For patients with newly diagnosed multiple myeloma, reaching minimal residual disease (MRD) negativity after treatment is associated with improved outcomes; however, the use of MRD to modulate therapy remains elusive. We present the final analysis of the MASTER trial of daratumumab, carfilzomib, lenalidomide, and dexamethasone (Dara-KRd) therapy in patients with newly diagnosed multiple myeloma, in which MRD status is used to modulate treatment duration and cessation. MASTER was a multicentre, single-arm, phase 2 trial conducted in five academic medical centres in the USA. Eligible participants were 18 years or older with newly diagnosed multiple myeloma (measurable by serum or urine protein electrophoresis or serum free light chains), a life expectancy of at least 12 months, and an Eastern Cooperative Oncology Group performance status of 0-2, and had received no previous treatment for multiple myeloma except up to one cycle of therapy containing bortezomib, cyclophosphamide, and dexamethasone. The study was enriched for participants with high-risk chromosome abnormalities (HRCAs). During the induction phase, participants received four 28-day cycles of Dara-KRd, each comprising daratumumab (16 mg/kg intravenously on days 1, 8, 15, and 22), carfilzomib (56 mg/m2 intravenously on days 1, 8, and 15), lenalidomide (25 mg orally on days 1-21), and dexamethasone (40 mg orally or intravenously on days 1, 8, 15, and 22); induction was followed by autologous haematopoietic stem-cell transplantation and up to two phases of consolidation with Dara-KRd. We assessed MRD by next-generation sequencing after or during each phase. The primary endpoint was reaching MRD negativity (<10-5). Participants who reached MRD negativity after or during two consecutive phases stopped treatment and began observation with MRD surveillance (MRD-SURE); participants who did not reach two consecutive MRD-negative results received maintenance lenalidomide. Secondary endpoints included progression-free survival and cumulative incidence of progression. All analyses were conducted in the intention-to-treat population. This trial is registered with ClinicalTrials.gov, NCT03224507, and is complete. Between Mar 21, 2018, and Oct 23, 2020, 123 participants were recruited to the study, of whom 70 (57%) were men, 53 (43%) were women, 94 (76%) were non-Hispanic White, 25 (20%) were non-Hispanic Black, and four (3%) were of another race or ethnicity. The median age of participants was 61 years (IQR 55-68), and 24 (20%) were aged 70 years or older. The median duration of follow up was 42·2 months (IQR 34·5-46·0). Of the 123 participants, 53 (43%) had no HRCAs, 46 (37%) had one HRCA, and 24 (20%) had two or more HRCAs. For 118 (96%) of 123 participants, MRD was evaluable by next-generation sequencing; the remaining five had an absence of sufficiently unique clonogenic sequences to enable tracking by the assay. Of these 118 participants, 96 (81%, 95% CI 73-88) reached MRD of less than 10-5 (comprising 39 [78%, 64-88] of 50 participants with no HRCAs, 38 [86%, 73-95] of 44 participants with one HRCA, and 19 [79%, 58-93] of 24 participants with two or more HRCAs) and 84 (71%, 62-79) reached MRD-SURE and treatment cessation. 36-month progression-free survival among all 123 participants was 88% (95% CI 78-95) for participants with no HRCAs, 79% (67-88) for those with one HRCA, and 50% (30-70) for those with two or more HRCAs. For the 84 participants reaching MRD-SURE, the 24-month cumulative incidence of progression from cessation of therapy was 9% (95% CI 1-19) for participants with no HRCAs, 9% (1-18) for those with one HRCA, and 47% (23-72) for those with two or more HRCAs. 61 participants (comprising 52% of 118 MRD-evaluable participants and 73% of 84 participants who reached MRD-SURE) remain free of therapy and MRD-negative as of Feb 7, 2023. The most common grade 3-4 adverse events were neutropenia (43 patients, 35%), lymphopenia (28 patients, 23%), and hypertension (13 patients, 11%). Three treatment-emergent deaths were recorded: two sudden deaths and one due to viral infection, none of which were judged to be treatment-related. This approach provided positive outcomes and a pathway for treatment cessation in most patients with newly diagnosed multiple myeloma. Outcomes for patients with ultra-high-risk multiple myeloma, defined as those with two or more HRCAs, remain unsatisfactory, and these patients should be prioritised for trials with early introduction of therapies with novel mechanisms of action. Amgen and Janssen Pharmaceuticals.