2025 EMN共识声明：功能性高危多发性骨髓瘤

ABSTRACT

Multiple myeloma(MM) is an incurable blood cancer characterized by clonal bone marrow plasmacytosis, hypercalcemia, renal failure,anemia, and osteolytic bone disease.Approximately 20% of NDMM patients, not predicted to have high-risk disease at di-agnosis, progress early, despite optimal induction+/-ASCT and lenalidomide maintenance, and are subsequently categorized as functional high-risk (FHR) disease. Standardized risk-stratification models incorporate biomarkers of tumor burden, existence of high-risk cytogenetics, with the presence/absence of plasma cell leukemia/extramedullary disease to attribute high-risk at di-agnosis; however, depth/duration of response to novel agent-based induction (NA-IND) as dynamic markers of disease risk have not been defined. However, irrespective of diagnostic risk-stratification, response to NA-IND may be the single most effective method of identifying patients whose FHR biology portends an unacceptably short overall survival (OS). In this EMN consensus statement, we define FHR-MM as disease progression within 18 months of commencement of first-line therapy in the absence of high-risk cytogenetics, discuss the underlying disease biology, and strategies to improve outcomes for these patients.

1| Introduction

In 2023, the International Myeloma Society convened an ex-pert panel to establish a consensus definition of high-risk MM as the presence of del(17p), with a cutoff of> 20% clonal frac-tion, and/or TP53 mutation; t(4;14), t(14;16), t(14;20), together with gain(1q21) and/or deletion(1p); monoallic del(1p32)and gain(1q) or bi-allelic del(1p32)[1]. Alternatively, the co-existence of 2 or more high-risk cytogenetic abnormalities as demonstrated by the Medical Research Council(MRC)

Myeloma IX and XI trials and more recently in a smaller UK real-world cohort receiving standard of care treatment simi-larly identifies patients as having ultra-high-risk MM[2-4]. A variation on this theme was described by the Myeloma Genome Project whereby biallelic inactivation of TP53 or amplification of CKS1B(1q21) on a background of International Staging System(ISS) Stage 3 disease was associated with a median overall survival of only 21 months[5]. It is widely recognized however, that these methods of risk-stratification may fail to identify FHR-MM, with a recent post hoc analysis of pooled

data from the pivotal CASTOR and POLLUX trials demon-strating that of patients enrolling in these darumatumab(D)-based studies after relapsing within 18 months of first-line therapy, few presented with any high-risk cytogenetics(D arm, 24.1%; control arm 24.6%) or ISS stage III disease(D arm, 11.7%; control arm, 18.3%)[6]. This subset of NDMM is associated with an unacceptably short OS and can only be re-liably identified following dynamic clinical assessment of the duration of response to front-line therapy[7]. So while there is an association between markers of disease burden(hi bone marrow plasma cell percentage, cytopenias, hypercalce-mia, renal insufficiency, high LDH) and advanced stage(ISS stage II/III and R-ISS stage III) with the presence of high-risk cytogenetics and early relapse, many patients defined as“standard-risk” at diagnosis will relapse within 18 months of commencement of front-line therapy[8-10]. Moreover, there are additional factors that compound the complexity of FHR disease. First, there is no currently accessible diagnostic assay that reliably identifies all cases of FHR disease at diagnosis to enable a risk-adapted therapeutic approach. Second, there is a paucity of published clinical data focused on FHR disease,leading to reliance on post hoc analyses derived from random-ized studies in early line relapsed and/or refractory(RR)MM to guide therapy. And while some of these analyses have been performed according to the proposed definition of FHR, oth-ers have not, producing data that cannot be extrapolated to guide management of this adverse-risk group, highlighting the unmet need for a uniform consensus definition of FHR.Furthermore, there is significant heterogeneity in standard front-line induction for both transplant eligible and ineligible patients in the real world, such that clinicians are faced with patients who are refractory to differing novel agents at the time of FHR disease occurrence.

2| Definition

While in its broadest terms, FHR disease encompasses both sub-optimal response to therapy and early progressive disease(in-clusive of primary refractory disease), we propose that the term FHR be confined to patients with the worst outcome, so those exhibiting early progressive disease be best defined, based on currently available data, as relapse within 18 months of initia-tion of any first-line therapy. This therefore represents the most age-inclusive definition that would inform studies into both the underlying biology of FHR and strategies to improve outcomes for FHR patients. Whilst it is acknowledged that others have defined FHR disease as relapse within 12-24 months of high-dose chemotherapy(HDT) conditioned front-line ASCT, this is highly problematic as it fails to capture transplant-ineligible(TIE) NDMM patients, likely in reality representing the major-ity of the FHR population, nor those TE-NDMM patients who demonstrate inadequate initial disease control owing to pri-mary drug resistance during or immediately after completion of induction therapy but prior to upfront ASCT, who then pro-ceed to alternative salvage therapy, foregoing ASCT. Moreover,given that conventional front-line ASCT is typically performed within 6 months of treatment initiation, relapse occurring within 12 months of ASCT would largely be within 18 months of treatment initiation. It is important to acknowledge that with the introduction of newer therapies, this proposed definition

of early progression will likely be extended beyond 18 months;however, the underlying disease kinetics, exemplified by early therapeutic resistance, that defines this subtype of MM will re-main unchanged.

Our proposed definition is based on 4 large landmark analy-ses of real-world and clinical trial data. The first, a landmark analysis from the Australian and New Zealand Myeloma and Related Diseases Registry(MRDR) of 1320 real-world NDMM patients, where early progression was defined as progression within 12 months of commencing front-line bortezomib-based induction(V-IND). In this study, 8.9%(118/1320) of patients experienced early progression and markedly inferior OS,compared to late progression(>12 months), median 20.2 vs.60.7 months(p<0.001) respectively, but also established a link between early progression, suboptimal response(p<0.001),de-fined as<MR after 2 induction cycles or<PR after 4 induction cycles, and primary refractory disease(p<0.001)[7]. Second,Corre et al.[11] in a landmark analysis of 2474 NDMM patients identified early progression≤18 months(n=496, 20%) to be independently associated with reduced OS despite intensive NA-based induction(HR 4.40,p<0.0001) and, consistent with the MRDR analysis, identified best response≥VGPR to be pro-tective, with lower risk of early progression(p<0.001). Third,an analysis of 926 NDMM patients enrolled in the prospective observational CoMMpass study and described by D'Agostino et al.[12] identified early progression≤18 months(n=191,20.6%) to be associated with lower ORR(p<0.001), depth of response(p<0.001) and refractoriness to PI(p<0.001) or PI and IMIDs(p<0.001). Landmark survival analysis validated the 18-month cut-off time point, demonstrating significantly worse OS in patients with early progression compared to those with later progression(p=0.004). Lastly, a recent landmark survival analysis from the Harmony Alliance platform of 9740 NDMM patients eligible for inclusion in 1 of 16 inter-national clinical trials identified 20.2%(n=1970) with early disease progression within 18 months of MM diagnosis and$$\text{ significantly inferior OS of} 48\text{ vs.} 115\text{ months for those with-}$$out early relapse, HR 3.32, p< 0.0001; but also highlighted a reduction in HFR disease occurrence with PI-IMID combina-tion therapy vs. PI or IMID alone(OR 0.48, p<0.0001)[13].Similarly, a recent updated analysis of 2682 real-world NDMM patients from the MRDR using an 18-month cut-off for early progression showed a statistically significant reduction in the incidence of FHR disease following PI-IMID induction with VRd, compared to VCD, 23% vs.28%, respectively(p=0.0011),with a median OS for FHR patients of approximately 3 years(MRDR, unpublished data) In keeping with these findings, an analysis of 3932 NDMM patients from the HONEUR network demonstrated a variation in the incidence of FHR(n=1022,26%), defined also as early progression within 18 months of commencement of first-line therapy, according to induction regimen: 32% following VCD/VMP, 24% following IMiD+PI combination(p<0.00001) and 13% following D-based induc-tion. Median OS was 34.0 vs. 73.5 months for FHR and con-trols, respectively(p<0.0001)[14].

