Daratumumab-based second line therapy improves outcomes after VRD induction, upfront autologous transplant and lenalidomide maintenance

Abstract

Background:

For eligible patients with newly diagnosed multiple myeloma (MM), induction with bortezomib, lenalidomide, and dexamethasone (VRD) used to be the standard of care until a few years ago, when it was gradually replaced by Daratumumab + VRD (Dara-VRD), followed by autologous stem cell transplantation (autoHCT) and lenalidomide maintenance. Most patients relapse and require second-line (2L) therapy.

Patients and Methods:

We conducted a retrospective, single-center study of MM patients who received VRD induction, upfront autoHCT between 2005 and 2021, and lenalidomide maintenance before relapse.

Results:

146 patients were included, with a median age of 60 years (range 32–80), and 39 (27%) patients had high-risk cytogenetics. Following first progression, 31% (n=45) received an immunomodulatory drug (IMiD) + proteasome inhibitor (PI)-containing triplet as 2L therapy, 21% (n=30) received an IMiD/PI-containing doublet, and 24% (n=35) received a daratumumab (Dara)-based 2L therapy. Patients receiving Dara-based regimens had higher rates of ≥VGPR (63%) compared to those receiving other regimens (IMiD+PI triplet: 35%, IMiD/PI doublet: 34%) (p=0.026). After a median follow-up of 35.8 months from 2L therapy, median PFS2 with Dara-based regimens was 59.9 months, versus 12.3 months and 11.5 months for doublets and triplets, respectively (p=0.007). Median OS from initiation of second-line therapy for the entire cohort was 52.6 months. In multivariable analysis, Dara-based therapy was associated with improved PFS2 (HR 0.35, p<0.001), while clinical progression (vs. biochemical) predicted worse PFS2 (HR 1.58, p=0.026). Progression ≥12 months post-autoHCT was associated with improved OS (HR 0.23, p<0.001); busulfan-melphalan conditioning (HR 4.06, p<0.001) and clinical progression (HR 1.99, p=0.006) were associated with inferior OS.

Conclusion:

Dara-based 2L regimens were associated with superior PFS2 approaching almost 5 years.

Microabstract

This single-center retrospective study examined outcomes of 146 patients with multiple myeloma who relapsed after receiving bortezomib, lenalidomide, and dexamethasone (VRD) induction, upfront autologous hematopoietic stem cell transplantation between 2005 and 2021, and lenalidomide maintenance. Recipients of daratumumab-based second line therapy exhibited superior progression-free survival compared to recipients of immunomodulatory drug and proteasome inhibitor-containing triplet or doublet therapies. These findings support the benefit of daratumumab-based second line therapy in anti-CD38-naïve patients.

Introduction

There has been a steady improvement in outcomes for patients with multiple myeloma (MM) over the past two decades, due to the availability of proteasome inhibitors (PIs), immunomodulatory drugs (IMiDs), monoclonal antibodies, and the use of autologous stem cell transplantation (autoHCT) in eligible patients¹. Until recently, bortezomib, lenalidomide, and dexamethasone (VRD) followed by autoHCT and lenalidomide maintenance was considered standard frontline treatment². It remains standard of care in places where anti-CD38-based regimens are not yet available. Despite these improvements, MM remains largely incurable, and most patients eventually relapse.

Treatment decisions in the relapsed/refractory setting are complex, and depend on prior therapies, duration and depth of response, and patient’s age and comorbidities. There is no universally accepted standard of care in the second-line (2L) setting. The Connect MM disease registry, a prospective observational study of U.S.-based community oncology practices, evaluated 2L regimens between 2009 and 2016 and found increasing use of triplet combinations and novel agents over time³. Patients who received novel therapies had improved median overall survival (OS) from the start of 2L therapy compared to those who did not (29.0 vs. 20.6 months; hazard ratio [95% CI] 1.29 [1.04–1.60]; p=0.022).

Similarly, a retrospective study from the Mayo Clinic assessed second- and third-line treatment patterns between 2003 and 2021⁴. Use of novel agents (carfilzomib, pomalidomide, daratumumab, elotuzumab) increased markedly over time, with 96.4% of patients receiving novel therapies between 2015–2021 compared to 78.2% between 2003–2008. Among 2L regimens, daratumumab, lenalidomide, and dexamethasone (DRD) was associated with best outcomes. However, fewer than half of patients in this study received upfront autoHCT, and only one-quarter received post-transplant maintenance therapy.

These studies with heterogeneous frontline treatments, and variable use of autoHCT and maintenance, limit the generalizability of their findings. In the current study, we sought to evaluate 2L regimens in a uniformly treated cohort of patients who received VRD induction, autoHCT, and post-transplant lenalidomide maintenance.

