Original Study Carfilzomib-Dexamethasone Versus Bortezomib- Dexamethasone in Relapsed or Refractory Multiple Myeloma: Updated Overall Survival, Safety, and Subgroups Robert Z. Orlowski,1 Philippe Moreau,2 Ruben Niesvizky,3 Heinz Ludwig,4 Albert Oriol,5 Wee Joo Chng,6 Hartmut Goldschmidt,7 Zhao Yang,8 Amy S. Kimball,9 Meletios Dimopoulos10 Abstract In this updated analysis of patients with relapsed/refractory multiple myeloma (RRMM) from the RandomizEd, OpeN Label, Phase 3 Study of Carfilzomib Plus DExamethAsone Vs Bortezomib Plus DexamethasOne in Pa- tients With Relapsed Multiple Myeloma (ENDEAVOR) trial, clinically meaningful overall survival improvements continue to be observed with carfilzomib 56 mg/m2 and dexamethasone (Kd56; n [ 464) versus bortezomib and dexamethasone (n [ 465), including in key patient subgroups. With longer-term data, the favorable benefit-risk profile of Kd56 continues to support its use as a standard-of-care in RRMM. Introduction: The phase III RandomizEd, OpeN Label, Phase 3 Study of Carfilzomib Plus DExamethAsone Vs Bor- tezomib Plus DexamethasOne in Patients With Relapsed Multiple Myeloma (ENDEAVOR) trial showed significantly improved progression-free survival and overall survival (OS) with carfilzomib (56 mg/m2) and dexamethasone (Kd56) versus bortezomib and Kd56 (Vd) in patients with relapsed or refractory multiple myeloma (RRMM). We report updated OS and safety data after 6 months of additional follow-up. Patients and Methods: Patients with RRMM (1-3 previous lines of therapy) were randomized 1:1 to Kd56 or Vd. Median OS was estimated using the KaplaneMeier method; OS was compared between treatment groups using Cox proportional hazards models. Results: As of July 19, 2017, median follow-up was 44.3 months for Kd56 and 43.7 months for Vd. Median OS was 47.8 months (Kd56) versus 38.8 months (Vd; hazard ratio, 0.76; 95% confidence interval, 0.633-0.915). OS was longer with Kd56 versus Vd within age and cytogenetic subgroups, and according to number of previous lines of therapy, previous bortezomib exposure, previous lenalidomide exposure, and lenalidomide-refractory status. Exposure-adjusted incidences per 100 patient- years of adverse events (AEs) were 1352.07 for Kd56 and 1754.86 for Vd; for Grade 3 AEs, these values were 162.31 and 175.90. Conclusion: With median follow-up of approximately 44 months, clinically meaningful improve- ments in OS were observed with Kd56 versus Vd, including in all subgroups examined. The Kd56 safety profile was consistent with previous analyses. Clinical Lymphoma, Myeloma & Leukemia, Vol. 19, No. 8, 522-30 ª 2019 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Keywords: Clinical outcomes, Efficacy, Phase III, Proteasome inhibitor 1 8 Department of Lymphoma and Myeloma, The University of Texas M.D. Anderson Department of Biostatistics, Amgen, Inc, Thousand Oaks, CA 9 Cancer Center, Houston, TX Department of Clinical Research, Amgen, Inc, Thousand Oaks, CA 2 10 Department of Hematology, University Hospital Hotel-Dieu, Nantes, France Department of Clinical Therapeutics, University Athens School of Medicine, Athens, 3 Department of Medical Oncology, Myeloma Center, New York Presbyterian Greece Hospital-Weill Cornell Medical Center, New York, NY 4 Department of Medicine I, Wilhelminen Cancer Research Institute, Wilhelminens- Submitted: Jan 10, 2019; Revised: Apr 11, 2019; Accepted: Apr 29, 2019; Epub: May pital, Vienna, Austria 2, 2019 5 Department of Clinical Hematology, Institut Catala d’Oncologia and Institut Josep Carreras, Hospital Germans Trias i Pujol, Barcelona, Spain Address for correspondence: Robert Z. Orlowski, MD, PhD, Department of Lym- 6 phoma and Myeloma, The University of Texas M.D. Anderson Cancer Center, 1515 Department of Hematology Oncology, National University of Singapore, Singapore 7 Holcombe Blvd, Unit 429, Houston, TX 77030 Department of Hematology Oncology, Heidelberg University Clinic and the National Center of Tumor Diseases, Heidelberg, Germany Fax: 713-563-5067; e-mail contact: ROrlowski@mdanderson.org - 2152-2650/ª 2019 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). 