Prognostic impact of MRD dynamics during maintenance therapy for patients with NDMM

Measurable residual disease (MRD) assessment is a key prognostic tool for patients with newly diagnosed multiple myeloma.¹ There is now growing potential for the use of serial MRD analysis rather than a one-off assessment to guide treatment decisions, particularly in relation to maintenance therapy and observation scenarios.¹ Relevant data surrounding these clinical settings however remains limited.¹

In order to bridge this knowledge gap, Bruno Paiva et al.¹ conducted a pooled analysis of two phase III trials investigating ixazomib maintenance therapy for patients with newly diagnosed multiple myeloma. The aim of the analysis was to evaluate the progression-free survival (PFS) in relation to MRD status, the variability of MRD status over time and its impact, and the PFS benefit with ixazomib maintenance compared with placebo.¹ We summarize the findings in the article below.

Study design

The two randomized phase III trials chosen for the pooled analysis were the TOURMALINE-MM3 (NCT02181413) and -MM4 (NCT02312258) studies. For more information on previous results from these trials, check out this Multiple Myeloma Hub article.

Both had similar study designs and were conducted in over 30 countries and across more than 160 clinical sites.

The study protocol for the MM3 and MM4 trials was as follows:

• Patients randomized at a 3:2 ratio (maintenance vs placebo)
• Ixazomib: Oral, 3 mg on Days 1, 8, and 15 (28-day cycle)
• Ixazomib dose increased to 4 mg from Cycle 5 if tolerated in previous cycles
• Placebo therapy matched ixazomib
• Bone marrow aspirates taken at randomization, Cycle 13, and at the end of treatment were used to assess patient MRD by eight-color flow cytometry

The primary endpoint for both studies was PFS. Secondary endpoints included the frequency of conversion from positive to negative MRD, sustained MRD negativity, and the correlation between MRD status and survival.

Results

A total of 1,362 patients were included in the pooled analysis and MRD status was available in 1,280 patients at the beginning of randomization:

• 20.5% had undetectable MRD
• 79.5% had detectable MRD

Patient characteristics, not including age, were generally well-balanced between patients with negative MRD and those with positive MRD at randomization before maintenance therapy (Table 1).

Table 1. Patient characteristics*

CR, complete response; IMiD, immunomodulatory drug; ISS, International Scoring System; MRD, measurable residual disease; PI, proteasome inhibitors; PR, partial response; VGPR, very good partial response. *Adapted from Paiva, et al.1 †Defined as the presence of any of the following three individual abnormalities: del(17), t(4;14), and t(14;16). ‡Defined as the presence of any of the following four individual abnormalities: del(17), t(4;14), t(14;16),and amp1q.
Characteristic, % (unless otherwise stated)	MRD negative		MRD positive
	Ixazomib (n = 820)	Placebo (n = 542)	Ixazomib (n = 820)	Placebo (n = 542)
Patients, n	161	101	606	412
Age
<75 years	90.1	90.1	77.6	76.2
≥75 years	9.9	9.9	22.4	23.8
Sex
Male	58.4	59.4	58.1	59.0
Female	41.6	40.6	41.9	41.0
Cytogenetic abnormalities at diagnosis
High risk†	13.0	26.7	17.2	17.0
Corresponding standard-risk	64.0	54.5	65.2	64.3
Unclassifiable	23.0	18.8	17.7	18.7
Expanded high risk‡	26.1	36.6	33.7	32.8
Corresponding standard risk	40.4	34.7	36.8	36.2
Unclassifiable	33.5	28.7	29.5	31.1
Preinduction ISS stage
I or II	67.1	68.3	68.5	67.2
III	32.9	31.7	31.5	32.8
PI exposed	90.1	89.1	84.3	83.0
IMiD exposed	39.8	43.6	34.3	35.0
Response status after transplant
CR or VGPR	95.0	92.1	61.1	63.6
PR	5.0	7.9	38.9	36.4

MRD status at randomization

Regardless of MRD status, patients treated with ixazomib experienced a higher median PFS than placebo (Figure 1). Overall, the median PFS for patients who were MRD negative was 38.6 months compared with 15.6 months for patients who were MRD positive (hazard ratio [HR], 0.47; 95% confidence interval [CI], 0.37–0.58; p = 0.019). Negative MRD status was associated with prolonged PFS in nearly all patient subgroups, including cytogenetic risk and prior treatment. Moreover, patients had a greater reduction in the risk of progressive disease (PD) or death if they were either enrolled in the MM4 trial, ≥75 years old, or had an International Staging System score of 3.

