Prevalence of BTK and PLCG2 Mutations in CLL Patients With Disease Progression on BTK Inhibitor Therapy: A Meta‐Analysis

Abstract

Bruton's tyrosine kinase inhibitors (BTKis) have revolutionized the treatment of chronic lymphocytic leukemia (CLL), with significantly improved outcomes for both treatment-naïve (TN) and relapsed/refractory (R/R) patients. BTKis bind irreversibly to the cysteine 481 (C481) residue of the BTK molecule inhibiting B-cell receptor (BCR)-mediated intracellular signals crucial for CLL-cell survival [1]. Despite its efficacy, long-term BTKi therapy often leads to therapy resistance with subsequent disease progression (DP) [2]. Resistance mechanisms predominantly relate to mutations in BTK gene, particularly the mutation at C481S, which disrupts covalent BTKi binding. Additionally, mutations in phospholipase Cγ2 (PLCG2), which encodes for a downstream effector of BCR signaling, are emerging as significant additional contributors to resistance [3].

To comprehensively assess the prevalence and significance of these resistance mechanisms, we conducted a systematic review and meta-analysis to quantify the prevalence of BTK and PLCG2 mutations in CLL patients who experience DP while treated with BTKis. We performed a comprehensive search of the PubMed database and manually reviewed abstracts from major hematology conferences (American Society Hematology [ASH] and European Hematology Association [EHA]) to identify relevant studies. The analysis adhered to Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines to ensure methodological rigor and transparency [4].

Studies were included if they reported upon CLL patients treated with covalent BTKis and reported assessments for BTK and/or PLCG2 mutations at DP. Two reviewers (SM and DG) independently performed data extraction, with discrepancies resolved by consensus. The primary outcome was the pooled prevalence of BTK and PLCG2 mutations amongst patients with DP while treated with BTKis. A separate analysis compared patients treated with first-generation BTKi (ibrutinib) versus second-generation BTKis (acalabrutinib, zanubrutinib). Cross study heterogeneity was assessed using the chi-squared (χ²) Q test and the I² statistic, with an I² value greater than 50% indicating substantial heterogeneity.

Seventeen studies with 724 patients provided data on BTK somatic mutations [2, 3] (Figure S1; Table S1; Supporting Information: References [1–11]). These studies included six post hoc analyses of Phase 3 clinical trials (ALPINE, ELEVATE R/R, RESONATE, RESONATE-2, ILLUMINATE, FLAIR), five Phase 2 trials (PCYC-1122, RESONATE-17, NCT02337829, NCT01500733, NCT03740529 [BRUIN]), and three retrospective multicenter analyses (French Innovative Leukemia Organization [FILO], European Research Initiative on CLL [ERIC], Hungarian Ibrutinib Resistance Analysis Initiative). Additionally, three retrospective monocentric studies from the MD Anderson Cancer Center (MDACC), Peter MacCallum Cancer Centre, and The Ohio State University (OSU) Comprehensive Cancer Center were included [2, 3] (Supporting Information: References [1–11]). The majority of patients (86.4%) received ibrutinib-based therapies (ibrutinib monotherapy, 78%; ibrutinib combined with rituximab [IR], 8.4%), while a smaller subset (13.5%) received second-generation BTKis (acalabrutinib, 8.4%; zanubrutinib, 5.1%).

Overall, 7.8% of patients discontinued BTKi therapy due to adverse events (AEs) or other reasons. Of those who discontinued due to DP (n = 667), 86.3% had R/R CLL, and 13.6% had TN CLL. The meta-analysis demonstrated that, amongst patients with progressive CLL, pooled prevalence of BTK mutations was 52% (95% CI: 39%–64%), but with substantial heterogeneity across studies (Q = 161.54, p < 0.001, I² = 91%) (Figure 1A).

FIGURE 1

Open in figure viewerPowerPoint

Meta-analysis on the prevalence of Bruton tyrosine kinase (BTK) (A) and phospholipase Cγ2 (PLCG2) (B) somatic mutations amongst patients with disease progression (DP) while treated with BTK inhibitor therapy.

Fourteen studies, which encompassed 620 patients, were analyzed for PLCG2 mutations [2, 3] (Figure S2; Table S1) (Supporting Information: References [1–5, 9–11]). These cohorts were the same as those analyzed for BTK mutations, except for the noninclusion of the patients from the MDACC, Peter MacCallum Cancer Centre, and the Hungarian Ibrutinib Resistance Analysis Initiative, which did not provide data on PLCG2 mutations. Most patients in the PLCG2 cohort were treated with ibrutinib-based therapies (ibrutinib-monotherapy, 80.3%; IR, 9.8%), while a smaller proportion (13.7%) received second-generation BTKis (acalabrutinib, 9.8%; zanubrutinib, 3.8%). Discontinuation of BTKi therapy due to AEs or other reasons was observed in 9.2% of patients. Amongst patients with progressive CLL (n = 563), 86% had R/R CLL, and 14% had TN CLL. The pooled prevalence of PLCG2 mutations was 11% (95% CI: 7%–17%), with notable heterogeneity across studies (Q = 40.77, p < 0.001, I² = 73%) (Figure 1B).

