Venetoclax and ibrutinib induces durable clinical responses in marginal zone lymphoma

Key Points

• •I+V induces high response rates, with eradication of MRD in 40% of patients with MZL.
• •These responses translate to a 56% progression-free survival estimate at 5 years and allowed for durable ETI.

Visual Abstract

Abstract

Oral ibrutinib and venetoclax (I+V) demonstrate activity as monotherapy in marginal zone lymphoma (MZL), with low complete response (CR) rates. I+V has shown good clinical activity with acceptable tolerability in other malignancies. We conducted a single-site, phase 2 trial of daily I+V in patients with MZL. Ibrutinib commenced at 560 mg daily; after 4 weeks, venetoclax commenced with weekly dose escalation to 400 mg daily. Combination therapy continued until disease progression or toxicity. The primary end point was week 16 CR. Minimal residual disease (MRD) was assessed by flow cytometry in bone marrow and peripheral blood. Patients achieving eradication of MRD could enter elective treatment interruption (ETI). Fifteen patients with MZL were treated; 14 are included in efficacy analysis, and 15 were evaluable for safety. Overall response was 79% (95% confidence interval [CI], 49-95) by F-fluorodeoxyglucose–positron-emission tomography (CR rate, 43%; 95% CI, 18-71). By computed tomography, 16-week CR rate was 29% (95% CI, 8-58). Best response within 56 weeks was 57% CR (95% CI, 37-80), which is higher than historic ibrutinib-monotherapy control (3% CR; P < .001). MRD clearance at week 56 was 40%. Six MRD-negative CR patients entered ETI, with 4 remaining disease-free at a median of 4 years. With a median follow-up of 5.5 years for the whole cohort, the 5-year progression-free survival estimate was 56% (95% CI, 27-78). I+V is safe and effective for MZL, with higher CR than with ibrutinib. Durable, ongoing CR was observed in MRD-negative patients. This trial was registered at www.clinicaltrials.gov as #NCT02471391.

Introduction

Marginal zone lymphoma (MZL) is an indolent B-cell lymphoma, comprising ∼10% of patients with low-grade lymphoma.¹ Three distinct subtypes of MZL are recognized: extranodal MZL of the mucosal-associated lymphoid tissue (50%-70% of cases), splenic MZL (20% of cases), and nodal MZL (10% of cases).^1,2 Although localized extranodal MZL treated with directed antimicrobial therapy or involved field radiation therapy results in durable remissions, the disease is incurable when advanced or in the relapsed or refractory (R/R) setting.³ In treatment-naïve patients for whom systemic therapy is required, approaches vary and include rituximab monotherapy,⁴ chlorambucil-rituximab,⁵ and bendamustine-rituximab.6, 7, 8, 9 Overall response rates (ORRs) with these regimens vary from 20% to 90%, depending on the line of therapy.4, 5, 6, 7, 8, 9 There is no established standard of care in R/R MZL, and until recently, efficacious treatment options were limited.10, 11, 12, 13, 14, 15

Attention has turned to the use of small molecule inhibitors that target intracellular pathways involved in MZL pathogenesis and cell survival. Aberrations in the nuclear factor κB signaling pathway, B-cell receptor signaling, and chronic antigenic stimulation are implicated in MZL pathogenesis.16, 17, 18 High levels of B-cell lymphoma 2 (BCL2) expression are also common in MZL.¹⁹ Ibrutinib is a covalent inhibitor of Bruton tyrosine kinase (BTK), a key component of B-cell receptor signaling.^20,21 As monotherapy at a continuous dose of 560 mg daily, a best ORR of 48% was demonstrated in a phase 2 study of patients with R/R MZL, with a complete response (CR) rate of 5% observed within 12 months and a median progression-free survival (PFS) of 15.7 months.¹⁰ Zanubrutinib has also demonstrated promising efficacy in R/R MZL, with an ORR of 68% and a 24-month PFS of 71%.²² Venetoclax is a BH3 mimetic that inhibits the antiapoptotic protein BCL2.²³ It induces apoptosis in malignant cells where survival is BCL2 dependent.²³ MZL cells are primed for apoptosis and show partial dependency on BCL2.²⁴ A phase 1 study of venetoclax monotherapy in R/R B-cell non-Hodgkin lymphoma included 3 heavily pretreated patients with MZL.¹¹ In this small subset, 2 of 3 patients achieved partial remission at doses between 600 and 1200 mg, with a PFS of 2.3 and 23.6 months.¹¹ Side effects in this study included gastrointestinal symptoms, anemia, neutropenia, and fatigue.¹¹

