Carfilzomib with immunomodulatory drugs for the treatment of newly diagnosed multiple myeloma

Carfilzomib, a selective proteasome inhibitor (PI), is approved for the treatment of patients with relapsed or refractory multiple myeloma (MM). Combination regimens incorporating a PI and immunomodulatory drug (IMiD) have been associated with deep responses and extended survival in patients with newly diagnosed MM (NDMM). Carfilzomib-based combinations with immunomodulators are being extensively studied in the frontline setting. The objective of this review was to describe efficacy and safety data for carfilzomib-based, PI/immunomodulatory combinations in NDMM. Information sources were articles indexed in PubMed and abstracts from key hematology/oncology congresses published between January 2012 and December 2018. PubMed and congresses were searched for prospective clinical studies assessing the combination of carfilzomib with an IMiD for NDMM treatment. Retrospective and preclinical reports, case reports/series, reviews, and clinical studies not evaluating carfilzomib–immunomodulator combinations in NDMM were excluded based on review of titles and abstracts. A total of nine articles and 72 abstracts were deemed relevant and included in the review. A total of six distinct carfilzomib-based, PI/immunomodulator combination regimens have been evaluated in 12 clinical trials. Overall, treatment with these regimens has resulted in deep responses, including high rates of negativity for minimal residual disease. These deep responses have translated to long progression-free survival and overall survival rates. Efficacy results for these regimens have generally been consistent across subgroups defined by age, transplant eligibility, and cytogenetic risk. The safety profile of carfilzomib in NDMM is consistent with that observed in the relapsed-refractory MM setting. Clinical studies have found that carfilzomib-based combinations with immunomodulators are highly active with a favorable safety profile in NDMM. The carfilzomib, lenalidomide, and dexamethasone (KRd) drug backbone is a promising foundation for treatment strategies aimed at achieving long-term, deep responses (functional cures) in the frontline setting. Several ongoing studies are evaluating KRd, with or without anti-CD38 monoclonal antibodies.


Introduction
There were 138,509 new multiple myeloma (MM) cases worldwide in 2016; in the United States, 30,770 projected cases occurred in 2018 [1,2]. The global incidence has increased sharply in recent decades, in part due to aging populations [1].
For newly diagnosed MM (NDMM) patients, the introduction of proteasome inhibitors (PIs) and immunomodulatory drugs (IMiDs) has improved treatment responses versus older therapies [3]. Several modern combination regimens incorporating a PI-IMiD backbone are considered recommended frontline regimens by the National Comprehensive Care Network and the European Society for Medical Oncology. These clinical regimens build off preclinical evidence showing that PIs and IMiDs have synergistic activity mediated by enhanced proteasome targeting, caspase activation, NF-κB inhibitory activity, and downregulation of CRBN/IRF4/MYC signaling and MCL1 [4][5][6]. These modern combination regimens have been associated with deep responses and improved progressionfree survival (PFS) and overall survival (OS) [7].
Given the established efficacy of carfilzomib in RRMM and the frontline potency of regimens incorporating a PI-IMiD backbone, recent NDMM clinical trials have evaluated combination regimens using carfilzomibbased, PI-IMiD combinations. This systematic review summarizes the efficacy and safety of these regimens for NDMM.

Methods
Studies were collected from a search of PubMed and abstracts from key hematology/oncology congresses published between January 2012 and December 2018. The following congresses were systemically searched between January 2012 and December 2018: American Society of Fig. 1 Mechanisms of action for carfilzomib [12][13][14][15][16][17]. ER endoplasmic reticulum, HLA human leukocyte antigen, MM multiple myeloma, NF-κB nuclear factor kappa-light-chain-enhancer of activated B cells, NK natural killer, UPR unfolded protein response PubMed and congresses were searched for prospective clinical studies that assessed the combination of carfilzomib with an IMiD in NDMM patients. Specific search terms included "carfilzomib," "Kyprolis," "MM," "thalidomide," "lenalidomide," "pomalidomide," "newly diagnosed," "transplant-eligible," and "transplant-ineligible." Three hundred and eighty-four articles were identified and were screened for relevancy to this review based on titles and abstracts. Retrospective and preclinical reports, case reports/ series, reviews, and clinical studies not evaluating frontline carfilzomib-immunomodulator combinations were excluded. Nine articles were deemed relevant for inclusion (Fig. 2a). Six hundred and fifty-five abstracts were identified, 72 of which were deemed relevant (Fig. 2b).

