MAIC Analysis of Zanubrutinib Vs Acalabrutinib Shows Superior Efficacy in CLL

The phase 3 ALPINE and ASCEND studies were used in the MAIC analysis assessing zanubrutinib vs acalabrutinib in relapsed/refractory chronic lymphocytic leukemia.

Zanubrutinib (Brukinsa) vs acalabrutinib (Calquence) demonstrated a significant improvement in progression-free survival (PFS) and complete response (CR) in patients with chronic lymphocytic leukemia (CLL), which was analyzed using a matching-indirect comparison (MAIC) analysis, presented at the 28th International Congress of Hematologic Malignancies.¹

After the baseline characteristics of the phase 3 ALPINE (NCT03734016) and ASCEND (NCT02970318) trial populations were matched for prognostic and predictive variables, zanubrutinib was associated with significant improvements in investigator-assessed PFS vs acalabrutinib in the unadjusted population (HR, 0.77; 95% CI, 0.55-1.07; P = .1213) and in the base case adjusted populations (HR, 0.68; 95% CI, 0.46-0.99; P = .0448). Moreover, overall survival (OS) outcomes potentially favored zanubrutinib post-matching in the unadjusted population (HR, 0.60; 95% CI, 0.37-0.97; P = .0354) and in the base case adjusted population (HR, 0.60; 95% CI, 0.35-1.02; P = .0575) in the base case adjusted populations.

“The CLL landscape is evolving rapidly, and this MAIC provides timely comparative effectiveness data for physicians, and [it] reinforces [zanubrutinib’s] role as a foundational CLL treatment via a robust evaluation of the efficacy in the ASCEND and ALPINE studies; the presented analysis not only accounts for differences in key patient characteristics, but also clarifies the impact COVID-19 may have had on study outcomes,” Mazyar Shadman, MD, MPH, study author and Innovators Network Endowed Chair, associate professor of Hematology and Oncology, Lymphoid Malignancies and Immunotherapy, at Fred Hutch Cancer Center and University of Washington, stated in a news release.²

To date, zanubrutinib is the only next-generation covalent BTK inhibitor to elicit superior PFS outcomes vs the first-generation BTK inhibitor ibrutinib (Imbruvica) in patients with relapsed/refractory CLL, as demonstrated in the ALPINE trial. The ALPINE data supported the 2023 FDA approval of zanubrutinib for patients with CLL or small lymphocytic lymphoma (SLL).³

Furthermore, in the ASCEND trial, the second-generation covalent BTK inhibitor acalabrutinib induced improved PFS outcomes compared with rituximab (Rituxan) in combination with idelalisib (Zydelig) or bendamustine in this patient population. The ASCEND data supported the 2019 FDA approval of acalabrutinib for adult patients with CLL or SLL.⁴ However, the phase 3 ELEVATE-RR trial (NCT02477696) showed that acalabrutinib was noninferior to ibrutinib (Imbruvica) in patients with relapsed/refractory disease with 17p or 11q deletions.¹

No trials have compared zanubrutinib head-to-head with acalabrutinib in patients with relapsed/refractory CLL. Thus, investigators performed a MAIC to determine the relative efficacy of these 2 agents.

Investigators matched individual patient data (IPD) from ALPINE with aggregated data from ASCEND and compared these data using unanchored MAIC methodology, as the 2 trials lacked a common comparator arm. ALPINE data were adjusted to account for the effects of COVID-19 due to the timing of the trial relative to the pandemic. A base case analysis population adjustment included all variables that were identified as prognostic factors or predictors of treatment efficacy. Investigators obtained pseudo PFS and OS IPD from the acalabrutinib arm of ASCEND by reconstructing the Kaplan-Meier curves from the published ASCEND data.

Investigators collected IPD from 327 patients enrolled in ALPINE and 155 patients enrolled in ASCEND. The ALPINE IPD had a September 2023 data cutoff and a median follow-up of 39 months. The ASCEND IPD had a data cutoff of October 2020 and a median follow-up of 36 months. After conducting a sensitivity analysis of scenarios to consider the effects of matching for different sets of variables, then matching, reweighing, and adjusting for variables, the ALPINE intent-to-treat population was filtered to include only patients with existing data regarding the selected baseline characteristics, excluding those with SLL, for a total of 308 patients. Population adjustment brought the ALPINE effective sample size to 184.8 patients, translating to 60% of the starting filtered population.

