Background: Assessing MRD has become a standard procedure in clinical trials to evaluate treatment efficacy. In accordance with its consistent prognostic value, the International Myeloma Working Group added MRD-negative criteria into response guidelines for its standardized use in clinical trials. That notwithstanding, the expectations for MRD as biomarker are to use it in routine clinical practice to help in treatment decisions, since in most clinical trials the therapeutic approach is defined upfront and does not vary according to patients' depth of response. However, the use of MRD in clinical practice is controversial and it remains unknown if tailoring treatment to achieve MRD-negativity is safe and improves patients' survival.

Aim: Compare in clinical practice, outcome and tolerability of a treatment strategy tailored to achieve sustained undetectable MRD by NGF and imaging, as compared to conventional treatment approaches that are not modified according to patients' depth of response.

Methods: This study was conducted in a single Hospital and included a total of 66 patients with newly-diagnosed MM from July 2014 to May 2019. All patients younger than 76 were prospectively included, whereas patients with high frailty score, severe senile dementia, other neoplasms, or with significant comorbidities in whom the therapeutic objective was only palliative care were excluded. In accordance to the local ethical committee and the Helsinki Declaration, all patients gave informed consent prior entering the study and were given the choice between the MRD and image driven (MRD-driven) and the conventional treatment (CT) approach. In the former, persistent MRD after the first-line of therapy was considered as treatment failure and patients received subsequent lines until achieving undetectable MRD by NGF and imaging (treatment endpoint). In the CT approach, subsequent lines of therapy were given upon progressive disease. The most commonly used first, second, and third line therapies in the MRD-driven approach were VBMCP/VBAD, VCD, and lenalidomide combinations, whereas in the CT cohort these were VCD for first-line, and lenalidomide combinations in second and third lines. Maintenance therapy (Interferon α2b + Prednisone for a year) was administered in 61% of patients treated according to the MRD-driven approach, and in 12% (bortezomib until progression) in the CT cohort. MRD was assessed in patients achieving complete remission using EuroFlow NGF, with a limit of detection of 2x10-6. Undetectable MRD by imaging was defined by negative PET/CT and by negative MRI of the spine and pelvis.

Results: Of the 66 patients enrolled thus far, 49 were treated with the MRD-driven and 17 with the CT approach. There were no significant differences between groups regarding patients' age (median, 62 years), the Revised-ISS (37.5%, 53% and 37.5% with R-ISS-I, -II and -III) or the usage of HDT/ASCT (85% vs 76%; P>.05). Approximately 80% of patients treated with the MRD-driven approach achieved undetectable MRD at 30 months. The median time from start of treatment to undetectable MRD was 24 months, after a average of 2.2 lines of therapy. By contrast, only 1 (6%) patient treated with CT showed undetectable MRD after first line of therapy. With a median follow-up of 29 months, progression-free survival (PFS) rates at 30 months were 92% for patients treated with the MRD-driven vs 28% for the CT approach (hazard ratio 0.10 [0.04-0.30]; p<0.0001). Such impact in PFS was similarly observed in sub analyses of patients treated with or without maintenance. A trend for prolonged overall survival (OS) was observed (92% vs 81% at 30-months, respectively; p=0.12). Of note, PFS and OS rates for patients treated with the MRD-driven strategy and achieving undetectable MRD were of 100% at 30 months. No significant differences were observed regarding the number of patients with adverse events treated with the MRD-driven vs CT (47% vs 59%), the number of adverse events per patient (mean of 1.9 vs 2.4), or in the number of adverse events per year of exposure (annual incidence of 1.2 vs 2.1).

Conclusions: We show that as compared to pre-specified therapies that are not modified according to patients' depth of response, an MRD-driven treatment approach reduces by 90% the risk of progression or death without increasing toxicity. These results support the use of undetectable MRD by NGF and imaging as treatment endpoint in routine clinical practice.

Disclosures

Paiva:Amgen, Bristol-Myers Squibb, Celgene, Janssen, Merck, Novartis, Roche, and Sanofi; unrestricted grants from Celgene, EngMab, Sanofi, and Takeda; and consultancy for Celgene, Janssen, and Sanofi: Consultancy, Honoraria, Research Funding, Speakers Bureau.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution