Selinexor Combined with Ibrutinib Demonstrates Tolerability and Safety in Advanced B-Cell Malignancies: A Phase I Study

For patients with chronic lymphocytic leukemia (CLL), mantle cell lymphoma (MCL), and refractory diffuse large B-cell lymphoma (DLBCL), cure is very unlikely. Chronic activation of the B-cell receptor signaling pathway is critical for the survival of CLL, DLBCL, and MCL cells (1–3). Bruton's tyrosine kinase (BTK) is a key component of the B-cell receptor signaling pathway. Treatment with ibrutinib, an oral BTK inhibitor, demonstrated efficacy and tolerability and has been approved for the treatment of CLL, MCL, and other B-cell malignancies (4–6). Ibrutinib must be given continuously to retain response and achievement of complete remission (CR) is rare (4–6). Resistance to ibrutinib therapy is common and leads to poor clinical outcomes (7, 8). There is significant clinical need to develop therapeutic strategies to induce deeper remissions. Rationally designed combination strategies are one approach to this problem.

Selinexor is a selective oral inhibitor of exportin-1 (XPO1)-mediated nuclear transport. XPO1 is an essential shuttle protein that identifies cargo molecules including ribonucleoprotiens (RNP) and tumor suppressor proteins (TSP), facilitating their transport from the nucleus into the cytoplasm thereby regulating the subcellular localization of these cargos and influencing their activity (9). XPO1 inhibition induces apoptosis and suppresses downstream effectors of B-cell activation, proliferation, and migration in CLL cells (10). XPO1 inhibition down-modulates BTK, and synergizes with ibrutinib in CLL cells cultured ex vivo, and in mouse models of acquired ibrutinib resistance. Selinexor's ability to overcome ibrutinib-mediated resistance was attributed to the drug's ability to target multiple B-cell receptor signaling nodes in a manner independent of BTK kinase activity, and thereby, preserving the ability to block adaptive signaling responses in ibrutinib-resistant subclones (11). In a phase I trial of relapsed/refractory (RR) lymphoid malignancies, selinexor was tolerable and resulted in an overall response rate (ORR) of 31% (12). The maximum tolerated dose (MTD) was 60 mg twice weekly (12). ORR was 28% in a phase II study of 127 patients with RR DLBCL (13). Median duration of response (DOR) was 9.3 months; 23 months in patients with CR (13). Selinexor was approved for patients with DLBCL.

With a goal of inducing deep remissions with synergistic dual blockade of BTK with ibrutinib and selinexor, we designed a phase I trial for patients with RR CLL and NHL. We hypothesized that selinexor could be combined with continuous ibrutinib at a safe and tolerable dose. Herein, we report the results of this study.

After obtaining written informed consent, patients with CLL, DLBCL, Richter transformation (RT), or MCL were enrolled at Ohio State University (Columbus, OH) and University of Utah (Salk Lake City, UT). Eligible patients were ≥18 years, had ≥1 prior therapy, and Eastern Cooperative Oncology Group (ECOG) performance status 0 to 1. Patients with CLL met criteria for therapy based on 2008 International Working Group for CLL (iwCLL) criteria. Patients received selinexor (per dose level) orally 1 to 2 times weekly (3 weeks of 4-week cycle) and daily oral ibrutinib (420 mg) starting cycle 1, day 8. To minimize nausea and weight loss previously associated with selinexor, dexamethasone (12 mg or equivalent) and ondansetron (8 mg or equivalent) were administered with each dose of selinexor. Dexamethasone could be tapered and discontinued per discretion of treating physician. Treatment cycles were repeated until disease progression, intolerance, death, or discontinuation of trial participation. The study was conducted in accordance with the Declaration of Helsinki and approved by local Institutional Review Boards.

Pharmacodynamic/pharmacokinetic analyses were performed as described in the Supplementary data.

