MAIC Analysis Shows PFS Favorability With Zanubrutinib vs Acalabrutinib/Venetoclax in CLL

Image Credit: © Bipul Kumar - stock.adobe.com

The relative efficacy of zanubrutinib (Brukinsa) vs acalabrutinib (Calquence) plus venetoclax (Venclexta; AV) demonstrated a significant progression-free survival (PFS) benefit compared with AV in patients with chronic lymphocytic leukemia (CLL), according to a matching-adjusted indirect comparison (MAIC) analysis, presented at the 2025 European Hematology Association Congress.1

The unadjusted comparison of investigator-assessed PFS (INV-PFS) for zanubrutinib in the phase 3 SEQUOIA trial (NCT3336333) compared with AV in the phase 3 AMPLIFY trial (NCT0386261) revealed that zanubrutinib demonstrated a significant benefit (HR, 0.47; 95% CI, 0.28-0.77; P = .003). In a similar trend, the population-adjusted INV-PFS demonstrated that zanubrutinib was also favored compared with AV (HR, 0.26; 95% CI, 0.13-0.54; P = .0003). Notably, the 36-month PFS rate was 85.6% for zanubrutinib before matching, and was 88.5% following matching in the base case, compared with AV at 76.5%.

“In the absence of head-to-head comparative trials, the indirect comparison statistical analyses were applied to compare the efficacy of zanubrutinib vs AV,” lead author Talha Munir, MBChB, PhD, and colleagues wrote in a poster presentation. “Results should be interpreted with considerations of inherent limitations of indirect comparison, such as MAIC model assumptions, [such as] the assumption that cross-trial differences in patient populations can be entirely explained by matching variables.”

Munir is a consultant hematologist at Leeds Teaching Hospitals NHS Trust and deputy chair of the United Kingdom (UK) National Cancer Research Institute Chronic Lymphocytic Leukemia Study Group in Leeds, UK.

Background of the SEQUOIA and AMPLIFY Trials

The open-label, randomized SEQUOIA trial evaluated the efficacy of zanubrutinib vs bendamustine plus rituximab (Rituxan; BR) in patients with treatment-naive CLL or small lymphocytic lymphoma, including patients without 17p deletions (Cohort 1), and those with 17p deletions (Cohort 2 and Cohort 3). Patients on the study were randomly assigned 1:1 to receive zanubrutinib (arm A) or BR (arm B).2 Furthermore, the global, multicenter, open-label, randomized AMPLIFY trial assessed the efficacy and safety of acalabrutinib plus venetoclax compared with chemoimmunotherapy, consisting of either fludarabine, cyclophosphamide, and rituximab (FCR) or BR, among patients with treatment-naive CLL without 17p deletions or TP53 mutation.3

Understanding the MAIC Methodology

Datasets for the MAIC from the SEQUOIA and AMPLIFY trials had similar follow-ups, with 43.7 months in the SEQUOIA study and 41.0 months in the AMPLIFY study.1 With BR and FCR/BR as assumed control arms in the respective studies, the comparison of zanubrutinib and AV was performed in the anchored MAIC. Of note, individual patient data of low-risk patients with CLL—without 17p deletions or TP53 mutations—from the SEQUOIA study were re-weighed to accurately match the main baseline patient characteristics in the AMPLIFY trial. Individual patient data for the AMPLIFY study were reconstructed and generated via digitized Kaplan-Meier curves of PFS. Sensitivity analyses were also performed in model scenarios of different matching variables.

“In the absence of head-to-head clinical trials, an anchored MAIC was conducted to investigate the comparative efficacy of zanubrutinib and AV in patients with low-risk treatment-naive CLL without 17p deletions or TP53 mutations,” the study authors wrote on the poster.

The MAIC included individual patient-level data from the intention-to-treat population of the SEQUOIA trial (n = 479) and published aggregate data from the AMPLIFY trial (AV, n = 291; FCR/BR, n = 290). Notably, identified variables that pose as potential treatment effect modifiers or prognostic factors for matching include age, sex, ECOG performance status, disease stage, 11q deletions, IGHV gene mutation status, geographic region, complex karyotype, cumulative illness rating scale, and creatinine clearance. Moreover, sensitivity analyses of model scenarios were performed to identify potential differences in matching values, independent review committee–assessed PFS (IRC-PFS), and unanchored MAIC of FCR and BR.

