CADTH Reimbursement Review

Zanubrutinib (Brukinsa)

Sponsor: BeiGene Canada ULC

Therapeutic area: Chronic lymphocytic leukemia

This multi-part report includes:

Clinical Review

Pharmacoeconomic Review

Stakeholder Input

Clinical Review

Executive Summary

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

Table 1: Background Information of Application Submitted for Review

NOC = Notice of Compliance.

Introduction

Chronic lymphocytic leukemia (CLL) is characterized by a proliferation and accumulation of small mature B-cells in the blood, bone marrow, lymph nodes, and lymphoid tissue.^1-3 Patients may present with B symptoms (features of lymphoma such as fever, chills, night sweats, and unintentional weight loss), fatigue, enlarged lymph nodes, or splenomegaly. However, clinical presentation is often asymptomatic in CLL.⁴

In Western countries, CLL is the most common type of leukemia, with 2018 Canadian cancer statistics showing an incidence of 6.0 per 100,000 population for newly diagnosed CLL (1,725 new cases).⁵ Statistics Canada estimated that the 2-year prevalence of CLL in Canada, excluding Quebec, in 2018 was 18.5 cases for males and 11.8 cases for females (total, 15.1 cases) per 100,000 population. In Canada, the diagnosis of CLL is guided by the International Workshop on Chronic Lymphocytic Leukemia (iwCLL) or WHO guidelines.^2,3

Treatment is generally not required for early asymptomatic CLL. Patients with early asymptomatic disease (e.g., Rai stage 0 or Binet stage A) are often followed with a watch-and-wait strategy and routine follow-ups to monitor their disease with physical examination that includes palpation of the lymph node areas, spleen, and liver, as well as complete and differential blood counts.^6,7 When treatments are indicated based on risk or disease symptoms, the treatment strategy should be personalized according to risk factors, age, fitness, and patient preferences.⁸ In Canada, physicians use 3 risk biomarkers (IGHV status, 17p deletion, and TP53 mutation) to guide personalized treatment.⁸ Very few patients are cured of CLL; therefore, the goals of therapy in most cases are to achieve effective and durable disease control (based on progression-free survival [PFS] and overall survival [OS]) with minimal toxicity and an acceptable quality of life.^6,9 Even though many patients achieve remission with appropriate treatment, relapse is common. Some patients with relapsed or refractory (r/r) disease require subsequent lines of treatment over the course their disease.^7,10 For treatment-naive (TN) patients, treatment options include fludarabine plus cyclophosphamide plus rituximab (FCR), chemoimmunotherapy combinations (e.g., bendamustine plus rituximab [BR], venetoclax plus obinutuzumab [VenG], chlorambucil plus obinutuzumab [GClb]), and Bruton tyrosine kinase (BTK) inhibitors.¹¹ In the r/r patients, treatment options include BTK inhibitors, venetoclax-based regimen, and idelalisib plus rituximab. Allogenic stem cell transplant is another potential option in the r/r setting.¹⁰

Zanubrutinib (Brukinsa) is a small-molecule BTK inhibitor that inhibits BTK activity by covalently binding to a cysteine residue in the BTK active site. Zanubrutinib has been observed to inhibit malignant B-cell proliferation and reduce tumour growth. Zanubrutinib is taken orally at doses of 320 mg (four 80 mg capsules) once daily or 160 mg (two 80 mg capsules) twice daily for the treatment of CLL in adults.¹² Zanubrutinib has been previously reviewed by CADTH for the treatment of Waldenström macroglobulinemia and mantle cell lymphoma who have received at least 1 prior therapy.^13,14 The sponsor’s reimbursement request aligns with the approved Health Canada indication (post–Notice of Compliance [NOC]). Zanubrutinib is indicated for the treatment of CLL and small lymphocytic lymphoma (SLL), a rare nodal form of CLL, in the US¹⁵ and for CLL in the European Union.¹⁶

The objective of this report is to review and critically appraise the evidence submitted by the sponsor on the beneficial and harmful effects of zanubrutinib capsules, 80 mg, oral, for the treatment of CLL in adult patients, as per the indication approved by Health Canada (post-NOC).

Stakeholder Perspectives

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

Patient Input

Patient input was provided by Lymphoma Canada (LC), a national charity with a mission to advocate and improve access to health care for patients affected by CLL and SLL in Canada. LC submitted input based on information collected from an anonymous online survey that was distributed in Canada and at international locations by email and social media, from November 2022 to February 2023. A total of 173 people (64 from Canada, 9 from the US, 1 from Costa Rica, and 99 from unknown locations) responded to the survey. Of the respondents, 149 had confirmed CLL, 23 had been diagnosed with SLL, and 1 had been newly diagnosed with an unknown lymphoma. CLL Canada assisted LC in distributing the survey and preparing the submission.

According to the survey, most patients with CLL or SLL were diagnosed through routine bloodwork and had been experiencing no or minor symptoms at the time of diagnosis. For the 122 respondents who rated the impact of their disease as highly negative (3 to 5 out of 5) at the time of diagnosis, the most frequent symptoms were fatigue (reported by 40% of respondents), night sweats (reported by 27%), and body aches and pains (reported by 20%). In terms of the psychosocial impact of CLL and SLL at the time of diagnosis, the most common factors reported by 109 respondents were anxiety and/or worry (reported by 61% of respondents) and stress of diagnosis (reported by 41%). Similarly, for the 109 respondents who reported currently experiencing effects that had a highly negative impact (3 to 5 out of 5), the most frequently reported symptoms were fatigue (reported by 44% of respondents), body aches and pains (reported by 25%), and night sweats (reported by 16%). Up to 75% of the 109 respondents with CLL experienced a negative impact on quality of life, such as anxiety and/or worry (reported by 61% respondents), stress of diagnosis (reported by 40%), and difficulty sleeping (reported by 37%). Of the 109 respondents who indicated that CLL had a negative impact on daily activities, the most frequently affected activities were travel (reported by 35% of respondents), volunteering (reported by 25%), and spending time with family and friends (reported by 24%). Seventy-six patients said that the following factors were extremely important when considering a novel therapy over their current treatment option(s): longer survival (reported by 85% of respondents), control of disease and symptoms (reported by 79%), longer remission (reported by 75%), and better quality of life (reported by 66%). Of the 77 patients who responded to a question about the importance of choice and options when deciding on a CLL treatment course, 60% said it is extremely important to have choice and 65% said it is extremely important to have a higher number of CLL and SLL treatment options available. When asked about the preference for route of administration (oral pill versus IV), 63 of 77 patients (82%) confirmed that they would prefer oral administration. Eleven patients (10 who had been previously treated) had experience with zanubrutinib for CLL treatment. Two patients said they are in remission (1 after 6 months and 1 after 1 to 2 years of zanubrutinib treatment) and 5 patients indicated that zanubrutinib controlled their CLL or SLL symptoms better than their previous treatments. Seven respondents are still on zanubrutinib treatment and 1 patient stopped. Four of 11 patients reported that they did not experience any side effects, and 8 patients reported that the side effects of zanubrutinib were less severe than those they had experienced with previous therapies. Symptoms reported were fatigue, easy bruising and/or bleeding, confusion or memory loss, diarrhea, muscle or joint pain, peripheral edema, hypertension, and localized infections. Two patients said that zanubrutinib negatively impacted their quality of life compared to other treatments.

Clinician Input

Input From the Clinical Expert Consulted by CADTH

The clinical expert consulted by CADTH indicated that the most important goals of treatment for patients with CLL is to reverse symptoms and control the disease for as long as possible with minimal toxicity and no significant negative impact on quality of life. The clinical expert stated that the biggest limitation to current treatments for patients with CLL is that tumour cell resistance usually occurs, and patients stop responding or relapse on therapy. Other limitations include toxicity and drug interactions, the requirement for continuous ongoing treatment, and the lack of curative treatments for patients with CLL. The clinical expert believed that the value of zanubrutinib would be incremental rather than transformative, as there are already 2 BTK inhibitors (i.e., ibrutinib and acalabrutinib) commonly used in clinical practice. The clinical expert speculated that zanubrutinib would be a welcome option for the first-line treatment of patients with CLL but would not be efficacious in patients who progress on other BTK inhibitors. According to the clinical expert, the patient population for zanubrutinib includes untreated patients aged 65 years or older with a good performance status with or without high-risk mutations (i.e., TP53 mutations, 11q mutations, or unmutated IGHV genes) or patients younger than 65 years who are not candidates for FCR, and r/r patients with CLL without transformation or central nervous system involvement. Although zanubrutinib is not ideal for patients who have received previous treatment with a BTK inhibitor or who have a bleeding disorder, the clinical expert consulted by CADTH indicated that zanubrutinib may be better tolerated by patients who need to stop other BTK inhibitors because of toxicity.

The clinical expert indicated that response to treatment is assessed by changes in peripheral blood counts. Disease progression, measured by an increasing lymphocyte count or worsening cytopenia, is a major reason for discontinuing treatment with zanubrutinib, per expert opinion. Toxicities that cannot be managed with dose reductions or a pause could also be a reason for stopping treatment. Zanubrutinib must be paused before various surgical procedures, due to the risk of bleeding. The clinical expert stated that zanubrutinib treatment should be managed by a specialist (i.e., hematologist or medical oncologist) who is familiar with this class of drug and can deal with toxicities and optimal dosing.

Clinician Group Input

LC, represented by 1 hematologist and the Ontario Health – Cancer Care Ontario (OH-CCO) Hematology Cancer Drug Advisory Committee, which consisted of 4 hematologists, submitted 2 clinician group inputs. In alignment with clinician group input, the clinical expert consulted by CADTH stated that patients who have been previously treated with a BTK inhibitor are not ideal candidates for zanubrutinib, but that patients with high-risk mutations (i.e., TP53 mutations, 11q mutations, or unmutated IGHV genes) older than 65 years and patients younger than 65 years who are not candidates for FCR are eligible for zanubrutinib treatment. Also, the clinical expert emphasized duration of response (DOR) or response to next treatment as important end points. Otherwise, the clinical expert and the 2 clinician groups generally agreed that zanubrutinib is a viable first-line option for TN patients with CLL and patients with r/r disease; important outcomes include a reduced symptom burden and an improved quality of life with minimal toxicity from treatment; routine blood counts and clinical exams should be used to measure response to therapy; patients with progressive disease and/or intolerable toxicity despite dose reductions should be considered for discontinuation of zanubrutinib treatment; and a specialist such as a hematologist, medical oncologist, or any other staff member who specializes in managing malignant hematological conditions and/or is familiar with this class of drug should be involved in the management of CLL with zanubrutinib.

Drug Program Input

The drug programs that participate in the CADTH reimbursement review process identified potential implementation issues related to relevant comparators for zanubrutinib; the accessibility of zanubrutinib for patients with high-risk genetic factors and patients who are unsuitable for IV therapy; the preferred dosing schedule for zanubrutinib in clinical practice; the eligibility of patients who are currently receiving ibrutinib or acalabrutinib and have not experienced disease progression; a change of place in therapy that zanubrutinib may cause for comparator drugs; dispensing issues due to storage restrictions; and the potential for drug-drug, drug-food, and drug-herb interactions with zanubrutinib that require assessment and/or intervention.

Clinical Evidence

Pivotal Studies and Randomized Controlled Trial Evidence

Description of Studies

Both the SEQUOIA and ALPINE trials are ongoing phase III, open-label, randomized controlled trials (RCTs). In the SEQUOIA trial, cohort 1 compared the efficacy and safety of zanubrutinib to BR in TN patients with CLL or SLL who were negative for 17p deletion and who were either older than 65 years or younger than 65 years with comorbid illnesses and at least 1 indication to treat. The SEQUOIA trial also included a single-arm study of 111 patients, cohort 2, which included TN patients with CLL or SLL who were positive for 17p deletion. The ALPINE trial compared the efficacy and safety of zanubrutinib to ibrutinib in r/r patients with CLL or SLL. Patients were randomized using an Interactive Response Technology (IRT) system to receive zanubrutinib or BR in the SEQUOIA trial and to receive ibrutinib in the ALPINE trial. The stratification factors were age, geographic region, genetic mutations, refractoriness to last therapy (in the ALPINE trial), and disease stage (in the SEQUOIA trial). A total of 479 patients in cohort 1 of the SEQUOIA trial were randomized in a 1:1 ratio to receive zanubrutinib (n = 241) or BR (n = 238), and 652 patients in the ALPINE trial were randomized in a 1:1 ratio to receive zanubrutinib (n = 327) or ibrutinib (n = 325). No Canadian sites were included in either trial.

The primary end point was PFS per independent review committee (IRC) in the SEQUOIA trial and overall response rate (ORR) per investigator assessment (IA) in the ALPINE trial. Other outcomes of interest included PFS per IA, ORR per IRC, OS, DOR (per IRC and IA), time to treatment failure, incidence of atrial fibrillation and flutter, and health-related quality of life (HRQoL).

In the SEQUOIA trial, the demographic and baseline characteristics were generally balanced between the zanubrutinib and BR arms in cohort 1, although zanubrutinib-treated patients were slightly more likely to be white than BR-treated patients (91.7% versus 86.6%). Most patients in cohort 1 (zanubrutinib versus BR) were enrolled at sites in Europe (72.2% versus 72.3%) and had an Eastern Cooperative Oncology Group Performance Status (ECOG PS) of 0 or 1 (93.8% versus 91.6%). The demographic and baseline characteristics were generally similar between the zanubrutinib arms in cohort 1 (without 17p deletion) and cohort 2 (with 17p deletion), except there were more patients from the Asia-Pacific region enrolled in cohort 2 than in cohort 1 (42.3% versus 13.7%).

In the ALPINE trial, demographic and baseline patient characteristics were similar in the zanubrutinib and ibrutinib arms in the intention-to-treat (ITT) analysis set (final ORR analysis). The median age was 67.0 years (range, 35 to 90 years) in the zanubrutinib arm and 68.0 years (range, 35 to 89 years) in the ibrutinib arm. Most patients (zanubrutinib versus ibrutinib) were enrolled at sites in Europe (60.6% versus 58.8%), were white (79.8% versus 83.1%), and had an ECOG PS of 0 or 1 (97.9% versus 96.0%). Demographic and baseline patient characteristics in the final PFS analysis ITT analysis set were similar to those in the ALPINE final ITT ORR analysis set.

Efficacy Results

A summary of key efficacy results is provided in Table 2 for the SEQUOIA trial and Table 4 for the ALPINE trial. Detailed outcome data are presented in Appendix 1.

Progression-Free Survival

PFS per IRC in the SEQUOIA Trial: As of the May 7, 2021, data cut-off, median PFS per IRC had not yet been reached in the zanubrutinib arm, and median PFS per IRC was 33.7 months (95% confidence interval [CI], 28.1 to not estimable [NE]) in the BR arm. Median follow-up time was 25.1 months (95% CI, 24.9 to 25.4 months) in the zanubrutinib arm and 24.6 months (95% CI, 22.8 to 25.2 months) in the BR arm. The hazard ratio (HR) for PFS per IRC comparing zanubrutinib with BR was 0.42 (95% CI, 0.28 to 0.63; P < 0.0001) in favour of zanubrutinib. Higher event-free rates were observed (zanubrutinib versus BR) at 12 months (94.5% versus 90.2%), at 24 months (85.5% versus 69.5%), and at 36 months (81.5% versus 40.8%). Subgroup analyses of PFS per IRC by age (< 65 years versus ≥ 65 years), sex, and ECOG PS (0 versus ≥ 1) were generally consistent with the primary analysis across all strata. However, inconsistent findings were reported in the subgroup analyses of high-risk genetic factors (IGHV mutation status [unmutated versus mutated] and TP53 mutation status [unmutated versus mutated]), cancer type (CLL versus SLL), disease stage (Binet stage A or B versus Binet stage C), and complex karyotype (< 3 abnormalities versus ≥ 3 abnormalities). Refer to Appendix 1 for the detailed subgroup analyses data. In addition, several prespecified sensitivity analyses based on the IRC assessment of PFS were included in the statistical analysis plan, including unstratified analysis, using the per-protocol analysis set, and changes to definitions of PFS and censoring events. The results were generally consistent with the results of the primary analysis (HR = 0.42; 95% CI, 0.28 to 0.63; P < 0.0001), and showed HR values ranging from 0.45 (95% CI, 0.31 to 0.67) to 0.34 (95% CI, 0.22 to 0.53).

In cohort 2, the median PFS by IRC was not reached in the zanubrutinib arm, and the event-free rates were 93.6% at 12 months, 88.9% at 24 months, and 84.9% at 36 months. A higher rate of progression was observed in patients with concurrent TP53 mutation than in those without (21.3% versus 8.1%). Consistent results were observed for investigator-assessed PFS, with event-free rates of 94.5% at 12 months, 87.0% at 24 months, and 82.6% at 36 months.

PFS per IA in the SEQUOIA Trial: The analysis of PFS per IA is the secondary outcome in cohort 1. A high concordance for PFS was observed in the IRC and investigator assessments (concordance rate for disease progression was 91.4.%), and the HRs for IRC-assessed and investigator-assessed PFS were also similar (HR = 0.42 [95% CI, 0.28 to 0.63] for PFS per IRC; HR = 0.42 [95% CI, 0.27 to 0.66] for PFS per IA).

In cohort 2, the median PFS by IA was not reached in the zanubrutinib arm, and the event-free rates were 94.5% at 12 months, 87.0% at 24 months, and 82.6% at 36 months.

PFS per IRC in the ALPINE Trial: The analysis of PFS per IRC in the ALPINE trial is a secondary outcome. Of note, this analysis was not part of the statistical hierarchy and was not adjusted for multiplicity. At the final PFS analysis cut-off date of August 8, 2022, IRC-assessed PFS events had occurred in 88 patients (26.9%) in the zanubrutinib arm and 120 patients (36.9%) in the ibrutinib arm (HR = 0.65; 95% CI, 0.49 to 0.86, nominal P = 0.0024). Median follow-up time was 32.9 months (95% CI, 27.8 to 33.1 months) in the zanubrutinib arm and 28.1 months (95% CI, 27.6 to 33.0 months) in the ibrutinib arm. Median PFS was not reach in the zanubrutinib arm, with the lower bound of the 95% CI at 34.3 months and median PFS was 35.0 months (95% CI, 33.2 to 44.3 months) in the ibrutinib arm.

Generally similar findings were observed in the interim efficacy set (data cut-off: December 31, 2020) and the final ORR analysis ITT analysis set (data cut-off: December 1, 2021).

PFS per IA in the ALPINE Trial: The analysis of PFS per IA in the ALPINE trial is the key secondary outcome and was adjusted for multiplicity in the final PFS analysis. At the final PFS analysis cut-off date of August 8, 2022, the investigators assessed that PFS events had occurred in 87 patients (26.6%) in the zanubrutinib arm and 118 patients (36.3%) in the ibrutinib arm. Patients in the zanubrutinib arm had a lower risk of PFS events per IA (HR = 0.65; 95% CI, 0.49 to 0.86), which was both noninferior (P < 0.0001 versus prespecified 1-sided significance level of 0.02498) and superior (P = 0.0024 versus prespecified 1-sided significance level of 0.02498). Median follow-up time was 31.4 months (95% CI, 27.7 to 33.1 months) in the zanubrutinib arm and 27.8 months (95% CI, 27.6 to 33.1 months) in the ibrutinib arm. Median PFS was not reached in the zanubrutinib arm, with the lower bound of the 95% CI at 34.3 months, and median PFS was 34.2 months (95% CI, 33.3 to NE) in the ibrutinib arm.

Generally similar findings were observed in the interim efficacy set (data cut-off: December 31, 2020) and the final ORR analysis ITT analysis set (data cut-off: December 1, 2021).

Overall Survival

SEQUOIA Trial: As of the final data cut-off of May 7, 2021, OS events had occurred in 16 (6.6%) patients in the zanubrutinib arm and 14 (5.9%) patients in the BR arm. The HR for OS comparing zanubrutinib with BR was 1.07 (95% CI, 0.51 to 2.22, P = 0.5672). Median OS was not reached in the zanubrutinib arm with a median follow-up time of 26.5 months, whereas in the BR arm, median OS was 37.8 months (95% CI, 37.8 to NE) with a median follow-up time of 25.1 months. Event-free rates (zanubrutinib versus BR) were 98.3% versus 96.4% at 12 months, 94.3% versus 94.6% at 24 months, and 92.3% versus 93.6% at 36 months.

In cohort 2, at the data cut-off date of May 7, 2021, there were 8 deaths (7.3%) reported in the zanubrutinib arm. Median OS was not reached in the zanubrutinib arm with a median follow-up time of 30.4 months. Event-free rates were 96.4% at 12 months, 93.6% at 24 months, and 90.7% at 36 months.

ALPINE Trial: The analysis of OS in the ALPINE trial is a secondary outcome. Of note, this analysis was not part of the statistical hierarchy and not adjusted for multiplicity.

At the final PFS analysis cut-off in the ITT analysis set (August 8, 2022), there were 48 deaths (14.7%) reported in the zanubrutinib arm and 60 deaths (18.5%) reported in the ibrutinib arm (HR = 0.76; 95% CI, 0.51 to 1.11, nominal P = 0.1533). Median OS was not reached in either arm at the median follow-up times of 32.9 months in the zanubrutinib arm and 32.7 months in the ibrutinib arm. Most patients were alive and on study at the data cut-off date (79.5% for zanubrutinib versus 73.8% for ibrutinib).

Generally similar findings were observed in the interim efficacy set (data cut-off: December 31, 2020) and the final ORR analysis ITT analysis set (data cut-off: December 1, 2021)

Overall Response Rate

ORR per IRC and per IA in the SEQUOIA Trial: Analyses of ORR per IRC and per IA (data cut-off: May 7, 2021) were secondary outcomes in cohorts 1 and 2, respectively, of the SEQUOIA study. Note that this analysis was not part of the statistical hierarchy and not adjusted for multiplicity. The ORR per IRC was 94.6% (95% CI, 91.0% to 97.1%) in the zanubrutinib arm and 85.3% (95% CI, 80.1% to 89.5%) in the BR arm. The majority of patients (zanubrutinib versus BR) achieved a partial response (PR) (85.5% versus 64.3%), followed by a complete response (CR) (6.6% versus 15.1%), a nodular PR (1.2% versus 5.9%), and a partial response with lymphocytosis (PRL) (1.2% versus 0.0%). Generally, ORR per IA was consistent with ORR per IRC. In cohort 2, ORR per IRC was 90.0% (95% CI, 82.8% to 94.9%), whereas ORR per IA was slightly higher, at 96.4% (95% CI, 91.0% to 99.0%).

ORR per IRC in the ALPINE Trial: In the ITT analysis set at the final PFS analysis cut-off of August 8, 2022, when ORR was assessed by the IRC, a higher response rate was observed in the zanubrutinib arm than in the ibrutinib arm (86.2% versus 75.7%). The majority of patients (zanubrutinib versus ibrutinib) achieved a PR (78.6% versus 69.8%), followed by a CR (6.7% versus 5.5%), a nodular PR (0.9% versus 0.0%), and a complete response with incomplete bone marrow recovery (CRi) (0.0% versus 0.3%). The analysis of ORR per IRC was the secondary outcome in the ALPINE study.

