GEM2012MENOS65 trial | Detection of measurable residual disease in MM using next-generation flow cytometry

The depth of response based on measurable residual disease (MRD) is one of the key prognostic factors in multiple myeloma (MM). Techniques used for detection of MRD, such as next-generation sequencing (NGS), next-generation flow cytometry (NGF), and positron-emission tomography/computerized tomography (PET/CT) vary in sensitivity levels. Therefore, MRD-negative definitions vary between studies depending on the evaluation technique and the MRD-negativity thresholds of detection used. In order to standardize assessment of MRD, the International Myeloma Working Group (IMWG) updated the response criteria in 2016.¹

Although the new IMWG MRD-negative response criterion has been evaluated in PET/CT and NGS, no prospective studies were conducted using NGF. Bruno Paiva and colleagues published in the Journal of Clinical Oncology results of a study assessing the value of MRD status using NGF in patients with MM.²

Methods

Study design

The open-label, phase III study PETHEMA (GEM2012MENOS65, NCT01916252) included 458 patients who received

• Induction with six cycles of bortezomib, lenalidomide, and dexamethasone (VRD)
• Conditioning with high-dose therapy (HDT)
  • busulfan and melphalan (Bu-Mel)
  • melphalan (Mel-200)
• Autologous stem cell transplantation (auto-HSCT)
• Consolidation with two cycles of VRD

Subsequently, patients were enrolled in the GEM2014MAIN (NCT02406144) study and randomized to

• 2 years of maintenance with
  • lenalidomide and low-dose dexamethasone (Rd)
  • Rd and ixazomib
• Patients with positive MRD status after 2 years continued with Rd for an additional 3 years

MRD assessment

MRD was evaluated using NGF at pre-specified times:

• After induction
• On Day 100 after HDT/auto-HSCT
• After consolidation
• At the time of suspected complete remission (CR)
• Annually during maintenance (the first 2 years were included in the analysis)

This clinical trial followed the EuroFlow Consortium standardization that defines NGF for MM as the analysis of two tubes with a pre-defined eight-color/antibodies panel. Ideally, each sample should contain > 5 million cells, which would mean sensitivity of at least 1×10^-5, although this might not be achieved with all samples obtained.³ For this reason, the limit of detection (LOD) needed to be determined for each sample according to the formula: (20/number of viable nucleated cells) × 100.

MRD was defined as

• Detectable if the percentage of phenotypically aberrant clonal plasma cells (PCs) ≥ LOD
• Undetectable if the phenotypically aberrant clonal PCs = 0 or < LOD

Results

Applicability and sensitivity of NGF

• In total, 1,114 (99.6%) out of 1,119 bone marrow (BM) aspirates tested after induction, HDT/auto-HSCT, or consolidation were evaluable
  • 14 of the evaluable samples were severely hemodiluted and inadequate for MRD assessment
• The IMWG considers adequate a threshold of < 10^-5 for prognostic purposes, and the median LOD reached by NGF in this set of samples was 2.9 × 10^-6
• The logarithmic ranges achieved by samples were
  • < 2 × 10^-6 in 1% of samples
  • 2 ×10^-6 to < 10^-5 in 88% of samples
  • ≥ 10^-5 to < 10^-4 in 99.9% of samples
  • ≥ 10^-4 in 100% of samples

MRD during treatment

• Percentage of patients with undetectable MRD increased over treatment; thus, higher rates were observed after HDT/auto-HSCT (42%) and consolidation (45%) compared to after induction (28%)
• Patients with ≤ partial response (PR) & MRD negativity after induction, those who remained MRD negative at later timepoints and at the end, all achieved CR except 2 patients. 32 of 33 patients with very good partial response (VGPR) & MRD negative after induction, remained MRD negative at later timepoints. At the end, 18 out of the initial 33 achieved a CR.

