Only Time Will Tell: Modeling the Kinetics of Covalent Inhibitors

Only Time Will Tell: Modeling the Kinetics of Covalent Inhibitors | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL British Journal of Pharmacology This is a preprint and has not been peer reviewed. Data may be preliminary. 10 December 2025 V1 Latest version Share on Only Time Will Tell: Modeling the Kinetics of Covalent Inhibitors Authors : Madeeha I. Ali and Peter J. Tonge [email protected] Authors Info & Affiliations https://doi.org/10.22541/au.176537376.66820079/v1 Published British Journal of Pharmacology Version of record Peer review timeline 879 views 229 downloads Contents Abstract Supplementary Material Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract The selection and optimization of drug leads is largely driven by equilibrium parameters such as IC50 values. However, this approach does not account for the time-dependence of drug-target interactions which is important given that drug and target concentrations fluctuate in the human body. A fundamental understanding of drug-target binding kinetics is particularly important when the formation and breakdown of the drug-target complex is slow compared to the rate of drug elimination and becomes critical when dealing with covalent drugs where the drug is irreversibly bound to the target. The parameters that define covalent inhibition, specifically kinact and KI, can be determined by quantifying target binding as a function of drug concentration and time. However, such assays are difficult to implement in early stages of drug discovery. Here, we review practical approaches to quantify irreversible inhibition including progress-curve kinetics, multi-timepoint IC50 fitting (EPIC) for scalable library triage, intact- and peptide-level mass spectrometry to confirm adducts and sites, label-free biophysics (SPR, NMR) for orthogonal validation, and washout/jump-dilution experiments to diagnose reversibility and measure residence time (τ). We then describe the translation of these microscopic rates to pharmacology through three system determinants: protein turnover, target vulnerability, and pharmacokinetics using case studies from BTK, JAK3, KRASG12C, and EGFR to illustrate how aligning chemistry with context yields durable efficacy. Finally, we propose reporting standards (IC50 with incubation time; kinact/KI; τ; turnover-driven occupancy) and a kinetics-first design paradigm to replace static potency with mechanism-anchored optimization of covalent inhibitors. Only Time Will Tell: Modeling the Kinetics of Covalent Inhibitors Madeeha I Ali 1,2 and Peter J Tonge 1,2,3* 1 Department of Chemistry and 2 Center for the Advanced Study of Drug Action, Stony Brook University, Stony Brook NY, 11794-3400, United States 3 Department of Biomedical Genetics, University of Rochester, Rochester, NY 14642, United States *Author to whom correspondence should be addressed: Email: [email protected] KEYWORDS Covalent inhibitor; binding kinetics; time-dependent drug activity; irreversible inhibition; non-equilibrium; slow-onset ABSTRACT The selection and optimization of drug leads is largely driven by equilibrium parameters such as IC 50 values. However, this approach does not account for the time-dependence of drug-target interactions which is important given that drug and target concentrations fluctuate in the human body. A fundamental understanding of drug-target binding kinetics is particularly important when the formation and breakdown of the drug-target complex is slow compared to the rate of drug elimination and becomes critical when dealing with covalent drugs where the drug is irreversibly bound to the target. The parameters that define covalent inhibition, specifically k inact and K I , can be determined by quantifying target binding as a function of drug concentration and time. However, such assays are difficult to implement in early stages of drug discovery. Here, we review practical approaches to quantify irreversible inhibition including progress-curve kinetics, multi-timepoint IC 50 fitting (EPIC) for scalable library triage, intact- and peptide-level mass spectrometry to confirm adducts and sites, label-free biophysics (SPR, NMR) for orthogonal validation, and washout/jump-dilution experiments to diagnose reversibility and measure residence time ( τ ). We then describe the translation of these microscopic rates to pharmacology through three system determinants: protein turnover, target vulnerability, and pharmacokinetics using case studies from BTK, JAK3, KRAS G12C , and EGFR to illustrate how aligning chemistry with context yields durable efficacy. Finally, we propose reporting standards (IC 50 with incubation time; k inact /K I ; τ ; turnover-driven occupancy) and a kinetics-first design paradigm to replace static potency with mechanism-anchored optimization of covalent inhibitors. 1  INTRODUCTION The first covalent drugs emerged in the late 19 th century with the discovery of acetylsalicylic acid (aspirin), which irreversibly acetylates Ser-529 in the active site of cyclooxygenase-1 (COX-1) to block the synthesis of pro-inflammatory prostaglandins (Vane, 1971). Although covalent inhibitors demonstrated that chemical modification of a biological target could lead to durable pharmacological activity, they were historically regarded with skepticism due to concerns over nonspecific reactivity and potential metabolic toxicity (Singh, Petter, Baillie, & Whitty, 2011). As a result, drug discovery programs predominantly focused on reversible inhibitors, and covalent strategies were mainly confined to natural products and a few successful drugs such as β-lactam antibiotics (De Cesco, Kurian, Dufresne, Mittermaier, & Moitessier, 2017; Nicola, Tomberg, Pratt, Nicholas, & Davies, 2010). However, the approval of targeted covalent inhibitors in oncology and immunology