Daratumumab plus lenalidomide and dexamethasone in relapsed/refractory multiple myeloma: extended follow-up of POLLUX, a randomized, open-label, phase 3 study

In POLLUX, daratumumab (D) plus lenalidomide/dexamethasone (Rd) reduced the risk of disease progression or death by 63% and increased the overall response rate (ORR) versus Rd in relapsed/refractory multiple myeloma (RRMM). Updated efficacy and safety after >3 years of follow-up are presented. Patients (N = 569) with ≥1 prior line received Rd (lenalidomide, 25 mg, on Days 1–21 of each 28-day cycle; dexamethasone, 40 mg, weekly) ± daratumumab at the approved dosing schedule. Minimal residual disease (MRD) was assessed by next-generation sequencing. After 44.3 months median follow-up, D-Rd prolonged progression-free survival (PFS) in the intent-to-treat population (median 44.5 vs 17.5 months; HR, 0.44; 95% CI, 0.35–0.55; P < 0.0001) and in patient subgroups. D-Rd demonstrated higher ORR (92.9 vs 76.4%; P < 0.0001) and deeper responses, including complete response or better (56.6 vs 23.2%; P < 0.0001) and MRD negativity (10–5; 30.4 vs 5.3%; P < 0.0001). Median time to next therapy was prolonged with D-Rd (50.6 vs 23.1 months; HR, 0.39; 95% CI, 0.31–0.50; P < 0.0001). Median PFS on subsequent line of therapy (PFS2) was not reached with D-Rd versus 31.7 months with Rd (HR, 0.53; 95% CI, 0.42–0.68; P < 0.0001). No new safety concerns were reported. These data support using D-Rd in patients with RRMM after first relapse.


Introduction
The development of firstand second-generation novel agents over the past decade has led to the acceptance of immunomodulatory drug (IMiD)-based or proteasome inhibitor (PI)-based doublet or triplet therapy as standard of care for newly diagnosed and relapsed or refractory multiple myeloma (RRMM) [1,2]. However, multiple myeloma (MM) remains an incurable disease, underscoring the need for new treatment strategies. Optimal combination and sequencing of these next-generation agents remain to be defined, particularly among various patient subgroups.
Daratumumab has demonstrated single-agent activity in heavily pretreated RRMM and in combination with standard-of-care regimens in RRMM after at least one prior therapy [10][11][12][13]. Daratumumab safety and efficacy are also established in combination with bortezomib, melphalan, and prednisone, and with lenalidomide and dexamethasone in patients with newly diagnosed MM (NDMM) who are transplantation ineligible [14,15].
In the prespecified interim analysis of the phase 3 POLLUX study (median follow-up, 13.5 months), daratumumab in combination with lenalidomide and dexamethasone (D-Rd) reduced the risk of disease progression or death by 63% (median progression-free survival [PFS] not reached vs 18.4 months; hazard ratio [HR], 0.37; 95% confidence interval [CI], 0.27-0.52; P < 0.001) and significantly increased the overall response rate (ORR) compared with lenalidomide and dexamethasone (Rd) alone (93 vs 76%; P < 0.001) in patients with at least one prior therapy [12]. In an updated analysis after a longer follow-up of 25.4 months, the PFS benefit of D-Rd was maintained compared with Rd, with median PFS still not being reached for D-Rd versus 17.5 months for Rd. In addition, deep and durable responses were achieved with D-Rd, and the PFS benefit of D-Rd versus Rd was consistently maintained regardless of the number of prior lines of therapy received, prior IMiD exposure, bortezomib refractoriness, time since last therapy, or cytogenetic risk [16].
We report the long-term efficacy and safety analyses of POLLUX after a median follow-up of more than 3.5 years.

Study design and patients
POLLUX is an ongoing, randomized, open-label, multicenter, phase 3 study in patients with RRMM (Clin-icalTrials.gov Identifier: NCT02076009). An independent ethics committee or institutional review board at each site approved the trial, and all patients provided written informed consent. The study protocol was conducted in accordance with the principles of the Declaration of Helsinki and the International Conference on Harmonisation Good Clinical Practice guidelines. The study design, primary results, and post hoc secondary analyses have been previously reported [12,16]. Briefly, eligible patients had progressive disease (according to International Myeloma Working Group [IMWG] criteria) [17,18] during or after their last regimen, received and responded to at least one prior line of therapy, and had a creatinine clearance ≥30 mL/min. Prior lenalidomide exposure was allowed, but patients with lenalidomide-refractory disease were excluded from participation. A total of 569 patients were randomly assigned (1:1) using an interactive web response system to Rd (lenalidomide: 25 mg orally on Days 1-21 of each 28day cycle; dexamethasone: 40 mg orally weekly) with or without daratumumab (16 mg/kg intravenous weekly for 8 weeks, every 2 weeks for 16 weeks, and every 4 weeks thereafter) until progression. The randomization was balanced by using randomly permuted blocks and was stratified by International Staging System (ISS), number of prior lines of therapy (1 vs 2 or 3 vs >3), and prior lenalidomide exposure. Treatment assignments were not blinded.