Although long-term survival data is pending, primary analysis of the PEREUS study of D-VRd vs. VRd induction in NDMM also suggests a potential reduction in the incidence of early dis-ease progression/FHR disease, with an estimated 48-month

TABLE 1| FHR-specific studies.

||FHR definition|# prior LOT|N|Regimen|Clinical outcome| |---|---|---|---|---|---| |ALLG MM17 (ACTRN12615000934549)|Primary refractory to VCD induction, defined as progressive disease during or within 60 days of last dose of front- line therapy|1|15 (30%)|4-6 cycles of KTd-ASCT-2 cycles of KTd-consolidation-Td maintenance|● 73% achieved ≥ PR prior to ASCT ● 33% were MRD(10-5) negative following 12 months of Td maintenance on ITT analysis| |ALLG MM21 (ACTRN12618001490268)|Primary refractory to VCD induction, defined as progressive disease during or within 60 days of last dose of front- line therapy|1|14 (28%)|4 cycles of Dara- Rd-ASCT-12 cycles Dara-Rd consolidation-R|● 64% achieved ≥ PR prior to ASCT ● 21% were MRD negative following 12 cycles of DRd consolidation| |MyDrug (NCT03732703)|Progressive disease within 18 months of 1L therapy in the TIE population and 36 months if received front-line ASCT and maintenance therapy|1-3|38|Dara-Ixa-Pom-dex|● ORR 92% ● ≥VGPR 50%| ||||17|Selnexor-Ixa-Pom-dex|● ORR 52.9%| |||||Pom-dex|● Median PFS 10.2 months|

PFS rate of 84.3% and 67.7% for D-VRd and VRd, respectively(p<0.0001)[15]. Taken together with real-world data, these re-sults support the notion that front-line CD38mAB-PI-IMID qua-drplets, where available, may delay the time to first progression in patients with FHR disease.

3| Current Landscape of FHR Disease

Despite the reported incidence of FHR disease in~20% of all NDMM when utilizing an 18-month cut-off, there have only been a few published clinical trials prospectively evaluating treat-ment options for these patients(Table 1). Conversely, several post hoc analyses of randomized studies, namely FIRST, CASTOR,POLLUX, and CANDOR, have described the outcomes of patients with 1 prior line of therapy(1LOT) who have relapsed within 18-24 months of commencement of induction, while others have included more heterogeneous patient populations(Table 2). Sub-analyses of ENDEAVOR, ASPIRE, IKEMA defined early relapse as within 12 months of the most recent line of therapy but were not restricted to patients with 1 LOT, including populations with 1-3 prior lines of therapy. This proposed consensus definition of FHR NDMM aims to ensure a consistent approach in these sub-analyses to enable data to be extrapolated to the real world.

4| Post Hoc Analyses of FHR Disease

4.1| Lenalidomide-Refractory FHR Disease

Lenalidomide(LEN)-refractory disease, defined as pro-gressive disease while on LEN or within 60 days of LEN cessation, has an ever-increasing prevalence in many

jurisdictions in real-world MM management. The IFM 2005-02, DETERMINATION, SWOG S0777, and FIRST trials have informed the adoption of LEN both pre-ASCT as part of bortezomib-LEN-dexamethasone(VRd)-based induction and as post-ASCT maintenance, and as part of VRd or LEN-dexamethasone(Rd) front-line therapy in the TIE-NDMM patient. There is, however, a paucity of data to guide the man-agement of LEN-refractory, FHR disease. The Phase 3 FIRST study, published in 2014, established Rd as a standard of care for induction in TIE-NDMM. This study compared continuous Rd with limited Rd for 18 cycles and melphalan-prednisolone-thalidomide(MPT) in 1623 patients. A recent post hoc analysis of Rd patients who experienced early progression, defined as<12 months(n=203) from treatment commencement vs. later relapse(n=186), demonstrated inferior outcomes with early progression, median OS(95% CI) of 26.8 months(21.9-32.8)versus 63.9 months(57.0-78.0), median OS from disease pro-gression to death 19.9 months versus 36.4 months, and median PFS from randomization to second progression(PFS2) 19.1 versus 42.1 months, respectively[16].

The MRC Myeloma XI trial was a Phase 3 response adaptive trial of 3894 NDMM patients. Subsequent post hoc analysis of 174(12.9%) patients who relapsed within 12 months of high-dose melphalan conditioned ASCT demonstrated markedly inferior survival outcomes characterized by a lack of salvageability de-spite NA-based second-line therapy, with a median PFS2 of only 18 months and OS of 26 months, compared to PFS2 of 85 months and OS of 91 months for those not experiencing early progres-sion[10]. Importantly, 64.4% of early progression patients had achieved≥VGPR to induction. There were no differences in sub-sequent outcomes, early vs. late relapse, according to whether patients received thalidomide or lenalidomide-based induction

→

TABLE 2 | Post hoc analyses of FHR disease.

||Definition of early progression (EP) versus late progression (LP)|# prior LOT N(EP)|Regimen|Clinical outcome|Early progression|Late progression|||| |---|---|---|---|---|---|---|---|---|---| ||Lenalidomide-refractory disease -FIRST study - phase 3|# prior LOT N(EP)|Regimen|0|Early progression|203|Median OS mo (95 CI)|26.8 (21.9-32.8)|63.9 (57.0-78.0)| |MRC Myeloma XI trial|EP or myeloma-related death < 12 months of high dose melphalan ASCT|0|174 (12.9%)|R-cyclo-d versus T-cyclo-d prior to high-dose melphalan ASCT-R maintenance versus observation|Median OS mo (95 CI)|19.9 (16.0-25.5)|36.4 (27.9-47.0)||| |CD38 mAB|EP < 18 months LP >18 months|1|228|D-Vd versus Vd|Median PFS mo (DVds vs. Vd)|15.4 versus 9.0 months (HR,0.51; p = 0.0488)|27.7 versus 7.9 months (HR, 0.2; p < 0.001)|27.7 versus 0.29 months (HR, 0.2; p < 0.001)|| |CANDOR|EP < 18 months LP >18 months|1|228|ORR (DVds vs. Vd)|79.3% versus 88.9% (13.3 versus 0%|95.6% versus 71.4% 22.8% versus 3.2%|95.6% versus 71.4% 22.8% versus 3.2%||| ||||||MRD neg (10 - 5) vs. Rd) mo:|36.9 versus 11.7 months (HR,0.41; p = 0.0002)|69.3 versus 29.7 months (HR, 0.53; p = 0.0007)||| |American Journal of Hermukology, 2023|CANDOR|EP < 18 months of initiation of front-line therapy or relapse < 12 months of front-line ASCT|1|210|D-Kd versus Kd|ORR (DRdvs.Rd) vs. Rd) vs. Rd) vs. RDs|93.6% versus 76.5% (PFS benefit D-Kd (95 CI) ORR (D-Kd vs. Kd)|93% versus 76.5% (HR 0.6 (0.3-1.2)|93% versus 13.8% HR 0.7 (0.3-1.4)|

References

+/- V-based consolidation(bortezomib-cyclophosphamide-dexamethasone) suggesting the outcome of FHR is likely to be agnostic to the type of PI-IMID induction therapy received.