The primary objective was to determine the MTD of the combination. Dose-limiting toxicity (DLT) was defined and evaluated as detailed in the Supplementary Data. Dose-finding was guided by a Bayesian continual reassessment method using a two-parameter logistic model and a target DLT rate of 30% (14). MTD was defined as the highest dose with model-estimated DLT rate less than 30% when the first of 10 to 12 patients were treated at a dose level or 36 patients were evaluated for DLT, at which 10 patients with CLL and 10 patients with NHL were to be treated in expansion cohorts. Adverse events (AE) were graded according to Common Terminology Criteria of Adverse Events (CTCAE) version 4 and were summarized for all patients who received treatment.

Response was assessed by iwCLL 2008 for patients with CLL or the 2007 International Working Group Response Criteria for patients with NHL. Additional statistical methods are included in the Supplementary Data.

The data generated in this study are available within the article and its Supplementary Data files.

Thirty-four patients were enrolled from June 2015 through July 2019. Included patients had diagnoses of CLL (n = 16) or NHL (n = 18; 9 RT, 6 DLBCL, and 3 MCL). Median age was 65 years (range = 50–80). Median number of prior therapies was 4 (range = 1–14), and 20 (59%) patients had previously received ibrutinib, including 11 (69%) of 16 patients with CLL. Baseline characteristics are detailed in Table 1.

MTD is 40 mg of selinexor weekly (days 1, 8, 15 of 28-day cycle) and 420 mg daily ibrutinib [dose level 1 (DL1)]. After the first 2 evaluable patients treated at DL1 did not have DLT, a second cohort of 2 patients was treated at DL2 with 30 mg of selinexor twice weekly and 420 mg daily ibrutinib. While the patients did not have grade ≥3 fatigue, both patients noted that fatigue was limiting their daily activities. While the grade of the toxicity did not officially meet DLT criteria, they were considered to have experienced DLT since they were not able to continue on the planned dose of the selinexor. As such, DL2 and all higher dose levels were deemed intolerable.

At data cut-off, 3 patients remain on therapy. Of the 31 patients who discontinued therapy, median duration of therapy was 3.2 months (range = 0.2–37 months). Most common reasons for therapy discontinuation were progressive disease (PD; 65%) and toxicity (23%). AEs leading to discontinuation were nausea and vomiting (n = 2), pneumonitis (n = 2), mouth sores, palpitations, and fatigue and abdominal cramping (n = 1 each). Hematologic AEs of any grade were common with 20 (59%), 19 (56%), and 13 (38%) patients experiencing thrombocytopenia, anemia, and neutropenia, respectively. Grade ≥ 3 hematologic AEs were uncommon with 8 (24%), 6 (18%), and 4 (12%) patients experiencing thrombocytopenia, anemia, and neutropenia, respectively. Most common nonhematologic AEs of any grade were fatigue (56%), nausea (53%), anorexia (41%), and diarrhea (41%). Thirty-five nausea events occurred in 18 patients during treatment: 40%, 14%, 6%, 14%, 6%, 3%, and 6% in cycles 1, 2, 3, 4, 5 to 7, 16, and 22, respectively. Grade ≥ 3 nonhematologic AEs were uncommon with 6 patients (18%) experiencing hypertension and 5 patients (15%) experiencing asymptomatic hyponatremia (Table 2). A complete list of AEs can be found in Supplementary Table S2. AEs of interest are detailed in Table 2. There were 4 patient deaths while on study; 2 related to disease progression and 2 related to infection. The range of time from last recorded dose of study treatment to death was 7 to 81 days.

