Title: Phase 1b trial of pembrolizumab monotherapy for relapsed/refractory multiple myeloma: <scp>KEYNOTE</scp>‐013
Abstract: Survival outcomes for patients with relapsed/refractory multiple myeloma (RRMM) have improved recently as a result of the development of new therapeutic options (Dimopoulos et al, 2016; Palumbo et al, 2016). Nevertheless, RRMM remains an incurable disease; long-term survival is rare, and most patients die after relapse (Neri et al, 2016). Pre-clinical evidence suggests anti-tumour potential of immune checkpoint inhibition in haematological malignancies, including multiple myeloma (MM) (Jelinek et al, 2017). MM is characterised in part by generalised immune suppression, including increased involvement of the programmed death 1 (PD-1, PDCD1)/programmed death ligand 1 (PD-L1, CD274) axis (Neri et al, 2016), rendering it a potential target for immune checkpoint inhibition. Pembrolizumab is a highly selective humanised antibody that blocks the interaction between PD-1 and its ligands and demonstrates robust anti-tumour activity and a favourable safety profile as a single agent in several malignancies (https://www.merck.com/product/usa/pi_circulars/k/keytruda/keytruda_pi.pdf). KEYNOTE-013 (ClinicalTrials.gov identifier, NCT01953692) is a multi-cohort multicentre, phase 1b study of single-agent pembrolizumab in patients with haematological malignancies, including RRMM. Results from the RRMM cohort of that study are presented herein. Patients received pembrolizumab 10 mg/kg every 2 weeks (Q2W) or 200 mg Q3W based on timing of enrolment in the study (see Data S1 for details). Response was assessed using the International Myeloma Working Group 2006 uniform response criteria (Durie et al, 2006). Efficacy and safety were analysed in all patients who received at least one dose of pembrolizumab. Additional details of the methodology are available in the supplement. At data cut-off (11 January 2018), all 30 patients (pembrolizumab 10 mg/kg Q2W, n = 20; pembrolizumab 200 mg Q3W, n = 10) had discontinued treatment because of progressive disease/clinical progression (n = 24; 80·0%), physician or patient decision to withdraw (n = 4; 13·3%) or adverse events (AEs; n = 2; 6·7%) (Figure S1). Data for the pembrolizumab doses were pooled for analyses. Baseline characteristics were representative of patients with RRMM (Table SI). Median age was 70·0 years. Patients had previously received a median of four lines of therapy. Sixteen (53·3%) patients previously underwent stem cell transplantation (SCT) and 20 (66·7%) had disease that was refractory to lenalidomide. After a median follow-up of 19·9 months (range, 2·0–44·0 months), the best overall response achieved with pembrolizumab monotherapy—the co-primary endpoint—was stable disease (n = 17; 56·7%) (Table 1). Median duration of stable disease was 3·7 months [95% confidence interval (CI), 2·7–18·85 months] (Figure S2). Of the 17 patients with stable disease, 10 (58·8%) had disease that was refractory to lenalidomide, five (29·4%) had double-refractory, 4 (23·3%) had triple-refractory and 1 (5·9%) had quadruple-refractory disease (Table SII). Baseline characteristics of these 17 patients were similar to those of the overall population (Tables SI and SII). Of the 27 patients with available data, 11 (40·7%) experienced reduction from baseline in maximum monoclonal protein or free light chains (Figure S3). Two patients had unconfirmed responses per laboratory assessment; however, both switched to a different therapy before confirmation of response. Median progression-free survival (PFS) was 2·7 months (95% CI, 1·4–3·7 months) (Figure S4A). Median OS was 20·2 months (95% CI, 14·1 to not reached) (Figure S4B). Median pembrolizumab exposure consisted of five administrations (range, 2–15) for 10 mg/kg Q2W and three administrations (range, 2–5) for 200 mg Q3W. In the safety analysis, the co-primary endpoint showed that 28 (93·3%) patients experienced at least one AE (any cause), most commonly anaemia (n = 13; 43·3%), asthenia (n = 11; 36·7%) and back pain (n = 8; 26·7%) (Table SIII). More than half the AEs were grade 1 or 2 (n = 15; 50·0%). The most common grade 3 AEs were anaemia (n = 4; 13·3%) and hypercalcaemia (n = 2; 6·7%). Grade 4 anaemia (any-cause AE) occurred in one patient. Twelve (40·0%) patients experienced at least one treatment-related AE (TRAE; Table 2), most commonly asthenia (n = 5; 16·7%) and arthralgia (n = 2; 6·7%). One (3·3%) patient experienced grade 3 treatment-related myalgia; none had grade 4/5 TRAEs. TRAEs led to no treatment interruptions and one (3·3%) discontinuation (myalgia). Immune-mediated AEs occurred in two (6·7%) patients (n = 1 each of pruritus, upper respiratory tract infection) (Table 2). Three patients died (two, myeloma; one, hypoxia) during the safety follow-up. Most TRAEs after pembrolizumab monotherapy were mild to moderate. Pembrolizumab was generally well tolerated in the RRMM patient cohort and showed a safety profile consistent with its safety profile in other cancers (https://www.merck.com/product/usa/pi_circulars/k/keytruda/keytruda_pi.pdf). The lack of objective response could potentially be overcome by combination modalities. Clinical benefit has been reported with