Abstract: The International Staging System (ISS) and the Revised International Staging System (R-ISS) are commonly used prognostic scores in multiple myeloma (MM). These methods have significant gaps, particularly among intermediate-risk groups. The aim of this study was to improve risk stratification in newly diagnosed MM patients using data from three different trials developed by the Spanish Myeloma Group. For this, we applied an unsupervised machine learning clusterization technique on a set of clinical, biochemical and cytogenetic variables, and we identified two novel clusters of patients with significantly different survival. The prognostic precision of this clusterization was superior to those of ISS and R-ISS scores, and appeared to be particularly useful to improve risk stratification among R-ISS 2 patients. Additionally, patients assigned to the low-risk cluster in the GEM05 over 65 years trial had a significant survival benefit when treated with VMP as compared with VTD. In conclusion, we describe a simple prognostic model for newly diagnosed MM whose predictions are independent of the ISS and R-ISS scores. Notably, the model is particularly useful in order to re-classify R-ISS score 2 patients in 2 different prognostic subgroups. The combination of ISS, R-ISS and unsupervised machine learning clusterization brings a promising approximation to improve MM risk stratification.

Abstract: Patients with multiple myeloma (MM) carrying standard- or high-risk cytogenetic abnormalities (CAs) achieve similar complete response (CR) rates, but the later have inferior progression-free survival (PFS). This questions the legitimacy of CR as a treatment endpoint and represents a biological conundrum regarding the nature of tumor reservoirs that persist after therapy in high-risk MM. We used next-generation flow (NGF) cytometry to evaluate measurable residual disease (MRD) in MM patients with standard- vs high-risk CAs (n = 300 and 90, respectively) enrolled in the PETHEMA/GEM2012MENOS65 trial, and to identify mechanisms that determine MRD resistance in both patient subgroups (n = 40). The 36-month PFS rates were higher than 90% in patients with standard- or high-risk CAs achieving undetectable MRD. Persistent MRD resulted in a median PFS of ∼3 and 2 years in patients with standard- and high-risk CAs, respectively. Further use of NGF to isolate MRD, followed by whole-exome sequencing of paired diagnostic and MRD tumor cells, revealed greater clonal selection in patients with standard-risk CAs, higher genomic instability with acquisition of new mutations in high-risk MM, and no unifying genetic event driving MRD resistance. Conversely, RNA sequencing of diagnostic and MRD tumor cells uncovered the selection of MRD clones with singular transcriptional programs and reactive oxygen species–mediated MRD resistance in high-risk MM. Our study supports undetectable MRD as a treatment endpoint for patients with MM who have high-risk CAs and proposes characterizing MRD clones to understand and overcome MRD resistance. This trial is registered at www.clinicaltrials.gov as #NCT01916252.

Abstract: Introduction: To discriminate different outcomes among patients in CR, the International Myeloma Working Group (IMWG )introduced more stringent CR (sCR) criteria by adding to the pre-existing CR parameters the requirement of a normal free-light chain ratio (sFLCr) plus the absence of clonal plasma cells (PCs) in bone marrow (BM) by immunohistochemistry (IHC). In 2011,the low-sensitivity cytometrycriteria were included as alternative methodology to IHC to define sCR. Aim: To validate the preliminary data of our previous study (Blood 2015. 