Impact of MRD on disease progression

After a median follow-up of 40 months

• Disease progression was significantly less likely in patients with undetectable MRD than in patients with persistent MRD after consolidation (7% vs 40%; p < 0.001)
• Undetectable MRD before maintenance was associated with an 82% reduction in the risk of progression or death and an 88% reduction in the risk of death (HR, 0.18; 95% CI, 0.11–0.30; p < 0.001 and HR, 0.12; 95% CI, 0.05–0.29; p < 0.001, respectively)
• The estimated 36-month progression-free survival (PFS) and overall survival (OS) rates were greater in patients with undetectable vs persistent MRD after consolidation (PFS 87% vs 50%, HR, 0.21; p < 0.001 and OS 96% vs 88%, HR, 0.36; p = 0.005)
  • The impact of MRD status on reducing the risk of progression or death was independent of treatment arm (Mel-200 or Bu-Mel) or achieving < CR (HR, 0.10; 95% CI, 0.01–0.72; p = 0.02)
  • Achieving or not a CR after consolidation in patients with positive MRD did not have a significant impact on PFS or OS
• Among patients with persistent MRD, all levels of positivity, even as low as ≥ 2 × 10^-6 and < 10^-5, were associated with significantly inferior PFS (p < 0.001) compared with patients who had undetectable MRD
• Among patients with undetectable MRD, the impact of negative status on survival was similar for patients who achieved it after induction, after auto-HSCT, or after consolidation. Survival rates when MRD was undetectable at early vs later time points were
  • PFS: 88% and 85%, respectively (p = 0.38)
  • OS: 94% and 99% (p = 0.17), respectively

MRD response and risk at diagnosis

• Undetectable MRD was associated with a significantly reduced risk of disease progression or death across all subgroups of the intent-to-treat population, including patients with high-risk cytogenetics
• There were no significant differences in the 36-month PFS rate when analyzing patients with undetectable MRD by the  Revised International Staging System (R-ISS), indicating that the achievement of undetectable MRD might overcome the poor prognosis of patients classified as R-ISS-III. Conversely, when MRD remained positive, patients presented worse outcomes according to their risk status (Table 1)
• After these observations, R-ISS and MRD status were identified as the variables with the highest predictive value for PFS and OS are presented in Table 2

Table 1. PFS rate at 36 months by R-ISS according to MRD status

Table 2. Multivariable analyses of PFS and OS, incorporating risk stratification at baseline according to the R-ISS and response after treatment according to MRD status

Impact of maintenance on the MRD status

MRD status after consolidation remained relatively stable during the first 2 years of maintenance

• 54% had sustained undetectable MRD
• 21% had persistent MRD
• 17% of patients changed from detectable to undetectable MRD
  • most conversions occurred during the first year of maintenance
• 7% of patients changed from undetectable to detectable MRD during maintenance, but none of them experienced disease progression thus far

Conclusion

Initial results from the GEM2012MENOS65 trial demonstrated the efficacy and safety of VRD as induction therapy for patients with newly diagnosed MM (NDMM) who are eligible for auto-HSCT. The investigators observed that with every additional VRD induction cycle and treatment phase, the depth of responses increased, and this trend has been further confirmed with MRD assessment by NGF.

With a median follow-up of 40 months, this analysis reported that almost 50% of patients achieved an undetectable MRD when treated with VRD followed by HDT/auto-HSCT and consolidation with VRD. The higher rates of undetectable MRD were translated into higher PFS and OS rates, reaching 90% at 3 years.

Patients presenting with an undetectable MRD after induction achieved similar survival outcomes to patients with MRD undetectable after HDT/auto-HSCT. Therefore, MRD after induction might become a tool to decide on early intensification of treatment, or whether to defer auto-HSCT until after disease progression in patients with undetectable MRD.

The study also validated the updated IMWG MRD-negative response criterion for NGF, which assesses the presence of malignant cells in the BM samples, and demonstrates that it is a highly applicable and sensitive way to evaluate treatment efficacy in MM outside clinical trials. Nevertheless, the authors support previous publications that advocate for defining MRD-negativity with a LOD of 10^-6 rather than the IMWG threshold of < 10^-5.