522 Clinical Lymphoma, Myeloma & Leukemia August 2019 https://doi.org/10.1016/j.clml.2019.04.018 Introduction At the time of the ENDEAVOR (NCT01568866) study design, As treatment options for multiple myeloma (MM) increase, standard therapy for MM that included a PI consisted of fixed- treatment paradigms are changing. The use of novel therapies has duration therapy. Fixed-duration therapy is appropriate when resulted in improved outcomes; however, patients with MM usually such therapy is curative, or when exposure above a threshold causes relapse1 and improving outcomes obtained with subsequent lines of toxicity. Because RRMM is incurable, and the dose-limiting toxicity therapy, particularly across the heterogeneous relapsed or refractory for the comparator arm is sensory neuropathy, it was considered MM (RRMM) patient population, remains a challenge. The “gold ethical to allow a continuous therapy design. Continuous therapy standard” therapeutic goal for any agent used in oncology is with carfilzomib (56 mg/m2) and dexamethasone (Kd56) versus extended overall survival (OS).2 To date, only 2 MM clinical trials, bortezomib and dexamethasone (Vd), with stratification on the basis the RandomizEd, OpeN Label, Phase 3 Study of Carfilzomib Plus of previous exposure to PI, was deemed balanced. The study design DExamethAsone Vs Bortezomib Plus DexamethasOne in Patients maximized treatment effect for each agent, allowing a true com- With Relapsed Multiple Myeloma (ENDEAVOR) trial and the parison of efficacy of these 2 PIs. The primary end point of CArfilzomib, Lenalidomide, and DexamethaSone versus Lenalido- progression-free survival (PFS) and secondary end point of OS were mide and Dexamethasone for the treatment of PatIents with met.3,4 In a prespecified interim analysis, the median OS was 47.6 Relapsed Multiple MyEloma (ASPIRE) trial, have shown a statis- months (95% confidence interval [CI], 42.5 to not estimable [NE]) tically significant improvement in OS beyond that observed using a in the Kd56 group versus 40.0 months (95% CI, 32.6-42.3) in the recent standard-of-care regimen containing either lenalidomide or Vd group (hazard ratio [HR], 0.791; 95% CI, 0.648-0.964; 1-sided the first-generation proteasome inhibitor (PI) bortezomib.3-5 P ¼ .010).3 Figure 1 Trial Profile Clinical Lymphoma, Myeloma & Leukemia August 2019 - 523 Updated Survival and Subgroup Analysis of ENDEAVOR consent, or death. Safety follow-up continued for all patients Figure 2 KaplaneMeier Curve of Overall Survival (Intent-to- Treat Population). The Curve Was Truncated at the receiving treatment and through 30 days after treatment Time Point When Only 10 Subjects (Kd56 and Vd discontinuation. Combined) Were at Risk The institutional review board/ethics committee at each partici- pating site approved the study, and the protocol conformed to Good Clinical Practice Guidelines and the principles of the Declaration of Helsinki. All participants provided written informed consent. Data Sharing Qualified researchers may request data from Amgen clinical studies. Complete details are available at: http://www.amgen.com/ datasharing Assessments Cytogenetic analyses were carried out using fluorescence in situ hybridization (FISH) to determine the risk status of patients. Those defined as being at “high risk” were patients with the genetic sub- types t(4;14) or t(14;16) in 10% of screened plasma cells, or with del(17p) in 20% of screened plasma cells. All other patients with available and known baseline cytogenetics were included in the “standard risk” subgroup. Patients with a FISH assessment whose data could not be analyzed or did not yield a definitive result were included in the “unknown/missing” cytogenetics subgroup.6 The severity of adverse events (AEs) was classified according to Abbreviations: HR ¼ hazard ratio; Kd56 ¼ carfilzomib (56 mg/m2) and dexamethasone; the National Cancer Institute Common Terminology Criteria for OS ¼ overall survival; Vd ¼ bortezomib and dexamethasone. Adverse Events version 