Generally similar findings were observed in the interim efficacy set (data cut-off: December 31, 2020) and the final ORR analysis ITT analysis set (data cut-off: December 1, 2021).

ORR per IA in the ALPINE Trial: In the final ORR analysis ITT analysis set (data cut-off: December 1, 2021), ORR per IA was higher in the zanubrutinib arm than in the ibrutinib arm (79.5% versus 71.1%). Most patients (zanubrutinib versus ibrutinib) achieved a PR (73.7% versus 68.3%), followed by a CR (33.7% versus 22.5%), a CRi (1.2% versus 0.3%), and a nodular PR (0.9% versus 0.0%). In this analysis, the response ratio for zanubrutinib to ibrutinib was 1.12 (95% CI, 1.02 to 1.22; superiority 2-sided nominal P = 0.0013).

Similar findings were observed in the interim analysis set ITT analysis set (data cut-off: December 31, 2020) and the final PFS analysis set ITT analysis set (data cut-off: August 8, 2022).

European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30

The European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ-C30) is a questionnaire developed to assess the quality of life of cancer patients. The EORTC QLQ-C30 includes 30 separate questions (items), resulting in 5 functional scales (physical functioning, role functioning, emotional functioning, cognitive functioning, and social functioning), 1 global health status scale, 3 symptom scales (fatigue, nausea and vomiting, and pain), and 6 single items (dyspnea, insomnia, appetite loss, constipation, diarrhea, and financial difficulties).¹⁷ Each raw scale score is converted to a standardized score that ranges from 0 to 100, with a higher score reflecting better function on the function scales, a worse state on the symptom and single-item symptom scales, and a better quality of life on global quality of life scale.

|||||||| |||||||| ||| |||||| |||| |||||||| || |||| || || ||| ||||||| ||| || ||||| | ||| ||||| || || ||| |||||| ||||| |||||||| ||| ||||||||| || |||| || ||| ||||| ||||||| |||||||| ||| ||| || ||| ||| || |||| |||||||||| ||||||| ||||||||| |||| ||||||||| || |||||| ||| ||| |||||| |||||| |||||||||||||| || ||||| |||||||| |||||||||||| ||| ||||||||||||||| ||||||| || ||| ||||||| |||||| ||||| ||||||||| |||||||||| ||||||| ||| ||||||||| |||| ||| ||| ||||||||||| || |||| || ||| ||||| ||||||| |||||||||||||| ||||||| ||||||| |||| ||| |||||||| ||| |||||| | || ||| ||||||| ||||||

EQ-5D-5L

SEQUOIA Trial: || |||||| || ||||||||| | |||||| |||||| |||||||||| || |||| |||||||| || ||| |||||||||||| ||| |||||||| || |||||||| || |||||||| |||||||| ||||||||| |||||||| |||||||||| |||| ||| || ||| || |||| ||| ||||| ||||||| ||||||| ||||||||| |||||||||| ||||| ||||||||||| ||| || |||||||||| || |||||||||||||||| ||| ||||||||||||||||||| ||||||| |||||||||||| || ||| |||| |||| ||||||| |||| |||||||| || ||| ||| ||||| || ||| |||||||| |||| |||||||| || ||| |||||||||||| ||| |||| |||||||| ||| ||| || ||| |||| |||||||| || |||| ||| || |||| || || ||| |||||||||||| ||| |||| |||||||| ||| ||| || ||| |||| |||||||| || |||| |||||||||||| ||||||| |||| ||| |||||||| ||| |||||| ||

ALPINE Trial: || ||| ||||| ||| |||||||| ||||||| ||||||| ||| |||||| || ||||| | ||| ||||| ||| ||||||||| | |||||||| |||||| ||||||||||| || |||||||| |||||||| || |||||||| || |||||||| ||||||| ||||||||| ||||||||| |||||||||| ||| ||||||||||||||| || ||| |||||||||||| ||| |||| ||| ||||||||| |||| ||||| || ||| ||||| |||||||||| ||| |||||||||||||||||| |||||||| | |||||||| ||||| ||||||||||| || |||||||| |||||||| || |||||||| || ||| |||||||||||| ||| |||| ||| ||||||||| |||| |||||||||||| || ||| |||| |||| ||||||| |||| |||||||| || ||| ||| ||||| || ||| |||||||| |||| |||||||| || ||| |||||||||||| ||| |||| ||||||| ||| ||| ||||||||| ||| |||| |||| ||| || ||||| || || |||| || || ||| |||||||||||| ||| |||| ||||||| ||| ||| ||||||||| ||| |||| ||||||| || ||||| ||||||||||| |||||||| |||| |||||||| || ||| ||||||| |||||||| ||| ||||| |||||||| |||||||| ||| ||||| ||| ||| ||||| ||| |||||||| ||| |||||||| ||| ||||| |||||||| |||||||| ||| ||||||||| ||||| ||||| ||| || ||||||||| ||||||| ||||||||| ||| ||| ||||||||| || |||||||| ||| || |||||||| |||| |||| |||| |||| ||||||||| ||||||| ||| |||||||||||| |||||||| || |||| ||| ||||||| ||| |||||| |||| |||||||||||

Harms Results

Incidence of Atrial Fibrillation and Flutter

SEQUOIA Trial: In cohort 1, the proportion of patients who had atrial fibrillation and flutter was 3.3% in the zanubrutinib arm and 2.6% in the BR arm. || |||||| || ||| |||||||||| || |||||||| ||| ||| ||||||| |||||| || |||||| |||||||||||||||||||| |||||||| ||| |||||

ALPINE Trial: Atrial fibrillation and flutter were tested as a key secondary end point, separated from the fixed-sequence hierarchical testing for the primary end point (ORR per IA), as zanubrutinib was found to be noninferior to ibrutinib in investigator-assessed ORR at the interim analysis. Multiplicity was controlled at the interim and final ORR analyses.

In the safety analysis set (data cut-off: December 31, 2020), atrial fibrillation and flutter were analyzed in the first 415 randomized patients, according to actual treatment received. In that analysis, the zanubrutinib arm had a significantly lower frequency of atrial fibrillation and flutter than the ibrutinib arm (2.5% versus 10.1%), which corresponded to a rate difference of –7.7% (95% CI, –12.3% to –3.1%; P = 0.0014).

In the ALPINE safety analysis set at the final ORR analysis cut-off (December 1, 2021), atrial fibrillation and flutter were less common in the zanubrutinib arm than in the ibrutinib arm (4.6% versus 12.0%), which corresponded to a rate difference of –7.4% (95% CI, –11.6% to –3.2%, P = 0.0006).

In the ALPINE safety analysis set at the final PFS analysis cut-off (August 8, 2022), atrial fibrillation and flutter were less common in the zanubrutinib arm (5.2%) than in the ibrutinib arm (13.3%), which corresponded to a rate difference of – 8.0% (95% CI, –12.4% to –3.6%, nominal P = 0.0004).

Adverse Events, Serious Adverse Events, Withdrawals due to Adverse Events, Mortality, and Notable Harms

The percentage of patients with any reported treatment-emergent adverse events (TEAEs) was 93.3% in the zanubrutinib arm and 96.0% in the BR arm in cohort 1 of the SEQUOIA trial, and 98.1% in the zanubrutinib arm and 99.1% in the ibrutinib arm in the ALPINE trial (final PFS analysis data cut-off: August 8, 2022). In SEQUOIA cohort 1, the most commonly reported adverse events (AEs) (occurring in ≥ 15% of patients) with a percentage of greater than 5% in the zanubrutinib arm (zanubrutinib versus BR) were contusion (19.2% versus 3.5%) and upper respiratory tract infection (17.1% versus 11.9%). In SEQUOIA cohort 2, 109 (98.2%) patients had at least 1 AE. The most commonly reported AEs in this arm were upper respiratory tract infection (20.7%), arthralgia and contusion (19.8% each), diarrhea (18.0%), nausea (16.2%), and constipation (15.3%).

In the ALPINE trial, COVID-19 was more commonly reported in the zanubrutinib arm than in the ibrutinib arm (23.1% versus 17.9%), as was upper respiratory infection (17.9% versus 12.7%). Serious adverse events (SAEs) were reported in 36.7% of patients in the zanubrutinib arm and 49.8% of patients in the BR arm in the SEQUOIA trial, and in 42.0% of patients in the zanubrutinib arm and 50.0% in the ibrutinib arm in the ALPINE trial (final PFS analysis data cut-off: August 8, 2022).

In the safety analysis set of cohort 1, TEAEs leading to treatment discontinuation was less common in the zanubrutinib arm than in the BR arm (8.3% versus 13.7%). In the ALPINE safety analysis set at the final PFS analysis cut-off (August 8, 2022), the incidence of TEAEs leading to treatment discontinuation was lower in the zanubrutinib arm than in the ibrutinib arm (15.4% versus 22.2%). In SEQUOIA cohort 1, death was recorded for 15 patients (6.6%) in the BR arm and 16 patients (6.7%) in the zanubrutinib arm. The most common cause of death was AEs in the BR arm (11 patients, or 4.8%) and in the zanubrutinib arm (11 patients, or 4.6%). In cohort 2, death was recorded for 8 patients (7.2%) in the zanubrutinib arm, which was most commonly related to disease progression (4 patients, or 3.6%) or AEs (3 patients, or 2.7%).

In the ALPINE safety analysis set at the final PFS analysis cut-off (August 8, 2022), a total of 108 deaths were reported. A lower proportion of patients died in the zanubrutinib arm than the ibrutinib arm (14.8% versus 18.5%). The most common causes of death were (zanubrutinib versus ibrutinib) TEAEs (9.0% versus 11.4%) and CLL and SLL (4.6% versus 5.6%; there were no detailed breakdown data reported). With regard to AEs of special interest (AESIs), in SEQUOIA cohort 1, the zanubrutinib and BR arms had similar overall rates of AESIs (82.9% versus 89.0%), although hemorrhage was reported more commonly in the zanubrutinib arm than in the BR arm (45.0% versus 11.0%), as was infection (62.1% versus 55.99%). In cohort 2, the most commonly reported AESIs were infections (79 patients, or 71.2%), hemorrhage (57 patients, or 51.4%), and second primary malignancies (24 patients, or 21.6%). In the ALPINE safety analysis set at the final PFS analysis cut-off (August 8, 2022), in general, the zanubrutinib and ibrutinib arms had similar overall rates of AESIs (90.7% versus 92.6%), except for atrial fibrillation and flutter (5.2% versus 13.3%), which were lower in the zanubrutinib arm than in the ibrutinib arm. Neutropenia was reported more commonly in the zanubrutinib arm than in the ibrutinib arm (29.3% versus 24.4%). The most common AESIs in the zanubrutinib arm were (zanubrutinib versus ibrutinib) infections (71.3% versus 73.1%) and hemorrhage (42.3% versus 41.4%). Generally similar findings were observed in the interim safety set (data cut-off: December 31, 2020) and in the final ORR analysis safety set (data cut-off: December 1, 2021).

Table 2: Summary of Key Results From the SEQUOIA Trial (ITT Analysis Set for Cohort 1, Safety Analysis Set for Cohort 2)

AE = adverse event; AESI = adverse event of special interest; BR = bendamustine plus rituximab; CI = confidence interval; EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; GHS = global health status; HR = hazard ratio; IA = investigator assessment; IRC = independent review committee; ITT = intention to treat; LS = least squares; NA = not applicable; NE = not estimable; NR = not reported; ORR = overall response rate; OS = overall survival; PFS = progression-free survival; QoL = quality of life; SAE = serious adverse event; SD = standard deviation; TEAE = treatment-emergent adverse event; VAS = visual analogue scale; vs. = versus.

Note: Data cut-off date was May 7, 2021.

^aComparative analysis of PFS as the primary end point was limited to zanubrutinib vs. BR in cohort 1.

^bAdjusted for multiplicity.

^cORR defined as the proportion of patients who achieved a complete response, complete response with incomplete hematopoietic recovery, nodular partial response, partial response, or partial response with lymphocytosis.

^dA positive value indicates improvement.

^eA negative value indicates improvement.

Source: SEQUOIA Clinical Study Report.¹⁸

Table 3: Summary of Key Results From the ALPINE Trial

AE = adverse event; AESI = adverse event of special interest; CI = confidence interval; EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; GHS = global health status; IA = investigator assessment; IRC = independent review committee; LS = least squares; NE = not estimable; NR = not reported; ORR = overall response rate; OS = overall survival; PFS = progression-free survival; QoL = quality of life; SAE = serious adverse event; SD = standard deviation; TEAE = treatment-emergent adverse event; VAS = visual analogue scale.

Notes: The PFS and OS analyses in the final ORR analysis (data cut-off: December 1, 2021) were not prespecified. Therefore, the P values for these analyses were not adjusted for multiple testing and are presented for descriptive purposes only.

Details from the table have been taken from the sponsor’s Summary of Clinical Evidence.⁴

^aInvestigator-assessed PFS was tested for noninferiority using a stratified Wald test and, if noninferiority was demonstrated, superiority was tested using a stratified log-rank test. Both analyses used 1-sided significance levels of 0.02498.

^bMultiplicity due to multiple end points and multiple tests was adjusted using fixed-sequence hierarchical testing.

^cOne-sided P value is calculated for noninferiority with the stratified test statistic against a null response ratio of 0.8558; the prespecified 1-sided significance level for ORR analysis was 0.005. Because the ALPINE trial met its primary end point at the prespecified interim analysis, P values for the final ORR analysis cut-off of December 1, 2021, are presented for descriptive purposes only.

^dTwo-sided P value for superiority is calculated with the stratified Cochran-Mantel-Haenszel test statistic; superiority testing with a 1-sided significance level of 0.0235 at the final ORR analysis (data cut-off: December 1, 2021) of ORR corresponds to chi-square P value cut-offs of 0.0469. Because the ALPINE trial met its primary end point at the prespecified interim analysis, P values for the final ORR analysis cut-off of December 1, 2021, are presented for descriptive purposes only.

^eResponders are defined as patients with a complete response, complete response with incomplete bone marrow recovery, partial response, or nodular partial response.

^fClopper-Pearson CI.

^gResponse ratio is the estimated ratio of the ORR in the zanubrutinib arm divided by that the ibrutinib arm.

^hA positive value indicates improvement.

ⁱA negative value indicates improvement.

Sources: ALPINE final ORR and final PFS Clinical Study Reports.^19,20

Critical Appraisal

Both the SEQUOIA and ALPINE trials were ongoing phase III, open-label RCTs. There was no particular concern with the methods of randomization or stratification. For the SEQUOIA trial, the CADTH review team considered the open-label design to be reasonable, given the distinct dosing regimens and administration routes between zanubrutinib and BR, which would likely allow investigators and patients to make inferences about treatment assignment regardless of blinding. In addition, cohort 2 in the SEQUOIA trial was designed as a single-arm study based on ethical considerations, as it is unethical to assign high-risk patients with 17p deletion to receive BR, which is associated with poor clinical outcomes and poor responses in this patient population. The CADTH review team would like to note that the open-label design of the SEQUOIA and ALPINE trials had the potential to introduce reporting bias in the assessment of subjective outcomes reported by patients, such as HRQoL and AEs. Disease response outcomes (PFS, ORR, DOR) were assessed by investigators and an IRC to help mitigate the biases associated with the open-label study design for both trials. Many of the outcomes used in the SEQUOIA and ALPINE trials (PFS, OS, ORR, DOR) are standard in oncology trials. As the SEQUOIA and ALPINE trials are ongoing, early reporting of the studies resulted in data immaturity at the primary efficacy analysis for the SEQUOIA trial, and at the interim and subsequent final ORR and PFS analyses for the ALPINE trial; median OS was not reached in the zanubrutinib group in the SEQUOIA trial or in either treatment group in the ALPINE trial. There were several critical protocol amendments that impacted the conduct of the trial after patients had first been randomized that may have biased the results and increased uncertainty because of increased heterogeneity in the patient population. The type I error rate was controlled for the primary and selected secondary outcomes in both studies. Several outcomes of interest to this review were tested and nominal P values were reported (e.g., PFS per IRC in the SEQUOIA trial; ORR per IRC, DOR per IRC and per IA), but any results with a P value less than the prespecified significance level should be interpreted with caution, considering the potentially inflated type I error rate. Although the subgroup analyses were prespecified, there is no evidence that the studies were powered to detect subgroups differences. In addition, there were imbalances in dose reductions, missing doses, and treatment exposure between treatment arms in the SEQUOIA trial, which bias the study results.

In terms of generalizability of the pivotal SEQUOIA and ALPINE studies, the clinical expert commented that the eligibility criteria for the SEQUOIA study were restricted and excluded the population of patients younger than 65 years who are healthy and have no comorbid illnesses, and who are often seen in patients with CLL treated in clinical practice in Canada. Thus, the study results may not be generalizable to younger patients with CLL who have no comorbid illnesses. In addition, SEQUOIA cohort 1 excluded patients without 17p deletion, which may compromise the generalizability of the study findings regarding the comparative efficacy of zanubrutinib to the general population of patients with CLL; however, a separate nonrandomized cohort was included to assess patients with TP53 deletions or mutations. In the SEQUOIA trial, the comparator was BR. Although it was considered the standard of care at the time of study design and study initiation (2017), BR was not a clinically relevant comparator, according to the clinical expert consulted by CADTH, as it is not commonly used in clinical practice currently. The majority of older patients have either nonmutated immunoglobulin variable regions or TP53 mutations that make them eligible for treatment with BTK inhibitors such as ibrutinib or acalabrutinib or VenG, which are preferred by most over BR. With regard to the choice of comparator in the ALPINE study, ibrutinib, the clinical expert commented that ibrutinib is a clinically relevant comparator in the r/r setting if patients have received first-line chemoimmunotherapy. Overall, there was no direct evidence available regarding the efficacy and safety of zanubrutinib relative to ibrutinib, acalabrutinib, or venetoclax plus obinutuzumab in the first-line setting, or to venetoclax plus rituximab in the r/r setting; thus, these results may not address the question of the most optimal treatment for these patients. At the time this report was prepared, the duration of follow-up was inadequate for assessment of OS. Symptom data from the SEQUOIA and ALPINE studies could not be generalized in a broader context due to the limited data available.

Long-Term Extension Studies

No long-term extension studies were identified for this review.

Indirect Comparisons

Description of Studies

The sponsor submitted a network meta-analysis (NMA) and a matching-adjusted indirect treatment comparison (MAIC) comparing zanubrutinib to relevant comparators in both the TN and r/r CLL settings. The sponsor-submitted NMA was informed by a systematic literature review (SLR) conducted to identify existing RCTs in adults with TN or r/r CLL. After completion of the NMA, the sponsor considered there to be notable uncertainty in the results related to the distance between nodes or the heterogeneity of patient populations and, therefore, conducted MAICs comparing zanubrutinib with both acalabrutinib and ibrutinib in the TN and r/r settings. The primary objective of the sponsor-submitted NMA and MAIC was to assess the efficacy (PFS, OS) and safety (AEs, SAEs, discontinuations due to AEs, and AEs by preferred term) of zanubrutinib in patients with TN or r/r CLL.

Efficacy Results

Network Meta-Analysis

In the TN CLL network, a total of 5 interventions — zanubrutinib (the SEQUOIA trial), ibrutinib (the ALLIANCE trial), BR (the SEQUOIA, ALLIANCE, and MABLE studies), rituximab plus chlorambucil (RClb) (the MABLE and CLL11 studies), and GClb (the CLL11 study) — were evaluated, and the only evaluable outcome was PFS. In the fixed-effects model of PFS, zanubrutinib was favoured over GClb (HR = 0.45; 95% credible interval [CrI], 0.23 to 0.86), over BR (HR = 0.42; 95% CrI, 0.27 to 0.66), and over RClb (HR = 0.22; 95% CrI, 0.12 to 0.41); however, there was no difference between zanubrutinib and ibrutinib (HR = 1.07; 95% CrI, 0.59 to 1.98) in terms of PFS.

In the r/r CLL network, a total of 5 interventions were evaluated: zanubrutinib (the ALPINE trial), ibrutinib (the ALPINE and ELEVATE-RR trials), acalabrutinib (the ELEVATE-RR and ASCEND trials), BR (the ASCEND and MURANO trials), and venetoclax plus rituximab (VenR) (the MURANO trial). Both PFS and OS were available for inclusion in the r/r CLL network meta-analysis. In the fixed-effects model of PFS, zanubrutinib was favoured over BR (HR = 0.13; 95% CrI, 0.06 to 0.26) and over acalabrutinib (HR = 0.52; 95% CrI, 0.30 to 0.89); however, there was no difference between zanubrutinib and VenR (HR = 0.69; 95% CrI, 0.32 to 1.46). In the fixed-effects model of OS, there was no difference between zanubrutinib and any of the other treatments.

Matching-Adjusted Indirect Comparison

|| ||| || ||| |||||||| | ||||| |||| ||||||||| ||||||||| |||||||||||| |||| ||| ||||||| ||||| || |||||||||||||| ||||| ||| |||||||||| ||||| |||||||||| ||| |||||||||| ||||||| ||| |||||| |||||||||| ||||| ||| || |||||||||| ||||||| |||||||||||| ||| ||||||||||||| |||| |||| |||| ||| |||| || ||||||| |||| |||| |||| |||| ||| |||| || ||||||| || ||||||||| |||| |||| |||| ||| |||| || ||||||| ||||||| |||||||| ||| |||||||||||| ||| |||||||||| ||| |||||||| ||| |||| |||||||| ||| |||| |||||||| || || ||||||||||| ||||||||| ||||| |||||||||| ||| |||||||||| ||||||| ||| |||||| |||||||||| ||||| ||| || |||||||||| ||||||| |||||||||||| ||| |||||||||||||| ||||| || ||||||||| ||| ||| ||| ||| ||| ||| |||||||| |||| ||| |||| ||||||||| |||||||||||| ||| ||||||||||||| |||| |||||||||| ||| ||||||||| |||||||| ||| ||||||||| ||||| |||| ||||||||| ||||||| ||| ||||||| |||||||| |||| ||| |||||| |||||| ||||| |||||||||| ||| |||||| |||||||||| ||||| ||| || |||||||||| ||||||| |||||||||||| ||| ||||||||||||| ||| ||| |||| |||| |||| ||| |||| || ||||||| |||||| ||| ||| |||| ||||| ||||||| ||| || |||| |||||||||| ||| |||||||| ||| |||| ||||||||| || ||||||||||| ||||||||| |||||||| ||||||| |||| ||| |||||||| || ||| ||||||||| |||||||

Harms Results

Network Meta-Analysis

Harms were not evaluated in the sponsor-submitted NMA.