4.2| CD38 Monoclonal Antibodies

Acknowledging that not all jurisdictions have access to front-line anti-CD38 targeting therapy, post hoc analyses have demonstrated PFS benefit with the use of CD38 targeting monoclonal antibodies in FHR disease. A post hoc analysis of the CASTOR(N=228) and POLLUX(N=289) daratumabum trials evaluated the outcomes of patients who had received only 1LOT according to the timing of relapse from initiation of front-line induction, early(<24 months) or late(>24 months);a secondary analysis, using an 18-month cut-off for early pro-gression, was also performed[6]. In CASTOR, median PFS with daratumabum-Vd(DVd) vs. Vd was 21.2 vs. 7.3 months(HR, 0.32; p<0.0001) in the early progression subgroup(<24 months), and 33.2 vs. 8.0 months(HR, 0.20; p<0.0001)in the late-relapse subgroup. In POLLUX, median PFS with D-LEN-dexamethasone(D-Rd) vs. Rd was 43.7 vs. 11.8 months(HR,0.37;p<0.0001) in the early progression subgroup, and 66.0 vs. 35.5 months(HR, 0.64; p=0.0351) in the late-relapse subgroup. Secondary analysis utilizing an 18-month cut-off was similar to that seen using the 24-month cut-off. So, while the utilization of D combined with V or LEN improved PFS when compared to control groups, it did not overcome the adverse association of early progression with shortened PFS when compared to late relapses; moreover, it should be noted that patients with LEN-refractory disease were excluded from the POLLUX trial and represented only a minority of patients in the CASTOR trial.

The Phase 3 CANDOR study evaluated D-carfizomib-d(D-Kd)versus Kd in RRMM with 1-3 prior lines of therapy(n=466). Of these, most had received prior V(D-Kd, 92% Kd 87%), many had also received LEN(D-Kd 39%, KD 48%) and approximately one-third were LEN refractory(D-Kd 32%; Kd 36%)[17]. A post hoc analysis evaluated outcomes in patients with early progression,defined in patients with 1LOT(n=210), as within 18 months of initiation of front-line therapy or relapse<12 months of front-line ASCT[18]. In the 1LOT subgroup, ORR was 86.4% vs. 57.6%for D-Kd vs. Kd with early progression and 93.9% vs. 88.9%without early progression, respectively. Improvement in PFS for D-Kd in the 1LOT patients was consistent across both early and late relapse subgroups, HR 0.6(95% CI 0.3-1.2) and 0.7(95%CI 0.3-1.4), respectively, supporting D-Kd salvage as a possible option for FHR disease in patients who have not received an-ti-CD38 therapy at first line.

The Phase 3 IKEMA study evaluated isatuximab-Kd(Isa-Kd)versus Kd in RRMM(n=302)[19]. A post hoc sub-analysis eval-uated outcomes in a heterogeneous patient population who had demonstrated early progression, defined in 1 LOT patients as re-lapse<18 months from initiation of front-line therapy or<12 months following front-line ASCT, but also in those with>1 prior line experiencing relapse<12 months from initiation of the most recent line of therapy[20]. In 1LOT patients, Isa-Kd achieved similar rates of minimal residual disease(MRD) neg-ative(<10-5) CR in early and late progressors, 30.0% vs. 34.5%.

Moreover, the depth of response benefit with Isa-Kd, regardless of the timing of relapse, was consistent in patients who were re-fractory to an IMID or PI. Although specific analysis of the effi-cacy of Isa-Kd in 1 LOT FHR patients was not reported, possibly owing to insufficient numbers, given that the depth of response,particularly MRD negativity, correlates with improved PFS and OS, this regimen warrants further consideration in LEN refrac-tory FHR disease[21-23].

4.3| Chimeric Antigen T-Cell Therapy(CAR-T)

The phase 3 CARTITUDE-4 study compared BCMA-targeted CAR-TCiltacabene autoleucel(cilta-cel) with either standard-of-care(SOC) regimes of D-pomalidomide-dexamethasone(D-Pd) or pomalidamide-bortezomib-dexamethasone(PVd), in lenalidomide-refractory patients with 1-3 prior lines of treat-ment. A post hoc subgroup analysis of patients with 1 prior line of therapy only(n=136, comprising 68 cilta-cel and 68 SOC patients), of which 58% were classified as having FHR disease(79/136), defined as early progression within 18 months of ASCT or in those who did not receive ASCT within 18 months of com-mencement of induction, was presented at the 2024 American Society of Clinical Oncology Annual Meeting(ASCO). Patients who received cilta-cel demonstrated superior ORR, 87.5%vs. 79.5%(p=0.01 95 CI, 1.3-8.4), MRD negativity 65% vs.10.3%(p<0.001 95 CI, 4.8-55.1), median PFS not reached vs.11.79 months(HR0.27(0.12-0.60) p=0.0006) and 12-month PFS rates of 77% vs. 49.1% compared to SOC, cilta-cel vs. SOC,respectively. Furthermore, there was no difference in ORR, me-dian PFS, 12-PFS rate(88.1 vs. 88.0 months, no-FHR vs. FHR,respectively) within the subset of patients with 1 prior LOT who received cilta-cel(n=60) whether they exhibited FHR disease or not, although clearly longer follow-up is needed.

5| Prospective Studies of FHR Disease

5.1| Chimeric Antigen T-Cell Therapy(CAR-T)

Cohort B of CARTITUDE-2, a phase 2 multi-cohort study of clita-cel included RRMM patients who had received PI-IMID triplet front-line induction and experienced early progression,defined as≤12 months from ASCT or≤12 months after the start of induction therapy in those who did not undergo ASCT. At a median follow-up of~29 months, 19 patients, of whom 79% were refractory to the last LOT, 16% triple-class refractory, and 79%with prior ASCT, had received cilta-cel. Despite this, cilta-cel achieved deep and durable responses, achieving MRD negativ-ity at 10-5 in 93% that was sustained≥6 months in 53%. Median PFS was not reached, and the 24-month PFS rate was 73%[24].

Similar findings were reported from Cohort 2a of KarMMa-2, a multi-cohort Phase 2 trial of BMCA-directed CAR T cell ther-apy with iecdatabagene vicleucel(ide-cel) that evaluated efficacy in patients(N=37) who had relapsed<18 months from initia-tion of front-line therapy. At a median follow-up of 21.5 months,ORR was 83.8% with MRD negativity of 70% at 12 months. Of these patients, 71% maintained MRD negativity≥18 months.Median PFS was 11.4 months with median OS not reached and with a similar incidence of cytokine release syndrome and

neurotoxicity to that seen in patients treated with ide-cel with more advanced stages of disease[25].

5.2| Bispecific Antibodies/T-Cell Engager Antibodies

Thus far, there have been no prospective studies or post hoc analyses evaluating bispecific antibodies/T-cell engagers in FHR MM.

6| Other FHR Studies

The ALLG MM17 trial(ACTRN12615000934549) evaluated se-quential therapy with 4 to 6cycles of carfilzomib-thalidomide-dexamethasone(KTd) salvage, ASCT, 2 cycles of KTd consolidation, and Td maintenance for 12 months in 50 patients who were primary refractory or had a suboptimal response,defined as<MR after 2 cycles or<PR after 4 cycles of VCD induction. The ORR was 78% with 34% Euro-flow MRD nega-tivity(10-5) in the intention-to-treat(ITT) population and 65%)in the evaluable population post 12 months of Td maintenance[26]. Further analysis of only primary refractory patients, N=15(30%), defined as having demonstrated PD during or within 60days of the last dose of 1L therapy and thereby fulfilling our proposed definition of FHR, demonstrated that 73% of patients achieved≥PR prior to ASCT and 33% were MRD negative fol-lowing 12 months of Td maintenance on ITT analysis.

