Safety and Efficacy of Ibrutinib in Combination with Rituximab and Lenalidomide in Previously Untreated Follicular and Marginal Zone Lymphoma: an Open Label, Phase II Study

Abstract

Follicular lymphoma (FL) and marginal zone lymphoma (MZL) are indolent non-Hodgkin lymphomas (iNHL). Median survival for iNHL is around 20 years. Because standard treatments are not curative, patients often receive multiple lines of therapy with associated toxicity—rationally designed, combination therapies with curative potential are needed. The immunomodulatory drug lenalidomide was evaluated in combination with rituximab for the frontline treatment of FL in the phase III RELEVANCE study. Ibrutinib, an oral Bruton tyrosine kinase inhibitor, is active in NHL and was evaluated in combination with lenalidomide, rituximab and ibrutinib (IRR) in a phase I study. We conducted an open-label, phase II clinical trial of IRR for previously untreated FL and MZL. The primary endpoint was progression free survival (PFS) at 24-months. Our study included 48 participants with previously untreated FL grade 1–3a (N=38), or MZL (N=10). Participants received 12, 28-day cycles of lenalidomide (15mg, days 1–21 cycle 1; 20mg cycles 2–12), rituximab (375mg/m2 weekly in cycle 1; day 1 cycles 2–12), and ibrutinib 560mg daily. With a median follow up of 65.3 months the estimated PFS at 24-months was 78.8% (95%CI 68.0–91.4%) and 60-month PFS was 59.7% (95%CI 46.6–76.4%). One death occurred unrelated to disease progression. Grade 3–4 adverse events were observed in 64.6%, including 50% with grade 3–4 rash. IRR is highly active as frontline therapy for FL and MZL. Compared to historical results with lenalidomide and rituximab, PFS is similar with higher grade 3–4 toxicity, particularly rash. The study was registered with ClinicalTrials.gov (NCT02532257).

Precis

After 5-years of follow from a phase II clinical trial of ibrutinib, lenalidomide and rituximab in previously untreated patients with follicular and marginal zone lymphoma the 5-year progression free survival rate was 59.7% and overall survival rate was 97.9%.

Introduction

Follicular lymphoma (FL) and marginal zone lymphoma (MZL) are subtypes of indolent Non-Hodgkin lymphomas (NHL). The majority of patients with indolent NHL present with advanced stage disease and are often asymptomatic. Prognosis is variable and impacted by clinical, biologic, and immunologic factors.^1–4

The typical frontline treatment for indolent NHL in the United States is rituximab in combination with chemotherapy.⁵ Despite high response rates,^{6, 7} and median progression free survival (PFS) exceeding 5 years with the addition of maintenance rituximab,⁸ chemoimmunotherapy is not considered curative. Furthermore, the side effect profile of cytotoxic therapy is an important consideration.⁹

Lenalidomide, a thalidomide derivative, is a second-generation immunomodulatory drug, proposed to have multiple mechanisms of action, including beneficial effects on both tumor and microenvironment. Lenalidomide has been associated with TNF-α inhibitory, T-cell costimulatory, and antiangiogenic activities.¹⁰ The molecular action of lenalidomide, involves its binding to protein targets cereblon, Ikaros, and Aiolos,^11–14 and subsequent effects on protein ubiquitination and degradation.¹⁵

The combination of lenalidomide plus rituximab has synergistic effects against NHL, as demonstrated in in vitro and animal models, by enhancing rituximab-induced apoptosis and rituximab-dependent NK cell-mediated cytotoxicity.^16–19 Based on the observed efficacy of lenalidomide combined with rituximab in relapsed/refractory (r/r) indolent NHL,²⁰ and the expectation of synergy between these agents, our group completed a phase II, single arm study to evaluate the efficacy and safety of lenalidomide and rituximab in patients with untreated, advanced-stage indolent NHL, demonstrating efficacy and safety of the combination.²¹ Similar findings were demonstrated by a phase II, multi-center study investigating the combination in previously untreated patients with indolent NHL.²² Based on these findings, the phase III, randomized international RELEVANCE study compared the efficacy of lenalidomide and rituximab versus chemoimmunotherapy for previously untreated FL patients.²³