demonstrated that covalent chemistry can be deployed selectively and safely (Kalgutkar & Dalvie, 2012). These inhibitors also offer extended clinical efficacy in disease models where reversible inhibitors have often failed, demonstrating how covalent drugs can sustain target occupancy despite declining plasma concentrations. For example, ibrutinib, the first-in class inhibitor of Bruton’s tyrosine kinase (BTK), established a therapeutic paradigm for B-cell malignancies and led to the development of numerous next-generation covalent inhibitors (Advani et al., 2013; Ouerdani et al., 2025). More recently, sotorasib, a covalent inhibitor targeting the KRAS G12C mutant, long considered “undruggable”, proved that covalent strategies can succeed where traditional approaches fall short (Canon et al., 2019; Janne et al., 2022). The clinical success of covalent inhibitors has transformed their development from a largely serendipitous process into one grounded in rational, mechanism-based design. The growing number of covalent drug discovery programs underscores the need for a deeper understanding of the mechanisms that govern covalent drug action. (De Vita, 2021; Potashman & Duggan, 2009). Despite the increased focus on covalent inhibitor discovery, a significant conceptual gap remains. Rational selection and optimization in medicinal chemistry remains rooted in equilibrium potency measurements typically performed over a range of drug concentrations. While conventional equilibrium-based measures of activity, such as IC 50 or K d values, are convenient and high throughput, these assays capture only the reversible binding step and represent a pseudo-equilibrium in the context of irreversible inhibitors. They therefore provide only a snapshot of inhibition at a fixed timepoint and do not report on the time-dependent progression from the noncovalent EI complex to the covalent EI complex they provide only a snapshot of inhibition at a fixed point in time (Strelow, 2017; Tonge, 2019). As a result, some compounds may appear to be weak binders based solely on equilibrium potency yet demonstrate robust cellular or in vivo activity due to favorable covalent kinetics. In contrast, high-affinity binders with slow inactivation rates may fail to achieve sufficient occupancy within the relevant pharmacokinetic window. The reliance on equilibrium parameters therefore risks both over and under-estimation of therapeutic potential (Tonge, 2018). Since time dependent inhibition is the defining pharmacological feature of covalent inhibition, rigorous analysis requires kinetic models that can deconvolute binding affinity from chemical reactivity. These models must distinguish between fast-acting and slow-acting agents and provide quantitative parameters related to pharmacokinetics, target turnover, and ultimately clinical response. Without such modeling, compound optimization can turn into a trial-and-error process, with little mechanistic rationale guiding progression (Copeland, 2013; Harris et al., 2018; Mons, Roet, Kim, & Mulder, 2022; Tonge, 2018). This Review aims to summarize approaches for modeling the kinetics of covalent inhibition and to illustrate how these approaches inform pharmacological translation ( Figure 1 ). First, we describe mechanistic frameworks that define covalent binding events, including irreversible and reversible-covalent inhibition. Next, we evaluate experimental strategies used to extract kinetic parameters, ranging from IC 50 measurements at single time points to progress-curve analysis and mass spectrometry-based quantification. We discuss the clinically validated inhibitors ibrutinib, ritlecitinib, and sotorasib to exemplify fast, intermediate, and slow covalent kinetics, respectively. Finally, we explain how protein turnover and target vulnerability modulate pharmacological outcomes, highlighting the importance of integrating biochemical kinetics with system-level determinants. By integrating in vitro assays and clinical efficacy, kinetic modeling offers a quantitative framework for the rational development of covalent inhibitors. Figure 1: Modeling kinetics of covalent inhibitors 2  MECHANISTIC FRAMEWORKS OF COVALENT INHIBITION The mechanism of covalent drug-target complex formation is discussed here in the context of enzyme inhibition. In addition, we assume that formation of the covalent adduct occurs via a two-step mechanism in which the initial reversible encounter complex (EI) is followed by the formation of a covalent bond between drug and target (E-I) and is analogous to the two-step induced-fit mechanism often observed for time-dependent reversible inhibitors. By convention, the microscopic rate constants for the two-step reaction are k 3 -k 6 where k 6 = 0 for an irreversible inhibitor ( Figure 2 ). Figure 2 . Inhibition mechanisms and analysis. (A) Two-step induced-fit mechanism for reversible inhibition. (B) Two-step mechanism for irreversible covalent inhibition. (C) Equations that describe irreversible inhibition: ( 1) K i is the dissociation constant of the initial encounter complex EI. ( 2 ) K I is the concentration of inhibitor that results in half the maximum rate of inactivation (k inact ). If k 4 >> k 5 then K I will be numerically the same as K i . The constants k inact and K I are analogous to k cat and K M for an enzyme catalyzed reaction. ( 3 ) k inact and K I can be determined by analyzing k obs as a function of [I], where k obs is the pseudo–first-order rate constant at which the enzyme is inactivated. (D) A k obs v [I] plot when [I]>>K I . The slope of the line gives k inact / K I . The plot was adapted from the reaction of ARS-1620 with KRAS G12C (Hansen et al., 2018). (E) A k obs v [I] plot showing saturation yields individual values for k inact and K I . The plot was adapted from the reaction of MRTX-849 (adagrasib) with KRAS G12C (Fell et al., 2020). In the two-step mechanism, the inhibitor binds reversibly to the enzyme to form a noncovalent complex (EI), defined by the equilibrium dissociation constant K i = k 4 /k 3 . Once bound, the complex can undergo covalent bond formation at a rate governed by the first-order rate constant k inact ( k 5 ). In this model, the macroscopic kinetic inhibition constant K I is defined as the inhibitor concentration that produces half of the maximal inactivation rate ( k inact ) and is distinct from the microscopic dissociation constant K i . The ratio k inact / K I is the second order rate constant that quantifies the kinetic efficiency of an inhibitor to convert free enzyme to covalent adduct and is equivalent to k cat / K M for an enzyme catalyzed reaction. If k 4 >> k 5 then K I will be numerically the same as K i , and formation of the final covalent complex can be separated into a binding event defined by K I followed by the rate of covalent bond formation. However, in most cases the values for the microscopic rate constants are not known and so it is common to use k inact /K I to drive inhibitor discovery and optimization. Note also that k inact is influenced by both the orientation of the reacting groups (e.g. electrophile and nucleophile) with respect to each other, as well as their intrinsic reactivity (sometimes quantified by k chem assessed using (e.g.) reactivity with glutathione). From a modeling perspective, the interplay of these rates determines the extent of steady-state occupancy and the speed of recovery after free drug is cleared. 2.1  Types of covalent inhibition: irreversible and reversible For irreversible covalent inhibitors, k 6 = 0 and time-dependent target occupancy is a function of drug concentration (PK) and the rate at which new protein is synthesized. Once the drug has eliminated, then the duration of action is linked directly to the rate of target turnover (ρ) (Tonge, 2018). For reversible inhibitors, the covalently modified enzyme E-I can regenerate free enzyme and inhibitor where the lifetime of the drug-target complex is quantified by the residence time (τ = 1/k off ), where k off is the rate constant for the formation of free enzyme from E-I and is often equivalent to the value for k 6 . The lifetime of the reversible covalent adduct can range from minutes to days such as the cyanoacrylamide inhibitors of BTK that form reversible Michael adducts with Cys-481 in the BTK active site (Bradshaw et al., 2015). Although reversible covalent adducts eventually breakdown, they may confer clinical durability when lifetimes exceed the dosing interval and also enhance safety if dissociation is rapid from off-targets (Bandyopadhyay & Gao, 2016). From a modeling standpoint, both classes are described by the same two-step framework (Patel, Huma, & Duncan, 2024). 2.2  Why equilibrium parameters fall short: what gets missed? Given the extent of covalent inhibition evolves with time as EI is converted to E-I, the equilibrium potency metrics, such as IC 50 , K d , K i , typically recorded at a particular time point under constant drug concentration, lack the mechanistic depth for covalent inhibitor optimization. Two inhibitors with identical K ᵢ values but different k inact values may appear indistinguishable in a single-point assay, yet their pharmacological performance may diverge sharply. A nanomolar affinity compound with slow intrinsic reactivity may never achieve meaningful occupancy within the pharmacokinetic window, while a weaker binder with rapid k inact can outperform it - albeit with greater off-target risk. Moreover, increasing k inact beyond a certain threshold often comes at the expense of K i , since structural modifications that accelerate chemistry can destabilize the reversible pre-complex. This trade-off reflects a physical ceiling, as k inact /K I cannot exceed the diffusion limit, ensuring that affinity and reactivity cannot be optimized independently. Indeed, it has been convincingly argued that the diffusion limit is 10 6 -10 7 M -1 s -1 rather than the more normally accepted value of 10 8 -10 9 M -1 s -1 (Srinivasan, 2025). Therefore, covalent inhibition must be evaluated kinetically, with early time points dominated by reversible occupancy ( K i ), and later time points approaching an asymptote defined by the k inact . Rational optimization requires balancing both parameters rather than pursuing either in isolation (Copeland, 2013; Faridoon, Zheng, Zhang, & Li, 2025; Srinivasan, 2025). 3 EXPERIMENTAL APPROACHES FOR MODELING COVALENT INHIBITION Time-dependent inhibition is not a new phenomenon (Morrison & Walsh, 1988), and although many slow-onset inhibitors with long residence times on their targets have been discovered, in many cases this has occurred serendipitously in part because of the reliance on the IC 50 values to drive the selection and optimization of drug leads. One principal reason for this is that many assays, such as those used in HTS, are single time-point assays that do not capture time-dependent binding. In addition, the two-step induced fit mechanism for time-dependent reversible inhibition involves a conformational change after formation of the initial EI complex, and so rational design of inhibitors that follow this mechanism requires knowledge of the transition state for the slow step on the binding reaction coordinate (Spagnuolo et al., 2017). However, the interest in covalent inhibitor discovery has forced the community to develop assays that account for the inherent time-dependence in formation of the final enzyme-inhibitor complex (E-I). In addition, while approaches for analyzing slow-onset reversible binding and covalent inhibition follow similar mechanistic frameworks, covalent adducts can be isolated and directly quantified which is a major advantage. Indeed, direct evidence for covalent bond formation is often a