Methods

We conducted a retrospective, single-center study of patients with MM who received VRD induction followed by autoHCT between 2009 and 2021 and post-transplant lenalidomide maintenance. Following disease progression they started 2L therapy. CONSORT flow diagram is presented in Supplementary Figure 1. Patients were identified through the institutional database and chart review. This study was approved by the Institutional Review Board (IRB) at The University of Texas MD Anderson Cancer Center and was conducted in accordance with the Declaration of Helsinki and the 1996 Health Insurance Portability and Accountability Act (HIPAA).

The primary endpoint was the second progression-free survival (PFS2), defined as the time from initiation of 2L therapy to second disease progression or death, whichever occurred first. Patients who were alive without second progression at the time of last follow-up were censored. Secondary endpoints included OS time, computed from the start of 2L therapy to last follow up, and hematologic response to 2L therapy. The International Myeloma Working Group (IMWG) criteria were used to assess response and progression⁵. First PFS (PFS1) time was computed from the date of autoHCT to the date of first disease progression or death, with censoring at last follow-up if no progression occurred. PFS1, PFS2, and OS were estimated using the Kaplan-Meier method, and differences between groups were evaluated using the log-rank test. Associations between clinical covariates (including cytogenetic risk, measurable residual disease [MRD] status, and type of 2L therapy) and survival outcomes were assessed using Cox proportional hazards regression models.

Comparative analyses of response and survival outcomes were restricted to the three most used second-line regimens: IMiD/PI doublets, IMiD+PI triplets, and daratumumab-based regimens. Patients receiving other regimens, including the unknown regimens, were included in descriptive analyses but excluded from subgroup comparison. MRD testing was performed using an eight-color next-generation flow cytometry (NGF) assay on bone marrow aspirates, with a sensitivity threshold of 1 in 10⁵ nucleated cells (0.001%), based on acquisition of at least 2 million events. When available, MRD was assessed at prespecified intervals, including post-induction, day +100 post-autoHCT, and thereafter as clinically indicated.

Cytogenetic data at diagnosis were obtained from fluorescence in situ hybridization (FISH) analysis of bone marrow aspirates performed at baseline, using validated probe sets for IGH::FGFR3 [t(4;14)], IGH::MAF [t(14;16)], TP53/CEP17 [del(17p)], and CDKN2C/CKS1B [1q21 gain or amplification]. Cytogenetic abnormalities were classified as high-risk if any of t(4;14), t(14;16), del(17p), or 1q21 gain/amplification were present. Institutional cut-offs for abnormal signal patterns were 4.9% for loss of one copy of CDKN2C, 5.8% for gain of one copy of CKS1B, 2% for gain of two copies of CKS1B, 0.2% for gain of three or more copies of CKS1B, 4.7% for loss of one copy of TP53, and 0.2% for t(4;14)/IGH::FGFR3. Next-generation sequencing (NGS) for TP53 was not routinely performed during the study period and those data, therefore, were unavailable for most patients.

Statistical analyses were performed using SAS version 9.4 for Windows (SAS Institute Inc., Cary, NC). All statistical tests were two-sided, with a significance level set at 0.05. No adjustments were made for multiple comparisons.

Results

Patient, Disease, and Treatment Characteristics

A total of 146 patients were included in the analyses, of whom 73 (50%) were males. The median age at autoHCT was 60.3 years (range, 31.5–79.6), 31 (21%) patients were Black, and most (n = 140, 96%) received autoHCT in 2010 or later. Fifty-three (36%) patients had Revised International Staging System (R-ISS) stage II and 8 (5%) patients had stage III. Thirty-nine (27%) patients had high-risk cytogenetic abnormalities. The median Hematopoietic Cell Transplantation-Comorbidity Index (HCT-CI) was 1 (range, 0–8), and 32 (22%) patients had HCT-CI score of >3. Melphalan alone was used as conditioning in 90% of the cohort. Baseline patient characteristics are shown in Table 1.

Table 1.

