Predictive role of sustained imaging MRD negativity assessed by diffusion‐weighted whole‐body MRI in multiple myeloma

Multiple Myeloma (MM) is still considered an incurable disease. However, the availability of novel and effective therapies is making the prospect of eradicating the malignant clone and converting MM into a curable disease realistic. In this context, the detection of minimal residual disease (MRD) has become of utmost importance. The high spatial genomic heterogeneity that characterizes MM patients explains the potential persistence of residual disease even in bone marrow (BM) MRD negative patients, highlighting the need of different techniques able to detect the disease within and outside of the BM.1 The prognostic usefulness of both MRD assessment and functional imaging response after treatment have been established.2 The combination of BM and imaging techniques has a direct impact on prognosis, with patients reaching double MRD negativity having the best outcome.3 Recently, sustained BM MRD-negativity and its association with an excellent outcome is emerging in clinical trials.4 Diffusion-weighted whole body MRI (DW-MRI) is increasingly used in the management of MM. It produces images in which the contrast between tissues is based on differences in the motion of water at a cellular level and permits to estimate cell density with the so-called apparent diffusion coefficient (ADC). The Myeloma Response Assessment and Diagnosis System (MY-RADS) recommendations have established criteria for Response Assessment Category (RAC) with a 5-point scale defining the range of imaging response after treatment from complete (RAC 1) to progressive disease (RAC 5).5 We have previously shown that RAC criteria were able to independently stratify patients with different outcomes after autologous stem cell transplantation (ASCT).6 Data regarding the predictive role of sustained imaging MRD-neg assessed by DW-MRI are lacking. With this aim, we implemented DW-MRI and RAC criteria in MM patients both after ASCT and at 1 year during maintenance therapy in order to evaluate the predictive information of sustained imaging MRD negativity on outcome. All transplant-eligible MM patients newly diagnosed at our institution between January 2016 and December 2019 who received maintenance therapy after ASCT were retrospectively analyzed. Treatment response according to IMWG criteria was assessed before maintenance. DW-MRI was performed at day +100 after ASCT, and thereafter at 1 year to monitor imaging-residual-disease. In patients receiving double ASCT, DW-MRI performed at day +100 after the second one was considered. Protocol details and application of RAC criteria in our external validation of MY-RADS guidelines in clinical practice have been previously published. BM samples were also collected for MRD assessment by 8-color multiparametric flow-cytometry (MFC, sensitivity 10−5) at day +100 after ASCT, and thereafter at 1-year in MRD-neg patients in order to assess sustained MRD negativity. We focused on the predictive role of sustained 1-year DW-MRI-neg on progression free survival (PFS) and overall survival (OS). PFS after second-line treatment was also evaluated. Kaplan–Meyer method was used to estimate survival curves and multivariate analysis (Cox proportional hazard models) was performed to evaluate the influence of prognostic factors on outcome. In patients with available 1-year BM MRD evaluation, the concordance between BM and DW-MRI results was calculated (Cohen's kappa statistics). This study has been approved at our institution by the local Ethics Committee and has been done in accordance with the Declaration of Helsinki. We identified 70 newly-diagnosed MM patients who underwent ASCT followed by maintenance therapy and had a DW-MRI scan at day +100 after ASCT and at 1-year. Induction regimens used were: VTD in 55 (79%), VRD in five (7%), Dara-VRD in six (9%), KRD in three (4%), KCD in one (1%). Forty patients (57%) received a single ASCT (MEL200), whereas double ASCT was performed in 30 patients (43%). All patients received lenalidomide maintenance, in combination with daratumumab in six (9%) and with carfilzomib in one (1%). Median follow-up duration was 46 months. Median age was 61 years (40–73), 28 (40%) patients were ISS stage III and 21 (30%) of them had high-risk cytogenetics (t[4;14] and/or t[14;16] and/or del17p and/or 1q gain/amp1q abnormalities). IMWG responses at day +100 after ASCT were: PR 6 (8%), VGPR 15 (21%), CR 38 (54%), and sCR 11 (16%). Complete imaging response according to MY-RADS was achieved after ASCT in 41 patients (RAC 1: 59%), whereas residual disease was observed in 29 patients (RAC >2:41%). BM MFC for MRD