Similarly, the ALLG MM21 trial(ACTRN12618001490268)evaluated 4 cycles of D-LEN-dexamethasone(D-Rd) salvage fol-lowed by ASCT and 12 cycles of D-Rd consolidation followed by R maintenance in 50 patients who were primary refractory or had a suboptimal response to VCD induction. ORD was 72% and Euro-flow MRD negativity(10-5) 46% in the ITT population and 79% in the evaluable population following 12 cycles of D-Rd con-solidation[27]. Further analysis of primary refractory patients only(N=14,28%) demonstrated that 64% achieved≥PR prior to ASCT and 21% were MRD negative following 12 cycles of D-Rd consolidation on ITT analysis.

MyDrug(NCT03732703) is an open-label, genomically-guided platform study evaluating a heterogenous patient population,including those who had received 1-3 prior therapies. Early pro-gression was defined as progression within 18 months of first line therapy in the TIE-NDMM population and 36 months if front-line ASCT and maintenance therapy had been received.The study is evaluating the addition of molecularly-targeted agents with an ixazomib-pomalidomide-dexamethasone(IPd)backbone for patients identified to have actionable mutations in the bone marrow as detected by next-generation sequenc-ing(NGS): KRAS/NRAS/BRAF(cobinetinib), FGFR3(enas idenib), IDH2(erdafitinib), CDKN2C(abemaciclib) or t(11;14)(venetoclax). Patients with no actionable mutation or mutational allelic burden< 25% were randomly assigned to one of the‘non-actionable' arms, of which Y1(Dara-IxaPd) and Y3(Selinexor-IPd) have been reported. Y1(n=38) achieved an ORR of 92.1%with 50% achieving a VGPR[28]. Y3(n=17) demonstrated an ORR of 52.9% with a median PFS of 10.2 months[29]. The first actionable cohort with KRAS/NRAS/BRAF mutations has 6

evaluable patients and reported cobimetinib-IPd to be safe in this population[30].

A recent phase I study of the anti-BCMA CAR-T Trocovabtagene autoleucal(trovo-cel, C-CAR088, NCT05632380) as consolida-tion, infused on day 3 post-ASCT demonstrated a favorable effi-cacy and safety profile in patients with ultra-high risk RRMM,in this instance defined as(1) genetic ultra-high risk: del(17p)≥60%; or≥2 high-risk cytogenetic abnormalities, and<CR after 4 cycles of first-line standard PI-IMID triplet induction;(2)'primary refractory' but consistent with our definition of subop-timal response:< MR after 2 cycles or< PR after 4 cycles of first-line standard induction;(3) early relapse: disease progression within 6 months from standard first-line therapy;(4) primary plasma cell leukemia and< CR after 4 cycles of first-line ther-apy. Of 12 patients recruited thus far, 50% were considered sub-optimal responders and 16.7% had demonstrated early relapse.The ORR was 100%, 11(91.7%) patients were in MRD negative CR, with 2 and 4 patients maintaining response at 18 months and 12 months post-ASCT, respectively. Median PFS and OS have not been reached at a median follow-up of 9.5 months from ASCT[31].

7| Disease Biology

7.1| Genomic Characterization

Whilst the identification of a high-risk gene expression profile by way of validated assays(GEP70/UAM70, EMC92/SKY92)is associated with a greater risk of early progression, these as-says alone cannot reliably detect FHR disease. A recent explor-atory analysis of ultra-high risk NDMM patients treated on the OPTIMUM/MUKnine(NCT03188172) trial reported that those with 3 or more of t(4;14), t(14;16), t(14;20), gain(1q), del(1p),del(17p), or del(17p) were associated with an increased risk of early progression, defined as within 18 months of treatment ini-tiation, despite intensive induction with D-cyclophosphamide-RVDh[32].

Conversely, an analysis of NDMM patients from the Multiple Myeloma Research Foundation CoMMpass dataset(N=512),utilizing a panel of established gene expression profiles of high-risk disease(EMC92, HMCL17, IFM15, UAMS70 and UAMS80) identified that most FHR patients did not demon-strate a high-risk profile. This analysis first divided patients into 3 groups:(1) genomic high-risk(GHR): t(4;14) or t(14;16)or complete loss of functional TP53(bi-allelic deletion of TP53 of mono-allelic deletion of 17p13(del17p13) and TP53 muta-tion) or 1q21 gain and ISS Stage 3;(2) FHR: refractory to in-duction therapy or early progression within 12 months but no genomic high-risk markers; and(3) standard-risk group who did not fulfill the criteria of GHR or FHR. There were no sig-nificant differences in gene expression profile between SR and FHR, nor between GHR and FHR. FHR patients did, however,demonstrate generally higher indices of chromosomal insta-bility(chromosomal instability index from sarcoma study CINSARC, centroson index CI and CIN70) and tumor ag-gression(proliferation index and PR, a gene expression profile associated with proliferation genes). Furthermore, transcrip-tomic, mutational signature, copy number, and somatic

mutational analysis did not identify a specific genomic'finger-print' distinguishing FHR from GHR and SR, possibly impli-cating external factors, such as immune microenvironmental dysregulation, to be of importance in FHR disease. However,mutational analysis did identify mutations affecting the IL-6/JAK/STAT3 pathway, associated with aberrant mitosis and DNA damage response, to be significantly enriched in FHR.Comparative RNA sequencing of the top 200 differentially ex-pressed genes in the 3 groups identified FHR to be enriched for genes linked to mitotic cell cycle and DNA replication,C2H2 zinc finger, and DNA repair[33, 34].

A study of 41 NDMM who either failed to respond or expe-rienced early progression(defined as<18 months) after a bortezomib-containing induction regimen(49% VCD, 17%VTD/VCDT, 22% RVD, 5% VD-PACE) utilized longitudinal single-cell RNA-sequencing analysis to identify a unique mo-lecular resistance signature(referred to as primary refractory MM-PRMM) characterized by hypoxia tolerance, protein folding,and mitochondrial respiration, with a very high clinical predictive value that was later validated using the CoMMpass database. Highly expressed genes included COX6C, COX7A2 associated with mitochondrial stress, endoplasmic reticulum,and unfolded protein-stress response(UPR) pathway genes PPIA, STMN1, oxidative stress SOD1, TXN, and the protea-some machinery genes PSMB4 and PSMA2. Survival analysis on CoMMpass patients with high expression of this PRMM molecular signature, compared to the rest of the population,revealed a HR of 3.9(2.22-6.87) with p=4.57x10-17, being mindful that although early progression was predicted by the PRMM signature, early progression still occurred in 41% of patients with the low-risk signature. Subsequently, the anal-ysis was repeated to separate responders and non-responders to D-KRD salvage, defining 2 major clusters of a total of 133 differentially expressed genes. Interestingly, CD38 was down-regulated, while proteasome genes PSMD4 and PSMB4 were upregulated, possibly underlying PI resistance, as were mito-chondrial stress genes, potentially implicating disruption of mitochondrial respiration as a factor promoting survival of malignant plasma cells[35].

Whilst these findings provide important insights into the under-lying FHR biology and potential therapeutic targets for future research, these assays remain unavailable. An accessible, com-mercially available method to reliably identify all FHR disease in a timely manner is needed.

7.2| Immune Cell Biology

Several groups have identified an association between im-mune reconstitution following ASCT and time to first relapse.The absolute lymphocyte content of the reinfused autologous graft(A-ALC) was first linked to higher post-transplant circu-lating lymphocyte counts and improved outcomes by Porrata et al.[36] in 267MM patients in 2014. An A-ALC≥0.5x109/kg vs.<0.5x109/kg predicted longer median post-transplant OS and time to progression, 58 vs.30 months, p=0.00022 and 22 vs.15 months,p<0.00012 respectively[36]. Natural killer cell reconstitution in particular, occurring early post-transplant, was subsequently identified to predict improved PFS. Rueff

et al. performed a retrospective flow cytometric analysis of CD4+B, CD8+T, and NK lymphocyte subset recovery in 114MM patients undergoing ASCT identifying a significant association between NK cell recovery 1 month post-ASCT and PFS but not OS[37]. Median PFS was 2.7 months com-pared to 15.8 months for patients with NK cell count below 100cells/μL vs.100-200cells/μL and for NK cells>200cells/ $\mu L$ ,10.8 months; p<0.001.