Baseline patient, disease, and treatment characteristics

Parameter	All (N=146) n (%)
Gender
Male	73 (50)
Female	73 (50)
Age at autoHCT (years)
Median (Range)	60.3 (31.5 – 79.6)
Race
Black	31 (21)
Other	110 (75)
Unknown	5 (3)
Year of autoHCT
< 2010	6 (4)
≥ 2010	140 (96)
R-ISS
I	42 (29)
II	53 (36)
III	8 (5)
Unknown	43 (29)
High-risk cytogenetic abnormality ¥
No	97 (66)
Yes	39 (27)
Unknown	10 (7)
HCT-CI
≤ 3	114 (78)
> 3	32 (22)
Conditioning regimen
Melphalan	129 (88)
Busulfan + Melphalan based	15 (10)
Melphalan based (other)	2 (1)
Response prior to autoHCT
CR	13 (9)
VGPR	72 (49)
PR	57 (39)
SD	4 (3)
MRD status prior to autoHCT
Negative	51 (35)
Positive	57 (39)
Unknown	38 (26)
Months from autoHCT to start of maintenance
Median (Range)	3.9 (1.1 – 11.5)
Duration of maintenance treatment (months)
Median (Range)	21.1 (1.0 – 112.3)
1st progression after autoHCT
< 12 months	19 (13)
≥ 12 months	127 (87)
Type of progression
Biochemical	75 (51)
Clinical	67 (46)
Unknown	4 (3)
Type of 2nd line regimen after progression
Daratumumab – containing regimen	35 (24)
IMID+PI – containing doublet	30 (21)
IMID+PI – containing triplet	45 (31)
Other	28 (19)
Unknown	8 (5)

^¥Defined as del17p, t(4;14), t(14;16) or 1q21 gain or amplification by fluorescence in situ hybridization (FISH).

Abbreviations: autoHCT = Autologous hematopoietic stem cell transplantation, CR = Complete response, HCT-CI = Hematopoietic Cell Transplantation-specific Comorbidity Index, IMID = Immunomodulatory drugs, MRD = Measurable residual disease, PI = Protease inhibitor, PR = Partial response, R-ISS = Revised International Staging System, SD = Stable disease, VGPR = Very good partial response.

Pre-transplant response to induction included ≥complete response (CR) in 13 (9%) patients and ≥very good partial response (VGPR) in 85 (58%) patients. MRD was negative in 51 (35%) patients and positive in 57 (39%) patients prior to autoHCT. Post-transplant, 74 (51%) patients achieved ≥ CR and 129 (88%) patients achieved ≥ VGPR. Post-transplant MRD assessment were unavailable for the majority of patients (115/146).

Disease Progression and Treatment Post-autoHCT

The median follow-up from autoHCT was 76.7 months (range, 8.6–180.8) and median PFS1 was 27.5 months (95% CI 24.3–32.9). First disease progression occurred ≥12 months after autoHCT in 127 (87%) patients. At progression, 75 (51%) patients had biochemical progression, and 67 (46%) patients had clinical progression. One hundred twenty (82%) patients progressed while on lenalidomide maintenance. Univariate and multivariable analyses for PFS1 are shown in Supplementary Tables 1 and 2, respectively.

Following progression, 35 (24%) patients received daratumumab-containing triplet or quadruplet regimens, 45 (31%) received an IMiD+PI triplet, and 30 (21%) received an IMiD/PI-based doublet. Patients in the daratumumab-containing group underwent autoHCT between 2012–2020, and those in the latter two groups were transplanted between 2009–2021. An additional 28 (19%) patients received other regimens, including 8 (5%) patients for whom the regimen details were not available. These patients were included in descriptive analyses only (Supplementary Table 3). Of those receiving daratumumab-containing triplet or quadruplet regimens, the most common combinations were daratumumab, pomalidomide, and dexamethasone (DPd) (n=17; 49%)or DRd (n=13; 37%), with 2 (6%) patients receiving daratumumab, lenalidomide, bortezomib, dexamethasone (DVRd). Following second-line therapy, 18 (12%) patients achieved ≥CR, 52 (36%) achieved ≥VGPR, 46 (32%) had a partial response, and 13 (9%) had progressive disease as their best response. Patients who received daratumumab-based 2L regimens had higher rates of ≥ VGPR (63%) at best response compared to those on IMiD+PI triplets (35%) or IMiD/PI doublets (34%) (p = 0.026).

Survival Outcomes

The median follow-up time from the start of second-line therapy was 35.8 months (range 0.6–139.2). The median PFS2 and OS from initiation of second-line therapy for the entire cohort were 12.4 months (95% CI 10.2–16.9) and 52.6 months (95% CI 45.0–86.3), respectively (Figure 1). Patients receiving daratumumab-based 2L regimens had significantly longer median PFS2 (59.9 months) compared to those receiving IMiD/PI doublets (12.3 months; hazard ratio [95% CI] 2.08 [1.10–3.94]; p=0.025) or IMiD+PI triplets (11.5 months; 2.50 [1.39–4.49]; p=0.002) (Figure 2). No statistically significant differences in OS were observed between patients receiving daratumumab-based regimens (median OS not reached), IMiD/PI doublets (51.9 months; 1.81 [0.81–4.07]; p=0.15), or IMiD+PI triplets (48.9 months; 2.00 [0.94–4.27]; p=0.07).

