Real-World Comparison of BCL2 Inhibitor and BTK Inhibitor Therapy in the Front-Line Treatment of CLL Including Patients with Del(17p)/TP53 Mutation

Background For patients (pts) with chronic lymphocytic leukemia (CLL), treatment with either continuous Bruton's tyrosine kinase inhibitor (BTKi) or time-limited B-cell Lymphoma 2 inhibitor (BCL2i) therapy is standard of care in the first-line (1L) setting. However, there are no reported prospective trials and limited retrospective data available evaluating the comparative efficacy of BTKi and BCL2i to guide therapeutic choice. Additionally, there are no head-to-head data evaluating efficacy or outcomes for pts with TP53 disrupted (TP53D) CLL, which is associated with poor prognosis and short median overall survival (OS). Most recommendations favor BTKi for TP53D CLL, largely extrapolated from cross-trial comparisons, given the perceived longer PFS and OS of pts with TP53D receiving 1L BTKi compared to BCL2i. In this study, we compare the real-world outcomes of pts with CLL, including those with TP53D, treated in the 1L setting with either BTKi or BCL2i therapy. Methods This study utilized the nationwide Flatiron Health electronic health record-derived de-identified database. Pts were included if they had a diagnosis of CLL and received 1L treatment between 1/2015 to 4/2024. Pts were categorized into either BTKi or BCL2 cohorts based on 1L treatment received. BTKi therapy included ibrutinib, acalabrutinib, or zanubrutinib. BCL2i therapy included venetoclax +/- rituximab or obinutuzumab. TP53D sub-cohort was defined by documented TP53 mutation and/or Del17p. Time to next treatment (TTNT), was defined as time from the start of 1L treatment to the initiation of second line treatment or death. TTNT and OS from 1L treatment initiation were estimated by Kaplan-Meier analysis and compared using adjusted Cox regression modeling. Results A total of 3,907 pts with CLL met inclusion criteria, of which 3,334 (85%) received BTKi and 573 (15%) received BCL2i as 1L therapy. The majority of pts were White (71%), male (63%), and had an ECOG of 0-1 (67%). The median age was 69 (62-76) for those receiving 1L BTKi, and 67 (58-73) for those receiving 1L BCL2i (p<0.001). The median follow-up for BTKi was 28 months (mos) vs. 19 mos for BCL2i, with a median duration of treatment (DOT) for BTKi of 19 mos vs. 17 mos for BCL2i (all p<0.001). There was no significant OS difference observed between 1L BTKi when compared to 1L BCL2i (p=0.08). At 2 years, the TTNT was significantly shorter with BTKi compared to BCL2i (67% vs 79%; p<0.001). On multivariate Cox regression, BTKi treatment remained a significant predictor for shorter TTNT (HR 1.57; CI 1.28-1.94, p<0.001). Increasing age (HR 1.03; CI 1.02-1.04, p<0.001) and ECOG ≥2 (HR 1.68; CI 1.42-1.98, p<0.001) were also predictors of shorter TTNT, while Del11q, Del13q, Trisomy 12, and IGHV mutational status had no significant association with TTNT. A total of 644 pts with TP53D were identified, with the majority receiving BTKi (n=577, 90%) compared to BCL2i (n=67, 10%), and no significant difference in median age (BTKi: 70 [63-78] vs. BCL2i: 74 [61-76], p=0.99). The median follow-up was 29 mos for BTKi vs. 24 mos for BCL2i. The median DOT for pts receiving BTKi was 20 mos vs. 19 mos for BCL2i (p=0.25). For pts with TP53D receiving BCL2i, the majority (66%) received therapy for >1 year. Importantly, there was no observed OS difference between pts treated with 1L BTKi as compared to 1L BCL2i (p=0.63). With regards to TTNT, there was no significant difference in 2-year TTNT between BTKi and BCL2i (p=0.63). On multivariate analysis, there was no significant difference in TTNT for TP53D pts treated with BTKi as compared to BCL2i (HR 1.01; CI 0.72-1.68, p=0.66). Additionally, increasing age (HR 1.01; CI 1.01-1.03, p=0.04) and ECOG ≥2 (HR 1.78; CI 1.27-2.55) were independent predictors of shorter TTNT in this subgroup. In contrast, IGHV mutated status (HR 0.58; CI 0.35-0.95, p=0.03) and presence of Del11q (HR 0.48; CI 0.28-0.83, p<0.01) predicted for longer TTNT, while Del13q and Trisomy 12 had no significant association. Conclusions We found that 1L treatment with BCL2i resulted in longer TTNT when compared to BTKi, without any significant difference in OS. For pts with TP53D, there was no significant difference in TTNT or OS between BTKi and BCL2i. However, the majority of pts with TP53D received BCL2i >1 year suggesting real-world use of continuous BCL2i rather than time-limited treatment. Prospective studies comparing these two treatment modalities would be necessary to confirm these findings.