An Assessment of Kinase Selectivity, Enzyme Inhibition Kinetics and in Vitro Activity for Several Bruton Tyrosine Kinase (BTK) Inhibitors

Abstract

Inhibitors of the Bruton’s tyrosine kinase (BTK) are of broad utility in the treatment of multiple diseases including several B-cell malignancies via effective blockade of oncogenic B-cell receptor (BCR) signaling. BTK is a cytoplasmic tyrosine kinase which harbors a targetable cysteine residue (Cys481) and the majority of BTK inhibitors are covalent modifiers directed at this position. Despite possessing a common mechanism of action, BTK inhibitors differ in key attributes including off-target kinome profiles, tolerability, pharmacokinetics and the underlying BTK inhibition kinetics. These characteristics play a significant role in the ultimate utility of these drugs. Herein, we compare several clinically active BTK inhibitors in biochemical and in vitro assays to gain a broader appreciation of the similarities and differences that govern the success of this important drug class. The combined datasets highlight that each agent has excellent on-target potency and good BTK selectivity. The data further suggests an association between optimized BTK inhibition kinetics and in vitro cytotoxicity profiles.

Bruton’s tyrosine kinase (BTK) has emerged as an important drug target for multiple indications ranging from diseases associated with chronic inflammation to broad use in B-cell malignancies. − While BTK is expressed in multiple cell types (i.e., T cells, macrophages), the well-established pathophysiology of BTK signaling in B-cells has generated widespread interest in BTK inhibitors as targeted therapies for hematological cancers including chronic lymphocytic leukemia (CLL), diffuse large B-cell lymphoma (DLBCL), mantle cell lymphoma (MCL), Waldenstrom macroglobulinemia/lymphoplasmacytic lymphoma (WM/LPL), marginal zone lymphoma (MZL) and follicular lymphoma (FL). ,, BTK plays important roles in B-cell development from maintenance of naïve and transitional B-cells to terminal differentiation of memory and regulatory B-cells. B-cell receptor (BCR) signaling is a hallmark of many B-cell malignancies that can trigger BTK phosphorylation and activation of downstream signaling cascades that stimulate survival and proliferation. , Functional genomic screens revealed core dependencies on BTK and downstream signaling effectors of the NF-ΚB pathway in cell models of DLBCL. ,, Soon after, the first-in-class BTK inhibitor ibrutinib was clinically tested in CLL with encouraging outcomes that eventually led to FDA approval. The early success of ibrutinib has led to the clinical development of multiple BTK inhibitors.

Kinases, in general, are a well-established class of drug targets. , Many kinases feature a targetable cysteine at/near the catalytic ATP binding site which is concurrently the most frequent binding site for small molecule inhibitors. BTK possesses a targetable cysteine (C481) and is one of the first kinases to be inhibited by a covalent, irreversible inhibitor in the clinic (i.e., ibrutinib). While many successful drugs work via the covalent inhibition of a druggable protein (e.g., penicillin, omeprazole), development of irreversible inhibitors historically slowed due to fears of broad reactivity within the proteome and potential toxicity issues. , In recent years, however, it has become clear that selective and safe irreversible inhibitors can be designed by optimizing the reversible binding configuration of the inhibitor-protein complex to facilitate the maximal rate of covalent bond formation between the intended targets reactive thiol and a cysteine-reactive functional group (most often a Micheal acceptor). , Thanks in part to the success of ibrutinib, new covalent kinase inhibitors have emerged, and creative scientific teams continue to expand our understanding of the design features that drive clinical success. ,

There are scientific and economic reasons that promising drug targets receive simultaneous attention from private sector entities hoping to merit a “best-in-class” designation for their drug. Sometimes derided as “me-too” drugs, the reality is that parallel efforts to drug the same target often produce important alternatives for heterogeneous patient populations that benefit from the distinctive properties that these drugs inherently possess. Ibrutinib does have several off-target liabilities that are associated with adverse events including inhibition of EGFR (skin rash), TEC (bleeding) and CSK (atrial fibrillation). − Further, the off-target profile of ibrutinib is thought to contribute to adverse drug reactions that can be limiting in combination settings, including diarrhea, hypertension, and cardiac-related side effects. As a result, several second generation BTK inhibitors sought to limit or modify off-target profiles including acalabrutinib, poseltinib, evobrutinib and zanubrutinib (Figure A). − Other agents were designed to improve or alter elements of the drugs binding and inhibition characteristics including branebrutinib and tirabrutinib (reaction rate), rilzabrutinib (covalent reversibility), fenebrutinib (noncovalent reversibility) and remibrutinib (targeting of the inactive BTK conformation). − Figure B (and Table S1) provides information on each of these agents highest achieved clinical phase as of 2025, the related indication, and notes on the binding modality and reported potency.

1.

(A). The chemical structure of clinically approved or investigational BTK inhibitors. (B). The most advanced regulatory or clinical phase, the primary indication according to clinicaltrials.gov and the binding/inhibition modality and potency for each listed BTK inhibitor.

