Results presented in a poster at the 2023 ASCO Annual Meeting demonstrated similar progression-free survival between acalabrutinib and zanubrutinib in patients with chronic lymphocytic leukemia.
Findings from an unanchored matching-adjusted indirect comparison of data from the phase 3 ASCEND (NCT02970318) and ALPINE (NCT03734016) trials showed that treatment with acalabrutinib (Calquence) and zanubrutinib (Brukinsa) led to similar investigator-assessed progression-free survival (PFS) rates in patients with relapsed/refractory chronic lymphocytic leukemia (CLL), respectively.1
Results were presented in a poster at the 2023 ASCO Annual Meeting. The primary efficacy analysis revealed that the unanchored matching-adjusted indirect comparison produced a similar hazard ratio (HR) for investigator-assessed PFS similar for acalabrutinib (n = 99) and zanubrutinib (n = 327; HR, 0.90; 95% CI, 0.60-1.36). The 12-month investigator-assessed PFS rate for acalabrutinib was 91% (95% CI, 84%-95%) and 76% (95% CI, 66%-84%) at 24 months. The 12-month investigator-assessed PFS rate for zanubrutinib was 92% (95% CI, 88%-94%) and 78% (95% CI, 73%-83%) at 24 months. Finally, the 12-month investigator-assessed PFS rate for ibrutinib (Imbruvica; n = 325) was 84% (95% CI, 79%-88%) and 66% (95% CI, 60%-71%) at 24 months.
Acalabrutinib and zanubrutinib were compared with ibrutinib in patients with relapsed/refractory CLL in the phase 3 ELEVATE-RR (NCT02477696) and ALPINE trials, respectively. However, differences in enrollment prevent anchored indirect treatment comparisons between the second-generation agents. As such, investigators performed an unanchored matching-adjusted indirect comparison of the safety and efficacy of acalabrutinib vs zanubrutinib with patient data from the ASCEND and ALPINE trials.
The ASCEND trial evaluated treatment with acalabrutinib vs chemotherapy and the ALPINE trial evaluated zanubrutinib vs ibrutinib, both in patients unrestricted by 17p or 11q deletions who had received 1 median prior line of therapy. The ELEVATE-RR trial evaluated acalabrutinib vs ibrutinib in patients with 17p and/or 11q deletions who had received a median of 2 prior lines of therapy.
In the unanchored MAIC, data with acalabrutinib (n = 155) from ASCEND were weighted to match aggregate data with zanubrutinib (n = 327) from ALPINE. Prognostic and predictive variables for investigator-assessed PFS, compiled from an exploratory multivariate Cox regression analysis of ASCEND, were also included in the matching. These variables included age, region of enrollment, sex, ECOG performance status, bulky disease, prior chemoimmunotherapy, 17p or 11q deletions, TP53 mutations without 17p deletions, IGHV mutational status, number of prior lines of therapy, and Rai stage.
Investigator-assessed PFS was assessed in all randomized patients with complete baseline data before and after matching. Pseudo–individual patient data for investigator-assessed PFS were derived from Kaplan-Meier curves using the algorithm by Guyot et al.2
Additional secondary efficacy analyses included the evaluation of individual patient data with acalabrutinib vs aggregate data from the ALPINE trial with ibrutinib (n = 325). Notably, the analysis used the same weights as the primary analysis and compared treatments in a similar patient population, differing from the population seen in the ELEVATE-RR trial.
Patient data from ASCEND were also pooled with individual patient data from the ELEVATE-RR trial in a sensitivity analysis to create a combined acalabrutinib arm (n = 406). This population
was matched with zanubrutinib data from ALPINE with the same variables as the primary analysis.
Finally, investigators evaluated odds ratios (ORs) of adverse effects (AEs) in treated patients with baseline data (acalabrutinib, n = 148; zanubrutinib, n = 324). An artificial cut-off date of February 21, 2020, was selected to match median treatment exposure for both agents. Additionally, the safety analysis also used the same matching variables as the primary analysis for consistency.
Additional results demonstrated that prior to matching, both the zanubrutinib and acalabrutinib populations were well balanced, except for region of enrollment. After matching, the effective sample size of the acalabrutinib arm was 99, which represented 66% of the original efficacy sample. A total of 65% of these patients were male and had a median age of 66 years. Notably, there we no differences between matched variables and differences between non-matched variables were reduced.
Following matching in the secondary analysis, acalabrutinib demonstrated superior investigator-assessed PFS to ibrutinib (HR, 0.60; 95% CI, 0.40-0.90). Moreover, in the sensitivity analysis, which reduced the effective sample size with acalabrutinib to 33% of the original pooled arm, no differences were reported in investigator-assessed PFS between the pooled acalabrutinib and zanubrutinib arms (HR, 0.92; 95% CI, 0.64-1.34).
Overall, matching had little impact on the safety results of acalabrutinib, and the safety profiles seen with zanubrutinib and acalabrutinib were similar, with some exceptions. This included the risk of a serious AE (OR, 0.61; 95% CI, 0.39-0.97), any grade hypertension (OR, 0.18; 95% CI, 0.09-0.37), grade 3 or higher hypertension (OR, 0.22; 95% CI, 0.09-0.54), any grade hemorrhage (OR, 0.54; 95% CI, 0.34-0.87), or an AE leading to dose reduction (OR, 0.30; 95% CI, 0.14-0.67), all of which occurred more often in patients treated with zanubrutinib vs acalabrutinib. Notably, pre-matching, lower cumulative incidence rates of any grade hypertension were seen in the acalabrutinib vs zanubrutinib arm. These resuls are considered to be hypothesis-generating, considering the limitations of MAIC anlayses.