4.03, and AEs were coded according to the Medical Dictionary for Regulatory Activities (MedDRA) version 20.0. For selected AEs of interest (acute renal failure, cardiac fail- The ENDEAVOR study remained open to fulfill a postmarket- ure, peripheral neuropathy, ischemic heart disease, and hemato- ing requirement that the sponsor submits a report of safety and poietic thrombocytopenia), standardized MedDRA query narrow efficacy outcomes with at least 3 years of follow-up data on long- scope (SMQN) grouped terms were applied. Patients who dis- term treatment. In July 2017, all ENDEAVOR patients had continued treatment before disease progression were followed-up received 3 years of follow-up. Herein we report updated exposure every 4 weeks for disease status until progression. After disease and survival data from the 3-year data cutoff. progression, patients were followed for survival status every 3 months. Patients and Methods Study Design and Patients Outcomes The ENDEAVOR trial design has been described previously.3,4 The primary end point of ENDEAVOR was PFS. Secondary end Briefly, ENDEAVOR was a phase III, randomized, open-label points were OS, overall response rate, duration of response, inci- study of adult patients with RRMM who had received 1 to 3 pre- dence of Grade 2 peripheral neuropathy, and safety. The final vious lines of therapy. PFS, overall response, and duration of response results were re- Patients were randomized 1:1 to receive Kd56 or Vd. Random- ported in the first interim analysis of ENDEAVOR (data cutoff ization was stratified in accordance with several baseline factors: November 10, 2014).4 Subsequently, final OS data were reported International Staging System stage (I vs. II-III); lines of previous from a prospectively planned interim OS analysis (data cutoff treatment (1 vs. 2 or 3 lines); previous PI treatment (previous vs. no January 3, 2017).3 Herein we report updated OS and safety data previous carfilzomib or bortezomib treatment); and planned route using a more recent data cutoff of July 19, 2017, including OS data of bortezomib administration (intravenous [IV] vs. subcutaneous). from subgroup analyses. Subgroups included patient age, cytoge- In the Kd56 group, patients received carfilzomib as an IV infu- netic risk group, previous therapy received, and number of lines of sion over 30 minutes on days 1, 2, 8, 9, 15, and 16 of 28-day cycles. previous treatment. The carfilzomib dose was 20 mg/m2 on days 1 and 2 of cycle 1, followed by 56 mg/m2 thereafter. Dexamethasone 20 mg was Statistical Analysis administered orally or IV on days 1, 2, 8, 9, 15, 16, 22, and 23. Median OS was estimated using the KaplaneMeier method. CIs Patients in the Vd group received bortezomib 1.3 mg/m2 as an IV for the median were estimated using the method by Klein and bolus or subcutaneous injection on days 1, 4, 8, and 11 of each 21- Moeschberger with log-log transformation.7 For the comparison of day cycle. Dexamethasone (20 mg; oral or IV) was administered on OS between treatment groups, HRs and corresponding 95% CIs days 1, 2, 4, 5, 8, 9, 11, and 12. Treatment was given until disease were estimated using stratified or unstratified Cox proportional progression, physician decision, unacceptable toxicity, withdrawal of hazards models for the primary intent-to-treat (ITT) population and 524 - Clinical Lymphoma, Myeloma & Leukemia August 2019 Robert Z. Orlowski et al Figure 3 KaplaneMeier Curves of Overall Survival (OS) According to Subgroup. (A) Age; (B) Cytogenetic Risk; (C) Number of Previous Lines of Therapy; (D) Previous Bortezomib Treatment; and (E) Previous Lenalidomide Treatment. The Curves Were Truncated at the Time Point When Only 10 Subjects (Kd56 and Vd Combined) Were at Risk Abbreviations: HR ¼ hazard ratio; Kd56 ¼ carfilzomib (56 mg/m2) and dexamethasone; NE ¼ not estimable; Vd ¼ bortezomib and dexamethasone. Clinical Lymphoma, Myeloma & Leukemia August 2019 - 525 Updated Survival and Subgroup Analysis of ENDEAVOR Figure 3 continued subgroup OS analyses, respectively. Exposure-adjusted patient in- Median