Time-limited acalabrutinib monotherapy in frail patients with previously untreated chronic lymphocytic leukemia: Primary endpoint analysis of the randomized STAIR trial.

Abstract Background Continuous therapy with covalent BTK inhibitors (BTKi) has become a cornerstone strategy for patients with chronic lymphocytic leukemia (CLL). However, unlimited administration of BTKi has raised concerns, including risk of clonal evolutionwith therapeutic resistance and accumulation of adverse events, which can be particularly detrimental in elderly or frail patients. We investigated an 18-month limited-duration therapy with the BTKi acalabrutinib (ACA) in frail patients with previously untreated CLL. Methods The STAIR study is a French investigator-sponsored (FILO) multicenter, randomized phase 2 trial evaluating the impact of ACA discontinuation on progression-free survival (PFS) in elderly and frail patients with untreated CLL (NCT04963946). Key inclusion criteria were : age >70 years, previously untreated CLL requiring treatment (iwCLL2018 criteria), CIRS score > 6 or impaired creatinine clearance (31-69 mL/min, Cockroft). Patients received continuous ACA at 100 mg BID for 18 months before being randomized (1:2) to the control arm A (continuing ACA) or the experimental arm B (discontinuing ACA), with stratification based on complex karyotype (> or = 3 abnormalities) and TP53 alterations. Patients in arm B who experienced CLL progression restarted ACA per protocol in case of relapse or progression with symptomatic disease (reviewed by an independent board). The primary endpoint was the PFS after randomization. The strategy was considered insufficiently promising if 1-year PFS rate was lower than 75% in the experimental arm. Results Among the 172 patients screened for eligibility between October 2021 and June 2023 across 29 centers in France, 160 were enrolled. One patient did not receive ACA and 38 (24%) discontinued ACA within the first 18 months and were not subsequently randomized. Reasons for discontinuation were toxicity (13 patients), disease progression as CLL (7 patients) or Richter transformation (2 patients), secondary primary malignancies (5 patients), deaths (4 patients) and patients' or physicians' decision (4 and 3, respectively). A total of 121 patients were randomized to continuing ACA (arm A, n=41) or discontinuing ACA (arm B, n=80). At randomization, median age was 77 years (range, 71-98), 57.9% were male, 57.9% had CIRS>6, and 79.3% had clearance <70mL/min. Regarding genetic features, 12.4% had TP53 mutations, 11.9% had del(17p) and 27% harbored complex karyotype. The majority of patients had unmutated IGHV (65.0%). At the date of data-cut off (June 1st, 2025), the median follow-up time from randomization was 14.2 months (95%CI, 11.3-16.7). In the control arm A (n=41), 7 patients experienced permanent ACA interruption, for toxicity in 6 cases. At data cut-off, rates of PFS and OS at 1 year after randomization were 96.3% (CI 76.5-99.5) and 100%, respectively. In the experimental arm B (n=80), 39 PFS events were observed, including 37 CLL progressions and 2 deaths. The rate of PFS at 1 year after randomization was 53.1% (CI 40.2-64.5) (one-sample logrank P>0.99). Among subgroups, patients with mutated IGHV had significantly higher 1-year PFS rate (90.4% (CI 65.9-97.6)) than those with unmutated IGHV (34.0% (CI 19.7-48.8)) (P<0.001). Of the patients who progressed after ACA discontinuation, 25 (67.6%) presented symptomatic disease requiring therapy per iwCLL2018 criteria. The rate of time-to-next-treatment at 1 year after randomization was 74.0% (CI 61.7-82.9). Twenty-four restarted ACA per protocol and 1 received venetoclax. Fourteen (77.8%) patients responded to ACA, 4 (22.2%) were stable and 7 were non evaluated at data cut-off. The rate of OS at 1 year after randomization was 96.8% (CI 87.9-99.2). In the overall population (n=160), the most common grade 3-4 treatment-emergent adverse event (AE) were haematologic : neutropenia (12%) and thrombocytopenia (3%). Atrial fibrillation (any grade) occured in 11 (6.9%) patients and grade 3-4 infections in 12 (7.5%). A total of 35 patients presented second primary malignancies including 23 non-melanoma skin cancers. After randomization, grade 3-4 AEs occured in 10 (24.4%) patients in arm A and 8 (13.8%) patients in arm B. Conclusion Limiting time of ACA therapy to 18 months results in a significant decreased rate of PFS at 1 year, particularly among patients with unmutated IGHV. Longer follow-up is warranted to investigate the further impact of ACA interruption on duration of response, AE burden and OS and to assess the response to ACA reinitiation.