126:858-62) regarding the lack of influence of an abnormal sFLCr in the outcome of MM patients, through the analysis of a more extensiveseries of newly diagnosed multiple myeloma (NDMM) patients in CR or sCR. Patients and Methods: This study is based on 459 NDMM patients who were transplant candidates and enrolled in the GEM2012MENOS65phase 3trial;evaluable patients were enrolled in a subsequent maintenance trial (NCT02406144).CR and sCR was defined according to the IMWG criteria. Agreeing to the protocol, patients with 〈 5% BM PCs and negative serum immunofixation (IF) but with unavailable urine IF should have been classified as VGPR, but after the results of recent analysis conducted by our group showing that these patients had the same outcome of those in CR (unpublish information), these patients were reclassified as CR. SFLCr (FREELITE assay) was stratified as normal (0.26-1.65) or abnormal ( 〈 0.26 if the patient was λ; 〉 1.65 if the patient was κ). BM aspirates were assessed for morphological enumeration of PCs and monitoring of minimal residual disease (MRD) using next-generation flow (NGF) according toEuroFlow SOPs. The median limit of detection was of 3x10-6. We classified as sCR all patients in CR with normal sFLCr and absence of clonal PCs by NGF with a reduced threshold of sensitivity to 10-4.The median follow-up was 40 months. Results: After ASCT,392 patients were evaluable for response; 239 (61%) reached ≥CR. Data from sFLCr and MRD was available in 225 and 221 patients, respectively. In 153 out of a total of 203 (74%) patients in CR in which complete information about FLC and MRD was available were categorized as sCR. The remaining 55 patients were consider in CR because of failure to accomplish 1 of the 2 criteria: abnormal sFLCr (n=49) or MRD+veby low sensitivity flow (n=11); 5patientsshared both criteria.In a landmark from ASCT, with a follow up of 27 months, sCRdidn't show significantly differences inPFS (2 years-PFS 90% vs 83%; P=.2) neither in OS (2 years-OS 96% vs 98%; P=.6) as compared to CR patients.Interestingly, patients with abnormal (n=51) vs normal (n=174) sFLCr showed superimposable PFS (2 years-PFS 86% vs 88%; P=.6) and OS (2 years-OS 95% vs 100%; P=.2).By contrast, in the 11 patients (out of the 221, 5%) with persistent MRD ( 〉 10-4) the PFS was significantly poorer as compared with MRD-ve cases (2-yearsPFS 91% vs48%; P=.001)but the OS was similar (2 years-OS 98% vs 96%; P=.3).As validation, we reproduced the analysis in the consolidation-2 end-point (figure 1), where375patients were evaluable for response assessment,267 of them (71%) reached ≥CR. Once again, in the landmark analysis, sCR didn't show significantly differences in PFS with respect to CR patients (2 years-PFS 88% vs 84%; P=.2) neither in OS (2 years-PFS 96% vs 90%; P=.3); moreover, patients with abnormal (n=55) vs normal (n=195) sFLCr showed superimposable PFS (2 years-PFS 84% vs 87%; P=.4) and OS (2 years-OS 89% vs 96%; P=.2).In the MRD analysis, patients with persistent MRD, had significantly inferior PFS (2-years PFS 87% vs 72%; P=.04 for 〉 10-4 MRDsensitivity). If we increase the sensitivity of the MRD to 10-6, the differences in PFS at 2 years are more evident (2 years-PFS 94% vs 67%; P 〈 .000001 for 〉 10-6 sensitivity). Conclusion: These results confirm our previous findings based on GEM05menos65/ GEM10mas65 clinical trials, indicating that for MM patients stringent CR criteria does not predict a different outcome as compared to standard CR. Specifically, the sFLCr doesn't identify patients in CR at distinct risk. If this essential criterion in the definition of sCR lacks connotations for the prognosis, is it not justified to maintain a response category whose real significance depends on the combination of the traditional CR criteria with a negative MRD status based on very low (IHC) or low resolution ( 〈 10-4) levels, which is outdated. Figure 1. Figure 1. Disclosures Martinez Lopez: Janssen: Research Funding, Speakers Bureau; Bristol Myers Squibb: Research Funding, Speakers Bureau; Novartis: Research Funding, Speakers Bureau; Celgene: Research Funding, Speakers Bureau. Rosinol:Janssen, Celgene, Amgen, Takeda: Honoraria. Puig:Takeda: Consultancy, Honoraria; Janssen: Consultancy, Honoraria, Research Funding; Celgene: Honoraria, Research Funding. Oriol:Celgene: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Amgen: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Takeda: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Janssen: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau. Ocio:Pharmamar: Consultancy; AbbVie: Consultancy; Seattle Genetics: Consultancy; BMS: Consultancy; Janssen: Consultancy, Honoraria; Takeda: Consultancy, Honoraria; Novartis: Consultancy, Honoraria; Sanofi: Research Funding; Amgen: Consultancy, Honoraria, Research Funding; Mundipharma: Research Funding; Celgene: Consultancy, Honoraria, Research Funding; Array Pharmaceuticals: Research Funding. De La Rubia:Ablynx: Consultancy, Other: Member of Advisory Board. Rios:Amgen, Celgene, Janssen, and Takeda: Consultancy. Mateos:Janssen: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; GSK: Consultancy, Membership on an entity's Board of Directors or advisory committees; Abbvie: Consultancy, Membership on an entity's Board of Directors or advisory committees; GSK: Consultancy, Membership on an entity's Board of Directors or advisory committees; Amgen: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Celgene: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Takeda: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Amgen: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees. San-Miguel:Sanofi: Consultancy; Takeda: Consultancy; Novartis: Consultancy; MSD: Consultancy; Janssen: Consultancy; Celgene: Consultancy; Brystol-Myers Squibb: Consultancy; Amgen: Consultancy; Roche: Membership on an entity's Board of Directors or advisory committees. Bladé:Amgen: Honoraria; Celgene: Honoraria; Janssen: Honoraria. Lahuerta:Takeda: Honoraria, Membership on an entity's Board of Directors or advisory committees; Amgen: Honoraria, Membership on an entity's Board of Directors or advisory committees; Celgene: Honoraria, Membership on an entity's Board of Directors or advisory committees; Janssen: Honoraria, Membership on an entity's Board of Directors or advisory committees.

Abstract: Background: The international criteria for definition CR, requires, among other parametres, a negative IF both in serum and urine; however, urine IF is not always performed. In the belief that this lack could bias the comparison between trials, the First Trial Independent Response Adjudication Committee (FTIRAC) recommended that patients who met all CR criteria except the availability of a urine IF should be classified as VGPR (Blood 2014; 124 [abstract 3460]) but this criteria is not always applied which may translate into differences in CR rates between trials. However, it is unknown (1) if this conversion has a real clinical basis, (2) if urine IF results alter the clinical meaning of CR, or (3) on the contrary, if patients in CR with and those without a documented negative urine IF have a similar prognosis, in which case this rule would underestimate the CR rates, increasing the biases and magnifying the problem that was intended to improve. Aim: To determine the value of urine negative IF in the definition of CR. Methods: 459 patients were enrolled into the GEM2012MENOS65phase3 trial and treated with 6 induction cycles of bortezomib, lenalidomide, and dexamethasone, HDT/ASCT and 2 consolidation courses. Evaluable patients were enrolled in a maintenance trail (NCT02406144). Excluding 6 patients who discontinued early, 453 were evaluable. At diagnosis, the M-component was detected exclusively in serum in 173 of these patients and in serum and urine in 212 patients; 68 patients had pure Bence-Jones M-protein (BJMM). The protein studies were performed in each cycle. At the time of negative IF, bone marrow aspirates were analysed for count of PCs and monitoring minimal residual disease (MRD) following EuroFlow SOPs (median limit of detection of 3x10-6).The response classifications were made according to the IMWG criteria, but we applied the FTIRAC criteria, and, patients with 〈 5% BM PCs and negative serum IF but with unavailable urine IF (or vice versa for patients with BJMM) were classified as VGPR. For the purpose of this study, we called these uncertain CR (uCR). Stringent Complete Responses were classified as CR.Median follow-up was 40 months. Results: Overall, 3774 protein evolution studieswere performed: 691 (18%) in