Promising efficacy and acceptable safety of venetoclax plus bortezomib and dexamethasone in relapsed/refractory MM

Investigation of novel agents with a unique mechanism of action may offer additional treatment options for patients with multiple myeloma (MM) who, despite recent advances, inevitably relapse or become refractory to therapy.^1,2 

The BCL-2 family of antiapoptotic (eg, BCL-2, BCL-X_L, and myeloid cell leukemia sequence 1 [MCL-1]) and proapoptotic (eg, BAX, BAK) proteins plays a critical role in regulating the intrinsic apoptosis pathway and cell survival.^3-5  Overexpression of antiapoptotic proteins is heterogeneous in myeloma cells. Both MCL-1 and BCL-2 have been shown to be overexpressed in a subset of myeloma cells and have been implicated in mediating their survival.⁶  The role of BCL-2 and MCL-1 in maintaining MM cell survival was recently confirmed by BCL-2 homology domain 3 profiling analysis.⁷ 

Venetoclax is an orally bioavailable, potent inhibitor of BCL-2.^8,9  Selective inhibition of BCL-2 restores the apoptotic pathway in malignant cells, and promising antitumor activity has been observed with venetoclax in numerous hematologic malignancies.^10-13  Venetoclax as a single agent has been shown to induce apoptosis in human MM cell lines and primary samples from patients with MM, particularly those carrying the t(11;14) translocation.¹⁴  Furthermore, single-agent venetoclax has shown activity in relapsed and refractory patients with t(11;14).^15,16  Similarly, targeting of MCL-1 genetically or pharmacologically in vitro and in vivo induced apoptosis of human MM cell lines.^17,18  Xenograft models that coexpressed BCL-X_L or MCL-1 with BCL-2 were resistant to venetoclax. This resistance was mitigated by combining venetoclax with bortezomib,¹⁹  a proteasome inhibitor that can inhibit MCL-1 indirectly via stabilizing the MCL-1–neutralizing protein NOXA.^20,21  Dexamethasone, a steroid commonly used in MM treatment, can upregulate the expression of the proapoptotic molecule BIM and increase its binding to BCL-2, which also results in increased sensitivity to venetoclax.²²  Thus, targeting BCL-2 and MCL-1 function to induce apoptosis through the combination of venetoclax, bortezomib, and dexamethasone is a compelling approach to treat MM.

Here, we report the safety, efficacy, pharmacokinetics, and exploratory biomarker results from an ongoing open-label phase 1b study of venetoclax in combination with a standard-of-care regimen of bortezomib and dexamethasone in patients with relapsed/refractory MM.

The open-label, multicenter, phase 1b M12-901 study (registered at www.clinicaltrials.gov as #NCT01794507) enrolled patients with relapsed/refractory MM from 8 study sites between November 2012 and May 2016. The data cutoff for this publication is 19 August 2016. Primary objectives of the study were to evaluate the safety and pharmacokinetics and determine maximum tolerated dose and recommended phase 2 dose (RP2D) of venetoclax when given with bortezomib and dexamethasone. Secondary objectives included overall response rate (ORR), time to disease progression (TTP), duration of overall response (DOR), and exploratory biomarkers.

Patients with relapsed/refractory MM meeting criteria for treatment (per physician assessment) were enrolled (detailed eligibility criteria are listed in the supplemental Methods, available on the Blood Web site). Patients had measurable disease at baseline, including monoclonal protein ≥1 g/dL in serum or ≥200 mg/24 hours in urine, or serum immunoglobulin free light chain ≥10 mg/dL. Patients had to have adequate bone marrow, renal, and hepatic function, as well as an Eastern Cooperative Oncology Group performance status of 0 or 1. Patients were excluded if they had prior treatment with bortezomib within 30 days of the first dose of venetoclax, grade 3/4 peripheral neuropathy, or systemic infection, fever, or neutropenia within 1 week of the first venetoclax dose. Patients were considered refractory to a prior line of therapy if at least a minimum response was not achieved while on that therapy or the patient had progressive disease while on therapy or within 60 days of last dose of that therapy.