Well-designed covalent inhibitors rely upon a two-step process involving affinity-based protein-inhibitor binding followed by covalent bond formation. , Development of these agents involves optimized potency (reflected by a lower K _I value) based on inhibitor-protein structural alignment and enhanced covalent bond formation (reflected by a higher k _inact value) driven by optimized alignment of the nucleophile and electrophile. Equally important is minimizing the intrinsic chemical reactivity (k _chem) of the electrophile to avoid undesired modification of off-target proteins or endogenous nucleophiles such as glutathione (GSH). , There is also growing recognition that noncovalent inhibitors benefit from enhanced kinetic selectivity whereby optimized binding kinetics (k _on, k _off, residence time) promote desired phenotypes while minimizing the impact of transiently inhibited off-targets that may cause unwanted side effects. − Most traditional biochemical methods for tracking enzyme activity and inhibitor potency do not provide details on these kinetic-based features of enzyme inhibition. Despite the growing appreciation of inhibition kinetics (covalent and noncovalent), few drug classes offer a large enough collection of optimized inhibitors to conduct a comparative survey of in vitro phenotypic outcomes alongside inhibition kinetics to gain insight into the driving features of drug efficacy. Here, we survey key selectivity data, inactivation kinetics and in vitro cytotoxicity measures for 10 advanced BTK inhibitors to better understand which attributes govern the activity of these agents in models of B-cell malignancies.

Results

The off-target profile of kinase inhibitors may contribute to the efficacy of an agent if the “off-target” impacts a relevant signaling pathway that is promoting disease phenotypes. Given our desire to relate the kinetic inhibition features of selected BTK inhibitors to cellular cytotoxicity, it was important to understand (if not control for), any potentially confounding off-target effects of these agents as they relate to known B-cell lymphoma signaling pathways. , Studies detailing the discovery of each BTK inhibitor profiled herein (see refs – ) often included a limited panel of kinome-profiling data. Most, however, focused on the deleterious off-targets associated with ibrutinib (e.g., EGFR, Src). Further, diverse methods were utilized to generate these data which complicated any comparative interpretation. To produce a disease-relevant and harmonized set of data, we profiled these agents across 24 kinase targets associated with B-cell malignancy signaling cascades through the BCR (e.g., BTK, CSK, LYN kinases), toll-like receptors (e.g., IRAK, TAK1 kinases) and cytokine receptors (e.g., JAK kinases) (Figure A). Profiles were generated via a functional radiometric biochemical assay utilizing γ-³³P ATP (Reaction Biology) using a standard concentration range (10 doses ranging from 10 μM to 0.5 nM), a static ATP concentration of 10 μM, and a 20 min preincubation period before starting the 2 h kinase reaction before analysis. The outcomes of this profile reveal that all agents tested are potent BTK inhibitors and most have appreciable activity versus both BMX and BLK (Figure B and Table S2). BMX and BLK are well documented off-targets for ibrutinib and other BTK inhibitors. Importantly, as members of the TEC kinase family, both have a targetable cysteine (C496 and C345, respectively) making avoidance of these targets difficult. Ibrutinib, zanubrutinib and poseltinib were the most promiscuous beyond BMX and BLK. The remaining agents (acalabrutinib, tirabrutinib, rilzabrutinib, remibrutinib, fenebrutinib, evobrutinib and branebrutinib) were remarkably free of any additional off target activity.

2.

(A). Wiring diagram of key signaling pathways associated with the pathophysiology of B-cell lymphoma with selected kinase targets highlighted. (B). Potency of selected BTK inhibitors versus the 24 highlighted kinase targets as determined by the “HotSpot” kinase assay system (Reaction Biology).

We next utilized a commercial NanoBRET (bioluminescence resonance energy transfer) intercellular kinase assay (Reaction Biology) to measure each agents BTK target engagement within the context of living cells. The assay measures the competitive displacement of a NanoBRET tracer which reversibly binds a NanoLuc luciferase-BTK fusion construct in HEK293 cells. Each BTK inhibitor was tested in duplicate (starting at 10 μM, 10-dose with 3-fold dilution) with an incubation time of 2 h. The data is reported in Table S3 with activities ranging from sub-nM to over 20 nM with rankings as follows: branebrutinib < ibrutinib < remibrutinib < zanubrutinib < poseltinib < fenebrutinib < rilzabrutinib < evobrutinib < acalabrutinib < tirabrutinib.

The broader appreciation of kinetic selectivity has compelled the research community to utilize new assessments of ligand–target interactions. Expanded use of TR-FRET based assays (or comparable technologies) offers a means to determine key binding kinetic data including on/off rates and residence time for reversible inhibitors and inactivation parameters for irreversible inhibitors. , To generate a comparative assessment of the covalent and noncovalent BTK inhibitors profiled herein, we utilized the COVALfinder and KINETICfinder platforms (Enzymlogic) that employ unique active-site directed fluorescent probes for individual kinases. A key assessment from the COVALfinder system involves the % binding of the BTK-directed fluorescent probe over time in the presence of each inhibitor across 20 distinct concentrations (1000 nM to 0.02 nM; 1.75 dilution factor). The reaction progress curves for ibrutinib are shown in Figure A (see Figure S1 for all other agents). These data can be utilized to calculate the observed rate constant (k _obs) (Figures B and S2 and Table S4), the inactivation constant (K _I) and the maximum potential covalent reaction rate (k _inact) as previously described (also described in the Methods Section). , The rate constant k _inact/K _I reflects the overall inactivation efficiency and is considered the appropriate metric to compare the activity of irreversible, covalent inhibitors. When evaluating irreversible inhibitors that follow a two-step mechanism, it is also important to consider that potency arises from both the binding affinity (K _d = k _off/k _on) of the initial, reversible noncovalent binding event and the k _inact. Together, these factors define the inactivation constant K _I, where K _I = (k _off + k _inact)/k _on, which should not be confused with the equilibrium dissociation constant K _d. , Finally, analysis of the dose–response curves over time (from 5 to 240 min) highlights how both drug concentration and exposure duration impact IC₅₀ (see Figures C and S3).