CR, 802 (21%) in uCR and 868 (23%)in VGPR.In all patients with M-component exclusively in serum at diagnosis butwith negative serum IF after treatment, and available urine IF (174 patients, 1476 protein studies), the urine paraprotein IF detection rate was 0%. In patients with a positive M-component in both serum and urine at diagnosis, but with negative serum IF after treatment, and available urine IF (212patients, 1763 protein assessment), 11 protein evaluations in 6 patients (2.8%) tested positive in urine; in other words: in 97,2% of the patients a negative serum IF predicts for negative urine.Since MRD is a robust subrogate of depth of responses (J ClinOncol. 2017;35:2900-10), we compared MRD-verates in patients achieving CR, uCR and VGPR. Interestingly, there were no significant differences in the post-consolidation MRD-verates among patients in CR vsthose in uCR (68% vs 77%, P=.1), whereas significant differences were seen when comparing CR and uCRvs VGPR (23% in the latter; P≤.0001). Accordingly, a landmark analysis performed at HDT/ASCT, - time pointselected to improve the PFS observation time- , showed 2-year PFS rates of 88%, 87% and 77% inpatients in CR, uCR and VGPR, no differences in 2-year PFS rates between patients in CR vsuCR (P=.6) while patients in VGPR showed inferior PFS compared with those in uCR (P=.04). With this landmark, the MRD-negativity ratios are similar to those described after consolidation. Conclusions: In MM patients with M-component exclusively in serum at diagnosis, urine IF follow-up is unnecessary, while in patients with paraprotein in both serum and urine,a negative serum IF response is accompanied by a negative urinary IF in 97.8% of patients. Moreover, patients with CR but without available IF in urine display similar MRD-veand PFS rates compared withthose in CR. By contrast, MRD-veand PFS values in these patients are significantly superior to those in VGPR. Thus, our results suggest that, except for those with pure BJMM, patients fulfilling CR criteria but with unavailable urine IF should be classified as CR instead of VGPR. This data discourages the application of the FTIRAC conversion criteria. Also, the IMWG criteria for CR should be reviewed. Disclosures Rosinol: Janssen, Celgene, Amgen, Takeda: Honoraria. Puig:Janssen: Consultancy, Honoraria, Research Funding; Takeda: Consultancy, Honoraria; Celgene: Honoraria, Research Funding. Garcia-Sanz:Affimed: Research Funding. Oriol:Amgen: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Celgene: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Takeda: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Janssen: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau. Rios:Amgen, Celgene, Janssen, and Takeda: Consultancy. De La Rubia:Ablynx: Consultancy, Other: Member of Advisory Board. Mateos:Janssen: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Celgene: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; GSK: Consultancy, Membership on an entity's Board of Directors or advisory committees; Takeda: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Amgen: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; GSK: Consultancy, Membership on an entity's Board of Directors or advisory committees; Abbvie: Consultancy, Membership on an entity's Board of Directors or advisory committees; Amgen: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees. San-Miguel:Brystol-Myers Squibb: Consultancy; Celgene: Consultancy; Novartis: Consultancy; Takeda: Consultancy; Amgen: Consultancy; MSD: Consultancy; Janssen: Consultancy; Sanofi: Consultancy; Roche: Membership on an entity's Board of Directors or advisory committees. Bladé:Celgene: Honoraria; Amgen: Honoraria; Janssen: Honoraria. Lahuerta:Takeda: Honoraria, Membership on an entity's Board of Directors or advisory committees; Amgen: Honoraria, Membership on an entity's Board of Directors or advisory committees; Celgene: Honoraria, Membership on an entity's Board of Directors or advisory committees; Janssen: Honoraria, Membership on an entity's Board of Directors or advisory committees.