MRD, measurable residual disease; PFS, progression-free survival.
*Adapted from Paiva, et al.¹

MRD dynamics

The evaluation of MRD dynamics was achieved by paired assessments at randomization and during maintenance. A total of 1,166 patients were analyzed and found that almost half of MRD-negative patients at randomization lost their negative status in <2 years (111/229, 48.5%). In contrast, a small group of patients who were MRD positive converted to MRD negative status during maintenance therapy (60/937, 6.4%). The MRD dynamics of patients during the maintenance period regardless of therapy is shown in Figure 2.

MRD, measurable residual disease.
*Adapted from Paiva, et al.^1
†Sustained MRD was defined as a positive or negative status held for more than 12 months.

The 14-month landmark analysis showed 2-year PFS rates to be higher in patients who had either converted to negative MRD or had sustained negative MRD status (Figure 3).

MRD, measurable residual disease; PFS, progression-free survival.
*Adapted from Paiva, et al.^1
†Sustained MRD was defined as a positive or negative status held for more than 12 months.

The risk of PD or death increased for patients who had converted from negative to positive MRD status compared with patients with sustained negative MRD (HR, 3.31; 95% CI, 1.77–6.20; p < 0.001). This was the same for patients with persistent positive MRD compared with those who had converted from positive to negative status (HR, 3.72; 95% CI, 1.85–7.46; p < 0.001).

Impact of ixazomib maintenance and placebo on MRD positive and negative patients

At the 14-month landmark analysis, patients treated with ixazomib showed a statistically significant difference in the rates of sustained MRD negativity compared with placebo at 76% and 60.3%, respectively (p = 0.04). There were no differences in the rate of conversions from MRD positive to MRD negative status.

At the 14-month landmark analysis, the 2-year PFS for patients who were MRD negative was higher compared with patients who were MRD positive (Figure 4). Higher rates of PFS were also observed with ixazomib in patients whose MRD levels had not increased from randomization to the 14-month analysis timepoint (HR, 0.39; 95% CI, 0.22–0.68; p < 0.001) compared with patients whose MRD level had increased (HR, 0.67; 95% CI, 0.22–2.07; p = 0.487). In conjunction with the marginal differences in the rate of conversions from MRD positive to MRD negative status with ixazomib versus placebo, these findings suggest that rather than erasing MRD, treatment was able to control the size of the clone for longer periods.

MRD, measurable residual disease; PFS, progression free survival.
*Adapted from Paiva, et al.¹

Future directions²

While this analysis provides many insights into the use of MRD to guide treatment decisions, especially in the maintenance setting, there remain several key areas for further investigation:

• Can we shorten maintenance duration in patients with sustained MRD negativity?
• Should we escalate therapy for patients with an MRD-positive status to convert to a negative MRD?
• Can blood-based MRD assays facilitate earlier detection of change in MRD?
• How do these marrow and blood dynamics correlate with imaging?
• Can we predict which patients will convert from MRD negative to positive?
• Should serial MRD assessment be prioritized over a one-off evaluation?

Conclusion

Results from this analysis highlight that MRD negative status is associated with prolonged PFS compared with patients who are MRD positive in almost all patient subgroups. Treatment with ixazomib was found to improve the PFS in patients with an MRD-positive status both at randomization and the 14-month landmark time point. However, 2-year ixazomib maintenance treatment was not effective in patients who were MRD negative at randomization or at the 14-month landmark. Interestingly, MRD status before maintenance therapy was found to be an independent prognostic factor for PFS. The clinical implications of MRD status over time are yet to be fully established; however, PFS results from this study indicate that MRD negativity as an endpoint for treatment could be extrapolated to a maintenance setting. Furthermore, persistent MRD was consistently associated with PD, and therefore, it is reasonable to suggest acting upon an MRD-positive status with treatment intensification rather than a negative MRD status.