critical step in distinguishing between covalent inhibition and reversible inhibitors with long residence times on their targets. Lastly, formation of the final E-I complex does not implicitly require any change in conformation of the target, greatly simplifying design principles for discovering covalent inhibitors. The quantitative modeling of covalent inhibitors must capture both the binding affinity of the initial reversible complex (EI) and the rate of the covalent adduct formation (E-I). This is achieved by individually determining k inact and K I . Here we summarize experimental methods used in covalent inhibitor development and evaluate how they model kinetic mechanisms that control the time-dependent target occupancy of covalent drugs. The methods are summarized in Table 1 . Method Analysis / Parameters n Stage o Comments Equilibrium binding assays a Dose response equation IC 50 , K I Hit ID Endpoint; High throughput Equilibrium; No time-dependence Activity assay b k obs vs [I]; Assume [I]>>[E] K I , k inact , k inact / K I Hit to lead Time-dependent; Deconvolutes binding and reactivity Low throughput; ideally use continuous format Direct adduct detection c %adduct formation vs time; Assume [I]>>[E] k inact / K I Hit to lead Direct confirmation of covalency, Site of modification Low throughput; Linear regime only yields ratio of k inact / K I COOKIE-Pro d Global fitting %TO v [I]; Assume [I]>>[E] K I , k inact , k inact / K I Hit to lead Proteome-wide cellular covalent kinetics Need clickable probe; Permeabilized cells –no protein turnover; IP enrichment; MS Sensitivity limits Rapid Fire- MS e Single-exponential fit of % occupancy vs [I] k inact / K I Hit to lead Direct quantitation of modified peptides; high throughput (~17s/sample) Cannot resolve k inact vs K I ; saturation at high [I]; precipitation limits dDRTC (Diagonal dose-response time course using intact MS) f Single-exponential fit of % covalent occupancy (t) vs [I]; assume [I]≫[E] k inact / K I Hit ID Efficient for covalent fragments; higher throughput than dose dependent time-course. Cannot deconvolute k inact vs K I ; linear regime only; K I ≥ 50 μM Dual time-point IC 50 shift g IC 50 (t1) vs IC 50 (t2) ; qualitative shift ratio Hit ID Rapid diagnostic of time-dependence No kinetic deconvolution; empirical threshold; cannot distinguish slow reversible vs covalent End-point preincubation IC₅₀ curves (EPIC) methods: EPIC-Fit, EPIC-CoRE h Global fit of IC 50 (t) vs [I] K I , k inact , k inact / K I (EPIC-Fit); K I , k 5 , k 6 , K I * (EPIC-CoRE) Hit to lead Full kinetic deconvolution for irreversible or reversible covalent mechanisms Requires numerical global fitting; many timepoints; competitive inhibitor/substrate assumptions SPR i k obs vs [I] in association-phase k inact / K I Hit to lead Activity-independent kinetics; regenerable chip Surface artifacts; immobilization bias; saturation issues Real-time NMR j Single-exponential decay in native peak intensity k mod (apparent modification rate) Hit to lead Direct measurement of covalent adduct formation in cells Cannot obtain k inact or K I ; low throughput; may require truncated protein ITC k Global kinetic competition model (binding + covalent step) from k obs vs [I] k inact / K I Hit to lead Direct catalytic heat-flow readout Low throughput; sample-intensive; limited for fast chemistry Jump dilution assays l Enzyme activity recovery after ≥100× dilution Reversibility (qualitative) Lead optimization; mechanistic Tests reversibility of the E–I complex; estimates adduct lifetime False “irreversible” if k off is slow; rebinding complicates analysis Washout recovery (cell-based) m Recovery of activity /occupancy vs time residence time τ =1/ k rev Lead optimization; mechanistic Tests reversibility in cells; measures duration of reversible adduct Confounded by turnover and rebinding; requires complete washout a (Strelow, 2017). b (Mons et al., 2022; Strelow, 2017). c (Campuzano et al., 2016). d (Lin et al., 2025). e (Li et al., 2022). f (Jeon et al., 2025). g (Berry & Zhao, 2008; Perloff et al., 2009). h (Mader & Keillor, 2024, 2025). i (Zhou, Ye, Pandey, & Du, 2025). j (Zhao, Haga, Tamura, Shimada, & Nishida, 2023). k (Hennecker et al., 2025). l (Bradshaw et al., 2015; Copeland, Basavapathruni, Moyer, & Scott, 2011; Tonge, 2018). m (Bradshaw et al., 2015; Tonge, 2018). n IC 50 is the half maximal inhibition response, k inact is the rate of inactivation, K I is the pseudo-equilibrium dissociation constant, k rev is the rate of dissociation of E-I complex, k off is the rate of dissociation of the EI complex, and ρ is the target turnover rate. o Stage. Some methods are only applicable to a single stage of the discovery pipeline, whereas others, particularly those that quantify binding kinetics, are applicable to multiple stages in agreement current recommendations that target engagement assays provide kinetic parameters that front-load information in early stages and guide compound progression (St John-Campbell & Bhalay, 2025). In the table we have indicated the stage where each method is likely to be most useful. 