CADTH Reimbursement Review

Zanubrutinib (Brukinsa)

Sponsor: BeiGene Canada ULC

Therapeutic area: Waldenström macroglobulinemia

This multi-part report includes:

Clinical Review

Pharmacoeconomic Review

Clinical Review

Executive Summary

An overview of the submission details for the drug under review is provided in Table 1.

Introduction

Waldenström macroglobulinemia (WM) is a rare, low-grade lymphoplasmacytic lymphoma characterized by the presence of clonal cells that secrete immunoglobulin M (IgM) in the bone marrow and other organs. Although described as indolent, WM can become a serious condition; features and symptoms include cytopenias, hyperviscosity, peripheral neuropathy, hemolytic anemia, hepatomegaly, splenomegaly, organomegaly, fatigue, weight loss, and recurrent fever and night sweats.¹ More than 90% of patients with WM have an activating mutation in myeloid differentiation factor 88 (MYD88^L265P), and approximately 30% have mutations in the chemokine receptor 4 (CXCR4) gene.² In Canada, the incidence of WM is 1 in 200,000 people per year.³ Currently, the diagnosis of WM is based on clinicopathological criteria, including bone marrow involvement by lymphoplasmacytic lymphoma cells, a serum IgM monoclonal paraprotein, and the presence of MYD88^L265P mutation.⁴ Once the diagnosis is established, the relationship between the patient’s symptoms and WM is confirmed, because therapy is generally reserved for symptomatic patients. Bone marrow involvement and serum levels of IgM, albumin, and beta2 microglobulin may be used to estimate the time until treatment initiation.⁴

Most patients presenting with symptomatic disease require treatment. The most important goals of therapy are to relieve lymphoma and paraprotein-related symptoms and delay disease progression by achieving prolonged remission. In patients who present with life-threatening complications related to hyperviscosity or cryoglobulinemia, plasmapheresis is used as a temporary measure until definitive treatment is initiated. The standard approach for first-line treatment in Canada is chemoimmunotherapy, most commonly bendamustine-rituximab (BR), followed by maintenance rituximab. Other regimens, including dexamethasone-rituximab-cyclophosphamide (DRC), bortezomib-rituximab, rituximab monotherapy, and chlorambucil monotherapy, are used for patients unable to tolerate BR. There is no standard of care for the treatment of relapsed/refractory (R/R) WM. Bortezomib-based chemotherapy is the most commonly used regimen (e.g., cyclophosphamide-bortezomib-dexamethasone [CyBorD]). Bruton tyrosine kinase (BTK) inhibitors (e.g., ibrutinib, acalabrutinib, zanubrutinib) are available only through compassionate access programs, and are currently most used in the R/R setting after failure of chemoimmunotherapy. However, none of these treatments is curative, and all patients are expected to relapse and require additional treatment.

Zanubrutinib (Brukinsa, 80 mg oral capsules) is a second-generation BTK inhibitor indicated for the treatment of adult patients with WM. It received a Notice of Compliance from Health Canada on March 1, 2021.³

The objective of this review was to evaluate the efficacy and safety of zanubrutinib 80 mg oral capsules for the treatment of adult patients with WM.

Stakeholder Perspectives

The information in this section is a summary of input provided by the patient groups who responded to CADTH’s call for patient input and from clinical experts consulted by CADTH for the purpose of this review.

Patient Input

Four patient groups provided input for the review of zanubrutinib in WM: the CanCertainty Coalition, Lymphoma Canada (LC) in collaboration with the Canadian Organization for Rare Disorders (CORD), and the WM Foundation of Canada (WMFC). The CanCertainty Coalition data were sourced through literature, Canadian prescription drug insurance coverage, population demographics, and previously conducted surveys. The CanCertainty data collection and submission were completed using CanCertainty resources and personnel and contract personnel exclusively. LC, CORD, and WMFC conducted anonymous online surveys of patients with WM between February 28, 2021 and May 10, 2021 registered through their respective databases and through social media outlets.

Symptoms of WM that most affected patients’ health-related quality of life (HRQoL) at diagnosis included fatigue (66%), night sweats (28%), neuropathy (24%), weight loss or loss of appetite (20%), and easy bruising or bleeding (20%). A total of 81% of respondents experienced at least 1 psychological and social impact of a WM diagnosis, including stress or anxiety (66%), difficulty sleeping (30%), impact on daily activities (28%), memory loss or concentration problems (19%), and depression (19%). In terms of treatment, 17% of patient respondents were receiving first-line treatment, 41% were in remission following a previous line of treatment, and 6% had relapsed following previous treatment and were waiting to begin another treatment. The most common treatments patients had received included chemotherapy monotherapy (55%), monoclonal antibodies (63%), and BTK inhibitors (36%). The most common side effects experienced by patients during treatment for WM included fatigue (72%), neutropenia (47%), nausea (39%), anemia (37%), peripheral neuropathy (37%), thrombocytopenia (30%), rash or itch (26%), back or joint pain (23%), mouth sores (22%), diarrhea (20%), headache (19%), and hair loss (17%). Patients noted that fatigue was particularly difficult to handle. Having a choice of treatment and enough treatment options were considered particularly important to patients. In terms of treatment outcomes, patients rated longer survival (75%), longer remission (76%), better HRQoL (70%), and fewer side effects (57%) as the most important.

Clinician Input

Input From Clinical Experts Consulted by CADTH

The clinical experts consulted by CADTH indicated that zanubrutinib would be used in the R/R setting after failure of standard chemoimmunotherapy because they expect it to be more efficacious (leading to more prolonged remission) and less toxic than a repeated round of chemoimmunotherapy. The clinical experts indicated that they would generally not consider zanubrutinib in the first-line treatment setting. All patients should be offered chemoimmunotherapy first, unless they are truly unfit for anything other than rituximab therapy or even oral chlorambucil. These patients have a defined treatment interval and can enjoy a prolonged remission after chemoimmunotherapy (with or without rituximab); as such, reserving zanubrutinib for later lines does not result in reduced survival. Zanubrutinib should be offered only to patients who have failed at least 1 line of therapy. The clinical experts consulted by CADTH also indicated that patients with asymptomatic disease should not be treated with zanubrutinib unless there is concern about impending hyperviscosity syndrome. Patients who are at very high risk for bleeding complications (e.g., those who cannot tolerate antiplatelet or anticoagulation equivalent) would be least suitable for treatment with zanubrutinib.

One clinical expert commented that WM is truly an orphan disease. Compared to other indolent lymphomas (e.g., follicular lymphoma), it affects a rare group of patients with unique clinical manifestations that do not respond as well to chemoimmunotherapy. There are few effective treatment options available at relapse and few or no new therapies available through clinical trials. Consequently, access to BTK inhibitors is imperative for this group of patients.

Clinician Group Input

Joint input was received from 2 registered clinicians on behalf of the Ontario Health (Cancer Care Ontario) Hematology Cancer Drug Advisory Committee for the review of zanubrutinib for the treatment of WM.

The clinicians stated that most patients demonstrate a good response to first-line BR and remain free of relapse for a few years. Contrary to the clinical experts consulted by CADTH, the clinician group advised that zanubrutinib may be used in the first-line setting or after relapse, given that there is currently no evidence to suggest the specific sequencing of treatment with zanubrutinib. However, patients with relapsed disease have a significant unmet need for additional treatment options, including BTK inhibitors. The clinicians indicated that the patients best suited to this treatment are those with symptomatic R/R WM.

Drug Program Input

The drug plans noted that in the ASPEN trial, zanubrutinib was compared to ibrutinib, which is not publicly funded in any jurisdiction in Canada. Ibrutinib, for the treatment of patients with WM who have received at least 1 prior therapy, was previously reviewed by CADTH and not recommended for reimbursement, Ibrutinib may be available for some patients (at no charge) through the sponsor’s patient support program. Relevant comparators for WM in Canadian jurisdictions include rituximab-based chemotherapy (BR, bortezomib-dexamethasone-rituximab, and DRC) for treatment-naive patients and those with R/R WM. Re-treatment with rituximab is funded for patients with a relapse-free interval (6 months to 12 months, depending on the jurisdiction) following the last dose of rituximab. In terms of prescribing considerations, the drug plans noted that zanubrutinib has the potential for drug-drug interactions, possibly increasing pharmacy resource use. However, in terms of care provision, the capsule strength of 80 mg (in bottles of 120 capsules) facilitates dispensing and dose adjustment without wastage. The drug plans had questions about patient eligibility criteria for treatment with zanubrutinib that were answered by the clinical experts consulted by CADTH.

Clinical Evidence

Pivotal Studies and Protocol Selected Studies

Description of Studies

The ASPEN trial is an ongoing, phase III, randomized, open-label, multi-centre study designed to compare the efficacy and safety of zanubrutinib and ibrutinib in patients with WM who required therapy. Between January 2017 and July 2018, 164 R/R and 37 unfit, treatment-naive patients with WM were recruited into cohort 1 (patients with MYD88 mutation) and randomized 1:1 to receive either ibrutinib (420 mg) or zanubrutinib (160 mg) in 28-day cycles. Cohort 2 was a non-randomized, no-comparator arm that included 28 patients with wild-type or unknown MYD88 mutation status, including 23 R/R and 5 unfit, treatment-naive patients, all of whom received zanubrutinib (160 mg). The primary efficacy end point was the proportion of patients in each arm of cohort 1 who achieved either complete response (CR) or very good partial response (VGPR), as determined by an Independent Review Committee (IRC) using an adaptation of the response criteria updated at the Sixth International Workshop on Waldenström’s Macroglobulinemia (IWWM).^5,6 Other end points included duration of response (DoR), progression-free survival (PFS), improvement in cancer-related symptoms, overall survival (OS), HRQoL, and medical resource utilization.

The most common indications (> 20%) for therapy initiation in cohort 1 were fatigue (57.2%), anemia (43.8%), B symptoms (systemic symptoms of fever, night sweats, and weight loss [30.3%]), hyperviscosity (26.9%), and peripheral neuropathy (22.4%). The median age of all patients was 70.0 years. The majority of patients were male (66.7%) and White (91.0%). In cohort 2, the median age was 72 years; 50% of patients were male and 96.4% were White. The most common indications for therapy initiation were fatigue (60.7%), B symptoms (35.7%), anemia (32.1%), hyperviscosity (21.4%), and peripheral neuropathy (10.7%).

A summary of the key results from the ASPEN trial is available in Table 2.

Efficacy Results

Cohort 1 – MYD88 ^L265P

The median follow-up time was 19.4 months in cohort 1. Nine patients in the ibrutinib arm and 6 patients in the zanubrutinib arm started non-protocol anticancer therapy. The median times to initiation of non-protocol anticancer therapy were 6.44 months in the ibrutinib treatment arm and 6.83 months in the zanubrutinib treatment arm. The median PFS had not been reached in either treatment arm. The event-free rates at 12 months for patients in the ibrutinib and zanubrutinib treatment arms were 87.2% (95% confidence interval [CI], 78.6% to 92.5%) versus 89.7% (95% CI, 81.7% to 94.3%), respectively, and 83.8% (95% CI, 74.5% to 89.9%) versus 85.0% (95% CI, 75.2% to 91.2%) at 18 months. In cohort 1, the median OS was not reached in either treatment arm. At the data cut-off date (August 31, 2019), 8 deaths occurred in the ibrutinib arm, and 6 deaths occurred in the zanubrutinib arm. The event-free rates for patients in the ibrutinib versus zanubrutinib treatment arms were 93.9% (95% CI, 86.8%, 97.2%) versus 97.0% (95% CI, 90.9% to 99.0%) at 12 months.

In cohort 1, the IRC-assessed CR or VGPR rates in the ibrutinib and zanubrutinib arm were 19.2% (95% CI, 12.0% to 28.3%) and 28.4% (95% CI, 19.9% to 38.2%), respectively. In R/R patients, the IRC-assessed CR or VGPR rates were 19.8% (95% CI, 11.7% to 30.1%) in the ibrutinib arm and 28.9% (95% CI, 19.5% to 39.9%) in the zanubrutinib arm (P = 0.11). In unfit, treatment-naive patients, the IRC-assessed CR or VGPR rates were 16.7% (95% CI, 3.6% to 41.4%) in the ibrutinib arm and 26.3% (95% CI, 9.1% to 51.2%) in the zanubrutinib arm. On average, in cohort 1, HRQoL (an exploratory end point) increased numerically during the trial observation period in both treatment arms. The least squares (LS) means for the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ-C30) global health status/quality of life (QoL) were 69.0 (standard error = 2.3) in the ibrutinib arm and 68.3 (standard error = 2.2) in the zanubrutinib arm, a difference of –0.69 (95% CI, –4.95 to 3.57). The mean changes in EuroQol 5-Dimensions (EQ-5D) scores from baseline were 9.0 (standard deviation [SD] = 17.90) in the ibrutinib arm and 13.7 (SD = 14.66) in the zanubrutinib arm at cycle 13, day 1.

Cohort 2 – MYD88^WT

The median follow-up time was 17.8 months in cohort 2. Three patients (1 unfit, treatment-native patient and 2 R/R patients) started non-protocol anticancer therapy with a median time to initiation of 3.61 months. In cohort 2, no patients achieved CR. The IRC-assessed CR or VGPR rate was 26.9% (95% CI, 11.6% to 47.8%).

Harms Results

In cohort 1, 97 ibrutinib-treated patients (99.0%) and 98 zanubrutinib-treated patients (97.0%) had at least 1 adverse event (AE); AEs of grade 3 or greater were reported in 62 patients (63.3%) and 59 patients (58.4%) in the ibrutinib and zanubrutinib treatment arms, respectively. Serious AEs (SAEs) were reported in 40 patients (40.8%) and 40 patients (39.6%) in the ibrutinib and zanubrutinib treatment arms, respectively. The most common SAE in the ibrutinib treatment arm was pneumonia (9 patients [9.2%]), followed by pyrexia and sepsis (3 patients [3.1%] each). The most common SAEs in the zanubrutinib treatment arm were febrile neutropenia, influenza, and neutropenia (3 patients [3.0%] each). Nine patients (9.2%) in the ibrutinib arm and 4 patients (4.0%) in the zanubrutinib treatment arm had AEs leading to study treatment discontinuation. A total of 7 patients (7.1%) in the ibrutinib treatment arm and 6 patients (5.9%) in the zanubrutinib treatment arm had died by the time of the data cut-off date; 5 patients (5.1%) in the ibrutinib arm and 1 patient (1.0%) in the zanubrutinib arm died within 30 days of the last dose of study drug.

Notable AEs included neutropenia, hemorrhage (minor and major bleeding), cardiovascular events, and second primary malignancy. In cohort 1, neutropenia was reported in 12 patients (12.2%) in the ibrutinib and 25 patients (24.8%) in the zanubrutinib arm. However, the higher incidence of neutropenia among zanubrutinib-treated patients did not translate to an increased occurrence of infections in the zanubrutinib arm. Fifty-eight patients (59.2%) in the ibrutinib arm and 49 patients (48.5%) in the zanubrutinib arm had hemorrhage (including minor bleeds involving mucous membranes and skin). Major hemorrhage was observed in 9 patients (9.2%) in the ibrutinib arm and 6 patients (5.9%) in the zanubrutinib arm. Atrial fibrillation or flutter was reported in 14 patients (14.3%) in the ibrutinib arm and 2 patients (2.0%) in the zanubrutinib treatment arm. Second primary malignancy was reported in 11 patients (11.2%) in the ibrutinib arm and 12 patients (11.9%) in the zanubrutinib arm.