3.

(A). The reaction progression curves for ibrutinib examined at 20 defined concentrations (1000 nM to 0.02 nM; 1.75 dilution factor) over 250 min. (B). The dependence of the observed rate constant (k _obs) on ibrutinib concentration. (C). Concentration dependent BTK inhibition by ibrutinib at 6 separate times of exposure (5–240 min). (D). Quadrant scale assessment of all covalent and covalent-reversible BTK inhibitors comparing rate of covalent bond formation (k _inact.(^s‑1)) against target affinity (K _i). (E). Enzyme inhibition kinetic analysis of all covalent, covalent-reversible and noncovalent BTK inhibitors.

Since modifying any part of an irreversible inhibitor may improve the rate of covalent bond formation (k _inact) at the expense of potency (K _I) or vice versa, the covalency quadrant serves as a valuable tool to discern which of these parameters primarily drives changes in inactivation efficiency. Within this classification, inhibitors are grouped as follows: quadrant I features low potency but high on-target reactivity; II = high potency, high on-target reactivity; III = high potency, low on-target reactivity; IV = low potency, low on-target reactivity (Figure D,E). From these data, remibrutinib (targeting the inactive BTK conformation), ibrutinib, and branebrutinib are positioned within quadrant III, reflecting high potency coupled with an optimized on-target reaction rate with BTK Cys481.

Our findings also reveal a wide range in inactivation efficiency with k _inact/K _I values from 2.5 × 10⁶ M^–1 s^–1 to 6.3 × 10⁴ M^–1 s^–1, representing a 40-fold span. Based on their BTK inactivation capacity, the eight irreversible covalent inhibitors can be ranked as follows: ibrutinib > remibrutinib > branebrutinib > zanubrutinib > poseltinib > tirabrutinib > evobrutinib > acalabrutinib. Notably, this significant difference in BTK inactivation efficiency is not primarily attributed to variations in k _inact, which exhibits only a 4-fold difference, but rather to differences in K _I values which range from 2.4 nM to 230 nM, representing a 96-fold change.

Rilzabrutinib is a reversible covalent inhibitor featuring a cyanoacrylamide warhead, characterized by prolonged residence time when bound to BTK, but rapid dissociation in the absence of proper stabilization of the covalent complex by the protein environment. Fenebrutinib is a noncovalent, active-site BTK inhibitor with a reported 18 h residence time as determined by a “jump dilution” biochemical experiment. To provide comparable kinetic selectivity data for these agents, we profiled them using the TR-FRET based KINETICfinder and COVALfinder assay systems to determine K _d, k _off and residence time values. These data demonstrate that rilzabrutinib possesses remarkable enzyme affinity (K _d 0.14 nM) and an extraordinarily stable BTK-inhibitor complex (residence time of 812 min) (Figure E and Table S4). Similarly good data was noted for fenebrutinib in both the phosphorylated (K _d 0.69 nM, residence time of 10 min) and nonphosphorylated (K _d 0.17 nM, residence time of 145 min) BTK isoforms (Figure S1I). For compounds exhibiting tight-binding behavior, such as fenebrutinib, discrepancies between residence times measured in this study and those previously reported using the jump dilution method can be attributed to differences in experimental conditions. Specifically, the jump dilution assay involves preincubating the target with the inhibitor at concentrations well above its IC₅₀, which may facilitate rebinding and prolonged apparent residence times.

The conformation of BTK is highly dynamic, with the protein adopting a range of active and inactive states depending on its phosphorylation status and microenvironment influenced protein complexes. The crystal structures of remibrutinib and fenebrutinib demonstrate their ability to stabilize an inactive (nonphosphorylated) conformation of BTK by occupying a selectivity pocket, also known as the H3 pocket, which is present exclusively in the nonphosphorylated form of the kinase. , By stabilizing BTK in the DFG-out conformation, both remibrutinib and fenebrutinib effectively prevent phosphorylation of the activation loop by upstream kinases. Consistent with these structural insights, the inactivation efficiency of remibrutinib is 140-fold higher for the nonphosphorylated form of BTK (k _inact/K _I = 1.9 × 10⁶ M^–1 s^–1) compared to the phosphorylated form (k _inact/K _I = 1.4 × 10⁴ M^–1 s^–1), with similar trends observed for fenebrutinib. The phosphorylation state of BTK is key to the binding kinetics of fenebrutinib, where interaction with the H3 pocket results in a 5-fold increase in residence time when binding to the nonphosphorylated form of the kinase. Overall, the combined datasets reveal that the majority of clinically active BTK inhibitors possess optimized enzyme inhibition kinetics.