Matching-Adjusted Indirect Comparison

||||||||| || ||| |||||| |||||| ||||||| |||| ||||| ||| ||| |||| ||| |||||| ||||||||| ||| ||| || ||| ||||| ||||| ||||||||||| ||||| ||| || |||||||||| ||||||| |||||||||||| ||| ||||||||||||| ||| |||| ||| ||||| ||| ||||| | | |||| ||||| ||| |||||||||||||||| ||| || |||| ||| ||| || ||||||||| ||||| ||||| ||||||||||| |||||||||||| ||| |||||||| |||| ||||||||||||| ||| |||||||| |||| |||| |||| ||| |||| || ||||||| |||||||| |||| |||| |||| ||| |||| || ||||||| |||||| |||| |||| |||| ||| |||| || ||||||| ||| |||||| |||| |||| |||| ||| |||| || |||||||||||| ||| || ||||||||| |||| |||| |||||||| || ||| |||||||| |||| ||||||| ||| |||||||||| ||| |||||| ||||||||| ||||||| ||| ||||||| ||| ||| |||||||||||| ||| ||| ||||| |||| |||||||||||| ||| |||||||| |||| ||||||||||||| ||| ||| ||| || |||||||| |||| |||| |||| ||| |||| || ||||||| ||| |||||||||| |||| |||| |||| ||| |||| || ||||||| ||||| ||||||||||||| ||| |||||||| ||| |||||||||||| |||| |||| |||| ||| |||| || |||||| ||||| ||||||||||| ||| ||||| ||| || |||||||||| ||||||| |||||||||||| ||| ||||||||||||| ||| ||||| | | ||||

Critical Appraisal

Network Meta-Analysis

The sponsor-submitted NMA was informed by a targeted literature review (TLR) and SLR, which included planned searches of multiple databases; however, clinical trial databases were not searched, and given the methodology of conducting a TLR followed by an SLR, it remains unclear if any relevant studies were missed. A quality assessment of the included studies was conducted; however, the results were not included. As part of the feasibility assessment for the NMA, a list of potential treatment-effect modifiers was developed from subgroups of the included trials, although these were not powered to detect differences and no formal search of potential effect modifiers was conducted.

Based on the results of the feasibility assessment, PFS was the only outcome evaluated in the TN CLL network meta-analysis, as OS was deemed too immature for comparison by NMA. For the TN CLL network meta-analysis, 3 of the trials included in the SLR (the RESONATE-2, ELEVATE-TN, and CLL14 trials) were excluded from the NMA because of substantial differences in effect modifiers across trials, which may have increased transitivity but reduced the robustness of the network. However, no sensitivity analysis was performed to determine the impact of excluding these trials.

Baseline characteristics between studies were generally similar, apart from the included populations; the SEQUOIA and ALPINE studies included patients with CLL and SLL, whereas all other studies included only patients with CLL. The proportion of patients with SLL in the SEQUOIA and ALPINE trials likely had little impact on the results, although this was not explored. No adjustments for differences in baseline characteristics were conducted.

For both PFS and OS (when reported) in the TN and r/r CLL network meta-analyses, results were mostly associated with wide 95% CrIs, suggesting notable imprecision. Although results of both NMAs suggested that zanubrutinib is favoured over most treatments, particularly for PFS, it should be noted that the results were produced using a fixed-effects model, and it is uncertain if the fixed-effects model was the appropriate model to use in these comparisons, due to the lack of reporting of model statistics. As a result, the superiority of zanubrutinib cannot be concluded from the NMA.

Matching-Adjusted Indirect Comparison

The choice to conduct an MAIC was justified, considering the lack of comparison included in the sponsor-submitted NMA for the relevant comparators of acalabrutinib and of VenG in the TN setting. As in the NMA, a major difference in populations was the inclusion of patients with either CLL or SLL in the zanubrutinib studies (SEQUOIA and ALPINE), whereas all comparator studies only included patients with CLL. Additionally, in the r/r MAIC, the population for the ELEVATE-RR study only included high-risk patients (patients with 17p deletion and/or 11q deletion); therefore, the population in the ALPINE study was also restricted to the subset of high-risk patients, which resulted in reduced sample sizes in the zanubrutinib and ibrutinib arms. The removal of patients who were not at high risk from the zanubrutinib studies may render the results for the r/r CLL matching-adjusted indirect comparison not generalizable to the r/r CLL population in Canada. ||||||||||||| |||||||| || ||| |||||||| |||||| ||||| || ||| |||||||||||| |||| ||| ||||||| ||| |||| ||| |||||||| |||||| ||| ||||||| |||||||| ||||||| |||| ||||| || ||||| || ||| || |||||||||| ||| ||||| ||| |||||||||||| ||| ||||| ||| ||||||||| || ||| ||| |||||||||||| || || ||||||| ||| |||| || |||| ||||||||| ||| ||| || ||||||||||||| || || |||| || ||||||||| ||| || ||| ||||||||| |||||||| |||| ||||||| || |||| || ||| ||| ||||| |||| ||||||||| |||||||||| |||||| ||||||||||| ||| |||| |||| || ||||||||||| |||||||| ||||||| |||||||||||| ||| ||| ||||||||||| || ||| || ||| |||||||| ||||||| || |||| ||| |||| ||||||| ||| || |||| |||||||||| ||||||||| ||| ||||| |||| ||| ||||||||| ||| |||| |||||||| ||| |||||| ||||||||| ||||| ||| || |||||||||| |||||| ||| || |||||||||||||| |||| |||| ||||||||||||||| ||| || |||| |||||||| ||||| |||| ||||| ||||||||||| |||||| ||||||||| |||||| ||||| |||||||||||| ||| |||||||| |||| ||||||||||||| ||| ||||||||| ||||||||| ||||||| ||| |||||| || ||| || |||||||| || ||| ||| |||||||| |||||||||||| ||| |||||||| |||| ||||||||||||| ||| |||||||| ||| |||||||||| |||| ||||| ||||||||||||| ||| |||||||| ||| ||||||||||||| || || ||||| |||||| |||| ||| |||||| |||| |||||||||| |||| |||| ||| |||| |||||||||| ||||||| ||||||||||| || ||||||||||| |||||||| ||| |||||| ||||||||||

Overall, there were multiple limitations of the sponsor-submitted MAIC, such as the reduction in sample sizes in both the TN and r/r populations, as well as the heterogeneity in baseline characteristics across studies leading to uncertainty about the overall generalizability of the results to the population in Canada and wide 95% CIs leading to imprecision and uncertainty in the results.

Studies Addressing Gaps in the Pivotal and RCT Evidence

A lack of evidence for zanubrutinib’s safety and effectiveness in previously treated patients with CLL who could not tolerate existing BTK inhibitors (ibrutinib and acalabrutinib) was identified as a gap in evidence.

Description of Studies

One ongoing phase II, multicentre, single-arm study evaluating the safety and efficacy of zanubrutinib in patients with previously treated B-cell malignancies, including CLL, who are intolerant of ibrutinib and/or acalabrutinib was submitted by the sponsor to address a gap in evidence. Of the estimated 90 participants, 67 patients were enrolled as of data cut-off date of September 8, 2021. In cohort 1, 57 patients had prior experience with ibrutinib, and in cohort 2, 10 patients had prior experience with acalabrutinib, alone or in addition to ibrutinib. Of the 67 patients enrolled, 43 (64.2%) were diagnosed with CLL.

Efficacy Results

Based on outcomes measured in 64 patients with a study duration of more than 90 days, disease was under control (i.e., stable disease or better) in 60 (93.8%) patients. In about two-thirds (64.1%) of patients, their disease condition improved while taking zanubrutinib. Two (3.1%) patients, 1 from each cohort, experienced progression on zanubrutinib as of the data cut-off date.

Harms Results

Overall, 34 of 57 (59.6%) patients on prior ibrutinib and 7 of 10 (70%) patients on prior acalabrutinib did not experience a recurrence of any intolerance event while taking zanubrutinib. Of note, 1 patient (1.5%) discontinued zanubrutinib due to a recurrence of a prior intolerance event (myalgia while taking acalabrutinib). As for severity, 25 of 38 grade 3 events (65.8%) that occurred in the ibrutinib group and 3 of 4 grade 3 events (75.0%) that occurred in the acalabrutinib group did not recur on zanubrutinib. None of the grade 4 intolerance events (2 cases of neutropenia, 1 case of alanine aminotransferase increase, and 1 case of aspartate aminotransferase increase) recurred. Among the intolerance events that did recur in patients on zanubrutinib, the recurrent events were mainly of lower severity (26 of 34 events [76.5%] for ibrutinib intolerance and 1 of 3 events [33.3%] for acalabrutinib intolerance), and none of the events recurred at a higher severity.

Critical Appraisal

The 2 pivotal trials, SEQUOIA and ALPINE, had exclusion criteria for patients with CLL who had been treated with a BTK inhibitor, but Study 215 addresses a gap in evidence by including such patients. However, there are a few limitations. As a single-arm trial, Study 215 does not address the comparative effectiveness of zanubrutinib. Second, as Study 215 is still ongoing, the interim data may overestimate the safety profile of zanubrutinib. In addition, a small sample size (N = 67) with a subgroup of patients with CLL (n = 43; 64.2%) introduces uncertainty in the results and issues with generalizability. Last, none of the study sites are in Canada, which may raise an issue about the external validity of the study results.

Conclusions

Patients and clinicians highlighted the need for new effective treatments that prolong life, control disease and symptoms, maintain quality of life, and reduce side effects compared to current treatments. According to 1 pivotal trial, zanubrutinib demonstrated a clinically meaningful improvement in PFS compared with BR in TN patients with CLL who were without 17p deletion. The results of the NMAs suggest that zanubrutinib was favoured in TN patients over all active comparators (BR, GClb, and RClb) except ibrutinib. In r/r patients with CLL, zanubrutinib demonstrated a clinically meaningful improvement in ORR compared with ibrutinib; the results of the NMAs indicated that zanubrutinib was favoured over acalabrutinib and BR, but not VenR, for PFS in r/r. |||||||| ||| ||||||| || ||| |||| ||||||| |||| ||||| ||| || |||||||||| ||||||| |||||||||||| ||| ||||| |||||||||| || ||| || ||| ||| ||| ||||||||| |||||||| || ||||||||| |||||||| ||||||||| ||| |||||||| || |||||||||||| || |||||||||| ||| ||| |||||||| ||||||| |||||||||||| OS data were considered immature and not interpretable at the time of the analysis, but the NMA results suggest that there were no differences in OS in the r/r population. ||| ||||||| || ||| |||| ||||||| |||| ||||| ||| || |||||||||| ||||||| |||||||||||| ||| ||||| |||||||||| || ||| || ||| ||| ||| ||||||||| ||| ||| |||| ||| |||||||| ||||||||||| ||||||||||| || ||||| |||||||||| ||| || ||||||||||| || ||| ||||||||||| |||||||| ||| |||| || |||||||||| ||| |||||||||| || |||||||||| ||||||||||| ||| || ||||| |||| ||| ||||| ||||||||| The pivotal study results were subjected to key limitations, such as imbalances in dose reductions, missing doses, and treatment exposure between treatment arms. In addition, limitations such as the exclusion of younger patients without comorbidities, the lack of comparative efficacy for TN patients with 17p deletion, and the use of a comparator treatment in low use in Canada were reported. Furthermore, there is uncertainty in the NMA and MAIC findings, due to the reduction in sample sizes in both the TN and r/r populations during the weighting process, the heterogeneity in baseline characteristics, and the wide CIs, which may limit the interpretability of the comparative efficacy and safety results and compromise the generalizability of the results to patients in Canada. No new safety signals were identified in either TN or r/r patients with CLL.

Introduction

The objective of this report is to review and critically appraise the evidence submitted by the sponsor on the beneficial and harmful effects of zanubrutinib, administered in 80 mg oral capsules, in the treatment of adult patients with CLL, per the indication approved by Health Canada (post-NOC).

Disease Background

The contents of this section have been informed by materials submitted by the sponsor and clinical expert input. The information has been summarized and validated by the CADTH review team.

CLL is characterized by the proliferation and accumulation of small, mature B-cells in the blood, bone marrow, lymph nodes, and lymphoid tissue.^1-3 Patients may present with B symptoms (e.g., fever, chills, night sweats), fatigue, enlarged lymph nodes, or splenomegaly. However, clinical presentation is often asymptomatic.⁴

In Western countries, CLL is the most common type of leukemia, with 2018 Canadian cancer statistics showing an incidence of 6.0 per 100,000 population for newly diagnosed CLL (1,725 new cases).⁵ Canadian mortality data from 2017 showed that 361 men and 250 women died from CLL (a total of 611 patients).²¹

In Canada, the diagnosis of CLL is guided by iwCLL or WHO guidelines.^2,3 Immunophenotyping of CLL cells will show that they co-express CD5, CD19, CD20, and CD23, with characteristically low expression of CD20 (compared with normal B-cells), and each clone is restricted to expressing kappa or lambda immunoglobulin light chains.² Immunophenotyping of peripheral blood may be sufficient, although lymph node or bone marrow biopsy may be helpful if the immunophenotyping results are not conclusive.⁶

Even though a specific companion diagnostic test is not expected to be required for zanubrutinib, similar to the 2 other BTK inhibitors (ibrutinib and acalabrutinib), genetic factors and related testing may play a role in treatment selection among the various options available for TN and r/r patients with CLL. These tests would be performed (where available) before starting other treatment options. Testing for the status of IGHV, TP53, and 17p deletion is useful for guiding personalized treatment. To test for 17p deletion or TP53 mutations and to confirm IGHV status (estimated to be mutated at ≥ 2% compared to germline), fluorescence in situ hybridization (FISH)¹⁰ and sequencing²² are performed, respectively. Retesting of IGHV status after disease progression is not necessary, as mutation status does not change over time.^10,23 However, the evolution of leukemia clones means that testing for 17p deletion and TP53 status should be repeated at each instance of disease progression if it was normal at the start of the last treatment.²³

Standards of Therapy

The contents of this section have been informed by materials submitted by the sponsor and clinical expert input. The information has been summarized and validated by the CADTH review team.

Treatment is generally not required for early asymptomatic disease. Patients with early asymptomatic disease (e.g., Rai stage 0 or Binet stage A) are often followed with a watch-and-wait strategy and routine follow-ups to monitor their disease with a physical examination that includes the palpation of lymph node areas, spleen, and liver, as well as complete and differential blood counts.^6,7 When treatments are indicated based on risk or disease symptoms, the treatment strategy should be personalized according to risk factors, age, fitness, and patient preferences.⁸ In Canada, physicians use 3 risk biomarkers (IGHV status, 17p deletion, and TP53 mutation) to guide personalized treatment.⁸ Very few patients are cured of CLL; therefore, the goals of therapy in most cases are to achieve effective and durable disease control (based on PFS and OS) with minimal toxicity and an acceptable quality of life.^6,9 Even though many patients achieve remission with appropriate treatment, relapse is common. Some patients with r/r disease require subsequent lines of treatment over the course their disease.^7,10

In Canada, CARE treatment algorithms have been published for TN CLL (Figure 1) and r/r CLL (Figure 2).

Figure 1: CARE Guideline for Treatment-Naive Patients With CLL

CLL = chronic lymphocytic leukemia; FCR = fludarabine plus cyclophosphamide plus rituximab; FISH = fludarabine plus cyclophosphamide plus rituximab; IGVH-M = IGVH mutation; IGVH-UM = unmutated IGVH; iwCLL = International Workshop on CLL; TP53-m = TP53 mutation.

Source: 2021 CARE guidelines.²⁴

Figure 2: CARE Guideline for Patients With r/r CLL

BCRi = B-cell receptor signalling pathway inhibitor; BR = bendamustine plus rituximab; CAR = chimeric antigen receptor; FCR = fludarabine plus cyclophosphamide plus rituximab; r/r = relapsed or refractory; VenR = venetoclax plus rituximab.

Source: 2021 CARE guidelines.²⁴

TN Population

TN patients can be classified in 3 practical subgroups based on their clinical characteristics and genetic risk factors.^11,24

For the first subgroup, which consists of younger, fit patients who do not have high-risk genetic factors (17p deletion, TP53 mutation, or unmutated IGHV), FCR is generally considered the standard treatment.^8,11 This subgroup is assumed to be relatively small,²⁵ as the median age at diagnosis of CLL is older than 70 years,^6,26 early-stage asymptomatic disease is managed with a watch-and-wait approach,^6,7 and many patients have high-risk genetic factors.²³

The second subgroup consists of patients who are not fit for FCR because of age or comorbidities and who do not have high-risk genetic factors (17p deletion, TP53 mutation, or unmutated IGHV). Various chemoimmunotherapy combinations (e.g., BR, VenG, GClb) have been used for this subgroup.

For the third subgroup of patients who have high-risk genetic factors (17p deletion, TP53 mutation, or unmutated IGHV), regardless of age and fitness, ibrutinib was historically considered the standard treatment, as those high-risk patients typically have a poor prognosis, fewer therapeutic options, and are likely to obtain the greatest relative clinical benefits from targeted BTK inhibitor treatment.¹¹

r/r Population

In the r/r population, treatment would be started for symptomatic patients, not simply based on progression.⁶ For those who have refractory disease or a short interval of symptomatic relapse (< 12 to < 36 months), a change of treatment class would be offered.^6,7,11 However, rechallenge using the previous treatment regimen might be considered in patients with a prolonged interval to symptomatic relapse.^6,7,11

In general, a BTK inhibitor would be considered for patients with r/r CLL who had received a fixed-duration regimen (e.g., VenG) in the TN setting.¹¹ Patients who received first-line venetoclax-based treatment and experienced a remission of at least 12 months could be eligible for rechallenge with a venetoclax-based regimen. Idelalisib plus rituximab is an infrequently used treatment option that would likely be reserved for patients with r/r CLL who are intolerant of a BTK inhibitor or relapse after several lines of therapy. Allogenic stem cell transplant is another potential option in the r/r setting.¹⁰

According to the clinical expert consulted by CADTH for the purpose of this review, patient age, cumulative illness rating scale (CIRS) score, and features of CLL such as V gene mutational status and cytogenetic profile are factors influencing first-line treatment choice. For younger patients with a good CIRS and no high-risk mutations, FCR can induce a very long remission, and perhaps even a cure. For younger patients with high-risk mutation features such as TP53 mutations, 11q mutations, or unmutated IGHV genes, continuous BTK inhibitors (mostly commonly used in Canada) and fixed-duration treatments with VenG (now being funded) are used. For older patients, fixed-duration treatments with mild alkylating drugs such as GClb can be used for those who are very frail, but BTK inhibitors are now accessible as a first-line option if such patients have unmutated IGHV genes, TP53 mutations, or other high-risk mutations. The clinical expert consulted by CADTH indicated that the most important goals of treatment for CLL are to reverse symptoms, control disease for as long as possible, minimize toxicity due to treatments, and avoid a significant negative impact on quality of life.

Drug Under Review

The key characteristics of zanubrutinib are summarized in Table 4, along with other treatments available for adults with CLL in Canada.

Zanubrutinib is taken orally at doses of 320 mg (four 80 mg capsules) once daily or 160 mg (two 80 mg capsules) twice daily for the treatment of CLL in adults.¹² Zanubrutinib is indicated for the treatment of adults with Waldenström macroglobulinemia, adults with mantel cell lymphoma who have received at least 1 prior therapy, and adults with marginal zone lymphoma who have received at least 1 prior anti-C20-based therapy.¹² Zanubrutinib has been previously reviewed by CADTH for the treatment of patients with Waldenström macroglobulinemia and patients with mantel cell lymphoma who have received at least 1 prior therapy.^13,14 The sponsor’s reimbursement request aligns with the approved Health Canada indication (post-NOC). Zanubrutinib is indicated for the treatment of CLL and SLL in the US¹⁵ and CLL in the EU.¹⁶

Zanubrutinib is a small-molecule BTK inhibitor that inhibits BTK activity by covalently binding to a cysteine residue in the BTK active site. BTK is a signalling molecule for the B-cell antigen receptor and plays a role in cytokine receptor pathways. Active BTK signalling leads to B-cell proliferation, trafficking, chemotaxis, and adhesion of B-cells. Zanubrutinib has been observed to inhibit malignant B-cell proliferation and reduce tumour growth. As a second-generation BTK inhibitor, zanubrutinib is a more selective BTK inhibitor than ibrutinib, with less off-target activity against other kinases, such as TEC, HER2, CSK, EGFR, and IL-2 inducible T-cell kinases. The more selective nature of zanubrutinib is hypothesized to lead to the fewer toxicities associated with the BTK inhibitor class, such as diarrhea, bleeding, atrial fibrillation, rash, and fatigue.¹²

Table 4: Key Characteristics of Zanubrutinib, Acalabrutinib, and Ibrutinib

BTK = Bruton tyrosine kinase; CBC = complete blood count; CLL = chronic lymphocytic leukemia; ECG = echocardiogram.

^aHealth Canada–approved indication.

Sources: Health Canada product monographs for zanubrutinib, acalabrutinib, and ibrutinib.^12,27,28

Stakeholder Perspectives

Patient Group Input

This section was prepared by CADTH staff based on the input provided by patient groups. The full original patient input(s) received by CADTH have been included in the stakeholder section at the end of this report.

LC is a national charity with a mission to advocate and improve access to health care for people in Canada affected by CLL and SLL. LC submitted input based on information collected from an anonymous online survey that was distributed in Canada and in international locations by email and social media from November 2022 to February 2023. A total of 173 people (64 Canadians, 9 Americans, 1 from Costa Rica, and 99 from unknown locations) responded to the survey. Of the respondents, 149 had confirmed CLL, 23 had been diagnosed with SLL, and 1 was newly diagnosed with unknown lymphoma. CLL Canada assisted LC in distributing the survey and preparing the submission.

According to the survey, most patients with CLL and SLL are diagnosed through routine bloodwork and experience no or minor symptoms at the time of diagnosis. For the 122 respondents who rated the impact of their disease as highly negative (3 to 5 out of 5) at the time of diagnosis, the most frequent symptoms were fatigue (reported by 40% of respondents), night sweats (reported by 27%), and body aches and pains (reported by 20%). In terms of the psychosocial impact of CLL and SLL at the time of diagnosis, the most common factors reported by 109 respondents were anxiety and/or worry (reported by 61% of respondents) and stress of diagnosis (reported by 41%). Similarly, for the 109 respondents who reported currently experiencing effects that had a highly negative impact (3 to 5 out of 5), the most frequently reported symptoms were fatigue (reported by 44% of respondents), body aches and pains (reported by 25%), and night sweats (reported by 16%). Up to 75% of the 109 respondents with CLL experienced a negative impact on quality of life, such as anxiety and/or worry (reported by 61% of respondents), stress of diagnosis (reported by 40%), and difficulty sleeping (reported by 37%). Of the 109 respondents who indicated that CLL had a negative impact on daily activities, the most frequently affected activities were travel (reported by 35% of respondents), volunteering (reported by 25%), and spending time with family and friends (reported by 24%). Seventy-six patients said that the following factors were extremely important when considering a novel therapy over their current treatment option(s): longer survival (reported by 85% of respondents), control of disease and symptoms (reported by 79%), longer remission (reported by 75%), and better quality of life (reported by 66%). Of the 77 patients who responded to a question about the importance of choice and options when deciding on a CLL treatment course, 60% said it is extremely important to have choice and 65% said it is extremely important to have a higher number of CLL and SLL treatment options available. When asked about a preference for route of administration (oral pill versus IV), 63 of 77 patients (82%) confirmed that they would prefer oral administration. Eleven patients (10 who had been previously treated) had experience with zanubrutinib for CLL treatment. Two patients said they are in remission (1 after 6 months and 1 after 1 to 2 years of zanubrutinib treatment) and 5 patients indicated that zanubrutinib controlled their CLL or SLL symptoms better than their previous treatments. Seven respondents are still on zanubrutinib treatment and 1 patient stopped. Four of 11 patients reported that they did not experience any side effects, and 8 patients reported that the side effects of zanubrutinib were less severe than those they had experienced with previous therapies. Symptoms reported were fatigue, easy bruising and/or bleeding, confusion or memory loss, diarrhea, muscle or joint pain, peripheral edema, hypertension, and localized infections. Two patients said that zanubrutinib had a negative impact on their quality of life compared to other treatments.