Critical Appraisal

The ASPEN trial was an open-label study. Therefore, important sources of bias from lack of blinding of patients and investigators to study treatments exist; patients’ knowledge of their treatment may have affected some safety end points; and different supportive care may have been offered to patients in the 2 treatment arms. The primary end point and key secondary end points were appropriate and adequately described. Data were immature for time-to-event outcomes, and median PFS and OS were not reached in either treatment arm. Given that the ASPEN trial is ongoing, future analyses may be more informative with respect to time-to-event outcomes. In addition to PFS and OS, time to next treatment was identified in the systematic review protocol as an important efficacy outcome; however, this was an exploratory outcome that limits the interpretation of results. Some other important outcomes, including OS and HRQoL, were also exploratory in the trial. Of note, the only outcome defined in the statistical testing hierarchy, CR or VGPR rate in the R/R patient population of cohort 1, did not reach statistical significance.

Ibrutinib is not the most relevant comparator for zanubrutinib in Canadian clinical practice. The most relevant comparators for WM in Canadian jurisdictions include rituximab-based chemotherapy for treatment-naive patients and those with relapsed disease. Therefore, relevance to the current clinical setting is limited, and the question of the comparative efficacy and safety of zanubrutinib to current standard of care in Canada cannot be answered. The inclusion criteria for the ASPEN study were generally reasonable, based on the intended patient population. However, the exclusion of patients with central nervous system (CNS) involvement in the ASPEN trial — while justified when the trial was designed, due to a lack of disease management guidelines — was not considered appropriate because these patients (i.e., patients with Bing Neel disease) may benefit from early BTK inhibitor treatment. The trial considered patients to be treatment-naive if they were unsuitable for chemoimmunotherapy due to age or the presence of comorbidities. This definition does not align with the standard definition of treatment-naive in oncology practice, which refers to patients who have not received prior anticancer therapy. Therefore, the trial evidence regarding the efficacy and safety of zanubrutinib compared to ibrutinib in truly treatment-naive patients is insufficient to guide treatment decisions in this patient population in clinical practice.

Indirect Comparisons

Description of Studies

The sponsor-submitted indirect treatment comparison (ITC), which was used to inform the pharmacoeconomic model, was appraised and summarized. A matching-adjusted indirect comparison (MAIC) was conducted based on a systematic literature review that compared the individual patient-level data (IPD) of the zanubrutinib arm of the ASPEN trial to match the populations of relevant trial reports for chemotherapy regimens in adult patients with treatment-naive or relapsed-refractory WM. The analysis was informed by a systematic literature review that identified 33 trials, mainly retrospective, that were subsequently excluded from the ITC. In total, 3 trials were included in the MAIC; 1 which included R/R WM patients, 1 that included treatment-naïve WM patients, and 1 that included a mixed R/R, and treatment-naive WM population. The interventions included zanubrutinib, BR, and DRC; however, DRC was used in the treatment-naive population, and BR was used in the R/R population. Three sets of pairwise MAICs were conducted. Two pairwise comparisons matched the overall zanubrutinib population (N = 102) to the BR (N = 71) and DRC (N = 72) populations separately. A subgroup analysis was conducted matching zanubrutinib patients with R/R disease to the BR population. No MAIC was conducted specifically to compare the unfit, treatment-naive subpopulation in ASPEN, given the small sample size of the unfit, treatment-naive patient population in the zanubrutinib arm of the ASPEN trial (n = 19). Several of the preidentified variables, including Eastern Cooperative Oncology Group Performance Status (ECOG PS), beta2 microglobulin concentration, and MYD88/CXCR4 mutation status, were not accounted for during weighting due to the limitations of available data. In the MAIC comparing zanubrutinib to BR, the variables included in the weighting process included age, prior lines of therapy, IgM concentration, International Prognostic Scoring System for Waldenström Macroglobulinemia (IPSSWM) score, and presence of extramedullary disease. In the MAIC comparing zanubrutinib to DRC, the variables included in the weighting were age, platelet count, hemoglobin count, and presence of extramedullary disease.

Efficacy Results

After weighting, the results of the MAIC comparing zanubrutinib to BR suggest that zanubrutinib is favoured over BR, including in the R/R subgroup for PFS and OS; however, the results lacked precision, showing wide 95% CIs. Zanubrutinib was associated with significantly longer PFS (hazard ratio [HR] = 0.37; 95% CI, 0.15 to 0.91) after weighting compared to BR. Compared to DRC, zanubrutinib was associated with significantly longer PFS (HR = 0.35; 95% CI, 0.14 to 0.86) after weighting. The HR for OS comparing zanubrutinib to BR indicated a statistically significantly longer OS in the overall population after weighting (HR = 0.29; 95% CI, 0.10 to 0.85]). ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||.

Harms Results

No indirect evidence was available for the safety or impact on HRQoL of zanubrutinib compared to relevant chemotherapy regimens.

Critical Appraisal

The ITC was informed by an appropriately conducted systematic review of the literature highlighting the relevant population and outcomes of interest for this review. Screening was conducted based on standard methods, with studies selected independently in duplicate, according to pre-specific criteria. No formal quality assessment of the included studies was conducted, which is an important limitation. The sponsor-submitted MAIC assumes that all effect modifiers and prognostic factors are accounted for in the model. A comprehensive list of prognostic factors and treatment-effect modifiers identified through appropriate channels was included in the report and — based on discussions with the clinical experts consulted by CADTH — these factors and modifiers were considered relevant; however, some of the factors, including ECOG PS, B2 microglobulin, and MYD88/CXCR4 mutation status, were not accounted for in the calculation of weight. This may result in bias because not all prognostic factors and effect modifiers that were originally identified were accounted for in the weights. Additionally, there were discrepancies between the cut-offs of identified variables and those available for weighting, potentially biasing the results further. In terms of external validity, the studies selected for indirect comparison included treatment with DRC in the treatment-naive population and with BR in the R/R population. In discussion with the clinical experts consulted by CADTH, the comparison to DRC in the treatment-naive, first-line population was considered irrelevant because it does not reflect clinical practice in Canada. No studies were identified in the systematic literature review (SLR) reporting results for BR in the treatment-naive population, which is the standard of care in Canada; thus, these were not included in the analysis for treatment-naive patients. Moreover, no studies were included in the treatment-naive population for patients for whom chemoimmunotherapy was considered unsuitable.

Conclusions

Based on clinical evidence from the ASPEN trial, the relative efficacy of zanubrutinib for the treatment of unfit, treatment-naive patients with R/R WM did not surpass that of the comparator, ibrutinib, another BTK inhibitor for the outcome of CR or VGPR in patients with R/R WM. The safety profiles of zanubrutinib and ibrutinib were similar in terms of occurrence of overall AEs and SAEs. Notable differences in toxicity between the 2 treatments included a higher incidence of atrial fibrillation in the ibrutinib arm and a higher incidence of neutropenia in the zanubrutinib arm. Ibrutinib is not publicly funded in Canada; it is currently only available for patients with WM through compassionate access programs. Given the lack of head-to-head studies evaluating zanubrutinib versus the most relevant comparators in Canada — and the important methodological limitations of the sponsor-submitted ITC — no conclusions could be drawn regarding the efficacy and safety of zanubrutinib compared with currently used chemoimmunotherapy regimens in patients with WM who are treatment-naive or R/R.

Based on input from clinicians consulted by CADTH, zanubrutinib is not expected to replace current standard of care first-line chemoimmunotherapy treatment regimens. The clinical experts indicated that all patients with WM will likely relapse after front-line chemoimmunotherapy. The results of re-treatment with chemoimmunotherapy for R/R disease are less optimal when compared to other indolent lymphomas; therefore, there is an unmet need for additional treatment options that prolong remission in patients with R/R WM. Given that patients become immunosuppressed with initial therapy, additional treatment options that minimize toxicity are desirable during relapse. The clinicians indicated that, based on clinical experience with BTK inhibitors, zanubrutinib may be more tolerable than the chemoimmunotherapy treatments currently used to treat patients with R/R WM.

Introduction

Disease Background

WM is a rare, low-grade, lymphoplasmacytic lymphoma characterized by the presence of IgM-secreting clonal cells in the bone marrow and other organs. Many patients who fulfill the criteria for a diagnosis are asymptomatic. Almost all patients diagnosed with WM have a preceding phase of IgM monoclonal gammopathy of undetermined significance (MGUS), but the clonal MGUS B-cells already contain the molecular signature of a malignant clone.⁸ Phenotypically, the lymphoplasmacytic cells of WM typically arise from CD25⁺, CD22^low, activated B lymphocytes, and express pan B-cell markers, CD19 and CD20.⁸ Although described as indolent, WM can become a serious, life-threatening disease, causing significant morbidity in the elderly. Morbidity and mortality in WM are associated with excess serum IgM rather than tumour infiltration, contrary to other lymphomas. The clinical manifestations of WM related to the overproduction of IgM include cytopenias, hyperviscosity, peripheral neuropathy, hemolytic anemia, hepatomegaly, splenomegaly, and organomegaly. Accompanying symptoms include fatigue, weight loss, recurrent fever, and night sweats.¹ Approximately 1 in 4 patients with WM have a family history of lymphoproliferative disorders, with first-degree relatives having a 20-fold higher risk of developing WM compared to the general population.^2,9 More than 90% of patients with WM have an activating mutation in the MYD88 gene (MYD88^L265P). Mutations in the CXCR4 gene are also common; these are observed in approximately 30% of cases.² Both of these mutations have prognostic significance and may be associated with clinical outcomes and response to targeted therapies.¹

The overall age-adjusted incidence of WM is 3.8 per million persons per year, with incidence increasing with age. The incidence of WM is twice as high in men as it is in women (5.4 million versus 2.7 per million, respectively). In Canada, the incidence of WM is 1 in 200,000 people per year.³ The 5- and 10- year PFS for IgM MGUS to WM is 90% and 81%, respectively.⁸ Median OS has been improving; from 1991 to 2000 and 2001 to 2010, median OS in patients diagnosed with WM in the US improved to 8 years from 6 years, respectively.¹⁰ Because patients with WM have an indolent disease course and are often of an advanced age, some patients ultimately succumb to other diseases of the elderly not related to WM; up to 40% of patients over 75 years of age with WM do not die of WM.¹¹ However, compared to the general population, patients with WM have a greater overall risk of second malignancies, including large cell lymphoma, myelodysplasia, and brain cancer.^8,12

Currently, the diagnosis of WM is based on clinicopathological criteria, including bone marrow involvement by lymphoplasmacytic lymphoma cells, a serum IgM monoclonal paraprotein, and the presence of MYD88^L265P mutation.⁴ Once the diagnosis is established, it is important to investigate the relationship between the patient’s symptoms and WM, because therapy should be reserved for symptomatic patients. Bone marrow involvement and serum levels of IgM, albumin, and beta2 microglobulin may be used to estimate the time until treatment initiation.⁴

Standards of Therapy

The treatment of a patient with WM should be highly personalized and consider their clinical presentation, comorbidities, and genomic profile, as well as the toxicity of the treatment regimens used to tailor treatment approaches.⁴ A number of treatment options for first and subsequent lines of therapy are identified in international guidelines; however, real-world treatment practices vary significantly, in part due to treatment availability. Common treatment options for patients with WM across all lines of therapy include alkylating drugs (bendamustine, cyclophosphamide), proteasome inhibitors (bortezomib, carfilzomib, ixazomib), anti-CD20 monoclonal antibodies (rituximab, ofatumumab), and BTK inhibitors (ibrutinib, zanubrutinib).⁴

Based on input from clinical experts consulted by CADTH for the purpose of this review, in Canada, therapeutic approaches for WM are based on watchful waiting or active surveillance for the approximately 25% of patients who are asymptomatic. Most patients presenting with symptomatic disease require treatment. In patients who present with life-threatening complications related to hyperviscosity or cryoglobulinemia, plasmapheresis is used as a temporary measure until definitive treatment is initiated. The standard approach for first-line treatment in Canada is chemoimmunotherapy, given that the vast majority of patients are good candidates for this treatment. The most commonly used chemoimmunotherapy regimen is BR, which is followed by maintenance rituximab. This regimen is associated with remissions that last longer than 5 years. For those unable to tolerate BR, other regimens have been used (DRC, bortezomib-rituximab, rituximab monotherapy, and chlorambucil monotherapy).

There is no standard of care for the treatment of R/R WM. Bortezomib-based chemotherapy is the most commonly used therapy (e.g., rituximab-CyBorD). Few patients are eligible for high-dose chemotherapy and autologous stem cell transplant, and even fewer are eligible for allogeneic hematopoietic stem cell transplant. If a patient has not had BR and has had a long remission (i.e., time to next treatment of more than 6 years to 7 years), then BR can be considered as second-line treatment. Currently, there is no publicly funded BTK inhibitor for this indication in Canada. Ibrutinib was reviewed by CADTH in September 2016 and was not recommended for reimbursement. In Canada, BTK inhibitors (e.g., ibrutinib, acalabrutinib, zanubrutinib) are available only through compassionate access programs, and are currently the most frequently used drugs in the R/R setting after failure of chemoimmunotherapy. BTK inhibitors are currently available through these programs, but access is temporary. BTK inhibitors are included in some provincial practice guidelines, such as Alberta’s, along with other non-funded options, including lenalidomide and everolimus.

None of the current treatments available for WM in any line of therapy can modify the underlying disease mechanism, and none is considered curative. All patients are expected to relapse and require additional treatment. The 2 most important goals of therapy are to relieve lymphoma- and paraprotein-related symptoms and to delay disease progression by achieving prolonged remission. Although WM is an indolent disease, it is associated with many potential symptoms (e.g., fatigue, aches, hyperviscosity) when it is active. The complications of paraproteinemia cause significant morbidity and can be life- and limb-threatening in some cases; severe hyperviscosity is an oncologic emergency. Because the patient population is generally older and “less fit,” treatments should be efficacious while minimizing toxicity. Improvement in HRQoL is always desired. With current therapies, patient QoL is good while patients are in remission, so most patients are willing to accept short-term symptoms (i.e., from chemotherapy) to achieve durable remission. In addition to delaying disease progression and controlling symptoms with minimal toxicity, the ideal treatment would minimize the hypogammaglobulinemia that is common with recurrent infections in patients with WM. More effective therapies that induce long-lasting remission would likely result in prolongation of OS; however, studies of WM rarely use OS as a primary end point.

Drug

Zanubrutinib (Brukinsa) is a small molecule inhibitor of BTK that, like other BTK inhibitors, forms an irreversible covalent bond at Cys481 within the adenosine triphosphate binding pocket of the BTK protein, preventing the proliferation and survival of malignant and normal B-cells. Zanubrutinib is a second-generation BTK inhibitor, designed to be more selective; it has more favourable pharmacokinetic and pharmacodynamic properties than the approved first-in-class BTK inhibitor, ibrutinib. Zanubrutinib has been shown to be more selective than ibrutinib for the inhibition of BTK in kinase inhibition and cell-based assays. Based on in vitro and in vivo studies, it was hypothesized that zanubrutinib would provide a deeper clinical response than ibrutinib as measured by response rate in patients with WM.