Abstract: Assessing measurable residual disease (MRD) has become standard with many tumors, but the clinical meaning of MRD in multiple myeloma (MM) remains uncertain, particularly when assessed by next-generation flow (NGF) cytometry. Thus, we aimed to determine the applicability and sensitivity of the flow MRD-negative criterion defined by the International Myeloma Working Group (IMWG). PATIENTS AND METHODS In the PETHEMA/GEM2012MENOS65 trial, 458 patients with newly diagnosed MM had longitudinal assessment of MRD after six induction cycles with bortezomib, lenalidomide, and dexamethasone (VRD), autologous transplantation, and two consolidation courses with VRD. MRD was assessed in 1,100 bone marrow samples from 397 patients; the 61 patients without MRD data discontinued treatment during induction and were considered MRD positive for intent-to-treat analysis. The median limit of detection achieved by NGF was 2.9 × 10 −6 . Patients received maintenance (lenalidomide ± ixazomib) according to the companion PETHEMA/GEM2014MAIN trial. RESULTS Overall, 205 (45%) of 458 patients had undetectable MRD after consolidation, and only 14 of them (7%) have experienced progression thus far; seven of these 14 displayed extraosseous plasmacytomas at diagnosis and/or relapse. Using time-dependent analysis, patients with undetectable MRD had an 82% reduction in the risk of progression or death (hazard ratio, 0.18; 95% CI, 0.11 to 0.30; P 〈 .001) and an 88% reduction in the risk of death (hazard ratio, 0.12; 95% CI, 0.05 to 0.29; P 〈 .001). Timing of undetectable MRD (after induction v intensification) had no impact on patient survival. Attaining undetectable MRD overcame poor prognostic features at diagnosis, including high-risk cytogenetics. By contrast, patients with Revised International Staging System III status and positive MRD had dismal progression-free and overall survivals (median, 14 and 17 months, respectively). Maintenance increased the rate of undetectable MRD by 17%. CONCLUSION The IMWG flow MRD-negative response criterion is highly applicable and sensitive to evaluate treatment efficacy in MM.

Abstract: Patients with multiple myeloma (MM) may show patchy bone marrow (BM) infiltration and extramedullary disease. Notwithstanding, quantification of plasma cells (PCs) continues to be performed in BM since the clinical translation of circulating tumor cells (CTCs) remains undefined. PATIENTS AND METHODS CTCs were measured in peripheral blood (PB) of 374 patients with newly diagnosed MM enrolled in the GEM2012MENOS65 and GEM2014MAIN trials. Treatment included bortezomib, lenalidomide, and dexamethasone induction followed by autologous transplant, consolidation, and maintenance. Next-generation flow cytometry was used to evaluate CTCs in PB at diagnosis and measurable residual disease (MRD) in BM throughout treatment. RESULTS CTCs were detected in 92% (344 of 374) of patients with newly diagnosed MM. The correlation between the percentages of CTCs and BM PCs was modest. Increasing logarithmic percentages of CTCs were associated with inferior progression-free survival (PFS). A cutoff of 0.01% CTCs showed an independent prognostic value (hazard ratio: 2.02; 95% CI, 1.3 to 3.1; P = .001) in multivariable PFS analysis including the International Staging System, lactate dehydrogenase levels, and cytogenetics. The combination of the four prognostic factors significantly improved risk stratification. Outcomes according to the percentage of CTCs and depth of response to treatment showed that patients with undetectable CTCs had exceptional PFS regardless of complete remission and MRD status. In all other cases with detectable CTCs, only achieving MRD negativity (and not complete remission) demonstrated a statistically significant increase in PFS. CONCLUSION Evaluation of CTCs in PB outperformed quantification of BM PCs. The detection of ≥ 0.01% CTCs could be a new risk factor in novel staging systems for patients with transplant-eligible MM.

Abstract: Background: Despite significant improvements in the treatment of MM, the outcome of patients with HR cytogenetics remains poor despite similar complete remission (CR) rates as compared to SR cases. Relapses among patients in CR are attributed to the persistence of MRD, but knowledge about the impact of MRD in patients with SR and HR cytogenetics, treated with modern therapies and monitored with next-generation techniques, is limited. Similarly, there is virtually no data about in vivo mechanisms of resistance in SR and HR MM; however, since MRD represents those very few cells that are resistant to treatment, it could be hypothesized that profiling MRD cells may shed light into the mechanisms of resistance in both SR and HR patients. Aim: To determine the clinical impact of MRD in MM patients with SR vs HR cytogenetics, and to identify transcriptional mechanisms determining MRD resistance by investigating the transcriptome of MRD cells in both patient subgroups. Methods: This study was conducted in a series of 390 patients enrolled in the PETHEMA/GEM2012 trial (6 induction cycles with VRD followed by ASCT and 2 courses of consolidation