Ibrutinib is an oral Bruton tyrosine kinase inhibitor which has activity in patients with NHL including r/r MZL.^24–26 In addition, synergistic downregulation of transcription factors such as IRF4 that amplify NF- kB signaling through concurrent use of lenalidomide and ibrutinib suggests this may be an attractive therapeutic combination.¹¹ A phase I study of ibrutinib in combination with rituximab and lenalidomide in previously untreated stage II-IV FL patients has been completed.²⁷ The study identified a recommended phase II dose and included 22 patients in whom the overall response rate (ORR) was 95%. Rash was a common toxicity with this regimen and was grade 3 in 36%. Given the potential for synergy with lenalidomide combined with ibrutinib, a revision to the dosing schedule during cycle 1 was pursued in an attempt to reduce the frequency and severity of rash in our study.

Here we report the results of a frontline, open-label, phase II clinical trial of ibrutinib, lenalidomide and rituximab for previously untreated patients with FL and MZL. This study builds on our experience with frontline lenalidomide and rituximab with the addition of ibrutinib which targets both malignant B-cells and the tumor microenvironment. The primary endpoint of our study was PFS rate at 24 months.

Patients and Methods

Eligibility criteria

Participants were ≥ 18 years of age and had histologically confirmed CD20+ FL, grade 1, 2, or 3a or MZL. Inclusion criteria included no prior systemic treatment for lymphoma, stage II, III, or IV disease, Eastern Cooperative Oncology Group (ECOG) performance status ≤ 2, and a need for therapy as determined by the treating physician. Additional criteria included adequate hematologic and organ function: absolute neutrophil count (ANC) ≥ 1,000/mm³ (independent of growth factor support), platelet counts ≥ 100,000/mm³ or ≥ 50,000/mm³ if bone marrow involvement with lymphoma, serum aspartate transaminase (AST) or alanine transaminase (ALT) < 3 x upper limit of normal (ULN), creatinine clearance (CrCl) >30 ml/min calculated by modified Cockcroft-Gault formula, and bilirubin < 1.5 x ULN unless bilirubin is due to Gilbert’s syndrome, documented liver involvement with lymphoma, or of non-hepatic origin, in which case bilirubin should not exceed 3g/dL. Women of childbearing potential and men who were sexually active were required to be practicing a highly effective method of birth control during and after the study. Key exclusion criteria included active central nervous system lymphoma, evidence of diffuse large B-cell transformation, grade 3B FL, history of human immunodeficiency virus, or active Hepatitis C Virus, or active Hepatitis B Virus infection, or any uncontrolled active systemic infection, clinically significant cardiovascular disease such as uncontrolled or symptomatic arrhythmias, congestive heart failure, or myocardial infarction within 6 months of screening, or any Class 3 (moderate) or Class 4 (severe) cardiac disease as defined by the New York Heart Association Functional Classification,²⁸ history of stroke or intracranial hemorrhage within 6 months prior to study entry, patients who required chronic treatment with strong CYP3A inhibitors or inducers, and concurrent systemic immunosuppressant therapy (e.g., cyclosporine, tacrolimus, etc., or chronic administration glucocorticoid equivalent of >10mg/day of prednisone) within 28 days of the first dose of study drug.

Study design

We performed an open-label, phase II, single center study designed to assess the efficacy and safety of ibrutinib combined with lenalidomide and rituximab in patients with previously untreated FL and MZL. Participants received 12 cycles of lenalidomide, 15mg orally daily on days 1–21 of cycle 1 (a lower dose was given with cycle 1 in an attempt to mitigate risk of rash), and 20mg orally daily on days 1–21 of cycles 2 through 12. Cycles were 28 days in length. Rituximab 375mg/m² was administered intravenously on days 1, 8, 15, and 22 of cycle 1, and day 1 of cycles 2 through 12. Ibrutinib 560mg orally daily was initiated on day 1 of cycle 1 and continued through 12 cycles. Therapy was continued for 12 cycles or until disease progression, unacceptable toxicity, or voluntary withdrawal. The dosing schedule for patients with renal dysfunction is included in the Supplemental methods.

The study was approved by the ethics committee and institutional review board at the University of Texas MD Anderson Cancer Center. Participants provided written informed consent consistent with federal and institutional guidelines. The study was registered with ClinicalTrials.gov (NCT02532257).