Zanubrutinib is indicated for the treatment of adult patients with WM. In the US, Brukinsa was granted an accelerated approval by the US FDA on November 14, 2019 for the treatment of adult patients with mantle cell lymphoma (MCL) who have received at least 1 prior therapy.¹³ In Canada, zanubrutinib was issued a Notice of Compliance by Health Canada on March 1, 2021.³ The indication is for the treatment of adult patients with WM.¹⁴ The sponsor has requested reimbursement criteria that align with the approved Health Canada indication.

Brukinsa is supplied in 80 mg oral capsules (size 0 hard gelatin capsules with a white to off-white opaque body and cap, marked in black ink with “ZANU 80”). The recommended total daily oral dose of zanubrutinib is 320 mg, which may be taken as either 320 mg (four 80 mg capsules) once daily or 160 mg (two 80 mg capsules) twice daily (12 hours apart). Per the Health Canada product monograph for zanubrutinib, treatment should continue until disease progression or unacceptable toxicity. Dose interruptions and reductions are recommended for non-hematological toxicities of grade 3 or higher, grade 3 febrile neutropenia, grade 3 thrombocytopenia with significant bleeding, and grade 4 neutropenia lasting more than 10 consecutive days. Discontinuation is recommended for grade 4 thrombocytopenia lasting more than 10 days.¹⁴

Stakeholder Perspectives

Patient Group Input

This section was prepared by CADTH staff based on the input provided by patient groups.

About the Patient Groups and Information Gathered

Four patient groups provided input for the review of zanubrutinib in WM: The CanCertainty Coalition and LC in collaboration with CORD and the WMFC. The CanCertainty Coalition is the united voice of more than 30 Canadian patient groups, cancer health charities, and caregiver organizations from across the country, joining together with oncologists and cancer care professionals to significantly improve the affordability and accessibility of cancer treatment. LC is a national Canadian registered charity that empowers the lymphoma community through education, support, advocacy, and research, collaborating with patients, caregivers, health care professionals, and other organizations and stakeholders to promote early detection, learn about the causes of lymphoma, find new and better treatments for lymphoma patients, help patients access those treatments, and work together to find a cure. CORD and WMFC are Canadian patient organizations with similar missions focused on their communities.

The data collected by the CanCertainty Coalition to inform this submission were sourced from literature, Canadian prescription drug insurance coverage, population demographics, and previously conducted surveys. The CanCertainty data collection and submission were completed using CanCertainty resources and personnel and contract personnel exclusively.

LC, CORD, and WMFC conducted an anonymous online survey of patients with WM between February 28, 2021 and May 10, 2021. The organizations reached patients through their respective databases and social media outlets (including Twitter, Instagram, and Facebook) and sent the survey to physicians to share with patients. The survey contained a combination of multiple choice, rating, and open‐ended questions. A total of 281 patients responded, of whom 109 had experience with a BTK inhibitor (87 with ibrutinib, 22 with zanubrutinib); 172 did not have this experience. Among all respondents, 47% lived in Canada, 56% were female, and 74% were 60 years of age or older.

Disease Experience

WM is considered a rare disease, which can make it a challenge to diagnose. Although 60% of patients in the survey received their diagnosis within 3 months of initial symptom presentation, 21% had to wait 6 months to 12 months, and 19% waited more than 1 year to receive a confirmed diagnosis. Symptoms of WM that most affected patients’ QoL at diagnosis included fatigue (66%), night sweats (28%), neuropathy (24%), weight loss or loss of appetite (20%), and easy bruising or bleeding (20%). A total of 81% of respondents experienced at least 1 psychological or social impact of a WM diagnosis, including stress and anxiety (66%), difficulty sleeping (30%), impact on daily activities (28%), memory loss or concentration problems (19%), and depression (19%). Similar symptom profiles and psychological or social impacts were observed between diagnosis and patients’ current status, which indicates that WM has consistent detrimental impacts on patients. Patients were asked to rate on a scale of 1 to 5 how WM had negatively affected various aspects of their lives (where 1 = no impact and 5 = significant negative impact). Patients indicated work, school, volunteering (3.62), and travel (3.04) as having been the most negatively affected.

Experience with Treatment

In the survey conducted by LC, CORD, and WMFC, 13% of patient respondents were still in the watch-and-wait phase following diagnosis and did not require treatment, and 40% of patients were currently receiving treatment. Of patients receiving treatment, 17% were receiving first-line treatment, 41% were in remission following a previous line of treatment, and 6% had relapsed following previous treatment and were waiting to begin re-treatment. The most common treatments included chemotherapy monotherapy (55%), monoclonal antibodies (63%), and BTK inhibitors (36%). In the later lines of therapy, BTK inhibitors were the top treatment choice.

The most common side effects of treatment for WM experienced by the surveyed patients included fatigue (72%), neutropenia (47%), nausea (39%), anemia (37%), peripheral neuropathy (37%), thrombocytopenia (30%), rash or itch (26%), back or joint pain (23%), mouth sores (22%), diarrhea (20%), headache (19%), and hair loss (17%). It was noted that many to all of these side effects were difficult to handle, particularly treatment-related fatigue. One patient said, “Anemia and fatigue continue to be the most challenging; recurring infection or susceptibility to infection is also a constant worry.”

Although not common, 20 patients confirmed side effects including infections or fever (30%), infusion-related reactions (20%), neutropenia (15%), cardiac complications (10%), and pneumonitis (10%), among others, as being the most difficult to tolerate because these resulted in hospitalization for management. It was noted that none of these side effects or hospitalizations was the result of BTK inhibitors. The side effects that patients experienced for longer than 2 years, or that appeared more than 2 years after treatment, included fatigue (36%), peripheral neuropathy (27%), and “chemo-brain” (21%). There were no long-term side effects reported by respondents related to BTK inhibitors.

When asked about the impact of various aspects of treatment (not including BTK inhibitors) on daily living, patients noted significant negative impacts due to treatment-related fatigue (30%), treatment side effects (27%), and infusion-related reactions or inability to tolerate treatment (19%). Patients said that previous treatments and side effects had further negatively affected their work, school, or volunteering activities (25%), daily activities (22%), and travel (27%). In contrast, patients indicated that BTK inhibitors did not negatively affect their mental health, work, school, or volunteer activities, relationships with family, friends, or intimate partners, ability to continue with daily activities, or personal image. In fact, BTK inhibitors actually had a positive impact in each of these categories.

The majority of patients in the LC survey were able to access treatment locally (78%); some of those who could not attributed this inability to living in a community without a cancer centre (9%) or to their treatment not being available at their local cancer centre (4%). As a result of not being able to access treatment locally, patients worried about their prognosis or survival (16%), required long and exhaustive trips to access treatment (13%), and experienced impacts to their daily activities (13%). Access to treatments such as oral BTK inhibitors (which do not involve travelling to a hospital or centre for administration) can limit the negative impacts related to treatment. CanCertainty noted that reimbursement of oral cancer drugs differs across Canadian provinces and territories: British Columbia, Alberta, Saskatchewan, Manitoba, Quebec, Northwest Territories, Yukon, and Nunavut reimburse oral cancer drugs for all in need, while Ontario and the Atlantic provinces do not. As a result, patients who do not have adequate insurance may have to pay out-of-pocket for medication and/or apply to funding assistance programs, which can take time and delay access to treatment. Financial impacts were also noted to be important to patients with WM. They cited the cost of medications (67%), parking (26%), and travel (21%). One patient said:

“During the treatment, I spent several hours 2 days a week getting therapy and then a couple of days to recover significantly which affected my available time for other things. When the protocol changed allowing my treatment to be given as an injection instead of IV, it cut down on the time which helped.”

As previously mentioned, 109 patients had experience with a BTK inhibitor, of whom 22 had experience with zanubrutinib accessed through a clinical trial (50%), private insurance (27%), or compassionate access program (3%). The most common side effects included easy bruising or bleeding (55%), diarrhea (18%), neutropenia (18%), rash or itch (8%), muscle or joint pain (18%), and muscle spasms (18%), with the most difficult to tolerate being diarrhea, bruising, and rash or itch. The majority of patients said zanubrutinib did not affect their work, school, travel, mental health, personal image, intimate or family relationships, or friendships, but did tend to positively affect their ability to continue with daily activities. When asked to describe their experience with zanubrutinib, 95% of patients said they had a good to excellent experience with the therapy; the remaining patients reported having a satisfactory experience, and said they would take this treatment option again if available and recommended by their doctor and would also recommend it to other patients. In comparison to patients who received other treatments, those who received zanubrutinib noted fewer side effects (64%), had a better and faster response rate (41%), and said it did not affect their QoL to the same extent as past treatments (36%).

One patient commented: “I am grateful for the opportunity to participate in a clinical trial for a drug that is less toxic than ibrutinib and that gives me such a deep response.” Another said, “It amazes me that people choose infusion therapy over the ease of zanubrutinib.” A third said, ““I like that it is an oral drug and that it has not produced any noticeable side effects.”

There were 6 patients in the LC survey who had been treated with both zanubrutinib and ibrutinib. According to the results, patients treated with both of these BTK inhibitors preferred their experience with zanubrutinib, citing less impactful side effects. Moreover, based on the summary results of the ibrutinib and zanubrutinib trials shared with the respondents, 45% of patients said they would use zanubrutinib over other BTK inhibitors, while 12% would use zanubrutinib after treatment with another BTK inhibitor has failed; 43% of patients were unsure. Reasons for choosing zanubrutinib over other BTK inhibitors included fewer side effects (63%) and a slightly better response rate (34%). A total of 62% of patients indicated they would also choose zanubrutinib if it was recommended by their doctor.

Improved Outcomes

Patients responding to the survey considered having a choice of treatment and enough treatment options to be important. To access new treatments for their WM, 67% of patients reported that they would be interested in participating in a clinical trial. In terms of treatment outcomes, patients rated longer survival (75%), longer remission (76%), better QoL (70%), and fewer side effects (57%) as the most important.

Patients stated that they would be likely to accept known, non–life-threatening risks or side effects for a new treatment. Very few patients would be willing to tolerate severe or long-term side effects. On a scale of 1 to 5, patients rated headache or cognitive changes (56%), changes in vision (58%), shortness of breath (46%), abdominal discomfort (nausea, vomiting, diarrhea, or constipation) (42%), and fatigue (42%) as the most important symptoms for new WM treatments to control.

Patient expectations for new treatment options included a “targeted oral treatment that will not cause secondary cancers or more discomfort than the disease itself” and would “provide [an] increase in quality of life and longevity with minimal side effects.”

Clinician Input

Input From Clinical Experts Consulted by CADTH

All CADTH review teams include at least 1 clinical specialist with expertise regarding the diagnosis and management of the condition for which the drug is indicated. Clinical experts are a critical part of the review team and are involved in all phases of the review process (e.g., providing guidance on the development of the review protocol; assisting in the critical appraisal of clinical evidence; interpreting the clinical relevance of the results; and providing guidance on the potential place in therapy). The following input was provided by 3 clinical specialists with expertise in the diagnosis and management of WM.

Unmet Needs

The clinical experts consulted by CADTH noted that patients with WM become resistant to current available treatment options and that all patients will eventually relapse after chemoimmunotherapy. After failing chemoimmunotherapy, results from re-treatment with chemoimmunotherapy are disappointing — particularly for patients who are refractory or early progressors — because re-treatment subjects patients to the toxicities of treatment without the hope of prolonged benefit. Moreover, remissions tend to be shorter with each subsequent round of chemoimmunotherapy. Many patients progress quickly after second-line treatment, and re-treatment with chemoimmunotherapy is generally of limited use. Given that patients often become immunosuppressed with initial therapy treatments, treatment options that minimize toxicity during relapse are very important. The only clear unmet need currently is for R/R patients. One clinical expert explained that before the availability of BTK inhibitors, patients would be offered complete palliation after failure of 2 lines of therapy because the toxicities of therapy outweighed the potential benefit. BTK inhibitors have consistently shown benefit in the relapsed setting for other lymphoproliferative disorders, namely MCL and chronic lymphocytic leukemia, particularly when used earlier in the course of disease. The clinical experts consulted by CADTH indicated that the same impact on efficacy has been demonstrated for WM, and that as such, BTK inhibitors have become a preferred treatment option (available through compassionate access programs) for relapsed WM because these drugs are generally well tolerated in the older, less fit and unfit population. In the experience of the clinical experts, these therapies have been life-changing and life-sustaining for patients with WM who have no other effective alternatives.

Place in Therapy

The clinical experts consulted by CADTH indicated that zanubrutinib would be used in the R/R setting after failure of standard chemoimmunotherapy. In the opinion of the clinical experts, zanubrutinib is expected to be more efficacious (i.e., associated with prolonged remission) and less toxic than a repeated round of chemotherapy. The clinical experts noted that this is particularly true for patients who fail chemotherapy early (i.e., while on rituximab maintenance or < 3 years after standard chemotherapy without maintenance rituximab). The clinical experts indicated that, based on the available evidence, they would generally not consider zanubrutinib in the first-line treatment setting because it does not provide sufficient benefit relative to standard chemoimmunotherapy to justify the added cost and low-level toxicity and inconvenience of indefinite first-line therapy.

All patients should be offered chemoimmunotherapy as first-line treatment unless they are considered truly unfit for anything other than rituximab therapy or even oral chlorambucil. These patients have a defined treatment interval and can enjoy a prolonged remission after chemoimmunotherapy (with or without rituximab); reserving zanubrutinib in later lines does not appear to result in reduced survival (based on experience with other BTK inhibitors in different disease settings). Therefore, there is no downside to postponing treatment with zanubrutinib to the relapsed setting. Patients should fail at least 1 line of therapy before being offered zanubrutinib, and the clinical experts noted that intolerance to treatment would be an insufficient reason for patients to not receive standard first-line treatment. For example, if a patient is unable to tolerate BR, other chemoimmunotherapy-based regimens should be attempted in its place (e.g., DRC or CyBorD). Even frail, elderly patients can trial dose-reduced BR or DRC. In the opinion of the clinical experts, zanubrutinib does offer an oral therapy with low toxicity that elderly patients may value, but that there are insufficient data to support its use in the treatment-naive population. The pivotal trial of zanubrutinib included only a small number of treatment-naive patients and did not define this population robustly. One of the clinical experts was of the view that zanubrutinib could be used in patients who had previously been treated with ibrutinib, but were intolerant, and that it should not be reserved for patients with contraindications to other therapies. The clinical experts consulted by CADTH indicated that if zanubrutinib was approved and funded, they would expect more patients to receive it as a second- or later-line therapy. One of the clinical experts stated that they would favour first-line chemoimmunotherapy and BTK inhibitors in second and later lines unless BTK inhibitors for indefinite use (i.e., continued until progression) were priced very reasonably.