We now understand that the expansion of immunosuppressive regulatory T cells(T-reg) and CD8+ T effector cell exhaustion/sescence, defined by the expression of inhibitory receptors TIGIT, PD-1, LAG-3, TIM-3, and downregulation of DNAM-1,is a major determinant of immunological escape, MM relapse,and PFS post-ASCT[38-40]; however, only a few analyses have focused on patients with FHR disease. One study of 79 patients evaluated changes in circulating CD4+FOXP3+T-reg cells by flow cytometry at 4 time points: before high-dose chemotherapy, engraftment, 6, and 12 months post-ASCT[41].Immunosuppressive CD4+FOXP3+T-reg cells were increased at the time of engraftment in patients who developed progres-sive disease within 12 months of ASCT(n=10) compared to non-progressors(n=50), 6.7% vs. 4.9%; p=0.026. Chung et al.[39] performed a retrospective analysis of immune reconstitu-tion in 55MM patients who had received bortezomib and/or LEN induction prior to ASCT and post-transplant LEN main-tenance Compared to patients who remained in continuous CR 12 months post-ASCT, patients who achieved a CR but sub-sequently relapsed early after ASCT had significantly higher percentages of exhausted/senescent T cells at relapse, but also at the 3-month mark, prior to clinical relapse. Exhausted/se-nescent T cells were characterized by downregulation of the costimulatory marker CD28, upregulation of CD57 indicating low proliferative capacity, and increased expression of PD-1.Moreover, the fold change in Tregs was significantly greater at relapse compared with non-relapsed patients at+12 months post-ASCT,1.65+/-0.25 vs.0.94+0.13(p<0.05) respectively.Lee et al.[38] correlated PFS, defined as the time from ASCT to relapse, with changes in specific T cell subsets, or mark-ers of immune regulation, activation, and proliferation(PD-1,LAG-3, ICOS, CTLA-4, GzmB and Ki-67) on T cell subsets at D+100 post-ASCT. Patients whose BM contained higher fre-quencies of Tregs, greater expression of PD-1 on CD4 effector cells, and LAG-3 on CD4 effector, CD8, and Treg cells, respec-tively, were at greatest risk of early relapse(p=0.031). Median PFS was 18 vs. 36 months for CD4PD-1high vs. CD4PD-1low(p=0.0042); 18 vs. 53 months for CD4LAG-3high vs. CD4LAG-3low(p=0.003) and 20 vs. 53 months for CD8LAG-3high vs.CD8LAG-3low, respectively.

Mass cytometry by time of flight(CyTOF) was performed on BM from 58 patients at day+60 and+100 post-ASCT, identi-fying a subset of patients who, at the time of immune reconsti-tution, had T cell subsets enriched with naive and terminally differentiated T cells and low-level TIGIT expression indic-ative of T-cell exhaustion, with inferior OS and time to pro-gression from ASCT, irrespective of high-risk FISH[42]. The presence of T cells at the opposite ends of the differentiation spectrum may suggest an inability of these T cells to engage and expand in response to MM antigens liberated following high-dose chemotherapy promoting disease relapse. Together,

these observations suggest that the pattern of immune recon-stitution as early as D+100 post-ASCT may provide predictive biomarkers for early relapse, irrespective of traditional risk stratification. Additional research is required to elucidate a novel biomarker to identify FHR disease, enhance risk strat-ification, and enable a risk-adapted approach to therapy at diagnosis.

8| Induction Regimens in the Real-World

The optimal induction regimen for NDMM continues to evolve.Based on the outcome of the Phase 3 PERSEUS trial, it is antic-ipated that D-VRd will replace VRd as the induction regimen of choice in a limited number of jurisdictions, contributing to the ever-increasing divide in global standards for real-world patients with NDMM. Consequently, clinicians seeking to offer FHR pa-tients the best available second-line treatment are faced with the conundrum of interpreting published data from patients treated with heterogeneous front-line regimens.

Analysis of published real-world data highlights extreme heterogeneity in global trends of induction therapy; more-over, while the use of front-line VRd induction is increasing in Northern America, Western Europe, and Australia, it is not homogenous. At a regulatory level, organizations such as the European Medicines Agency(EMA), Food and Drug Administration(FDA) and Therapeutic Goods Administration(TGA), to name a few, ensure available treatments within their jurisdiction are both safe and efficacious. However, access to novel agent-based induction therapy incorporating prote-some inhibitors(PIs), immunomodulatory drugs(IMIDs),and, in some countries, CD38mAB, is largely limited by cost and availability of funded access. Globally, upfront bortezo-mib in combination with either cyclophosphamide(VCD) or thalidomide(VTD) remains the most commonly prescribed regimens for TE-NDMM. Systematic real-world analyses evaluating the impact of PI-IMID triplet induction and now D-PI-IMID quadruplet induction are needed to determine the effect, if any, on the incidence and clinical outcomes of FHR disease. Moreover, it is unclear whether there remains a role for CD38mAB-based regimens following front-line D-based induction.

9| Approach to the Patient With FHR Disease

9.1| Scoring Systems to Quantify Risk of Early Relapse

Four scoring systems have been proposed to identify patients at high risk of early progression. The Centre for Blood and Marrow Transplant Research(CIBMTR) scoring system,Simplified Early Relapse in MM(S-ERMM) score, Dynamic S-ERMM(DS-ERMM) score, and European Bone Marrow Transplant Early Relapse(EBMT) scoring system are summa-rized in Table 3[43-45]. There are important caveats to con-sider when determining the applicability of these models to clinical practice. First, whilst some of the models were estab-lished following retrospective analyses of real-world registry datasets, others utilized data from clinical trial participants

treated with diverse induction regimens prior to FHR occur-rence, as summarized in Table 4. Second, each scoring sys-tem considers distinct clinical and disease-related variables to quantify risk of early relapse, utilizing a variety of quite different clinical end-points, including 3-year PFS and 3-year OS, median OS and PFS-2, and a percentage estimate of PFS at 12 months post-ASCT. And finally, and perhaps most im-portantly, 2 of the scoring systems were derived entirely from patients receiving ASCT as part of first-line therapy and there-fore likely of limited applicability to the majority of NDMM patients. CIBMTR and S-ERMM incorporate biochemical markers with high-risk cytogenetics and bone marrow plasma cell burden(BMPC)≥10% pre-ASCT and BMPC>60% at diagnosis, for CIBMTR and S-ERMM, respectively, but no measures of response to therapy. The dynamic S-ERMM(DS-ERMM) combined the S-ERMM score with achievement of≥VGPR. The EMBT score, published in 2023, incorporated disease status at ASCT, ISS, and Karnofsky performance score.

The CIBMTR aims to predict outcomes following ASCT but was not specifically designed to identify FHR patients, despite demonstrating a positive correlation between the CIBMTR score and 3-year PFS and OS. The CIBMTR internal validation cohort identified low versus intermediate versus high-risk patients to have a median PFS of 42 vs. 35 vs. 14 months(p<0.001), a 3-year PFS estimate of 68% vs. 49% vs. 31%(p<0.001) with a HR for death of 3.13(p=0.0001) and median OS of 49 months(p<0.001) in the high-risk group. The CIBMTR score was ex-ternally validated in a cohort of 313 patients, of which only 12(3.8%) were classified as high risk, with a risk of relapse at 1-year(41.7% vs. 18.2% vs. 10.5%, p=0.0043), 2-year(58.3% vs. 38.2%vs. 17.8%, p=0.0001) and 3-year(66.7% vs. 50.9% vs. 25.7%,p<0.0001) compared to intermediate and low-risk groups, re-spectively, and a median PFS of 14.1 months.