Of all patients (n = 34), the ORR was 32%. ORRs for CLL, DLBCL, MCL, and RT were 44%, 33%, 33%, and 11%, respectively. An additional 16 (47%) patients maintained stable disease (SD; Supplementary Fig. S5). Figure 1 depicts the DOR. Of the 16 patients with SD, 15 had more than 6 months of follow-up, and 20% (3/15) had maintained SD for more than 6 months. In the CLL cohort, 94% (15/16) achieved at least SD as the best response. The patient who achieved CR with no detectable residual disease received only one prior therapy, and discontinued selinexor after maintaining CR for almost 24 months. The patient who achieved complete response with incomplete count recovery (CRi) had previously received ibrutinib with a known BTK resistance mutation and a complex karyotype. This patient discontinued both agents secondary to presumed ibrutinib-related cardiac dysrhythmia toxicity while in CRi after a duration of remission of 3 months. Of 9 evaluable patients with CLL with known BTK mutations, 2 responded to therapy [1 CRi, 1 partial response (PR)], and 7 had SD. Median range of DOR or SD in this group was 3.7 (range = 0.2–13.4) months. Of the 6 patients with DLBCL, 1 achieved CR. The patient with CR had received one prior therapy and benefited from a prolonged 22-month DOR. One of 2 patients with DLBCL with SD maintained SD for 36 months. Of 9 patients with RT, 1 achieved CR and 5 achieved SD. The patient with RT who achieved CR had three prior therapies, including ibrutinib. This patient discontinued therapy at 1 month after achieving CR to receive allogeneic hematopoietic stem cell (HSC) transplantation. In the 5 patients with RT who achieved SD, the median time from first achieving SD to progression or death was 1.3 months (range = 0.4–4.5). In the 11 patients on study who achieved at least PR, only 3 had progressed or died, and the median DOR was not reached (NR; 95% CI, 3.0 months–NR).

Median progression-free survival (PFS) for patients with CLL and NHL were 8.9 (95% CI, 3.9–16.1) and 2.7 (95% CI, 0.7–5.4) months, respectively (Fig. 2A). Median PFS for patients with CLL who received prior ibrutinib or did not was 4.6 months (95% CI, 3.1–12.9) and NR (95% CI, 8.9–NR), respectively. For patients with CLL who received ≥3 prior therapies, 91% (10/11) had PD. All but 1 of these patients had previously received ibrutinib. Only 1 of 5 patients (20%) who received 1 to 2 prior therapies progressed, and that patient had previously received ibrutinib. With median follow-up of 33.3 (range = 12.9–45.3) and 24.2 (range = 17.7–50.3) months among survivors, median OS for patients with CLL and NHL was 58.9 (95% CI, 19.6–58.9) and 5.1 (95% CI, 3.0–14.8) months, respectively. Estimated 2-year OS was 73.7% (95% CI, 44.1–89.2) and 27.8% (95% CI, 10.1–48.9) for patients with CLL and NHL, respectively (Fig. 2B).

A detailed summary of the pharmacodynamic and pharmacokinetic analyses can be found in the Supplementary Data. Notably, in the 1 patient with RT who achieved CR, at least 2-fold nuclear accumulation was observed at the C1D1-posttimepoint for c-Myc, FOXO3a, IκBα, and BCL2 proteins (Supplementary Fig. S2). At the C1D2, a reduction in Myc mRNA expression was observed in patients with better clinical response, whereas a significant increase in Myc mRNA was variably observed in patients with poor response.

The pharmacokinetic data suggests there is no major drug interaction between selinexor and ibrutinib.

In this study, the combination of ibrutinib and selinexor was tolerable and responses were observed in this heavily pretreated population of patients with CLL and NHL. Efficacy may be greater in earlier lines of therapy. Further investigation of this combination is needed to optimize response rates and durability.

When compared with studies of single-agent ibrutinib or selinexor in patients with RR lymphoid malignancies, the combination resulted in slightly more toxicity than what was seen with single-agent ibrutinib (15, 16). The most common grade 1/2 AEs demonstrated in this phase 1 combination study were fatigue (56%), nausea (50%), anorexia (41%), and diarrhea (35%). These toxicities were similar to what was seen with single-agent selinexor with 49%, 65%, 56%, and 32% experiencing grade 1/2 fatigue, nausea, anorexia, and diarrhea, respectively (12). In a phase II study of patients with relapsed DLBCL treated with single-agent selinexor, similar rates of grade 1/2 fatigue (36%), nausea (52%), anorexia (33%), and diarrhea (32%) were observed (13). As expected, the same toxicities were slightly more than what was seen with single-agent ibrutinib with 37%, 41%, 42%, and 30% of patients experiencing grade 1/2 nausea, fatigue, diarrhea, and anorexia, respectively (15). The comparable toxicity seen in this combination study is likely related to the lower and less frequent dosing of selinexor than what was used in the single-agent phase 1 selinexor study (60 mg in 2 weekly doses) and the aggressive prophylactic measures implemented on this study to minimize nausea and weight loss associated with the drug. Dexamethasone and ondansetron were administered with each dose of selinexor. The incidence of nausea events decreased after cycle 1, and most patients were able to taper or discontinue the dexamethasone over time indicating that tolerance of nausea improved over time for the patients on this combination study. Of note, there are now data to support that the second-generation BTK inhibitors acalabrutinib and zanubrutinib cause less toxicity than ibrutinib (17, 18). In addition, third generation BTK inhibitors such as pirtobrutinib and ARQ-531 have very favorable toxicity profiles (19, 20). A combination of one of these newer agents with selinexor could potentially lead to less toxicity than demonstrated in this study.