Clinician Input

Input From Clinical Experts Consulted by CADTH

All CADTH review teams include at least 1 clinical specialist with expertise in 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 1 clinical specialist with expertise in the diagnosis and management of adults with CLL or SLL.

Unmet Needs

According to the clinical expert consulted by CADTH, the first-line treatment options for adults with CLL depend on the patient’s age, CIRS score, and features of the CLL, including the V gene mutational status and cytogenetic profile. For younger patients with a good CIRS score without high-risk mutations, chemoimmunotherapy with FCR can induce a very long mission and perhaps even a cure. For younger patients with higher-risk mutations such as TP53 mutations, 11q mutations, or unmutated IGHV genes, continuous BTK inhibitors are most commonly used in Canada. In addition, fixed-duration treatments with VenG are now funded and being used for this patient population. For older patients, fixed-duration treatments with mild alkylating regimens, such as GClb, are occasionally offered. The clinical expert also stated that many of these patients will have better results with less toxicity with a BTK inhibitor, which is now accessible in the first-line setting.

The clinical expert consulted by CADTH indicated that the most important goals of treatment in patients with CLL is to reverse symptoms and control the disease for as long as possible with treatments that have minimal toxicity and do not have a significant negative impact on quality of life. The clinical expert stated that the biggest limitation of current treatments for patients with CLL is that tumour cell resistance usually occurs and patients stop responding or relapse on therapy. Other limitations include toxicity and drug interactions, the requirement for continuous ongoing treatment, and the fact that there are no curative treatments for patients with CLL.

Place in Therapy

The clinical expert indicated that zanubrutinib is an enhanced BTK inhibitor with increased kinase selectivity. The clinical expert believes that the value of zanubrutinib is incremental rather than transformative, as there are already 2 BTK inhibitors (i.e., ibrutinib and acalabrutinib) commonly used in clinical practice. The clinical expert speculated that zanubrutinib would be a welcome option for the first-line treatment of patients with CLL. The clinical expert does not expect that zanubrutinib will be efficacious in patients who progress on other BTK inhibitors.

Patient Population

According to the clinical expert consulted by CADTH, the patient population for zanubrutinib includes untreated patients aged 65 years and older with a good performance status with or without high-risk mutations (i.e., TP53 mutations, 11q mutations, or unmutated IGHV genes), patients younger than 65 years who are not candidates for FCR, and r/r patients with CLL without transformation or central nervous system involvement. Although patients who have had previous treatment with a BTK inhibitor or who have had a bleeding disorder are not ideal for treatment with zanubrutinib, the clinical expert consulted by CADTH indicated that zanubrutinib may be better tolerated by patients who need to stop other BTK inhibitors because of toxicity.

Assessing the Response Treatment

The clinical expert consulted by CADTH indicated that response to treatment is assessed by changes in peripheral blood counts, which can easily be documented by clinicians looking after patients. The clinical expert stated that repeat bone marrow biopsies are not often performed in clinical practice, but were required as part of the clinical trial formal response criteria. The clinical expert stated that the DOR or response to next treatment are important end points used by clinicians to choose appropriate treatments and to inform prognosis. The clinical expert stated that objective responses often correlate with improvements in cytopenia, which may result in decreased transfusion requirements or decreased risk of infection.

Discontinuing Treatment

According to the clinical expert consulted by CADTH, disease progression, measured by increasing lymphocyte count or worsening cytopenia, is a major reason for discontinuing treatment with zanubrutinib. Enlarged peripheral nodes or an enlarged spleen on treatment could indicate transformation to a more aggressive lymphoma, which would require a change in treatment. Toxicities that cannot be managed with dose reductions or a transient drug being paused could also be a reason for stopping treatment. Zanubrutinib must be transiently paused before various surgical procedures, owing to the risk of bleeding.

Prescribing Considerations

The clinical expert indicated that the diagnosis of CLL requires peripheral blood samples for FISH and V gene sequence analysis, both of which have become validated tests and are available at most cancer centres or can be performed centrally by a specialist. The clinical expert stated that zanubrutinib treatment should be managed by a specialist (i.e., hematologist or medical oncologist) who is familiar with this class of drug and can deal with toxicities and optimal dosing.

Clinician Group Input

This section was prepared by CADTH staff based on the input provided by clinician groups. The full original clinician group input(s) received by CADTH have been included in the stakeholder section at the end of this report.

Two clinician groups submitted input. LC, a national, not-for-profit organization for patients with lymphoma and CLL, submitted input collected by hematologists specialized in CLL treatment across Canada through email exchanges and discussion. The OH-CCO Hematology Cancer Drug Advisory Committee, which provides guidance on Provincial Drug Reimbursement Programs and the Systemic Treatment Program, represented by 4 clinicians, submitted information gathered through video conferences and emails.

Unmet Needs

The LC clinician group stated that despite its excellent efficacy, ibrutinib (the first in class BTK inhibitor) has a number of side effects, which result in nearly 20% of patients discontinuing the drug due to intolerance. The group stated that acalabrutinib (a second-generation BTK inhibitor) has become the BTK inhibitor of choice in Canada because its efficacy is equal to that of ibrutinib but it has fewer side effects; however, it has drug-drug interactions with proton pump inhibitors. In addition, some patients are also intolerant of acalabrutinib. Therefore, the clinician group believes that zanubrutinib would provide an additional choice for those who are intolerant of and/or have safety concerns related to current BTK inhibitors. Last, the LC group said that some patients might prefer the once-daily dosing of zanubrutinib over the twice-daily dosing of acalabrutinib. The OH-CCO Hematology Cancer Drug Advisory Committee added that goals of treatment are to improve blood counts, lessen symptoms, improve organomegaly and adenopathy, and improve quality of life. The OH-CCO Hematology Cancer Drug Advisory Committee stated that unmet needs in the CLL population include treatments with a favourable toxicity profile (especially compared to the cardiac toxicity of ibrutinib); treatments that improve PFS and OS for patients in the 17p deletion subgroup; a BTK inhibitor option for low-risk patients in the first-line setting; treatments that have convenient dosing (e.g., once daily); and treatments with fewer drug interactions (e.g., with proton pump inhibitors).

Place in Therapy

According to the LC clinician group, BTK inhibitors are a standard-of-care therapy for patients with CLL. They are the frontline therapy for patients with a poor prognosis in some provinces and for those not fit for intensive fludarabine-based chemoimmunotherapy in other provinces, and are an unrestricted therapy for patients with r/r disease. The LC group said that zanubrutinib would replace 1 of the currently funded BTK inhibitors, but would not significantly replace other CLL therapies. For example, the LC group said it anticipates that zanubrutinib will replace ibrutinib in some of patients who are still receiving ibrutinib and will be used instead of acalabrutinib in patients who initiate therapy (and that zanubrutinib will replace other BTK inhibitors for those starting therapy and provide a switching option for those intolerant of other BTK inhibitors). Last, the LC group said it does not expect that zanubrutinib will change treatment sequencing or guidelines in Canada. The OH-CCO Hematology Cancer Drug Advisory Committee stated that, in Ontario, BTK inhibitors are only used as first-line therapy in high-risk patients and in an r/r setting, and ibrutinib and acalabrutinib are used under the Exceptional Access Program. The OH-CCO Hematology Cancer Drug Advisory Committee added that zanubrutinib would be another BTK inhibitor that could be used in the first-line setting and in patients with r/r CLL.

Patient Population

According to the clinicians, any patient currently eligible for a BTK inhibitor should be eligible for zanubrutinib. This includes all patients with r/r CLL who have not progressed on a prior ibrutinib or acalabrutinib; patients intolerant of a prior BTK inhibitor; patients with high-risk CLL (17p deletion or TP53 mutation, and/or unmutated IGHV) of any age in the frontline setting,; and 4) older patients or those unfit for fludarabine-based therapy (in place of chemoimmunotherapy). The OH-CCO Hematology Cancer Drug Advisory Committee added that in first-line and r/r settings, all patients with symptomatic CLL would be suited for zanubrutinib treatment.

Assessing Response to Treatment

Based on LC clinician group input, simple blood tests and physical examinations would be sufficient to determine response to zanubrutinib. The group stated that visits would be every 1 to 3 months at the start of therapy and would be every 3 to 6 months for those achieving long-lasting remission over many years. The group emphasized that no special tests or visits are required. The OH-CCO Hematology Cancer Drug Advisory Committee added that usual response measures for CLL, such as blood counts, lymph nodes, and spleen size, would be used to determine response.

Discontinuing Treatment

The LC clinician group stated that zanubrutinib, like other BTK inhibitors, is provided until disease progression (determined clinically by an increase in palpable lymph nodes or palpable spleen and/or an increase in lymphocytes in simple blood tests) or until unacceptable toxicity. The group emphasized that this approach is already a standard of care in Canada with other BTK inhibitors and would not require any new learning or testing. The OH-CCO Hematology Cancer Drug Advisory Committee added that progressive disease and significant intolerance despite a dose reduction would be considered when deciding whether to discontinue zanubrutinib therapy.

Prescribing Conditions

The clinicians said that only hematologists or oncologists who treat hematologic cancers should prescribe zanubrutinib. Additionally, the LC clinician group said that general practitioners in oncology and other associated professionals working in the care of patients with malignant hematology would be able to prescribe zanubrutinib. The OH-CCO Hematology Cancer Drug Advisory Committee added that hematologists in all settings would be appropriate to diagnose, treat, and monitor patients receiving zanubrutinib.

Additional Considerations

The LC clinician group hopes that zanubrutinib will create more competition in BTK inhibitor class and possibly lower costs. Also, the group expects that patients will have the same access to zanubrutinib as to ibrutinib and acalabrutinib, giving them more choice.

Drug Program Input

The drug programs provide input on each drug being reviewed through CADTH’s reimbursement review processes by identifying issues that may impact their ability to implement a recommendation. The implementation questions and corresponding responses from the clinical expert consulted by CADTH are summarized in Table 5.

Table 5: Summary of Drug Plan Input and Clinical Expert Response

BTK = Bruton tyrosine kinase; CIRS = cumulative illness rating scale; CLL = chronic lymphocytic leukemia; pERC = CADTH pan-Canadian Oncology Drug Review Expert Review Committee.

Clinical Evidence

The objective of CADTH’s Clinical Review Report is to review and critically appraise the clinical evidence submitted by the sponsor on the beneficial and harmful effects of zanubrutinib oral capsules, 80 mg, for the treatment of CLL in adults. The focus will be placed on comparing zanubrutinib to relevant comparators and identifying gaps in the current evidence.

A summary of the clinical evidence included by the sponsor in the review of zanubrutinib is presented in 3 sections, and CADTH’s critical appraisal of the evidence is included after each section. The first section, the Systematic Review, includes pivotal studies and RCTs that were selected according to the sponsor’s systematic review protocol. The second section includes indirect evidence from the sponsor. The third section includes additional studies that were considered by the sponsor to address important gaps in the pivotal and RCT evidence.

Included Studies

Clinical evidence from the following sources is included in the CADTH review and appraised in this document:

• 2 pivotal studies, which are ongoing international, phase III, open-label, randomized trials (SEQUOIA and ALPINE)

• 2 indirect treatment comparison (ITCs): an NMA, and an MAIC

• 1 additional study addressing gaps in evidence (Study 215, an ongoing phase II, multicentre, single-arm study).

Pivotal Studies and RCT Evidence

The contents of this section have been informed by materials submitted by the sponsor. The information has been summarized and validated by the CADTH review team.

Description of Studies

Two pivotal trials (SEQUOIA and ALPINE) met in the inclusion criteria for the systematic review conducted by the sponsor, and their characteristics are summarized in Table 6.

SEQUOIA Study (TN Population)

The ongoing international, phase III, open-label, randomized SEQUOIA study (BGB-3111 to 304, NCT03336333) included 4 main cohorts (1, 1a, 2, and 3) of patients with untreated CLL or SLL who required treatment and who were at least 65 years of age or were younger than 65 years but unsuitable for FCR treatment. An overview of the study schematic for the SEQUOIA trial is presented in Figure 3.

Only cohort 1 and cohort 2 of the SEQUOIA trial are summarized in this submission. Cohort 3 was excluded because enrolment is ongoing, and that cohort is receiving zanubrutinib plus venetoclax (a regimen not included in the current reimbursement request). Cohort 1a (patients enrolled only at centres in China) was excluded because the efficacy data were immature at the data cut-off.³⁷ Detailed information about cohorts 1 and 2 is summarized as follows.

• Cohort 1 consisted of 479 patients who were negative for 17p deletion based on central FISH and were randomized in a 1:1 ratio to receive zanubrutinib (n = 241) or 6 cycles of bendamustine plus rituximab (n = 238). Central randomization in cohort 1 was performed using IRT and stratified by age (< 65 years versus ≥ 65 years), Binet stage (C versus A or B), IGHV mutational status (mutated versus unmutated), and geographic region (North America versus Europe versus Asia-Pacific).

• Cohort 2 consisted of 111 patients who were positive for 17p deletion and received zanubrutinib.

The objectives in cohort 1 were to compare the efficacy and safety of zanubrutinib with BR in patients who did not have 17p deletion as a high-risk factor. The objectives in cohort 2 were to evaluate the efficacy and safety of zanubrutinib in patients who had 17p deletion as a high-risk factor. In cohort 1, the primary end point was PFS assessed by the IRC. Secondary end points included ORR (defined as the combined proportions of CRs and PRs) and DOR. Descriptive outcomes are available for the single zanubrutinib arm in cohort 2. Efficacy and safety data were evaluated at a planned interim analysis through a cut-off date of May 7, 2021, and in the final OS analysis through the data cut-off date of October 31, 2022. At the investigator’s discretion, patients who received BR in cohort 1 could cross over and receive zanubrutinib after IRC-confirmed disease progression.

Patients were enrolled at 153 centres in 14 countries and 1 region. No Canadian sites were included in the SEQUOIA trial. Enrolment ultimately exceeded the planned targets in cohort 1 (N = 479 versus 450 planned) and in cohort 2 (N = 111 versus 100 planned).³⁷ Patients were followed during a posttreatment phase that started the day after the last dose of study medication was taken and continued until IRC-confirmed disease progression. A long-term follow-up phase started the day after IRC-confirmed disease progression and continued until the study ended or the patient died (whichever came first). The final analysis was planned after 118 PFS events had taken place, although superiority was met for zanubrutinib at a planned interim analysis (107 PFS events) and unblinding of the study for efficacy was recommended by the data monitoring committee.

Figure 3: Schema for the SEQUOIA Study

B + R = bendamustine plus rituximab; CLL = chronic lymphocytic leukemia; del17P = 17p deletion; SLL = small lymphocytic lymphoma.

^a Randomization for cohort 1 was stratified by age (< 65 years vs. ≥ 65 years), Binet stage (C vs. A or B), IGHV mutational status (mutated vs. unmutated), and geographic region (North America vs. Europe vs. Asia-Pacific).

^b Crossover for patients in arm B to receive next-line zanubrutinib is allowed after IRC-confirmed disease progression.

^c The same randomization stratification factors used for cohort 1 were used for cohort 1a, except for geographic region.

^d Cohort 2 (arm C) was closed to enrolment when the arm C sample size (approximately 100 patients) was reached.

^e Cohort 3 (arm D) was opened for enrolment in selected countries and/or sites after arm C closed. Arm D was closed to enrolment when the arm D sample size was reached.

Source: SEQUOIA Clinical Study Report.¹⁸

ALPINE Study (r/r Population)

ALPINE, which is an ongoing international, phase III, open-label, randomized study (BGB-3111 to 305, NCT03734016), compared zanubrutinib to ibrutinib in patients with r/r CLL or SLL (≥ 1 previous treatment). An overview of the study schematic for the ALPINE trial is presented in Figure 4. The primary end point was investigator-assessed ORR, and secondary end points included ORR by IRC, PFS by IA and IRC, and OS. Efficacy and safety data were evaluated at a planned interim analysis through a cut-off date of December 31, 2020,⁴⁰ in the final ORR analysis through a cut-off date of December 1, 2021, and in the final PFS analysis with a cut-off date of August 8, 2022. Results are presented for all cut-off dates. The objective of the ALPINE trial is to compare the efficacy and safety of zanubrutinib with ibrutinib.

This study is currently being conducted at 113 study centres in 15 countries. No Canadian sites were included in the ALPINE trial. Central randomization was performed using IRT and stratified by age (< 65 years versus ≥ 65 years), geographic region (China versus non-China), refractory to last therapy (yes or no), and 17p deletion and/or TP53 mutation status (present or absent).

Enrolment ultimately exceeded the planned target (N = 652 versus 600 planned), and the primary efficacy end point (investigator-assessed ORR) was evaluated at a planned interim analysis in the first 415 patients who were randomized to the zanubrutinib arm (n = 207) or the ibrutinib arm (n = 208). At the interim analysis, the study met its primary end point based on the superior efficacy of zanubrutinib compared to ibrutinib in the investigator-assessed ORR, and consistent results were observed in the final ORR and PFS analyses.

Figure 4: Schema for the ALPINE Study

CLL = chronic lymphocytic leukemia; SLL = small lymphocytic lymphoma; vs. = versus.

Note: Randomization was stratified by age (< 65 vs. ≥ 65 years), region (China vs. non-China), refractory status (yes vs. no), and 17p deletion and/or TP53 mutation status (present vs. absent).

Sources: ALPINE interim and final ORR Clinical Study Reports.^20,29

Table 6: Details of the Pivotal Studies and RCT Evidence Identified by the Sponsor

BTK = Bruton tyrosine kinase; CLL = chronic lymphocytic leukemia; CR = complete response; CRi = complete response with incomplete bone marrow recovery; CYP3A = cytochrome P450, family 3, subfamily A; ECOG PS = Eastern Cooperative Oncology Group Performance Status; EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; FCR = fludarabine, cyclophosphamide, and rituximab; FISH = fluorescence in situ hybridization; IRC = independent review committee; iwCLL = International Workshop on CLL; ORR = overall response rate; OS = overall survival; PFS = progression-free survival; PR = partial response; PRL = partial response with lymphocytosis; SLL = small lymphocytic lymphoma.

Note: Details from the table have been taken from the sponsor’s Summary of Clinical Evidence.⁴

Sources: SEQUOIA Clinical Study Report,³⁷ ALPINE final ORR and final PFS Clinical Study Reports.^19,20

Protocol Amendment

SEQUOIA Trial

There were 10 protocol amendments reported for the SEQUOIA trial. Of these 10, the following were particularly of note and impactful. Protocol amendment 1.0, made November 27, 2018, added eligibility for patients with a history of localized prostate cancer to enrol in the SEQUOIA trial. Protocol amendment 2.0, made April 1, 2019, removed eligibility for patients with active and/or ongoing autoimmune anemia and/or autoimmune thrombocytopenia. Protocol amendment 4.0, made February 10, 2021, allowed patients in arm B of cohort 1 or 1a to cross over to receive next-line treatment with single drug zanubrutinib following disease progression confirmed by IRC.

ALPINE Trial

There were 7 protocol amendments reported for the ALPINE trial. Of these 7, the following was particularly of note and impactful. Protocol amendment 1.0, made August 29, 2019, removed eligibility for patients with active and/or ongoing autoimmune anemia and/or autoimmune thrombocytopenia.

Populations

Inclusion and Exclusion Criteria

The key inclusion and exclusion criteria applied to the SEQUOIA and ALPINE trials are summarized in Table 6.

SEQUOIA Trial

Patients included in cohort 1 and cohort 2 of the SEQUOIA study had previously untreated CLL or SLL, required treatment, and were older (≥ 65 years) or not eligible for FCR treatment (patients < 65 years old with 1 or more of the following: impaired creatinine clearance [< 70 mL/min]; a CIRS score > 6; or a history of severe infection or multiple infections in the previous 2 years).³⁷ Cohort 2 included patients with 17p deletion as a high-risk factor; these patients were excluded from cohort 1, although cohort 1 included other high-risk patients with mutated TP53 or unmutated IGHV.³⁷ Patients with any history of prolymphocytic leukemia or Richter’s transformation, any currently active clinically significant cardiovascular disease, any active infection (including hepatitis B, hepatitis C, or HIV), or requiring ongoing treatment with a strong CYP3A inhibitor or inducer were excluded from the SEQUOIA trial.

ALPINE Trial

Patients included in the ALPINE study were adults (≥ 18 years old) who had r/r CLL or SLL after at least 1 prior systemic therapy, with or without high-risk genetic factors, and who required treatment based on progressive bone marrow failure, splenomegaly, lymphadenopathy, progressive lymphocytosis, and/or constitutional symptoms (unintentional weight loss, significant fatigue, fever or night sweats without evidence of infection).^38,40 Patients with any history of prolymphocytic leukemia or Richter’s transformation, any currently active clinically significant cardiovascular disease, malignancy in the previous 3 years (some exceptions), any history of severe bleeding disorders, a recent history of stroke or intracranial hemorrhage, any active infection (including hepatitis B, hepatitis C, or HIV), prior treatment with a BTK inhibitor, an ongoing need for corticosteroid treatment, a need for treatment with warfarin or other vitamin K antagonist, or a need for ongoing treatment with a strong CYP3A inhibitor or inducer were excluded from the ALPINE trial.