follow-up time for OS was 44.3 months in the Kd56 group cidences per 100 patient-years were calculated. Total person-time in and 43.7 months in the Vd group. The median OS for the ITT each treatment group was the sum of the time to first treatment- population was 9.0 months longer in the Kd56 group (median, 47.8 emergent AE for each patient, or the entire time of exposure to months; 95% CI, 41.9-NE) versus the Vd group (median, 38.8 study drug if a patient had no event. months; 95% CI, 31.7-42.7), resulting in an HR of 0.76 (95% CI, 0.63-0.92; 1-sided P ¼ .0017; Figure 2). Results Patients Subgroup Analyses of OS Between June 20, 2012, and June 30, 2014, 929 eligible patients KaplaneMeier curves of OS according to patient subgroups are were enrolled in ENDEAVOR and randomly assigned to the Kd56 shown in Figure 3. Median OS was longer in the Kd56 versus Vd group (n ¼ 464) or the Vd group (n ¼ 465; Figure 1). groups, regardless of patient age. In patients aged 75 years, me- dian OS was 36.1 versus 23.9 months (HR, 0.78; 95% CI, 0.51- Overall Survival 1.20; Kd56, n ¼ 77; Vd, n ¼ 66). For patients aged 65 to 74 years, The number of patients alive at the cutoff on July 19, 2017, was median OS was 49.0 versus 36.2 months (HR, 0.71; 95% CI, 214 (46.1%) in the Kd56 group and 168 (36.1%) in the Vd group. 0.52-0.96; Kd56, n ¼ 164; Vd, n ¼ 189), and for those aged <65 526 - Clinical Lymphoma, Myeloma & Leukemia August 2019 Robert Z. Orlowski et al Figure 3 continued years, median OS was 47.8 versus 42.2 months (HR, 0.79; 95% Cytogenetic risk status was evaluated for 381/464 (82%) patients CI, 0.60-1.03; Kd56, n ¼ 223; Vd, n ¼ 210). To assess potential in the Kd56 group and 404/465 (87%) patients in the Vd group. reasons for the greater difference in median OS between treatment When assessed according to cytogenetic risk status, median OS was arms in the older patient subgroups compared with the <65-year- longer in the Kd56 group versus the Vd group for high-risk patients old subgroup, subsequent salvage regimens were evaluated in each (median OS, 28.0 vs. 22.7 months; HR, 0.81; 95% CI, 0.58-1.14; age subgroup. The percentage of patients treated with at least 1 Kd56, n ¼ 97; Vd, n ¼ 113) and standard-risk patients (median subsequent antimyeloma salvage therapy was similar between OS, NE vs. 43.5 months; HR, 0.79; 95% CI, 0.62-1.01; Kd56, Kd56- versus Vd-treated patients in each age subgroup n ¼ 284; Vd, n ¼ 291). (see Supplemental Table 1 in the online version). The percentage of Median OS was longer in the Kd56 group, regardless of the patients treated with specific antimyeloma salvage therapies, number of lines of previous therapy received. In patients in the including lenalidomide, cyclophosphamide, and pomalidomide, Kd56 (n ¼ 231) versus Vd (n ¼ 229) groups who had received 1 were also similar between Kd56- versus Vd-treated patients across previous line of therapy, the median OS was 51.3 versus 43.7 age subgroups. Bortezomib was the only subsequent antimyeloma months, respectively (HR, 0.77; 95% CI, 0.58-1.02). Similarly, in therapy that showed a >10% difference between the Kd56 and Vd patients previously treated with 2 to 3 lines, median OS was 39.5 arms within each age group, but these differences were similar versus 28.4 months in the Kd56 (n ¼ 233) versus Vd (n ¼ 236) across age groups. groups, respectively (HR, 0.75; 95% CI, 0.59-0.96). Clinical Lymphoma, Myeloma & Leukemia August 2019 - 527 Updated Survival and Subgroup Analysis of ENDEAVOR Table 1 Summary of Adverse Events in the Safety Population Kd56 (n [ 463) Vd (n [ 456) Any Adverse Event, n (%) 457 (98.7) 451 (98.9) Exposure-Adjusted Incidence of Adverse Events, Rate Per 100 Patient-Years (95% CI) 1352.07 (1233.62-1481.89) 1754.86 (1600.15-1924.53) Any Grade 3 or Higher Adverse Event, n (%) 379 (81.9) 324 (71.1) Exposure-Adjusted Incidence of Grade 3 or Higher Adverse Events, Rate Per 100 162.31 (146.77-179.50) 175.90 (157.75-196.13) Patient-Years (95% CI) Any Serious Adverse Event, n (%) 279 (60.3) 183 (40.1) Any Adverse Event Leading to Carfilzomib or