The ORR derived from the combination for patients with DLBCL was 33%, which was similar to the ORR demonstrated in patients with DLBCL treated with single-agent selinexor in the phase II study (28%; ref. 13). As there were only 6 patients with DLBCL on our study, limited data precludes determination of whether cell of origin affects response rates. Notably, in the 2 patients with DLBCL who achieved SD, 1 maintained SD status for 36 months and the other for 2 months. Maintaining stable disease control with this regimen could allow patients to be bridged to another therapy, such as chimeric antigen receptor T-cell therapy.

Similarly, patients with RT typically have aggressive disease and 5 of 9 were able to achieve SD with this combination therapy. The median duration of SD was 1.3 months (range = 0.4–4.5), which could prolong survival long enough to allow for these patients to be bridged to another therapy or clinical trial. Notably, the 1 patient with RT who achieved CR demonstrated a robust pharmacodynamic response following initiation of selinexor treatment, where inhibiting XPO1-mediated nuclear export resulted in significant nuclear accumulation of proteins including p53, FOXO3a, MYC, and BCL2. This patient was then able to transition to more definitive therapy with an allogeneic HSC transplantation. Interestingly, while variably observed in this limited sample set, a significant increase in Myc mRNA expression following selinexor treatment was observed in patients with poor clinical response, whereas a decrease in Myc expression was predominantly observed in patients with better response. Aberrant Myc expression has been linked to CLL progression and RT, warranting further evaluation of Myc as a prognostic indicator following ibrutinib/selinexor combination therapy.

The responses demonstrated by the combination were notable in two CLL populations: those with known BTK mutations and those with minimal prior therapy. Two of 10 patients with CLL with known BTK mutations had response to therapy and 7 had SD. This indicates that the dual inhibition of BTK may be able to overcome ibrutinib resistance. All 3 patients with CLL who received only one line of prior therapy responded to this combination; 1 achieved CR with no detectable residual disease, the other 2 had PR. The patient with CR discontinued selinexor after achieving response and has a current DOR of almost 24 months. This response is notable as ibrutinib alone rarely leads to CR and even more rarely leads to remission with no residual disease. This indicates that early dual inhibition of BTK with this combination may deepen responses seen with ibrutinib. As such, a planned expansion cohort aims to study this combination in patients with CLL with only one prior line of therapy.

In summary, this study combining selinexor (40 mg weekly) and ibrutinib (420 mg daily) induced responses and demonstrated tolerability in patients with RR B-cell malignancies. The combination held similar toxicities to what has been demonstrated in studies with single-agent selinexor. Patients with DLBCL and RT maintained disease control with the regimen that may allow for bridging to other therapies. Responses were seen in patients with CLL who had minimal prior therapy. Future study of this regimen will focus on efforts to deepen remissions in patients with CLL receiving ibrutinib therapy.

