Acalabrutinib ± Obinutuzumab Vs Obinutuzumab + Chlorambucil in Treatment-Naive Chronic Lymphocytic Leukemia: 6-Year Follow-up of Elevate-TN

Background: Previous reports of ELEVATE-TN (NCT02475681) at median follow-up of up to 58.2 months (mo) demonstrated superior efficacy of acalabrutinib (A) ± obinutuzumab (O) compared with O + chlorambucil (Clb) in patients (pts) with treatment-naive (TN) chronic lymphocytic leukemia (CLL). Herein, updated results at 74.5 mo of follow-up are reported.

Methods: Pts received A monotherapy, A+O, or O+Clb. Pts who progressed on O+Clb could cross over to A monotherapy. Investigator (INV)-assessed progression-free survival (PFS), INV-assessed overall response rate (ORR), overall survival (OS), and safety were evaluated. All analyses are ad-hoc and P-values are descriptive.

Results: A total of 535 pts (A, n=179; A+O, n=179; O+Clb, n=177) were randomized: median age was 70 years, 63% had unmutated immunoglobulin heavy chain variable region genes (uIGHV), and 14% had del(17p) and/or TP53 mutation (TP53m). At a median follow-up of 74.5 mo (range, 0.0–89.0; data cutoff March 3, 2023), median PFS was not reached (NR) for A+O and A vs 27.8 mo for O+Clb (hazard ratio [HR] vs O+Clb: 0.14 and 0.23, respectively; P<0.0001 for both; HR A+O vs A: 0.58; P=0.0229); estimated 72-mo PFS rates were 78%, 62%, and 17%, respectively (Figure 1). For the 79 pts who crossed over from O+Clb to A, median PFS2 (time to second disease progression or death) was NR; estimated 72-mo PFS2 rate was 54%. Median OS was NR in any treatment arm and was significantly longer with A+O vs O+Clb (HR: 0.62; P=0.0349); estimated 72-mo OS rates were 84% for A+O, 76% for A, and 75% for O+Clb (Figure 2). In 337 pts with uIGHV, median PFS was NR in both the A+O and A arms vs 22.2 mo for O+Clb (HR: 0.08 and 0.12, respectively; P<0.0001 for both), and estimated 72-mo PFS rates were 75%, 60%, and 5%, respectively, with median OS NR in all treatment arms and estimated 72-mo OS rates of 84%, 76%, and 74%, respectively. In 73 pts with del(17p) and/or TP53m, median PFS was 73.1 mo for A+O and NR for A vs 17.5 mo for O+Clb (HR: 0.28 and 0.23, respectively; P≤0.0009 for both), and estimated 72-mo PFS rates were 56%, 56%, and 18%, respectively. Median OS was NR in both the A+O and A arms vs 74.9 mo for O+Clb (HR: 0.53 and 0.46, respectively, neither statistically significant) and estimated 72-mo OS rates were 68%, 72%, and 53%, respectively. ORR was significantly higher with A+O (96%; 95% confidence interval [CI]: 92–98; P<0.0001) and A (90%; 95% CI: 85–94; P=0.0499) vs O+Clb (83%; 95% CI: 77–88). Combined complete response (CR) plus CR with incomplete hematologic recovery rates were higher with A+O (37%) and A (19%) vs O+Clb (14%; P≤0.0499 for both).

Median treatment exposure was 74.4 mo and 72.0 mo for A in the A+O and A arms, respectively; 5.5 mo and 5.6 mo for O in the A+O and O+Clb arms, respectively; and 5.5 mo for Clb in O+Clb arm. Adverse events (AEs) for the O+Clb arm have been previously reported. The most common (≥5% of pts) grade ≥3 AEs for A+O and A, respectively, were neutropenia (31% and 12%), thrombocytopenia (8% and 3%), diarrhea (6% and 1%), COVID-19 (9% and 7%), pneumonia (7% and 6%), syncope (5% and 2%), and hypertension (4% and 5%). For events of clinical interest, grade ≥3 atrial fibrillation, hypertension, and secondary primary malignances were reported in 2%, 4%, and 10% of pts treated with A+O and 2%, 5%, and 5% of pts treated with A, respectively. Treatment is ongoing in 54% (A+O; n=96) and 47% (A; n=84) of pts; the most common reasons for treatment discontinuation were AEs, observed in 21% (n=38) of pts treated with A+O and 18% (n=32) of pts treated with A, and progressive disease, observed in 6% (n=10) of pts treated with A+O and 14% (n=25) of pts treated with A. For pts who crossed over from O+Clb to A, 41% (n=32) discontinued A due to AEs in 13% (n=10) of pts and progressive disease in 16% (n=13) of pts.

Conclusions: With a median follow-up of 74.5 mo, the efficacy and safety of A+O and A monotherapy were maintained in pts with TN CLL, including in pts with high-risk genetic features. At 6 years of follow-up, PFS was significantly longer in pts treated with A+O vs A. Median OS was NR in any treatment arm and was significantly longer in pts treated with A+O vs O+Clb.