Bortezomib Dose Reduction, n (%) 138 (29.8) 226 (49.6) Any Adverse Event Leading to Treatment Discontinuation, n (%) 137 (29.6) 121 (26.5) Any Adverse Event Leading to Death, n (%) 32 (6.9) 22 (4.8) Abbreviations: Kd56 ¼ carfilzomib (56 mg/m2) and dexamethasone; Vd ¼ bortezomib and dexamethasone. Median OS was longer in the Kd56 group versus the Vd previous lenalidomide treatment for MM. Median OS for the group, regardless of previous exposure to bortezomib. Median OS Kd56 group versus the Vd group was 29.2 versus 21.4 months for the Kd56 group versus the Vd group was 41.8 versus 32.7 (HR, 0.86; 95% CI, 0.62-1.18) for lenalidomide-refractory months (HR, 0.85; 95% CI, 0.67-1.08) for patients with pre- patients (Kd56, n ¼ 113; Vd, n ¼ 123) and NE versus 42.8 vious bortezomib exposure (Kd56, n ¼ 250; Vd, n ¼ 252) and months (HR, 0.74; 95% CI, 0.59-0.93) for those who were not NE versus 42.2 months (HR, 0.66; 95% CI, 0.50-0.88) for refractory to a previous lenalidomide treatment (Kd56, n ¼ 351; patients without previous bortezomib exposure (Kd56, n ¼ 214; Vd, n ¼ 342). Vd, n ¼ 213). Median OS was longer in the Kd56 group compared with the Vd Safety group regardless of previous exposure to lenalidomide. Median OS Overall, 457 (98.7%) patients in the Kd56 group and 451 for the Kd56 group versus the Vd group was 35.4 versus 29.4 (98.9%) patients in the Vd group experienced an AE (Table 1). months (HR, 0.88; 95% CI, 0.67-1.16) for patients with previous Exposure-adjusted patient incidence per 100 patient-years of AEs lenalidomide exposure (Kd56, n ¼ 177; Vd, n ¼ 178) and 51.3 overall were 1352.07 (95% CI, 1233.62-1481.89) for the Kd56 versus 42.2 months (HR, 0.68; 95% CI, 0.54-0.88) for those group and 1754.86 (95% CI, 1600.15-1924.53) for the Vd without previous lenalidomide exposure (Kd56, n ¼ 287; Vd, n ¼ group. Grade 3 AEs occurred in 379 (81.9%) patients in the 287). Median OS was also longer in the Kd56 group versus the Vd Kd56 group and 324 (71.1%) in the Vd group (exposure-adjusted group regardless of whether or not patients were refractory to any patient incidence per 100 patient-years were 162.31 [95% CI, Table 2 Treatment-Emergent Adverse Events in the Safety Population Kd56 (n [ 463) Vd (n [ 456) All Grades Grade 3 or Higher All Grades Grade 3 or Higher a Most Common Events, n (%) Anemia 202 (43.6) 80 (17.3) 130 (28.5) 46 (10.1) Diarrhea 170 (36.7) 19 (4.1) 185 (40.6) 40 (8.8) Pyrexia 151 (32.6) 14 (3.0) 70 (15.4) 3 (0.7) Hypertension 150 (32.4) 69 (14.9) 46 (10.1) 15 (3.3) Fatigue 149 (32.2) 32 (6.9) 140 (30.7) 35 (7.7) Dyspnea 149 (32.2) 29 (6.3) 62 (13.6) 10 (2.2) Events of Interest, n (%) Cardiac failure (SMQN) 51 (11.0) 28 (6.0) 16 (3.5) 9 (2.0) Ischemic heart disease (SMQN) 18 (3.9) 12 (2.6) 9 (2.0) 7 (1.5) Peripheral neuropathy (SMQN) 97 (21.0) 11 (2.4) 249 (54.6) 44 (9.6) Acute renal failure (SMQN) 50 (10.8) 27 (5.8) 29 (6.4) 16 (3.5) Hematopoietic thrombocytopenia (SMQN) 148 (32.0) 58 (12.5) 123 (27.0) 67 (14.7) Neutropenia (PT) 29 (6.3) 12 (2.6) 26 (5.7) 10 (2.2) Abbreviations: Kd56 ¼ carfilzomib (56 mg/m2) and dexamethasone; PT ¼ preferred term; SMQN ¼ standardized Medical Dictionary for Regulatory Activities query, narrow scope; Vd ¼ bortezomib and dexamethasone. a Adverse events (preferred terms) are included if reported in 30% of patients in either treatment group. 528 - Clinical Lymphoma, Myeloma & Leukemia August 2019 Robert Z. Orlowski et al Table 3 Treatment-Emergent Adverse Events Over Time in the Safety Population Month 0 to £12 Month 12 to £24 Month 24 to £36 Month >36 Kd56 Vd Kd56 Vd Kd56 Vd Kd56 Vd (n [ 463) (n [ 456) (n [ 212) (n [ 104) (n [ 89) (n [ 41) (n [ 50) (n [ 25) Any Grade 3 or 343 (74.1) 306 (67.1) 112 (52.8) 44 (42.3) 42 (47.2) 15 (36.6) 27 (54.0) 7 (28.0) Higher Adverse Event, n (%) Abbreviations: Kd56 ¼ carfilzomib (56 mg/m2) and dexamethasone; Vd ¼ bortezomib and dexamethasone. 