Outcomes

The primary endpoint was PFS at 24 months as assessed by the investigator. Secondary endpoints included best and 120 week complete response rate (CRR) and ORR by Lugano classification,²⁹ duration of response, event free survival, time to next anti-lymphoma treatment, overall survival (OS) and safety. Exploratory endpoints included immunophenotyping of PBMCs to determine alteration in immune cell subsets, identification of signaling pathways or biomarkers that predicted sensitivity or resistance by gene expression profiling, and assessment of cytokines and chemokines.

Exome sequencing

DNA and RNA were extracted from pre-treatment FFPE tissue using AllPrep FFPE kits (Qiagen) according to the manufacturer’s protocol. DNA was repaired using NEBNext FFPE DNA Repair Mix (NEB) then 50–150ng used for library preparation with KAPA HyperPrep Plus kits (Roche). Samples were pooled, exonic sequences enriched with Nimblegen Exome v3.0 probes (Roche), and the enriched libraries sequenced on a HiSeq4000 instrument (Illumina) at the MD Anderson Advanced Technology Genomics Core. Data were analyzed using our previously described pipeline for detecting variants without matched germline control.³⁰ The m7-FLIPI risk score was calculated as previously described.³¹

Cytokine and chemokine analysis

Serum was isolated from peripheral blood and analyzed by ELISA using the following kits: V-PLEX Chemokine Panel 1 (Meso Scale Discovery; Eotaxin, Eotaxin-3, IL-8, IP-10, MCP-1, MCP-4, MDC, MIP-1α, MIP-1β, TARC), V-PLEX Proinflammatory Panel 1 kit (Meso Scale Discovery; IFNγ, IL-1β, IL-2, IL-4, IL-6, IL-8, IL-10, IL-12p70, IL-13, TNFα) and V-PLEX Plus Human IL17A kit (Meso Scale Discovery; IL17A), Human BAFF/BLyS/TNFRSF13B Quantikine ELISA kit (R&D Systems; BAFF), and Human CXCL13/BLC/BCA-1 Quantikine ELISA Kit (R&D Systems; CXCL13).

Multiplex imaging assay for macrophage characterization

Both pretreatment and post-progression incisional core biopsies were utilized for macrophage characterization. We optimized and validated a multiplexed immunofluorescence panel using CD19, CD47, CD14, CD68, CD163, CD115 (CSF1R), CD172a (SIRPα), and CD274 (PD-L1). CD19 and CD47 were selected as general B-cell lymphoma markers, CD14 and CD68 as general monocyte/macrophage markers, and CD115, CD163, CSF1R and PD-L1 were selected as markers of protumoral phenotype, based on available FL literature.^32–37 Each antibody was assessed by a uniplex immunofluorescence using the Opal 9 kit (catalogue #NEL797001KT; Akoya Biosciences, Marlborough, MA), according to the following clones and dilutions: CD19 (clone LE-CD19, Dako, 1:25 double), CD47 (clone D3O7P, CST, 1:25 double), CD14 (clone SP192, Abcam, 1:100), CD68 (clone PG-M1, Agilent, 1:50), CD163 (clone 10D6, Leica, 1:100), CSF1R (clone EPR20754, Abcam, 1:25 double), SIRPα (clone EPR22930–163, Abcam, 1:25), and PD-L1 (clone E1L3N, CST, 1:100). The slides were imaged using the Vectra Polaris spectral imaging system (Akoya Biosciences, Marlborough, MA) using the fluorescence protocol at 10 nm λ from 420 nm to 720 nm. Both germinal center and interfollicular areas from lymph nodes with reactive lymphoid hyperplasia were used as a control. Six regions of interest were selected in each case. Each marker was analyzed at single cell level and a supervised algorithm for phenotyping was built for each marker. Cell density for each marker and all possible combinations were consolidated using Spotfire software (TIBCO Spotfire). The nearest neighbor analysis was performed using R version 4.2.1