Venetoclax was administered orally once daily. To evaluate the risk of tumor lysis syndrome, a 1-week lead-in period with 50 mg venetoclax alone followed by venetoclax combined with bortezomib and dexamethasone was implemented in the first cohort. Patients in subsequent cohorts received the designated cohort dose of venetoclax once daily (100, 200, 300, 400, 500, 600, 800, 1000, and 1200 mg), as determined by continual reassessment method,^23,24  in combination with bortezomib and dexamethasone, starting with cycle 1 day 1 (no lead-in period). Bortezomib (1.3 mg/m² subcutaneous injection) was given on days 1, 4, 8, and 11 during cycles 1 to 8 and days 1, 8, 15, and 22 during cycles 9 to 11.²⁵  Dexamethasone (20 mg oral) was given on days 1, 2, 4, 5, 8, 9, 11, 12 during cycles 1 to 8 and on days 1, 8, 15, and 22 during cycles 9 to 11. Patients who remained on the study beyond cycle 11 received venetoclax monotherapy. The supplemental Methods provide further details on treatment.

Safety assessments included adverse event (AE) monitoring, vital signs, physical examination, 12-lead electrocardiography, multiple-gated acquisition/2-dimensional echocardiogram, and clinical laboratory tests. AEs were graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events v4.03.²⁶ 

Clinical responses as well as disease progression were assessed by the investigator, using the International Myeloma Working Group criteria, based on central or local laboratories^27,28  (supplemental Table 2).

Pharmacokinetic assessments and results, including dose–response analyses, are described in the supplemental Methods.

Interphase fluorescence in situ hybridization analysis was performed on CD138-selected bone marrow mononuclear cells (BMMCs) using probes for t(11;14), t(4;14), del(17p), del(13q), and chromosomes 5, 9, or 15. Expression of BCL2 (BCL-2), BCL2L1 (BCL-X_L), and MCL1 (MCL-1) messenger RNA in purified CD138-positive BMMCs was determined by droplet digital polymerase chain reaction. Additional details are provided in the supplemental Methods.

The study was designed jointly by sponsors and investigators according to Good Clinical Practice guidelines, the Declaration of Helsinki, and applicable regulations, with institutional review board approval for all study sites. Data analyses were conducted by personnel employed by the sponsor, and investigators had full access to data for review. The first draft of the manuscript was written by a medical writer employed by AbbVie, with author input. Subsequent drafts were prepared by all authors and the medical writer. Authors made the decision to submit the manuscript for publication and vouch for adherence to the study protocol and accuracy of data reported. All patients provided written informed consent.

Patients who received at least 1 dose of venetoclax were included in the analyses. Efficacy was also evaluated by subgroups, including nonrefractory (ie, naive or sensitive) or refractory to prior therapies, number of prior lines of therapy (1-3, 4-6, and >6), and cytogenetics. SAS software (SAS Institute, Inc., Cary, NC) was used for statistical summaries and analyses; additional details are shown in the supplemental Methods.

Sixty-six patients were enrolled. The median age was 64 years (range, 38-79), and 39 (65%) patients had an International Staging System stage of II/III (Table 1). Patients had cytogenetic abnormalities as follows: 9 (14%) had t(11;14) translocation, 5 (8%) had t(4;14) translocation, 15 (23%) had del(17p) deletion, 30 (45%) had del(13q) deletion, and 30 (45%) had hyperdiploidy. Patients had received a median of 3 prior therapies (range, 1-13), with 80% of patients previously treated with bortezomib (39% refractory), 73% previously treated with lenalidomide (53% refractory), and 59% having had a prior stem cell transplant.