Patient Population

Zanubrutinib may be offered to patients with R/R WM, particularly those with MYD88 mutations, who require treatment after chemoimmunotherapy and to those who obtained a poor response to chemoimmunotherapy or had chemoimmunotherapy more than once. There is no standard of care for these patients, and repeated rounds of chemotherapy carry toxicity while offering a low likelihood of prolonged benefit. The clinical experts noted that there are no specific disease characteristics that make patients more or less suitable for treatment with zanubrutinib. The drug is expected to be more effective in earlier lines of treatment; as such, it would be considered routinely in second-line therapy or beyond. Treatment is indicated in the R/R setting for patients with symptomatic disease only (i.e., a period of observation at relapse can and should be done because a return of paraproteinemia by itself is not an indication for treatment). When patients develop symptomatic disease, then second-line treatment should be offered.

In terms of how patients best suited for treatment with zanubrutinib can be identified, the clinical experts noted that WM is not challenging to diagnose, but determining whether a patient’s symptoms or findings are caused by WM requires expertise. For example, patients are often sent for reassessment or re-treatment, and may be determined to have alternative diagnoses, such as a second malignancy or progressive anemia from iron deficiency. Diagnosis of WM is both a clinical and pathological 1. Molecular techniques to confirm MYD88 mutational status can help distinguish WM from other lymphoproliferative disorders, particularly marginal zone lymphoma. While MYD88 testing is not performed routinely, it can be done if there is diagnostic uncertainty between marginal zone lymphoma and lymphoplasmacytic lymphoma. One of the clinical experts stated that MYD88 mutation should be confirmed before embarking on BTK inhibitor treatment; however, testing may not be available in all provinces. Although MYD88 mutation may be found in other hematologic B-cell malignancies, certain clinical features — such as the presence of paraproteinemia (which is essential) and the absence of other classic lymphoma features (e.g., lymphadenopathy) — can help to confirm the diagnosis of WM.

The clinical experts consulted by CADTH also indicated that patients with asymptomatic disease should not be treated with zanubrutinib unless there is concern about impending hyperviscosity syndrome. Patients who are at very high risk for bleeding complications (e.g., those who require antiplatelet or anticoagulation equivalent) would also be least suitable for treatment with zanubrutinib. Patients who have previously progressed on a BTK inhibitor should not be eligible for zanubrutinib, whereas patients who are intolerant of ibrutinib could be considered. One of the clinical experts noted that the ASPEN trial excluded patients with CNS involvement, but Bing Neel syndrome is, in fact, a situation where BTK inhibitors, including zanubrutinib, may be particularly valuable due to CNS penetration; thus, CNS involvement should not be used as a reason not to offer zanubrutinib. Ibrutinib is a well-established therapy for Bing Neel syndrome.

Assessing Response to Treatment

Regarding how and when patients eligible to receive zanubrutinib should be assessed to determine if they are benefiting from the treatment, the clinical experts consulted by CADTH indicated that there are currently no data on non-responders or how to identify them for zanubrutinib or any of the BTK inhibitors. Response to treatment is measured by assessing disease status after the initiation of therapy, and response is assessed as either CR, partial response (PR), stable disease, or progressive disease (PD), as in clinical trials. If a patient has stable disease or PD, the current treatment is discontinued, and another treatment is initiated. A PR would be acceptable and is generally the norm for patients with WM on any currently available treatment. Patients are then monitored until progression; time to next treatment would be the next time point. However, response rates are not sufficient; a more substantial measure, such as PFS, is required as a minimum for clinicians to adopt a new treatment in practice. Time to next treatment is also important because if the first-line treatment delays the initiation of a second treatment, this is particularly useful information. In indolent lymphoma (in which patients can receive multiple therapies and live for many years, similarly to patients with myeloma), it is difficult to assess OS; as such, OS is less often considered for WM or studies of other indolent lymphomas.

The clinical experts considered that a clinically meaningful response to treatment would include: hematological response (e.g., resolution of cytopenias, splenomegaly), reduction in or elimination of paraprotein, resolution of lymphoma-related symptoms (e.g., neuropathy with WM), and prolonged DoR — the longer the better, given that therapies are limited for this disease in regard to the magnitude of response (ideally a CR, but a PR would be the expected result, and many patients can enjoy prolonged PFS even with a PR).

Response to treatment is generally assessed every 3 months to 6 months, or more frequently when a therapy is newly initiated (i.e., every cycle). This would also apply to patients on zanubrutinib, in whom disease status is often assessed after the initial 3 months to 6 months and every 3 months thereafter.

Discontinuing Treatment

The parameters described by the clinical experts that could be used to identify patients who are no longer responding to or benefiting from the treatment include: clinically symptomatic disease progression, new lymphadenopathy or splenomegaly, progressive anemia from marrow infiltration, or progressive IgM increase (not just minimal change in monoclonal protein); and severe toxicity (particularly grade 3 or higher) that cannot be managed through dose reduction.

Prescribing Conditions

The clinical experts consulted by CADTH considered any setting, including community or academic, to be appropriate for providing treatment with zanubrutinib, provided the prescribing clinicians understand how to prescribe and monitor the therapy. However, zanubrutinib would generally be expected to be provided in an outpatient clinic by a hematologist or oncologist.

Additional Considerations

One of the clinical experts commented that WM is truly an orphan disease, affecting a rare group of patients with unique clinical manifestations who do not respond as well to chemoimmunotherapy as patients with other indolent lymphomas (e.g., follicular lymphoma). There are few treatment options at relapse, and those available are generally ineffective. As well, there are few or no new therapies available through clinical trials. Consequently, access to BTK inhibitors is imperative for this group of patients.

Clinician Group Input

This section was prepared by CADTH staff based on the input provided by patient groups.

Clinician input was received from 2 registered clinicians on behalf of the Ontario Health (Cancer Care Ontario) Hematology Cancer Drug Advisory Committee for the review of zanubrutinib for the treatment of WM. The Ontario Health-Cancer Care Ontario Drug Advisory Committees provide evidence-based clinical and health system guidance on drug-related issues, including those related to the provincial drug reimbursement programs and the Systemic Treatment Program.

Current Treatments

Treatments for first-line therapy for WM include BR and ibrutinib-rituximab (accessible through private pay). Treatments for patients who have relapsed include re-treatment with BR or treatment with ibrutinib-rituximab, other rituximab chemotherapy combinations, or palliative chlorambucil.

Unmet Needs

The clinicians stated that current treatments are not curative, and patients often become refractory. Additionally, some patients are unable to tolerate ibrutinib due to its toxicity profile. The goals of treatment for patients with WM are to delay the progression of disease, prolong survival, prevent end-organ effects related to hyperviscosity, and improve overall HRQoL. Most patients demonstrate a good response to first-line BR and remain free of relapse for a few years. It is patients with relapsed disease who have a significant unmet need for a drug like zanubrutinib.

Place in Therapy

The clinicians advised that zanubrutinib may be used in the first-line setting or after relapse. There is currently no evidence to suggest the specific sequencing of treatment with zanubrutinib, given that the ASPEN study enrolled newly diagnosed patients as well as patients who had been previously treated. There is also no evidence to suggest whether zanubrutinib should be prescribed to patients who fail or are intolerant to ibrutinib.

Patient Population

The patients best suited to this treatment are those with symptomatic R/R WM. These patients are identified as per the routine clinical diagnosis for WM. The clinicians explained that the patients least suited for zanubrutinib are those with prior BTK inhibitor exposure, given that they were excluded from the ASPEN study. It is not possible for the clinicians to identify the patients most likely to exhibit a response to zanubrutinib because the ASPEN study did not identify specific subgroups of patients likely to benefit the most.

Assessing Response to Treatment

Outcomes that can indicate whether a patient is responding to treatment include response rates based on blood count, IgM level, and routine imaging, as per clinical practice. The prevention of end-organ effects and, at minimum, a PR to treatment would be considered clinically meaningful responses to zanubrutinib. Response to treatment should be assessed approximately every 1 month to 3 months, as per clinical practice.

Discontinuing Treatment

Progression of disease, lack of clinically meaningful responses, and the occurrence of treatment-related toxicities are good indicators to help decide if treatment with zanubrutinib should be discontinued.

Prescribing Conditions

The clinicians explained that zanubrutinib would be prescribed in community settings because it is an oral drug that can be taken at home.

Drug Program Input

The drug programs provide input on each drug being reviewed through CADTH’s reimbursement review processes by identifying issues that may affect their ability to implement a recommendation. The drug plans noted that in the ASPEN trial, zanubrutinib was compared to ibrutinib, which is not publicly funded in any jurisdiction in Canada. Ibrutinib, for the treatment of patients with WM who have received at least 1 prior therapy, was previously reviewed by CADTH and not recommended for reimbursement. Ibrutinib may be available for some patients (at no charge) through the sponsor’s patient support program. Relevant comparators for WM in Canadian jurisdictions include rituximab-based chemotherapy (BR, bortezomib-dexamethasone-rituximab, and DRC) for treatment-naive patients and those with R/R WM. Re-treatment with rituximab is funded for patients with a relapse-free interval (6 months to 12 months, depending on jurisdiction) following the last dose of rituximab. In terms of prescribing considerations, the drug plans noted that zanubrutinib has the potential for drug-drug interactions, possibly increasing pharmacy resource use. However, in terms of care provision, the capsule strength of 80 mg (in bottles of 120) facilitates dispensing and dose adjustment without wastage. Some system and economic issues were noted by the drug plans. The submitted budge impact analysis (BIA) includes ibrutinib, which is not publicly funded in Canada for WM, potentially affecting the BIA results. However, a revised BIA was submitted by the sponsor that no longer included any market share or costs for ibrutinib. The drug plans also noted that a confidential negotiated price exists for biosimilar rituximab (pan-Canadian Pharmaceutical Alliance) and subcutaneous rituximab; and bendamustine and bortezomib are available in a generic format.

The implementation questions and corresponding responses from the clinical experts consulted by CADTH are summarized in Table 3.

Clinical Evidence

The clinical evidence included in the review of zanubrutinib is presented in 2 sections. The first section, the Systematic Review, includes pivotal studies provided in the sponsor’s submission to CADTH and Health Canada, as well as those studies that were selected according to an a priori protocol. The second section includes indirect evidence from the sponsor and indirect evidence selected from the literature that met the selection criteria specified in the review.

Systematic Review (Pivotal and Protocol Selected Studies)

Objectives

To perform a systematic review of the beneficial and harmful effects of zanubrutinib 80 mg oral capsules for the treatment of adult patients with WM.

Methods

Studies selected for inclusion in the systematic review included pivotal studies provided in the sponsor’s submission to CADTH and Health Canada, as well as those meeting the selection criteria presented in Table 4. Outcomes included in the CADTH review protocol reflect outcomes considered to be important to patients, clinicians, and drug plans.

The literature search was performed by an information specialist using a peer-reviewed search strategy according to the PRESS Peer Review of Electronic Search Strategies checklist.¹⁵

Published literature was identified by searching the following bibliographic databases: MEDLINE All (1946‒) through Ovid and Embase (1974‒) through Ovid. All Ovid searches were run simultaneously as a multi-file search. Duplicates were removed using Ovid deduplication for multi-file searches, followed by manual deduplication in Endnote. The search strategy comprised both controlled vocabulary, such as the National Library of Medicine’s MeSH (Medical Subject Headings), and keywords. The main search concept was Brukinsa (zanubrutinib). Clinical trials registries were searched: the US National Institutes of Health’s clinicaltrials.gov, WHO’s International Clinical Trials Registry Platform (ICTRP) search portal, Health Canada’s Clinical Trials Database, and the European Union Clinical Trials Register.

No filters were applied to limit the retrieval by study type. Retrieval was not limited by publication date or by language. Conference abstracts were excluded from the search results. See Appendix 1 for the detailed search strategies.

The initial search was completed on June 21, 2021. Regular alerts updated the search until the meeting of the pERC on October 13, 2021.

Grey literature (literature that is not commercially published) was identified by searching relevant websites from the CADTH Grey Matters: A Practical Tool for Searching Health-Related Grey Literature checklist. Included in this search were the websites of regulatory agencies (US FDA and European Medicines Agency). Google was used to search for additional internet-based materials. See Appendix 1 for more information on the grey literature search strategy.

Two CADTH clinical reviewers independently selected studies for inclusion in the review based on titles and abstracts, according to the predetermined protocol. Full-text articles of all citations considered potentially relevant by at least 1 reviewer were acquired. Reviewers independently made the final selection of studies to be included in the review, and differences were resolved through discussion.

A focused literature search for network meta-analyses (NMA) dealing with WM or Brukinsa (zanubrutinib) was run in MEDLINE All (1946–) on June 18, 2021. No limits were applied. Articles were screened by 1 researcher for ITCs that met the patient, intervention, comparator, and outcome criteria listed in Table 4.

Findings From the Literature

A total of 107 studies were identified; 103 were excluded, while 4 potentially relevant citations were retrieved for full-text screening.^16-19 Three potentially relevant reports from other sources were also identified that included regulatory approvals (i.e., Health Canada and FDA). Two studies met the inclusion criteria.^16,18 The details of the included study (the ASPEN trial) are summarized in Table 5. Excluded studies are listed in Appendix 2. Information relevant to this report was derived from the submission to CADTH, Health Canada, and the FDA (Figure 1).^3,7,13

Figure 1: Flow Diagram for Inclusion and Exclusion of Studies

Description of the ASPEN Trial

The ASPEN trial is an ongoing, phase III, randomized, open-label, multi-centre study designed to compare the efficacy and safety of zanubrutinib to ibrutinib in patients with WM who require therapy according to the consensus panel criteria from the 7th International Workshop on Waldenström Macroglobulinemia.^7,20 The study consisted of an initial screening phase, a treatment phase, and a follow-up phase. The study is being conducted at 60 centres in 12 countries (Australia, Czech Republic, France, Germany, Greece, Italy, Netherlands, Poland, Spain, Sweden, UK, and US). Ibrutinib was chosen as the comparator because it was approved by the European Medicines Agency for the treatment of WM in adults who have received prior treatment for their disease and in previously untreated patients for whom treatment with chemoimmunotherapy is not suitable. The FDA approved zanubrutinib for the treatment of adult patients with WM in August 2021.

Randomization and treatment allocation: Based on MYD88 gene sequencing, patients were enrolled into either cohort 1 (MYD88^L265P) or cohort 2 (MYD88^WT). Patients with either missing or inconclusive MYD88 gene-sequencing results were assigned to cohort 2 by default. Using an interactive response technology system, cohort 1 patients were randomized 1:1 to receive either zanubrutinib (arm A) or ibrutinib (arm B). Stratification factors included CXCR4 mutational status (CXCR4^WHIM versus CXCR4^WT versus missing) and the number of prior therapies for WM (0 versus 1 to 3 versus > 3). A computer-generated randomization list, including stratification factor values and treatment arm assignments, was produced, reviewed, and approved by an independent statistician. Cohort 2 patients were assigned to receive zanubrutinib (arm C, non-randomized) by the interactive response technology system (Figure 2).⁷

Blinding: This was an open-label study. The IRC was blinded to study treatment, but the independent Data Monitoring Committee was not.