The S-ERMM estimated the odds ratio(OR) for early relapse, de-fined as within 18 months of front-line therapy, high versus low risk, of 5.59(p<0.001) and 4.87(p=0.001) in the training and validation datasets, respectively. S-ERMM high patients had a median PFS of 13.2 months, median PFS-2 of 19.8 months, and median OS of 31.5 months. However, prediction of early progres-sion with S-ERMM was not superior to the ISS, R-ISS, and R2-ISS when applied to an external validation CoMmpass dataset(N=476).

The EBMT score system, published in 2023, utilized a point-based scoring system to predict early progression,≤12months of front-line ASCT, based on disease status at ASCT, ISS at diagnosis, and Karnofsky score≤70, and has now been exter-nally validated in a large cohort of patients enrolled in 5MM transplantation registries. High-risk patients had a 12-month PFS rate of 57.1%(N=195/7228) in the large real-world training Mel200 ASCT cohort and 56.2% in a second validation cohort who received Mel140 ASCT.

These scoring systems are attractive despite associated lim-itations, particularly in the absence of a simple diagnostic assay. In practice, however, clinical assessment of the rapid-ity, depth, and duration of response, utilizing conventional serological methods and radiographic assessment for progres-sive lytic or extramedullary disease, remains the only reliable

TABLE 3| Scoring systems to identify patients at high risk of early progression.

|Score|Variables|Clinical outcome| |---|---|---| |CIBMTR[43]|● Album in diagnosis<3.5g/dL:+2 points ● Standard risk CG:+1 point ● High-risk CG:+4 points) ● Pre-ASCT bone marrow plasma cells(BMPCs): +4 points):≥10% ● No CG abnormality, Pre-ASCT BMPCs<10%, albumin at diagnosis≥3.5g/dL: 0 points|● Low risk(0-3) ● Intermediate risk(4-8) ● High risk(9-10) ● 3-year PFS: 58% versus 49% versus 31%(p<0.001) ● 3-year OS:88% versus 81% versus 64%;(p<0.001)| |S-ERMM [44]|● LDH>ULN:+5 points ● Presence of t(4;14):+5 points ● Presence of del17p:+3 points ● Abnormal albumin:+3 points ● BMPC>60%:+3 points ● Lambda free light chain:+2 points|● Low risk(≤5) ● Intermediate risk(6-10) ● High risk(≥11) Overall Risk of ER18: Intermediate versus Low (OR=2.39,p<0.001) High versus low (OR=5.59,p<0.001) Median OS: NR versus 59.5 versus 31.5 months(Low vs. Int vs. High) Median PFS2:62.3 versus 40.0 versus 19.8 months Median PFS 31.6 versus 17.3 versus 13.2 months. Low risk(≤0) Intermediate(1-5) High:≥6| |DS-ERMM|● Achievement of VGPR:-4 points|Overall risk of ER18: High:≥6 Overall risk of ER18: Int versus Low (OR=2.36,p<0.001) High versus Low (OR=6.34,p<0.001) Median OS: NR versus 57.3 months(p<0.001) Median PFS-2 NR versus 53.8 versus 40.2 months (p<0.001)| |EBMT score [45]|ISS stages: ● I:+0 point ● II:+1 point ● III:+2 points Disease status at ASCT: ● CR/VGPR:+0 point ● PR:+1 point ● SD/MR:+2 points ● Relapse/Progression:+4 points Karnofsky≤70:+1 point|PFS at 12 months: 91.7% versus 57.1% for Low risk (score 0) versus high risk(score 4).|

means of identifying all FHR patients. Currently, there is no clinical risk assessment tool that allows for prediction of func-tional high-risk disease.

10| Clinical Trials

Enrolment for newly identified patients to a clinical trial spe-cifically targeting FHR disease is best practice; however, there are limited trial options that focus on FHR. The Australasian Myeloma Research Consortium(AMARC) sponsored IBIS study(ACTRN12621001037897) incorporates the CELMOD Iberdomide(IB),CD38mABisatuximab(IS) and dexamethasone

<EndOfPara>as salvage for patients relapsing $\leq 18$ months of initiating 1 L and will complete accrual by mid-2025. IBIS is the forerunner for a Centralized platform for FHR-MM(ZEPFHR-MM), a multi-domain study with Bayesian statistical design enabling rapid read-out of response, at completion of cycle 4 of any given stra-tum. ZEPFHR will evaluate numerous novel drug combina-tions, including bispecific T cell engagers, and will open in early 2025, again for patients relapsing within 18 months of initiating 1 L. Correlative analysis will enable evaluation of the genomic,immunological, proteomic, and transcriptomic signature in FHR MM to elucidate novel biomarkers and also therapeutic targets in this disease. To our knowledge, this is the first study to prospectively evaluate bispecific antibodies in FHR disease.

TABLE 4| Patient populations evaluated.

|Score|Population| |---|---| |CIBMTR|N=2548| ||CIBMTR transplant| ||registry(USA)| ||2008-2017| |S-ERMM|N=2086| ||Training set: N=844| ||Median age=66| ||IST-CAR-506(phase II), IST-CAR-561(phase I/II),| ||EMN01(Phase III), RV-| ||MM-EMN-441(Phase III)| ||Validation set: N=374| ||Median age=57| |EBMT|FORTE NCT02203643| ||| ||EBMT registry(Europe)| ||2014-2019| ||Training set: N=7227| ||200 mg/m2 Mel-ASCT| ||Validation:| ||N=5616 Mel200;| ||N=1523140mg/m2 Mel-ASCT|

A recent online search of ClinicalTrials.gov did not identify ad-ditional FHR-MM studies. In the absence of a suitable FHR clin-ical trial, clinicians should consider a CD38mAB-IMID based regimen or CART cell therapy, if available.

11| Conclusion

With the emergence of newer therapies, it will likely be nec-essary to modify how we define FHR. Specifically, the defi-nition of early progression will need to be pushed out to later time points, as irrespective of what induction is being utilized,those patients that progress earlier rather than later will al-most certainly manifest shorter OS. Moreover, the possibility exists that whatever the underlying biology is that drives FHR,it may well, to a greater or lesser extent, be agonistic to our currently available classes of therapeutics. At present, we pro-pose a definition of FHR disease that is simple and applicable to all NDMM patients. This consensus definition will inform clinical trials that focus on FHR-MM patients to determine the best therapeutic strategies to improve clinical outcomes while providing a correlative platform to progress our under-standing of the adverse biology underpinning FHR disease.To date, very limited data exist to inform the best approach to FHR as we have defined; however, there is evidence that while CD38mAB-based approaches provide a benefit, they do not entirely mitigate the impact of FHR disease biology, even in those that are CD38mAB naïve at the time of early progression.Enrolment in a clinical trial that specifically focuses on FHR disease remains best practice.

Ethics Statement

The authors have nothing to report.