D.M. Stephens reports grants from Karyopharm during the conduct of the study, as well as personal fees from Eli Lilly and Company, Karyopharm, Pharmacyclics, Innate Pharma, Beigene, Epizyme, TG Therapeutics, Adaptive, AstraZeneca, AbbVie, Celgene, CSL Behring, BMS, and Genentech outside the submitted work. A.S. Ruppert reports other support from Telios Pharma, Inc. outside the submitted work; in addition, A.S. Ruppert reports employment at Eli Lilly and Company as of January 3, 2022, though all work for this manuscript was completed before that date. K.A. Rogers reports grants from Genentech, AbbVie, Novartis, and Janssen, as well as personal fees from AbbVie, Genentech, AstraZeneca, Pharmacyclics, Innate Pharma, Beigene, and Acerta Pharma outside the submitted work. S. Bhat reports personal fees from Acerta and Beigene, as well as grants from AstraZeneca and TG Therapeutics outside the submitted work. S.M. Jaglowski reports grants and personal fees from Kite, Novartis, Juno/BMS, Takeda, and CRISPR outside the submitted work; in addition, S.M. Jaglowski has a patent for U.S. Patent 15/538,279: “BTK Inhibitor Combinations and Dosing Regimen”, pending. J.A. Woyach reports grants from NCI, Leukemia and Lymphoma Society, and Karyopharm during the conduct of the study. J.A. Woyach also reports grants and personal fees from AbbVie, Pharmacyclics, and Janssen; personal fees from AstraZeneca, Beigene, Arqule, Loxo, and Newave; and grants from Schrodinger and Morphosys outside the submitted work. No disclosures were reported by the other authors.

D.M. Stephens: Conceptualization, resources, data curation, formal analysis, funding acquisition, investigation, writing–original draft, project administration, writing–review and editing. Y. Huang: Conceptualization, data curation, formal analysis, validation, investigation, methodology, writing–original draft, writing–review and editing. A.S. Ruppert: Conceptualization, data curation, software, formal analysis, validation, investigation, methodology, writing–original draft, writing–review and editing. J.S. Walker: Conceptualization, resources, data curation, formal analysis, investigation, methodology, writing–original draft, writing–review and editing. D. Canfield: Resources, data curation, formal analysis, validation, investigation, methodology, writing–review and editing. C.B. Cempre: Resources, data curation, formal analysis, validation, investigation, methodology, writing–review and editing. Q. Fu: Data curation, formal analysis, validation, investigation, methodology, writing–review and editing. S. Baker: Data curation, formal analysis, validation, investigation, methodology, writing–review and editing. B. Hu: Data curation, formal analysis, investigation, writing–review and editing. H. Shah: Data curation, formal analysis, investigation, writing–review and editing. R. Vadeboncoeur: Data curation, formal analysis, investigation. K.A. Rogers: Data curation, formal analysis, funding acquisition, investigation, writing–review and editing. S. Bhat: Data curation, formal analysis, investigation, writing–review and editing. S.M. Jaglowski: Data curation, formal analysis, investigation, writing–review and editing. H. Lockman: Data curation, formal analysis, investigation, writing–review and editing. R. Lapalombella: Conceptualization, resources, formal analysis, funding acquisition, validation, investigation, methodology, writing–original draft, writing–review and editing. J.C. Byrd: Conceptualization, resources, data curation, formal analysis, supervision, funding acquisition, validation, investigation, methodology, writing–original draft, writing–review and editing. J.A. Woyach: Conceptualization, resources, data curation, formal analysis, supervision, funding acquisition, validation, investigation, methodology, writing–original draft, writing–review and editing.

The authors would like to thank the following funding sources for research conducted resulting in the preparation of this manuscript:

• (i)  NIH/NCI grant number K23 CA212271 to D.M. Stephens

• (ii)  NIH/NCI grant number NCATS TL1 TR002735 to J.S. Walker

• (iii)  NIH/NCI grant number R35 CA197734 to J.C. Byrd

• (iv)  NIH/NCI R01 CA192928 to R. Lapalombella, J.A. Woyach, and J.C. Byrd

• (v)  Leukemia and Lymphoma Society Scholar in Clinical Research grant number CDP 2331–20 to K.A. Rogers

Images were generated using instruments and services at the Campus Microscopy and Imaging Facility (CMIF) at The Ohio State University. This facility is supported in part by NCI grant number P30 CA016058–45.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.