With a uniformed dataset informing on the kinetic enzyme inhibition of the BTK inhibitors profiled in this study, we next sought to establish comparative in vitro activity (cytotoxicity) profiles in relevant cell models of lymphoma. For most of these models, ibrutinib as a single agent possesses limited cellular cytotoxicity or apoptosis induction potential across relevant concentration ranges (up to 10 μM) at standard assay time points (48–72 h). , However, when combined with other drugs that target codependencies in lymphoma (e.g., BCL-2 targeting agents including venetoclax), ibrutinib demonstrates profound cytotoxicity and pro-apoptotic activity. Thus, we utilized previously reported methods examining the combination potential of each BTK inhibitor with venetoclax in 3 relevant cell models (two DLBCL models: OCI-Ly10 and HBL1 and one MCL model: REC-1). , We first judged the comparative activity of each BTK inhibitor by ascertaining the degree of synergistic cell killing across a wide range of drug concentrations. In the DLBCL models, each BTK inhibitor was examined at dose ranges from 2000 nM to 7.8 nM (1:2 dilution factor) and venetoclax was examined from 1000 nM to 3.9 nM (1:2 dilution factor). In the MCL cell model, venetoclax was examined from 10000 nM to 39 nM (1:2 dilution factor). We have previously shown that the combination of ibrutinib and venetoclax is highly synergistic in these cell models using multiple methods for assessing synergy. The results from this study confirm these outcomes with synergistic cell killing observed across multiple drug concentration ranges as judged by the ΔBliss summation (ΔBlissSum) method (Figure A). In the OCI-Ly10 model, the ΔBlissSum values demonstrate that multiple BTK inhibitors (e.g., remibrutinib, rilzabrutinib and zanubrutinib) possess similar, broad synergistic cell killing across multiple concentration ranges when combined with venetoclax (Figures B, S4 and Table S5A). The venetoclax combination outcomes for a few agents (poseltinib and acalabrutinib) were more focal in nature, occurring only at the higher concentration ranges of both drugs. Similar rankings were noted for the HBL1 and REC-1 cell models (Figures C,D, S5, S6 and Table S5B,C).

4.

(A). Heatmap view of ibrutinib + venetoclax 10 × 10 block in the OCI-Ly10 cell line. Both % response and ΔBliss heatmaps are shown. (B). Bar chart comparison of the ΔBlissSum values for all BTK inhibitors profiled in this study in combination with venetoclax in the OCI-Ly10 cell line with representative visuals of selected ΔBliss heatmaps. (C). Bar chart comparison of the ΔBlissSum values for all BTK inhibitors profiled in this study in combination with venetoclax in the HBL1 cell line with representative visuals of selected ΔBliss heatmaps. (D). Bar chart comparison of the ΔBlissSum values for all BTK inhibitors profiled in this study in combination with venetoclax in the REC-1 cell line with representative visuals of selected ΔBliss heatmaps. (E). Heatmap view of ibrutinib + venetoclax 10 × 10 block in the OCI-Ly10 cell line showing the % response values and isolating the potentiation response data for venetoclax as affected by ibrutinib at a single concentration (7.8 nM). (F). Bar chart comparison of the IC₅₀ values for venetoclax values as affected by all BTK inhibitors profiled in this study at a single concentration (7.8 nM) in the OCI-Ly10 cell line. (G). Bar chart comparison of the IC₅₀ values for venetoclax values as affected by all BTK inhibitors profiled in this study at a single concentration (7.8 nM) in the HBL1 cell line. (H). Bar chart comparison of the IC₅₀ values for venetoclax values as affected by all BTK inhibitors profiled in this study at a single concentration (7.8 nM) in the REC-1 cell line.

Beyond the analysis of synergy outcomes across multiple concentration ranges, it is also instructive to consider the impact of BTK inhibitor exposure at a static low dose. In each of the cell models tested, venetoclax had no single agent activity at the concentrations tested (consider the far-right column in Figure E). Evaluating the impact of adding a BTK inhibitor at its lowest tested dose (7.8 nM) on the venetoclax cytotoxicity yields an affected concentration response curve and IC₅₀ value. The addition of ibrutinib (7.8 nM), for instance, stimulates dose–response activity for venetoclax across the tested concentration range (1000 nM to 3.9 nM (1:2 dilution factor)) resulting in a IC₅₀ value of 62 nM (Figure E). Using this analysis method to compare each of the BTK inhibitors tested herein reveals those agents with strong venetoclax potentiation even at this low exposure concentration in all 3 cell models (Figure F–H and Supporting Information Tables 6–C). These data consistently show that exposure to 7.8 nM concentrations of remibrutinib, ibrutinib, branebrutinib and zanubrutinib stimulate the lowest venetoclax IC₅₀ values.

