Beyond mechanisms: toward patient-centered choices in BTKi-venetoclax fixed-duration therapy for CLL
Year: 2026
Type: article
Abstract: We read with interest the timely perspective recently published in Leukemia by Tedeschi et al. addressing fixed-duration (FD) combination therapies of venetoclax with Bruton tyrosine kinase inhibitors (BTKis) in chronic lymphocytic leukemia (CLL) [ 1 ]. The authors examine the evolving landscape of time-limited regimens and present a mechanistic rationale to guide readers in preferring one BTKi over another in combination with venetoclax [ 1 ]. The authors’ major claims are that the support for ibrutinib–venetoclax (I+V) is based on indirect, cross-trial comparisons across heterogeneous patient populations enrolled in the GLOW, CAPTIVATE, and AMPLIFY studies [ 2 , 3 , 4 ]. Given the relevance of the issue in clinical practice, the authors’ conclusions warrant a more cautious interpretation in the absence of direct evidence before privileging one BTKi partner over another in FD treatment strategies. First, as detailed in a previous analysis, these trials differ significantly in baseline characteristics, including number of patients, median age, fitness status, comorbidity burden, and the prevalence of high-risk genomic abnormalities such as del(17p) and TP53 mutations [ 5 ]. Moreover, these cross-trial comparisons are inherently constrained by differing median follow-up durations (~3.4 years for AMPLIFY, 5.3 years for GLOW, and 5.5 years for CAPTIVATE), which may bias estimates of efficacy [ 2 , 3 , 4 ]. Another critical consideration is the varying applicability of trial findings to broader patient populations. The majority of evidence supporting FD venetoclax-BTKi regimens originates from clinical trials enrolling predominantly fit patients (e.g., CAPTIVATE, AMPLIFY) [ 3 , 4 ]. The seminal CLL17 trial addresses this issue only partially. Specifically, in the I+V arm, the median age was 66 years (IQR 59–73), and the median CIRS score was 3 (IQR 1–5), suggesting limited representation of older or more comorbid individuals [ 6 ]. Caution is warranted when extrapolating the results of these trials to elderly patients or those with substantial comorbidities, who constitute a significant subset of the CLL population. The GLOW trial provides valuable outcome data for older, comorbid patients treated with the I+V combination, demonstrating substantial efficacy [ 2 ]. However, in this study, a higher incidence of severe cardiotoxicity and sudden deaths—presumed to be cardiac in origin—was observed [ 2 ]. These safety concerns ultimately prevented FDA approval of this combination in the United States. Despite the lack of FDA approval, the 2026 NCCN guidelines for the management of CLL include fixed-duration or measurable residual disease (MRD)-guided treatment with I+V as one of the “other therapeutic approaches,” as an alternative to the preferred front-line fixed-duration options: V+obinutuzumab (V+O) and the V+acalabrutinib (V+A) regimen, with or without obinutuzumab [ 7 ]. Second, the use of undetectable MRD (uMRD) as a surrogate endpoint for progression-free survival (PFS) in CLL requires careful, nuanced evaluation. In patients treated with I+V combinations, evidence supports an individual-level surrogate relationship between MRD status at the end of therapy (EoT) and PFS calculated from that landmark, linking deeper remissions to prolonged progression-free intervals. Updated CAPTIVATE data, presented at the 2025 European Hematology Association (EHA) meeting, reinforce this association: uMRD identified a favorable trajectory, with 5.5-year PFS of 71% in the uMRD group versus 47% in the detectable MRD (dMRD) group [ 3 ]. The GLOW study similarly demonstrated an MRD-dependent distinction in early PFS, with 2-year PFS of 93% in the uMRD group versus 79.6% in the dMRD group [ 2 ]. By contrast, the AMPLIFY trial revealed attenuation of the MRD–PFS relationship in the acalabrutinib–venetoclax (A+V) arm in the intention-to-treat analysis (uMRD at EoT, 45%; 3-year PFS, 76.5%), attributable to high treatment discontinuation rates—driven largely by adverse events—and further confounded by the SARS-CoV-2 pandemic [ 4 ]. These findings, potentially limited by shorter follow-up, were replicated in the I+V arm of the CLL17 trial (uMRD at EoT, 47.2%; 3-year PFS, 79.4%) [ 6 ]. A longer follow-up will clarify whether a correlation between MRD measured at EOT and PFS will emerge in these studies. A critical limitation of EoT PFS for validating MRD as a reliable surrogate is that it fails to account for therapy-induced alterations in the MRD–PFS relationship occurring before this landmark, thereby obscuring treatment effects from trial onset [ 8 ]. Trial-level analyses using PFS from therapy initiation more reliably capture regimen-specific MRD–PFS correlations across studies. In our trial-level meta-analysis based on these principles, the magnitude of MRD–PFS correlations showed limited variability across BTKi–venetoclax combinations. Moreover, at the trial level analysis, the attenuation of the MRD–PFS correlation was not observed solely in the AMPLIFY and CLL17 trials; it has also been observed in the CAPTIVATE and GLOW studies [ 2 , 3 , 4 ]. Although the primary focus was on BTKi–venetoclax combinations, we included, for comparative