Page 19 - Haematologica-April 2018

Editorials
Bendamustine plus rituximab in chronic lymphocytic leukemia: is there life in the old dog yet?
Clemens-Martin Wendtner
University of Cologne and Klinikum Schwabing, Academic Teaching Hospital of the University of Munich, Germany
E-mail: clemens.wendtner@uni-koeln.de doi:10.3324/haematol.2018.188862
The results of the multicenter international MABLE trial published in this issue of Haematologica repre- sent the first randomized phase III data comparing two frequently used chemoimmunotherapies, chlorambu- cil plus rituximab (Clb-R) versus bendamustine plus ritux- imab (BR), in patients with chronic lymphocytic leukemia (CLL) and concomitant comorbidities.1 Michallet and col- leagues report a significant benefit for BR with respect to complete response (CR) rate, progression-free survival (PFS) and minimal residual disease (MRD) negativity rate, while the safety profiles of both chemoimmunotherapies were quite similar. A dominant inclusion criterion was that patients had to be ineligible for treatment with fludarabine due to comorbidities. Altogether, 357 patients were ran- domized while 241 of them were treatment-naive. In this frontline population the primary endpoint of the trial was met, i.e., higher CR rates after 6 cycles of treatment in favor of BR (24% vs. 9%, respectively). As secondary end- points, PFS (30 months vs. 20 months, respectively) and the MRD negativity rate (66% vs. 36%, respectively) were also significantly superior in the BR arm, whereas overall response rate (ORR) and overall survival (OS) did not differ between arms.
While the differences in efficacy endpoints between both treatment arms are quite impressive, some critical points need further reflection. The entire MABLE trial contains a quite heterogeneous and not well-defined group of CLL patients, i.e., first-line and second-line patients, while the aforementioned analysis carried out by Michallet et al. focuses only on the subgroup of treat- ment-naive CLL patients. The recruitment of second-line patients had been stopped by an amendment of the trial, caused by slow accrual. Furthermore, comorbidity is not well-characterized as an inclusion criterion for this trial, and seems rather subjective when given the discriminator “fludarabine-ineligibility”. It would have been desirable if a comorbidity scoring, based, for example, on the cumu- lative illness rating scale (CIRS), as has been used in sim- ilar trials (e.g., COMPLEMENT-1, CLL11, etc.), had also been applied for the MABLE trial population.2,3 In addi- tion, the dose of bendamustine that was chosen for first- line use (90mg/m2) attests to a reasonably fit patient pop- ulation (70 mg/m2 is the typical standard for unfit patients) and makes a comparison with the COMPLE- MENT-1 trial (Clb vs. Clb plus ofatumumab) or the CLL11 trial (Clb vs. Clb-R vs. Clb plus obinutuzumab [Obi]) difficult. In the initially reported first-line data for BR at a dose of 90 mg/m2 based on a phase II trial (CLL2M trial) of the German CLL Study Group (GCLLSG), only a minority of patients was comorbid and/or over 70 years of age.4 In a phase III trial comparing BR against fludarabine, cyclophosphamide, and ritux- imab (FCR) in fitter patients, it was shown that the 90
mg/m2 bendamustine dose was quite toxic in patients
over the age of 65, inducing severe infections of grade III
5
and IV in more than 20% of patients. Therefore, an inter-
national consensus panel recommended a lower dose of bendamustine (70 mg/m2) in elderly patients.6 Excepting the question of the adequately-dosed chemotherapy backbone, there is a need to discuss whether the anti- CD20 monoclonal antibody (mAb) rituximab should still be the standard-of-care for unfit CLL patients nowadays. Within the CLL11 trial of the GCLLSG a combination of chlorambucil plus the type II mAb obinutuzumab has been shown to be superior to the doublet of Clb-R, at least with respect to PFS.
Besides the problematic comparison of MABLE data to other chemoimmunotherapy trials focusing on a less fit CLL population, there must be a critical discussion as to whether the question regarding which chemoim- munotherapy is superior as a frontline approach in CLL patients is still relevant nowadays, given the fact that many other therapeutic options have become available over the last few years. We have learned from the RES- ONATE-II trial that the Bruton’s tyrosine kinase (BTK) inhibitor, ibrutinib, is very effective as a first-line therapy in CLL patients, regardless of the risk factors, e.g., an unmutated IGHV status.7 Although a direct comparison of ibrutinib monotherapy to a chemoimmunotherapy standard, such as Clb-R or BR, is lacking thus far, it is dif- ficult to imagine that one of the main players of the MABLE trial could beat ibrutinib, which induces durable responses in the frontline setting (median PFS not reached after a median observation time of 18.4 months). However, data from the ILLUMINATE trial performed by the UK CLL Study Group, which is comparing chloram- bucil plus obinutuzumab versus ibrutinib plus obinu- tuzumab, are pending. In addition, the U.S. based trial A041202 (Alliance) will answer the question of whether BR still has a role compared to ibrutinib or ibrutinib plus
Table 1. Efficacy of different first-line treatment options in less fit/elderly CLL patients.
Clb-R (MABLE)1
BR(MABLE)1 Clb-Obi(CLL11)3
Ibrutinib
(RESONATE-II)7 Venetoclax-Obi
(CLL14)8
Median age (yrs)
72
72 74
73 75
ORR CR MRD negativity PFS (%) (%) (%) (months)
75 9 13 30
74 24 41 40 77 22 38 27
86 4 0 NR 100 58 92 NR
ORR: overall response rate; CR: complete response; MRD: minimal residual disease; PFS: pro- gression-free survival; Clb: chlorambucil; R: rituximab; BR: bendamustine plus rituximab; Obi: obinutuzumab; NR: not reached.
haematologica | 2018; 103(4)
563