A phase 1 study of ibrutinib in combination with R-ICE in patients with relapsed or primary refractory DLBCL

The standard treatment of chemosensitive-to-salvage therapy (ST) relapsed or refractory (rel/ref) diffuse large B-cell lymphoma (DLBCL) is high-dose chemotherapy followed by autologous stem cell transplantation (HDT-ASCT).¹  In the postrituximab era, all ST regimens are equally disappointing with ∼50% of patients achieving either partial remission (PR) or complete remission (CR) enabling procession to HDT-ASCT.^2,3  Depth of remission to ST by modern functional imaging criteria⁴  is singularly prognostic to post–HDT-ASCT outcomes.⁵  Clearly, improvement in ST would translate into more cures of rel/ref DLBCL.

The Bruton tyrosine kinase inhibitor ibrutinib has recently demonstrated favorable tolerability and activity in rel/ref DLBCL, particularly of non-germinal center (non-GC) phenotype.⁶  Given these data, and the unmet need of improving ST for rel/ref DLBCL, we report a phase 1 study of ibrutinib in combination with standard rituximab, ifosfamide, carboplatin, and etoposide (R-ICE) for patients with rel/ref DLBCL.

This study was approved by the Memorial Sloan Kettering Cancer Center Institutional Review Board. Transplant-eligible patients with biopsy-proven CD20⁺ rel/ref DLBCL to a single line of rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP)–like therapy were eligible, including those with transformed histologies. Cell of origin (COO) was determined by immunohistochemical (IHC) Hans criteria.⁷  Double-expressor lymphomas were defined as MYC expression ≥40% and BCL-2 expression ≥50% of the tumor cells by IHC. All patients underwent pretreatment bone marrow biopsy and staging studies per modern functional imaging criteria.⁴  Patients were required to have normal end-organ function. Patients were ineligible if they had central nervous system involvement or serologic evidence of viral hepatitis. HIV⁺ patients were eligible if they were on a stable antiretroviral regimen and had a CD4⁺ ≥429/µL.

Study treatment consisted of 3 cycles of R-ICE, as previously reported,⁸  in combination with ibrutinib days 1 to 21 of each cycle. Ibrutinib was dosed 3 by 3; dose level (DL) 1 was 420 mg, DL 2 was 560 mg, and DL 3 was 840 mg daily. All patients received growth factor support with either pegylated granulocyte colony-stimulating factor or daily granulocyte colony-stimulating factor with CD34⁺ hematopoietic progenitor cell (HPC) harvest allowable after cycle 2 or cycle 3. Patients underwent interim restaging evaluation with computed tomography chest/abdomen/pelvis following 2 cycles of protocol therapy to document lack of progression of disease. Full restaging was performed at the end of 3 cycles of therapy.⁴  Deauville 1 to 3 responses were CR, Deauville 4 were PR, and Deauville 5 were stable disease or progression of disease. Patients were evaluable for response if they completed 1 full cycle of protocol therapy.

Dose-limiting toxicities (DLTs) were defined by Common Terminology Criteria for Adverse Events (CTCAE; version 4.0) as either grade 4 hematological toxicities or at least grade 3 nonhematologic toxicity requiring >1 week delay of treatment, or unresolved ibrutinib-related at least grade 2 nonhematologic toxicities within cycle 1. The study included an expansion cohort of patients at the declared maximum tolerated dose DL 3. Kaplan-Meier methodology was used to estimate overall survival (OS), progression-free survival (PFS), and event-free survival (EFS). Events included death, progression, and/or any reason for discontinuation of study treatment. All analyses were conducted using R (v3.3.2).