Conflicts of Interest

S.L. reports honoraria from Amgen, Janssen, BMS; support for attend-ing meetings from Janssen and Pfizer. E.T. reports grants from Amgen,Janssen, GSK, Takeda, and Sanofi; payment or honoraria for lectures,presentations, speaker bureaus, manuscript writing, or educational events from Amgen, ASTRA, BMS, GSK, Integris, Pfizer, Sanofi, and Takeda; support for attending meetings or travel from Amgen, Takeda,and Eusa Pharma; and participation on a data safety monitoring board or advisory board from BMS, GSK, Sanofi, Takeda, Eusa Pharma,Janssen, ASTRA, and Amgen. F.G. reports payment or honoraria for lectures,presentations, speaker bank, GSK, Sanofi, Takeda, Eusa Pharma,and Eusa Pharma; and participation on a data safety monitoring board, eruda-tional events from Janssen, Karyopharm, and Pfizer. M.K. reports honoraria from Takeda, Celgene, Amgen, Janssen Oncology, Consulting or Advisory Role for Janssen Oncology, Celgene, Bristol Myers Squibb,Takeda, Amgen, AbbVie, GlaxoSmithKline, Seattle Genetics, Pfizer,Research Funding: Celgene(Inst), Travel, Accommodations, Expenses from Takeda,Janssen,Bristol Myers Squibb/Celgene. M.B. reports hon-aria from Sanofi, Celgene, Amgen, Janssen, Novartis, Bristol Myers Squibb, and AbbVie; participation in an advisory board for Janssen and GSK; and research funding from Sanofi, Celgene, Amgen, Janssen, Novartis, Bristol Myers Squibb, and AbbVie; participation in an advisory board for Janssen and GSK; and research funding from Sanofi, Celgene, Amgen, Janssen, Novartis, Bristol Myers from Janssen, BMS, Haemalogix, Antengene; consultancy for Janssen,Haemalogix; speakers bureau, BMA, Pfizer, Haemalogix, Antengene and Sanofi.The other authors declare no conflicts of interest.

Data Availability Statement

Data sharing is not applicable for this article as no new data were cre-ated or analyzed in this work.

References

1."International Myeloma Society Consensus on Genomic Definition of High-Risk Myeloma," Journal of Clinical Oncology In press.

2.A.Panopoulou,S.Easdale,M.Ethell,et al.,“Impact of Ultra High-Risk Genetics on Real-World Outcomes of Transplant-Eligible Multiple Myeloma Patients," Hema 7,no.2(2023):e831.

3.K.D.Boyd,F.M.Ross,L.Ciechcio,et al.,“A Novel Prognostic Model in Myeloma Based on Co-Segregating Adverse FISH Lesions and the ISS: Analysis of Patients Treated in the MRC Myeloma IX Trial," Leuke-mia 26, no.2(2012):349-355.

4.V.Shah,A.L.Sherborne,B.A.Walker,et al.,“Prediction of Outcome in Newly Diagnosed Myeloma: A Meta-Analysis of the Molecular Pro-files of 1905 Trial Patients," Leukemia 32, no.1(2018):102-110.

5. B. A. Walker, K. Mavrommatis, C. P. Wardell, et al.,“A High-Risk,Double-Hit, Group of Newly Diagnosed Myeloma Identified by Ge-nomic Analysis," Leukemia 33, no. 1(2019): 159-170.

6.A.Spencer,P.Moreau,M.-V.Mateos,et al.,“Daratumumab for Pa-tients With Myeloma With Early or Late Relapse After Initial Therapy:Subgroup Analysis of CASTOR and POLLUX," Blood Advances 8, no.2(2024):388-398.

7. A. Spencer, P. Mollee, H. A. Blacklock, et al.,“Real-World Outcome for Newly Diagnosed Patients With Functional High-Risk Myeloma- a

Myeloma and Related Diseases Registry Analysis," Blood 134, no. Sup-plement_1(2019): 269.

8.F.Gay,G.Bertuglia, and R. Mina,“A Rational Approach to Func-tional High-Risk Myeloma," Hematology. American Society of Hematol-ogy. Education Program 2023, no. 1(2023): 433-442.

9.E.Kastritis, M. Roussou, E. Eleutherakis-Papaikouov, et al.,“Early Relapse After Autologous Transplant Is Associated With Very Poor Survival and Identifies an Ultra-High-Risk Group of Patients With My-eloma," Clinical Lymphoma Myeloma and Leukemia 20, no. 7(2020):445-452.

10.C.Bygrave,C.Pawlyn,F.Davies,et al.,“Early Relapse After High-Dose Melphalan Autologous Stem Cell Transplant Predicts Inferior Survival and Is Associated With High Disease Burden and Genetically High-Risk Disease in Multiple Myeloma," British Journal of Haematol-ogy 19,3.(2021):551-555.

11.J.Corre,L.Montes,E.Martin,et al.,“Early Relapse After Autolo-gous Transplant for Myeloma Is Associated With Poor Survival Regard less of Cytogenetic Risk," Haematologica 105, no. 9(2020): e480-e483.

12. M.D'Agostino, M. G. Zaccaria, B. Ziccheddu, et al.,“Early Relapse Risk in Patients With Newly Diagnosed Multiple Myeloma Character-ized by Next-Generation Sequencing," Clinical Cancer Research 26, no.18(2020):4832-4841.

13.E.K.M.,A. Manubens,M.D'Agostino,et al.,“Early Relapse in Pa-tients With Newly-Diagnosed Multiple Myeloma: A European Myeloma Network/Harmony Alliance Analysis Including 10843 Patients," Blood 144, no. Supplement 1(2024): 898.

14.N. M. Yael Cohen, I. Avivi, T. Shragai, et al.,“CHARACTERISICS and Outcomes of Patients With Functional High Risk Multiple MyLoma From the HONEUR Federated Data Network," Hema 8, no. S1(2024):1687.

15.P. Sonneveld, M. A. Dimopoulos, M. Boccadoro, et al.,“Daratu-mumab, Bortezomib, Lenalidomide, and Dexamethasone for Multiple Myeloma," New England Journal of Medicine 390, no.4(2024): 301-313.

16.S. Manier, M. Dimopoulos, C. Hulin, et al.,“Predictors of Lenallo-mide Refractory Relapse Timing With Newly Diagnosed Multiple My-eloma: A FIRST Trial Subanalysis," Clinical Lymphoma Myeloma and Leukemia 23, no. 9(2023): e297-e306.

17.S.Z. Usmani,H. Quach,M.V.Mateos,et al.,“Carfilzomib, Dexa-methasone, and Daratumumab Versus Carfilzomib and Dexametha-asone for Patients With Relapsed or Refractory Multiple Myeloma(CANDOR): Updated Outcomes From a Randomised Multicentre,Open-Label, Phase 3 Study," Lancet Oncology 23, no.1(2022):65-76.

18.K. Weisel, G.F. Geils, L. Karlin, et al.,“Carfilzomib, Dexametha-sone, and Daratumumab Versus Carfilzomib and Dexamethaas in Relapsed or Refractory Multiple Myeloma: Subgroup Analysis of the Phase 3 Candor Study in Patients With Early and Late Relapse," Blood 136, no. Supplementary 1(2020): 37-38.

19.P.Moreau,M.A. Dimopoulos,J. Mikhail,et al.,“Isatuximab, Car-filzomib, and Dexamethasone in Relapsed Multiple Myeloma: Car and Dexamethasone in Relapsed Multiple Myeloma: & Multimaxite, C.2021:371-29.

20.T.Facon,P.Moreau,R.Baker,et al.,“Isatuximab Plus Carfilzomib and Dexamethasone in Patients With Early Versus Late Relapsed Mul-tiple Myeloma: Ikema Subgroup Analysis," Blood 140, no. Supplement 1(2022): 1820-1822.

21.T.Facon,P.Moreau,R.Baker,et al.,“Isatuximab Plus Carfilzomib and Dexamethasone in Patients With Early Versus Late Relapsed Mul-tiple Myoloma: IKAMEa Subgroup Analysis," Haematologica 109, no. 2(2024):604-616.

22.R. Szalat,K. Anderson, and N. Munshi,"Role of Minimal Residual Disease Assessment in Multiple Myeloma," Haematologica 109, no. 7(2024):2049-2059.

23. N. Moukalled, F. Malard, A. Bazarbachi, and M. Mohty,“Minimal Residual Disease in Multiple Myeloma," La Presse Medicale 54(2024):104261.