The median time on study was 5.9 months (range, 0.3-29) for all patients (N = 66), 2.8 months (range, 0.3-9.5) for those who were considered refractory to prior bortezomib (n = 26), and 9.1 months (range, 0.5-29) for those nonrefractory to bortezomib (n = 40). Forty-six (70%) patients discontinued the study. The primary reason for discontinuation was disease progression in 36 patients, AEs for 5, consent withdrawal for 2, and 3 for reasons not specified. AEs leading to discontinuation were respiratory and cardiac failure (n = 2), lung adenocarcinoma, Guillain-Barré syndrome, and sepsis; none were considered by the investigator as related to venetoclax, though 1 case of dose-limiting cardiac failure in the 300-mg cohort was deemed possibly related to dexamethasone by the investigator, and the Guillain-Barré syndrome in the 1200-mg cohort was considered possibly related to bortezomib. Six (9%) patients required venetoclax dose reductions due to AEs, and 27 (41%) temporarily interrupted treatment due to AEs. Five deaths were reported during the study, with 4 due to disease progression and 1 due to respiratory syncytial virus infection (supplemental Table 4; Figure 1 [time on study]).

The most common AEs were gastrointestinal toxicities (diarrhea [46%], constipation [41%], and nausea [38%]) (Table 2). The most common grade 3/4 AEs were cytopenias: thrombocytopenia (29%), anemia (15%), and neutropenia (14%) (Table 2). Mean platelet counts decreased over the course of each cycle with bortezomib, recovering prior to initiation of the next cycle, and then remained stable once monotherapy with venetoclax was started after cycle 11 (supplemental Figure 2). The median time to first event of grade 3/4 thrombocytopenia was 15 days (range, 3-267), and 13 out of 19 patients received platelet transfusions for grade 3/4 thrombocytopenia. The thrombocytopenia reported in this study was similar to most veliparib-containing regimens (ie, nadir at the end of cycle 2 and a trend to less severe nadirs afterward). Time to first event of grade 3/4 neutropenia was 29 days (range, 21-171), and 7 out of 9 patients with grade 3/4 events received growth factor support. Few patients temporarily interrupted dosing or reduced their dose, and no patients discontinued the study due to these grade 3/4 cytopenias (supplemental Table 4). Serious AEs occurring in ≥2 patients were pneumonia (n = 5), sepsis (3), pyrexia (3), influenza (3), febrile neutropenia (3), thrombocytopenia (2), cardiac failure (2), lower respiratory tract infection (2), acute kidney injury (2), respiratory failure (2), embolism (2), and hypotension (2) (Table 3). The maximum tolerated dose of venetoclax in combination with bortezomib and dexamethasone was not reached. Dose-limiting toxicities were a grade 3 cardiac failure in the 300-mg cohort deemed possibly related to dexamethasone, and a grade 3 thrombocytopenia event during the first cycle in the safety expansion. No events of laboratory or clinical tumor lysis syndrome were reported in the study.

Forty-four (67%) of 66 patients achieved an overall response (PR or better), with 28 (42%) patients achieving a very good partial response or better (≥VGPR; 3 stringent complete response, 10 complete response, and 15 VGPR) (Figure 2). Six (9%) patients had stable disease and 14 had disease progression as the best response. Changes in monoclonal protein as per central laboratory are shown in supplemental Figure 3. For all patients, median TTP was 9.5 months (95% confidence interval [CI], 5.7, 10.4), and among responders, the median DOR was 9.7 months (95% CI, 7.4, 15.8) (supplemental Figure 4). Median TTP and DOR by response category are provided in supplemental Figure 5 and supplemental Table 5.

Patients who were nonrefractory to bortezomib (ie, naive or sensitive) were more likely to respond (ORR, 90% [35/39]; ≥VGPR, 64% [25/39]) than those whose disease was refractory to prior bortezomib (ORR, 31% [8/26]; ≥VGPR, 8% [2/26]). Three out of the 8 (38%) bortezomib-refractory responders were t(11;14) positive. Median TTP (11.3 [95% CI, 10.2, –] vs 1.8 [95% CI, 1.2, 4.9] months) and DOR (10.2 [95% CI, 8.8, –] vs 4.2 [95% CI, 2.3, 8.1] months) were longer for bortezomib-nonrefractory patients than bortezomib-refractory patients (Figure 3A). Response rates appeared less dependent on prior exposure to lenalidomide, with overall responses seen in 60% (21/35) of the lenalidomide-refractory subgroup and 72% (21/29) in the lenalidomide-nonrefractory subgroup. The ORR was 84% (16/19) for those nonrefractory to bortezomib but refractory to any immunomodulatory drug and 95% (18/19) for patients who were nonrefractory to bortezomib and any immunomodulatory drug.