For covalent inhibitors and inhibitors with long residence times, drug pharmacodynamics can be decoupled from drug pharmacokinetics if exposure levels allow for saturating inhibition. In these instances, target inhibition persists until new protein is synthesized. For BTK, studies show that protein replacement is heterogeneous across disease states and subjects with rates ranging from 3 to 40% after 24 h. − Achieving saturating enzyme inhibition requires both optimized inhibition kinetics and sufficient drug exposure following dosing. We reviewed key pharmacokinetic data from the published literature for the drugs profiled in this study including the highest serum concentration value (C _max) and the time required for the serum concentration to decrease by 50% (T _1/2) (Figure A,B and Table S1 (and references therein)). It is important to note that the data captured and presented represents exposure values following a defined or assumed recommended dose for oncology indications and that multiple reports exist for many of the drugs studied herein. Alternate values exist from those we show and the dosing choices may diverge for nononcology indications. For instance, the data listed herein for remibrutinib is based on a 400 mg QD dose. In alternate indications (e.g., Chronic Spontaneous Urticaria), remibrutinib is dosed chronically at 25 mg BID. It is also important to note that C _max values are for total drug. The fraction unbound (FU) values, which can influence drug exposure at the disease site, are only reported for selected agents including ibrutinib (2.7% FU), acalabrutinib (2.5% FU), zanubrutinib (6% FU), tirabrutinib (8.3% FU) and branebrutinib (1.2% FU). While imperfect, the relative exposure levels for the BTK inhibitors profiled herein can provide an estimation of the physiological likelihood that each agent achieves a high level of BTK occupancy and inhibition. These data highlight a wide range of pharmacokinetic values for the BTK inhibitors listed. For each drug, the dosing and resulting exposure profiles are influenced by both tolerability and pharmacodynamic considerations. The PK data for ibrutinib, for instance, represents the exposure following a 560 mg dose. Previous work has shown that ibrutinib achieves >95% BTK occupancy in CLL cells in situ after 4 h following doses as low as 2.5 mg/kg (175 mg dose for a 70 kg subject). Similar data is available for many of the agents profiled herein with similar conclusions (i.e., achievement of full BTK inhibition at the prescribed dose and pharmacokinetic exposures).

5.

(A). A generic pharmacokinetics drug concentration vs time (PK) plot highlighting C _max values for all BTK inhibitors profiled in this study at a stated or assumed oncology dose level. (B). A generic pharmacokinetics drug concentration vs time (PK) plot highlighting T _1/2 values for all BTK inhibitors profiled in this study at a stated or assumed oncology dose level. (C). Plot of each agents C _max value (x-axis) versus each agents rate constant k _inact/K _I (y-axis). (D). Plot of each agents T _1/2 value (x-axis) versus each agents rate constant k _inact/K _I (y-axis). (E). Plot of each agents C _max values (x-axis) versus each agents affected IC₅₀ value for venetoclax (y-axis). (F). Ranking of the % apoptotic cells (CaspaseGlo) for HBL1 cells treated with a cocktail of venetoclax, prednisolone and each BTK inhibitor profiled in this study at set concentrations for each agent (20.5 nM) over 8 h.

The pharmacokinetic differences of these agents are meaningful and often yield contrasting benefits. It is instructive, therefore, to consider how each drugs inactivation kinetics are impacted by their pharmacokinetic exposure realities. The lower k _inact/K _I value for acalabrutinib, for instance, may be less problematic in vivo given this agent’s higher C _max value (1.78 μM) (Figure C). The rapid clearance of branebrutinib (T _1/2 = 1.6) coupled with its promising BTK inactivation potential may prove beneficial as the drug is removed rapidly, mitigating potential tolerability issues, after effectively covalently inhibiting BTK (Figure D). We also evaluated the BTK inhibitor affected venetoclax IC₅₀ values alongside each agent’s exposure (e.g., C _max) (Figure E). Ibrutinib and branebrutinib, for instance, promote outstanding affected venetoclax IC₅₀ values but have relatively low C _max values. Conversely, remibrutinib and zanubrutinib also had exceptional affected venetoclax IC₅₀ values but possess relatively high C _max values. The significance of these data is unclear. The dogmatic view often defaults to an assumption that higher levels of drug exposure are preferred. Pharmacokinetic drug concentrations are captured via serum levels. Outside the circulatory system, diseased cells may be exposed to less drug. Thus, drugs with higher exposure concentrations and a broader volume of distribution may have a greater impact on remote diseased cells and a greater potential for improved therapy responses. These realities favor agents with good inactivation kinetics and increased exposure profiles.

The timing of drug response should also be considered. We have previously shown that the pro-apoptotic effects of ibrutinib as a single agent are marginal at reasonable concentrations (up to 10 μM) and time points (24 h) in multiple lymphoma cell models. , The combination of ibrutinib, venetoclax and prednisolone, however, elicits rapid (<12 h) apoptosis induction at low concentrations (<200 nM for each agent). To compare the BTK inhibitors examined in this study, we explored the pro-apoptotic potential of a cocktail of BTKi + venetoclax + prednisolone at a static concentration of each agent (20.5 nM) at an 8 h time point. For those agents with rapid clearance (acalabrutinib, branebrutinib), the 8 h time point represents a zone where selected individuals may begin to recover functional BCR signaling through BTK enzyme replacement. Rapid, therapy-stimulated apoptosis could overcome the role of BTK replenishment on therapy response. Each BTK inhibitor, in combination with venetoclax and prednisolone, had a pro-apoptotic effect above baseline (DMSO control) (Figure F and Table S7). Several agents (ibrutinib, zanubrutinib) stimulated cellular apoptosis close to the level of the control (bortezomib (20 μM)). Importantly, for those agents with longer half-lives (ibrutinib, fenebrutinib), drug will still be present if BTK enzyme replacement occurs with the possibility of further extending therapy effectiveness.