Interventions

SEQUOIA Trial

The SEQUOIA study was designed as an open-label trial because of the different routes of administration of zanubrutinib (oral) and BR (IV). The BR regimen was selected as the comparator in cohort 1 because it was recommended at the time as first-line therapy for patients with CLL who were at least 65 years old or who were younger and had significant comorbidities.⁴⁶ The FCR regimen was also a standard first-line treatment at the time, but its use is generally restricted to young and fit patients.²⁷ In cohort 1 (for the zanubrutinib arm) and cohort 2, zanubrutinib was administered orally at 160 mg (two 80 mg capsules) twice daily, with treatment continuing until disease progression, unacceptable toxicity, death, withdrawal of consent, loss to follow-up, or study termination.³⁷ In cohort 1, patients randomized to the comparator arm received 6 cycles of BR, with bendamustine administered intravenously at 90 mg/m² per day on the first 2 days of each cycle and rituximab administered intravenously at 375 mg/m² for cycle 1 and 500 mg/m² for cycles 2 to 6. A comparator arm was not used for cohort 2 because patients with 17p deletion are not indicated to receive chemoimmunotherapy, owing to the poor response observed in this subgroup. Zanubrutinib dose interruptions and modifications were prespecified for hematologic and nonhematologic toxicities. At the first occurrence of a toxicity, zanubrutinib was restarted at the original dose of 160 mg twice daily, with subsequent reductions at the second toxicity occurrence (restarted at 80 mg twice daily) and the third toxicity occurrence (restarted at 80 mg once daily). Zanubrutinib was to be discontinued at the fourth occurrence of a toxicity. Patients should not receive other anticancer therapies (cytotoxic, biologic, or immunotherapy) while on treatment in this study. Other anticancer therapy should not be administered until disease progression (per clinical practice standards at the study centre), unmanageable toxicity, or until no further clinical benefit occurs, which requires permanent discontinuation of the study drug.³⁷

ALPINE Study

The ALPINE study was designed as an open-label trial. Ibrutinib was selected as the comparator. Oral treatment using zanubrutinib (160 mg [as two 80 mg capsules] twice daily) or ibrutinib (420 mg once daily based on local prescribing guidelines) was continued until disease progression, unacceptable toxicity, death, withdrawal of consent, loss to follow-up, or study termination.^38,40 Dose interruptions and modifications for both zanubrutinib and ibrutinib were prespecified for hematologic and nonhematologic toxicities, although local prescribing guidelines always took precedence. The dose reduction instructions for zanubrutinib in the ALPINE study were the same as in the SEQUOIA study. Patients could not receive other anticancer therapies (including but not restricted to chemotherapy, immunotherapy, corticosteroids for the treatment of CLL, experimental therapy, radiotherapy, and herbal medications) while on treatment in this study. Other anticancer therapies could not be administered until disease progression (per clinical practice standards at the study centre), unmanageable toxicity, or until no further clinical benefit occurs, which requires permanent discontinuation of the study drug.^38,40

Permitted medications for patients in the ALPINE and SEQUOIA trials were blood product transfusion and growth factor support, per standard of care and institutional guidelines; corticosteroids for non-CLL indications, except patients could not receive treatment with systemic corticosteroids other than intermittently to control or prevent infusion reactions or for short durations (< 2 weeks) to treat non-CLL-related conditions (e.g., to treat a flare of chronic obstructive pulmonary disease). Chronic systemic corticosteroid use was not permitted, except for adrenal replacement therapy to reduce symptoms, per standard of care and institutional guidelines.^37,38,40

Outcomes

The efficacy end points assessed in this Clinical Review Report are provided in Table 7 and subsequently summarized. Summarized end points are those included in the sponsor’s Summary of Clinical Evidence, as well as any identified as important to this review, according to stakeholders (for example, the clinical expert, clinician groups, or patient groups).

Table 7: Outcomes Summarized From the Pivotal Studies and RCT Evidence Identified by the Sponsor

EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; IA = investigator assessment; IRC = independent review committee; ORR = overall response rate; OS = overall survival; PFS = progression-free survival; RCT = randomized controlled trial.

Note: Details from the table have been taken from the sponsor’s Summary of Clinical Evidence.⁴

^aStatistical testing for these end points was adjusted for multiple comparisons (e.g., O’Brien-Fleming–type Lan-DeMets alpha spending function). For OS, EORTC QLQ-C30, and EQ-5D-5L in the SEQUOIA trial, adjustment was performed using the fixed-sequencing Bonferroni method.

^bMultiplicity due to multiple end points and multiple tests will be handled per the graphical approach described by Maurer and Bretz (2013)⁴⁰ using fixed-sequence hierarchical testing with a study-wide 1-sided significance level of 0.025. Under this procedure, secondary end points will be tested only if the primary end point is significant.

^cIf the noninferiority of ORR per IA is statistically significant, the key secondary end point of atrial fibrillation and flutter incidence will be tested at the interim and final analyses of ORR with the same 1-sided significance levels as ORR, but will be tested separately from the fixed-sequence hierarchical testing.

Sources: SEQUOIA Clinical Study Report,³⁷ ALPINE final ORR and final PFS Clinical Study Reports.^19,20

Progression-Free Survival

In SEQUOIA cohort 1, PFS was the primary end point and was defined as the time from randomization to the earlier of disease progression or death due to any cause, using IRC assessment. Investigator-assessed PFS was the secondary end point for SEQUOIA cohort 1. In cohort 2, PFS by IA and by IRC were secondary end points.

The duration of PFS will be right-censored for patients in the SEQUOIA trial who met 1 of the following conditions:

• no baseline disease assessments, date of randomization (censored)

• starting a new CLL-related or SLL-related therapy before documentation of disease progression or death, date of last disease assessment before the start of a new CLL- or /SLL-related treatment (censored)

• death or disease progression immediately after 2 or more missed consecutive disease assessments, date of last disease assessment with documented nonprogression (censored)

• alive without documentation of disease progression before the data cut-off date, date of last disease assessment (censored).

In the ALPINE trial, PFS by IA was a key secondary end point and PFS by IRC was a secondary end point. The definition of PFS was the same as in the SEQUOIA trial.

Censoring rules for PFS for patients in the ALPINE trial were as follows:

• no baseline disease assessments, date of randomization (censored)

• progressive disease or death more than 6 months after the last disease assessment (more than 12 months if a patient is on the disease assessment schedule of every 24 weeks), date of the last disease assessment before death or progressive disease (censored)

• alive without documentation of progressive disease, date of last disease assessment (censored).

Overall Survival

In SEQUOIA cohorts 1 and 2 and in the ALPINE trial, OS was a secondary end point. In both the SEQUOIA and ALPINE trials, OS was defined as the time from randomization to the date of death (whatever the cause). Patients who are alive or lost to follow-up as of the data analysis cut-off date will be censored at the date the patient is last known to be alive.

Overall Response Rate

The secondary end points of SEQUOIA cohorts 1 and 2 were ORR assessed by IRC and by the investigators. ORR was defined as the crude proportion of patients in each treatment group who achieved a CR, CRi, PR, nodular PR, or PRL, in accordance with the modified 2008 iwCLL guidelines with modification for treatment-related lymphocytosis for patients with CLL.^1,35

In the ALPINE trial, ORR per IA was the primary end point and ORR per IRC was the secondary end point. ORR was defined as the proportion of patients in each treatment group who achieved a CR, CRi, PR, or nodular PR, in accordance with the modified 2008 iwCLL guidelines with modification for treatment-related lymphocytosis for patients with CLL.^1,35

Duration of Response

In SEQUOIA cohorts 1 and 2 and the ALPINE trial, DOR assessed by IRC and by IA were secondary end points. In both the SEQUOIA and ALPINE trials, DOR was defined as the date that response criteria are first met to the date that disease progression is objectively documented or death, whichever occurs first. DOR was determined by IRC and by IA using the 2008 iwCLL criteria with modification for treatment-related lymphocytosis for patients with CLL.^1,35

Time to Treatment Failure

In the ALPINE trial, the analysis of time to treatment failure, defined as the time from randomization to discontinuation of the study drug for any reason, was the secondary outcome. Time to treatment failure was censored at the data cut-off for the patients who did not discontinue study treatment.

European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30

In the SEQUOIA trial, HRQoL measured by EORTC QLQ-C30 was the secondary end point for cohort 1 and an exploratory end point for cohort 2. In the ALPINE trial, HRQoL measured by EORTC QLQ-C30 was the secondary end point. The EORTC QLQ-C30 is a questionnaire developed to assess quality of life in patients with cancer. The EORTC QLQ-C30 includes 30 separate questions (items) resulting in 5 functional scales (physical functioning, role functioning, emotional functioning, cognitive functioning, and social functioning), 1 global health status scale, 3 symptom scales (fatigue, nausea and vomiting, and pain), and 6 single items (dyspnea, insomnia, appetite loss, constipation, diarrhea, and financial difficulties).¹⁷ The recall period is 1 week (the past week). The EORTC QLQ-C30 has been widely used among cancer patients in general, and specifically among patients with non-Hodgkin lymphoma. A high score on a functional scale represents a high level of functioning, whereas a high score on a symptom scale or single item represents a high level of symptomatology. Refer to Table 8 for the psychometric properties and the minimally important difference.

EQ-5D-5L

In the SEQUOIA trial, HRQoL measured by EQ-5D-5L was a secondary end point for cohort 1 and an exploratory end point for cohort 2. In the ALPINE trial, HRQoL measured by EQ-5D-5L was a secondary end point. The EQ-5D-5L is a standardized instrument used to measure health outcomes.³⁶ Patients self-rate their current state of mobility, self-care, usual activities, pain and/or discomfort, and anxiety and/or depression by choosing 1 of 5 possible responses that record the level of severity (no problems, slight problems, moderate problems, severe problems, or extreme problems) for each dimension. The questionnaire also includes a visual analogue scale to self-rate general health state on a scale from “the worst health you can imagine” to “the best health you can imagine.” Refer to Table 8 for the psychometric properties and the minimally important difference.

Safety

Incidence of Atrial Fibrillation and Flutter

SEQUOIA Trial

The incidence of atrial fibrillation and flutter was not considered an end point in the SEQUOIA trial; however, it was assessed in the safety analysis.

ALPINE Trial

The incidence of atrial fibrillation and flutter, defined as the incidence of TEAEs related to either atrial fibrillation or atrial flutter, was the key secondary end point in the ALPINE trial.

Table 8: Summary of Outcome Measures and Their Measurement Properties

CLL = chronic lymphocytic leukemia; EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; HRQoL = health-related quality of life; MID = minimally important difference; QoL = quality of life; VAS = visual analogue scale.

AEs, SAEs, and Notable Harms

For both the SEQUOIA and ALPINE trials, TEAEs included any AE with an onset date on or after the first dose of a study drug up to 30 days after study drug discontinuation or the start of a new anticancer therapy, whichever comes first. SAEs included any event that resulted in death, was life-threatening, required hospitalization or prolongation of existing hospitalization, resulted in disability and/or incapacity, was a congenital anomaly and/or birth defect, or was determined to be a significant medical AE by the investigator, based on medical judgment (e.g., that may jeopardize the patient or may require medical and/or surgical intervention to prevent 1 of the outcomes listed previously).

In both the SEQUOIA and ALPINE trials, nonhematologic AEs and SAEs were assessed and graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events version 4.03.³⁷ Hematological toxicities were graded based on the grading scale for hematologic toxicity in CLL studies.⁵ The investigators assessed the severity of each reported AE and its potentially causal relationship with the study drug. In both the SEQUOIA and ALPINE trials, the definition of AESIs was AEs that are known to be associated with BTK inhibitor treatment (hemorrhage, atrial fibrillation and flutter, hypertension, second primary malignancies, tumour lysis syndrome, infection, and cytopenias [neutropenia, thrombocytopenia, and anemia]). For patients receiving zanubrutinib, all AEs and SAEs were reported until the latest occurrence of 1 of the following: 30 days after the last dose of zanubrutinib, disease progression, or the start of a new CLL therapy in the absence of progression.

Statistical Analysis

Sample Size and Power Calculation

SEQUOIA Trial

The sample size calculation for cohort 1 is based on the primary efficacy analysis of PFS by ICR, which compares arms A and B in cohort 1. Assuming that the PFS HR (arm A/arm B) in cohort 1 is 0.58, 118 events are required to achieve 83.5% power at a 2-sided alpha of 0.05 to reject the null hypothesis, and 1 interim analysis is planned after 73% of the target number of events at the final analysis. If 450 patients are enrolled to cohort 1 and randomized in a 1:1 ratio to arms A and B over a 25-month period, 118 PFS events are expected to be accumulated at 41 months from the study start. This assumes a median PFS in arm B of 42 months and that PFS follows an exponential distribution.⁴¹ Approximately 710 patients will be enrolled; of those, 450 patients without the 17p deletion mutation in cohort 1 will be available for the primary efficacy analysis, as will approximately 80 additional patients from Chinese sites without the 17p deletion mutation in cohort 1a, approximately 100 patients with the 17p deletion mutation in cohort 2, and approximately 80 patients with the 17p deletion mutation in cohort 3.

ALPINE Trial

The sample size calculation is based on the primary efficacy analyses for the primary end point of ORR per IA. Assuming a response ratio (zanubrutinib arm/ibrutinib arm) of 1.03 (72%/70%), 600 patients will provide more than 90% power to demonstrate the noninferiority of zanubrutinib to ibrutinib at the noninferiority margin of 0.8558 (response ratio) and a 1-sided alpha level of 0.025 when there is 1 interim analysis at 69% information fraction. The response rate for ibrutinib is approximated from published clinical data.⁴² Assuming an HR of 0.9 (zanubrutinib arm/ibrutinib arm), 205 PFS events are required to achieve 80% power at a 1-sided alpha of 0.025 to demonstrate the noninferiority of zanubrutinib to ibrutinib at the noninferiority margin of 1.3319 for an HR for the key secondary end point of PFS per IA. If the 600 patients are randomized in a 1:1 ratio to the 2 arms over a 24-month period, including a 9-month ramp-up period, before reaching peak enrolment of 33 patients per month with a 0.0017 per month hazard rate for dropout, 205 PFS events are expected to accumulate in 45 months from the study start. A median PFS of 47 months for ibrutinib and an exponential distribution for PFS are also assumed.

Justification of the Noninferiority Margin for ORR

A noninferiority margin of 0.8558 in the response ratio was derived using the 95% to 95% fixed-margin approach.⁴³ In the RESONATE trial, the ibrutinib effect over ofatumumab represented by the ratio of response rate (PR or higher) was 10.43 with a 95% CI of 5.2 to 21.0 based on the IRC assessment.⁴⁴ In the RESONATE-2 trial, the ibrutinib effect over chlorambucil represented by the ratio of response rate (PR or higher) was 2.33 with a 95% CI of 1.83 to 2.97 based on the IRC assessment.⁴⁵ In a fixed-effects meta-analysis of the 2 studies using inverse variance weighting, the ibrutinib effect in the response rate ratio is estimated to be 2.7392 with a 95% CI of 2.1781 to 3.4450. Thus, the control arm effect is 2.1781, the lower bound of the 95% CI. Because the effect sizes of ibrutinib are overactive controls in both studies (ofatumumab and chlorambucil, respectively), rather than placebo, the choice of the control arm effect is very conservative and results in a narrow margin.^44,45 Requiring 80% of the control arm effect to be retained (on the log scale) in zanubrutinib to demonstrate noninferiority generates a noninferiority margin of 0.8558 (for the response ratio), which is within the clinically acceptable limit.

Statistical Test or Model

The statistical analyses in the SEQUOIA and ALPINE trials are summarized in Table 9.

SEQUOIA Trial

The primary end point (PFS assessed by the IRC in cohort 1) was analyzed using a log-rank test stratified by randomization stratification factors (age [< 65 years versus ≥ 65 years], Binet stage [C versus A or B], and IGHV mutational status [mutated versus unmutated]). The null and alternative hypotheses for testing the superiority of zanubrutinib to BR were H₀ HR = 1 and H_a HR = 0.58. The HR and its 2-sided 95% CI were estimated from a stratified Cox regression model. The distribution of PFS, including median PFS and PFS rate at selected time points, such as 12 and 24 months, was estimated using the Kaplan-Meier method for each arm. Censoring was performed according to FDA Guidance for Industry.⁴⁶ All inferential statistics were based on a comparison of zanubrutinib and BR in cohort 1 (patients without 17p deletion); summary statistics are reported for zanubrutinib in cohort 2 (patients with 17p deletion). No adjustment was planned for covariates or baseline factors.

Analyses summarized in this review are based on data collected through a data cut-off of May 7, 2021, for the planned interim analysis. The interim analysis had been planned when 86 PFS events were reported by the investigators in cohort 1, although there were 107 PFS events per the IRC assessment at the interim analysis (i.e., 107 of 118, or 91%, of planned events for the final analysis). Based on the interim analysis, the external data monitoring committee determined that superiority was achieved for PFS, and recommended unblinding of the study for efficacy. The final analysis for PFS was not performed because the superiority boundary was met at the interim analysis.

ALPINE Trial

The primary end point (investigator-assessed ORR) was tested for noninferiority using a stratified Wald test and, if noninferiority was demonstrated, superiority was tested using a stratified Cochran-Mantel-Haenszel test. The analyses were adjusted for randomization stratification factors (age [age < 65 versus ≥ 65 years], geographic regions [China versus non-China region], refractory status [yes versus no], and 17p deletion and/or TP53 mutated status [yes versus no]). The monitoring boundaries for the noninferiority and superiority tests were based on the O’Brien-Fleming boundary approximated by the Lan-DeMets spending function. An interim analysis (the first 415 patients randomized), a final ORR analysis, and a final PFS analysis were planned for all end points. The interim analysis occurred approximately 12 months after the first 415 patients had been randomized, whereas the final ORR analysis occurred approximately 12 months after 600 patients had been randomized. The final PFS analysis occurred when 205 PFS events per IA have happened. Analyses summarized in this review are based on data collected through a data cut-off of December 1, 2021, for the planned final ORR analysis and a data cut-off of August 8, 2022, for the planned final PFS analysis. The number of patients at the final ORR and PFS analyses was 652 in both cases.

Superiority testing was performed with a 1-sided significance level of 0.005 at the interim analysis and 0.0235 at the final analysis of ORR, which correspond to chi-square distribution P value cut-offs of 0.0099 and 0.0469, respectively. If zanubrutinib was noninferior and superior to ibrutinib in the investigator-assessed ORR, investigator-assessed PFS was tested for noninferiority using a stratified Wald test and, if noninferiority was demonstrated, superiority was tested using a stratified log-rank test. Both analyses used 1-sided significance levels of 0.02498. The distribution of PFS was estimated for each treatment arm using the Kaplan-Meier method.

If zanubrutinib was noninferior to ibrutinib in the investigator-assessed ORR, the superiority of zanubrutinib to ibrutinib in atrial fibrillation and flutter was tested separately from the fixed-sequence hierarchical testing of ORR and PFS. The analysis of atrial fibrillation and flutter was performed using an unstratified chi-square distribution (based on ≥ 5 patients in the 2 × 2 contingency table) with a 1-sided significance level of 0.005 at the interim analysis and a planned 1-sided significance level of 0.0235 at the final analysis.

Interim Analysis

SEQUOIA Trial

There was 1 interim analysis of PFS by IRC in cohort 1. The O’Brien-Fleming boundary approximated by the Lan-DeMets spending function was implemented for efficacy, and the Haybittle-Peto method was implemented for futility. This analysis was scheduled to occur after approximately 73% of the targeted total PFS events from arms A and B in cohort 1 were reported, which is anticipated to occur approximately 33 months after the first patient was randomized.

ALPINE Trial

There was 1 interim analysis for the noninferiority (and superiority if noninferiority is met) testing of ORR. The interim analysis was performed approximately 12 months after the randomization of 415 patients. The monitoring boundaries for the interim and the final analyses for the noninferiority and superiority tests were based on the O’Brien-Fleming boundary approximated by the Lan-DeMets spending functions.

Multiplicity

SEQUOIA Trial

Multiplicity due to testing of multiple hypotheses for primary end point (PFS by IRC) was adjusted by an O’Brien-Fleming–type Lan-DeMets alpha spending function. Only select secondary end points (OS and patient-reported outcomes) were to be tested in cohort 1 if the primary end point was statistically significant. No inferential testing was done for other secondary end points, including ORR and DOR.³⁷ The planned interim analysis of OS was not expected to have enough power to identify a significant difference between the 2 arms and was conducted with a 1-sided alpha level of 0.00005 to detect statistical significance. Multiplicity in testing OS and patient-reported outcomes was adjusted with the fixed-sequencing Bonferroni method, which, unlike the regular fixed-sequencing method, allows for lower ranked end points to be tested even when higher ranked end points are not statistically significant. The significance level was 0.025 and evenly distributed to OS and patient-reported outcomes. If 1 was positive at the 0.0125 significance level, the other outcome would inherit the significance level and be tested at 0.025 (e.g., if patient-reported outcomes had the smaller P value [< 0.0125], then OS would be tested at a significance level of 0.025).

ALPINE Trial

To control for the study-wide type I error, individual significance levels were adjusted for the tests of the primary end point of ORR per IA (noninferiority and superiority) and the key secondary end point of PFS per IA (noninferiority and superiority). Multiplicity due to multiple end points and multiple tests was handled with the graphical approach, using fixed-sequence hierarchical testing.⁴⁷ Under this procedure, secondary end points were tested only if the primary end point was significant. If the noninferiority of ORR per IA is statistically significant, the key secondary end point of atrial fibrillation and flutter incidence was tested at the interim and final analyses of ORR with the same 1-sided significance levels as ORR, but was tested separately from the fixed-sequence hierarchical testing.

Hypothesis testing was performed according to the multiplicity adjustment flow chart shown in Figure 5. The study-wide 1-sided significance level of 0.025 was passed to subsequent hypothesis tests in the sequence and was distributed for each of the following potential analysis time points based on the known correlation of the interim and final test statistics and corresponding alpha spending function.

1. Interim analysis of ORR

   1. The noninferiority of ORR per IA is tested at a 1-sided significance level of 0.005 based on the O’Brien-Fleming alpha spending function with an information fraction of 64% (415 divided by 652, the total number of randomized patients); if this is not statistically significant, do not conduct any of the remaining hypothesis tests at this analysis time point and continue to the final analysis of ORR.

   2. If the noninferiority of ORR per IA is statistically significant, the superiority of ORR per IA will be tested at a 1-sided significance level of 0.005 based on the O’Brien-Fleming alpha spending function with an information fraction of 64%; if this is not statistically significant, continue to the final analysis of ORR for additional hypothesis testing, starting with the superiority of ORR per IA.

   3. If the superiority of ORR per IA is statistically significant at the interim analysis, PFS per IA will be compared between the 2 treatment arms for descriptive purposes only, not for statistical inference (for either noninferiority or superiority), at this interim analysis. A 1-sided 0.00001 significance level will be spent to account for the increased false-positive rate due to this descriptive analysis.

2. Final analysis of ORR

   1. If the noninferiority of ORR per IA was not statistically significant at the interim analysis of ORR:

      1. The noninferiority of ORR per IA will be tested at a 1-sided significance level of 0.0235; if this is not statistically significant, the study does not meet the primary objective and no additional hypothesis testing will be performed in this study.

      2. If the noninferiority of ORR per IA is statistically significant, the superiority of ORR per IA will be tested at a 1-sided significance level of 0.0235; if this is not statistically significant, no additional hypothesis testing will be performed in this study.

      3. If the superiority of ORR per IA is statistically significant, PFS per IA will be compared between the 2 treatment arms for descriptive purposes only, not for statistical inference (for either noninferiority or superiority). A 1-sided 0.00001 significance level will be spent to account for the increased false-positive rate due to this descriptive analysis.

   2. If the noninferiority of ORR per IA was statistically significant but the superiority of ORR per IA was not significant at the interim analysis of ORR:

      1. The superiority of ORR per IA will be tested at a 1-sided significance level of 0.0235; if this is not statistically significant, no additional hypothesis testing will be performed in this study.

      2. If the superiority of ORR per IA is statistically significant, PFS per IA will be compared between the 2 treatment arms for descriptive purposes only, not for statistical inference (for either noninferiority or superiority). A 1-sided 0.00001 significance level will be spent to account for the increased false-positive rate due to this descriptive analysis.