3.1  Progress Curves from Enzyme Assays: Determination of k inact , K I and k inact /K I Time-dependent enzyme inhibition is ideally quantified by measuring the formation of the final E-I complex as a function of time and inhibitor concentration. This can be accomplished using progress-curve kinetics. Under conditions where the enzyme-catalyzed rate is linear in the absence of inhibitor, then curvature observed in the presence of inhibitor can be used to assess the time-dependence of the reaction (Copeland, 2013; Tonge, 2019). The progress curves are typically fit to a single-exponential function to obtain k obs , the apparent first-order rate constant for conversion of EI to E-I. Plotting k obs v [I] then yields a hyperbolic curve that allows determination of k inact and K I ( Figure 2 ) As with all assays, there are caveats: (i) it may not be possible to monitor the reaction at sufficiently high concentrations of inhibitor (i.e. that spans K I ), so that the k obs v [I] plot is linear and only the ratio of k inact /K I can be extracted; (ii) while progress curve analysis is most commonly performed using activity assays, alternative methods must then be employed to ensure irreversible binding as it is difficult to distinguish an irreversible inhibitor from a reversible inhibitor that dissociates very slowly ( k 6 is small); and (iii) for very rapid inactivators, methods with higher temporal resolution, such as stopped-flow, are needed to capture fast inactivation events. Overall, the strength of this method is the quantitative measurement of inactivation efficiency that enables rank-ordering of SAR beyond IC₅₀ values , whereas the method is relatively low-throughput as continuous assays are often restricted to single compounds or small sets. 3.2  MS-Based Progress Curves: Determination of k inact , K I and k inact /K I Mass spectrometry (MS), through intact-protein MS and peptide-mapping LC–MS/MS analyses, provides direct confirmation of covalent bond formation and site selectivity. These methods have been instrumental in KRAS G12C covalent inhibitor programs, including ARS-510 and ARS-1620, where integrating progress-curve enzymology with mutational and GDP/GTP state controls revealed preferential modification of the GDP bound form (Canon et al., 2019). However, their low-throughput format presents a practical bottleneck as collecting time courses across multiple concentrations is labor-intensive and instrumentally demanding. Consequently, such experiments have typically been limited to qualitative assessments or a few timepoints, rather than a detailed mechanistic characterization (Hallenbeck et al., 2018). Advances in direct MS-based readouts have begun to overcome this limitation by enabling quantitative measurement of kinetic constants from samples analyzed across multiple concentrations and timepoints (Harris et al., 2018). One example is Rapid Fire-MS, in which intact protein is captured on a solid-phase extraction cartridge, desalted, and eluted directly into the mass spectrometer within ~20 s/sample. Instead of peptide mapping, the intact-protein mass is monitored, enabling rapid quantification of covalent adduct formation across hundreds of electrophilic compounds (Campuzano et al., 2016). When multiple inhibitor concentrations and time points are analyzed, the resulting global progress curves (% adduct vs time vs [I]) can be fit to the standard two-step covalent model to extract k inact , K I , and k inact /K I (Mader & Keillor, 2024). Furthermore, the MS approach has recently been adapted to measuring covalent inhibition in digitonin-permeabilized cells using COOKIE-Pro ( CO valent O ccupancy KI netic E nrichment via Pro teomics). In this method, desthiobiotin-probe pulldown of unreacted proteins following the labeling step quantifies kinetic occupancy, enabling direct determination of k inact , K I , and k inact /K I values without the need for purified proteins (Lin et al., 2025). 3.3  MS-Based Determination of k inact /K I Collecting sufficient data to determine k inact and K I is resource intensive and also may face practical hurdles such as the inability to obtain data at [I] >> K I . Thus, to minimize the workflow, methods have been reported that yield just k inact /K I values. For instance, the Rapid Fire-MS intact-protein workflow was paired with targeted multiple reaction monitoring (MRM) to quantify adduct formation with KRAS G12C (Li et al., 2022). MRM involved short LC–MS/MS runs focused on diagnostic peptides containing Cys-12, providing confirmation that the observed intact adducts mapped to the intended nucleophile. Together, this two-tier workflow enabled determination of k inact / K I at a scale feasible for library triage, while maintaining the ability to confirm modification of the target nucleophile. Importantly, the k obs v [I] analysis assumes that [I] >> [E] – i.e. that there is no ligand depletion during the experiment, and the method described by Li et al. does not require this assumption provided that k inact is sufficiently rapid so that the fraction of the reversible EI complex present at any time is very low relative to E-I. In addition, an endogenous cellular target engagement workflow has been developed that uses immunoenrichment followed by UPLC–MRM MS to quantify loss of the KRAS G12C peptide, providing in vitro and in cellulo target-engagement kinetics with increased throughput in a probe-free format (Kantae et al., 2022). This approach reports k mod / K I app values (as a proxy for k inact / K I ) without requiring purified protein, although it depends on the availability of a suitable MS-detectable peptide for the modification site, sufficient target abundance, and high-quality antibodies for immunoenrichment. More recently, another method has been described in which k inact / K I is estimated from the dose and time that yields 50% target occupancy (Jeon et al., 2025). The diagonal dose-response time-course (dDRTC) method is particularly appropriate for early-stage covalent drug discovery when fragments bind weakly to the enzyme i.e. when K I >> 50μM and k inact / K I values range from 1 to 10,000 M⁻¹s⁻¹. 3.4  k inact /K I from Endpoint Assays As discussed previously, equilibrium IC 50 measurements are simple and high throughput but mechanistically misleading for covalent drugs unless the incubation time is reported (Harris et al., 2018; Strelow, 2017). The Dual-timepoint IC 50 shift method: The dual-timepoint method requires IC 50 measurements at two defined pre-incubation times and was standardized in P450 time-dependent inhibition assays (Berry & Zhao, 2008; Perloff et al., 