The combination of ibrutinib and venetoclax (I+V) has already been evaluated in patients with mantle cell lymphoma (MCL)^25,26 and chronic lymphocytic leukemia (CLL),^27,28 demonstrating substantial synergistic efficacy with acceptable safety and tolerability profiles, consistent with preclinical predictions. The single-agent activity of each drug,^10,11 together with distinct mechanisms of action that generate synergistic efficacy in other mature B-cell lymphomas, and acceptable overlapping toxicity profiles suggest I+V as an attractive combination in the treatment of MZL. Here, we investigate the efficacy and safety of the combination I+V in patients with MZL and compare the results to an ibrutinib-treated historical cohort.²⁹

Methods

Study design and patient population

This study was a single-site, open-label, phase 2, investigator-initiated trial conducted at the Peter MacCallum Cancer Centre (PMCC) and The Royal Melbourne Hospital. Adult patients were eligible to participate if they had R/R MZL or treatment-naïve MZL (because they were inappropriate candidates for treatment with cytotoxic chemotherapy). A prior histologic diagnosis of MZL per World Health Organization 2008 criteria was required. Eligible patients needed treatment based on investigator opinion and had at least 1 site of measurable disease. Eligible patients had a neutrophil count ≥0.75 × 10⁹/L and a platelet count ≥50 × 10⁹/L, with a lower acceptable threshold if the thrombocytopenia was due to bone marrow (BM) infiltration. An Eastern Cooperative Oncology Group performance status score of 0, 1, or 2 was required. Patients with clinically significant cardiovascular disease and any history of ventricular tachyarrhythmia were excluded. Full eligibility criteria are listed in the supplemental Appendix.

Treatment

As with our original protocol for patients with MCL,³⁰ all patients commenced oral ibrutinib monotherapy 560 mg daily for the first 4 weeks, aiming to reduce disease burden and risk of tumor lysis syndrome (TLS) with venetoclax, while potentially sensitizing MZL cells to venetoclax (see Figure 1). Venetoclax was added to ibrutinib from week 5, in a weekly stepwise fashion initially over 6 weeks to a target dose of 800 mg daily, based on the phase 1 study demonstrating venetoclax monotherapy activity in MZL at doses of 600 to 1200 mg daily (noting that no recommended phase 2 dose in MZL was determined).³¹ After the first 4 recruited patients attained this target dose, the protocol was amended to revise the venetoclax target dose to 400 mg daily for all patients due to a reported I+V drug interaction in a similar study (ClinicalTrials.gov identifier: NCT02910583), in which the venetoclax dose was reduced to 400 mg per day.³²

Figure 1.

Study schema. Patients commenced ibrutinib at a dose of 560 mg/d for 4 weeks. Then weekly stepwise dose escalation of venetoclax to a target 800 mg/d began. A later protocol amendment reduced the target venetoclax dose to 400 mg/d. Both study drugs were continued until disease progression or unacceptable toxicity occurred. Disease reassessment was performed at weeks 4, 16, 28, 40, and 56 (arrows) by means of CT, with PET at baseline, at weeks 16 and 56, by BM examination, and by multiparametric flow cytometry assessment of MRD in PB.

TLS risk stratification was adapted from that used in studies of venetoclax in CLL.³³ The supplemental Appendix details TLS risk stratification and protocolized monitoring and includes study amendments.