Discussion and Conclusion

BTK inhibitors represent an important and exciting class of drugs with potential to effectively treat a range of human diseases. , Given this promise, numerous approved and investigational BTK inhibitors have been developed including both covalent and noncovalent modifiers of BTK. The collective availability of these agents offers a rare opportunity to study both kinase selectivity and enzyme inactivation parameters as governing features of in vitro cellular response. We utilized biochemical kinome profiling and kinetic selectivity measurements alongside in vitro cytotoxicity and apoptosis evaluations in cellular models of lymphoma (DLBCL and MCL) to decipher which properties promote promising phenotypic responses following drug exposure. We also reviewed the concentration and timing of phenotypic responses relative to the reported pharmacokinetics of each of the agents profiled as a potential insight into clinical outcomes.

The biochemical evaluation of selectivity focused on 24 kinase targets of relevance in lymphoma. The first-in-class BTK inhibitor ibrutinib was the most promiscuous across this panel (Figure B). Next-generation agents remibrutinib, fenebrutinib, tirabrutinib and acalabrutinib were the most selective. Across all cell-based phenotypic surveys of cytotoxicity and apoptosis (in combination with venetoclax and prednisolone), ibrutinib consistently ranked among the most active agents tested. However, several of the most selective BTK inhibitors (e.g., remibrutinib and fenebrutinib) were also ranked highly in these studies. These data suggest that the on-target inhibition of BTK by these agents is the primary contributor to lymphoma cell cytotoxicity. It should be noted that these data do not provide context on the relative occupancy of different targets over time and, thus, the pharmacokinetic impact on in vivo selectivity.

To assess key features of the BTK inactivation kinetics of these agents, we utilized TR-FRET based assays keyed toward both covalent and noncovalent inhibitors (Figures and S1–S3). From these studies, the vast majority of tested BTK inhibitors possessed exceptional binding and inhibition kinetics with remibrutinib, ibrutinib, zanubrutinib, fenebrutinib, rilzabrutinib and branebrutinib having the best overall profiles. These outcomes aligned reasonably well with NanoBRET assessments of BTK inhibition and the various in vitro phenotypic assessments which consistently ranked these agents among the most active. Notably, for irreversible inhibitors, inactivation efficiency correlates reasonably well with in vitro phenotypic (cytotoxicity and apoptosis) outcomes, underscoring the impact of inactivation kinetics on therapeutic efficacy.

We also collected key pharmacokinetics data for each of the agents examined herein using stated or assumed dosing levels for an oncology indication. There exists a wide range of exposure concentrations (C _max from 0.02 to 3.66 μM) and times (T _1/2 from 1.1 to 9.5 h). The various data presented must be viewed cautiously as not every agent has a reported fraction unbound value (providing for free drug concentrations) or volume of distribution. When aligning these data with the kinetic selectivity and phenotypic outcomes presented in this study, it is clear that all the agents profiled are well optimized BTK inhibitors with exciting clinical potential. Several agents, including remibrutinib, zanubrutinib, rilzabrutinib and fenebrutinib, do stand out as possessing remarkable selectivity, kinetic inhibition, phenotypic activity and favorable pharmacokinetics.

Continued clinical assessment of these agents will be the ultimate arbiter of each drug’s potential. Ibrutinib, acalabrutinib, zanubrutinib and tirabrutinib have demonstrated clinical utility to the degree that they have earned regulatory approval. Several other agents (remibrutinib, rilzabrutinib and fenebrutinib) are the subject of recent, successful clinical studies. , Others (branebrutinib and poseltinib) have been discontinued for undisclosed reasons. Whether successful or not, each of these agents represents remarkable effort and insight into the expanding appreciation of enzyme inhibition kinetics. Hopefully, through careful comparative analyses, these drugs will continue to refine our understanding and appreciation of the parameters that govern success for kinase inhibitor therapies and beyond.

Methods

HotSpot and NanoBRET Kinase Screening

Biochemical assessment of each BTK inhibitor versus 24 targets of relevance in lymphoma was performed using the commercial “HotSpot” kinase assay system (Reaction Biology). Protocol details are provided by the commercial vendor (https://www.reactionbiology.com/services/kinase-assays/kinase-screening/) and as previously reported. Briefly, BTK enzyme and substrate are preincubated with the individual inhibitors for 20 min followed by γ-³³P ATP addition, incubation for 2 h and radioisotopic P81 filter-binding measurement. The NanoBRET target engagement assay employs an energy transfer technique designed to measure molecular proximity in living cells. The assay measures the apparent affinity of test compounds by competitive displacement of the NanoBRET tracer, reversibly bound to a NanoLuc luciferase-kinase fusion construct in cells. Protocol details are provided by the commercial vendor (Reaction Biology) (https://www.reactionbiology.com/services/kinase-assays/nanobret-intracellular-kinase-assay/).