Statistical considerations

Descriptive statistics including mean, standard deviation, median, and range for continuous variables such as age, frequency counts and percentages for categorical variables such as stage and response status are provided. The best response rates and their 95% confidence intervals using the exact method are reported. Fisher’s exact test was used to evaluate the associations between best response and patient characteristics. Difference in continuous variables between patient groups was evaluated using the Wilcoxon rank sum test. Kaplan-Meier method was used to estimate OS, PFS, DOR, EFS, and TTNT. All survival outcomes were assessed from initiation of study drug, OS was defined as time to death; PFS as time to progression date or death whichever came first; DOR was defined as time from earliest response to progression date or death whichever came first; EFS as new treatment, progression date or death whichever came first and TTNT as time to next treatment. Statistical software SAS 9.4 (SAS, Cary, NC) and TIBCO Spotfire S+ 8.2 (TIBCO Software Inc., Palo Alto, CA) were used for all the analyses.

Results

We enrolled 48 patients at the University of Texas MD Anderson Cancer Center between 2016 and 2018. Patient baseline characteristics are shown in Table 1. The majority of patients were male (n=32), had FL (n=38) and advanced stage disease (n=45).

Table 1.

Patient characteristics (n=48). FLIPI denotes follicular lymphoma international prognostic index; GELF denotes Groupe d’Etude des Lymphomes Folliculaires.

Variable	Category	Frequency count	Percentage
Sex	Female	16	33.3%
	Male	32	66.7%
Race/ethnicity	Asian	4	8.3%
	African American	1	2.1%
	White	38	79.2%
	Hispanic	5	10.4%
Diagnosis	Follicular lymphoma	38	79.2%
	Grade 1/2	35	92.1%
	Grade 3a	2	5.3%
	Indeterminant	1	2.6%
	Marginal zone lymphoma	10	20.8%
	Mucosa-associated lymphoid tissue lymphoma	3	6.3%
	Nodal marginal zone lymphoma	4	8.3%
	Splenic marginal zone lymphoma	3	6.3%
Stage	II	3	6.3%
	III	12	25.0%
	IV	33	68.8%
FLIPI score	0–1	7	14.6%
	2	22	45.8%
	≥3	19	39.6%
GELF group	High	29	60.4%
	Low	19	39.6%

After a median follow up of 65.3 months (interquartile range, 63.3 – 69 months) the estimated PFS at 24 months was 78.8% (95% CI 68.0 – 91.4%) and median PFS was 70.2 months (95% CI 57.7 – not reached [NR]) in all study participants (Figure 1). Median PFS in patients with FL (n=38) was 71.2 months (95% CI: 57.7 – NR) and in MZL (n=10) was 54.2 months (95% CI: 25.7 – NR). Median event free survival and time to next treatment were similar to PFS. The median duration of response (n=46) was 71.2 months (95% CI 57.7 – NR). At last follow up only one death had occurred, which was due to COVID pneumonia in a patient who was in remission nearly 5-years after treatment. Three patients (6.3%) experienced histologic transformation to diffuse large b cell lymphoma. These events occurred at 7, 21 and 23 months from initiation of study.

Figure 1.

Progression free survival (A), event free survival (B), time to next treatment (C), overall survival (D), progression free survival in patients with follicular lymphoma (E) and marginal zone lymphoma (F).

The best ORR (CR/PR) was 95.8% (95% CI: 85.7 – 99.5%) and CRR was 83.3% (95% CI: 69.8 – 92.5%). Among the 46 evaluable patients, the CRR at 120 weeks was 63.0% (95% CI: 47.5 – 76.8%). The median time to response among the 46 responders (CR/PR) was 2.76 months (min-max: 1.25 – 14.16 months; interquartile range: 2.73 – 2.99 months). There was no statistically significant association between baseline patient characteristics, including sex, race/ethnicity, diagnosis, stage, grade, FLIPI score, GELF group and either CRR or ORR (Supplemental table 1).

Grade 1–2 adverse events (AE) occurred in 97.9% of participants, grade 3–4 AE were observed in 64.6% with no grade 5 events. Commonly reported AE by grade are shown in Table 2. Eight (16.7%%) patients discontinued therapy due to treatment related AE, 3 due to recurrent grade 3 rash, 2 due to pneumonitis (1 grade 2, 1 grade 3), 1 due to pneumonia (grade 3), 1 due to ventricular arrhythmia (grade 4) and 1 due to upper gastrointestinal bleeding (grade 3). 89.6% had one or more treatment delays and 35.4% one or more dose reductions. The most common grade ≥3 AE were rash (50.0%), neutropenia (14.6%), and diarrhea (12.5%). The most common grade 1–2 AE included myalgias (77.1%), fatigue (72.9%), and diarrhea (47.9%). One patient experienced atrial fibrillation (grade 2) during cycle 4 and was able to restart ibrutinib at 420 mg.