After completion of venetoclax dose escalation, patients continued I+V daily until progressive disease (PD) or unacceptable adverse events were encountered. With data emerging regarding treatment discontinuation without progression in CLL³⁴ and MCL,²⁵ a protocol amendment subsequently allowed elective treatment interruption at investigator discretion for patients with confirmed minimal residual disease (MRD)–negative CR, after a minimum of 56 weeks of treatment.

Study end points and assessments

The primary end point of the study was the CR rate at week 16, per International Working Group criteria.³⁵ The secondary end points were ORR at 16 weeks, MRD clearance, PFS, time to progression, duration of response, overall survival, and the tolerability of the combination, including the frequency and severity of adverse events. Exploratory end points were CR by International Working Group with F-fluorodeoxyglucose–positron-emission tomography (PET), evaluation of MRD-negative CR with PET, and MRD assessment by flow cytometry in peripheral blood (PB) and BM. We also evaluated the rate of elective treatment interruption and durability of sustained responses after treatment cessation.

Disease reassessment was performed at weeks 4, 16, 28, 40 and 56 by computed tomography (CT), BM examination, and MRD assessment. Concurrent PET reassessment of disease was performed at weeks 16 and 56. BM examinations were conducted until MRD was ≤1 cell in 10⁴ by flow cytometry. Response criteria for imaging assessments are provided in the supplemental Appendix. Testing for MRD was performed by 8-color flow cytometry in BM, with a minimum sensitivity of 1 cell in 10⁻³ to 10⁻⁴.³⁶ Disease assessment beyond 56 weeks was at investigator discretion, and follow-up data were collected by review of medical records.

Responses were centrally reviewed by the principal investigator and a PET physician with expertise in lymphoma. Adverse events were graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events version 4.0.³⁷ The mucosa-associated lymphoid tissue International Prognostic Index, splenic marginal zone lymphoma International Prognostic Index, and IPI were calculated as previously published.38, 39, 40

Eligibility to enter elective treatment interruption was determined by confirmation of MRD-negative CR on 2 separate occasions at least 3 months apart, as evaluated by PET and BM MRD, with a minimum sensitivity 1 malignant cell in 10⁴. After treatment cessation, patients were monitored with PB flow cytometry monthly and CT scan every 2 months until 4 months, then every 3 months until 13 months, and every 6 months thereafter. Disease recrudesce warranting reintroduction of both drugs was defined either as unequivocal detection of MRD in either PB or BM on 2 consecutive occasions at least 2 weeks apart or radiological disease progression.

Study oversight

This study protocol was developed by the authors S.M.H., A.W.R., J.F.S., and C.S.T. The study was conducted with approval from the PMCC Human Research Ethics Committee. The study adhered to the principles of the Declaration of Helsinki and the Good Clinical Practice Guidelines of the International Conference on Harmonization. All patients provided written informed consent. Overall responsibility for the conduct of this study was provided by PMCC as the sponsor. The authors confirm adherence to the study protocol and vouch for the accuracy and completeness of data.

Statistical methods

The sample size (n = 15) was pragmatic, dictated by constraints of feasible recruitment and funding. It was anticipated that if 10 patients demonstrated CR at 16 weeks, the primary end point CR rate of 67% (95% confidence interval [CI], 38-88) would justify further investigations of efficacy of I+V in MZL in a larger study. Other end points were analyzed in an exploratory fashion, including comparison to a historic control.²⁹ Statistical analyses were performed with IBM SPSS Statistics version 28.0.1.1 (14). The cutoff date for data reported herein is 31 May 2024. Patients who did not have an event by this date were censored at the time of last clinical assessment. Time-to-event analyses were performed using Kaplan-Meier methods. Response rates and survival estimates are reported with 95% CIs with binomial distribution.