2009). A shift ratio > 1.5 is used to flag time-dependent inhibition, as the apparent potency increases with longer pre-incubation times. Such behavior can arise from either slow-binding reversible interactions or covalent bond formation. However, there is no empirical ΔIC 50 threshold above which a covalent mechanism can be concluded. In kinase systems such as BTK and EGFR, progressive leftward shifts in IC 50 values, when supported by orthogonal evidence such as mass spectrometric confirmation of adduct formation, have been interpreted as consistent with covalent rather than equilibrium binding (Harris et al., 2018). The caveat is that the magnitude of the change in IC 50 depends on both intrinsic reactivity and target recognition, and can reveal qualitative time dependence only when k inact is sufficiently small. Nevertheless, because this method measures the composite effect of binding and inactivation, it cannot yield absolute kinetic parameters such as k inact or K I (Strelow, 2017). Endpoint Preincubation IC 50 Curves (EPIC): If IC₅₀ measurements are made at a series of defined pre-incubation times, the multi-timepoint IC 50 datasets can be globally fit to the two-step covalent model using mathematical tools such as EPIC-Fit ( E ndpoint P re-incubation IC 50) (Mader & Keillor, 2024) and EPIC-CoRE ( E̲ ndpoint P̲ re-incubation I̲C̲ 50- C̲o̲ valent R̲ e̲versible) (Mader & Keillor, 2025). Here, IC 50 values collected over a time course are fit numerically (e.g., in MATLAB, Python, KinTek or DynaFit) to extract k inact , K i , and k inact / K I values (Daryaee & Tonge, 2019; Strelow, 2017). This method preserves much of the mechanistic insight of progress-curves while retaining the scalability of IC 50 screens, enabling time-dependent pre-incubation endpoint assays to be included in the medicinal chemist’s toolbox for rigorous characterization of covalent inhibitors. Overall, the best practice is to state the incubation time for any IC 50 value, and where possible, to obtain multi-timepoint curves for global analysis. (Thorarensen et al., 2021). 3.5  Biophysical Techniques: SPR, NMR and ITC In addition to enzyme assays or MS-based methods, additional methods are available for determining the kinetics of E-I formation. Label-free methods, such as surface plasmon resonance (SPR) has been adapted to covalent inhibitors, where the association phase yields k obs values that can then be analyzed as a function of [I] to give k inact / K I (Zhou et al., 2025). In addition, real-time NMR has been used to track loss of Cys-12 resonances and the growth of adduct peaks during KRAS G12C adduct formation, providing kinetic validation both in vitro and in live cells (Zhao et al., 2023). Lastly, isothermal titration calorimetry (ITC) kinetic competition formats have been developed to extract covalent kinetic parameters from heat-flow profiles generated during catalysis (Hennecker et al., 2025). These biophysical assays are lower throughput than enzymatic progress curves but offer powerful orthogonal confirmation. 3.6  Recovery Assays: Washout or Tracer Displacement Functional washout ( in-cellulo ) or jump-dilution ( in-vitro ) experiments can be used to directly probe reversibility, for example by monitoring the recovery of enzyme activity or the ability of a probe to displace a bound compound from the protein target. These assays can therefore distinguish reversible from irreversible covalent mechanisms (Copeland, 2013; Tonge, 2018). For instance, proteosomes inhibited by bortezomib gradually recover within 48 h after washout, consistent with a slowly reversible covalent mechanism, whereas carfilzomib shows no recovery, supporting irreversible adduct formation (Curran & McKeage, 2009; Kubicki et al., 2022). Additionally, for reversible covalent inhibitors, such assays can quantify covalent residence time (τ = 1/k rev ). Rilzabrutinib, a reversible covalent BTK inhibitor, was characterized using biochemical dissociation and cell washout assays to establish that it forms a covalent but reversible adduct at Cys-481 with a residence time of ~9.6 days. The intact inhibitor was regenerated following proteolysis, confirming reversibility, while cellular washout demonstrated durable yet recoverable BTK occupancy (Bradshaw et al., 2015; Langrish et al., 2021). Nonetheless, because compounds with long residence time or insufficient dilution of the EI complex can appear irreversibly bound under practical assay conditions, careful optimization of assay duration, dilution and controls is essential to correctly evaluate reversibility. In summary, medicinal chemistry campaigns demand assays that balance throughput with mechanistic resolution, and the appropriate method depends on the stage of discovery. During hit identification, rapid and scalable assays such as dual time-point IC 50 shifts, diagonal dose-response time-course (dDRTC), or RapidFire MS/MRM are useful to diagnose covalency and time dependence. These approaches trade mechanistic precision for speed but allow large fragment libraries to be triaged based on k inact / K I rather than static IC 50 values (Mader, Borean, & Keillor, 2024; Mader & Keillor, 2024, 2025). In the hit-to-lead phase, more quantitative frameworks, including enzymatic activity progress curves, intact-protein or cellular MS workflows such as COOKIE-Pro, and multi-timepoint IC 50 fitting (EPIC-Fit or EPIC-CoRe), can resolve k inact and K I independently and support structure-kinetic optimization. Although determination of both k inact and K I is ideal, k inact / K I remains a practical and highly informative measure for ranking chemical series at early stages. Lead optimization then draws on functional assays, reversibility measurements, and proteomics-based turnover analyses (ρ) to relate biochemical kinetics to pharmacological activity. These measurements motivate the broader question of how covalent engagement translates into pharmacological effect, which is the focus of the next section. 