with VRD). FISH was analyzed on CD138 purified PCs at diagnosis. MRD was predefined to be prospectively assessed following induction, transplant and consolidation, using next-generation flow (NGF) according to EuroFlow. In 40 patients [28 with SR and 12 with HR cytogenetics: i.e., t(4;14), t(14;16) and/or del(17p)], diagnostic and MRD tumor cells persisting after VRD-induction were isolated by FACS according to patient-specific aberrant phenotypes. Due to the small number of sorted MRD cells (median of 25,600) we used a 3' end RNAseq method optimized for generating libraries from low-input starting material (MARSeq). Differential expression analyses were performed with DESeq2 R package. Results: At the latest time-point in which MRD was assessed, MRD-positive rates progressively increased (p =.006) from SR patients (148/300, 49%) to cases with t(4;14) (24/42, 57%) and del(17p) (29/38, 76%). Furthermore, MRD levels were significantly superior in patients with del(17p) compared to SR FISH (0.02% vs 0.006%, p =.009), while MRD levels in patients with t(4;14) (0.004%) were similar to those in SR MM. Only 10 patients had a t(14;16) and 4 were MRD-positive. Among patients achieving MRD-negativity ( 〈 2x10-6), 3-year progression-free survival (PFS) rates were similar for those with SR FISH, t(4;14) and del(17p) (90%, 100% and 89%; p 〉 .05). Conversely, 3-year PFS rates for MRD-positive patients decreased from those having SR FISH to those with t(4;14) and del(17p) (59%, 46% and 24%, respectively), with statistically significant differences between the first and the latest subgroups (p 〈 .001). Since clearance of MRD notably lowered the risk of relapse and persistence of MRD significantly shortened the PFS in each cytogenetic group (p ≤.001), we investigated the unique features of MRD cells persisting after VRD-induction by comparing their transcriptome to that of patient-matched tumor cells at diagnosis (n=40). Accordingly, MRD cells showed 763 genes significantly deregulated (Padj 〈 .05), including a cluster of proteasome subunits and proteasome related genes (i.e. PSMB5, PSMC3IP, BTRC, HUWE1, FBXL20 and TRIM69). Gene set enrichment analysis unveiled biologic determinants of MRD resistance such as the IL6-JAK-STAT signaling pathway in SR patients and the ROS pathway in HR patients (FDR 〈 0.1). Interestingly, the number of genes deregulated in MRD cells of SR patients was 9-fold higher than HR cases suggesting that, whereas in SR MM, a few tumor cells with specific gene regulatory networks may have higher probability to persist VRD induction, the presence of HR cytogenetic alterations is associated per se, with a transcriptional program that allows a few MRD cells to persist treatment. Conclusions: This is one of the largest studies integrating patients' cytogenetics and MRD status. Our results, based on intensive treatment and MRD monitoring using NGF, unveil that achieving MRD-negativity may overcome the poor prognosis of HR cytogenetics. By contrast, persistent MRD significantly reduces PFS rates, particularly in patients with del(17p). Interestingly, MRD cells from SR and HR patients may have different transcriptional mechanisms leading to VRD resistance, and further understanding of these could provide knowledge on how to eradicate MRD in both patient subgroups. Disclosures Puig: Takeda: Consultancy, Honoraria; Celgene: Honoraria, Research Funding; Janssen: Consultancy, Honoraria, Research Funding. Garcia-Sanz:Affimed: Research Funding. Martinez-Lopez:BMS: Research Funding; Pfizer: Research Funding; Vivia: Honoraria; Celgene: Honoraria, Research Funding; Janssen: Honoraria, Research Funding; Novartis: Research Funding. Oriol:Celgene: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Amgen: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Takeda: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Janssen: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau. Rios:Amgen, Celgene, Janssen, and Takeda: Consultancy. De La Rubia:Ablynx: Consultancy, Other: Member of Advisory Board. Mateos:GSK: Consultancy, Membership on an entity's Board of Directors or advisory committees; Abbvie: Consultancy, Membership on an entity's Board of Directors or advisory committees; GSK: Consultancy, Membership on an entity's Board of Directors or advisory committees; Takeda: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Celgene: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Amgen: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Janssen: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Amgen: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees. Lahuerta:Janssen: Honoraria, Membership on an entity's Board of Directors or advisory committees; Celgene: Honoraria, Membership on an entity's Board of Directors or advisory committees; Amgen: Honoraria, Membership on an entity's Board of Directors or advisory committees; Takeda: Honoraria, Membership on an entity's Board of Directors or advisory committees. Bladé:Janssen: Honoraria. San-Miguel:Amgen: Honoraria; BMS: Honoraria; Novartis: Honoraria; Sanofi: Honoraria; Celgene: Honoraria; Roche: Honoraria; Janssen: Honoraria.