Results

Patient characteristics

From June 2017 through October 2019, a total of 15 consecutive patients were enrolled in the study. One patient (patient 1) had a diagnostic revision from nodal MZL to follicular lymphoma based on an independent pathology reviewer assessment, prompted by the emergence of a discordant PET-evaluated nodal response. This patient was excluded from efficacy analysis but is included in safety analysis (see the supplemental Appendix for case history). Among the 14 patients evaluated for efficacy, 9 had R/R disease, with a median of 1 prior line of therapy (range, 1-3) at study entry (see Table 1 for the 14 efficacy-evaluable patients). Five patients had previously untreated disease. All treatment-naïve patients had significant comorbidities and were considered by investigators to be inappropriate for DNA-damaging chemotherapy. The median age of all patients was 67 years (range, 38-79). Twelve patients had stage 4 disease. In the 9 patients with R/R disease, the median time from last treatment was 24.4 months (range, 4.0-72.8); 3 had disease refractory to their most recent prior therapy.

Table 1.

Characteristics of 14 patients with MZL at study entry

Characteristic	Value
Age, median (range), y	67 (38-79)
Age >70 years, n (%)	5 (36)
Sex, n (%)
Female	6 (43)
Male	8 (57)
ECOG performance status score, n (%)
0	10 (71)
1	4 (29)
B symptoms, n (%)	1 (7)
Ann Arbor stage, n (%)
Stage IE	1 (7)
Stage III	1 (7)
Stage IV	12 (86)
BM involvement, n (%)	10 (71)
Baseline cytopenias, n (%)	5 (36)
Hemoglobin $\le$110g/L	4 (29)
Neutrophils ≤1.5 × 109/L	2 (14)
Platelets ≤100 × 109/L	2 (14)
Bulky∗ nodal and extranodal sites of disease, n (%)	6 (43)
Paraprotein, n (%)	4 (29)
MZL subtype, extranodal, n (%)	7 (50)
MALT-IPI score, n/N (%)
0	1/7 (14)
1	3/7 (43)
2	1/7 (14)
3	2/7 (29)
Splenic	5 (36)
SMZL-IPI score, n/N (%)
0	1/5 (20)
1	2/5 (40)
2	2/5 (40)
Nodal	2 (14)
IPI∗ score, n/N (%)
3	1/2 (50)
4	1/2 (50)
Previous treatment for MZL, n (%)
Yes	9 (64)
No	5 (36)
No. of previous therapies among patients who received therapy, median (range)	1 (1-3)
Time from last treatment among patients who received therapy, median (range), mo	24.4 (4.0-72.8)
Disease refractory or stable after most recent therapy, n/N (%)	3/9 (33)
Previous therapy, n/N (%)
Chemoimmunotherapy	7/9 (78)
Involved site radiation therapy	2/9 (22)

MALT-IPI, mucosa-associated lymphoid tissue International Prognostic Index; SMZL-IPI, splenic marginal zone lymphoma International Prognostic Index.

^∗Bulky denotes ≥5 cm nodal or nonnodal mass.

In the efficacy population, the median duration on study treatment excluding elective treatment interruption was 24.2 months (range, 2-79). The median follow-up was 66 months (range, 38-78). One patient did not attain the target dose of venetoclax due to ibrutinib-related grade 4 drug-induced hepatitis, which was proved to be related to the investigational product on single-agent rechallenge. As of 31 May 2024, a total of 8 patients remained on study, including 4 patients in ongoing response on study drug and 4 patients in elective treatment interruption (Figure 2). Two of 6 patients who commenced elective treatment interruption experienced progression events: 1 asymptomatic nodal progression on CT surveillance, which did not warrant re-treatment; and 1 asymptomatic cutaneous relapse, subsequently treated with local radiation. Of the other 4 efficacy-evaluable patients, 3 patients discontinued treatment due to disease progression, and 2 of these subsequently died: one due to PD with large cell transformation; and the other due to infectious complications after allogeneic stem cell transplant. One patient discontinued treatment due to toxicity (grade 4 drug-induced hepatitis).