pembrolizumab in combination with pomalidomide and dexamethasone in patients with RRMM, suggesting that pembrolizumab may act in synergy with other antimyeloma therapies (Badros et al, 2017). However, on 3 July 2017, based on interim data presented to the data monitoring committee, the US Food and Drug Administration halted two phase 3 pembrolizumab-based combination studies in MM (KEYNOTE-183 and KEYNOTE-185; ClinicalTrials.gov identifiers, NCT02576977 and NCT02579863) (The ASCO Post, 2017). These trials compared pomalidomide or lenalidomide and low-dose dexamethasone, with and without pembrolizumab, respectively, in patients with RRMM who had previously received at least two lines of therapy and in treatment-naive patients with newly-diagnosed MM who were ineligible for autologous SCT. Based on these interim data, the benefit-risk profile for combination regimens with pembrolizumab, pomalidomide/lenalidomide and dexamethasone was unfavourable for RRMM. Long-term safety and survival analyses are ongoing. In conclusion, the best response achieved by patients with RRMM receiving pembrolizumab monotherapy is stable disease. The safety profile of pembrolizumab in RRMM is consistent with its safety profile in other cancers. However, results from studies involving a small number of patients and without a comparator group should be interpreted with caution. Nevertheless, the current findings, along with results of other studies, in this patient population suggest that the future of PD-1 inhibition, in combination with other agents as treatment for RRMM, should be carefully and comprehensively explored. The authors thank the patients and their families and all investigators and site personnel. The authors also thank Jennifer Park, Karen Giallella, Brook Burton, Seth Thompson, Kathleen Fox (employees of Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Kenilworth, NJ, USA) and Roger Dansey (former employee of Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Kenilworth, NJ, USA) for their contributions to the development of the manuscript. Medical writing and/or editorial assistance was provided by Jacqueline Kolston, PhD, Doyel Mitra, PhD and Matthew Grzywacz, PhD, of the ApotheCom pembrolizumab team (Yardley, PA, USA). This assistance was funded by Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Kenilworth, NJ, USA. Funding for this research was provided by Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Kenilworth, NJ, USA. Vincent Ribrag: Consulting/Advisory Role: Gilead, Infinity, Bristol-Myers Squibb, NanoString, Incyte; Research Funding: ArgenX; Expert Testimony: Servier; Travel, Accommodations, Expenses: Roche, Bristol-Myers Squibb. David E. Avigan: Consultant/Advisory Role: Seattle Genetics, Celgene, Bristol-Myers Squibb, Juno, Partners Therapeutics, Parexel, Kymriah; Research Funding: Pharmacyclics, Celgene. Damian J. Green: Consultant/Advisory Role: June Therapeutics. Trisha Wise-Draper: Research Funding: Merck & Co., Inc. Juan G. Posada: Research Funding: Merck & Co., Inc. Ravi Vij: Honoraria: Amgen, Takeda, Celgene, Bristol-Myers Squibb, Janssen, Karyopharma, AbbVie; Consultant/Advisory Role: Amgen, Takeda, Celgene, Bristol-Myers Squibb, Janssen, Karyopharma, AbbVie; Research Funding: Amgen, Takeda. Ying Zhu: Employment: Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Kenilworth, NJ, USA. Mohammed Z. H. Farooqui: Employment, Stock: Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Kenilworth, NJ, USA. Patricia Marinello: Employment, Stock and Travel Expenses: Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Kenilworth, NJ, USA. David S. Siegel: Honoraria: Celgene, Takeda, Amgen, Bristol-Myers Squibb; Speaker's Bureau: Celgene, Takeda, Amgen, Bristol-Myers Squibb; Travel Expenses: Celgene, Takeda, Amgen, Bristol-Myers Squibb. Conception and design: David E. Avigan, Patricia Marinello, David S. Siegel. Collection and assembly of data: David E. Avigan, Damian J. Green, Juan G. Posada, Mohammed Z. H. Farooqui. Data analysis and interpretation: Vincent Ribrag, David E. Avigan, Damian J. Green, Trisha Wise-Draper, Ravi Vij, Ying Zhu, Mohammed Z. H. Farooqui, Patricia Marinello, David S. Siegel. Provision of study materials/patients: Juan G. Posada. Manuscript writing: All authors. Final approval of manuscript: All authors. Accountable for all aspects of the work: All authors. Data S1. Supplementary material. Table SI. Patient baseline characteristics. Table SII. Baseline characteristics of patients with stable disease as best response to pembrolizumab (10 mg/kg Q2W and 200 mg Q3W groups pooled). Table SIII. Adverse events occurring in at least two patients. Figure S1. CONSORT (study flow) diagram. Q2W, every 2 weeks; Q3W, every 3 weeks. Figure S2. Kaplan-Meier estimates of duration of stable disease based on International Myeloma Working Group uniform response criteria per site review. Figure S3. Change from baseline in maximum M-protein or free light chain in the 27 of 30 patients for whom data were evaluable. Figure S4. Kaplan-Meier estimates of (A) progression-free survival and (B) overall survival based on International Myeloma Working Group uniform response criteria per site review. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.