Patients who received 1 to 3 prior lines of therapy had an ORR of 89% (33/37), 65% (24/37) had a ≥VGPR. In patients with 4 to 6 prior lines of therapy, ORR was 50% (10/20), and 20% (4/20) had a ≥VGPR. One of 9 patients who had >6 prior therapies achieved PR. Median TTP and DOR were longer with 1 to 3 prior therapies than with >3 prior therapies (Figure 3B).

The subset of patients who were nonrefractory to bortezomib and had 1 to 3 prior therapies had an ORR of 97% (29/30; Figure 2), with the highest ORR (100% [11/11]) observed in patients who were bortezomib naive and had 1 to 3 prior lines of therapy.

A total of 9 patients had the t(11;14) translocation, a subset of MM reported to respond to venetoclax monotherapy (S.K., Jonathan L. Kaufman, Cristina Gasparetto, Joseph Mikhael, Ravi Vij, Brigitte Pegourie, Lofti Benboubker, T.F., Martine Amiot, P.M., E.A.P., Stefanie Alzate, Martin Dunbar, Tu Xu, Suresh K. Agarwal, S.H.E., J.D.L., J.A.R., P.C.M., M.V., and C.T., manuscript submitted June 2017). Of the 9 patients with t(11;14), 6 received a target dose <800 mg, of whom 4 had at least a PR. The other 3 patients received doses ≥800 mg, and all achieved at least a PR. Therefore, clinical responses were observed in 78% (7/9) of these patients. Some of the responses reported in more-refractory patients were in the t(11;14) subgroup, where 4 patients were double refractory to bortezomib and lenalidomide, and 1 achieved CR and 2 achieved PR as a best response (1 received 600 mg, 1 received 800 mg, and 1 received 1200 mg daily venetoclax). The 1 double-refractory patient who did not respond to therapy received 50 mg daily venetoclax. Also, 4 patients with t(11;14) had >3 prior lines, with 1 achieving VGPR and 2 achieving PR. However, responses to the combination therapy were similar to patients without t(11;14), with an ORR of 65% (37/57).

Baseline bone marrow aspirate samples were available from 52 patients, of which 45 were evaluable for BCL2 (BCL-2), BCL2L1 (BCL-X_L), and MCL1 (MCL-1) gene expression analysis. A broad range of BCL2 expression was observed (log2 median: 2.68 [range, −3.17 to 10.09]). Higher BCL2 levels were observed in patients who achieved PR or better (log2 median, 3.01 vs 0.87; P < .01) (Figure 4) and also in patients who had 1 to 3 prior therapies (log2 median, 3.03 vs 0.94; P < .05) (supplemental Figure 6). No association was observed between BCL2L1 or MCL1 gene expression and response or number of prior therapies (Figure 4; supplemental Figure 6). Based on BATTing (bootstrapping and aggregating thresholds from trees),²⁹  a threshold value of log2 BCL2 ≥3.0 was determined to provide optimum selection of patients likely to have a response. This cutoff value was consistent with the preclinical threshold associated with venetoclax sensitivity (log2 BCL2 ≥ 2.7).²²  Seventeen of 18 (94%) patients with high BCL2 expression achieved at least PR, with 12 patients (66%) achieving ≥VGPR (Figure 5). Sixteen of 27 patients with low BCL2 expression achieved at least PR (ORR, 59%), with 6 patients (22%) achieving ≥VGPR (Figure 5). Median TTP (11.6 vs 5.7 months, P = .0044) and DOR (10.2 vs 7 months, P = .0550) were longer for patients with high BCL2 expression than those with low BCL2 expression.