Figure 2.

I+V treatment, elective treatment interruption, disease response with PET, and survival for 14 patients with MZL. Individual patient data are shown in lanes, ordered by sequential recruitment on study. CR was documented by PET; MRD-neg CR with PET; and PR with PET. Arrows at the end of each lane indicate ongoing treatment on study or ongoing surveillance in elective treatment interruption. ETI, elective treatment interruption; IB, ibrutinib; MRD Neg, MRD negative; VEN, venetoclax.

Efficacy

Responses, with and without PET assessments, are shown in Table 2. At week 16, the primary end point of CR rate without PET (with standard CT assessment) was 29% (95% CI, 8-58). Using PET, the ORR at week 16 was 79% (95% CI, 49-95), with a CR rate of 43% (95% CI, 18-71) and a partial response (PR) rate of 36% (95% CI, 13-65). Three patients with PR at week 16 subsequently attained CR by week 28. Overall, 10 patients (71%; 95% CI, 42-92) had a PET-evaluated CR as their best response (Table 2).

Table 2.

Response to combination therapy with I+V

Overall response	Without PET (N = 14)	95% CI	With PET (N = 14)	95% CI
Response at week 4, n (%)
CR	1 (7%)	1-29	1 (7%)	1-29
PR	5 (36%)	15-62	9 (64%)	39-85
Stable disease	8 (57%)	32-80	1 (7%)	1-29
Progressive disease	0	0-16	0	0-16
Could not be evaluated	0		3 (21)∗	6-47
Response at week 16, n (%)
CR	4 (29%)	8-58	6 (43%)	18-71
PR	6 (43%)	18-71	5 (36%)	13-65
Stable disease	2 (14%)	3-39	1 (7%)	1-29
Progressive disease	0	0-16	1 (7%)	1-29
Could not be evaluated	2 (14%)	3-39	1 (7%)	1-29
Best response, n (%)
CR	8 (57%)	29-82	10 (71%)	42-92
PR	4 (29%)	11-55	3 (21%)	6-47
Stable disease	2 (14%)	3-39	1 (7%)	1-29
Progressive disease	0	0-16	0	0-16
Could not be evaluated	0		0

The percentages may not total 100 because of rounding.

^∗Three patients were not reassessed by PET because this was not protocol mandated at that time point.

Because a previous trial of continuous ibrutinib monotherapy assessed best response rates within 56 weeks with CT, we also assessed response in the same manner for indirect comparison with this previous study.¹⁰ Within 56 weeks, the best CR rate by CT was 8 of 14 (57%; 95% CI, 29-82), which was >3 of 63 (5%; 95% CI, 1-10: P < .001) reported in the previous study of ibrutinib monotherapy. The best ORR at 56 weeks with CT was also higher in this study, at 12 of 14 (86%; 95% CI, 62-97), than in the ibrutinib monotherapy trial, which reported 30 of 63 (48%; 95% CI, 34-58; P = .017).

Initial single-agent ibrutinib for 4 weeks reduced the tumor bulk, downgrading the risk category for TLS in 3 of the 6 patients with baseline tumor bulk ≥5 cm.

Of the 14 efficacy-evaluable patients, 10 had evaluable disease in PB and BM at baseline for MRD testing. Of those evaluable, all remained MRD positive at 4 weeks. By week 16, of the 10 patients, 1 became MRD negative, whereas 4 became MRD negative by week 56. Of 6 patients attaining CR with PET evaluation at week 16, two were MRD negative. Of 8 patients attaining CR with PET evaluation at week 56, five were MRD negative.

At a median follow-up of 66 months, the median PFS had not been reached, with estimated rate of survival without either progression or death at 5 years of 56% (95% CI, 27-88; Figure 3A). In total, 6 patients had PD, with 3 occurring within the first year after attaining an initial response (PR = 2; CR = 1; Figure 2). At the time of analysis, the median overall survival had not been reached and was estimated to be 86% (95% CI, 54-96) at 5 years (Figure 3B). The median duration of response was not yet reached, with 53% (95% CI, 23-75) of responding patients estimated to be progression free at 5 years.