Endpoints and assessments
The primary efficacy endpoint was PFS. Secondary efficacy endpoints included ORR, rates of very good partial response (VGPR) or better and complete response (CR) or better, minimal residual disease (MRD), time to response, duration of response, and overall survival (OS). PFS on subsequent line of therapy (PFS2) was an exploratory endpoint and was defined as the time from randomization to progression after the next line of subsequent therapy or death.
Exploratory post hoc secondary analyses evaluated patient subgroups according to prior lines of therapy (1 and 1-3), prior lenalidomide treatment, refractoriness to bortezomib, and achievement of CR or better. The number of prior lines of therapy was determined by investigators according to the IMWG consensus guidelines [18]. PFS, ORR, and MRD negativity were assessed for each subgroup.
MRD was assessed at the time of suspected CR and at 3 and 6 months after confirmed CR (and every 12 months thereafter if CR was maintained) using clonoSEQ® V2.0 (Adaptive Biotechnologies, Seattle, WA, USA). To allow for stringent, unbiased MRD evaluation, the entire intent-totreat (ITT) population was evaluated, and patients were considered MRD positive if they had MRD-positive test results or no MRD assessment.

Statistical analyses
Statistical analyses have been described previously [12]. Approximately 560 patients were randomized to observe 295 PFS events to detect an HR of 0.7 for the D-Rd group relative to the Rd group with 85% power at a two-sided significance level of 0.025, using a group sequential testing design.
PFS was compared between treatment groups based on a stratified log-rank test. HRs and 95% CIs were estimated using a stratified Cox regression model with treatment as the sole explanatory variable, and the Kaplan-Meier method was used to estimate the distributions. Stratified Cochran-Mantel-Haenszel tests were used to test treatment differences in ORRs and rates of VGPR or better and CR or better. MRD-negative rates were compared between groups using a Fisher's exact test and the likelihoodratio test.

Results
Between June 16, 2014 and July 14, 2015, 569 patients at 135 sites in 18 countries across North America, Europe, and the Asia Pacific region were randomly assigned in POLLUX; 286 patients were assigned to D-Rd and 283 patients were assigned to Rd. Baseline patient demographics, prior treatment history, and other clinical and cytogenetic characteristics have been previously published [12,16] and are summarized in Supplementary Table 1. Approximately half of patients (52%) had received one prior line of therapy, 18% had received prior lenalidomide, 44% had received prior IMiD and PI, and 21% were refractory to bortezomib.  Fig. 1e). The PFS benefit of D-Rd versus Rd was also maintained in patients who received two or three lines of prior therapy, and in subgroups based on cytogenetic risk status, age, type of MM, ISS disease stage, Eastern Cooperative Oncology Group performance status score, baseline renal or hepatic function, prior treatment exposure, and refractory status (Fig. 2).