KINETICfinder Screening

KINETICfinder detects the binding and displacement of a fluorescent probe at the active site using TR-FRET and a labeled antitag antibody (Life Technologies). Assays were conducted in black 384-well microplates (Greiner) containing either 0.05 nM full-length human BTK (inactive, Life Technologies) or ATP-activated BTK (Carna Biosciences). The reaction mixture included 10 nM fluorescent probe, 2 nM terbium-labeled antibody and assay buffer (50 mM HEPES, pH 7.5, 10 mM MgCl₂, 0.01% Brij-35, 1 mM DTT, and 1% DMSO). Test compounds were prepared in DMSO as 100X concentrated solutions and diluted in a 4-point, 10-fold series. The kinetic assays were read continuously at RT in a PHERAstar FSX plate reader (BMG LABTECH) and the specific signals were fitted to the Motulsky-Mahan’s eq (eq ). The affinity (K _d), association rate constant (k _on), dissociation rate constant (k _off) and residence time (T) of each test compound were calculated using KINPy software. In this setting, we report as % binding.

$$K_{\mathrm{A}}=k_1[L]+k_2$$

$$K_{\mathrm{B}}=k_3[I]+k_4$$

$$S=\sqrt{({(K_{\mathrm{A}}-K_{\mathrm{B}})}^2+4\times k_1\times k_3\times L\times I\times {10}^{-18})}$$

$$K_{\mathrm{F}}=0.5\times (K_{\mathrm{A}}+K_{\mathrm{B}}+S)$$

$$K_{\mathrm{S}}=0.5\times (K_{\mathrm{A}}+K_{\mathrm{B}}-S)$$

$$\mathrm{D}\mathrm{I}\mathrm{F}\mathrm{F}=K_{\mathrm{F}}-K_{\mathrm{S}}$$

$$Q=\frac{B_{\max }\times k_1\times L\times {10}^{-9}}{\mathrm{D}\mathrm{I}\mathrm{F}\mathrm{F}}$$

$$Y=Q\times (\frac{k_4+\mathrm{D}\mathrm{I}\mathrm{F}\mathrm{F}}{K_{\mathrm{F}}\times K_{\mathrm{S}}}+\frac{k_4-K_{\mathrm{F}}}{K_{\mathrm{F}}}{e}^{(-K_{\mathrm{F}}\times X)}-\frac{k_4-K_{\mathrm{S}}}{K_{\mathrm{S}}}{e}^{(-K_{\mathrm{S}}\times X)})$$ 1

COVALfinder Screening

COVALfinder monitors the binding and displacement of an active-site-directed fluorescent probe via TR-FRET detection. The assay utilizes a labeled antitag antibody and includes either inactive BTK or active BTK at 0.05 nM, combined with 10 nM fluorescent probe and 2 nM terbium antibody in assay buffer. For all experiments, 100X concentrated test compounds were serially diluted (1.75-fold, 20 points) in DMSO. Kinetic measurements were continuously recorded at RT and the specific signals were fitted to a single-exponential equation to determine k _obs. A secondary plot of k _obs versus compound concentration enabled the calculation of kinetic constants and inhibition mechanisms, distinguishing between two-step irreversible covalent (eq ) and two-step reversible covalent (eq ). In the irreversible model, k _inact represents the maximum inactivation rate at infinite compound concentration, while K _I denotes the concentration required to achieve half-maximal inactivation (1/2 × k _inact), with inactivation efficiency expressed as k _inact/K _I. The half-life for inactivation at this infinite concentration is given by t _1/2 = 0.693/k _inact. In the reversible covalent model, K _d represents the equilibrium dissociation constant of the initial encounter complex, while k ₃ and k ₄ define the forward and reverse rate constants for covalent bond formation. The final high-affinity complex dissociation constant (K _d*) is determined using eq . Additionally, dose–response curves were generated at each time point to calculate IC₅₀ values, facilitating the assessment of time-dependent inhibition. In this setting, we report as %binding.

$$k_{\mathrm{o}\mathrm{b}\mathrm{s}}=\frac{k_{\mathrm{i}\mathrm{n}\mathrm{a}\mathrm{c}\mathrm{t}}\times X}{K_{\mathrm{I}}+X}$$ 2

$$k_{\mathrm{o}\mathrm{b}\mathrm{s}}=k_4+\frac{k_3\times X}{K_{\mathrm{d}}+X}$$ 3

$$K_{\mathrm{d}}^{*}=\frac{K_{\mathrm{d}}}{1+\left(\frac{k_3}{k_4}\right)}$$ 4

Cell-Lines and Reagents

All the DLBCL cell-lines used for this study were cultured in advanced RPMI 1640 medium (Thermo Fisher) supplemented with penicillin/streptomycin and 5% fetal bovine serum (Tet tested, Atlanta Biologicals). All the cell-lines were maintained/expanded in a standard tissue culture incubator at 37 °C in a 5% CO₂ environment.