The characteristics of the 21 patients enrolled and treated in the study are listed in Table 1. There were no DLTs seen at any DL, and DL 3 was expanded to n = 15. One patient with non-GC DLBCL was removed from study per the treating physician’s decision with new grade 3 atrial arrhythmia on cycle 1 day 13, and thus was not evaluable for response. Ninety-five percent of patients experienced expected and transient grade ≥3 hematologic toxicities with recovery prior to each cycle. The median number of platelet transfusions per patient over 3 cycles was 2 (range, 0-11). Fifteen of the 20 patients evaluable for response underwent R-ICE–primed CD34⁺ HPC apheresis procedures on study; 14 collected HPCs with a median of 5.5 × 10⁶ CD34⁺/kg (range, 1.7-8.6). The only patient who failed to collect HPCs was the sole HIV⁺ patient on study at DL 3 in the setting of febrile illness. There were 4 episodes of febrile neutropenia on study, which compares similarly to 8 episodes of febrile neutropenia in the original R-ICE study (n = 36 total patients).⁸  Additionally, 4 grade 3 infections (influenza, parainfluenza, Clostridium difficile, and a staphylococcal catheter-associated bloodstream infection) were observed. The n = 12 at least grade 3 nonhematologic toxicities on study can be found in the supplemental Table (available on the Blood Web site). Grade 3 events of potential cardiovascular interest were observed on study. These included 2 sinus tachycardia events experienced by patients in the setting of grade 3 anemia during cycle 3 and C difficile infection 3 weeks following completion of protocol therapy, respectively. Two grade 3 syncopal events occurred: 1 in a patient during cycle 1 without consequence, and 1 in a patient in the setting of a new influenza diagnosis. Lastly, 1 patient, with large tumor burden Richter-transformed disease, experienced grade 3 hypotension on cycle 1 day 1 rituximab infusion followed by tumor lysis, which subsequently resolved following inpatient management. Other notable events on study include: 1 patient removed for asymptomatic pneumatosis coli on interim restaging imaging, and 1 patient who self-discontinued study for grade 2 nausea. Both of these patients achieved CR following 2 cycles, proceeded to HDT-ASCT, and remain progression-free.

Overall response rate (ORR) was 90% (68%-99%) in 20 patients (CR = 11; PR = 7). By intent to treat (ITT), 8 of 9 patients with non-GC DLBCL achieved a CR (89% [52%-99%]); n = 1 aforementioned patient was removed for atrial arrhythmia during cycle 1. Six of 9 non-GC patients were double-expressor lymphoma by IHC. All 5 patients with rel/ref Richter transformation achieved a response (CR = 2; PR = 3) with 2 of these patients demonstrating 17p deletion on cytogenetic studies. Table 2 includes response rates according to COO/histologic subtype. In univariate analysis, rel/ref non-GC DLBCL demonstrated a significantly increased rate of CR compared with other subtypes (P = .008). At a median follow-up of 14 months for survivors, the 1-year EFS, PFS, and OS by ITT was 0.51 (95% confidence interval [CI], 0.33-0.79), 0.62 (95% CI, 0.43-0.9), and 0.65 (95% CI, 0.46-0.93), respectively (Figure 1A). Of the 18 patients who achieved a response on study, n = 14 and n = 2 proceeded to HDT-ASCT and allogeneic hematopoietic transplantation, respectively. The 1-year PFS and OS of transplanted patients on study are 0.73 (95% CI, 0.51-0.99) and 0.82 (95% CI, 0.62-0.99), respectively (Figure 1B).

Ibrutinib at doses up to 840 mg daily plus R-ICE demonstrated favorable tolerability and an encouraging signal of efficacy with 90% of patients responding to protocol therapy and >50% achieving a CR. All patients with non-GC DLBCL completing at least 1 cycle of therapy achieved a metabolic CR. We have recently reported that FDG-PET CR, per International Conference on Malignant Lymphoma criteria, is the single greatest prognostic factor toward OS in patients proceeding to HDT-ASCT in the postrituximab era. Recent large prospective randomized studies have shown equivalency of all platinum-containing ST programs^2,3  ∼50% ORR of previously rituximab-exposed patients. We selected R-ICE in combination with ibrutinib given Memorial Sloan Kettering Cancer Center’s experience in developing R-ICE, as well as reduced toxicity and myelosuppression of R-ICE in contrast to R-DHAP (rituximab, dexamethasone, cytarabine, cisplatin).³  Importantly, we have demonstrated that additive myelosuppression with ibrutinib in combination with R-ICE does not prohibit effective chemotherapy-primed mobilization of HPCs.