Activity of carfilzomib-IMiD combinations in NDMM
To date, six distinct carfilzomib-IMiD combinations have been assessed in 12 NDMM clinical trials . Key study information and efficacy data (including minimal residual disease [MRD] and MRD limit of detection) are summarized in Tables 1 and 2 and described below. All studies assessed twice-weekly carfilzomib unless stated otherwise.

Carfilzomib-lenalidomide-dexamethasone
In NDMM, KRd has been examined in six distinct clinical studies. In 2012, the MM Research Consortium (MMRC) reported findings from a phase 1/2 study of KRd (n = 53; transplant-eligible or -ineligible) [27]. Patients received KRd induction until disease progression or unacceptable toxicity for up to eight cycles (carfilzomib 20, 27, or 36 mg/m 2 in phase 1; 36 mg/m 2 in phase 2). Transplanteligible patients who achieved ≥partial response (PR) after cycle 4 had stem cell collection (SCC) and the option to proceed with autologous stem cell transplantation (ASCT), but per protocol, were to resume KRd after SCC. After eight cycles, patients received KRd maintenance for a maximum of 24 total KRd cycles. After four cycles, 38% of patients achieved ≥near complete response (nCR; defined as no detectable M protein on electrophoresis independent of immunofixation status, stable bone disease, and normal serum calcium concentration; phase 2 primary endpoint). At a median follow-up of 13 months (median treatment duration, 12 cycles), overall response rate (ORR) was 98% including 42% with a stringent complete response (sCR). Approximately two-thirds of patients who completed ≥8 cycles achieved sCR. Among patients with a complete response (CR/suspected CR), the MRDnegativity rate was 91% (multiparametric flow cytometry [MFC]). Notably, deep responses were obtained with KRd alone as the seven patients who proceeded to ASCT were censored for best response at the time of proceeding to transplant because they did not resume KRd. PFS rates were 97% and 92% at 12 and 24 months, respectively. There were no major efficacy differences between standard-and high-risk cytogenetics subgroups, although numbers were limited (standard risk, n = 34; high risk, n = 17). In a subgroup analysis of the MMRC study evaluating elderly patients (age ≥65 years; n = 23), the ≥very good PR (VGPR) rate was 91%; the sCR rate was 65% [28]. Updated MMRC study results (median follow-up, 4 years) showed that extended KRd treatment increased sCR rates from 42 to 51% [29]. The 2-and 4-year PFS rates were 92% and 69%, respectively [29]. The rating scheme is as follows: 1: properly powered and conducted randomized clinical trial; 2: Well-designed controlled trial without randomization    [30]. Patients received eight KRd cycles then lenalidomide maintenance. Carfilzomib (36 mg/m 2 ) was administered for the first 3 weeks of a 4-week cycle. Deep responses were observed, with 62% (NDMM) and 100% (SMM) of patients achieving ≥nCR [30]. The 1-year PFS rate in NDMM was 95% and all NDMM patients remained alive [30]. Thirtyfour of 45 enrolled NDMM patients (76%) achieved MRDnegative status (MFC). MRD-negativite ≥nCR rates by MFC were 100% (NDMM) and 92% (SMM); MRDnegativity rates by next-generation sequencing (NGS) were 67% and 75%, respectively [30]. MRD negativity was observed In hematopoietic progenitor cell grafts from 29/30 patients [31]. In a 5-year follow-up of the study (NDMM cohort), the MRD-negative CR rate was 62%; median time to progression was 67.3 months and the 6-year OS rate was 84% [32]. Depth of response and PFS were consistent regardless of age (≥65 vs <65 years) or cytogenetic risk [30,32]. The NIH Clinical Center study was recently reopened for enrollment with expanded sample size. Based on results from the MMRC and NIH Clinical Center studies, KRd is a treatment option for transplant-eligible and -ineligible patients according to the National Comprehensive Care Network guidelines.
Although extended KRd treatment without transplant in the MMRC study showed strong clinical benefit, some patients experienced progression [27]. Therefore, a second phase 2 MMRC study added ASCT to KRd to evaluate if further improvement of outcomes was possible [33,34]. In this study (n = 76), patients received four 28-day cycles of KRd induction followed by ASCT. Following ASCT, patients received four cycles of KRd consolidation and then KRd maintenance for 10 cycles. Carfilzomib was administered at 20/36 mg/m 2 . Response rates were 91% ≥VGPR, 78% ≥CR, and 75% sCR (primary endpoint) [33,34]. Postconsolidation MRD-negative ≥CR rates were 67% (NGS) and 95% (MFC) [34]. High rates of MRD negativity after consolidation were found to be sustained through to cycle 18 and after lenalidomide maintenance; this sustained MRD negativity (cycle 18) correlated with 3-year PFS and OS rates [34]. The 3-year PFS and OS rates for sustained MRD negativity patients by NGS were 94 and 100%, which were higher than those in the overall population (PFS, 86%; OS, 93%) [34]. Similar findings were observed in patients with high-risk disease (n = 27) [34].
The high rates of MRD negativity observed with KRd and integration of MRD assessments into standardized response criteria [3] provided rationale for an ongoing phase 1/2 study that is the first to evaluate MRD status as a guide for KRd treatment duration in NDMM patients [36]. During phase 1, patients receive carfilzomib (45 or 56 mg/m 2 ) with lenalidomide-dexamethasone to determine the maximum tolerated dose (MTD) of carfilzomib. Patients who achieve MRD negativity after any cycle receive two additional KRd cycles before treatment discontinuation, whereas MRD-positive patients continue therapy for ≤12 cycles, or until disease progression or unacceptable toxicity. The phase 2 primary endpoint is MRD negativity at the MTD. Phase 1 results established the carfilzomib MTD at 56 mg/m 2 . The phase 2 portion of the study is ongoing.
Inspired by the NIH Clinical Center study, the ongoing phase 2 GEM-CESAR trial is assessing KRd induction and consolidation with ASCT for SMM at high risk of progression. Preliminary post-ASCT efficacy results (69% ≥CR; 58% MRD negative) demonstrate the promise of this approach [39]. An EMN-MSKCC randomized phase 2 study comparing KRd versus Rd in high-risk SMM has recently started.
Overall, the safety profile of KRd in NDMM patients has been consistent with the profile observed in the ASPIRE study. In the FORTE trial, rates of treatment discontinuation due to adverse events (AEs) and fatal AEs were 8 and 1%, whereas in ASPIRE these rates were 15% and 7%, respectively (see Tables 3 and 4 for additional details) [24,38,40].
Carfilzomib-thalidomide-dexamethasone CARTHADEX was a phase 1/2 trial that evaluated carfilzomib-thalidomide-dexamethasone (KTd) as induction and consolidation therapy in transplant-eligible NDMM patients. During KTd induction, patients received four cycles of carfilzomib 20/27, 20/36, 20/45, or 20/56 mg/m 2 . In the overall population (N = 91), KTd induction resulted in an ORR of 90%, including a CR in 25% of patients and a ≥VGPR (primary endpoint) in 68% of patients [41]. Response rates improved with transplant and KTd consolidation (Table 2). After consolidation, patients with standard-and high-risk disease showed no difference in ≥VGPR rates and had similar CR rates. At a median followup of 23 months, the median PFS was not reached and the 3-year PFS rate was 72%. A subsequent update of the CARTHADEX trial assessed the impact of prolonged KTd induction therapy (eight cycles) at carfilzomib 56 mg/m 2 on depth of response [42]. Patients treated with prolonged KTd induction (n = 26) had higher ≥VGPR and ≥CR rates at end of induction compared with those treated with four KTd induction cycles at 56 mg/m 2 (n = 20). However, depth of response between the two groups was noted to be similar after completion of ASCT and consolidation [42]. Another update of the CARTHADEX trial found that postconsolidation sCR and VGPR rates were numerically higher for the 36-56 versus 27 mg/m 2 carfilzomib dose levels, but that PFS/OS were similar between these doses [43].
The safety of carfilzomib-thalidomide combinations (KTd and CYKLONE) was consistent with previously reported safety of both agents. Lack of overlapping toxicities (especially neuropathy) between these two agents allowed a favorable safety profile for these regimens.