Overall response and ≥VGPR response rates increased with increasing venetoclax dose through 800 mg across all patients (Figure 6A). Within the subgroup of patients who were bortezomib nonrefractory and who had 1 to 3 prior therapies, ORR was at or near 100% across all dose ranges (Figure 6B), while ≥VGPR response rates increased with venetoclax dose through 800 mg. Grade 3/4 neutropenia appeared to increase at doses above 800 mg (supplemental Table 6). An association between higher venetoclax exposures and higher neutropenia rates was confirmed in an exposure–response analysis.³⁰  Based on both safety and efficacy data, the RP2D of venetoclax in combination with bortezomib and dexamethasone was determined to be 800 mg.

Peak venetoclax concentrations were attained at 6 to 8 hours postdose (supplemental Table 7 and supplemental Figure 7). Venetoclax half-life could not be estimated in the study due to the limited sampling after the time to maximum concentration observed. Estimated venetoclax pharmacokinetic parameters were consistent with those reported in literature for venetoclax alone,^12,31  indicating that bortezomib did not affect venetoclax pharmacokinetics.

Dysregulation of apoptotic pathways in plasma cells, often via aberrant BCL-2 and/or MCL-1 overexpression, is thought to play a major role in the development and progression of MM.^3,6  Antagonizing BCL-2 and MCL-1 function to induce apoptosis is thus a compelling therapeutic approach in MM that can be achieved with the combination of venetoclax with bortezomib and dexamethasone.

In this phase 1b study, high response rates (68% ORR and 40% ≥VGPR) were reported among patients with relapsed/refractory MM treated with the combination of venetoclax, bortezomib, and dexamethasone. These response rates were observed despite most patients (39/66) receiving a dose of venetoclax that was inferior to the RP2D of 800 mg/day. Patients who were nonrefractory to bortezomib and who had received 1 to 3 prior lines of therapy had 97% ORR, with 73% of these patients achieving a ≥VGPR response. These patients also experienced more-durable responses and longer TTP compared with patients who were bortezomib refractory or who had received more prior lines of therapy. Though no direct comparisons can be made between trials, the 94% response rate for patients who were bortezomib non-refractory and who received 1 to 3 prior therapies appears to be encouraging based on recent studies combining other novel drugs with bortezomib and dexamethasone, such as panobinostat or daratumumab, as well as carfilzomib with dexamethasone.^32,33  Furthermore, these results are superior to those achieved in patients who received bortezomib and dexamethasone with elotuzumab.³⁴  Prior therapy with lenalidomide had a lesser effect on response rates, as 60% of the lenalidomide-refractory and 72% of lenalidomide-nonrefractory patients achieved PR or better. In those patients refractory to any immunomodulatory drug but nonrefractory to bortezomib, 84% had a clinical response. Conversely, only a third of the patients refractory to bortezomib had an overall response, suggesting that venetoclax cannot fully overcome resistance to bortezomib and potentially requires additional MCL-1 downregulation by bortezomib for broader activity in the non-t(11;14) population.

A higher ratio of BCL2 to MCL1 and BCL2L1 (BCL-X_L) expression is commonly found in MM cells harboring the t(11;14) translocation, which correlates with sensitivity to venetoclax in vitro.^14,22  Venetoclax monotherapy was evaluated in a phase 1 study (#NCT01794520) in which almost all clinical responses were in patients with relapsed/refractory t(11;14) MM (S.K., Jonathan L. Kaufman, Cristina Gasparetto, Joseph Mikhael, Ravi Vij, Brigitte Pegourie, Lofti Benboubker, T.F., Martine Amiot, P.M., E.A.P., Stefanie Alzate, Martin Dunbar, Tu Xu, Suresh K. Agarwal, S.H.E., J.D.L., J.A.R., P.C.M., M.V., and C.T., manuscript submitted June 2017). Furthermore, biomarker analyses in that study showed a correlation between clinical response and high BCL2:BCL2L1 and BCL2:MCL1 ratios.

In the present study, clinical responses were observed regardless of t(11;14) status. ORRs of 78% and 65% were observed in patients with and without t(11;14) MM, respectively. This broader activity of venetoclax beyond t(11;14) in the combination regimen was anticipated, as modulation of BCL-2 family members by bortezomib and dexamethasone sensitizes MM cells to venetoclax.²²  Interestingly, higher BCL2 levels were observed in patients who achieved at least a PR or better. In contrast, no association was observed between BCL2L1 or MCL1 gene expression and clinical response with venetoclax when combined with bortezomib and dexamethasone.