Quantitative High-throughput Combination Screening (Pairwise Matrix Format)

For pairwise drug-combination assessment in matrix format, compounds (10 nL) were acoustically dispensed into 1536-well white tissue culture-treated plates with an Echo 655 acoustic liquid handler (Labcyte). A 9-point custom concentration range with 1:2 dilution between points was used for pairwise drug-combination assessments in 10 × 10 matrix format. Dry-spotted plates were sealed and stored at −80 °C until needed. After thawing the plates, cells were added to compound-containing plates at a density of 500-cells/well in a final volume of 5 μL media, using a using a Multidrop Combi dispenser (ThermoFisher). Plates were incubated for the desired time at standard incubator conditions, covered by a stainless steel gasketed lid to prevent evaporation. For cell viability measurements, 3 μL of Cell Titer Glo (Promega) were added to each well and plates were incubated at room temperature for 15 min with the stainless-steel lid in place. Luminescence readings were taken using a Viewlux imager (PerkinElmer) with a 2″ exposure time per plate. For apoptosis activation measurements, 3 μL of Casp3/7 Glo (Promega) were added to each well and plates were incubated at room temperature for 15 min with the stainless-steel lid in place. Luminescence readings were taken using a Viewlux imager (PerkinElmer) with a 10″ exposure time per plate. Because both Cell Titer Glo and Casp3/7 Glo are lysis-based reagents, multiple sets of identical plates were used to collect each individual cell-line/readout/time-point. Measurement of drug-to-drug interactions was performed as previously described , and the ΔBliss metric was used to quantitatively compare/contrast synergy/antagonism/additivity across cell-lines.

Multi-Component, Time-Course (MCTC) Screening

We established fixed dose levels (20.5 nM) for the three-drug cocktail of prednisolone, venetoclax and the defined BTK inhibitor and used an Echo 655 acoustic liquid handler (Labcyte) to dispense each agent into an individual wells of a single 1536-well white bottom, polystyrene tissue culture-treated, Corning plate. Each drug-treatment set was plated four times (N = 4) and a full column (32 wells) of DMSO (30 nL) was used for negative control and background subtraction. A full column (32 wells) of Bortezomib (20 μM) was used as a positive control for apoptosis induction. Assay plates were incubated for the desired time at standard incubator conditions, covered by a stainless steel gasketed lid to prevent evaporation. For apoptosis activation measurements, 3 μL of Casp3/7 Glo (Promega) were added to each well and plates were incubated at room temperature for 15 min with the stainless-steel lid in place. Luminescence readings were taken using a Viewlux imager (PerkinElmer) with a 10″ exposure time per plate. Pro-apoptotic activity was read using normalized relative fluorescent units (RFU) of individual drug-treatment sets at 8 h.

Glossary

Abbreviations

BTK

Bruton’s tyrosine kinase

BCR

B-cell receptor

Cys

cysteine

CLL

chronic lymphocytic leukemia

DLBCL

diffuse large B-cell lymphoma

MCL

mantle cell lymphoma

WM/LPL

Waldenstrom macroglobulinemia/lymphoplasmacytic lymphoma

MZL

marginal zone lymphoma

FL

follicular lymphoma

FDA

food and drug administration

EGFR

epidermal growth factor receptor

TEC

tec protein tyrosine kinase

CSK

C-terminal Src kinase

GSH

glutathione

Src

proto-oncogene tyrosine-protein kinase Src

CSK

tyrosine-protein kinase CSK

LYN

tyrosine-protein kinase Lyn

IRAK

interleukin receptor associated kinase

TAK1

mitogen-activated protein kinase kinase kinase 7

JAK

Janus kinase

ATP

adenosine triphosphate

BMX

cytoplasmic tyrosine-protein kinase BMX

BLK

tyrosine-protein kinase BLK

TEC

tyrosine-protein kinase TEC

BRET

bioluminescence resonance energy transfer

TR-FRET

time-resolved fluorescence energy transfer

IC₅₀

half-maximal inhibitory concentration

QD

quaque die (once daily)

BID

bis in die (twice daily)

DMSO

dimethyl sulfoxide

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsptsci.5c00412.

• Figure S1: the reaction progression curves examined at 20 defined concentrations. Figure S2: the dependence of the observed rate constant (kobs) on concentration. Figure S3: concentration dependent BTK inhibition at 6 separate times of exposure. Figure S4: heatmap view of venetoclax 10 × 10 combination blocks in the OCI-Ly10 cell line. Figure S5: heatmap view of venetoclax 10 × 10 combination blocks in the HBL1 cell line. Figure S6: heatmap view of venetoclax 10 × 10 combination blocks in the REC-1 cell line. (PDF)

• SI Table 1: BTK inhibitor listing SI Table 2: kinase profiling versus key lymphoma targets SI Table 3: kinase profiling via BTK NanoBRET assay for target engagement in HEK293 SI Table 4: kinetic inhibition profiling versus BTK. SI Table 5A/B/C: combination assessment in OCI-Ly10 (A), HBL1 (B), ReC1 (C). SI Table 6A/B/C: BTKi potentiation of venetoclax in OCI-Ly10 (A), HBL1 (B), ReC1 (C). SI Table 7: Pro-apoptosis following BTKi + venetoclax + prednisolone at 8 h in HBL1 (XLSX)

#.

A.C. and X.Z. contributed equally. Data acquisition: A.C., X.Z., P.A. C.K.-T., C.MK., D.H., K.W. Data Analysis: A.C., P.A. P.J.M, C.M., J.D.P., D.H., S.B.H., M.R., P.J.T., W.H.W., M.C., L.M.S., C.J.T. Writing and editing: A.C., P.A. P.J.M, C.M., J.D.P., M.R., P.J.T., W.H.W., M.C., L.M.S., C.J.T. Conceptualization: A.C., P.A. M.C., L.M.S., C.J.T.

The authors declare no competing financial interest.