Carfilzomib-cyclophosphamide-lenalidomidedexamethasone
In the phase 3 UK NCRI Myeloma XI trial, transplanteligible patients were randomized to the upfront PI-based quadruplet induction regimen of carfilzomib-cyclophosphamide-lenalidomide-dexamethasone (KCRd), or triplet induction therapy (cyclophosphamide-lenalidomidedexamethasone [CRd] or cyclophosphamide-thalidomidedexamethasone [CTd]) [48]. Patients in the triplet induction arms could have received a second induction with bortezomib-cyclophosphamide-dexamethasone (VCd) before ASCT depending on post first induction maximum response (≥VGPR, no second induction; PR or minor response, randomized to VCd or nothing; stable or progressive disease, VCd as second induction). Carfilzomib 20/36 mg/m 2 was administered twice weekly for the first 3 weeks of each 4-week cycle. A total of 1056 patients underwent induction randomization (CTd, n = 265; CRd, n = 265; KCRd, n = 526). Deeper responses with carfilzomib-based quadruplet versus IMiD-based triplet treatment were observed at the end of first induction: the ≥VGPR rate at end of initial induction was 82.3% for KCRd versus 52.8 and 64.9% for CTd and CRd, respectively. This difference in depth of response was sustained until ≥100 days post-ASCT.
The quadruplet combination was well tolerated in this large phase 3 trial [48]. Grade ≥3 neutropenia occurred in 16% of KCRd patients, compared with 13% for CTd and 22% for CRd. Grade ≥2 sensory neuropathy rates were lower in the KCRd and CRd arms compared with the CTd arm (~2% vs~10%). The incidences of anygrade cardiac failure and hypertension in the KCRd arm were both <2%. Carfilzomib-lenalidomide-dexamethasone plus daratumumab The phase 1b MMY1001 study evaluated KRddaratumumab in NDMM patients [49,50]. Patients were treated with this combination for ≤ 13 treatment cycles or until discontinuation for ASCT. Carfilzomib (20/70 mg/m 2 ) was administered once weekly. This combination (n = 21) resulted in an ORR of 100% with 91 and 57% of patients achieving ≥VGPR and ≥CR, respectively. The 1-year PFS rate was 95%. An ongoing phase 2 trial is evaluating MRD using blood and bone marrow-based assays after KRd-daratumumab treatment for NDMM. Preliminary results showed the feasibility of longitudinal MRD tracking in three patients who completed the MRD assays [51].
The phase 2 ASCENT trial (NCT03289299) is assessing the feasibility of KRd-daratumumab induction for highrisk SMM.