The primary outcomes of this analysis were investigator-assessed PFS and OS, which were compared between the 2 studies using a weighted Cox proportional hazard model, and CR, which was compared using a weighted logistic regression model.

Baseline characteristics were similar between the ASCEND and post-matching ALPINE populations. In the ASCEND population, 21.9% of patients were 75 years of age or older, most patients (69.7%) were male, and 37.4% of patients had an ECOG performance status (PS) of 0. Most patients (84.5%) were from Europe. Regarding genomic status, 16.2%, 17.4%, 25.2%, and 25.2% of patients had IGHV mutations, 17p deletions, 11q deletions, and TP53 mutations, respectively. Furthermore, 32.4% and 49.0% of patients had a complex karyotype and bulky disease, respectively. A total of 77.4% of patients had beta2-microglobulin levels higher than 3.5 mg/L, and 58.1% of patients had Rai stage 0 to II disease or Binet A/B disease. In total, 25.8%, 11.0%, and 10.3% of patients had received 2, 3, and at least 4 prior therapies, respectively; which consisted of an anti-CD20 antibody (83.9%), alkylators other than bendamustine (85.8%), previous treatments consisted of (30.3%), and a purine analog (70.3%). Patients had a median absolute lymphocyte count of 48.9 x 109 cells/L, a median absolute neutrophil count of 3.8 x 109 cells/L, and a median platelet count of 119.5 x 109 cells/L.

In the post-matching ALPINE population, 21.9% of patients were 75 years of age or older, most patients (69.7%) were male, and 37.4% of patients had an ECOG PS of 0. Most patients (84.5%) were from Europe. Regarding genomic status, 16.2%, 17.4%, 25.2%, and 25.2% of patients had mutated IGHV, 17p deletions, 11q deletions, and TP53 mutations, respectively. Furthermore, 28.6% and 49.0% of patients had a complex karyotype and bulky disease, respectively. A total of 62.8% of patients had beta2-microglobulin levels higher than 3.5 mg/L, and 58.1% of patients had Rai stage 0 to II disease or Binet A/B disease. In total, 25.8%, 11.0%, and 10.3% of patients had received 2, 3, and at least 4 prior therapies, respectively, which included an anti-CD20 antibody (83.9%), alkylators other than bendamustine (85.8%), bendamustine (30.3%), and a purine analog (70.3%). Patients had a median absolute lymphocyte count of 49 x 109 cells/L, a median absolute neutrophil count of 4 x 109 cells/L, and a median platelet count of 119.0 x 109 cells/L.

CR outcomes favored zanubrutinib in both the unadjusted population (odds ratio [OR], 2.88; 95% CI, 1.18-7.02; P = .0198) and the base case adjusted population (OR, 2.90; 95% CI, 1.13-7.43; P = .0270), respectively. PFS, OS, and CR results from the 6 sensitivity analysis populations were consisted with the base case adjusted population results.

The study authors noted several limitations of this research, including the potential for bias resulting from the assumption that differences between trials can be completely explained by variables selected for matching; the lack of an analysis of PFS per independent review committee in the MAIC due to data unavailability; and the lack of a safety comparison between zanubrutinib and acalabrutinib because of differing treatment exposure times across the 2 trials.

“Head-to-head randomized clinical trials are the gold standard when it comes to evaluating the potential impact of individual treatments for patients. MAICs are intended to be hypothesis-generating, provided they are conducted with appropriate rigor to minimize potential biases,” Shadman concluded.

References

1. Shadman M, Brown JR, Williams R, et al. Efficacy of zanubrutinib versus acalabrutinib in the treatment of relapsed or refractory chronic lymphocytic leukemia (R/R CLL): a matching-adjusted indirect comparison (MAIC). Presented at: 28th International Congress on Hematologic Malignancies; February 29 to March 3, 2024. Miami, Florida.
2. BeiGene announces new efficacy analysis comparing Brukinsa vs acalabrutinib in relapsed or refractory chronic lymphocytic leukemia. News release. BeiGene. February 29, 2024. Accessed March 1, 2024. https://shorturl.at/aeAGT
3. FDA approves zanubrutinib for chronic lymphocytic leukemia or small lymphocytic lymphoma. FDA. January 19, 2023. Accessed March 1, 2024. https://shorturl.at/drQ29
4. FDA takes second action under international collaboration, approves new treatment option for patients with chronic lymphocytic leukemia. FDA. November 21, 2020. Accessed March 1, 2024. https://shorturl.at/fjkO3