Table 2.

Common adverse events (AE) observed in >10% of study participants by grade.

	Grade 1	Grade 2	Grade 3	Grade 4
Non-hematologic AE	N (%)	N (%)	N (%)	N (%)
Arthralgia	11 (22.9)	2 (4.2)
Blurred vision	5 (10.4)
Chills	12 (25)	3 (6.3)
Constipation	13 (27.1)
Cough	5 (10.4)	4 (8.3)
Diarrhea	17 (35.4)	6 (12.5)	6 (12.5)
Dizziness	4 (8.3)	1 (2.1)
Dry mouth	5 (10.4)
Dry skin	9 (18.8)	1 (2.1)
Dyspnea	5 (10.4)	2 (4.2)	1 (2.1)
Edema, limbs	13 (27.1)	5 (10.4)
Fatigue	23 (47.9)	12 (25)	2 (4.2)
Fever	9 (18.8)	2 (4.2)
Headache	16 (33.3)	1 (2.1)
Mucositis	9 (18.8)	3 (6.3)
Myalgia	30 (62.5)	7 (14.6)
Nausea	13 (27.1)	3 (6.3)
Numbness	7 (14.6)
Pruritis	6 (12.5)	1 (2.1)	1 (2.1)
Rash, maculopapular	13 (27.1)	3 (6.3)	24 (50)
Vomiting	3 (6.3)	2 (4.2)
Hematologic AE
Anemia	2 (4.2)	2 (4.2)	1 (2.1)
Leukopenia			2 (4.2)
Neutropenia	2 (4.2)	1 (2.1)	3 (6.3)	4 (8.3)

A total of 20 patients had data available from exomes which achieved a coverage of 50X or more, which were combined with clinical data to calculate an m7-FLIPI score,³¹ which was not significantly associated with CRR (Figure 2). CREBBP mutations were present in 4 out of 4 patients that did not achieve CR at end of treatment and 37.5% of the patients (6 out of 16) that did achieve CR (Fisher p=0.0867). CARD11 mutations have been implicated as an escape mechanism from ibrutinib in FL,³⁸ but were not present in this dataset. CD79B, MYD88 or TNFAIP3 mutations were also not present. There was a single patient with a detectable CD79A mutation, and this patient achieved a CR.

Figure 2.

Oncoplot of m7-FLIPI and other mutations along with clinical response and toxicity data in participants with follicular lymphoma (A), proinflammatory cytokines (B), anti-inflammatory cytokines (C) and chemokines (D) as assessed on day 1 of cycles 1, 2 and 7 and at end of treatment.

Cytokines were assessed on day 1 of cycles 1, 2 and 7 and at the end of treatment (Figure 2). No difference in proinflammatory (TH1) cytokines were observed between patients who did or did not achieve a CR. The anti-inflammatory cytokines (TH2) IL2 and IL4 were numerically higher in patients who achieved a CR at weeks 2 and 7 but not at the end of treatment. Chemokines were also assessed and EOTAXIN3 was numerically higher in patients who achieved a CR. There was no association with grade 3–4 rash and cytokines (Supplemental figure 1).

Pretreatment tissue biopsies were available for 14 patients, and serial tissue biopsies were collected at time of relapse or progression in 11 patients which were utilized to assess macrophage populations. A mean tissue area of 0.14 mm² (range, 0.03–0.16) and a mean number of 28,004 cells (range, 12,974–53,435 cells) were analyzed. Not including single-marker outputs, 55 combinations were identified for macrophage markers.

A greater mean density of pre-treatment CD14+ CD163+ SIRPα+ PD-L1+ cells were observed in patients with progression within 24 months as compared to those without (1.55 vs 1.15 cells/mm², p=0.03) and a shorter median PFS was observed among patients with a higher density of pre-treatment CD68+ CSF1R+ SIRPα + cells (40 months vs NR, p=0.03; Supplemental figure 2). No significant association between macrophage subpopulations and grade 3–4 skin toxicity was observed. When comparing pre-treatment biopsies to progression samples, a greater density of CD68+ CD163+ SIRPα+ PD-L1+ cells was observed at time of progression (2.2 vs 0 cells/mm²).