Study phases: The screening phase consisted of screening evaluations that were performed within 35 days before randomization, with the exception of a fresh bone biopsy, which could be performed up to 42 days before randomization as long as no intervening therapy had been administered. A fresh bone marrow aspirate was required for flow cytometry and the MYD88 and CXCR4 mutational analyses at screening. After treatment assignment, the first dose of ibrutinib or zanubrutinib was administered at cycle 1, day 1. A treatment cycle consisted of 28 days; treatment continued until progression or unacceptable toxicity. In all study arms, patients were to return approximately 30 days after the last dose of the study drug for safety follow-up visit(s) for the collection of information about AEs and SAEs that may have occurred after the patient discontinued the study. Information on new anticancer therapies given after the last dose of the study drug continued to be collected after discontinuation of study drug.

Protocol Amendments

The protocol was amended 5 times before the data cut-off date. In the original statistical analysis plan (SAP), the hierarchical testing procedure of the primary end point (VGPR or CR rate) included a noninferiority test. The noninferiority test was removed from the planned analyses in the final SAP before unblinded analyses being performed (by the sponsor), but the test was performed as a post hoc analysis. In addition, the noninferiority margin for the key secondary end point of major response rate (MRR) was changed to 12% from 8%. Other major changes to the conduct of the study included: a background update, with new zanubrutinib results and data on the role of MYD88 mutation in responsiveness of WM to BTK inhibitors; a change of the primary objective to the proportion of patients who achieved VGPR or CR, based on the clarification of the primary study hypothesis that stemmed from the updated zanubrutinib data; the addition of MRR and VGPR or CR (by investigator assessment) as secondary end points; the addition of antitumour activity and safety of zanubrutinib in MYD88^WT patients with WM as exploratory end points; the addition of QoL and medical resource utilization as exploratory end points; and the identification of patients with MYD88^L265P WM as the primary population for randomization and study analyses (cohort 1).

Figure 2: ASPEN Trial Design

BID = twice daily; CXCR4 = chemokine receptor 4; MYD88MUT = mutated MYD88 gene; MYD88^WT = wild-type MYD88 gene; PD = progressive disease; QD = once daily; R = randomization; RR = relapsed/refractory; TN = treatment-naive; UNK = unknown; WHIM = warts, hypogammaglobulinemia, immunodeficiency, and myelokathexis syndrome; WT = wild type.

Note: TN indicates unsuitability for chemoimmunotherapy (up to 20% of overall population).

Source: Clinical Study Report for Brukinsa.⁷

Populations

Inclusion and Exclusion Criteria

In the ASPEN trial, eligible patients had R/R WM after at least 1 prior line of therapy or were treatment-naive and considered by their treating physician to be unsuitable for standard chemoimmunotherapy regimens, based on comorbidities and risk factors (hereafter referred to as unfit, treatment-naive). Study investigators were required to document the organ system(s) and specific reason(s) for which they considered the patient unsuitable for standard chemoimmunotherapy. Patients with no prior therapy (i.e., unfit, treatment-naive) comprised no more than 20% of the patients in cohort 1. Relapsed patients were defined as whose who previously achieved a CR or VGPR or PR, but showed PD after a period of greater than or equal to 6 months. Refractory patients were defined as those who experienced prior treatment failure or disease progression within 6 months of therapy initiation. Patients were required to have measurable disease and adequate end-organ function. Patients with prior BTK inhibitor exposure, disease transformation, clinically significant cardiovascular disease, or active CNS lymphoma were not eligible to participate in the trial. The inclusion and exclusion criteria used in the trial are shown in Table 5.

Baseline Characteristics

Demographic Characteristics and Medical History

Between January 2017 and July 2018, 164 R/R and 37 unfit, treatment-naive patients with WM were recruited into cohort 1 (MYD88^L265P). Two R/R patients were randomized, but never dosed (1 patient in the ibrutinib arm had a CNS lymphoma identified before dosing, and 1 patient in the zanubrutinib arm had acute kidney injury). The most common indications (> 20%) for therapy initiation were fatigue (57.2%), anemia (43.8%), B symptoms (30.3%), hyperviscosity (26.9%), and peripheral neuropathy (22.4%). The median age of all patients was 70.0 years. The majority of patients were male (66.7%), White (91.0%), and randomized at sites in Europe (59.7%) and Australia or New Zealand (30.8%) (Table 6). Overall, 8% of patients in the ibrutinib arm and 11% of patients in the zanubrutinib arm had a CXCR4^WHIM mutation. Approximately 85% were in the intermediate- or high-risk prognostic category, and 77% had CT evidence of extramedullary disease.²¹ Cohort 2 (MYD88^WT) included 28 patients (23 R/R, 5 unfit, treatment-naive) (Table 7). The most common indications for therapy initiation were fatigue (60.7%), B symptoms (35.7%), anemia (32.1%), hyperviscosity (21.4%), and peripheral neuropathy (10.7%). The median age was 72 years; 50% of patients were male and 96.4% were White.

In cohort 1, 37 unfit, treatment-naive patients were considered unsuitable for chemoimmunotherapy (Table 8): 27 patients (73.0%) were deemed unsuitable due to age (range = 64 years to 89 years); 8 patients (21.6%) were deemed unsuitable due to cardiac conditions (e.g., hypertension, ischemic heart disease, dilated cardiomyopathy, CABG); 4 patients (10.8%) were deemed unsuitable due to renal conditions (e.g., inadequate renal function for standard chemotherapy, chronic kidney disease); and 2 patients (5.4%) were deemed unsuitable due to infection (e.g., recurrent bacterial infections and sinusitis). Other miscellaneous reasons for unsuitability included lack of central venous access, allergic reaction to rituximab, and high IgM levels that, in the treating physician’s judgment, were a contraindication to chemoimmunotherapy.

In cohort 1, prior and/or concomitant medical conditions were reported in 197 patients (98.0%) overall (98 patients [99.0%] in the ibrutinib treatment arm and 99 patients [97.1%] in the zanubrutinib treatment arm). All 37 unfit, treatment-naive patients (100%) and 160 R/R patients (97.6%) had prior and/or concomitant medical conditions. The most common prior and/or concomitant medical conditions reported in greater than or equal to 10% of patients overall included hypertension (40.8%), anemia (20.9%), fatigue (15.9%), benign prostatic hyperplasia (13.4%), gastroesophageal reflux disease (11.9%), and insomnia (10.9%). In cohort 2, prior and/or concomitant medical conditions were reported in 27 patients (96.4%) overall, and included hypertension (35.7%), anemia (17.9%), and benign prostatic hyperplasia, cholecystectomy, chronic obstructive pulmonary disease, or hypercholesterolemia (14.3% each).

Prior Anticancer Therapies

For the 164 patients in cohort 1 and the 23 patients in cohort 2 with R/R disease, the median number of prior anticancer therapies was 1. In cohort 1, the most common prior anticancer therapies were rituximab (rituximab or ofatumumab; 149 patients [90.9%]); alkylating drugs (139 patients [84.8%]), and corticosteroids (110 patients [67.1%]); the use of these was generally comparable between the ibrutinib and zanubrutinib treatment arms, except for the following for R/R patients: prior alkylators (88.0% for zanubrutinib versus 81.5% for ibrutinib), steroid use (72.3% for zanubrutinib versus 61.7% for ibrutinib), and vinca alkaloids (27.7% for zanubrutinib versus 22.2% for ibrutinib). In addition, there were 3 patients with a prior history of stem cell transplant in the zanubrutinib treatment arm versus only 1 such patient in the ibrutinib arm. Rituximab monotherapy was used as part of prior treatments in 18% of cohort 1 overall (22 patients [22%] in the ibrutinib arm and 14 patients [14%] in the zanubrutinib arm). In cohort 2, the most common prior anticancer therapies were alkylating drugs and rituximab (22 patients [95.7%]) followed by corticosteroids (17 patients [73.9%]). Only 1 patient in cohort 2 had received rituximab monotherapy before study enrolment (Table 9).

Interventions

Zanubrutinib or ibrutinib was dispensed by the study centre personnel to patients at scheduled study visits to ensure adequate drug supply for administration at home throughout the treatment phase. Patients randomized or assigned to zanubrutinib (arms A and C) were instructed to take 160 mg (80 mg × 2 capsules) orally with a glass of water twice daily at approximately the same time each day. The time difference between 2 consecutive doses should have been at least 8 hours. Patients randomized to ibrutinib (arm B) were instructed to take 420 mg (140 mg × 3 capsules or in other applicable dose forms) orally with a glass of water once daily at approximately the same time each day. Patients were not required to fast before or after administration of either zanubrutinib or ibrutinib. Zanubrutinib or ibrutinib was to be taken as prescribed from cycle 1, day 1 until disease progression, unacceptable toxicity, death, withdrawal of consent, loss to follow-up, or termination of the study by the sponsor. Compliance with study drug administration was measured by reviewing patient diaries and tablet counts at each study visit.

Outcomes

The primary efficacy end point was the proportion of patients in each arm of cohort 1 achieving either CR or VGPR, as determined by the IRC using an adaptation of the response criteria updated at the Sixth IWWM.^5,6

The main secondary efficacy end points for cohort 1 were as follows:

• MRR as assessed by the IRC, defined as the proportion of patients achieving CR, VGPR, or PR

• DoR as assessed by the IRC, defined as the time from first determination of response (CR, VGPR, or PR) (per modified IWWM criteria) until first documentation of progression (per modified IWWM criteria) or death, whichever comes first

• PFS as assessed by the IRC, defined as the time from randomization to the first documentation of progression (per modified IWWM criteria) or death, whichever occurs first

• Resolution of treatment-precipitating symptoms, defined as the absence of the symptoms that triggered the initiation of study treatment (per the IWWM treatment guidelines) at any point during study treatment

The main exploratory end points included:

• Time to next treatment, defined as the time from the date of randomization until the start date of a new anticancer therapy other than study medications in patients with MYD88^L265P WM (cohort 1)

• OS, defined as the time from the date of randomization until the date of death from any cause in patients with MYD88^L265P WM (cohort 1)

• MRR according to CXCR4 mutation status (CXCR4^WHIM versus CXCR4^WT) in patients with MYD88^L265P WM (cohort 1)

• Change in QoL as assessed using the EORTC QLQ-C30 and EQ-5D in patients with MYD88^L265P WM (cohort 1) (further detail, including information on the validity and reliability of these instruments, is presented in Appendix 3)

• Medical resource utilization as assessed by the number of hospitalizations, lengths of hospital stays, and supportive care in patients with MYD88^L265P WM (cohort 1)

• Anticancer activity of zanubrutinib (i.e., CR or VGPR rate, MRR, overall response rate, PFS, DoR, and OS as assessed by the IRC and by the investigator) in patients with MYD88^WT WM (cohort 2)

Assessments

Bone marrow aspiration and biopsy information was collected at baseline and week 48, and as clinically indicated thereafter, including for confirmation of CR. Bone marrow samples taken at baseline were assayed for MYD88 and CXCR4 mutations before randomization. Quantitative serum immunoglobulins, M-paraprotein, and beta2-microglobulin levels were measured at baseline, the start of each cycle until cycle 12, and every 3 cycles thereafter. MRI or contrast-enhanced CT scans were performed at baseline. Patients with extramedullary disease underwent follow-up scans every 3 cycles until cycle 12 and every 6 cycles thereafter until disease progression. HRQoL assessments were collected at baseline, every 3 cycles until cycle 12, and every 6 cycles thereafter (Table 9). All patients were followed for AEs for 30 additional days after the last dose of the study drug. All treatment-related AEs and SAEs were followed until resolution or stabilization.

Safety Outcomes

The incidence, timing, and severity of AEs were assessed using Version 4.3 of the National Cancer Institute Common Terminology Criteria for Adverse Events. A treatment-emergent adverse event (TEAE) was defined as an AE with an onset time or increase in severity level on or after the first dose of study drug and within 30 days after the last dose of study drug or before the initiation of a new anticancer therapy, whichever occurred first. Unless otherwise stated, all AE summaries are of TEAEs. A treatment-related AE was an AE that was assessed by the investigator as related to the study drug or for which an assessment of the causal relationship was missing. AEs of special interest included hemorrhage (including minor bleeding, such as contusion and petechiae), major hemorrhage (defined as serious or ≥ grade 3 bleeding at any site or CNS bleeding of any grade), atrial fibrillation or flutter, hypertension, second primary malignancies, tumour lysis syndrome, infections (opportunistic), neutropenia, thrombocytopenia, and anemia.

Statistical Analysis

Sample Size Calculations

The sample size calculation was based on the comparison of the primary end point of CR or VGPR rate in the R/R analysis set in cohort 1. Assuming that RR_A equals 0.35 and RR_B equals 0.15, where RR_A and RR_B denote the CR or VGPR rate in arm A and arm B, respectively, 75 patients per arm (150 in total) provided a power of 0.814 in testing RR_A versus RR_B in the R/R analysis set in cohort 1 using a normal approximation to binomial test with a 2-sided significance of 0.05. Assuming that MRR_A equals 0.90 and MRR_B equals 0.80, the power of demonstrating noninferiority of zanubrutinib in the R/R analysis set in cohort 1 was 96.8% when a noninferiority margin of 12% was used. In addition to the 150 R/R patients, approximately 20% (n = 38) of unfit, treatment-naive patients with MYD88^L265P were enrolled in cohort 1. Assuming MYD88^L265P mutation was present in 90% of the enrolled patients, a total of approximately 210 patients were enrolled in cohorts 1 and 2 combined.

Analysis Sets

The intention-to-treat (ITT) analysis set included all randomized patients assigned to a treatment in cohort 1 (MYD88^L265P).

The R/R analysis set (a subset of the ITT analysis set) included all randomized patients with at least 1 prior line of therapy. This was the primary analysis set used for efficacy analyses.

The efficacy analysis set in cohort 2 included all patients who received any dose of zanubrutinib and were centrally confirmed to have MYD88^WT.

The safety analysis set included all patients who received any dose of zanubrutinib or ibrutinib. This was the analysis set used for all safety analyses.

Analyses of Outcomes

In the original SAP, the hierarchical testing procedure of the primary end point (VGPR or CR rate) included a noninferiority test with a noninferiority margin of –4.5% to be conducted before the superiority test. The margin of –4.5% was determined before the unblinded analysis (of the sponsor) being performed, using the 95% to 95% fixed margin approach on the estimated treatment effect of ibrutinib based on 2 ibrutinib monotherapy studies.^22,23 After discussions with the FDA before the unblinded analysis, the noninferiority test was removed from the planned analyses, and was instead performed as a post hoc analysis, as specified in the final SAP.

The primary efficacy analysis was planned to be performed approximately 12 months after the last R/R patient was randomized. Comparison between ibrutinib and zanubrutinib for the primary efficacy end point (cohort 1) was based on a hierarchical fixed-sequence procedure to adjust for multiplicity. The analysis of the superiority of zanubrutinib compared to ibrutinib in patients with R/R WM was performed first. If the comparison was statistically significant, further testing was performed using the ITT population (including 38 treatment-naive patients in addition to the R/R patients with MYD88^L265P). The superiority of the primary end point of CR or VGPR rate was tested using the Cochran-Mantel-Haenszel test stratified by the CXCR4 status (CXCR^WHIM versus CXCR^WT/missing), prior line of therapy (1 to 3 versus > 3 for R/R patients and 0 versus > 3 in the ITT analysis set), and age group (≤ 65 years versus > 65 years) at a 1-sided significance level of 0.025. If the 2-sided P value was less than 0.05 and the estimated risk difference was positive, it would be concluded that the VGPR or CR rate for zanubrutinib was significantly greater than the VGPR or CR rate for ibrutinib, and the primary objective of superiority would be met.

