CRISTALLO: Results from a Phase III Trial of Venetoclax–Obinutuzumab Versus Fludarabine, Cyclophosphamide and Rituximab or Bendamustine–Rituximab in Patients with Untreated Chronic Lymphocytic Leukemia without Del(17p) or TP53 Mutations

Background:

Time-limited therapies, such as venetoclax (Ven)–obinutuzumab (O), fludarabine–cyclophosphamide–rituximab (FCR), and bendamustine–rituximab (BR), have shown significant rates of undetectable (u) minimal residual disease (MRD) in patients (pts) with chronic lymphocytic leukemia (CLL). MRD status at end of treatment (tx) strongly correlates with progression-free survival (PFS) and overall survival and is often used in trials as an early indicator of clinical outcome; however, it has not been widely used as a standalone primary efficacy endpoint. We present, for the first time, primary analysis results from the CRISTALLO Phase III trial (NCT04285567), performed in fit pts with tx-naïve CLL, with MRD as the sole primary endpoint.

Methods:

CRISTALLO is an international, open-label, randomized trial, comparing VenO vs FCR/BR. Pts aged ≥18 years with a cumulative illness rating scale score ≤6, creatinine clearance ≥70mL/min, without del(17p) or TP53 mutations, and in need of tx were randomized 1:1 to 6 cycles of VenO plus 6 cycles of Ven or 6 cycles of FCR/BR (randomization stratified by Binet stage, immunoglobulin heavy-chain variable region gene [IGHV] mutational status, and age). Oral daily Ven was initiated with a 5-week ramp-up period on Cycle (C) 1 Day (D) 22; 400mg was administered from C3D1 onwards. Intravenous O was administered for 6 cycles on C1D1/2, 8, and 15, and on D1 of remaining cycles.

Primary endpoint was uMRD rate (<1 CLL cell in 10,000 leukocytes, <10^-4) in peripheral blood (PB) using next-generation sequencing at Month 15 (intent-to-treat population). Key secondary endpoints included uMRD rates in PB and bone marrow (BM) at end of tx (EOT), complete response (CR) rate, and PFS. All p-values for secondary endpoints (except PFS) were descriptive. Exploratory endpoints included uMRD rates at deeper cutoffs. Data cutoff was when all pts reached Month 15.

Results:

Pts were enrolled from May 28, 2020 to Jan 23, 2023. At data cutoff (Mar 19, 2024), 166 pts were randomized to VenO (n=80) or FCR/BR (n=86). Baseline demographics and clinical characteristics were generally balanced (VenO vs FCR/BR): median age was 62 vs 61 years, 33.8% vs 33.7% of pts were aged >65 years, 68.8% vs 66.3% were male, and 56.3% vs 55.8% had mutated IGHV. Overall, 52.4% of pts had Binet Stage B disease, followed by Stage C (34.4%), and Stage A (13.3%). Median duration of follow-up was 32 months; 83.8% completed 12 cycles of Ven tx and 75.6% completed FCR/BR tx.

The primary endpoint was met: 81.3% vs 54.7% of pts treated with VenO vs FCR/BR achieved uMRD in PB at Month 15 (difference: 26.6, 95% CI: 12.3–40.9; p=0.0004); these results were consistent across baseline demographics and clinical characteristics subgroups. uMRD rates in PB at EOT were higher with VenO (81.3%) vs FCR/BR (60.5%) (difference: 20.8, 95% CI: 6.7–34.9; p=0.0053). Similarly, 70.0% vs 38.4% achieved uMRD in BM at EOT (difference: 31.6, 95% CI: 16.6–46.7; p<0.0001). CR rates in pts treated with VenO and FCR/BR were 47.5% and 30.2%, respectively. At Month 15, 63.8% (VenO) and 27.9% (FCR/BR) achieved PB uMRD at a <10^-6 cutoff.

At data cutoff, PFS was immature; fewer pts had progressed or died with VenO vs FCR/BR (hazard ratio: 0.49, 95% CI: 0.2–1.3; p=0.13), and 2-year PFS rates were 95.7% and 90.4%, respectively.

Tx-related adverse events (AEs) occurred in 97.4% and 89.4% of pts treated with VenO and FCR/BR, respectively; related AEs leading to tx discontinuation were lower with VenO (6.5%) vs FCR/BR (15.3%). Rate of serious AEs was higher with VenO (61.0%) vs FCR/BR (49.4%); those leading to tx withdrawal were lower with VenO (7.8%) vs FCR/BR (11.8%). Common (≥10%) AEs in the VenO vs FCR/BR arms included infusion-related reaction (70.1% vs 42.4%), COVID-19 infection (45.5% vs 35.3%), thrombocytopenia (39.0% vs 21.2%), diarrhea (35.1% vs 12.9%), and nausea (20.8% vs 40.0%). Six pts died (3 in each arm); these were considered unrelated to tx. Laboratory tumor lysis syndrome (TLS) occurred in 10.4% vs 2.4% of pts treated with VenO vs FCR/BR, respectively; in the VenO arm, all TLS events occurred during O debulking and before Ven initiation.

Conclusions:

At the primary analysis, VenO demonstrated superiority over FCR/BR in driving deep molecular remissions; uMRD rate at Month 15 in PB was significantly higher in pts treated with VenO vs FCR/BR. There was a numerical but non-significant PFS benefit at this immature timepoint. No new safety signals were identified with VenO.