After matching, reweighing, and adjusting variables, the zanubrutinib low-risk population from the SEQUOIA trial (n = 389) was identified, and after population adjustments, the effective sample size for SEQUOIA was 126 patients.

Baseline Patient Characteristics

Baseline patient characteristics were measured for patients from the AMPLIFY trial and low-risk patients from the SEQUOIA trial pre- and post-matching.

Among patients in the AMPLIFY trial (n = 581), 26.8% were 65 years of age or older, and the majority were male (64.5%) and had an ECOG performance status of 0 or 1 (91.8%). Binet stage AB or Rai 0 to II was observed in 56.2% of patients. Of note, 17.6% of patients had 11q deletions and 58.6% had unmutated IGHV.

In the SEQUOIA study pre- and post-matching, 78.7% and 26.8% of patients were 65 years of age or older, respectively, and 61.7% and 64.5% were male. ECOG performance status of 0 or 1 was observed in 93.3% and 91.8% of patients, and 70.2% and 56.2% of patients had Binet stage AB or Rai 0 to II. Additionally, 19.5% and 17.6% of patients had 11q deletions, and 52.7% and 58.6% of patients had unmutated IGHV.

Additional PFS Outcomes

Landmark PFS rates were established based on the 2 studies and treatment arms at 12, 24, 36, and 48 months. In the SEQUOIA study, the 12-, 24-, 36-, and 48-month INV-PFS rates zanubrutinib arm were 97%, 94.2%, 88.5%, and 86.7%, respectively; these INV-PFS rates in the BR arm were 83.6%, 71.8%, 47.8%, and 32.2%. Furthermore, in the AMPLIFY study, the INV-PFS rates AV arm at the respective time points were 96%, 91%, 79%, and 67%; these rates in the FCR/BR arm were 88%, 79%, 66%, and 52%. The IRC-PFS rates for the AV arm in the AMPLIFY study at the same time points were 95%, 88%, 77%, and 64%; these rates for the FCR/BR arm were 88%, 79%, 67%, and 49%.

Notably, the sensitivity analyses demonstrated consistent results based on model scenarios of different matching variables, along with IRC-PFS (HR, 0.23; 95% CI, 0.12-0.48; P < .0001) and in unanchored MAIC without the FCR/BR control arm assumption (HR, 0.44; 95% CI, 0.12-0.89; P = .0220)

Disclosures: Dr Munir reported serving in consultancy and advisory roles for AbbVie, Alexion Pharmaceuticals, AstraZeneca, BeOne, Janssen-Cilag, Lilly, Roche, and SOBI; participation in speakers’ bureau for AbbVie, Alexion Pharmaceuticals, AstraZeneca, Gilead Sciences, Janssen-Cilag, and SOBI; and travel, accommodations, or expenses from AbbVie, Alexion Pharmaceuticals, AstraZeneca, and Janssen-Cilag.

References

1. Munir T, Yang K, Xu S, Williams R, Shadman M. Comparative efficacy of zanubrutinib versus fixed-duration acalabrutinib plus venetoclax for first-line treatment of chronic lymphocytic leukemia (CLL): a matching-adjusted indirect comparison (MAIC). Presented at: 2025 European Hematology Association Congress; June 12-15, 2025; Milan, Italy. Abstract PS1581.
2. A study comparing zanubrutinib with bendamustine rituximab in participants with previously untreated CLL or SLL (SEQUOIA). ClinicalTrials.gov. Updated March 7, 2025. Accessed June 16, 2025. https://clinicaltrials.gov/study/NCT03336333
3. Study of acalabrutinib (ACP-196) in combination with venetoclax (ABT-199), with and without obinutuzumab (GA101) versus chemoimmunotherapy for previously untreated CLL (AMPLIFY). ClinicalTrials.gov. Updated December 27, 2024. Accessed June 16, 2025. https://clinicaltrials.gov/study/NCT03836261