Figure 3.

Key survival outcomes for patients with MZL. (A-D) Kaplan-Meier plots, with 95% CIs (dashed lines), for PFS (A), overall survival (B), duration of response (C), and time to progression (D). Tick marks represent censoring at last follow-up before data cutoff for patients without an event.

Six patients entered elective treatment interruption in MRD-negative CR. This small subgroup had a median age of 68 years and had received stable doses of ibrutinib 560 mg and venetoclax 400 mg. All patients had stage 4 disease. Four patients had extranodal MZL, and 2 had splenic MZL. Four patients were treatment naïve, and 2 had been previously exposed to a single line of therapy, each with a median time from last treatment of 42 months. The median time to commencement of elective treatment interruption was 17.4 months (range, 13.0-32.0), and the median time on elective treatment interruption was 36 months (range, 6.4-63). There have been 2 progression events in this group: 1 patient had asymptomatic nodal progression detected by surveillance imaging after 6.4 months on elective treatment interruption, and 1 patient had symptomatic cutaneous progression after 23 months on elective treatment interruption. PFS for the 6 patients are 20, 47, +55, +64, +65, and +83 months, respectively.

Safety

The safety analysis was performed for all 15 patients recruited to the study. Treatment-emergent adverse events are summarized in Table 3. The most common adverse events were gastrointestinal, mainly diarrhea; grade 1 in 53% and grade 2 in 27%. No patients discontinued either drug due to diarrhea, although some required dose reductions. Neutropenia occurred in 80%; grade ≥3 in 40%. Granulocyte colony-stimulating factor use was permitted per protocol, at the clinician’s discretion. Febrile neutropenia occurred less commonly (20%).

Table 3.

Adverse events and serious adverse events

Event	Any grade (n = 15)	Grade ≥3 (n = 15)
Any adverse event	No. of patients with event (%)
Hematologic toxicity
Neutropenia	12 (80)	6 (40)
Thrombocytopenia	6 (40)	1 (7)
Anemia	3 (20)	0
Nonhematologic toxicity
Diarrhea	12 (80)	0
Bruising	11 (73)	0
Nausea	7 (47)	0
Gastroesophageal reflux	7 (47)	0
Fatigue	6 (40)	0
Upper respiratory tract infection	6 (40)	1 (7)
Skin/soft tissue infection	5 (33)	0
Urinary tract infection	4 (27)	0
AF	4 (27)	1 (7)
Hypertension	4 (27)	3 (20)
Rash	4 (27)	0
Arthritis	3 (20)	0
Cramps	3 (20)	0
Sensory peripheral neuropathy	3 (20)	0
Constipation	3 (20)	0
Cough	3 (20)	0
Epistaxis	3 (20)	0
Oral mucositis	3 (20)	0
Noncardiac chest pain	3 (20)	0
Brittle nails	3 (20)	0
Any serious adverse event
G3 lower respiratory tract infection	3 (20)	—
G3 febrile neutropenia	3 (20)	—
G3 skin/soft tissue infection	2 (13)	—
G2 seizure	2 (13)	—
G3 febrile neutropenia	2 (13)	—
G3 upper respiratory tract infection	1 (7)	—
G3 diarrhea	1 (7)	—
G3 small bowel obstruction	1 (7)	—
G3 acute cholecystitis	1 (7)	—
G3 second-degree AV block	1 (7)	—
G4 drug-induced hepatitis	1 (7)	—

AV, atrio-ventricular; G3, grade 3.

Other common adverse events included bruising (73%), nausea (47%), gastroesophageal reflux (47%), fatigue (40%), upper respiratory tract infection (40%), thrombocytopenia (40%), and skin/soft tissue infection (33%).

