD 10-6). BM aspirates enriched for CD138+ cells were analyzed by clonoSEQ to achieve MRD 10-7 sensitivity. We evaluated the incidence of disease resurgence and progression-free survival (PFS), stratified by 10-7 status. Forty-seven patients discontinued maintenance after a median of 36 months. Baseline MRD ≥ 10-7 was observed in 19% (9/47). The median follow-up post-discontinuation was 30 months. Disease resurgence (MRD 10 ≥ -6) occurred in 11 patients, including 5 disease progressions. One patient died from a second cancer. The estimated 3-year cumulative incidence of disease resurgence was 20% for patients with baseline MRD < 10-7 compared to 75% for MRD ≥ 10-7 (HR 7.8, 95% CI 2.2-27.6, p = 0.001). Baseline MRD ≥ 10-7 was associated with inferior PFS compared to MRD < 10-7 (HR 10.1, 95% CI 1.6-62.3; 3-year PFS 49% vs 92%). Maintenance discontinuation in patients with MM and MRD < 10-6 led to low rates of disease resurgence. MRD < 10-7 may be a superior cessation threshold, requiring further validation.© 2024. The Author(s).