146.77-179.50] and 175.90 [95% CI, 157.75-196.13], receiving treatment with Kd56 versus Vd.3,4 In the present analysis respectively). with longer follow-up (median follow-up of approximately 44 The most commonly occurring AEs and AEs of interest are months), OS results continued to show improvement with Kd56 shown in Table 2. Grade 3 cardiac failure (SMQN) occurred in versus Vd (median OS, 47.8 vs. 38.8 months; HR, 0.76; 95% CI, 28 (6.0%) patients in the Kd56 group and 9 (2.0%) in the Vd 0.63-0.92; 1-sided P ¼ .0017). group. Cardiac failure (SMQN) led to carfilzomib or bortezomib In the interim OS analysis, treatment with Kd56 also improved dose reduction in 8 (1.7%) patients in the Kd56 group and no OS across several patient subgroups,3 and this continues with an patient in the Vd group. Cardiac failure (SMQN) led to discon- additional 6 months of follow-up. For patient subgroups according tinuation of any treatment in 18 (3.9%) patients in the Kd56 group to previous lines of therapy, previous bortezomib exposure, previous and 4 (0.9%) patients in the Vd group. lenalidomide exposure, and lenalidomide-refractory status, there was Grade 3 hypertension was reported in 69 (14.9%) patients in the a 12% to 34% reduction in the risk of death for the Kd56 group Kd56 group and 15 (3.3%) patients in the Vd group. Hypertension versus the Vd group. Across all age and cytogenetic subgroups, an led to reduction of carfilzomib dose in 11 (2.4%) patients in the OS benefit was shown for Kd56 versus Vd, including in elderly Kd56 group and no patient in the Vd group. No patient in either patients (75 years) and patients with high-risk cytogenetics. treatment group discontinued carfilzomib or bortezomib treatment Late phase clinical trials include measures of clinical benefit that because of hypertension. support regulatory approval. Although measurement of OS requires Grade 2 peripheral neuropathy (SMQN) was reported in 32 prolonged trials, and OS outcomes might be confounded by the (6.9%) patients in the Kd56 group and 159 (34.9%) in the Vd group. availability of active therapeutics used in later lines, OS remains the Peripheral neuropathy (SMQN) led to carfilzomib dose reduction in oncology end point preferred by the European Medicines Agency, 9 (1.9%) patients in the Kd56 group and bortezomib dose reduction the Food and Drug Administration, and other health authorities for in 142 (31.1%) patients in the Vd group. Peripheral neuropathy regulatory approval. (SMQN) led to discontinuation of any treatment in 2 (0.4%) patients The safety data obtained during this analysis were consistent with in the Kd56 group and 40 (8.8%) patients in the Vd group. those previously reported from the first and final analyses of Infections and infestations (system organ class) were reported in ENDEAVOR.3,4 The exposure-adjusted incidence per 100 patient- 368 (79.5%) patients in the Kd56 group and 319 (70.0%) patients years of AEs was lower in patients treated with Kd56 than Vd in the Vd group. Grade 3 AEs in this class were reported in 148 (1352.07 vs. 1754.86). Although the number of patients with re- (32.0%) patients and 94 (20.6%) patients, respectively. ported Grade 3 AEs was higher with carfilzomib than with bor- The incidence of Grade 3 AEs decreased over time in both tezomib (379 [81.9%] and 324 [71.1%] for Kd56 vs. Vd, treatment groups (Table 3). In total, 343 (74.1%) patients of 463 respectively), the exposure-adjusted incidence of Grade 3 AEs was receiving Kd56 experienced a Grade 3 AE between 0 to 12 also lower in the Kd56 group than in the Vd group (162.31 vs. months, 112 (n ¼ 212; 52.8%) within 12 to 24 months, 42 (n ¼ 175.90). Overall, the long-term safety profile for Kd56 was 89; 47.2%) between 24 to 36 months, and 27 (n ¼ 50; 54.0%) acceptable, and there has been no evidence of cumulative toxicity beyond 36 months of treatment. A total of 306 (67.1%) of 456 with extended treatment. patients receiving Vd reported grade 3 AEs from 0 to 12 Because of the relapsing nature of MM, and the availability of months, 44 (n ¼ 104; 42.3%) between 12 to 24 months, 15 (n ¼ well tolerated agents, prolonged courses of maintenance therapy are 41; 36.6%) between 24 to 36 months, and 7 (n ¼ 25; 28.0%) now a standard of care in myeloma. Because maintenance