Figure 1.

Progression-free survival and overall survival of entire cohort.

Figure 2.

Progression-free survival by type of second line treatment regimen.

Multivariable Analysis

In multivariable analysis for PFS2, daratumumab-containing second-line regimens were associated with significantly improved PFS2 (0.35 [0.20–0.61]; p<0.001), while clinical (vs. biochemical) progression at first relapse was associated with worse PFS2 (1.58 [1.06–2.37]; p=0.026) (Table 2). In multivariable analysis for OS, progression ≥12 months after autoHCT was associated with improved survival (0.23 [0.11–0.46]; p<0.001), whereas busulfan and melphalan conditioning (4.06 [1.87–8.81]; p<0.001) and clinical progression (vs. biochemical) (1.99 [1.22–3.25]; p=0.006) were associated with inferior survival (Table 3). Univariate analyses for PFS2 and OS are presented in Supplementary Tables 4 and 5, respectively.

Table 2:

Multivariable analysis of progression-free survival following second progression

	Progression Free survival
	Hazard Ratio (95% CI)	p-value
Race
Non-black vs. Black	1.39 (0.81–2.39)	0.23
R-ISS
II vs. I	1.39 (0.84–2.32)	0.20
III vs. I	2.47 (0.94–6.52)	0.07
Cytogenetic risk
High vs standard	1.28 (0.76–2.17)	0.36
MRD/Response prior to autoHCT
MRD negative CR vs other	1.32 (0.64–2.70)	0.46
1st Progression after autoHCT
≥12 months vs <12 months	0.81 (0.41–1.61)	0.56
Type of progression
Clinical vs. biochemical	1.58 (1.06–2.37)	0.026
Progression on lenalidomide
Yes vs. no	1.58 (0.87–2.89)	0.14
Type of 2nd line regimen after progression
Dara-based triplet/quadruplet vs. other	0.35 (0.20–0.61)	<0.001

Abbreviations: autoHCT= autologous hematopoietic stem cell transplantation, CI= Confidence Interval, CR= Complete Response, dara= daratumumab, MRD= Measurable Residual Disease, R-ISS= Revised Multiple Myeloma International Staging System,

Table 3:

Multivariable analysis of overall survival.

	Overall Survival
	Hazard Ratio (95% CI)	p-value
Cytogenetic Risk
High vs. standard	1.50 (0.84–2.68)	0.17
Conditioning regimen
Busulfan + melphalan-based vs melphalan alone	4.06 (1.87–8.81)	<0.001
MRD/Response prior to autoHCT
MRD negative CR vs. other	0.59 (0.27–1.29)	0.19
1st Progression after autoHCT
≥ 12 months vs. <12 months	0.23 (0.11–0.46)	<0.001
Type of Progression
Clinical vs. biochemical	1.99 (1.22–3.25)	0.006

Abbreviations: autoHCT= autologous stem cell transplantation, CI= Confidence interval, CR= Complete Response, MRD= Measurable residual disease

Discussion

In this study we showed that in 146 patients who had disease progression after VRD induction, autoHCT and lenalidomide maintenance, and received 2L treatment, median PFS was 12.4 months, and median OS 52.6 months. Notably, patients treated with daratumumab-based 2L regimens at relapse had significantly longer PFS2, with median PFS2 approaching 5 years.

Several real-world studies have evaluated second-line treatment patterns, including the Connect MM registry³ and a retrospective analysis from the Mayo Clinic⁴. While the Connect MM registry highlighted evolving use of novel agents, it did not report details of induction, autoHCT, or maintenance, and daratumumab-based regimens were too infrequent to allow meaningful comparison with non-daratumumab approaches. The Mayo Clinic analysis identified DRd as associated with the longest time to next treatment (TTNT) (26 months) and unreached OS; however, their cohort was more heterogeneous, with only 21% receiving PI and IMiD induction, 25% receiving post-autoHCT maintenance, and 14% receiving lenalidomide maintenance. In contrast, our uniformly treated population with VRD induction, autoHCT, and continuous lenalidomide maintenance enables a more direct comparison of 2L regimens and demonstrates that daratumumab-based therapy can achieve longer PFS2, even in a setting of near-universal prior lenalidomide exposure.