purposes, trials of venetoclax–obinutuzumab (VO) combinations in our analysis. Notably, MRD–PFS correlations with VO were stronger than those observed for BTKi–venetoclax combinations, likely reflecting differential, compartment-specific efficacy between BTK inhibitors and anti-CD20 monoclonal antibodies (Fig. 1 ). These findings challenge the universal applicability of MRD as a prognostic marker across CLL treatments, given treatment-dependent kinetics and inter-compartmental variability. Fig. 1 Data refers to the correlation between undetectable measurable residual disease (uMRD) in peripheral blood assessed at the end of therapy (EoT) and 3-year progression-free survival (PFS) within specific targeted-treatment contexts: venetoclax plus an anti-CD20 antibody (as in CLL13, CLL14, CLL17), BTK inhibitor (BTKi) plus venetoclax with anti-CD20 (as in AMPLIFY, CLL13), and BTKi–venetoclax regimens without anti-CD20 (as in CAPTIVATE, GLOW, AMPLIFY, CLL17). Full size image Cross-trial comparisons of time to next treatment (TTNT) suggest relatively similar outcomes for BTKi–venetoclax combination regimens. Three-year TTNTs were 88.5% in the AV arm of AMPLIFY and 89%, 92%, and 92% in the I+V arms of CAPTIVATE, GLOW, and CLL17, respectively [ 2 , 4 , 6 , 9 ]. Longer-term follow-up will be needed to see if clear differences with respect to safety and efficacy emerge between different fixed-duration regimens. TTNT is a surrogate endpoint used in comparative trials because it can reflect patient-centered outcomes and inform pragmatic clinical decision-making. However, definitions of TTNT vary: some analyses define TTNT as death from any cause or initiation of subsequent therapy, while others censor deaths. Furthermore, TTNT may have limited utility as an efficacy surrogate in trials of continuous BTKi therapy, where intolerance-driven discontinuation is common; its interpretive value is greater for fixed-duration (FD) regimens. Nevertheless, re-treatment decisions are influenced by physician judgment, so TTNT may not precisely measure the efficacy of prior therapy and should be interpreted with caution as a surrogate endpoint [ 2 , 4 , 6 , 9 ]. Third, the mechanistic rationale presented in favor of ibrutinib’s superiority—particularly its effect on “compartment shift” and sensitization of CLL cells to venetoclax-induced apoptosis—merits also a balanced evaluation [ 1 ]. These hypotheses stem largely from preclinical models and retrospective analyses [ 10 , 11 ]. Data from retrospective observational studies highlighting different lymphocytosis values between the two agents are confounded by imbalanced risk factors, including varying frequencies of del(17p), del(11q), and NOTCH1 mutations [ 12 ]. Head-to-head pharmacodynamic and pharmacokinetic studies are needed to establish whether these mechanistic differences materially impact clinical efficacy. What is only implicitly suggested in the Leukemia perspective but warrants more explicit emphasis is the superior target selectivity exhibited by second-generation BTKi like acalabrutinib [ 1 ]. Studies investigating continuous acalabrutinib as a single agent reported a reduced incidence of cardiovascular adverse events compared to ibrutinib in the ELEVATE R/R trial, which employed a hierarchical design [ 13 ]. Consistent with these findings, a post-hoc safety analysis of the AMPLIFY trial indicates an 9.3% incidence of cardiac events and only 0.7% of atrial fibrillation cases with A+V combination [ 12 ]. Notably, in the I+V arm of the CLL17 trial, cardiac disorders were reported in 23.8% of patients, with atrial fibrillation accounting for 12.5% of these events [ 6 ]. Finally, it is important to emphasize that FD BTKi–venetoclax combinations constitute only one therapeutic modality within the broader treatment paradigm for CLL [ 7 ]. Currently, continuous BTKi monotherapy remains the standard of care for a relevant subset of patients with CLL, particularly those harboring TP 53 mutations, or deemed ineligible for venetoclax due to frailty, logistical constraints, lack of caregiver support, severely impaired renal function and or increased risk for tumor lysis syndrome [ 7 ]. Numerous effective frontline V+BTKi regimens are now available for CLL, and the therapeutic landscape continues to expand with combinations of venetoclax and other BTKis, including covalent agents such as zanubrutinib and non-covalent inhibitors such as pirtobrutinib [ 14 , 15 ]. Longer-term follow-up will be needed to determine whether clear differences in safety and efficacy emerge among the different V+BTKi options. Moreover, selecting the optimal BTKi partner for V in fixed-duration regimens should require prospective, head-to-head clinical trials. However, given the conceptual and practical constraints, direct head-to-head comparisons between V+I and V+A with or without obinutuzumab are unlikely. Real-world data from unselected patient populations are likely to better inform practice. In the interim, optimal BTKi for venetoclax should be guided by a careful evaluation of treatment-safety profiles, the characteristics of CLL, the burden of comorbidities, and patient preferences.
Source: Leukemia
Language: English
Subfield: Genetics
Field: Medicine
Domain: Health Sciences
Open Access status: hybrid
Estimated APC paid: $5,190
OpenAlex ID: W7140185972