3. Final analysis of PFS

   1. If the superiority of ORR per IA is statistically significant at either the interim or final analysis of ORR, PFS per IA will be followed until 205 PFS events per IA have occurred for the final PFS analysis; noninferiority of PFS per IA will be first tested at a 1-sided significance level of 0.02498.

   2. If the noninferiority of PFS per IA is statistically significant, the superiority of PFS per IA will be tested at a 1-sided significance level of 0.02498.

Figure 5: Flow Chart for the Multiplicity Adjustment

ORR = overall response rate; PFS = progression-free survival.

Source: ALPINE statistical analysis plan.⁴⁸

Data Imputation Methods

Missing data for the HRQoL, measured by EORTC QLQ-C30 and EQ-5D-5L in the SEQUOIA and ALPINE trials, were analyzed using a repeated-measures mixed model to account for missing data under the missing-at-random assumption.

Subgroup Analyses

SEQUOIA Trial

In cohort 1, subgroup analyses were provided for the primary end point of PFS per IRC and selected secondary efficacy end points; however, no statistical analysis was planned in the SEQUOIA trial. Subgroups reported that were relevant to our protocol included age (< 65 versus ≥ 65 years), sex, disease stage (Binet stage A or B versus Binet stage C and Ann Arbor stage I or II bulky versus Ann Arbor stage III or IV), ECOG PS (0 versus ≥ 1), and high-risk genetic factors (IGHV mutation status [unmutated versus mutated], 17p deletion [present versus absent]), and TP53 mutation status [unmutated versus mutated]).

ALPINE Trial

Subgroup analyses for the primary end point of ORR per IA and selected secondary end points were reported; however, no statistical analysis was planned in the ALPINE trial. Subgroups reported that were relevant to our protocol included age (< 65 versus ≥ 65 years), sex, disease stage (Binet stage of A or B and versus Binet stage C and Ann Arbor stage I or II bulky versus Ann Arbor stage III or IV), ECOG PS (0 versus ≥ 1), and high-risk genetic factors (IGHV mutation status [unmutated versus mutated], 17p deletion [present versus absent]), and TP53 mutation status [unmutated versus mutated]).

Sensitivity Analyses

SEQUOIA Trial

For cohort 1, in the analysis of the primary end point of PFS by ICR, alternative censoring rules, such as not censoring for new anticancer therapy, was used, and the primary analysis was repeated as a sensitivity analysis. PFS by ICR was analyzed using the per-protocol population as well.

ALPINE Trial

Progression-Free Survival

The noninferiority of zanubrutinib to ibrutinib for PFS was tested under the noninferiority margin of 1.3319 (for the HR of zanubrutinib to ibrutinib) using a stratified log-rank test based on 4 randomization stratification factors: age (< 65 years versus ≥ 65 years), geographic region (China versus non-China), refractory status (yes or no), and 17p deletion and/or TP53 mutation status (present versus absent). The null and alternative hypotheses to test the noninferiority are as follows:

• for H_0NI, the HR (zanubrutinib to ibrutinib) was at least 1.3319

• for H_aNI, the HR (zanubrutinib to ibrutinib) was less than 1.3319.

There was a single analysis of PFS planned for the purpose of inference when approximately 205 PFS events had occurred; however, a 1-sided significance level of 0.00001 was applied to the analysis of PFS at the time when ORR was analyzed to compensate for the potential type I error increase from the descriptive analysis. The 205 PFS events are expected to accrue 45 months after the study start. If the P value from the stratified log-rank test for noninferiority is significant, the noninferiority of zanubrutinib to ibrutinib in terms of PFS is demonstrated. Further testing of superiority in terms of PFS was performed in this case. The noninferiority margin of 1.3319 was derived using the 95% to 95% fixed-margin method based on a meta-analysis of the RESONATE and RESONATE-2 studies. In the RESONATE-2 study, the estimated PFS HR for ibrutinib versus chlorambucil is 0.16, with a 95% CI of 0.09 to 0.28. In the updated RESONATE results, the estimated PFS HR for ibrutinib versus ofatumumab is 0.106, with a 95% CI of 0.073 to 0.153.38. In a fixed-effects meta-analysis, the pooled HR is estimated as 0.120, with a 95% CI of 0.088 to 0.163. Therefore, the control arm effect is –0.163 in HR and 1.814 in log HR. With the requirement of 84.2% of the control arm effect to be retained in zanubrutinib, a noninferiority margin of 1.3319 for the HR (zanubrutinib to ibrutinib) is generated. The HR for PFS and its 95% CI was estimated from a stratified Cox regression model. The distribution of PFS, including median and other quartiles, and the PFS rate at selected time points were estimated using the Kaplan-Meier method for each arm. For PFS, alternative censoring rules, such as censoring for new anticancer therapy, will be used in the sensitivity analyses.

ORR per IRC

ORR per IRC was analyzed using the methods employed for ORR by IA.

Overall Survival

OS was analyzed using the methods employed for PFS by IA.

Duration of Response

The distribution of DOR by ICR, including median and other quartiles, was estimated using the Kaplan-Meier method for each treatment group. There was no treatment arm comparison for DOR. The same analysis was performed for DOR by IA. The censoring rule used in the PFS analysis were used in the analysis of DOR.

Time to Treatment Failure

The HR for time to treatment failure and its 95% CI was estimated from a stratified Cox regression using the 4 randomization stratification factors (age [< 65 years versus ≥ 65 years], geographic region [China versus non-China], refractory status [yes or no], and 17p deletion and/or TP53 status [present versus absent]). The Kaplan-Meier method was used to estimate the distribution of time to treatment failure for each treatment group. Time to treatment failure was calculated as the date of randomization to the date of discontinuation of the study treatment for any cause. Time to treatment failure was censored at the data cut-off for the patients who did not discontinue the study treatment.

Health-Related Quality of Life

HRQoL measured by the EORTC QLQ-C30 questionnaire was summarized for each assessment time point for each treatment group. The EORTC QLQ-C30 global health status/quality of life score was compared between treatment arms using a linear mixed model for repeated measures at cycle 7 (24 weeks) and cycle 13 (48 weeks). Clinically meaningful changes from baseline in global health status/quality of life and functional domains were summarized as improved, stable, or worsened, and compared between arms A and B. The data were analyzed using a repeated-measures mixed model to account for missing data under the missing-at-random assumption. Changes in EQ-5D-5L scores were summarized descriptively.

Atrial Fibrillation and Flutter Incidence

If the noninferiority of zanubrutinib to ibrutinib in ORR is statistically significant, then the superiority of zanubrutinib to ibrutinib in the key secondary end point of atrial fibrillation and flutter was tested, but separately from the fixed-sequence hierarchical testing that includes ORR and PFS. The final analysis was performed on the safety analysis set, according to the actual treatment received. The monitoring boundaries for the superiority test are based on the O’Brien-Fleming boundary approximated by the Lan-DeMets spending function, with an overall 1-sided significance level of 0.025. If hypothesis testing for the superiority of the rate of atrial fibrillation and flutter is performed at the interim analysis, a 1-sided significance level of 0.005 (equivalent to a chi-square distribution P value cut-off of 0.0099) was allocated. If hypothesis testing for the superiority of the rate of atrial fibrillation and flutter is performed at the final analysis, a 1-sided significance level of 0.0235 (equivalent to a chi-square distribution P value cut-off of 0.0469) was allocated. Hypothesis testing on the rate of atrial fibrillation and flutter was performed using an unstratified chi-square distribution if the expected count in the 2 × 2 contingency table (treatment arm by atrial fibrillation and flutter status) is at least 5 patients. If any expected count in the 2 × 2 contingency table is less than 5 patients, then hypothesis testing was performed using Fisher’s exact test.

Table 9: Statistical Analysis of Efficacy End Points

CI = confidence interval; CLL = chronic lymphocytic leukemia; DOR = duration of response; EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; IA = investigator assessment; IRC = independent review committee; MAR = missing at random; MMRM = mixed model for repeated measures; ORR = overall response rate; OS = overall survival; PFS = progression-free survival; PR = partial response; PRL = partial response with lymphocytosis.

Note: Details from the table have been taken from the sponsor’s Summary of Clinical Evidence.⁴

Sources: SEQUOIA Clinical Study Report,³⁷ ALPINE final ORR and final PFS Clinical Study Reports.^19,20

Analysis Populations

The analysis sets for the SEQUOIA and ALPINE trials are summarized in Table 10.

SEQUOIA Trial

The ITT analysis set was used for efficacy analyses in cohort 1 (zanubrutinib versus BR in patients without 17p deletion), and a descriptive summary of zanubrutinib efficacy was performed using the ITT analysis set in cohort 2 (zanubrutinib in patients with 17p deletion). A sensitivity analysis was performed using the per-protocol analysis set in cohort 1. Safety analyses were performed using all patients who received any dose of a study drug, grouped according to whether they received zanubrutinib (cohort 1 or cohort 2) or BR (only cohort 1).

ALPINE Trial

Interim efficacy analyses (data cut-off date: December 31, 2020) were performed according to the ITT principle using the first 415 patients who were randomized (interim efficacy set). The interim efficacy analysis was performed approximately 12 months after the randomization of 415 patients. Interim safety analyses were performed for patients who were exposed to at least 1 dose of a study drug among the first 415 patients (interim safety set). Final efficacy analyses (data cut-off dates: December 1, 2021, and August 8, 2022) were performed according to the ITT principle using all 652 randomized patients (ITT analysis set). Final safety analyses were performed for all patients who were exposed to at least 1 dose of a study drug among the randomized patients.

Table 10: Analysis Populations From the SEQUOIA and ALPINE Trials

DOR = duration of response; IA = investigator assessment; IRC = independent review committee; IRT = Interactive Response Technology; ITT = intention to treat; ORR = overall response rate.

Note: Details from the table have been taken from the sponsor’s Summary of Clinical Evidence.⁴

Sources: SEQUOIA Clinical Study Report,³⁷ ALPINE final ORR and final PFS Clinical Study Reports.^19,20

Results

Patient Disposition

Patient disposition for the SEQUOIA and ALPINE trials are summarized in Table 11.

SEQUOIA Trial

At the data cut-off date of May 7, 2021, 479 patients were randomized in cohort 1 to the zanubrutinib arm (N = 241) or the BR arm (N = 238). Twelve patients were randomized in cohort 1 but did not receive any study treatment (1 patient in the zanubrutinib arm and 11 patients in the BR arm). In the zanubrutinib arm, 34 patients (14.1%) discontinued study treatment, which was most commonly related to AEs (20 patients, or 8.3%). In the BR arm (fixed-duration therapy), all 227 treated patients had discontinued therapy at the data cut-off date, with 188 patients (79.0%) discontinuing because they completed the prescribed therapy and 31 patients (13.0%) discontinuing because of AEs. Most patients in the BR arm completed treatment before disease progression and 15 patients (6.3%) in the BR arm initiated next-line crossover therapy with zanubrutinib. The median follow-up times were 26.35 months in the zanubrutinib arm and 25.92 months in the BR arm.

In cohort 2, all randomized patients received zanubrutinib treatment, although 18 patients (16.2%) discontinued treatment, most commonly because of disease progression (10 patients, or 9.0%). Death was the most common reason for study discontinuation (8 patients, or 7.2%).

Impact of COVID-19 on Patient Disposition

COVID-19 affected patient disposition in the study, although it did not significantly affect the primary end point. In cohort 1, 5 patients (2.1%) in the zanubrutinib arm discontinued treatment because of COVID-19-related AEs. No dose modifications were observed in the BR arm because the COVID-19 pandemic began after patients in that arm had concluded treatment.³⁷ More patients discontinued the study because of fatal COVID-19-related AEs in the zanubrutinib arm than in the BR arm (5 patients [2.1%] versus 1 patient [0.4%]). No patients in cohort 2 discontinued treatment or the study because of COVID-19-related AEs.

ALPINE Trial

In the final ORR analysis ITT analysis set (data cut-off: December 1, 2021) and the final PFS analysis ITT analysis set (data cut-off: August 8, 2022), 652 patients had been randomized to receive zanubrutinib (n = 327) or ibrutinib (n = 325) and 324 patients in each arm had received the treatment as intended. Of the 4 patients who did not receive treatment despite randomization (3 patients in the zanubrutinib arm and 1 patient in the ibrutinib arm), 1 patient had an AE (chickenpox), 2 patients withdrew consent in the zanubrutinib arm, and 1 patient was withdrawn by the investigator (thrombocytopenia) in the ibrutinib arm. Fewer patients discontinued treatment in the zanubrutinib arm than in the ibrutinib arm (19.3% versus 33.2% in the final ORR analysis ITT analysis set; 26.3% versus 41.2% in the final PFS analysis ITT analysis set), which was primarily due to AEs (13.8% versus 18.2% in the final ORR analysis ITT analysis set; 16.2% versus 22.8% in the final PFS analysis ITT analysis set) and disease progression (4.0% versus 9.8% in the final ORR analysis ITT analysis set; 7.3% versus 12.9% in the final PFS analysis ITT analysis set)

Fewer patients discontinued the study in the zanubrutinib arm than in the ibrutinib arm (14.4% versus 20.9% n the final ORR analysis ITT analysis set; 20.5% versus 26.2% in the final PFS analysis ITT analysis set), which was primarily due to death (10.1% versus 12.3% in the final ORR analysis ITT analysis set; 14.7% versus 18.5% in the final PFS analysis ITT analysis set). The median follow-up times were 24.3 months (range, 0.1 to 34.1 months) in the zanubrutinib arm and 23.8 months (range, 0.1 to 37.0 months) in the ibrutinib arm in the final ORR analysis ITT analysis set, and 32.0 months (range, 0.1 to 41.8 months) in the zanubrutinib arm and 27.9 months (range, 0.1 to 45.2 months) in the ibrutinib arm in the final PFS analysis ITT analysis set.

Impact of COVID-19 on Patient Disposition

Fewer patients discontinued treatment due to COVID-19-related AEs in the zanubrutinib arm than in the ibrutinib arm (2.4% versus 3.4% in the final ORR analysis ITT analysis set; 4.3% versus 5.2% the final PFS analysis ITT analysis set). Similarly, fewer patients discontinued the study due to COVID-19-related death in the zanubrutinib arm than in the ibrutinib arm (2.4% versus 3.4% in the final ORR analysis ITT analysis set; 4.9% versus 6.2% in the final PFS analysis ITT analysis set).

Generally similar findings were observed in the interim efficacy set (data cut-off: December 31, 2020), albeit with lower discontinuation rates that were likely related to the shorter follow-up period in the interim analysis.

Table 11: Summary of Patient Disposition From the SEQUOIA and ALPINE Trials Submitted by the Sponsor (ITT Analysis Set)

BR = bendamustine plus rituximab; ITT = intention to treat; NA = not applicable; NR = not reported; ORR = overall response rate; PFS = progression-free survival; SD = standard deviation.

Note: Details from the table have been taken from the sponsor’s Summary of Clinical Evidence.4

^aOne patient in cohort 2 was not included in the ITT analysis set or the per-protocol analysis set because they did not have a 17p deletion and thus were incorrectly enrolled in cohort 2 (patients with 17p deletion). However, that patient was included in the safety analysis set because they received zanubrutinib.

^bThe per-protocol sets excluded 4 patients who did not receive any dose of a study drug and 1 patient who had a critical protocol deviation.

Sources: SEQUOIA Clinical Study Report,³⁷ ALPINE final ORR and final PFS Clinical Study Reports.^19,20

Baseline Characteristics

A summary of baseline patient demographics, disease characteristics, and treatment history in the SEQUOIA and ALPINE trials is shown in Table 12. The baseline characteristics outlined in the table are limited to those that are most relevant to this review or were felt to impact the outcomes or interpretation of the study results.

SEQUOIA Trial

The demographic and baseline characteristics were generally balanced between the zanubrutinib and BR arms in cohort 1, although zanubrutinib-treated patients were slightly more likely to be white (91.7% versus 86.6%). Most patients in cohort 1 (zanubrutinib versus BR) were enrolled at sites in Europe (72.2% versus 72.3%) and had an ECOG PS of 0 or 1 (93.8% versus 91.6%). The demographic and baseline characteristics were generally similar between the zanubrutinib arms in cohort 1 (without 17p deletion) and cohort 2 (with 17p deletion), except fewer patients from the Asia-Pacific region were enrolled in cohort 1 than cohort 2 (13.7% versus 42.3%).

Medical histories were generally comparable between the zanubrutinib and BR arms in cohort 1 and the zanubrutinib arm in cohort 2. Prior and/or concurrent medical conditions were reported by the data cut-off date at similar rates in patients who received BR (98.7%) or zanubrutinib (98.8% for cohort 1; 99.1% for cohort 2). The most commonly reported condition was hypertension (55.6% versus 54.6% for zanubrutinib versus BR in cohort 1; 50.5% for zanubrutinib cohort 2).

Disease histories were also generally comparable between the zanubrutinib and BR arms in cohort 1 and the zanubrutinib arm in cohort 2, with the exception of 17p deletion status (cohort 1 consisted of patients without 17p deletion; cohort 2 consisted of patients with 17p deletion). Enrolment errors led to 2 patients with 17p deletion being included in cohort 1 and 1 patient without 17p deletion being included in cohort 2; these patients were treated as protocol deviations.

All patients in all arms had signs and/or symptoms of CLL. Most patients had CLL as their cancer type (91.7% versus 91.6% for zanubrutinib versus BR in cohort 1, 90.1% for zanubrutinib in cohort 2), which at study entry was most commonly Binet stage B (57.0% versus 56.9% for zanubrutinib versus BR in cohort 1; 49.0% in cohort 2) or Binet stage C (29.4% versus 30.3% for zanubrutinib versus BR in cohort 1; 37.0% in cohort 2). The median time from diagnosis to randomization was 31.28 months (range, 0.7 to 231.9 months) in the zanubrutinib cohort 1 arm, 28.67 months (range, 0.9 to 231.4 months) in the BR arm, and 21.39 months (range, 1.1 to 323.8 months) in the zanubrutinib cohort 2 arm.

ALPINE Trial

Demographic and baseline patient characteristics were similar in the zanubrutinib and ibrutinib arms in the ITT analysis set (final PFS analysis). The median age was 67.0 years (range, 35 years to 90 years) in the zanubrutinib arm and 68.0 years (range, 35 to 89 years) in the ibrutinib arm. Most patients (zanubrutinib versus ibrutinib) were enrolled at sites in Europe (60.6% versus 58.8%), were white (79.8% versus 81.5%), and had an ECOG PS of 0 or 1 (97.9% versus 96%). Demographic and baseline patient characteristics in the final ORR analysis ITT analysis set were similar those in the ALPINE final PFS analysis ITT analysis set.

In the ITT analysis set (final PFS analysis), prior and/or concurrent medical conditions were reported by most patients and were generally similar in the 2 arms (95.7% versus 96.6% for zanubrutinib versus ibrutinib). The most frequent medical condition was hypertension (48.9% versus 48.3% for zanubrutinib versus ibrutinib). Other common medical histories were reported at similar frequencies in the 2 arms (zanubrutinib versus ibrutinib), such as metabolism and nutrition disorders (48.9% versus 45.5%), gastrointestinal disorders (37.0% versus 39.7%), musculoskeletal and connective tissue disorders (36.1% versus 39.4%), infections and infestations (36.1% versus 36.3%), and respiratory, thoracic, and mediastinal disorders (27.5% versus 32.3%). Prior and/or concurrent medical conditions in the final ORR analysis ITT analysis set were similar to those in the final PFS analysis ITT analysis set.

Disease histories were also similar in the 2 arms and in the final PFS analysis ITT analysis set and the final ORR analysis ITT analysis set. In the final PFS analysis ITT analysis set, the median time from initial diagnosis to randomization was 83.5 months (range, 1 to 346 months) in the zanubrutinib arm and 82.0 months (range, 1 to 326 months) in the ibrutinib arm. The majority of patients (zanubrutinib versus ibrutinib) had CLL (96.0% versus 95.1%) at stage B (45.3% versus 47.4%) or stage C (40.7% versus 36.9%), whereas only a few patients had SLL (4.0% versus 4.9%). Genetic mutations were similar in the zanubrutinib and ibrutinib treatment arms, including 17p deletion (13.8% versus 15.4%), 11q deletion (27.8% versus 27.1%), TP53 mutations (15.3% versus 13.8%), and unmutated IGHV (73.1% versus 73.5%). Disease histories in the final ORR analysis ITT analysis set were similar to those in the final PFS analysis ITT analysis set.

In the ITT analysis set (final PFS analysis), all patients had received at least 1 prior line of systemic therapy. The most commonly used prior therapy (zanubrutinib versus ibrutinib) was anti-CD20 antibodies (83.8% versus 82.8%), followed by alkylators other than bendamustine (83.8% versus 79.7%) and chemoimmunotherapy (79.5% versus 76.0%). In the final ORR analysis ITT analysis set, patients had numbers and types of prior systemic therapies similar to those in the final PFS analysis ITT analysis set.

Table 12: Summary of Baseline Characteristics for the SEQUOIA and ALPINE Trials (ITT Analysis Set)

BR = bendamustine plus rituximab; CLL = chronic lymphocytic leukemia; ECOG PS = Eastern Cooperative Oncology Group Performance Status; iMiD = immunomodulatory imide drug; ITT = intention to treat; LDH = lactate dehydrogenase; NA = not applicable; NR = not reported; ORR = overall response rate; PFS = progression-free survival; SD = standard deviation; SLL = small lymphocytic lymphoma; ULN = upper limit of normal.

Note: Details from the table have been taken from the sponsor’s Summary of Clinical Evidence.⁴

^aPercentages are based on number of patients with CLL.

^bCytopenia: anemia (hemoglobin ≤ 110 g/L) or thrombocytopenia (platelet count ≤ 100 10⁹/L) or neutropenia (absolute neutrophil count ≤ 1.5 10⁹/L).

^cInadvertent inclusion of these patients in the zanubrutinib arm of cohort 1.

^dOne patient without 17p deletion was included in this cohort due to site error. This patient was not included in the efficacy analysis.

^eBased on monosomy 13q mutation results.

^fComplex karyotype is defined as 3 or more abnormalities.

^gSamples not yet evaluated.

Sources: SEQUOIA Clinical Study Report,³⁷ ALPINE final ORR and final PFS Clinical Study Reports.^19,20

Prior Medication

Prior medications in the SEQUOIA and ALPINE trials are summarized in Table 13.

SEQUOIA Trial

The most common prior medication classes in cohort 1 (zanubrutinib versus BR) were drugs that act on the renin-angiotensin system (44.2% versus 37.9%), antithrombotic drugs (30.0% versus 32.6%), beta-blockers (26.3% versus 29.5%), and drugs for acid-related disorders (22.5% versus 21.1%). The most common prior medication classes in cohort 2 were drugs that act on the renin-angiotensin system (40.5%), lipid-modifying drugs (30.6%), analgesics (29.7%), antithrombotic drugs (28.8%), vitamins (19.8%), drugs for acid-related disorders (18.0%), and beta-blockers (13.5%).