Safety of carfilzomib-IMiD combinations in NDMM
The safety profile of carfilzomib has been well established in RRMM, where it has shown to have manageable AEs [53] and a favorable benefit-risk profile given the observed PFS/OS advantage. In this setting, the most common grade 3/4 AEs associated with single-agent carfilzomib were primarily hematologic [54]. Cardiovascular events have also been associated with carfilzomib, with hypertension (mainly grades 1 and 2) being the most common type. In a pooled analysis of phases 1-3 RRMM clinical studies (n = 2044 carfilzomib-exposed patients), any-grade and grade ≥3 hypertension rates were 18.5% and 5.9%, respectively, and any-grade and grade ≥3 cardiac failure rates were 6.7% and 4.4%, respectively [55]. Carfilzomib-associated cardiovascular events have been found to be generally reversible and manageable, as reflected by relatively low carfilzomib discontinuation rates across studies [55][56][57]. Furthermore, the efficacy benefit of carfilzomib in terms of reduction in the risk of progression/death has been found to outweigh the risks observed with carfilzomib (including cardiac failure or hypertension) among RRMM patients [55].
Carfilzomib has been associated with low rates of peripheral neuropathy (PN) in RRMM, an AE associated with bortezomib that may impair QoL and limit treatment duration [58]. The low rate of PN with carfilzomib is consistent with preclinical results showing that in contrast to bortezomib, carfilzomib has minimal activity against nonproteasomal proteases, including the stress-induced, prosurvival protease HtrA2/Omi [8]. This selectivity may explain the difference in PN incidence between the carfilzomib and bortezomib arms in ENDEAVOR [19]. In ASPIRE, addition of carfilzomib to Rd did not increase PN rates [24]. The safety profile of carfilzomib-IMiD combination therapy in NDMM patients has been consistent with the known safety profile of carfilzomib. Below, we summarize the above AEs of interest observed in the large, randomized NDMM studies (FORTE and Myeloma XI). General safety and AEs of interest for all NDMM studies are shown in Tables 3 and 4.

Hematologic adverse events
Hematologic events were the most common AEs observed in frontline clinical trials of carfilzomib-IMiD combinations ( Table 4). The largest datasets come from the FORTE and Myeloma XI trials. In the FORTE trial, 18-22% of KRd patients had ≥1 grade ≥3 or serious hematologic AE [38]. Hematologic AE rates were comparable for KRd and KCd (18-22% vs 17%). In the KCRd arm of the Myeloma XI trial, rates of grade ≥3 neutropenia and thrombocytopenia were 16.4%, and 8.4%, respectively [48].

Peripheral neuropathy
Consistent with findings in RRMM, carfilzomib-based combination therapy has been associated with a low incidence of grade ≥2 PN in NDMM and a low rate of carfilzomib discontinuation or dose reductions due to PN (Table 4). In the KCRd arm of the Myeloma XI trial, grade ≥2 motor and sensory neuropathy rates were~3% each [48]. PN rates have not been reported for the FORTE trial.