Atrial fibrillation (AF) and hypertension were observed in 4 patients (27%) each. This rate of AF was higher than the 8% rate of AF seen in a prior ibrutinib continuous monotherapy study for R/R MZL.¹⁰ The increased rate of AF may be reflective of a higher cardiovascular risk cohort, in which 10 patients had adequately controlled cardiovascular comorbidities at enrollment, including 5 in frailer treatment-naïve group considered inappropriate for frontline treatment with chemoimmunotherapy.

Serious adverse events occurred in 8 patients (53%). Infections caused most of these events, and none were grade 4 or 5. One patient developed an invasive pulmonary Aspergillus infection. Drug-induced hepatitis, confirmed on liver biopsy and rechallenge with single-agent ibrutinib, was noted in 1 patient, necessitating withdrawal from the study (Figure 2 patient 5; refer to the supplemental Data for further information). Liver function abnormalities resolved fully over time after treatment withdrawal. A seizure event was seen in 1 patient with a history of epilepsy and another patient with a sleep deprivation trigger on the background of prior neurosurgery; neither was considered related to the study drugs, and treatment was continued. Neither laboratory nor clinical TLS was observed.

Within the first 56 weeks of trial treatment, 13 of 15 patients remained on venetoclax 400 mg, 1 patient was on 200 mg, and 1 had discontinued venetoclax because of grade 4 drug-induced hepatitis caused by ibrutinib. Beyond 56 weeks, venetoclax dose intensity was maintained in most patients, with 10 of 13 stable on 400 mg daily, 1 on 300 mg daily, and 1 on 200 mg daily. The remaining patient discontinued venetoclax treatment because of persistent grade 3 neutropenia. Ibrutinib required more frequent dose modifications. Of the 15 patients during the first 56 weeks, 10 remained on 560 mg daily of ibrutinib, 4 were on 420 mg daily, and 1 had discontinued ibrutinib due to grade 4 drug-induced hepatitis. Beyond 56 weeks, of the 13 patients on treatment, 5 remained on 560-mg ibrutinib, 7 on 420 mg, and 1 additional patient discontinued treatment due to persistent grade 2 pruritus.

Discussion

Although this study holds promise of the safety and efficacy of I+V, it is limited by the small cohort evaluated. The cohort was heterogeneous for MZL subtype, treatment-naïve and R/R status, and venetoclax target dosing. Therefore, the results of this study are to be interpreted with caution.

This study is, to our knowledge, the first description of I+V for the treatment of patients with MZL, demonstrating a complete response rate of 29% (95% CI, 8-58) by CT assessment and 43% (95% CI, 18-71) by PET at week 16. Similar to prior studies of this combination in MCL^25,26 and CLL,^27,28 the synergistic activity of the combination likely underpins the improvements in efficacy shown in our study compared to ibrutinib monotherapy. Best responses occurred after week 16 and typically within 1 year on treatment, likely due to the indolent growth kinetics of MZL.

The best response within 56 weeks demonstrated a CR of 57%, with an ORR of 86%, and attainment of MRD negativity in 40% of evaluable patients. Evaluating best response at this time point allowed for the clearest comparison with the previous phase 2 study of ibrutinib in R/R MZL, in which CR as best response within 56 weeks was 5% and best ORR at this time point was 48%.¹⁰ By comparison, the combination I+V had higher rates of attainment of both CR and ORR. Comparing our results to a treatment-naïve cohort, noting different patient populations and disease characteristics, the I+V combination appears more durable, with PFS estimated at 5 years of 56%, compared to a median PFS of 15.7 months in patients treated with ibrutinib monotherapy.¹⁰