ALPINE Trial

In the safety analysis set (final PFS analysis; data cut-off date: August 8, 2022), prior medication classes were comparably prevalent in patients in the 2 treatment arms. The most common prior medication classes (zanubrutinib versus ibrutinib) were drugs that act on the renin-angiotensin system (32.1% versus 29.0%), antigout preparations (25.9% versus ibrutinib 25.6%), lipid-modifying drugs (28.1% versus 22.8%), beta-blockers (20.7% versus 23.5%), and antithrombotic drugs (including anticoagulants and antiplatelet drugs) (21.0% versus 21.0%).

In the final ORR analysis safety analysis set (data cut-off date: December 1, 2021), patients had numbers and types of prior medication similar to those in the final PFS analysis safety analysis set.

Table 13: Summary of Prior Medication in the SEQUOIA and ALPINE Trials (Safety Analysis Set)

BR = bendamustine plus rituximab; ORR = overall response rate; PFS = progression-free survival.

Sources: SEQUOIA Clinical Study Report,³⁷ ALPINE final ORR and final PFS Clinical Study Reports.^19,20

Postbaseline Anticancer Systemic Therapies

Postbaseline anticancer systemic therapies in the ALPINE trial are summarized in Table 14.

SEQUOIA Trial

No results for postbaseline anticancer systemic therapies were reported for the SEQUOIA trial.

ALPINE Trial

In the final PFS analysis ITT analysis set (data cut-off date: August 8, 2023), a lower proportion of patients in the zanubrutinib arm than in the ibrutinib arm received subsequent anticancer treatment for CLL (7.3% versus 13.8%). The most common postbaseline anticancer systemic therapies (zanubrutinib versus ibrutinib) were rituximab (3.1% versus 4.3%), venetoclax (2.4% versus 6.8%), cyclophosphamide (1.5% versus 1.8%), doxorubicin (1.5% versus 0.9%), vincristine (1.5% versus 0.9%), ibrutinib (0.9% versus 2.2%), and acalabrutinib (0.6% versus 1.8%). The median time from the last dose of the study treatment to the initiation of anticancer systemic therapy was 0.72 months (range, 0.1 to 14.2 months) for the 24 zanubrutinib-treated patients who received postbaseline anticancer therapy and 0.95 months (range, 0.0 to 19.9 months) for the 45 ibrutinib-treated patients who received postbaseline anticancer therapy.

In the final ORR analysis ITT analysis set (data cut-off date: December 1, 2021), patients had numbers and types of postbaseline anticancer systemic therapies similar to those in the final PFS analysis ITT analysis set. No median time from the last dose of the study treatment to the initiation of anticancer systemic therapy was reported for the final ORR analysis.

Table 14: Summary of Postbaseline Anticancer Systemic Therapies in the ALPINE Trial (ITT Analysis Set)

ITT = intention to treat; ORR = overall response rate; PFS = progression-free survival.

Sources: ALPINE final ORR and final PFS Clinical Study Reports.^19,20

Exposure to Study Treatments

SEQUOIA Trial

Patient exposure data for the SEQUOIA trial are summarized in Table 15. Based on the fixed duration of BR therapy, the median duration of exposure was substantially longer for zanubrutinib (26.1 months for cohort 1, 30.0 months for cohort 2) than for bendamustine (5.5 months) of for rituximab (5.6 months). However, median relative dose intensities were similar for zanubrutinib (98.0% for cohort 1, 97.9% for cohort 2), bendamustine (96.5%), and rituximab (98.7%). The most common reason for dose reduction was AEs in the zanubrutinib arms (8.3% for cohort 1, 5.4% for cohort 2) and in the bendamustine arm (37.4%). Most patients in the zanubrutinib arms had missed doses (64.6% for cohort 1, 74.8% for cohort 2), which were commonly related to AEs (38.3% for cohort 1, 37.8% for cohort 2) or procedures (26.7% for cohort 1, 45.9% for cohort 2).

ALPINE Trial

Patient exposure data for the ALPINE trial are summarized in Table 16. In the final PFS safety analysis set (cut-off date: December 1, 2021), the median overall treatment duration was longer in the zanubrutinib arm than in the ibrutinib arm (28.4 months versus 24.3 months). Fewer patients had dose reductions or interruptions in the zanubrutinib arm than in the ibrutinib arm (16.0% versus 18.5% and 55.6% versus 63.0%, respectively), and fewer patients in the zanubrutinib arm than in the ibrutinib arm had dose reductions or interruptions related to AEs (11.1% versus 16.4% and 46.0% versus 52.8%, respectively). The median duration of dose interruption was slightly shorter in the zanubrutinib arm than in the ibrutinib arm (22.0 days versus 21.0 days). Generally consistent findings were observed in the final ORR analysis safety set (data cut-off: December 1, 2021).

Table 15: Summary of Patient Exposure in the SEQUOIA Trial (Safety Analysis Set)

NA = not applicable; NR = not reported; SD = standard deviation.

Notes: Details from the table have been taken from the sponsor’s Summary of Clinical Evidence.⁴

Data cut-off was May 7, 2021.

Source: SEQUOIA Clinical Study Report.³⁷

Table 16: Summary of Patient Exposure in the ALPINE Trial (Safety Analysis Set)

ORR = overall response rate; PFS = progression-free survival; SD = standard deviation.

^aTreatment duration (months) was calculated as (last dose date – first dose date + 1)/30.4375, where the data cut-off date is used as last dose date for ongoing patients.

^bRelative dose intensity is defined as the ratio of the actual dose intensity (mg/day) to the planned dose intensity (mg/day).

^cA patient may be counted in more than 1 row. Multiples of the same reason were counted once per patient in each row.

^dDuration was calculated for patients with dose interruption only.

Note: Details from the table have been taken from the sponsor’s Summary of Clinical Evidence.⁴

Sources: ALPINE final ORR and final PFS Clinical Study Reports.^19,20

Concomitant Medications

Concomitant medications in the SEQUOIA and ALPINE trials are summarized in Table 17.

SEQUOIA Trial

Almost all patients in cohort 1 received at least 1 concomitant medication (99.2% for the zanubrutinib arm and 99.1% for the BR arm). The most common concomitant medications in the zanubrutinib and BR arms were antibacterials for systemic use (69.6% versus 75.3%). The following medications were reported less frequently in the zanubrutinib arm than in the BR arm: analgesics (41.3% versus 84.6%), antigout preparations (47.5% versus 63.9%), corticosteroids for systemic use (20.4% versus 82.4%), antihistamines for systemic use (18.8% versus 82.8%), antivirals for systemic use (35.4% versus 63.4%), and immunostimulants (10.8% versus 58.1%). The use of vaccines was more common in the zanubrutinib arm than in the BR arm (31.3% versus 3.5%).

The most common concomitant medications used in the zanubrutinib arm in cohort 2 were antibacterials for systemic use (74.8%), analgesics (54.1%), drugs that act on the renin-angiotensin system (46.8%), antigout preparations (40.5%), antithrombotic drugs (37.8%), drugs for acid-related disorders (34.2%), antivirals for systemic use (30.6%), vaccines (30.6%), and lipid-modifying drugs (29.7%).

ALPINE Trial

Almost all patients in the final PFS analysis safety analysis set (data cut-off date: August 8, 2022) (zanubrutinib versus ibrutinib) received at least1 concomitant medication (98.8% versus 98.5%), which were most commonly antibacterials for systemic use (73.1% versus 75.6%), antigout preparations (50.3% versus 59.0%), and antivirals for systemic use (52.8% versus 55.9%).

In the final ORR analysis safety analysis set (data cut-off date: December 1, 2021), patients had numbers and types of concomitant medication and anticancer treatment similar to those in the final PFS analysis safety analysis set.

Table 17: Summary of Concomitant Medication in the SEQUOIA and ALPINE Trials (Safety Analysis Set)

BR = bendamustine plus rituximab; ORR = overall response rate; PFS = progression-free survival.

Sources: SEQUOIA Clinical Study Report,³⁷ ALPINE final ORR and final PFS Clinical Study Reports.^19,20

Protocol Deviation

SEQUOIA Trial

In cohort 1, a numerically higher proportion of patients reported at least 1 type of important protocol deviation in the zanubrutinib arm than in the BR arm (4.1% versus 0.8%). The most common protocol deviation in the zanubrutinib and BR treatment arms was related to the assessment of safety (2.9% versus 0.4%). Critical protocol deviations were reported in 3 (1.2%) patients in the zanubrutinib arm and 1 (0.4) patient in the BR arm. The investigators stated that the protocol deviations were not expected to affect interpretation of the results, considering the limited number of reported cases.

In cohort 2, important protocol deviations were reported in 2 (1.8%) patients in the zanubrutinib arm. One patient (0.9%) in the zanubrutinib arm reported a critical protocol deviation.

ALPINE Trial

At the data cut-off date of December 1, 2021, more patients in the zanubrutinib arm than in the ibrutinib arm reported at least 1 type of important protocol deviation (1.5% versus 0.6%). One patient had a critical protocol deviation (prohibitive medication or treatment) in the zanubrutinib arm. The investigators stated that the protocol deviations were not expected to affect interpretation of the results, considering the limited number of reported cases.

At the data cut-off date of August 8, 2022, a higher proportion of patients in the zanubrutinib arm than in the ibrutinib arm reported at least 1 type of important protocol deviation (2.4% versus 1.2%). One patient had a critical protocol deviation (prohibitive medication or treatment) in the zanubrutinib arm, and 1 patient had a critical protocol deviation (dosing and administration) in the ibrutinib arm. The investigators stated that the protocol deviations were not expected to affect interpretation of the results, considering the limited number of reported cases.

Measurements of Treatment Compliance

SEQUOIA Trial

Individual data for study drug administration were monitored and recoded. The details of treatment compliance measurements were not reported.

ALPINE Trial

Compliance with study drug administration was assessed using patient diaries, tablet counts, and verbal patient reports at each study visit in the ALPINE trial. The details of treatment compliance measurements were not reported.

Efficacy

Unless otherwise specified, the key efficacy results of the SEQUOIA and ALPINE trials are summarized in Table 18 and Table 19, respectively.

Progression-Free Survival

PFS per IRC in the SEQUOIA Trial

The analysis of the primary outcome of PFS per IRC in SEQUOIA cohorts 1 and 2 is summarized in Table 18, Figure 6, and Figure 7, respectively.

As of May 7, 2021, in cohort 1, PFS events per IRC had occurred in 36 (14.9%) patients in the zanubrutinib arm and 71 (29.8%) patients in the BR arm. Median PFS per IRC had not yet been reached in the zanubrutinib arm, whereas median PFS per IRC was 33.7 months (95% CI, 28.1 to NE) in the BR arm. The HR for PFS per IRC comparing zanubrutinib with BR was 0.42 (95% CI, 0.28 to 0.63; P < 0.0001). Median follow-up time based on the reverse Kaplan-Meier method was 25.1 months (95% CI, 24.9 to 25.4 months) in the zanubrutinib arm and 24.6 months (95% CI, 22.8 to 25.2 months) in the BR arm. Higher event-free rates were observed in the zanubrutinib arm than in the BR arm (94.5% versus 90.2%) at 12 months, 24 months (85.5% versus 69.5%), and 36 months (81.5% versus 40.8%).

Subgroup analyses of PFS per IRC by age (< 65 versus ≥ 65 years), sex, and ECOG PS (0 versus ≥ 1) were generally consistent with the primary analysis across all strata. However, inconsistent findings were reported in the subgroup analyses of high-risk genetic factors (IGHV mutation status [unmutated versus mutated] and TP53 mutation status [unmutated versus mutated]), cancer type (CLL versus SLL), disease stage (Binet stage of A or B versus Binet stage C), and complex karyotype (< 3 abnormalities versus ≥ 3 abnormalities).

Refer to Appendix 1 for detailed subgroup analyses data. Several prespecified sensitivity analyses based on the IRC assessment of PFS were included in the statistical analysis plan, including unstratified analysis, using the per-protocol analysis set, and changes to definitions of PFS and censoring events. The results were generally consistent with results of the primary analysis (HR = 0.42; 95% CI, 0.28 to 0.63; P < 0.0001), and showed HR values ranging from 0.45 (95% CI, 0.31 to 0.67) to 0.34 (95% CI, 0.22 to 0.53). High concordance for PFS per IRC and PFS per IA was observed (concordance rate for disease progression = 91.4.%), and HRs for PFS per IRC (0.42; 95% CI, 0.28 to 0.63) and PFS per IA were also similar (0.42; 95% CI, 0.27 to 0.66).

In cohort 2, median PFS by IRC was not reached in the zanubrutinib arm, and the event-free rates were 93.6% at 12 months, 88.9% at 24 months, and 84.9% at 36 months. Subgroup analyses showed that PFS appeared to be preserved across most high-risk subgroups, including age, Binet stage, ECOG PS, bulky disease, IGHV mutational status, baseline cytopenia, and complex karyotype status. A higher rate of progression was observed in patients with concurrent TP53 mutation than in those without (21.3% versus 8.1%). Consistent results were observed for PFS per IA, with event-free rates of 94.5% at 12 months, 87.0% at 24 months, and 82.6% at 36 months.

Figure 6: Kaplan-Meier Plot of PFS by IRC in Cohort 1 of the SEQUOIA Trial (ITT Analysis Set)

BR = bendamustine plus rituximab; IRC = independent review committee; ITT = intention to treat; PFS = progression-free survival.

Note: Data cut-off was May 7, 2021.

Source: SEQUOIA Clinical Study Report.¹⁸

Figure 7: Kaplan-Meier Plot of PFS by IRC in Cohort 2 of the SEQUOIA Trial (Safety Analysis Set)

CI = confidence interval; IRC = independent review committee; PFS = progression-free survival.

Note: Data cut-off was May 7, 2021.

One patient was excluded from the efficacy analysis because they did not have a 17p deletion and were enrolled in cohort 2 in error.

Source: SEQUOIA Clinical Study Report.¹⁸

PFS per IA in the SEQUOIA Trial

The analysis of the secondary outcome of PFS per IA in SEQUOIA cohort 1 is summarized in Table 18, Figure 8, and Figure 9, respectively. Of note, the analysis of PFS per IA was not adjusted for multiplicity.

As of the May 7, 2021 data cut-off, in cohort 1, PFS events per IA had occurred in 29 (12.0%) patients in the zanubrutinib arm and 57 (23.9%) patients in the BR arm. The HR for PFS per IA comparing zanubrutinib with BR was 0.42 (95% CI, 0.27 to 0.66; nominal P < 0.0001). Median follow-up time was 22.8 months (95% CI, 22.6 to 23.8 months) in the zanubrutinib arm and 22.6 months (95% CI, 22.4 to 22.9 months) in the BR arm. Median PFS time was not reached in the zanubrutinib arm, whereas median PFS was 33.7 months (95% CI, 28.4 to 33.7 months) in the BR arm. Higher event-free rates were observed in the zanubrutinib arm than in the BR arm (95.8% versus 91.2%) at 12 months and 24 months (87.7% versus 76.5%).

In cohort 2, the median PFS by IA was not reached for the zanubrutinib arm, and the event-free rates were 94.5% at 12 months, 87.0% at 24 months, and 82.6% at 36 months.

Figure 8: Kaplan-Meier Plot of PFS per IA in Cohort 1 of the SEQUOIA Trial (ITT Analysis Set)

BR = bendamustine plus rituximab; IA = investigator assessment; ITT = intention to treat; PFS = progression-free survival.

Note: Data cut-off was May 7, 2021.

Source: SEQUOIA Clinical Study Report.¹⁸

Figure 9: Kaplan-Meier Plot of PFS per IA in Cohort 2 of the SEQUOIA Trial (Safety Analysis Set)

CI = confidence interval; IA = investigator assessment; PFS = progression-free survival.

Note: Data cut-off was May 7, 2021.

Source: SEQUOIA Clinical Study Report.¹⁸

PFS per IRC in the ALPINE Trial

The analysis of PFS per IRC in the ALPINE trial is a secondary outcome (summarized in Table 19, Figure 10, and Figure 11). Of note, this analysis was not part of the statistical hierarchy and was not adjusted for multiplicity.

At the final ORR analysis cut-off date of December 1, 2021, IRC-assessed PFS events had occurred in 60 patients (18.3%) in the zanubrutinib arm and 87 patients (26.8%) in the ibrutinib arm. Patients in the zanubrutinib arm had a lower risk of PFS per IRC (HR = 0.61; 95% CI, 0.44 to 0.86; nominal P = 0.0038). Median follow-up time was 22.1 months in both the zanubrutinib and ibrutinib arms. Median PFS was not reached in either arm. Clear separation of the Kaplan-Meier curves for PFS was also observed based on data from the final ORR analysis cut-off (December 1, 2021).

At the final PFS analysis cut-off date of August 8, 2022, IRC-assessed PFS events had occurred in 88 patients (26.9%) in the zanubrutinib arm and 120 patients (36.9%) in the ibrutinib arm. Patients in the zanubrutinib arm had a lower risk of PFS per IRC than those in the ibrutinib arm (HR = 0.65; 95% CI, 0.49 to 0.86, nominal P = 0.0024). Median follow-up time was 32.9 months (95% CI, 27.8 to 33.1 months) in the zanubrutinib arm and 28.1 months (95% CI, 27.6 to 33.0 months) in the ibrutinib arm. Median PFS was not reach in the zanubrutinib arm with the lower bound of the 95% CI of 34.3 months; median PFS was 35.0 months (95% CI, 33.2 to 44.3 months) in the ibrutinib arm. Clear separation of the Kaplan-Meier curves for PFS was also observed based on data from the final PFS analysis cut-off (August 8, 2022).

Generally similar findings were observed in the interim efficacy set (data cut-off: December 31, 2020), Refer to Table 45 in Appendix 1 for detailed outcome data.

Figure 10: Kaplan-Meier Plot of PFS by IRC in the ALPINE Final ORR Analysis (ITT Analysis Set)

IRC = independent review committee; ITT = intention to treat; No. = number; ORR = overall response rate; PFS = progression-free survival.

Note: Data cut-off was December 1, 2021.

Source: ALPINE final ORR analysis Clinical Study Report.²⁰

Figure 11: Kaplan-Meier Plot of PFS by IRC in the ALPINE Final PFS Analysis (ITT Analysis Set)

IRC = independent review committee; ITT = intention to treat; PFS = progression-free survival.

Note: Data cut-off was August 8, 2022.

Source: ALPINE final PFS analysis Clinical Study Report.¹⁹

PFS per IA in the ALPINE Trial

The analysis of PFS per IA is the key secondary outcome in the ALPINE trial and was adjusted for multiplicity in the final PFS analysis (summarized in Table 19).

At the final ORR analysis cut-off date of December 1, 2021, PFS events per IA had occurred in 58 patients (17.7%) in the zanubrutinib arm and 91 patients (28.0%) in the ibrutinib arm. Patients in the zanubrutinib arm had a lower risk of PFS per IA (HR = 0.55; 95% CI, 0.39 to 0.76; nominal P = 0.0004), although a slightly lesser difference was observed for PFS per IRC, as previously mentioned. Median follow-up time was 22.1 months in both the zanubrutinib and ibrutinib arms. Median PFS was not reached in either arm; the lower bound of the 95% CI was 29.6 months in the zanubrutinib arm. Clear separation of the Kaplan-Meier curves for PFS was also observed based on the data from the final ORR analysis cut-off (December 1, 2021) (Figure 12).

At the final PFS analysis cut-off date of August 8, 2022, the investigators assessed that PFS events had occurred in 87 patients (26.6%) in the zanubrutinib arm and 118 patients (36.3%) in the ibrutinib arm. Patients in the zanubrutinib arm had a lower risk of PFS per IA (HR = 0.65; 95% CI, 0.49 to 0.86), which was both noninferior (P < 0. 0001 versus prespecified 1-sided significance level of 0.02498) and superior (P = 0.0024 versus prespecified 1-sided significance level of 0.02498). Median follow-up time was 31.4 months (95% CI, 27.7 to 33.1 months) in the zanubrutinib arm and 27.8 months (95% CI, 27.6 to 33.1 months) in the ibrutinib arm. Median PFS was not reached in the zanubrutinib arm with the lower bound of the 95% CI of 34.3 months; median PFS in the ibrutinib arm was 34.2 months (95% CI, 33.3 to NE). Clear separation of the Kaplan-Meier curves for PFS was also observed based on data from the final PFS analysis cut-off (August 8, 2022) (Figure 13).

Generally similar findings were observed in the interim efficacy set (data cut-off: December 31, 2020), albeit with lower PFS event rates that were likely related to the shorter follow-up period in the interim analysis. Refer to Table 45 in Appendix 1 for detailed outcome data.

Figure 12: Kaplan-Meier Plot of PFS by IA in the ALPINE Final ORR Analysis (ITT Analysis Set)

IA = investigator assessment; ITT = intention to treat; No. = number; ORR = overall response rate; PFS = progression-free survival.

Note: Data cut-off was December 1, 2021.

Source: ALPINE final ORR analysis Clinical Study Report.²⁰

Figure 13: Kaplan-Meier Plot of PFS by IA in the ALPINE Final PFS Analysis (ITT Analysis Set)

IA = investigator assessment; ITT = intention to treat; No. = number; PFS = progression-free survival.

Note: Data cut-off was August 8, 2022.

Source: ALPINE final PFS analysis Clinical Study Report.¹⁹

Overall Survival

SEQUOIA Trial

The analysis of the hierarchically tested key secondary outcome of OS in SEQUOIA cohort 1 and SEQUOIA cohort 2 is summarized in Table 18.

As of the May 7, 2021, data cut-off, death events had occurred in 16 (6.6%) patients in the zanubrutinib arm and 14 (5.9%) patients in the BR arm. The HR for OS comparing zanubrutinib with BR was 1.07 (95% CI, 0.51 to 2.22; P = 0.5672). Median OS was not reached in the zanubrutinib arm with a median follow-up time of 26.5 months, whereas in the BR arm, median OS was 37.8 months (95% CI, 37.8 to NE) with a median follow-up time of 25.1 months. Event-free rates (zanubrutinib arm versus BR arm) were 98.3% versus 96.4% at 12 months, 94.3% versus 94.6% at 24 months, and 92.3% versus 93.6% at 36 months. Most patients were alive and censored at the data cut-off date (Figure 14).

In cohort 2, at the data cut-off date of May 7, 2021, there were 8 deaths (7.3%) reported in the zanubrutinib arm. Median OS was not reached in the zanubrutinib arm with a median follow-up time of 30.4 months. The event-free rates were 96.4% at 12 months, 93.6% at 24 months, and 90.7% at 36 months.³⁷ Most patients were alive and censored at the data cut-off date.

Figure 14: Kaplan-Meier Plot of OS in Cohort 1 of the SEQUOIA Trial (ITT Analysis Set)

BR = bendamustine plus rituximab; ; ITT = intention to treat; No. = number; OS = overall survival.

Note: Data cut-off was May 7, 2021.

Source: SEQUOIA Clinical Study Report.¹⁸

ALPINE Trial

The analysis of OS in the ALPINE trial is a secondary outcome (summarized in Table 19, Figure 15, and Figure 16). Of note, this analysis was not part of the statistical hierarchy and was not adjusted for multiplicity.