The key secondary end point of MRR by IRC was to be tested only if any of the superiority tests for the primary end point were statistically significant. If the primary end point of VGPR or CR rate was superior in the R/R analysis set only, the key secondary end point of MRR was tested for noninferiority in the R/R analysis set at a 1-sided significance level of 0.025. If the primary end point of VGPR or CR rate was superior in both the R/R and ITT analysis sets, MRR would be tested for noninferiority in the R/R and ITT analysis sets, respectively, at a 1-sided significance level of 0.025. The study-wide type I error would be controlled at a 1-sided 0.05 level. For the other secondary end points assessed, including PFS, the statistical tests performed were descriptive without multiplicity adjustment. The MRR by IRC test for noninferiority of zanubrutinib compared to ibrutinib was conducted with a noninferiority margin of 12% (H₀: MRR_A-MRR_B ≤ –12%, and H_a: MRR_A-MRR_B > –12%, where MRR_A is the MRR in the zanubrutinib arm and MRR_B is the MRR in the ibrutinib arm). The 95% CI for the Cochran-Mantel-Haenszel common risk difference was to be constructed with normal approximation and standard error based on Sato (1989) with strata similar to those described earlier for the primary end point of VGPR or CR. If the lower bound of the CI was greater than the noninferiority margin, the null hypothesis would be rejected, concluding that the MRR in zanubrutinib is noninferior to the MRR in ibrutinib. In addition, as a sensitivity analysis, the Cochran-Mantel-Haenszel common risk difference was to be estimated using the null variance estimator (Klingenberg [2013]).

Time to next treatment was summarized descriptively using the Kaplan–Meier (KM) method. Time to next treatment for patients without subsequent anticancer therapy was censored at the date of the patient’s last available information. PFS was analyzed at the time of the primary analysis of VGPR or CR rate, which was approximately 4 years after the first patient was randomized. The KM method was used; PFS was right-censored for patients who met 1 of the following criteria: no baseline disease assessment; started a new anticancer therapy before disease progression/death; experienced disease progression or death immediately after or more than 6 months since the last disease assessment (more than 12 months if a patient was on a response assessment schedule of every 24 weeks); alive without documentation of disease progression. Two-sided 95% CIs for median PFS were estimated using the Brookmeyer and Crowley method. OS was analyzed using methods similar to those described for PFS. Patients who remained alive as of the data cut-off date or who discontinued the study due to reasons other than death were right-censored at the date on which the patient was last known to be alive. The analysis of DoR conducted similarly to that of PFS. DoR was not compared between the 2 treatment arms. The difference in the resolution of any and all treatment-precipitating symptoms between zanubrutinib and ibrutinib was tested using a Chi-square test. The number and percentage of patients with the resolution of each and all symptoms were summarized.

The EORTC QLQ-C30 and EQ-5D were summarized for each assessment point. For the EORTC QLQ-C30, the percentage of patients with clinically meaningful changes from baseline in “global health status/QoL” and functional domains was summarized as “improved,” “stable,” or “worsened,” according to the scale’s scoring manual. The minimally important difference (MID) used to determine these categories was not reported. In addition to descriptive analysis, the QLQ-C30 global health status/QoL scale scores were compared between treatment groups in cohort 1 using a linear mixed effects model for repeated measures. LS means and standard errors for the difference between treatment arms were reported. The EuroQol 5-Dimensions 5-Levels (EQ-5D-5L) comprised a descriptive system and the EuroQol Visual Analogue Scale (EQ VAS) with 5 dimensions (mobility, self-care, usual activities, pain/discomfort, and anxiety/depression). The EQ VAS was summarized descriptively. Descriptive statistics were also performed, including the number and percentage of patients reporting each level of problem on each dimension of the EQ-5D.

Medical resource utilization, including the number of hospitalizations, planned and unplanned hospital visits, lengths of hospital stays, and supportive care (e.g., transfusions, growth factor support, IV antibiotics) were summarized at each cycle for each treatment arm in cohort 1.

All analyses were performed on data collected through a cut-off date of August 31, 2019.

Subgroup Analyses

The proportion of patients in cohort 1 who achieved a VGPR or CR were assessed in the following subgroups: sex (male versus female), age (≤ 65 years versus > 65 years; > 75 years versus ≤ 75 years), geographic region (Australia or New Zealand versus Europe versus North America), number of prior lines of therapy (0 versus 1 to 3 versus ≥ 3 and R/R versus treatment-naive), baseline ECOG PS (0 versus ≥ 1), baseline CXCR4 mutation status by Sanger method (warts, hypogammaglobulinemia, infections, and myelokathexis versus wild type or missing), baseline IgM level (≤ 40 g/L versus > 40 g/L), baseline beta2-microglobulin level (≤ 3 mg/L versus 3 mg/L), baseline hemoglobin concentration (≤ 110 g/L versus > 110 g/L), baseline platelet count (≤ 100 × 109/L versus > 100 × 109/L), baseline presence of extramedullary disease (yes versus no), and IPSSWM score (low versus intermediate versus high).

Handling of Dropouts or Missing Data

Missing data were not imputed unless otherwise specified. Missing dates or partially missing dates were not imputed as data level for prior or concomitant medications or procedures, new anticancer therapies, AEs, and deaths.

Results

Patient Disposition

In cohort 1 (MYD88^L265P), 201 patients were randomized: 99 in the ibrutinib arm and 102 in the zanubrutinib arm. A total of 37 patients (18.4%) were unfit, treatment-naive (18 in the ibrutinib treatment arm and 19 in the zanubrutinib treatment arm), and 164 patients (81.6%) were R/R (81 in the ibrutinib treatment arm and 83 in the zanubrutinib treatment arm). Two R/R patients were randomized but not treated: 1 in the zanubrutinib treatment arm due to an AE (unrelated to screening procedures) and 1 in the ibrutinib treatment arm due to PD (Bing Neel syndrome). As of the data cut-off date, 158 patients (78.6%) were continuing study treatment (77 patients [77.8%] in the ibrutinib treatment arm and 81 patients [79.4%] in the zanubrutinib treatment arm).

The most common reasons for discontinuation of study treatment were AEs (9 ibrutinib-treated patients [9.1%] versus 4 zanubrutinib-treated patients [3.9%]) and PD (5 ibrutinib-treated patients [5.1%] versus 7 zanubrutinib-treated patients [6.9%]). At the time of the primary analysis, 92% of the patients in the R/R analysis set had at least 15 months of follow-up. Overall, the median follow-up times on study for patients treated with ibrutinib and zanubrutinib were 19.38 months and 19.47 months, respectively (Table 10).

Of the 28 patients enrolled in cohort 2 (MYD88^WT) and treated with zanubrutinib (5 unfit, treatment-naive and 23 R/R), 17 patients (60.7%) remained on treatment as of the data cut-off date. The most common reasons for discontinuation of study treatment were PD (6 patients [21.4%]) and AEs (2 patients [7.1%]). The median follow-up time on study was 17.8 months in this group (Table 11).

Exposure to Study Treatments

In cohort 1, the overall median treatment durations were 18.5 months and 18.7 months for the ibrutinib and zanubrutinib treatment arms, respectively; the median relative dose intensities were 98.1% and 97.6%, respectively. For R/R patients, the median treatment durations were 17.99 months and 18.00 months in the ibrutinib and zanubrutinib treatment arms, respectively, with median relative dose intensities of 98.14% and 97.73%, respectively. The median treatment durations for treatment-naive patients were 20.73 months and 21.45 months in the ibrutinib and zanubrutinib treatment arms, respectively, with median relative dose intensities of 98.76% and 97.58%.

Concomitant Medications

In cohort 1, almost all patients received greater than or equal to 1 concomitant medication(s) (98% in the ibrutinib arm and 96% in zanubrutinib arm). The most common concomitant medications in both arms were antibacterial drugs for systemic use (74 patients [75.5%] in the ibrutinib treatment arm and 69 patients [68.3%] in the zanubrutinib treatment arm) and analgesics (44 patients [44.9%] in the ibrutinib treatment arm and 46 patients [45.5%] in the zanubrutinib treatment arm). This was followed by drugs for acid-related disorders (37 patients [37.8%] in the ibrutinib arm and 35 patients [34.7%] in the zanubrutinib arm) and antithrombotic drugs (32 patients [32.7%] in the ibrutinib arm and 37 patients [36.6%] in the zanubrutinib arm). Antidiarrheals and intestinal anti-inflammatory or anti-infective drugs were used by 19 patients (19.4%) in the ibrutinib arm and by 3 patients (3.0%) in the zanubrutinib arm, consistent with the increased frequency of diarrhea in the former group. Antianemic preparations were used by 14 patients (14.3%) and 32 patients (31.7%) in the ibrutinib and zanubrutinib treatment arms, respectively, consistent with lower hemoglobin levels at baseline in the zanubrutinib-treated patients.

Protocol Violations

Overall, 7 patients (3.5%) had major protocol violations: study assessment or procedures (n = 5), hepatitis B or C testing not assessed (n = 5), or disallowed medication (n = 1).

Efficacy

The median follow-up time of patients was 19.4 months in cohort 1 and 17.8 months in cohort 2.

Time to Next Treatment

Cohort 1: MYD88^L265P

In cohort 1, 9 patients in the ibrutinib arm and 6 patients in the zanubrutinib arm started non-protocol anticancer therapy. The median time to initiation of non-protocol anticancer therapy were 6.44 months in the ibrutinib treatment arm and 6.83 months in the zanubrutinib treatment arm (Table 12).

Cohort 2: MYD88^WT

In cohort 2, 3 patients (1 unfit, treatment-native and 2 R/R) started non-protocol anticancer therapy with a median time to initiation of 3.61 months.

PFS

Cohort 1: MYD88^L265P

In cohort 1, the median IRC-assessed PFS was not reached in either treatment arm. The event-free rates at 12 months for patients in the ibrutinib and zanubrutinib treatment arms were 87.2% (95% CI, 78.6% to 92.5%) versus 89.7% (95% CI, 81.7% to 94.3%), respectively, and were 83.8% (95% CI, 74.5% to 89.9%) versus 85.0% (95% CI, 75.2% to 91.2%) at 18 months. For patients with R/R disease in the ibrutinib and zanubrutinib treatment arms, the event-free rates at 12 months were 85.9% (95% CI, 75.9% to 91.9%) versus 92.4% (95% CI, 83.8% to 96.5%), respectively, and 81.7% (95% CI, 71.1% to 88.8%) versus 85.9% (95% CI, 73.7% to 92.7%) at 18 months (Table 13 and Figure 3).

The median PFS assessed by the investigator was not reached in either treatment arm. The event-free rates at 12 months for patients in the ibrutinib and zanubrutinib treatment arms were 90.4% (95% CI, 82.4% to 94.9%) versus 93.9% (95% CI, 87.0% to 97.2%) respectively, and 87.1% (95% CI, 78.4% to 92.5%) versus 88.9% (95% CI, 79.3% to 94.2%) at 18 months.

Cohort 2: MYD88^WT

In cohort 2, the event-free rate was 72.4% (95% CI, 50.6% to 85.8%) at 12 months and 68.1% (95% CI, 46.2% to 82.6%) at 24 months (Table 14 and Figure 4). In patients overall in cohort 2, the event-free rate as assessed by the investigator was 69.0% (95% CI, 47.5% to 83.2%) at 12 months and 64.7% (95% CI, 43.0 to 79.9%) at 18 months.

Figure 3: Kaplan–Meier Plot of IRC-Assessed PFS (Cohort 1: MYD88^L265P) (ITT Analysis Set)

IRC = Independent Review Committee; ITT = intention to treat; no. = number.

Source: Clinical Study Report for Brukinsa.⁷

Figure 4: Kaplan–Meier Plot of IRC-Assessed PFS (Cohort 2: MYD88^WT) (Efficacy Analysis Set)

CI = confidence interval; IRC = Independent Review Committee; no. = number.

Source: Clinical Study Report for Brukinsa.⁷

Overall Survival

Cohort 1: MYD88^L265P

In cohort 1, the median OS was not reached in either treatment arm (Table 15 and Figure 5). By the data cut-off date, 8 deaths had occurred in the ibrutinib arm, and 6 deaths had occurred in the zanubrutinib arm. The event-free rates for patients in the ibrutinib versus zanubrutinib treatment arms were 93.9% (95% CI, 86.8% to 97.2%) versus 97.0% (95% CI, 90.9% to 99.0%) at 12 months, and 92.8% (95% CI, 85.5% to 96.5%) versus 97.0% (95% CI, 90.9% to 99.0%) at 18 months.

Cohort 2: MYD88^WT

In cohort 2, the median OS was not reached (Table 16 and Figure 6). The event-free rates were 96.2% (95% CI, 75.7% to 99.4%) at 12 months and 87.8% (95% CI, 66.7% to 95.9%) at 18 months.

Figure 5: Kaplan–Meier Plot of Overall Survival (Cohort 1: MYD88^L265P) (ITT Analysis Set)

ITT = intention to treat; no. = number.

Source: Clinical Study Report for Brukinsa.⁷

Figure 6: Kaplan–Meier Plot of Overall Survival (Cohort 2: MYD88^WT) (Efficacy Analysis Set)

CI = confidence interval; no. = number.

Source: Clinical Study Report for Brukinsa.⁷

Overall Response

Cohort 1: MYD88^L265P

No patients achieved a CR. In cohort 1, the IRC-assessed VGPR or CR rates in the ibrutinib and zanubrutinib arms were 19.2% (95% CI, 12.0% to 28.3%) and 28.4% (95% CI, 19.9% to 38.2%), respectively. In R/R patients, the IRC-assessed VGPR or CR rate was 19.8% (95% CI, 11.7% to 30.1%) in the ibrutinib arm and 28.9% (95% CI, 19.5% to 39.9%) in the zanubrutinib arm (P = 0.11). In unfit, treatment-naive patients, the IRC-assessed VGPR or CR rate was 16.7% (95% CI to 3.6%, 41.4%) in the ibrutinib arm and 26.3% (95% CI, 9.1% to 51.2%) in the zanubrutinib arm (Table 17).

A post hoc analysis showed that noninferiority was demonstrated under the noninferiority margin of –4.5%, which was determined before the unblinded analysis being performed. The lower bound of the 95% CI for the VGPR or CR rate difference assessed by the IRC was –2.5% in the R/R analysis set in cohort 1. If the study end point was not changed from noninferiority of VGPR or CR to superiority of VGPR or CR, noninferiority of VGPR or CR in zanubrutinib-treated patients compared to ibrutinib would have been met.

Cohort 2: MYD88^WT

In cohort 2, no patients achieved a CR. The IRC-assessed VGPR or CR rate was 26.9% (95% CI, 11.6% to 47.8%) (Table 18).