The combination treatment was generally tolerable; however, frequent dose reductions of ibrutinib beyond 56 weeks were required. The safety profile of the combination was as expected when evaluating the toxicity profiles of these drugs as single agents and comparable with the side effect profiles of the combination used in other diseases.^10,11 Gastrointestinal toxicity and neutropenia were the most common toxicities. Febrile neutropenia (20%), similar to other venetoclax studies, was much lower than rates of neutropenia. AF (27%) and hypertension (27%) were common. This was higher than the 8% AF rate in the prior continuous ibrutinib study in R/R MZL¹⁰ and the 2.9% AF rate in the continuous zanubrutinib study in R/R MZL.²² Although concerning, it is difficult to determine whether this a reflection of a study with a small denominator or a true indicator of a higher cardiovascular risk group exposed to ibrutinib. Evaluation of combinations that use a second-generation covalent BTK inhibitor, in place of ibrutinib, with a BCL2 inhibitor are likely to improve the cardiovascular risk of this treatment approach. TLS, major hemorrhage, ventricular arrhythmias, and cardiac failure were not seen.

Time-limited combination I+V has been evaluated in CLL^27,28 and MCL.^25,26 As an exploratory objective in this study, time-limited therapy was evaluated, and 6 patients entered elective treatment interruption after attaining MRD-negative CR. Four of these patients remain in ongoing deep response with a median 36 months off study treatment, establishing the proof of principle of the feasibility of a time-limited novel agent treatment program in MZL.

The treatment landscape of MZL spans localized treatments, rituximab monotherapy, chemoimmunotherapy, targeted agents, and immunotherapy including chimeric antigen receptor T cells.¹⁵ I+V, as a combination targeted therapy, may challenge the place of continuous BTK inhibitor monotherapy in the R/R setting, with deeper responses that are durable and the potential for time-limited treatment. It may also be a suitable option for patients requiring systemic treatment but who are unfit for chemoimmunotherapy or chimeric antigen receptor T-cell therapy. In this study, discontinuation rates due to tolerability and particularly toxicities associated with ibrutinib are problematic. Future studies evaluating second-generation BTK inhibitors with BCL2 inhibitors may overcome toxicity and tolerability concerns.

This study demonstrates the safety and efficacy of the combination I+V in both R/R and treatment-naïve MZL. The CR rate and PFS are highly encouraging when considered against historical ibrutinib monotherapy, confirming combinatorial benefit. Moreover, there is evidence that time-limited treatment is feasible and can afford excellent ongoing disease control. A larger study, including a time-limited approach in its design, is warranted.

Conflict-of-interest disclosure: S.M.H. reports honoraria from BeiGene, AstraZeneca, and Janssen; and advisory board fees from BeiGene. P.A.T. reports honoraria from AbbVie, Ascentage, AstraZeneca, BeiGene, Genentech, Genmab, Janssen, Lilly, and Merck; travel support from AbbVie, Genmab, Merck, and Roche; and speaker fees from Adaptive Biotechnologies. D.R. reports honoraria from Bristol Myers Squibb (BMS), Takeda, Novartis, Amgen, and Merck Sharp & Dohme; and research funding from BMS, Takeda, Novartis, and Amgen. A.W.R. is an inventor on a patent related to venetoclax; is an employee of Walter and Eliza Hall Institute, which received milestone and royalty income related to venetoclax that is shared with employees; and reports research funding for this research from AbbVie and Janssen and for past research from BeiGene. J.F.S. reports consultancy fees from Genor Biopharma and TG Therapeutics; research funding from AbbVie, BMS, and Roche; speaker’s bureau fees from AbbVie, Roche, and AstraZeneca; and honoraria from AbbVie, Roche, AstraZeneca, BeiGene, and Janssen. C.S.T. reports honoraria from AstraZeneca, BeiGene, Gilead, AbbVie, and Janssen; and research funding from AbbVie and Janssen. M.A.A. reports honoraria from Roche, Takeda, Novartis, Gilead, AstraZeneca, Janssen, and AbbVie; and is a current employee of the Walter and Eliza Hall Institute, which received milestone and royalty payments related to venetoclax that are shared with employees. The remaining authors declare no competing financial interests.