Safety
No new safety concerns were reported in either treatment group with longer follow-up. The most common treatmentemergent AE (TEAE) was neutropenia, occurring in 63.3% of patients treated with D-Rd and 48.0% of patients who received Rd ( Table 2). The most common (≥5%) grade 3/4 TEAEs observed with D-Rd and Rd included neutropenia, febrile neutropenia, anemia, thrombocytopenia, lymphopenia, pneumonia, diarrhea, fatigue, hypokalemia, and cataracts ( Table 2). The percentage of patients with TEAEs leading to treatment discontinuation was similar between groups (D-Rd, 14.8; Rd, 14.6%). The most common TEAEs (≥1%) leading to treatment discontinuation with D-Rd versus Rd were pneumonia (1.8 vs 1.1%), pulmonary embolism (0 vs 1.1%), septic shock (1.1 vs 0%), and general physical health deterioration (1.1 vs 0%), respectively. The incidence of second primary malignancies was similar between groups, occurring in 8.5% of patients who received D-Rd and 8.9% of patients who received Rd (Table 3).    [17,18]. The analysis included patients who had a confirmed diagnosis of MM and measurable disease at baseline or screening. In addition, patients had received at least one administration of trial treatment and had at least one disease assessment after the baseline visit.
b Intent-to-treat population. c P value was calculated using the Cochran-Mantel-Haenszel chisquare test. d Clinical benefit includes all patients with minimal response, partial response, VGPR, CR, and stringent CR. e Criteria for a stringent CR include the criteria for a CR plus a normal free light-chain ratio and the absence of clonal plasma cells as assessed by immunohistochemical or immunofluorescence analysis or by flow cytometry. f Includes patients who achieved a minimal response. g P value was calculated using the Fisher's exact test.
Patients with one prior line of therapy gained the greatest clinical benefit with D-Rd, resulting in a 58% reduction in the risk of disease progression or death compared with Rd. Consistent findings were observed in CASTOR, in which patients who received one prior line of therapy demonstrated the greatest clinical benefit with daratumumab plus bortezomib and dexamethasone (D-Vd) versus bortezomib and dexamethasone (Vd) alone (78% reduction in the risk of disease progression or death), regardless of prior treatment with either lenalidomide or bortezomib [19].
In POLLUX, patients who were refractory to lenalidomide were excluded from the study. However, D-Rd demonstrated improved efficacy outcomes, including prolonged PFS and improved depth of response in patients who received prior lenalidomide but were not refractory to the drug. Furthermore, D-Rd prolonged PFS versus Rd in poor prognostic patient subgroups, including those with ISS stage III disease, patients who were refractory to their last prior line of therapy, and patients with high cytogenetic risk abnormalities, although to a lesser extent in comparison with other patient subgroups evaluated.
Although cross-study comparisons must take into account differences in study population and design, the median PFS observed with D-Rd (44.5 months) is unprecedented in the RRMM treatment setting. In the phase 3 ASPIRE study of carfilzomib plus Rd (KRd) compared with Rd alone in patients with relapsed MM and one to three prior treatments, median PFS was 26.1 versus 16.6 months with KRd and Rd, respectively (HR, 0.66; 95% CI, 0.55-0.78; P < 0.001), at a median follow-up of 48.8 months for KRd and 48.0 months for Rd [20]. With longer follow-up (median 67.1 months), median OS was 48.3 months for KRd versus 40.4 months for Rd, resulting in 21% reduction in the risk of death (HR, 0.79; 95% CI, 0.67-0.95; P = 0.0045) [20]. In the phase 3 TOURMALINE-MM1 study in patients with RRMM and one to three prior therapies, median PFS was 20.6 months with ixazomib in combination with Rd (IRd) versus 14.7 months with Rd alone (HR, 0.74; 95% CI, 0.59-0.94; P = 0.01), at a median follow-up of 14.8 versus 14.6 months in the IRd and Rd groups, respectively [21]. Median PFS in the phase 3 ELOQUENT-2 study in patients with RRMM and one to three prior therapies was 19.4 months with elotuzumab plus Rd compared with 14.9 months with Rd alone (HR, 0.73; 95% CI, 0.60-0.89; P = 0.0014), with 3 years of extended follow-up [22]. D-Rd induced deep and durable responses that continued to deepen over time with longer follow-up. At the most recent clinical cutoff, rates of VGPR or better (80.4 vs 49.3%) and CR or better (56.6 vs 23.2%) with D-Rd versus Rd were higher than those observed at primary analysis (VGPR or better: 75.8 vs 44.2%; CR or better: 43.1 vs 19.2%) [12] and at~2 years of follow-up (VGPR or better: 78.6 vs 47.8%; CR or better: 51.2 vs 21.0%; Supplementary  Fig. 1a) [16]. MRD-negative rates have also continued to deepen with D-Rd versus Rd over time (interim analysis: 22.4 vs 4.6% [12]; 2-year follow-up: 26.2 vs 6.4% [16]; current analysis: 30.4 vs 5.3%; Supplementary Fig. 1b), while the rates of MRD negativity with Rd have remained relatively constant with longer follow-up. It is important to note that an updated next-generation sequencing assay with improved calibration rate was used to determine MRD negativity in the current study. PFS was prolonged with PFS2 may serve as a surrogate endpoint for OS when survival data are not available [23]. The use of PFS2 has been suggested as a preferred endpoint, particularly for studies investigating long-term maintenance treatment [23][24][25]. The findings presented here demonstrate that D-Rd significantly prolongs the time to subsequent therapy and PFS2 versus Rd, conferring a 47% reduction in the risk of disease progression or death on the next line of therapy. These data indicate that treatment with D-Rd does not negatively impact on patient outcomes on subsequent therapy. At the time of this analysis, 17 more deaths occurred with Rd (121/283) compared with D-Rd (104/ 286). Follow-up for OS in POLLUX is ongoing, with the final analysis planned after 330 deaths.   With longer follow-up, the safety profile of D-Rd and Rd remains largely consistent with the known safety profiles of daratumumab [26] and of Rd [20,27,28]. Despite the higher rates of neutropenia and infections (upper respiratory tract infection and pneumonia), the rates of grade 3 or 4 infections were similar between treatment groups and were managed according to local institutional treatment standardof-care protocols. Consistent tolerability was also observed with daratumumab in the phase 3 MAIA study of D-Rd versus Rd in patients with NDMM who are ineligible for transplantation (median follow-up, 28 months) [15].
Taken together, the results from >3.5 years of median follow-up demonstrate that D-Rd continues to provide significant PFS benefit and induces deeper and more durable responses, including a greater than five-fold increase in the rate of MRD negativity versus Rd alone in patients with RRMM. No new safety concerns were observed following a median of 34 months of D-Rd exposure. These updated findings continue to support the use of D-Rd in patients with RRMM after first relapse.