Résumé
Introduction
Venetoclax (Ven) is a highly selective oral inhibitor of BCL2, a key regulator of the intrinsic apoptotic pathway, which is constitutively overexpressed in CLL. Efficacy and safety of VenR given for a fixed duration in R/R CLL was evaluated in the randomized Phase III MURANO study compared with standard bendamustine + rituximab (BR). A first pre-planned analysis, when the majority of patients (pts) were still on study treatment (median follow-up 23.8 mo), established a superior PFS of VenR over BR (Seymour et al. NEJM 2018). With all pts having completed therapy, we analysed long-term outcome with a median follow up of 36.0 mo.
Methods
Pts were randomized to receive 6 cycles of VenR followed by Ven 400mg once daily for a total of 2 yrs, or 6 cycles of BR. Disease status was assessed by CT scan at screening, Cycle (C) 4 and 2−3 mo after end of combination therapy (EOCT). During Ven single agent and at follow-up visits, response was determined by clinical assessment every 3 mo until 3 yrs of follow-up, then every 6 mo thereafter or until PD. PFS status was based on investigator assessment. Peripheral blood (PB) minimal residual disease (MRD) was analysed centrally by ASO-PCR and/or flow cytometry at C4, EOCT and every 3/6 mo thereafter. All p values are descriptive.
Results
In total, 389 pts were enrolled in VenR (n=194) or BR (n=195) arms. As of May 8 2018, all pts were off treatment. For BR, 154 (79%) had completed 6 cycles. In the VenR arm, 174 (90%) completed the VenR combination phase and 130 (67%) had completed 2 yrs of Ven. The remainder had PD (11%), died (1%), or withdrew due to AEs (15%) or other reason (6%). Median Ven exposure duration and relative dose intensity were 24.4 (range 0-27.9) mo and 97.4% (26-100%), respectively, overall and 17.7 (0.5-21.9) mo and 99% during Ven single-agent therapy.
With a median time off Ven after 2 yrs treatment of 9.9 (1.4-22.5) mo, PFS with VenR remains superior to BR (HR, 0.16 [95% CI 0.12, 0.23]; p<0.0001; median not reached vs 17.0 mo; Figure 1). 3-year PFS estimate was 71.4% (95% CI 64.8%, 78.1%) vs 15.2% (95% CI 9.1%, 21.4%) respectively. Consistent magnitude of treatment effect on PFS with VenR was observed in all clinical and biological subgroups (Figure 2). For pts who completed 2 yrs of Ven (n=130), 6 and 12 mo PFS estimates were 92% (95% CI 87.3, 96.8) and 87% (95% CI 81.1, 93.8), respectively (Figure 3). 16/130 pts developed PD after completion of Ven; of these, 14 were MRD-positive at >1% in PB at Mo 24 (when Ven single agent was scheduled to cease) and 10/16 pts had del(17p)/TP53 mutation at baseline. Clinical and cytogenetic risk factors in pts with and without PD after Mo 24 are in Table 1. In this analysis, OS improvement was seen with VenR over BR (HR, 0.50 [95% CI 0.30, 0.85]; p=0.0093; 3-yr rate: 87.9% vs 79.5%; Figure 4).
Subsequent CLL-directed treatment was given after PD in 91 pts in the BR arm. 71/91 (78%) BR arm pts received novel targeted agents, including 45 who had ibrutinib and 7 who had Ven. 27/194 (14%) pts in the VenR arm received subsequent therapy: 13/27 (44%) had novel targeted agents as next treatment, including 8 pts who had ibrutinib and 3 who were re-treated with Ven.
See Table 2 for a safety overview for VenR combination and Ven single-agent periods. During Ven single agent: 17/171 pts (10%) had an AE leading to drug withdrawal; 7/171 pts (4%) had an AE leading to dose reduction; 44/171 pts (26%) had a Ven dose interruption due to an AE; 7/171 pts (4%) had a fatal AE (4 other cancers, 2 cardiac, 1 pneumonia). Grade 3-4 AEs occurred in 59/171 pts (35%); the most frequent were neutropenia (20 pts, 12%), anemia (5 pts, 3%), and thrombocytopenia (3 pts, 2%). 12/171 (7%) pts had a grade 3-4 infection in the Ven single-agent phase. The total number of Richter transformation events was 7 with VenR and 6 with BR.
Conclusions
With all pts off treatment and 3 yrs' median follow-up, continued substantial benefit was observed with VenR, with PFS and OS superior to BR. There were no new safety signals; most pts were able to complete treatment. The rate of CLL progression in the first 12 mo after Ven completion was modest (13%), supporting the feasibility and safety of a time-limited VenR duration. The protocol has been amended to include assessment of response and durability of disease control with VenR reintroduction at PD. This updated analysis of this Phase III global randomized study demonstrates clinically meaningful benefit of the VenR chemotherapy-free regimen with a fixed duration in all pts with R/R CLL.