Cardiovascular events
Cardiovascular events have been reported with carfilzomib-IMiD combinations in the frontline setting (Table 4). In the FORTE trial, the rate of grade 3/4 or serious cardiac events was 2-3% in the KRd arms and 3% in the KCd arm [38]. Hypertension (grade 3/4, or serious in severity) was reported in 8% of patients who received 12 KRd cycles and in 3% patients who received KRd or KCd with ASCT [38]. In the KCRd versus CRd versus CTd arms of the phase 3 Myeloma XI study, any-grade arrhythmia, heart failure, and hypertension were reported in approximately 2.3% versus 1% versus 1.5%, 1% versus 0% versus 0%, and <1% versus <1% versus <1% of patients [48].
Although cardiovascular events have been observed with carfilzomib, they appear to be manageable and reversible in the majority of cases [56]. Importantly, grade ≥3 heart failure rates are relatively low considering carfilzomib efficacy. Clinical experience with carfilzomib has aided development of strategies for managing and mitigating cardiovascular events [56,57]; clinical guidelines have been developed in collaboration with cardiologists to minimize cardiotoxicity with carfilzomib treatment [59].

Conclusions
Carfilzomib-based, PI-IMiD combination therapies have been shown to be highly efficacious with a favorable safety profile in NDMM, as evidenced by the data from several studies (single-arm and randomized) reviewed above. These regimens elicit deep and durable responses, including MRD negativity, which have translated to high PFS and OS rates. Furthermore, a consistent benefit for these regimens has been observed across patient segments, including in both transplant-eligible and -ineligible patients, across age subgroups, and regardless of cytogenetic risk. To date, results have been reported from only one randomized phase 3 study of carfilzomib-based IMID combination therapy in NDMM, and there are currently no phase 3 data for how these therapies directly compare with the most recent frontline standards of care. Future data from such studies will be important for evaluating the role of carfilzomib-based, PI-IMiD combination therapy in NDMM. Bortezomib-based, PI-IMiD triplet combinations are considered standards of care in the frontline setting. For example, VRd is considered a preferred regimen by the NCCN for transplant-eligible and -ineligible patients based on data showing that this regimen improved PFS and OS compared with Rd. Despite the absence of controlled and randomized trials, carfilzomib-based, PI-IMiD combinations may offer several advantages relative to this standard of care. For some patients, PN associated with bortezomib may interrupt treatment, resulting in suboptimal response and QoL. These patients may benefit from a carfilzomibbased PI-IMiD regimen. Although there are currently no head-to-head data of carfilzomib-versus bortezomib-based, PI-IMiD therapy in NDMM, KRd (without or without ASCT) is one of the most active frontline regimens observed in NDMM studies [27-36, 38-40, 60]. Preliminary data from the observational CoMMpass study (NCT01454297) suggest that improved response rates and longer event-free survival are associated with KRd versus VRd in NDMM [61], while other preliminary data from CoMMpass and the Flatiron observational study suggest longer treatment duration with VRd [62]. However, these findings need to be validated in a prospective manner. The ongoing, randomized phase 3 ENDURANCE study (NCT01863550; N = 1000+), which is evaluating KRd versus VRd, will provide important head-to-head data regarding the benefit-risk profiles of carfilzomib-versus bortezomib-based PI-IMiD therapy in NDMM, with the limitation of capping carfilzomib treatment duration to nine cycles.
Carfilzomib has been associated with manageable and reversible cardiovascular events in RRMM, and these events merit monitoring in NDMM. In a meta-analysis of NDMM and RRMM patients (N = 2594), any-grade and grade ≥3 congestive heart failure rates were 4.1% and 2.5%, respectively [63].
The introduction of new antimyeloma agents has significantly improved the outlook of MM patients; however, MM remains an incurable disease. To achieve long-term remissions or even functional cure for NDMM patients, the most potent agents need to be used upfront. Based on favorable clinical outcomes of carfilzomib therapies in the relapsed setting, we believe that a carfilzomib-IMiD combination will also benefit NDMM patients. Ongoing frontline studies are evaluating the impact of adding new drugs with novel mechanisms, such as monoclonal antibodies, to the carfilzomib-IMiD combination to achieve functional cures in MM. from Amgen, ABBVIE, BMS, Celgene, Janssen, Karyopharm, Millennium, Takeda, Sanofi, and SkylineDx. Dr Mohty has no disclosures to report. Dr Iskander and Dr Mezzi are employees of and own stock in Amgen, Inc. Dr Siegel reports honoraria and consulting or advisory role fees for Celgene, Amgen, Merck, Janssen, BMS, Takeda, and Karyopharm; speakers' bureau participation for Celgene, Amgen, Merck, Janssen, BMS, and Takeda; and research funding from Celgene.
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons. org/licenses/by/4.0/.