Importantly, this combination also had an acceptable safety profile in patients with relapsed/refractory MM. Mild gastrointestinal toxicities were the most common AEs reported, and cytopenias were the most common grade 3/4 AEs; these were manageable and did not lead to study discontinuation. Neutropenia could be managed with the use of growth factors.

Overall, the clinical results seen with this combination are encouraging and support an ongoing phase 3 trial (#NCT02755597) of this regimen in patients with relapsed/refractory MM. Agents that target these BCL-2 family members, including venetoclax and bortezomib, provide a novel therapeutic approach for MM based on a strong mechanistic rationale. Additionally, biomarker data on BCL2 expression indicate that venetoclax may offer a unique biologic-driven approach in MM.

The authors thank the patients and their families, as well as the study coordinators and support staff. Medical writing support was provided by Sharanya Ford, statistical programming support was provided by Ruiling Zhang and Srikanth Birru, biomarker analysis support was provided by Yan Sun, data analysis support was provided by Ming Zhu, and clinical trial operational support was provided by Stefanie Alzate; all are employees of AbbVie.

Venetoclax is being developed in a collaboration between AbbVie and Genentech. AbbVie and Genentech provided financial support for the study and participated in the design, study conduct, analysis and interpretation of data, as well as the writing, review, and approval of the manuscript. This study was supported by research funding from AbbVie and Genentech to P.M., A.C.-K., A.W.R., A.B.A., T.F., S.K., C.T., and S.J.H.

Contribution: P.M., A.W.R., S.K., C.T., J.A.R., P.C.M., S.J.H., E.A.P., J.D.L., S.H.E., and M.V. designed research; P.M., A.W.R., S.K., C.T., J.A.R., P.C.M., S.J.H., A.C.-K., A.B.A., and T.F. collected data; P.M., A.W.R., S.K., C.T., J.A.R., P.C.M., S.J.H., A.C.-K., A.B.A., T.F., E.A.P., J.D.L., S.H.E., M.V., J.C., W.M., J.J., A.H.S., and K.J.F. analyzed and interpreted data, and wrote the manuscript.

Conflict-of-interest disclosure: P.M. has received honoraria from Celgene, Janssen-Cilag, Takeda, Novartis, and Amgen and has served as a consultant for Celgene, Janssen, Takeda, Novartis, and Amgen. A.W.R. is an employee of Walter and Eliza Hall Institute of Medical Research, which receives milestone and royalty payments related to venetoclax, and has received research funding from AbbVie, Servier, Janssen, Merck, and BeiGene. A.B.A. is an employee of Celgene, has served as a consultant for Amgen and Millennium, and has participated in a speakers’ bureau for Onyx, Janssen, and Novartis. T.F. has served as a consultant and has participated in a speakers’ bureau for Janssen. S.K. has received honoraria from Skyline Diagnostics, Kesios Therapeutics, and Noxxon Pharma; has served as a consultant for Takeda, Janssen Oncology, Amgen, Bristol-Myers Squibb, and AbbVie; and has received research funding from Celgene, Takeda, AbbVie, Novartis, Sanofi, Janssen Oncology, and Merck. C.T. has received research funding from AbbVie. E.A.P. is an employee of and owns stock in Genentech. J.C., W.M., J.J., K.J.F., J.D.L., S.H.E., J.A.R., and M.V. are employees of and own stock in AbbVie. A.H.S. and P.C.M. are employees of, own stock in, and have patents with AbbVie. S.J.H. has participated in a speakers’ bureau for Janssen-Cilag and has received research funding from Janssen-Cilag and AbbVie. A.C.-K. declares no competing financial interests.

Correspondence: Philippe Moreau, Hematology Department, University Hospital Hôtel-Dieu, Place Ricordeau, 44093 Nantes, France; e-mail: philippe.moreau@chu-nantes.fr.