Abstract: Background: MRD is an established biomarker to evaluate treatment efficacy, define patients at risk based on persistent MRD, and eventually, act as surrogate for prolonged survival based on sensitive MRD-negative definitions. Accordingly, the IMWG has developed criteria for MRD-negativity defined by next-generation sequencing, NGF or PET/CT, and has recommended their inclusion in clinical trials. Notwithstanding, most flow cytometry results have been obtained using less sensitive methods and in fact, there is no data about the impact of NGF-based MRD assessment in clinical trials. Aim: To define the feasibility, sensitivity and clinical impact of NGF-based MRD assessment in the phase III PETHEMA/GEM2012 trial. Methods: A total of 458 patients were enrolled into the PETHEMA/GEM2012 trial. MRD was predefined to be prospectively assessed at three time-points: after six induction cycles with bortezomib, lenalidomide, and dexamethasone (VRD), after HDT/ASCT, and after two courses of consolidation with VRD. MRD monitoring was performed blinded for clinical outcomes in four PETHEMA/GEM laboratory cores, and data was centralized for MRD analyses. MRD assessment was performed following EuroFlow SOPs in a total of 1,134 bone marrow (BM) samples from 419 patients. The 39 cases without MRD assessment had suboptimal response to induction and were thus considered as MRD+ for intention-to-treat analyses. Noteworthy, in 14 BM samples with undetectable MRD, B-cell precursors, erythroblasts and mast cells represented & lt;0.01% of BM cells, and these samples were thus considered as hemodiluted and inadequate for MRD assessment. The limit of detection (LOD) was determined for each of the 1,117 BM samples representative for MRD assessment, according to the formula: (20/nucleated viable cells) x 100; the median LOD achieved by NGF in the PETHEMA/GEM2012 trial was of 3x10-6. Results: Overall, 225/458 (49%) patients had undetectable MRD at the latest time-point in which MRD was assessed and were thus classified as MRD-. Conversely, 233/458 (51%) cases remained MRD+: 28% with ≥10-4 MRD, 12% with 10-5 MRD, and 11% with 10-6 MRD. Detailed analyses of MRD kinetics in 320 patients with available MRD results at all three time-points, showed that the percentage of MRD- patients increased from 35% into 54% and 58% after induction, HDT/ASCT and consolidation, respectively. Furthermore, a restricted analysis among MRD+ patients showed that whereas after induction only 8% of them had MRD levels as low as 10-6, subsequent intensification with HDT/ASCT and consolidation could reduce MRD levels down to 10-6 in 32% of MRD+ cases. Progression-free survival (PFS) rates at 3-years were of 92%, 70%, 54% and 44% for patients being MRD-negative, MRD+ 10-6, 10-5 and ≥10-4, respectively (P & lt;.001; Figure). Thus far, only 6/225 (3%) MRD- patients have relapsed; strikingly, all 6 cases had extramedullary plasmacytomas at diagnosis, all relapsed with extramedullary plasmacytomas, and only 2 had concomitant serological relapse. The favorable outcome of MRD- patients encouraged us to investigate the impact of MRD negativity in both standard- and high-risk patients defined by FISH [i.e.: t(4;14), t(14;16), and/or del(17p)]. Even though MRD- rates were significantly inferior in patients with high- vs standard-risk FISH (37% vs 50%, respectively; P=.03), 3-year PFS rates were similar between patients with high- and standard-risk FISH reaching MRD-negativity (94% and 91%, respectively; P=.56); by contrast, MRD+ cases with high- and standard- risk FISH had median PFS of 27 and 35 months, respectively (P=.025). Conclusions: This is the largest study of MRD monitoring in MM based on the total number of samples analyzed (n=1,134). Our results show that NGF-based MRD assessment is feasible in large multicenter clinical trials, is highly-sensitive, and allows the identification of hemodiluted BM samples inadequate for MRD assessment. Risk of relapse among MRD-negative patients was remarkably reduced (3%), and was particularly related to the reappearance of extramedullary plasmacytomas, which urges the need for combined cellular and imaging MRD monitoring in these patients; by contrast, even