Groups have investigated modifying platinum-based ST backbones with the goal of improving ORR. The recently reported multicenter, international randomized phase 3 ORCHARRD study failed to demonstrate improvement of ORR, PFS, or OS with substitution of ofatumumab for rituximab.⁹  Other groups have investigated adding agents to R-ICE chemotherapy for rel/ref DLBCL. The group from Hackensack University combined lenalidomide with R-ICE in a phase 1 study with 9 of 15 patients responding to protocol therapy.¹⁰  Interestingly, improved response rates in GC patients were seen despite the previously reported relatively exclusive single-agent activity of lenalidomide in non-GC patients.¹¹  A phase 1 of vorinostat in combination with R-ICE was reported in 2013 in a multicenter study with an ORR of 70% and CR <33%.¹²  Lastly, a randomized phase 2 study of R-ICE plus the anti-CD40 monoclonal antibody dacetuzumab vs R-ICE plus placebo demonstrated no difference in response rate in the 2 groups.¹³  This study was stopped short of planned accrual due to interim futility analysis.

In summary, this phase 1 study confirms that ibrutinib in doses up to 840 mg daily combined with R-ICE is well tolerated with no observed DLTs and preserved ability to mobilize HPCs. Early signal of activity, especially in non-GC patients, warrants further investigation in this subgroup. A phase 2 multicenter study is currently being planned.

The authors thank the staff of the Lymphoma and Adult BMT Services at Memorial Sloan Kettering Cancer Center for their enduring dedication to patient care.

This work was supported by the Memorial Sloan Kettering Cancer Center Core grant (P30 CA008748) from the National Institutes of Health, National Cancer Institute; the National Institutes of Health, National Cancer Institute Cancer Therapeutic Evaluation Program; and the Steven A. Greenberg Aggressive Lymphoma Research Fund.

Contribution: C.S.S. and C.H.M. designed the study; C.S.S., M.J.M., H.S., S.M.D., A.Y., and C.H.M. interpreted the data; and C.S.S., M.J.M., P.D., J.G., A.K., A.N., M.L.P., C.S.P., D.J.S., A.D.Z., S.J.M., S.T.M., A.I.C., A.Y., and C.H.M. collected and analyzed the data and wrote the manuscript.

Conflict-of-interest disclosure: C.S.S. received research support from Pharmacyclics. M.J.M. received research support and honoraria from Genentech. P.D. received honorarium and served on an advisory board for Seattle Genetics. J.G. received honoraria and served on an advisory board for Samus Therapeutics, Gilead, AbbVie Pharmaceuticals, Genentech, Bayer, Aratana, Arcus Medica, Incyte, Mass Medical International, Merck, Orexo, and Royal Bank of Canada. A.K. received research funding from AbbVie Pharmaceuticals, Adaptive Biotechnologies, Seattle Genetics, Celgene, and Pharmacyclics, and served on an advisory board for Celgene. A.N. received research funding, speaker’s fees, and travel fees from Pharmacyclics. A.D.Z. consulted and served on an advisory board for Genentech, Gilead, Celgene, Janssen, Amgen, Novartis, and Adaptive Biotechnologies; consulted for Pharmacyclics; received research funding from Roche, Gilead, and MEI Pharmaceuticals; and is the Data Monitoring Committee Chair for Beigene. A.Y. received honorarium from Seattle Genetics, Takeda Millennium, Janssen, AbbVie Pharmaceuticals, and Pharmacyclics. C.H.M. consulted for, served on an advisory board for, and received research funding from Seattle Genetics. The remaining authors declare no competing financial interests.

Correspondence: Craig S. Sauter, Memorial Sloan Kettering Cancer Center, Box 276, 1275 York Ave, New York, NY 10065; e-mail: sauterc@mskcc.org.