We further explored whether the prevalence of BTK or PLCG2 mutations differed between patients treated with the first-generation BTKi ibrutinib and those treated with the second-generation inhibitors acalabrutinib or zanubrutinib. The prevalence of BTK mutations was 56% (95% CI: 38%–74%) amongst ibrutinib-treated patients and 51% (95% CI: 26%–77%) amongst those treated with acalabrutinib or zanubrutinib (Figure S3). PLCG2 mutations were detected in 13% (95% CI: 6%–23%) of ibrutinib-treated patients, compared with 9% (95% CI: 0%–25%) in patients exposed to acalabrutinib or zanubrutinib (Figure S4).

Finally, we investigated whether pre-existing TP53 mutations (median 49.5%; range 22.9%–100%) or the duration of BTKi therapy (median 40.5 months; range 27.7–78 months) correlated with the development of BTK and PLCG2 mutations.

A positive correlation was found between the presence of PLCG2 mutations and the TP53 mutational burden (r² = 0.872, p = 0.001), as well as the duration of BTKi therapy (r² = 0.539, p = 0.02) (Figures S5 and S6). However, no significant correlation was observed between these factors and the presence of BTK mutations.

Our meta-analysis demonstrates that resistance to BTKis in CLL is multifaceted, with BTK mutations prevalent in over half of patients who experience DP. While PLCG2 mutations are less common, their association with a high TP53 mutation burden and prolonged BTKi exposure underscores their emergent role in secondary resistance mechanisms. These results may suggest that fixed-duration therapies could mitigate these resistance pathways, with potential improvement in long-term outcomes for patients [5, 6].

Given the limited representation of TN CLL patients in our dataset, these findings primarily pertain to patients with R/R CLL. Notably, most R/R patients were likely exposed to prior chemotherapy. In our aggregate analysis, however, we could not differentiate between R/R patients with and without prior chemotherapy exposure. Therefore, the specific contribution of prior chemotherapy to BTKi resistance mediated by BTK or PLCG2 mutations could not be assessed in detail.

Much of our knowledge in relation to the prevalence of BTK or PLCG2 mutations is derived from studies of ibrutinib-treated patients [2, 3] (Supporting Information: References [1–11]). A recent post hoc analysis of the ELEVATE-R/R trial, which compared acalabrutinib with ibrutinib in R/R high-risk CLL patients, revealed that the rate of emergent BTK mutations at relapse was significantly lower in those treated with ibrutinib as compared to acalabrutinib (37% vs. 69%) (Supporting Information: Reference [2]). However, due to the limited sample size associated with this analysis, the clinical significance of these findings remains uncertain. Additionally, the presence of higher risk genomic features amongst patients in the ELEVATE-RR trial, which contributes to genetic instability, limits the generalizability of these results (Supporting Information: Reference [2]). In our meta-analysis of trials involving ibrutinib and second-generation BTKis, we observed similar rates of BTK and PLCG2 mutations amongst patients with progressive disease, regardless of the specific BTKi used.

An important consideration in the transition from ibrutinib to second-generation BTKis is the emergence of additional BTK mutations that mediate resistance. In a post hoc analysis of the ELEVATE-R/R trial, T474I gatekeeper mutations were observed in 29% of patients exhibiting resistance to acalabrutinib (Supporting Information: Reference [2]). In a separate study examining patients treated with zanubrutinib, L528W mutations, known to impair kinase function, were detected at rates comparable to those of the C481S mutation (Supporting Information: Reference [3]). Our analysis confirms that the canonical C481S mutation remains the predominant mutation, accounting for approximately 60%–100% of BTK mutations in patients treated with ibrutinib or second-generation BTK inhibitors (Table S2).

Finally, it is important to acknowledge the limitations of our current meta-analysis. While our current study points to somatic genetic mutations as the primary drivers of resistance, one-third of the patients in our analysis lacked either detectable BTK or PLCG2 mutations. In these patients, resistance to BTKis may develop through nongenetic adaptive mechanisms that activate compensatory pro-survival pathways. Specifically, the activation of PI3K/AKT/ mTOR, NF-κB, and MAPK pathways, along with the upregulation of BCL2, MYC, and XPO1 or downregulation of PTEN, may enable B cell survival even in the presence of BTK inhibition [7]. Additionally, resistance may be further supported by microenvironmental factors, including chemokine and integrin signaling via upregulation of CXCR4 and VLA4 [7]. Such mechanisms may also be pathogenic mediators of BTKi therapy resistance, concepts, which challenge the predominant focus on genetic factors in this field of study.

Other limitations of this study include the inconsistent reporting of variant allele frequencies (VAFs), the reliance on peripheral blood samples that may not capture clonal evolution in lymph nodes and the lack of longitudinal data mapping mutation dynamics over time [8].

In summary, this meta-analysis provides a comprehensive insight into the prevalence of BTK and PLCG2 somatic mutations in CLL, which has lost sensitivity to BTKi therapy. Our report may lead to a further appreciation of genetic mechanisms of resistance to BTKis in CLL treatment. Ultimately, the heterogeneity of resistance mechanisms necessitate further research also into nongenetic factors driving DP.