detection was available before maintenance in 61 patients and was negative in 39 (64%) patients. At 1-year DW-MRI complete imaging response (RAC 1) was observed in 53 (76%) patients, whereas residual disease was documented in 17 patients (24%). BM MRD evaluation at 1-year was available in 35 patients and was negative in 29 (83%). The concordance between 1-year DW-MRI and BM MFC MRD results was high (85%, kappa 0.46: 8% both positive, 77% both negative). Median PFS was significantly longer for patients with imaging MRD-neg at 1-year (RAC 1) compared to patients with residual disease on DW-MRI (RAC ≥2) (median PFS: 55.4 vs. 28.4 months; 3-years PFS: 91% vs. 30%, respectively [HR 0.12; 95% CI: 0.04–0.35; p < 0.0001]) (Figure 1A). Survival of imaging MRD-neg patients at 1-year was significantly longer for patients with RAC1 versus RAC ≥2 (median OS: NR vs. 63 months; 3-years OS: 100% vs. 82%, respectively [HR 0.13; 95% CI: 0.03–0.66; p 0.0007]) (Figure 1B). Upgrade of imaging response from RAC ≥2 at maintenance start to RAC1 at 1-year was observed in 13 patients (19%) Their outcome was superimposable to that of other MRD-neg patients at 1-years. Fifteen out of 17 RAC ≥2 patients at 1-year after ASCT (88%) were treated with the following second-line therapies at biochemical or clinical relapse: DVD: 6 (40%), KD reinduction and ASCT: 6 (40%), KPD:1 (7%), and PVD in 2 (13%) patients. Their median PFS was 13 months. Notably, of 29 patients in sustained MFC MRD-neg at 1-year, two had imaging-residual-disease on DW-MRI. Both developed aggressive extramedullary relapse, one with CNS involvement. Among the prognostic parameters available at 1-year after transplant in all 70 patients (IMWG response, cytogenetic risk profile, ISS stage and RAC score at 1-year), multivariable analysis showed that both PFS and OS were independently affected only by DW-MRI persistent disease (RAC ≥2): p < 0.001, HR 0.12 (95% CI: 0.05–0.30) for PFS, p = 0.032, HR 0.20 (95% CI: 0.05–0.87) for OS. MRD evaluation, an essential step toward achievement of operational cure in MM, has evolved from a single timepoint assessment to repeated monitoring of MRD over time, since sustained BM MRD-neg appears a better predictor of longer PFS and OS. Few data have been reported to date regarding sustained imaging MRD-negativity during maintenance therapy using PET and ancillary studies are under investigations in clinical trials. To our knowledge, this is the first report describing the prognostic value of sustained imaging MRD-negativity assessed by DW-MRI in MM. Our results highlight the ability of DW-MRI to identify a significant proportion of patients who fail to achieve sustained imaging MRD-negativity at 1-year during maintenance therapy (RAC ≥2: 24%), whose outcome is dismal despite being responsive according to IMWG criteria. Compared to RAC1 patients their PFS was halved and OS was significantly shorter. Moreover, their PFS after salvage with the most effective standard second-line combination therapies available was very short. The retrospective nature of our observations, the relatively low number of patients and the lack of BM MFC evaluation at 1-year in the entire cohort of patients represent major limitations of our study, which requires further confirmation. Nevertheless, they lend support to the use of functional imaging techniques to allow a better dynamic risk assessment in order to identify patients at increased risk of relapse and particularly poor prognosis. For patients failing to achieve sustained DW-MRI MRD-negativity, an earlier use of novel emerging salvage treatments, such as CAR-T cells or bispecific antibodies may be considered. Angelo Belotti designed the research and analyzed data and wrote the paper, Rossella Ribolla collected data and performed research, Claudia Crippa collected data, Marco Chiarini performed the research, Viviana Giustini performed the research, Samantha Ferrari collected data, Annalisa Peli collected data, Chiara Cattaneo collected data, Aldo Roccaro supervised research, Barbara Frittoli performed research, Luigi Grazioli performed research, Giuseppe Rossi analyzed data and wrote the paper, Alessandra Tucci analyzed data and wrote the paper. We thank the patients and staff of the Department of Hematology, Spedali Civili, Brescia. We also thank the Department of Radiology, Spedali Civili, Brescia. AMR: Fondazione AIRC; Fondazione Regionale per la Ricerca Biomedica, ERA-NET TRANSCAN-2. The authors declare no conflict of interest to disclose. The data that support the findings of this study are available from the corresponding author upon reasonable request; the data are not publicly available due to privacy and ethical restrictions.