Duration of Response

Cohort 1: MYD88^L265P

In cohort 1, the median durations of CR or VGPR and MRR had not been reached overall or for R/R patients in either treatment arm who achieved a response to study treatment. Four events occurred in patients with VGPR or CR in the ibrutinib arm, and 1 event occurred in patients with VGPR or CR in the zanubrutinib arm. Among patients who achieved a major response, 9 events occurred in the ibrutinib arm, and 6 events occurred in the zanubrutinib arm. The event-free rates at 12 months and 18 months for patients in the ibrutinib arm who achieved a major response were 87.9% (95% CI, 77.0% to 93.8%) and 87.9% (95% CI, 77.0% to 93.8%), respectively. For those in the zanubrutinib arm, the event-free rates at 12 months and 18 months were 94.4% (95% CI, 85.8% to 97.9%) and 85.2% (71.7% to 92.6%), respectively (Table 19).

Cohort 2: MYD88^WT

In cohort 2, in patients who achieved a response to study treatment, the median duration of major response had not been reached. The event-free rates at 12 months and 18 months for patients who achieved a major response were 62.3% (95% CI, 27.7% to 84.0%) and 62.3% (95% CI, 27.7% to 84.0%), respectively (Table 20).

Health-Related Quality of Life

In cohort 1, HRQoL measures on average increased numerically during the trial observation period in both treatment arms (Figure 7 and Figure 8). The LS mean for the EORTC QLQ-C30 global health status/QoL was 69.0 (standard error = 2.3) in the ibrutinib arm and 68.3 (standard error = 2.2) in the zanubrutinib arm (difference = –0.69; 95% CI, –4.95 to 3.57). The mean change in EQ-5D score from baseline was 9.0 (SD = 17.90) in the ibrutinib arm and 13.7 (SD = 14.66) in the zanubrutinib arm (at cycle 13, day 1).

Figure 7: EORTC QLQ-C30 Global Health Status: Change From Baseline Over Time (Cohort 1: MYD88^L265P) (ITT Analysis Set)

CI = confidence interval; EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; ITT = intention to treat.

Note: A high score for global health status represents a high quality of life.

Source: Clinical Study Report for Brukinsa.⁷

Figure 8: EQ-5D Score — Change From Baseline Over Time (Cohort 1: MYD88^L265P) (ITT Analysis Set)

CI = confidence interval; EQ-5D = EuroQol 5-Dimensions; ITT = intention to treat.

Source: Clinical Study Report for Brukinsa.⁷

Cancer-Related Symptoms

Cohort 1: MYD88^L265P

In cohort 1, the proportions of patients with resolution of all treatment-precipitating symptoms in the ibrutinib and zanubrutinib treatment arms were 68.0% and 68.8%, respectively. Most patients had resolution of any treatment-precipitating symptoms (92.8% and 94.8%, respectively). The proportions of patients in the zanubrutinib versus ibrutinib treatment arm with WM-related symptoms at baseline who achieved a resolution were 79.3% versus 86.0% for fatigue, 95.8% versus 87.5% for hemoglobin of less than or equal to 10 g/dL, 97.1% versus 96.2% for B symptoms, 96.3% versus 96.3% for hyperviscosity, 58.3% versus 47.6% for peripheral neuropathy, 83.3% versus 44.4% for amyloidosis, and 75.0% versus 58.3% for and platelet count of less than 100 × 10⁹/L. In R/R patients, 64.6% and 70.5% of patients in the ibrutinib and zanubrutinib treatment arms, respectively, had resolution of all treatment-precipitating symptoms.

Cohort 2: MYD88^WT

In cohort 2, 68.0% of patients overall had resolution of all treatment-precipitating symptoms. Most patients had resolution of any treatment-precipitating symptoms (92.0%). Overall, resolution of fatigue, hyperviscosity, and peripheral neuropathy due to WM were obtained in 86.7%, 83.3%, and 33.3% of patients, respectively.

Health Care Resource Utilization

In cohort 1, 23.9% of patients had emergency room and hospital admissions. Planned admissions for 17.9% of patients overall were for reasons that are expected and commonly encountered in elderly patients with WM. A total of 220 hospital visits were reported with a mean duration of 6.4 days (5.8 days and 6.9 days for the ibrutinib-treated and zanubrutinib-treated patients, respectively) (Table 21). Seventeen patients in the ibrutinib arm and 19 patients in the zanubrutinib arm had 35 and 43 planned admissions, respectively. (Two events of granulocyte colony-stimulating factor self-administration for zanubrutinib were not considered admissions.) Of these planned admissions, 20 (57%) for patients in the ibrutinib arm and 31 (72%) for patients in the zanubrutinib arm resulted in discharge on the day of admission. The remaining planned admissions were generally short and/or in line with the reason for admission, such as for knee replacement surgery and rehab. Forty-three admissions (55%) were for (or, on 1 occasion, involved) red blood cell transfusions, platelet transfusions, growth factor administration, and/or IV immunoglobulin infusion. Six admissions (8%) can be considered disease-related (such as for plasmapheresis or bone marrow and trephine), and 30 admissions (38%) (1 involving a platelet transfusion) were for other reasons that can be considered expected and typical for elderly patients in general, such as bladder neoplasm or total hip arthroplasty.

Subgroups

Results of subgroup analyses were consistent with the primary analysis for VGPR or CR. The proportions of patients in cohort 1 who achieved a VGPR or CR were similar in each treatment arm for the subgroups of interest identified in the systematic review protocol (Table 22).

Harms

In cohort 1, nearly all patients (97 [99.0%] of ibrutinib-treated patients and 98 [97.0%] of zanubrutinib-treated patients) had at least 1 AE or TEAE; AEs greater than or equal to grade 3 were reported in 62 patients (63.3%) and 59 patients (58.4%) in the ibrutinib and zanubrutinib treatment arms, respectively (Table 22). Among all zanubrutinib-treated patients (cohort 1 and cohort 2), 122 (94.6%) had at least 1 AE, including 77 patients (59.7%) who had AEs greater than or equal to grade 3 (Table 24).

Serious Adverse Events

In cohort 1, SAEs were reported in 40 patients (40.8%) the ibrutinib treatment arm and in 40 patients (39.6%) in the zanubrutinib treatment arms (Table 22). The most common SAE in the ibrutinib treatment arm was pneumonia (9 patients [9.2%]), followed by pyrexia and sepsis (each reported by 3 patients [3.1%]). The most common SAEs in the zanubrutinib treatment arm were febrile neutropenia, influenza, and neutropenia (each reported by 3 patients [3.0%]). Among all zanubrutinib-treated patients, pneumonia was more common in cohort 2 compared with cohort 1 (10.7% versus 1.0%). Of note, cohort 2 had an older patient population and a smaller sample size than cohort 1 (Table 23).

Withdrawals Due to Adverse Events

In cohort 1, 9 patients (9.2%) in the ibrutinib arm and 4 patients (4.0%) in the zanubrutinib treatment arm had AEs leading to study treatment discontinuation. Five patients (5.1%) in the ibrutinib treatment arm had AEs leading to study treatment discontinuation that were assessed as related to ibrutinib (1 each of drug-induced liver injury, hepatitis, interstitial lung disease, pneumonia, and pneumonitis); and 2 patients (2.0%) in the zanubrutinib treatment arm had AEs leading to study treatment discontinuation that were assessed as related to zanubrutinib (1 each of neutropenia and cardiomegaly). In cohort 2, 2 patients (7.1%) had AEs leading to study treatment discontinuation that were assessed as related to zanubrutinib treatment (1 each of subdural hemorrhage and diarrhea).

Deaths

In cohort 1, a total of 7 patients (7.1%) in the ibrutinib treatment arm and 6 patients (5.9%) in the zanubrutinib treatment arm died (as recorded in the death electronic case report form) at the time of the data cut-off date. Five patients (5.1%) in the ibrutinib arm and 1 patient (1.0%) in the zanubrutinib arm died within 30 days of the last dose of study drug. PD was the most common cause of death in the zanubrutinib treatment arm, reported in 3 patients (3.0%). Deaths due to AEs occurred in 4 ibrutinib-treated patients (4.1%) and 1 zanubrutinib-treated patient (1.0%); all 5 deaths due to AEs occurred within 30 days of the last dose date. The 4 deaths due to AEs in the ibrutinib arm were attributed to bacterial sepsis, acute cardiac failure, sepsis, and, in 1 case, unknown cause; none of these were judged to be related to ibrutinib treatment. The 1 death due to an AE in the zanubrutinib arm was assessed as related to zanubrutinib treatment. The death occurred in an |||||||||||||||||| patient with R/R WM and was from cardiomegaly approximately 3 months after initiating zanubrutinib. The patient was reported to have had a history of hypertension, aortic stenosis, chronic inflammatory demyelinating polyneuropathy, and multiple plasmapheresis.

In cohort 1, the most common AEs in the ibrutinib arm were diarrhea (31 patients [31.6%]), upper respiratory tract infection (28 patients [28.6%]), and contusion and muscle spasms (23 patients [23.5%] each). In the zanubrutinib arm, the most common AEs were neutropenia (25 patients [24.8%]), upper respiratory tract infection (24 patients [23.8%]), and diarrhea (21 patients [20.8%]). The following AEs were greater (> 10% difference) in the ibrutinib treatment arm versus the zanubrutinib treatment arm: muscle spasms (23.5% versus 9.9%), atrial fibrillation (14.3% versus 2.0%), diarrhea (31.6% versus 20.8%), contusion (23.5% versus 12.9%), peripheral edema (19.4% versus 8.9%), pneumonia (12.2% versus 2.0%). Neutropenia was observed in a higher proportion of patients in the zanubrutinib arm versus the ibrutinib treatment arm (24.8% versus 12.2%). Among all zanubrutinib-treated patients, the incidences of AEs were generally comparable between cohort 1 and cohort 2. Exceptions (> 10% difference) in cohort 1 included neutropenia (24.8% versus 14.3%, respectively), nausea (14.9% versus 3.6%), and dyspnea (13.9% versus 3.6%); exceptions (> 10% difference) in cohort 2 included pneumonia (2.0% versus 14.3%), respiratory tract infection (5.9% versus 17.9%), and decreased appetite (4.0% versus 14.3%) (Table 25).

Notable Harms

Neutropenia: In cohort 1, neutropenia was reported in 12 patients (12.2%) in the ibrutinib arm and 25 patients (24.8%) in the zanubrutinib arm. In cohort 2, neutropenia was observed in 4 patients (14.3%). Neutropenia was observed in 29 out of 129 patients (22.5%) among all zanubrutinib-treated patients in both cohorts. Neutropenia greater than or equal to grade 3 was reported in 8 patients (8.2%) in the ibrutinib arm and 16 patients (15.8%) in the zanubrutinib treatment arms. Among all zanubrutinib-treated patients in both cohorts, neutropenia greater than or equal to grade 3 was reported in 19 out of 129 patients (14.7%).

Hemorrhage (including minor and major bleeding): In cohort 1, 58 patients (59.2%) in the ibrutinib arm and 49 patients (48.5%) in the zanubrutinib arm had hemorrhage (including minor bleeds involving mucous membranes and skin). The predominant events reported in the ibrutinib treatment arm versus the zanubrutinib treatment arm were mild or moderate mucocutaneous bleeding (i.e., contusion [23.5% versus 12.9%], epistaxis [19.4% versus 12.9%], hematuria [10.2% versus 6.9%], hematoma [7.1% versus 5.0%], petechiae [3.1% versus 6.9%], and purpura [6.1% versus 3.0%]). No patients in the ibrutinib treatment arm and 1 patient (1.0%) in the zanubrutinib treatment arm discontinued study treatment due to a hemorrhagic event. Forty-nine of the 58 ibrutinib-treated patients (84%) and 36 of the 49 zanubrutinib-treated patients (73%) had hemorrhage events that were considered related to treatment by the investigator.

Major hemorrhage, defined as serious (or ≥ grade 3) bleeding at any site or CNS bleeding of any grade, was observed in 9 patients (9.2%) in the ibrutinib arm and 6 patients (5.9%) in the zanubrutinib arm. The only major hemorrhages reported in more than 1 patient were hematuria and retinal hemorrhage (each occurring in 2 ibrutinib-treated patients). Eight ibrutinib-treated patients (8.2%) and 6 zanubrutinib-treated patients (5.9%) had major hemorrhage (grade ≥ 3). Six ibrutinib-treated patients (6.1%) and 5 zanubrutinib-treated patients (5.0%) had serious events of major hemorrhage. None of the major hemorrhage events were fatal. One patient in the zanubrutinib treatment arm discontinued study treatment for subdural hemorrhage. No patients in either treatment arm had a dose reduction for a major hemorrhagic event. Three patients (2 in the zanubrutinib arm [1.2%] and 1 in the ibrutinib arm [1.0%]) required transfusions for the management of major hemorrhage. The major hemorrhage events were considered related to treatment by the investigator in 6 out of 9 ibrutinib-treated patients and in 2 out of 6 zanubrutinib-treated patients. The median time to the first event of major hemorrhage was 258.0 days (range = 3 days to 539 days) for the ibrutinib arm and 289 days (range = 171 days to 537 days) for the zanubrutinib arm.

Cardiovascular events: Atrial fibrillation or flutter was reported in 14 patients (14.3%) in the ibrutinib arm and 2 patients (2.0%) in the zanubrutinib treatment arm. Atrial fibrillation greater than or equal to grade 3 was reported in 3 ibrutinib-treated patients (3.1%) and in 0 zanubrutinib-treated patients. None of the atrial fibrillation or flutter events led to death. Four patients (4.1%) in the ibrutinib arm had atrial fibrillation or flutter greater than or equal to grade 3; 0 patients in the zanubrutinib arm had this. No patient discontinued treatment due to atrial fibrillation or flutter. In the ibrutinib arm, 2 patients (2.0%) had dose reduction due to atrial fibrillation or flutter.

Second primary malignancy: In cohort 1, a second primary malignancy was reported in 11 patients (11.2%) in the ibrutinib arm: basal cell carcinoma (n = 2), squamous cell carcinoma (n = 4), Bowen’s disease (n = 1), skin cancer (n = 2), bladder transitional cell carcinoma (n = 2), and chronic myeloid leukemia (n = 1). In the zanubrutinib arm, 12 patients (11.9%) were observed to have a second primary malignancy, including basal cell carcinoma (n = 4), squamous cell carcinoma (n = 2), Bowen’s disease (n = 1), chronic myelomonocytic leukemia (n = 1), metastatic colorectal cancer (n = 1), endometrial adenocarcinoma (n = 1), malignant lung neoplasm (n = 1), malignant melanoma (n = 1), stage I malignant melanoma (n = 1), plasma cell myeloma (n = 1), and skin cancer (n = 1).

Critical Appraisal

Internal Validity

The objective of the ASPEN trial was to assess the efficacy of zanubrutinib compared to ibrutinib in R/R and treatment-naive patients with WM. If the study end point had not been changed from noninferiority of VGPR or CR to superiority of VGPR or CR, noninferiority of VGPR or CR in zanubrutinib-treated patients compared to ibrutinib-treated patients would have been met. Although this supports the primary efficacy analysis, the post hoc nature of this analysis is an inherent limitation. Thus, it can be considered as exploratory only.