MRD levels as low as 10-5 and 10-6 conferred significantly inferior PFS. Overall, this study defines MRD-negativity as the most relevant clinical endpoint for both standard- and high-risk transplant-eligible MM patients. Figure Figure. Disclosures Paiva: Sanofi: Consultancy, Honoraria, Research Funding; Janssen: Honoraria, Membership on an entity's Board of Directors or advisory committees; Celgene: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Amgen: Honoraria; Merck: Honoraria; Novartis: Honoraria; Takeda: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; EngMab: Research Funding. Oriol: Amgen: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Other: sponsored symposia, Speakers Bureau; Janssen: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Other: sponsored symposia, Speakers Bureau; Takeda: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Other: sponsored symposia; Celgene: Speakers Bureau. de la Rubia: Janssen: Other: Honoraria; Amgen: Other: Honoraria; Celgene: Other: Honoraria. Rosinol: Celgene: Honoraria; Janssen: Honoraria. Mateos: Amgen: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Celgene: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Takeda: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Janssen: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees. Lahuerta: Amgen: Honoraria; Celgene: Honoraria; Janssen: Honoraria. San Miguel: Roche: Membership on an entity's Board of Directors or advisory committees; Bristol-Myers Squibb: Consultancy, Membership on an entity's Board of Directors or advisory committees; MSD: Consultancy, Membership on an entity's Board of Directors or advisory committees; Amgen: Consultancy, Membership on an entity's Board of Directors or advisory committees; Novartis: Consultancy, Membership on an entity's Board of Directors or advisory committees; Takeda: Consultancy, Membership on an entity's Board of Directors or advisory committees; Celgene: Consultancy, Membership on an entity's Board of Directors or advisory committees; Sanofi: Consultancy, Membership on an entity's Board of Directors or advisory committees; Janssen: Consultancy, Membership on an entity's Board of Directors or advisory committees.

Abstract: Undetectable measurable residual disease (MRD) is a surrogate of prolonged survival in multiple myeloma. Thus, treatment individualization based on the probability of a patient achieving undetectable MRD with a singular regimen could represent a new concept toward personalized treatment, with fast assessment of its success. This has never been investigated; therefore, we sought to define a machine learning model to predict undetectable MRD at the onset of multiple myeloma. Experimental Design: This study included 487 newly diagnosed patients with multiple myeloma. The training (n = 152) and internal validation cohorts (n = 149) consisted of 301 transplant-eligible patients with active multiple myeloma enrolled in the GEM2012MENOS65 trial. Two external validation cohorts were defined by 76 high-risk transplant-eligible patients with smoldering multiple myeloma enrolled in the Grupo Español de Mieloma(GEM)-CESAR trial, and 110 transplant-ineligible elderly patients enrolled in the GEM-CLARIDEX trial. Results: The most effective model to predict MRD status resulted from integrating cytogenetic [t(4;14) and/or del(17p13)], tumor burden (bone marrow plasma cell clonality and circulating tumor cells), and immune-related biomarkers. Accurate predictions of MRD outcomes were achieved in 71% of cases in the GEM2012MENOS65 trial (n = 214/301) and 72% in the external validation cohorts (n = 134/186). The model also predicted sustained MRD negativity from consolidation onto 2 years maintenance (GEM2014MAIN). High-confidence prediction of undetectable MRD at diagnosis identified a subgroup of patients with active multiple myeloma with 80% and 93% progression-free and overall survival rates at 5 years. Conclusions: It is possible to accurately predict MRD outcomes using an integrative, weighted model defined by machine learning algorithms. This is a new concept toward individualized treatment in multiple myeloma. See related commentary by Pawlyn and Davies, p. 2482