At the final ORR analysis cut-off in the ITT analysis set (December 1, 2021), there were 33 deaths (10.1%) reported in the zanubrutinib arm and 40 deaths (12.3%) reported in the ibrutinib arm (HR = 0.80; 95% CI, 0.50 to 1.28; nominal P = 0.3561). Median OS was not reached in either arm at median follow-up times of 24.9 months in the zanubrutinib arm and 24.6 months in the ibrutinib arm. Most patients were alive and on study at the data cut-off date (85.6% versus 79.1% for zanubrutinib versus ibrutinib).

At the final PFS analysis cut-off in the ITT analysis set (August 8, 2022), there were 48 deaths (14.7%) reported in the zanubrutinib arm and 60 deaths (18.5%) reported in the ibrutinib arm (HR = 0.76; 95% CI, 0.51 to 1.11; nominal P = 0.1533). Median OS was not reached in either arm at median follow-up times of 32.9 months in the zanubrutinib arm and 32.7 months in the ibrutinib arm. Most patients were alive and on study at the data cut-off date (79.5% versus 73.8% for zanubrutinib versus ibrutinib).

Generally similar findings were observed in the interim efficacy set (data cut-off: December 31, 2020), albeit with lower OS event rates that were likely related to the shorter follow-up period in the interim analysis. Refer to Table 45 in Appendix 1 for detailed outcome data.

Figure 15: Kaplan-Meier Plot of OS in the ALPINE Final ORR Analysis (ITT Analysis Set)

ITT = intention to treat; No. = number; ORR = overall response rate; OS = overall survival.

Note: Data cut-off was December 1, 2021.

Source: ALPINE final ORR analysis Clinical Study Report.²⁰

Figure 16: Kaplan-Meier Plot of OS in the ALPINE Final PFS Analysis (ITT Analysis Set)

ITT = intention to treat; OS = overall survival; PFS = progression-free survival.

Note: Data cut-off was August 8, 2022.

Source: ALPINE final PFS analysis Clinical Study Report.¹⁹

Overall Response Rate

ORR per IRC and per IA in the SEQUOIA Trial

The analysis of ORR per IRC and per IA was the secondary outcome in SEQUOIA cohorts 1 and 2. Note that this analysis was not part of the statistical hierarchy and was not adjusted for multiplicity. The ORR per IRC was 94.6% (95% CI, 91.0% to 97.1%) in the zanubrutinib arm and 85.3% (95% CI, 80.1% to 89.5%) in the BR arm, with an odds ratio (OR) favouring zanubrutinib (OR, 3.162; 95% CI, 1.608 to 6.220; P = 0.0006). Based on the IRC assessment, the majority of patients (zanubrutinib versus BR) had achieved a PR (85.5% versus 64.3%), followed by a CR (6.6% versus 15.1%), a nodular PR (1.2% versus 5.9%), and a PRL (1.2% versus 0). The ORR per IA was 97.5% (95% CI, 94.7% to 99.1%) in the zanubrutinib arm and 88.7% (95% CI, 83.9% to 92.4%) in the BR arm, with an OR favouring zanubrutinib (OR, 5.22; 95% CI, 2.08 to 13.08; P = 0.0001). In the analysis of investigator-assessed best overall response, the majority of patients (zanubrutinib versus BR) had achieved a PR (84.6% versus 62.6%), followed by a CR (9.1% versus 18.1%), a nodular PR (2.1% versus 7.6%), a PRL (1.7% versus 0.0%), and a CRi (0.0% versus 0.4%). In cohort 2, the ORR per IRC was 90.0% (95% CI, 82.8% to 94.9%), whereas ORR per IA was slightly higher, at 96.4% (95% CI, 91.0% to 99.0%).

ORR per IRC in the ALPINE Trial

The analysis of ORR per IRC was the secondary outcome in the ALPINE study. Note that this analysis was not part of the statistical hierarchy and was not adjusted for multiplicity.

In the ITT analysis set at the final ORR analysis cut-off of December 1, 2021, when ORR was assessed by the IRC, a higher response rate was observed in the zanubrutinib arm than in the ibrutinib arm (80.4% versus 72.9%), with a response ratio of 1.14 (95% CI, 1.05 to 1.22; nominal P = 0.0264). The majority of patients (zanubrutinib versus ibrutinib) had achieved a PR (76.1% versus 70.5%), followed by a CR (4.0% versus 2.5%), and a nodular PR (0.3% versus 0.0%).

In the ITT analysis set at the final PFS analysis cut-off of August 8, 2022, when ORR was assessed by the IRC, a higher response rate was observed in the zanubrutinib arm than in the ibrutinib arm (86.2% versus 75.7%), with a response ratio of 1.14 (95% CI, 1.05 to 1.22; nominal P = 0.0007). The majority of patients (zanubrutinib versus ibrutinib) had achieved a PR (78.6% versus 69.8%), followed by a CR (6.7% versus 5.5%), a nodular PR (0.9% versus 0.0%), and a CRi (0.0% versus 0.3%).

Generally similar findings were observed in the interim efficacy set (data cut-off: December 31, 2020). Refer to Table 45 in Appendix 1 for detailed outcome data.

ORR per IA in the ALPINE Trial

The analysis of ORR per IA was the key primary outcome in the ALPINE study and was adjusted for multiplicity in the prespecified interim analysis.

In the interim analysis ITT analysis set (data cut-off: December 31, 2020), ORR was analyzed in the first 415 randomized patients from the safety analysis set and according to actual treatment received. ORR per IA was higher in the zanubrutinib arm than in the ibrutinib arm (78.3% versus 62.5%). The majority of patients (zanubrutinib versus ibrutinib) had achieved a PR (75.8% versus 61.1%), followed by a CR (1.4% versus 1.4%), a nodular PR (0.5% versus 0.0%), and a CRi (0.5% versus 0.0%). In this analysis, the response ratio for zanubrutinib to ibrutinib was 1.25 (95% CI, 1.10 to 1.41), which was both noninferior (P < 0. 0001 versus prespecified 1-sided significance level of 0.005) and superior (P = 0.0006 versus prespecified 2-sided significance level of 0.0099).

In subgroup analyses, no major inconsistency was identified. In addition, several prespecified sensitivity analyses based on the ORR per IA were included in the statistical analysis plan, such as accounting for disease progression due to study drug interruption, accounting for death due to COVID-19 using the per-protocol analysis set, and changing the definition of best overall response that counted assessments of a PRL that were subsequently followed by a PR or a higher response as confirmed best overall responses of PR for patients with CLL. The results were generally consistent with results of the primary analysis for the first 415 randomized patients in the interim analysis ITT analysis set and showed response ratio values ranging from 1.11 (95% CI, 0.99 to 1.24) to 1.26 (95% CI, 1.11 to 1.43). Response ratios were similar, at 1.17 (95% CI, 1.04 to 1.33) for ORR per IRC and 1.25 (95% CI, 1.10 to 1.41) for ORR per IA. The noninferiority of zanubrutinib to ibrutinib was demonstrated against a noninferiority margin of 0.8558 with the 1-sided P < 0.0001 versus the prespecified 1-sided significance level of 0.005; the 2-sided P value was 0.0121 for superiority, which was higher than the prespecified 2-sided significance level of 0.0099. Refer to Table 45 in Appendix 1 for detailed outcome data.

The results in the final ORR analysis set ITT analysis set (data cut-off: December 1, 2021) were generally consistent with results from the interim efficacy set. Investigator-assess ORR was higher in the zanubrutinib arm than in the ibrutinib arm (79.5% versus 71.1%). In this analysis, the response ratio for zanubrutinib to ibrutinib was 1.12 (95% CI, 1.02 to 1.22; superiority 2-sided nominal P = 0.0013). Because the ALPINE trial met its primary end point at the prespecified interim analysis, P values for the final ORR analysis cut-off of December 1, 2021, were presented for descriptive purposes only. The majority of patients (zanubrutinib versus ibrutinib) had achieved a PR (73.7% versus 68.3%), followed by a CR (3.7% versus 2.5%), a CRi (1.2% versus 0.3%), and a nodular PR (0.9% versus 0). Furthermore, in subgroup analyses, the response rate favoured zanubrutinib across almost all analyzed subgroups, including difficult-to-treat patients with 17p deletion and/or TP53 mutations (rate difference = 21.3%; 95% CI, 7.0% to 35.7%). Refer to Table 45 in Appendix 1 for detailed outcome data.

The results in the final PFS analysis set ITT analysis set (data cut-off: August 8, 2022) were generally consistent with results from the interim efficacy set. Investigator-assess ORR was higher in the zanubrutinib arm than in the ibrutinib arm (83.5% versus 74.2%). In this analysis, the response ratio for zanubrutinib was 1.12 (95% CI, 1.04 to 1.22; superiority 2-sided nominal P = 0.0013). Because the ALPINE trial met its primary end point at the prespecified interim analysis, P values for the final PFS analysis cut-off of August 8, 2022, were presented for descriptive purposes only. The majority of patients (zanubrutinib versus ibrutinib) had achieved a PR (74.6% versus 64.9%), followed by a CR (6.1% versus 4.0%), a nodular PR (1.8% versus 0.0%), and a CRi (0.9% versus 0.9%).

Duration of Response

SEQUOIA Trial

The analysis of DOR per IRC and per IA was the secondary outcome in SEQUOIA cohorts 1 and 2. Note that this analysis was not part of the statistical hierarchy and was not adjusted for multiplicity.

DOR per IRC

There were 228 responders in the zanubrutinib arm and 203 responders in the BR arm based on IRC assessment. Among patients who achieved overall responses to treatment, median follow-up time per IRC was 22.1 months in both arms. Median duration of response per IRC was not reached in the zanubrutinib arm but was 30.6 months (95% CI, 25.5 to NE) in the BR arm. Event-free rates per IRC in the zanubrutinib arm versus the BR arm were 91.7% versus 81.3% at 18 months and 87.5% versus 70.3% at 24 months. In cohort 2, as of the data cut-off date of May 7, 2021, there were 99 responders in the zanubrutinib arm. Median follow-up time was 25.1 months. The median DOR was not reached in cohort 2. Event-free rates were 93.8% at 18 months and 91.6% at 24 months.

DOR per IA

There were 235 responders in the zanubrutinib arm and 211 responders in the BR arm based on IA. Among patients who achieved objective responses to treatment, median follow-up time per IA was 19.8 months in both arms. Median duration of response per IA was not reached in the zanubrutinib arm but was 30.6 months (95% CI, 26.2 to NE) in the BR arm. Event-free rates per IA in the zanubrutinib arm versus the BR arm were 93.1% versus 87.0% at 18 months and 88.1% versus 75.2% at 24 months. In cohort 2, there were 106 responders in the zanubrutinib arm. Median follow-up time was 24.9 months. The median DOR was not reached in cohort 2. Event-free rates were 89.3% at 18 months and 86.9% at 24 months.

ALPINE Trial

The analysis of DOR per IRC and per IA was the secondary outcome in the ALPINE study (Table 19). Note that this analysis was not part of the statistical hierarchy and was not adjusted for multiplicity.

DOR per IRC

At the final ORR analysis cut-off (December 1, 2021), there were 263 responders in the zanubrutinib arm and 237 responders in the ibrutinib arm based on IRC assessment. Events were observed for 34 patients (12.9%) in the zanubrutinib arm and 39 patients (16.5%) in the ibrutinib arm. Median follow-up time per IRC was 16.4 months in the zanubrutinib arm and 13.8 months in the ibrutinib arm. The median DOR was not reached in either arm. Event-free rates per IRC in the zanubrutinib arm versus the ibrutinib arm were 91.6% versus 86.4% at 12 months, and 76.0% versus 71.2% at 24 months.

At the final PFS analysis cut-off (August 8, 2022), there were 282 responders in the zanubrutinib arm and 246 responders in the ibrutinib arm based on IRC assessment. Events were observed for 60 patients (21.3%) in the zanubrutinib arm and 69 patients (28.0%) in the ibrutinib arm. Median follow-up time per IRC was 22.3 months in the zanubrutinib arm and 21.7 months in the ibrutinib arm. The median DOR was not reached in the zanubrutinib arm with the lower bound of the 95% CI of 31.3 months, and median follow-up time per IRC was 33.9 months (95% CI, 32.2 to 41.4 months) in the ibrutinib arm. Event-free rates per IRC in the zanubrutinib arm versus the ibrutinib arm were 98.6% versus 97.5% at 12 months, 77.4% versus 67.8% at 24 months, and 54.7% versus 28.3% at 36 months.

Generally similar findings were observed in the interim efficacy set (data cut-off: December 31, 2020), Refer to Table 45 in Appendix 1 for detailed outcome data.

DOR per IA

At the final ORR analysis cut-off (December 1, 2021), there were 260 responders in the zanubrutinib arm and 231 responders in the ibrutinib arm based on IA. Events were observed for 28 patients (10.8%) in the zanubrutinib arm and 40 patients (17.3%) in the ibrutinib arm. Median follow-up time per IA was 16.6 months in the zanubrutinib arm and 13.8 months in the ibrutinib arm. The median DOR per IA was not reached in either arm, but the lower bound of 95% CI was 24.6 months in the ibrutinib arm. Event-free rates per IA in the zanubrutinib arm versus the ibrutinib arm were 92.2% versus 85.8% at 12 months, and 78.6% versus 70.1% at 24 months.

At the final PFS analysis cut-off (August 8, 2022), there were 273 responders in the zanubrutinib arm and 241 responders in the ibrutinib arm based on IA. Events were observed for 53 patients (19.4%) in the zanubrutinib arm and 62 patients (25.7%) in the ibrutinib arm. Median follow-up time per IA was 22.2 months in the zanubrutinib arm and 19.4 months in the ibrutinib arm. The median DOR per IA was not reached in the zanubrutinib arm with the lower bound of the 95% CI of 31.3 months, whereas median follow-up time per IRC was 33.9 months (95% CI, 32.2 to NE) in the ibrutinib arm. Event-free rates per IA in the zanubrutinib arm versus the ibrutinib arm were 91.9% versus 86.1% at 12 months, 79.5% versus 71.3% at 24 months, and 60.6% versus 21.5% at 36 months.

Generally similar findings were observed in the interim efficacy set (data cut-off: December 31, 2020), albeit with lower DOR per IA event rates that were likely related to the shorter follow-up period in the interim analysis. Refer to Table 45 in Appendix 1 for detailed outcome data.

Time to Treatment Failure

SEQUOIA Trial

Time to treatment failure was not assessed in the SEQUOIA trial.

ALPINE Trial

The analysis of time to treatment failure was the secondary outcome in the ALPINE study. Note that this analysis was not part of the statistical hierarchy and was not adjusted for multiplicity.

At the final ORR analysis cut-off (December 1, 2021), the proportion of patients with treatment failure was lower in the zanubrutinib arm than in the ibrutinib arm (19.3% versus 33.2%). With a median follow-up time of 25.1 months in each treatment arm, the median time to treatment failure was not reached in either arm. The HR of treatment failure comparing the zanubrutinib arm to the ibrutinib arm was 0.50 (95% CI, 0.36 to 0.68). The 24-month event-free rate was 79.5% in the zanubrutinib arm and 63.8% in the ibrutinib arm.

At the final PFS analysis cut-off (August 8, 2022), the proportion of patients with treatment failure was lower in the zanubrutinib arm than in the ibrutinib arm (26.3% versus 41.2%). With the median follow-up time of 33.2 months in the zanubrutinib arm and 33.4 months in the ibrutinib arm, the median time to treatment failure was not reached in either arm, whereas the lower end of 95% CI was 34.4 in the ibrutinib arm. The HR of treatment failure comparing the zanubrutinib arm to the ibrutinib arm was 0.54 (95% CI, 0.41 to 0.72; nominal P < 0.0001). The 24-month event-free rate was 79.9% in the zanubrutinib arm and 65.0% in the ibrutinib arm.

Generally similar findings were observed in the interim efficacy set (data cut-off: December 31, 2020), albeit with lower treatment failure event rates that were likely related to the shorter follow-up period in the interim analysis. Refer to Table 45 in Appendix 1 for detailed outcome data.

European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 ||||| ||||||| ||||||| ||| ||| ||||||| ||| |||||| |||||| |||| |||||||||| || ||||| || ||| ||||| ||| |||||||||||||

SEQUOIA Trial

|| |||||| || || ||| |||| ||||||| || ||| ||| ||||| || ||| |||||||||||| |||| ||| |||| |||| ||||||| |||| |||||||| || |||| || || |||||| |||||| ||||||| |||||||| |||||||||||| |||| |||||||||||| |||||||| |||||||||||||||| ||||||||| ||| |||| |||| ||| ||||||| ||| ||||||| ||| ||||||| |||| |||||||| |||| |||||||| |||| |||||||| ||| |||| |||||||| ||||||||||||| || ||| || |||| ||| |||| |||| ||||||| |||| |||||||| || |||| || || |||||| |||||| ||||||| |||||||| |||||||||||| |||| |||||||||||| |||||||| |||||||||||||||| ||||||||| ||| |||| |||| |||| ||||||| |||| ||||||| |||| ||||||| |||| |||||||| ||| ||||||||| ||| |||||||| ||| |||| ||||||||| ||||||||||||| |||||| ||||||| ||||||| |||| ||| |||||||| ||| |||||| ||

ALPINE Trial

||| |||| ||||||| |||| |||| |||||||| || ||| ||| |||||||| |||||| |||||| | ||| ||||| ||| || ||| ||||| ||| |||||||| ||| ||||| ||| |||||||| |||| |||||||||| || ||||| ||| ||| ||| ||||| ||| |||||||| ||||||| ||||||||| ||| |||||| || ||| |||||||||||| |||| ||| |||| |||| ||||||| |||| |||||||| || ||||| || || |||||| |||||| ||||||| |||||||| |||||||||||| |||| |||||||||||| |||||||| |||||||||||||||| ||||||||| ||| |||| |||| |||| |||||||| |||| |||||||| |||| |||||||| |||||| |||||||| ||||| ||||||| ||||| |||||||| ||| ||||| |||||||| ||||||||||||| || ||| ||||||||| |||| ||| |||| |||| ||||||| |||| |||||||| || ||||| || || |||||| |||||| ||||||| |||||||| |||||||||||| |||| |||||||||||| |||||||| |||||||||||||||| ||||||||| ||| |||| |||| |||| |||||||| |||| |||||||| |||| |||||||| ||||| |||||||| ||||| |||||||| |||| |||||||| ||| ||||| |||||||| ||||||||||||| ||||||||| ||||||| |||||||| |||| |||||||| || ||||| || ||| ||| ||||| ||| |||||||| ||||||| ||||||| ||| |||||| || ||| |||||||||||| |||| ||| |||| |||| ||||||| |||| |||||||| || ||||| || || |||||| |||||| ||||||| |||||||| |||||||||||| |||| |||||||||||| |||||||| |||||||||||||||| ||||||||| ||| |||| |||| |||| |||||||| |||| |||||||| |||| |||||||| |||||| |||||||| | |||| ||||||| ||||| ||||||||| ||| ||||| |||||||| ||||||||||||| || ||| ||||||||| |||| ||| |||| |||| ||||||| |||| |||||||| || ||||| || || |||||| |||||| ||||||| |||||||| |||||||||||| |||| |||||||||||| |||||||| |||||||||||||||| ||||||||| ||| |||| |||| |||| |||||||| |||| |||||||| |||| |||||||| ||||| |||||||| ||||| |||||||| |||| |||||||| ||| ||||| |||||||| ||||||||||||| ||||||||| ||||||| |||||||| |||| |||||||| || ||||| |||||||||||| ||||||| |||||||| |||| |||||||| || ||| ||||||| |||||||| ||| ||||| |||||||| |||||||| ||| |||||| |||||| ||||| || ||||| || || ||| |||||||| | ||| |||||||| ||||||| |||||

EQ-5D-5L

SEQUOIA Trial

|| |||||| || ||||||||| | |||||| |||||| |||||||||| || |||||||| || ||| |||||||||||| ||| |||||||| || |||||||| || |||||||| |||||||| ||||||||| |||||||| |||||||||| |||| ||| || ||| || |||| ||| ||||| ||||||| ||||||| ||||||||| |||||||||| ||||| ||||||||||| ||| || |||||||||| || |||||||||||||||| ||| ||||||||||||||||||| ||| |||| |||| ||||||| |||| |||||||| || ||| ||| ||||| || ||| |||||||| |||| |||||||| || ||| |||||||||||| ||| |||| |||||||| ||| ||| || ||| |||| |||||||| || |||| ||| || |||| || || ||| |||||||||||| ||| |||| |||||||| ||| ||| || ||| |||| |||||||| || |||| |||||||||||| ||||||| |||| ||| |||||||| ||| |||||| ||

ALPINE Trial

|| ||| ||||| ||| |||||||| ||||||| ||||||||| ||| |||||| || ||||| | ||| ||||| ||| ||||||||| | |||||||| |||||| ||||||||||| || |||||||| |||||||| || |||||||| || |||||||| ||||||| ||||||||| ||||||||| |||||||||| ||| ||||||||||||||| || ||| |||||||||||| ||| |||| ||| ||||||||| |||| ||||| || ||| ||||| |||||||||| ||| |||||||||||||||||| |||||||| | |||||||| ||||| ||||||||||| || |||||||| |||||||| || |||||||| || ||| |||||||||||| ||| |||| ||| ||||||||| |||| |||||||||||| || ||| |||| |||| ||||||| |||| |||||||| || ||| ||| ||||| || ||| |||||||| |||| |||||||| || ||| |||||||||||| ||| |||| ||||||| ||| ||| ||||||||| ||| |||| |||| ||| || ||||| || || |||| || || ||| |||||||||||| ||| |||| ||||||| ||| ||| ||||||||| ||| |||| ||||||| || ||||| |||||| ||| ||||| ||| |||||||| ||||||| ||||||| ||| |||||| || ||||| | ||| ||||| ||| ||||||||| | |||||||| |||||| ||||||||||| || |||||||| |||||||| || |||||||| || |||||||| ||||||| ||||||||| ||||||||| |||||||||| ||| ||||||||||||||| || ||| |||||||||||| ||| |||| ||| ||||||||| |||| ||||| || ||| ||||| |||||||||| ||| |||||||||||||||||| |||||||| | |||||||| ||||| ||||||||||| || |||||||| |||||||| || |||||||| || ||| |||||||||||| ||| |||| ||| ||||||||| |||| |||||||||||| || ||| |||| |||| ||||||| |||| |||||||| || ||| ||| ||||| || ||| |||||||| |||| |||||||| || ||| |||||||||||| ||| |||| ||||||| ||| ||| ||||||||| ||| |||| |||| ||| || ||||| || || |||| || || ||| |||||||||||| ||| |||| ||||||| ||| ||| ||||||||| ||| |||| ||||||| || ||||| ||||||||||||| ||||||| |||||||| |||| |||||||| || ||| ||||||| |||||||| ||| ||||| |||||||| |||||||| ||| |||||| |||||| ||||| || ||||| || || ||| |||||||| | ||| |||||||| ||||||| |||||

Table 18: Summary of Key Efficacy Results From the SEQUOIA Trial (ITT Analysis Set for Cohort 1, Safety Analysis Set for Cohort 2)