Given that WM is a rare disease, the sample size was acceptable. Statistical power calculations were reported, and the target sample size (210 in cohorts 1 and 2 combined) was achieved. However, the number of treatment-naive patients was limited (37 in cohort 1). Randomization was stratified, and stratification was based on relevant prognostic factors, which included CXCR4^WHIM mutational status and prior lines of therapy; these were identified as subgroups of interest by the clinical experts consulted. Cohort 2 (MYD88^WT) was a non-randomized exploratory single arm; as such, the relative efficacy of zanubrutinib compared to ibrutinib in this population (which represents approximately 10% of the population of patients with WM) cannot be determined. The study was generally well balanced with respect to patient demographics and disease characteristics.

Because the ASPEN trial was an open-label study, access to aggregated data summaries with actual study treatment assignment of the randomized arms while the study was ongoing may have introduced unwanted bias due to the possibility of inconsistent queries among patients with different treatments, or over-interpretation of immature, accruing data. A Data Integrity Protection Plan was put in place to describe the steps taken before database lock for the primary analysis of efficacy to minimize these potential biases for the randomized portion of the study.⁷ The primary end point was assessed by an IRC, which reduces bias related to outcome assessment. Additionally, there are other important sources of bias that may result from lack of blinding of patients and investigators to study treatments. Patients’ knowledge of their assigned treatment may have affected some safety end points, and different supportive care may have been offered to patients in the 2 treatment arms.

The primary end point and key secondary end points were appropriate and adequately described. Data were immature for time-to-event outcomes, and median PFS and OS were not reached in either treatment arm. Given that the ASPEN trial is ongoing, future analyses may be more informative with respect to time-to-event outcomes. In addition to PFS and OS, time to next treatment and HRQoL were also identified in the systematic review protocol as important efficacy outcomes. However, these were studied as exploratory outcomes in the ASPEN trial, which limits the interpretation of results. Furthermore, the instruments used to assess HRQoL (QLQ-C30 and EQ-5D) have not been validated in the WM patient population, and no information regarding their reliability, responsiveness, or MID could be identified in this population (Appendix 3). It is unclear what threshold was used to determine clinically meaningful change in HRQoL in the ASPEN trial. Overall, the exploratory nature of important end points is a major limitation of the presented evidence.

External Validity

The clinical experts, clinician groups, and drug plans noted that ibrutinib is not an appropriate comparator for zanubrutinib in Canadian clinical practice. Relevant comparators for WM in Canadian jurisdictions include rituximab-based chemotherapy for treatment-naive patients and those with relapsed disease. Re-treatment with rituximab is funded for patients with a relapse-free interval (6 months to 12 months, depending on jurisdiction) following the last dose of rituximab. Therefore, relevance to the current clinical setting is limited, and the question of the comparative efficacy and safety of zanubrutinib versus current standard of care in Canada cannot be answered (see the Indirect Evidence section).

The ASPEN trial defined treatment-naive patients as those unsuitable for standard chemoimmunotherapy, based on the judgment of the study investigators; no explicit criteria were used to define this patient population. This definition of treatment-naive does not align with the standard definition of treatment-naive in oncology research and practice. A treatment-naive patient is generally defined as a patient with no prior anticancer therapy. It is not based on the suitability of a patient to receive treatment. The reasons cited for considering patients unsuitable for standard chemoimmunotherapy (e.g., age, hypertension, ischemic heart disease) were not considered by the clinical experts consulted by CADTH to be key factors that guide treatment decision in this population, particularly age. The clinical experts consulted by CADTH deemed the methods and criteria use in the ASPEN trial to determine suitability for standard chemoimmunotherapy to be insufficient for deciding which patients are truly ineligible for chemoimmunotherapy. Therefore, the trial evidence regarding the efficacy and safety of zanubrutinib compared to ibrutinib in unfit, treatment-naive patients is insufficient to guide treatment decisions in truly treatment-naive patients in clinical practice.

The inclusion criteria used in the ASPEN trial were generally reasonable, based on the intended patient population. However, the clinical experts consulted by CADTH noted the exclusion of patients with CNS involvement in the ASPEN trial. According to the clinical experts, patients with CNS involvement, namely Bing Neel syndrome, may in fact benefit from zanubrutinib early in the course of their disease. Indeed, ibrutinib is a well-established therapy for Bing Neel syndrome. The exclusion of patients with Bing Neel syndrome was justified during the planning phase of the trial because no comprehensive guidelines existed at the time for the diagnostic and therapeutic approach or response assessment of Bing Neel syndrome. (Guidelines were published in 2017, around the time the first patient in the trial was dosed.) Given that Bing Neel syndrome was considered a separate disease entity within WM, with a unique clinical course and management challenges, patients with CNS involvement were excluded to ensure the enrolment of a fairly homogenous patient population. Other criteria to define trial populations — including R/R — and definitions of outcomes (e.g., CR, PR) were comparable to the standard definitions used in clinical practice.

Indirect Evidence

Objectives and Methods for the Summary of Indirect Evidence

The objective of this section is to provide an appraisal and summary of indirect evidence from the sponsor-submitted ITC comparing zanubrutinib to chemotherapy regimens. There is no standard of care for the treatment of R/R WM. BTK inhibitors — such as ibrutinib, the comparator in the ASPEN trial — are most frequently used in Canadian clinical practice through compassionate access for the treatment of WM. As such, the sponsor submitted an ITC to provide an assessment of the relative efficacy of zanubrutinib compared to chemotherapeutic regimens currently funded by Canadian public plans.²⁴

The literature search identified 1 ITC; however, it did not meet the eligibility criteria. The sponsor-submitted ITC, which was used to inform the pharmacoeconomic model, was appraised and summarized.²⁴

Description of Indirect Comparison(s)

The sponsor submitted a MAIC based on an SLR that compared the IPD of the zanubrutinib arm of the ASPEN trial to the populations of relevant trial reports for chemotherapy-based regimens in adult patients with treatment-naive or R/R WM.²⁴

Methods of the Sponsor-Submitted ITC

Objectives

The objective of the sponsor-submitted report was to conduct a systematic review of the literature and perform an ITC to provide an assessment of the relative efficacy of zanubrutinib compared to chemotherapeutic regimens currently funded by Canadian public plans.²⁴

Study Selection Methods

A systematic literature search was conducted in September 2020 using Embase, MEDLINE and MEDLINE In-Process (through ProQuest), and the Cochrane Library (using wiley.com and including the Cochrane Database of Systematic Reviews, the Database of Abstracts of Reviews of Effectiveness, and the Cochrane Central Register of Controlled Trials). The study selection criteria for the review are summarized in Table 26. Briefly, eligible patients were informed by the ASPEN trial, and included treatment-naive (i.e., first-line) patients with WM for whom chemoimmunotherapy is unsuitable, or patients with R/R WM who had received at least 1 prior therapy (i.e., second-line patients). Relevant comparators were identified through a systematic search of clinical practice guidelines, a prior CADTH review, and consultation with clinical experts.²⁴

Relevant outcomes for the ITC were narrower than for the systematic review, which included a broad range of efficacy and safety outcomes. PFS and OS were the primary outcomes of interest for the ITC.²⁴

No date restriction was applied to the search.²⁴

Primary screening of study titles and abstracts was performed by 2 independent reviewers, who applied the basic study selection criteria (population, intervention, and study design). Full-text articles were obtained for potentially relevant studies identified by primary screening, and secondary screening was performed by 2 independent reviewers against the same eligibility criteria. Uncertainty regarding the inclusion of studies during primary or secondary screening was checked and judged by a third reviewer.²⁴

Data extraction was performed by a single reviewer using a predefined data extraction template, and data were quality checked by a separate reviewer against the source publication. For baseline patient characteristics and AE incidence, summary mean estimates were extracted from comparator trial publications whenever available. Individual patient-level event and censoring times for survival were derived through a 2-step process for OS and PFS KM curves. First, the numerical values of the curves (i.e., time on the x-axis and proportion of patients alive on the y-axis) were obtained through graphical digitization using WebPlotDigitizer . Second, the number of events and censoring at each time point were manually calibrated to create a “simulated” trial population that would reproduce the KM curves presented in trial publications, based on the reported number of patients at risk and/or the marker for censoring on the KM curves.²⁴

No formal approach was taken to assess the risk of bias in the included studies.²⁴

ITC Analysis Methods

A feasibility assessment was conducted to determine whether an NMA was possible by determining the appropriate approach based on clinical input and availability of data, including: anchored comparisons (i.e., common comparator arms); evaluable PFS and OS KM curves; baseline patient characteristics; study design features (prospective, sample size); and geographic location. Study design features considered the size of the trial (larger, better), whether an anchored comparison was possible, and whether outcomes were collected prospectively. Geographic location considered whether treatment patterns or populations would be expected to be similar to the Canadian setting. The final study selection for the ITC was conducted through an informal assessment by Canadian and clinical experts and the sponsor personnel.²⁴

A summary of the analysis methods for the MAIC is shown in Table 26. A MAIC approach was used for this comparison because an NMA based on aggregate-level data was determined to be infeasible due to a lack of network connectivity. First, the patients from the ASPEN trial who did not qualify for comparable studies were excluded based on the selection criteria from the comparator study. Second, the IPD for zanubrutinib obtained from the ASPEN trial were reweighted such that the weighted mean baseline characteristics were similar to those reported in the comparator publications. Finally, the effect estimates for outcomes of interest were weighted using the generated weights. In the process of adjustment, each patient was assigned a weight representing the inverse of the odds of being in the ASPEN trial’s zanubrutinib arm versus being in a specific comparator trial. Baseline characteristics of patients before and after weighting were provided (Table 31 and Table 32).

Based on input from the clinical experts consulted by the sponsor — and from the ASPEN trial and other published literature — the following were considered to be potential prognostic factors or effect modifiers: age (≤ 75 years versus > 75 years; ≤ 65 years, and 66 to 75 years versus > 75 years); number of prior therapies (0 lines to 3 lines versus > 3 lines; 1 line to 3 lines versus 3 lines); ECOG PS (0 to 1 versus > 1); MYD88/CXCR4 mutation status; IgM concentration (≤ 40 g/L versus > 40 g/L); beta2-microglobulin concentration (≤ 3 versus > 3 mg/L); platelet count (≤ 100 versus > 100 × 10⁹/L); hemoglobin concentration (≤ 110 g/L versus > 110 g/L); presence of extramedullary disease; and WM IPSS. Not all variables were included in each comparison, and the reason for variable exclusion was not provided; however, it was presumed to be due to lack of available data.

For the unanchored MAIC, inferences on the comparisons were performed using the “sandwich” package in R.²⁴

Two MAIC analyses using zanubrutinib IPD from the ASPEN trial were conducted during this analysis: 1 comparing zanubrutinib to the single-arm trial (Tedeschi et al. [2015]) for BR and another comparing zanubrutinib to the single-arm trial for DRC (Dimopoulos et al. [2007] and Kastritis et al. [2015]). (The Dimopoulos and Kastritis publications describe the same study) The primary outcome measures were investigator-assessed PFS and OS. OS and PFS with zanubrutinib compared to other relevant therapies were assessed by estimating HRs using Cox proportional hazard (PH) models. The PH assumption was assessed through visual inspection of the log-cumulative hazard plots for PFS and OS.²⁴ No further information on the approach used to estimate the parameters of the model, or the choice of outcomes for the MAIC, was provided.

Results of ITC

Summary of Included Studies

In total, 1,351 records were identified from database searches. After the removal of duplicates, 1,118 abstracts were screened for eligibility, resulting in 1,021 exclusions. A total of 97 publications were assessed for eligibility for the SLR based on full texts, with 33 publications included.²⁴ No quality assessment of the included studies was conducted.

Of the 33 studies included in the systematic review, 7 consisted of ibrutinib monotherapy and did not inform comparisons for the ITC, and 2 publications were for zanubrutinib compared to ibrutinib (i.e., the ASPEN trial; however, the ibrutinib arm was not included in the ITC). Of the 24 remaining trials, 2 were included in the MAIC: Tedeschi et al. (2015), which evaluated BR, and Dimopoulos at al. (2007) and Kastritis et al. (2015), which concern the same trial evaluating DRC.²⁴ A detailed summary of the 24 eligible studies is provided in Table 28.

Of the 4 studies identified for BR, Tedeschi et al. (2015) was considered by the sponsor to be the most suitable for inclusion in the MAIC because²⁴:

• It reported evaluable PFS KM and OS KM for incorporation into the cost-effectiveness analysis.

• The results in this study are representative of other studies of BR, which have consistently shown it to be at least as effective as other chemoimmunotherapeutic regimens.

• It reported baseline patient characteristics, which supported further comparability assessment and potential matching adjustments.

• It had the largest sample size.

Of the 5 articles identified for DRC, 2 articles (Dimopoulos et al. [2007] and Kastritis et al. [2015]) describing the same study were considered by the sponsor for inclusion in the MAIC because²⁴:

• The study was the only prospective study.

• It reported an evaluable OS KM, over the long-term, for incorporation into the cost-effectiveness analysis.

• It reported baseline patient characteristics, which supported further comparability assessment and potential matching adjustments.

• It had the largest sample size.

Of the 6 studies identified for fludarabine-cyclophosphamide-rituximab or fludarabine-rituximab, no study was considered by the sponsor for inclusion in the ITC because of the relatively small sample sizes and a lack of reporting of OS KM curves or PFS KM curves. For cladribine-rituximab and best supportive care, no study was identified in the SLR. One study of rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP) was identified; however, it was a small cohort of R/R patients with WM (n = 10/13). Therefore, these comparators, although listed in the final scope, were not included in the ITC or the cost-effectiveness analysis.²⁴

For comparators for adults with WM whose disease is untreated and for whom chemoimmunotherapy is unsuitable, several studies were identified. However, these trials were not considered for the ITC because of the small sample size of treatment-naive patients (unsuitable for chemoimmunotherapy) in the zanubrutinib arm in the ASPEN trial (n = 19), which made it infeasible to match to the comparator populations.²⁴ Thus, patients with treatment-naive disease were not analyzed in a separate MAIC. Instead, they were included in the overall ASPEN trial population in both MAIC scenarios.

Study and baseline characteristics of the 3 included studies are summarized in Table 28. Study phase, countries included, dosing regimen and administration, and follow-up times varied across the studies. Both R/R and treatment-naive patients were evaluated in the included studies; where the ASPEN trial included both R/R and treatment-naive patients, Tedeschi et al. (2015) included only R/R patients with WM, and Dimopoulos et al. (2007) and Kastritis et al. (2015) included only treatment-naive patients. Included interventions were all different, with differences in administration method and dosing schedule. The ASPEN trial enrolled more patients than the other 2 studies (n = 102 [zanubrutinib arm only] versus n = 71 versus n = 72). Follow-up times were similar between the ASPEN and the Tedeschi et al. (2015) study, at 19.47 months and 19 months, respectively; however, the Dimopoulos et al. (2007) and Kastritis et al. (2015) study had follow-ups of 23.4 months and 8 years, respectively.²⁴ Many important baseline characteristics for the Tedeschi et al. (2015) and Dimopoulos et al. (2007) and Kastritis et al. (2015) studies were not reported. As such, there are numerous sources of possible clinical heterogeneity between the studies.