Our findings align with real-world data from the Canadian Myeloma Research Group (CMRG) registry, which reported variable outcomes across 2L regimens in a predominantly transplant-ineligible cohort⁶. DRd, used by 10% of patients, achieved the highest rate of ≥VGPR (66%) and the longest median PFS (27.6 months) compared to 44% and 17.7 months in the overall cohort. Daratumumab, bortezomib, and dexamethasone (DVd) showed notably inferior outcomes, with a ≥VGPR rate of 40% and median PFS of 11.2 months. These differences may partly be driven by restricted access to certain drugs in Canada: 95% of DVd patients had prior lenalidomide exposure, while 95% of DRd patients had prior bortezomib exposure. Nonetheless, lenalidomide-exposed patients still had longer median PFS with DRd than with DVd (12.0 vs 5.4 months). Notably, of the 35 patients in our study who received daratumumab-containing 2L therapy, 15 received a combination with lenalidomide, despite prior lenalidomide exposure or resistance.

Our results with daratumumab-based regimens compare favorably to landmark phase 3 trials of daratumumab-based salvage regimens. Median PFS was 44.5 months with DRd in POLLUX ^7,8, 16.7 months with DVd in CASTOR ^9,10, 28.4 months with daratumumab, carfilzomib, and dexamethasone (DKd) in CANDOR ^11,12, and 24.4 months with DPd in APOLLO¹³. Notably, only 55% of patients in POLLUX were IMiD-exposed (18% to lenalidomide), and none were refractory, while 32% were lenalidomide-refractory in CANDOR compared to 79% in APOLLO and 82% in our study. The superior outcomes in our cohort may reflect differences in patient characteristics, including a younger median age (60 vs. 65, 64, 64 and 67 years in POLLUX, CASTOR, CANDOR, and APOLLO respectively) and fewer prior lines of therapy, as approximately half or more of the patients in these trials had received at least two prior regimens.

Median PFS from autoHCT (PFS1) in our cohort was 27.5 months, shorter than reported in the IFM2009¹⁴ (50 months) and DETERMINATION (67.5 months)¹⁵ trials, which established VRD induction followed by autoHCT and lenalidomide maintenance as standard frontline therapy. This discrepancy may partly be explained by selective inclusion of patients who had relapsed and required 2L therapy, thereby excluding those with longer remission. Even with a relatively shorter PFS1 in our patients, daratumumab-based regimens at relapse showed durable PFS2.

Our study has the expected limitations of a retrospective, single-center analysis, including selection bias, a small sample size, and heterogeneity of 2L treatments. MRD assessment was limited in our cohort, with post-transplant data unavailable for most patients, precluding meaningful evaluation of its impact on outcomes. It should be noted that all patients in our cohort were anti-CD38–naive at 2L therapy. In contemporary practice, patients receiving daratumumab as part of frontline induction, consolidation, or maintenance may experience different outcomes with daratumumab-based second-line regimens, and our findings may not directly apply to those settings16,17. Furthermore, we did not include the option of the CART cilta-cel, which has recently been approved for 2L treatment following the CARTITUDE-4 trial18. During earlier years of the study period, lenalidomide, daratumumab and pomalidomide were available primarily through clinical trials or compassionate use programs, before their general availability after FDA approval.

In conclusion, this study showed that in this relatively homogeneous population, daratumumab-based 2L regimens were associated with durable responses, and support daratumumab-based combinations as the preferred second-line option for anti-CD38–naive patients. Since all patients were naïve to anti-CD38 agents at relapse, the improved outcomes with daratumumab- containing regimens might also demonstrates the advantage of adding a new class of drug, as opposed to using same-class drugs, such as a different PI or IMiD.

Clinical Practice Points.

• Retrospective, single-center study of 146 multiple myeloma patients who received bortezomib, lenalidomide, and dexamethasone (VRD) induction, upfront autologous hematopoietic stem cell transplantation between 2005 and 2021, and lenalidomide maintenance before relapse.

• Following first progression, patients received either an immunomodulatory drug (IMiD) + proteasome inhibitor (PI)-containing triplet (31%) or doublet (21%), or daratumumab-based (24%) second line therapy.

• Recipients of daratumumab-based second line therapy had a higher rate of achieving at least a very good partial response (63%) compared to recipients of IMiD+PI triplet (35%) or doublet (34%) second line therapy.

• With a median follow-up of 35.8 months, median second progression-free survival was 59.9 months with daratumumab-based second line therapy, compared to 12.3 and 11.5 months with IMiD+PI triplet or doublet second line therapy.

• Findings support daratumumab-based combinations as preferred second line option for anti-CD38–naïve patients.

Data Availability Statement:

The data of this study are available upon request from the corresponding author.