Discussion

The combination of ibrutinib, lenalidomide and rituximab demonstrated significant clinical activity in this single center phase II trial of patients with untreated FL and MZL. The estimated PFS at 24 months was 78.8% and median PFS was 70.2 months. Response rates were also high with an ORR of 95.8% including 83.3% CRR. The CRR at 120 weeks was 63%. After a median follow up of approximately 5-years, only one death had occurred amongst the 48 study participants.

Compared to historical results in patients on the lenalidomide and rituximab arm of the phase III REVLEVANCE trial,²³ both the CRR and duration of response with the triplet of ibrutinib, lenalidomide and rituximab appear favorable. In patients treated with lenalidomide and rituximab in the RELEVANCE trial the ORR was 61% and CRR was 48%. CRR at 120 weeks was 48%. The 6-year PFS rate in the RELEVANCE trial was 60%,³⁹ which is similar to the PFS seen in this study. In the absence of a randomized study, it is hard to discern whether the addition of ibrutinib to the backbone of lenalidomide and rituximab resulted in a clinically meaningful impact. Given the favorable efficacy and toxicity observed with lenalidomide and rituximab in frontline FL, there is interest in exploring triplet combinations, however, as in our experience, it is challenging to generate synergism and not observe marginal additive benefit.

In regard to toxicity, grade 3–4 AE were observed in 64.6% and eight patients discontinued therapy due to treatment related AE, including 3 due to recurrent grade 3 rash. In patients randomized to lenalidomide and rituximab enrolled in the RELEVANCE trial grade 3–4 AE were observed in 32%, 7% experienced grade 3–4 rash and 2.2% discontinued therapy due to AE.²³ A phase I study of the ibrutinib, lenalidomide and rituximab triplet in untreated FL found high rates of grade 3 rash, 36%.²⁷ Similarly, a phase Ib study of the triplet for the treatment of r/r central nervous system NHL observed grade 3 rash in 20% (n=3). The dose of lenalidomide was reduced in cycle 1 of our study in an effort to reduce rates of rash, but this does not appear to have been an effective mitigating strategy.

Neither the m7FLIPI score nor individual mutations were associated with clinical response, though we note that this analysis was only performed in 20 of the study participants. Similarly, no differences in circulating cytokines or chemokines, except for EOTAXIN3 which was higher in patients with CR, were detected when comparing patients who did to patients who did not achieve a CR.

In the tumor microenvironment we found increased expression of the immune checkpoints (SIRPα and PD-L1) on infiltrating macrophages to be associated with early FL progression. Previous reports have demonstrated monocyte/macrophage related genetic signatures and high expression of SIRPα on macrophages in the tumor microenvironment of FL patients to be associated with worse survival.^40–42

In conclusion, the combination of ibrutinib, lenalidomide and rituximab is highly active as frontline therapy for patients with FL and MZL. The CRR at 120 weeks with the triplet appears favorable when compared to historical results with lenalidomide and rituximab, but with similar PFS rates and higher grade 3–4 toxicity, particularly rash. In the absence of a randomized trial, it is difficult to compare our results directly to lenalidomide and rituximab without ibrutinib. The similar clinical efficacy paired with the toxicity profile indicate this triplet is a low priority for further clinical research in patients with untreated indolent NHL. Notable was the correlative studies performed, despite the limited sample size. Future studies may further exploit the tumor microenvironment with either more potent immune modulators or CD20 targeting agents, strategies aimed at targeting macrophages, or incorporation of immune checkpoint inhibitors. Predictive biomarkers of both efficacy and toxicity are needed to inform combination studies with curative potential for these diseases and priority should be placed on inclusion of correlative studies. The number of novel therapeutic options for the treatment of indolent NHL continues to increase providing ample opportunity to explore rationale combinations. However, given the rarity of FL and limited trial participants, as a community we need to prioritize studies most likely to be practice changing.

Data availability statement:

Data will not be available.