24. J. Hillengis, A. D. Cohen, M. E. Agha, et al.,“The Phase 2 CARTI-TUDE-2 Trial: Updated Efficacy and Safety of Ciltacabtagene Autoleu-cel in Patients With Multiple Myeloma and 1-3 Prior Lines of Therapy(Cohort A) and With Early Relapse After First Line Treatment(Cohort B)," Blood 142, no. Supplement 1(2023): 1021.

25.S. Usmani,K.Patel,P.Hari,et al.,“KarMMa-2 Cohort 2a: Efficacy and Safety of Idecabtagene Viculeuc in Clinical High-Risk Multiple Myeloma Patients With Early Relapse After Frontline Autologous Stem Cell Transplantation,” Blood 140, no. Supplement 1(2022):875-877.

26.R. Turner,H.Quach,N.Horvath,et al.,“Response Adaptive Salvage With Kt and ASCT for Functional High-Risk Multiple Myeloma-The Australasian Leukemia and Lymphoma Group(ALLG) MM17 Trial,"British Journal of Haematology 202(2023): 530-538.

27.S.Lim,J.Reynolds,H.Quach,et al.,“Response Adaptive Salvage Treatment With Daratumumab-Lenalidomide-Dexamethasone for Newly Diagnosed Transplant-Eligible Multiple Myeloma Patients Fail-ing Front-Line Bortone Zembib-Based Induction Therapy-ALLG MM21,"British Journal of Haematology 205(2024): 900-914.

28.S.K.Kumar,C.C.Hofmeister,N.Biran,et al.,“The Mydrug Plat-form Trial: Efficacy and Safety of Daratumumab, Ixazomib, Pomalid-omide, and Dexamethasone in Patients With Functionally High-Risk Multiple Myeloma, Blood 140, no. Supplement 1(2022): 4434-4435.

29. J. Richter, J. A. Zonater, N. Nathwani, et al.,“Selinexor, Ixazombi,Pomalidomide and Dexamethasone in Functionally High-Risk Multi-ple Myeloma: Results From the Myeloma Developing Regimens Using Genomics(MyDRUG) Sub-Protocol Y3," Blood 142, no. Supplement 1(2023): 3387.

30.S.Kumar,G.Bianchi,N.Biran,et al.,“Myeloma Developing Reg-miens Using Genomics(MyDRUG) Trial: Results From the RAS Mu-tation Targeting Arm," Journal of Clinical Oncology 40, no. 16 suppl(2022):8055.

31.Y.Xu,W.Li,G.An,et al.,“Promising Safety and Efficacy of Trovocatagene Autoleuc(C-CAR088) Followed ASCT in Ultra High-Risk Multiple Myeloma(UHR-MM) Patients," Blood 144(2024):7236.

32.M.F. Kaiser, R. Philip, A. Hall, et al.,“Ultra High-Risk Multiple Myeloma Patients With Multi-Hit Tumours and SKY92 High Risk Sig-nature Are at Increased Risk of Early Relapse Even When Treated With Extended Intensified Induction and Consolidation-Results From the Optimum/Mukine Trial," Blood 142, no. Supplement 1(2023): 881.

33.C.Y.Soekojo,T.H.Chung,M.S.Furqan,and W.J.Chng,"Genomic Characterization of Functional High-Risk Multiple Myeloma Patients,"Blood Cancer Journal 12, no.1(2022): 24.

34.C.Y. Soekojo,T.-H.Chung,M.S.Furqan,and W.-J.Chng,"Identi-fying the Genomic Profile of Functional High-Risk Multiple Myeloma Patients,"Patients,"Blood 136, no. Supplement 1(2020): 51-52.

35.Y.C. Cohen, M. Zadla, S. Y. Wang, et al.,“Identification of Resis-tance Pathways and Therapeutic Targets in Relapsed Multiple Myeloma Patients Through Single-Cell Sequencing," Nature Medicine 27, no. 3(2021):491-503.

36.L.F. Porrata, M. A. Gertz, S. M. Geyer, et al.,“The Dose of Infused Lymphocytes in the Autograft Directly Correlates With Clinical Out-come After Autologous Peripheral Blood Hematopoietic Stem Cell Transplantation in Multiple Myeloma," Leukemia 18, no. 6(2004):1085-1092.

37.J. Rueff, M. Medinger, D. Heim, J. Passweg, and M. Stern,“Lympho-cyte subset Recovery and Outcome After Autologous Hematopoietic Stem Cell Transplantation for Plasma Cell Myeloma," Biology of Blood and Marrow Transplantation 20, no. 6(2014): 896-899.

13

38. L. Lee, N. Alrasheed, G. Khandelwal, et al., “Increased ImmuneRegulatory Receptor Expression on Effector T Cells as Early Indicators of Relapse Following Autologous Stem Cell Transplantation for Multiple Myeloma,” Frontiers in Immunology 12 (2021): 618610.

39. D. J. Chung, K. B. Pronschinske, J. A. Shyer, et al., “T- Cell Exhaustion in Multiple Myeloma Relapse After Autotransplant: Optimal Timing of Immunotherapy,” Cancer Immunology Research 4, no. 1 (2016): 61–71.

40. S. A. Minnie and G. R. Hill, “Autologous Stem Cell Transplantation for Myeloma: Cytoreduction or an Immunotherapy?,” Frontiers in Immunology 12 (2021): 651288.

41. E. V. Batorov, M. A. Tikhonova, N. V. Pronkina, et al., “Increased Circulating CD4(+)FOXP3(+) T Cells Associate With Early Relapse Following Autologous Hematopoietic Stem Cell Transplantation in Multiple Myeloma Patients,” Oncotarget 9, no. 43 (2018): 27305–27317.

42. H. Parmar, M. Gertz, E. I. Anderson, S. Kumar, and T. V. Kourelis, “Microenvironment Immune Reconstitution Patterns Correlate With Outcomes After Autologous Transplant in Multiple Myeloma,” Blood Advances 5, no. 7 (2021): 1797–1804.

43. B. Dhakal, A. D'Souza, N. Callander, et al., “Novel Prognostic Scoring System for Autologous Hematopoietic Cell Transplantation in Multiple Myeloma,” British Journal of Haematology 191, no. 3 (2020): 442–452.

44. G. M. Zaccaria, L. Bertamini, M. T. Petrucci, et al., “Development and Validation of a Simplified Score to Predict Early Relapse in Newly Diagnosed Multiple Myeloma in a Pooled Dataset of 2,190 Patients,” Clinical Cancer Research 27, no. 13 (2021): 3695–3703.

45. M. Beksac, S. Iacobelli, L. Koster, et al., “An Early Post- Transplant Relapse Prediction Score in Multiple Myeloma: A Large Cohort Study From the Chronic Malignancies Working Party of EBMT,” Bone Marrow Transplantation 58, no. 8 (2023): 916–923.

相关指南

• 2025+EBMT建议：同种异体造血细胞移植治疗皮肤T细胞淋巴瘤 · 血液内科 2025
• 2025+SIECVI共识声明：脓毒性心肌病的超声检查 · 心血管内科 2025
• 2025 EAS共识声明：临床分期指导代谢紊乱及其后遗症的治疗 · 综合与多学科 2025
• 2026 APSC共识声明：经导管缘对缘修复在治疗三尖瓣和二尖瓣反流中的应用 · 心血管内科 2026
• 中国慢性髓系白血病个体化全程管理专家共识 · 血液内科
• 胃肠外直接凝血酶抑制剂在特殊人群中的临床应用专家共识 · 血液内科
• （2026.V1）NCCN临床实践指南：直肠癌 · 肿瘤内科 2026
• 2026 FIGO最佳实践建议：绝经期女性的心理健康 · 内分泌与代谢 2026