lenali- beyond 36 months from Vd start date. domide confers a PFS and possibly an OS advantage compared with At 36 months, 35 patients (7.5%) in the Kd56 group and 21 placebo in treatment of first-line MM8-10 and when lenalidomide- patients (4.5%) in the Vd group were still receiving treatment. dexamethasone is compared with dexamethasone,11,12 lenalidomide-dexamethasone is becoming a standard first-line Discussion treatment, and therefore might not be an option for patients at The phase III ENDEAVOR trial evaluated the efficacy and safety relapse.10 The ENDEAVOR trial was designed to offer RRMM of Kd56 versus Vd in patients with RRMM. The initial publications patients an alternate doublet therapy until progression or intoler- from ENDEAVOR reported significant improvements in OS (21% ance. These results, showing tolerability and a survival benefit, reduction in the risk of death), PFS, and responses for patients support the emerging role of prolonged PI therapy. Clinical Lymphoma, Myeloma & Leukemia August 2019 - 529 Updated Survival and Subgroup Analysis of ENDEAVOR Conclusion consulting or advisory role for Amgen, Celgene, Janssen, and With additional follow-up, clinically meaningful OS improve- Takeda, and serving on the speakers’ bureaus for Amgen and ments continue to be observed with Kd56 versus Vd, including in Janssen. W.J.C. reports honoraria from Amgen, Celgene, Janssen, key patient subgroups. The benefit-risk profile of Kd56 with longer- Novartis, Roche, and Takeda, and research funding from Celgene, term data continued to provide support for its use as a standard of Janssen, Merck, Novartis, and Roche. H.G. reports research support care in patients with RRMM. from Amgen, Bristol-Myers Squibb, Celgene, Chugai, Janssen, Mundipharma, Novartis, Sanofi, and Takeda, advisory board Clinical Practice Points participant for Adaptive Biotechnology, Amgen, Bristol-Myers  Extending OS is a major therapeutic goal for MM agents. PIs, Squibb, Celgene, Janssen, Sanofi-Aventis, and Takeda, and hono- such as bortezomib or carfilzomib, are a backbone therapy in raria for participant in speakers’ bureaus for ArtTempi, Bristol- MM treatment. Myers Squibb, Celgene, Chugai, Janssen, and Novartis. Z.Y. re-  In a prespecified interim analysis (data cutoff January 3, 2017) of ports employment with and stock holdings from Amgen. A.S.K. the randomized phase III ENDEAVOR study, Kd56 signifi- reports employment with Amgen and stock holdings from Amgen cantly improved OS over Vd in patients with RRMM. and WindMIL. M.D. reports consulting or advisory role fees from  Herein we report updated ENDEAVOR OS and safety data Amgen, Celgene, Janssen, Novartis, and Takeda, research funding using a more recent data cutoff of July 19, 2017, when all pa- from Genesis Pharma, and honoraria from Amgen, Celgene, Jans- tients had received at least 3 years of follow-up. sen, Novartis, and Takeda.  Median OS was extended by 9 months with Kd56 versus Vd. These OS benefits were consistent across all subgroups exam- Supplemental Data ined, including patient age, cytogenetic risk, previous lenalido- The supplemental table accompanying this article can be found in mide exposure, and lenalidomide-refractory status. the online version at https://doi.org/10.1016/j.clml.2019.04.018.  Safety data were consistent with previous analyses.  These longer-term data continue to provide support for the use of Kd56 as a standard of care in patients with RRMM. References 1. Laubach J, Garderet L, Mahindra A, et al. Management of relapsed multiple myeloma: recommendations of the International Myeloma Working Group. Acknowledgments Leukemia 2016; 30:1005-17. 2. Pazdur R. Endpoints for assessing drug activity in clinical trials. Oncologist 2008; Onyx Pharmaceuticals, Inc, an Amgen subsidiary, funded the 13(suppl 2):19-21. ENDEAVOR study and was involved in study design; collection, 3. Dimopoulos MA, Goldschmidt H, Niesvizky R, et al. Carfilzomib or bortezomib in relapsed or refractory multiple myeloma (ENDEAVOR): an interim overall analysis, and interpretation of data; in the writing of this report; and survival analysis of an open-label, randomised, phase 3 trial. Lancet Oncol 2017; 18: in the decision to submit this article for publication. Medical 1327-37. 4. Dimopoulos MA, Moreau P, Palumbo A, et al. Carfilzomib and dexamethasone vs. writing assistance was provided by BlueMomentum, an Ashfield bortezomib and dexamethasone for patients with relapsed or refractory multiple company, part of UDG Healthcare PLC, and Guerry Cook and Yin myeloma (ENDEAVOR): a randomised, phase 3, open-label, multicentre study. Lancet Oncol 2016; 17:27-38. C. Lin of Amgen, Inc, and funded by Amgen. 5. Siegel DS, Dimopoulos MA, Ludwig H, et al. Improvement in overall survival with carfilzomib, lenalidomide, and dexamethasone in patients with relapsed or refractory multiple myeloma. J Clin Oncol 2018; 36:728-34. Disclosure 6. Chng WJ, Goldschmidt H, Dimopoulos MA, et al. Carfilzomib-dexamethasone R.Z.O. reports consultancy or advisory board participation for vs. bortezomib-dexamethasone in relapsed or refractory multiple myeloma by cytogenetic risk in the phase 3 study ENDEAVOR. Leukemia 2017; 31:1368- Amgen, Bristol-Myers Squibb, Celgene, Janssen, Kite Pharma, 74. Sanofi-Aventis, and Takeda, and grant or research support from 7. Klein JP, Moeschberger ML. Survival Analysis: Techniques for Censored and Truncated Data. New York: Springer; 1997. Amgen, BioTheryX, and Spectrum Pharma. P.M. reports honoraria 8. Benboubker L, Dimopoulos MA, Dispenzieri A, et al. Lenalidomide and dexa- from Amgen, Bristol-Myers Squibb, Celgene, Janssen, Millennium, methasone in transplant-ineligible patients with myeloma. N Engl J Med 2014; 371:906-17. Novartis, Onyx, and Takeda, and consulting or advisory role fees 9. McCarthy PL, Holstein SA, Petrucci MT, et al. Lenalidomide maintenance after from Amgen, Bristol-Myers Squibb, Celgene, Janssen, Millennium, autologous stem-cell transplantation in newly diagnosed multiple myeloma: a meta-analysis. J Clin Oncol 2017; 35:3279-89. Novartis, Onyx, and Takeda. R.N. reports consultancy for Amgen, 10. Pulte ED, Dmytrijuk A, Nie L, et al. FDA approval summary: lenalidomide as Bristol-Myers Squibb, Celgene, Janssen, and Takeda. H.L. reports maintenance therapy after autologous stem cell transplant in newly diagnosed multiple myeloma. Oncologist 2018; 23:734-9. serving in a consulting or advisory role for Amgen, Cilag-Janssen, 11. Dimopoulos M, Dmytrijuk A, Nie L, et al. Lenalidomide plus dexametha- and Takeda, serving on the speakers’ bureau for Amgen, Bristol- sone for relapsed or refractory multiple myeloma. N Engl J Med 2007; 357: 2123-32. Myers Squibb, Celgene, Cilag-Janssen, and Takeda, and receiving 12. Weber DM, Chen C, Niesvizky R, et al. Lenalidomide plus dexamethasone for research funding from Amgen and Takeda. A.O. reports serving in a relapsed multiple myeloma in North America. N Engl J Med 2007; 357:2133-42. 530 - Clinical Lymphoma, Myeloma & Leukemia August 2019 Robert Z. Orlowski et al Supplemental Data Supplemental Table 1 Subsequent Antimyeloma Therapies According to Age Group in ‡10% of Patients (ITT Population) <65 Years 65-74 Years ‡75 Years Kd56 (n [ 223) Vd (n [ 210) Kd56 (n [ 164) Vd (n [ 189) Kd56 (n [ 77) Vd (n [ 66) Patients Treated 150 (67.3) 143 (68.1) 101 (61.6) 125 (66.1) 40 (51.9) 33 (50.0) With at Least 1 Antimyeloma Therapy, n (%) Dexamethasone 110 (49.3) 115 (54.8) 71 (43.3) 92 (48.7) 28 (36.4) 23 (34.8) Lenalidomide 77 (34.5) 87 (41.4) 44 (26.8) 62 (32.8) 16 (20.8) 16 (24.2) Cyclophosphamide 53 (23.8) 65 (31.0) 34 (20.7) 35 (18.5) 12 (15.6) 14 (21.2) Pomalidomide 43 (19.3) 50 (23.8) 29 (17.7) 47 (24.9) 10 (13.0) 10 (15.2) Bortezomib 59 (26.5) 32 (15.2) 44 (26.8) 20 (10.6) 15 (19.5) 5 (7.6) Melphalan 29 (13.0) 32 (15.2) 19 (11.6) 20 (10.6) 5 (6.5) 4 (6.1) Thalidomide 27 (12.1) 34 (16.2) 9 (5.5) 23 (12.2) 3 (3.9) 3 (4.5) Prednisone 8 (3.6) 11 (5.2) 10 (6.1) 20 (10.6) 4 (5.2) 5 (7.6) Abbreviations: ITT ¼ intent-to-treat; Kd56 ¼ carfilzomib (56 mg/m2) and dexamethasone; Vd ¼ bortezomib and dexamethasone. Clinical Lymphoma, Myeloma & Leukemia August 2019 - 530.e1