4  PHARMACOLOGICAL RELEVANCE OF COVALENT DRUG ACTIVITY A common misconception is that irreversible inhibition inherently guarantees durable pharmacological effect; however, the persistence of pathway suppression depends on the alignment of several rate processes. The translation of covalent inhibition into pharmacological response can be understood by considering how quickly an inhibitor inactivates its target, whether covalent occupancy can accumulate within the pharmacokinetic window, how rapidly new protein enters the system, and what level of target engagement is required for efficacy. These questions map to measurable system parameters that connect microscopic rate constants to macroscopic pharmacology: inactivation efficiency ( k inact /K I ), protein resynthesis rate (ρ), drug pharmacokinetics (PK), and a target vulnerability function that relates occupancy to effect to predict the pharmacological susceptibility of a pathway to partial or complete inhibition (Daryaee & Tonge, 2019; Tonge, 2018) ( Figure 3 ). A central requirement for covalent drug action is that inactivation proceeds rapidly enough for significant occupancy to accumulate before the compound is cleared. For instance, ritlecitinib (PF-06651600) employs transition-state stabilization that accelerates covalent bond formation ( k inact = 2.32 s -1 and k inact /K I = 3.7 x 10 5 M -1 s -1 ) and enables ~ 65-72% of JAK3 in vivo occupancy that results in inhibition of pSTATE5 signaling by 98%, with recovery over 8-24 h following 50 mg once-daily dosing in alopecia areata (Blair, 2023; Martin et al., 2024; Thorarensen et al., 2017). When the rate of inactivation is modest, clearance may outpace covalent bond formation. The KRAS G12C inhibitor sotorasib has k inact /Kᵢ = 9.9 × 10³ M⁻¹ s⁻¹ and a ~ 5 h half-life; consequently, sustained target engagement relies on daily dosing to accumulate sufficient levels of target occupancy (Canon et al., 2019; Li et al., 2022). By contrast, adagrasib ( k inact /Kᵢ 35 mM⁻¹ s⁻¹) achieves durable engagement because its prolonged systemic exposure (~23 h with six- fold accumulation at steady state) compensates for a kinetic efficiency that is only a few fold higher than sotorasib, enabling sustained KRAS G12C occupancy (Janne et al., 2022). Another critical determinant in the translation of extended target occupancy to prolonged drug activity is the rate of target turnover. Covalent inhibition is only as durable as the time required for new protein to replace the modified pool. Kinetic simulations demonstrate that when turnover is rapid, covalent occupancy decays rapidly even for irreversible inhibitors, whereas slow turnover produces sustained inhibition long after drug elimination ( Figure 3 ) (Daryaee & Tonge, 2019). Despite increasing recognition that the apparent “labeled fraction” in covalent cellular profiling assays at any given time reflects a composite of chemical modification and the net influx of unmodified protein, de novo protein synthesis is rarely quantified in covalent programs (Heinzlmeir & Muller, 2022; Yang, Tallman, Porter, & Liebler, 2015). Where turnover has been quantified, the diversity is substantial, as shown in Table 2 : JAK3, MYC, and TYK2 renew rapidly, whereas EGFR, KRAS, and RIPK2 exhibit markedly longer half-lives. Protein turnover is shaped by sequence features, cellular context, disease state, and assay methodology, and can vary widely even within the same protein family (Zhang et al., 2025). BTK exemplifies this biological variability. It turns over slowly in Ramos cells (>24 h), and it is also slow in CLL patient samples, yet the patient-derived resynthesis rate spans nearly an order of magnitude (3.6-31.4%/day) (Alsadhan et al., 2020). This variation reflects inter-individual variability in protein renewal, which can directly influence the pharmacodynamic duration of covalent inhibitors in vivo. Even within the “slow-turnover” regime, several-fold differences across patients may alter how long covalent occupancy is maintained following drug washout above the threshold needed for the therapeutic effect. Such variability underscores the need to measure protein turnover directly in the specific cellular or clinical context used to evaluate a covalent inhibitor, rather than extrapolating rates across systems or individuals. This consideration is important when designing live-cell covalent assays, interpreting occupancy-based chemoproteomic data, or performing multi-hour incubations in intact cellular or patient-derived systems, where turnover-driven heterogeneity may shape the observed duration of action. Figure 3 : The impact of turnover on target occupancy and target vulnerability. (A) Time-dependent target occupancy for an irreversible covalent inhibitor. Once drug has been eliminated, the fractional target occupancy will decrease as new target is synthesized. The simulation is taken from (Daryaee & Tonge, 2019) and assumes that target turnover occurs linearly at different rates (ρ) of 0, 0.005, 0.05 and 0.5 h −1 . (B) Hypothetical target vulnerability plots for low and high vulnerability targets. The vulnerability function is defined by the minimum level of engagement required for any effect to be observed (TO min ) and the level of engagement that leads to the maximal efficacy (TO max ). The third parameter required to define the function is the Hill coefficient or slope factor that determines the steepness of the effect response between TO min and TO max . For the low vulnerability target, the full physiological effect of the drug requires close to 100% target engagement, whereas only ∼35% engagement is needed for the high vulnerability target. The Hill coefficients for the two functions are 4.6 (high) and 16.4 (low). Adapted from (Daryaee & Tonge, 2019). Target Protein family Biological Role Cell type Half-life (t 1/2 ) MYC a Transcription factor Proliferation; oncogenic Various cell lines ~20-30 min MDM2 b RING domain-bearing proteins (MDM2 gene family) E3 ubiquitin ligase; regulator of p53 tumor‐suppressor pathway; tumor angiogenesis hTERT-RPE1 cells 24 min JAK3 c nRTK (JAK family) Cytokine receptor signaling Human PBMCs ~3.5 h TYK2 d nRTK (JAK family) Type I IFN/IL-12 cytokine signaling HeLa cells ~2 h EGFR e RTK Growth factor signaling A431 cells 27.5 h KRAS f Small GTPase (Ras family) Proliferation signaling HCT116 ~22 h BTK g nRTK (Tec-family) B-cell receptor signaling Ramos B cells >24 h RIPK2 h Serine/threonine kinase Immune signaling HEK293 ~50 h a (Ahmadi, Rahimi, Zarandi, Chegeni, & Safa, 2021). b (Fulcher, Sobajima, Batley, Gibbs-Seymour, & Barr, 2025) c (Telliez et al., 2016). d (Siewert, Muller-Esterl, Starr, Heinrich, & Schaper, 1999). e (Greig et al., 2015). f (Hong et al., 2020). g (Alsadhan et al., 2020; Harris et al., 2018). h (Mathieson et al., 2018). Ultimately, the central question becomes what fraction of target must be inactivated to produce effect, the concept of target vulnerability. High-vulnerability targets respond to partial engagement, whereas low-vulnerability ones require near-complete suppression (Tonge, 2018). Therefore, similar inactivation efficiency may yield very different clinical outcomes depending on whether the target is “difficult” (low vulnerability) or “easy” (high vulnerability). BTK exemplifies a low-vulnerability target: >90% occupancy is required for maximal efficacy in a rat collagen-induced arthritis model using the covalent inhibitor CC-292 (Daryaee et al., 2017; Podoll et al., 2019; Smith et al., 2017). Since BTK turns over slowly, irreversible inhibitors such as ibrutinib ( k inact /K ᵢ = 1.0 × 10 6 M⁻¹ s⁻¹) and acalabrutinib ( k inact /K ᵢ = 3.0 × 10 4 M⁻¹ s⁻¹) maintain suppression long after plasma clearance (Liclican et al., 2020). Thus, PD becomes decoupled from PK, but only because both turnover and vulnerability favor durable engagement. In summary, integrating inactivation efficiency, protein turnover, pharmacokinetic exposure and target vulnerability can enable covalent discovery programs to prioritize compounds based not only on affinity or potency but on the quantitative integration of chemistry and system biology. 5  CONCLUSION Covalent inhibition is inherently time-dependent, requiring explicit characterization of both binding (K I ) and chemical reactivity ( k inact ), as well as reversibility when relevant ( k rev ). But these microscopic rates alone do not dictate outcome. Durable efficacy emerges only when chemistry, turnover, vulnerability, and PK exposure are aligned: rapid chemistry for fast-turnover proteins, prolonged exposure for slow inactivators, or low turnover to completely decouple PK from PD. Thus covalent inhibitor discovery programs must move beyond static IC 50 values and adopt kinetic frameworks that report k inact /K I , residence time, and turnover-driven occupancy. At a minimum, incubation time should be reported together with IC 50 values; ideally, multi-timepoint assays, progress-curve fits, intact-MS confirmation, and washout recovery are combined to build a mechanistic profile. Without such rigor, both false negatives (discarding weak binders with rapid chemistry) and false positives (advancing high affinity but kinetically poor compounds) will continue to undermine translation. In sum, the modern paradigm for covalent drugs is not “covalent versus reversible” but “kinetics versus equilibrium”. By embracing kinetic modeling and integrating it with turnover, vulnerability, and PK, the next generation of covalent inhibitors can be designed with precision, achieving durable efficacy where static potency metrics would have failed. Ultimately, kinetic modeling is not an accessory to covalent drug discovery; it is the framework that can translate chemistry into pharmacology. AUTHOR CONTRIBUTIONS M. Ali: Conceptualization (equal);writing—original draft (equal); writing—review and editing (equal). P. Tonge: Conceptualization (equal);writing—original draft (equal); writing—review and editing (equal). ACKNOWLEDGEMENTS The research was supported by the National Institutes of Health grant GM149297 to PJT and T32GM136572 to MIA. CONFLICT OF INTEREST STATEMENT The authors declare no conflicts of interest. DATA AVAILABILITY STATEMENT N/A-Review. ORCID Madeeha Ali https://orcid.org/0009-0005-9329-3475 Peter Tonge https://orcid.org/0000-0003-1606-3471 REFERENCES Advani, R. H., Buggy, J. J., Sharman, J. P., Smith, S. M., Boyd, T. E., Grant, B., … Fowler, N. H. (2013). Bruton tyrosine kinase inhibitor ibrutinib (PCI-32765) has significant activity in patients with relapsed/refractory B-cell malignancies. J Clin Oncol, 31 (1), 88-94. doi:10.1200/JCO.2012.42.7906Ahmadi, S. E., Rahimi, S., Zarandi, B., Chegeni, R., & Safa, M. (2021). MYC: a multipurpose oncogene with prognostic and therapeutic implications in blood malignancies. J Hematol Oncol, 14 (1), 121. doi:10.1186/s13045-021-01111-4Alsadhan, A., Cheung, J., Gulrajani, M., Gaglione, E. M., Nierman, P., Hamdy, A., … Wiestner, A. (2020). Pharmacodynamic Analysis of BTK Inhibition in Patients with Chronic Lymphocytic Leukemia Treated with Acalabrutinib. Clin Cancer Res, 26 (12), 2800-2809. doi:10.1158/1078-0432.CCR-19-3505Bandyopadhyay, A., & Gao, J. (2016). Targeting biomolecules with reversible covalent chemistry. Curr Opin Chem Biol, 34 , 110-116. doi:10.1016/j.cbpa.2016.08.011Berry, L. M., & Zhao, Z. (2008). An examination of IC50 and IC50-shift experiments in assessing time-dependent inhibition of CYP3A4, CYP2D6 and CYP2C9 in human liver microsomes. Drug Metab Lett, 2 (1), 51-59. doi:10.2174/187231208783478407Blair, H. A. (2023). Ritlecitinib: First Approval. Drugs, 83 (14), 1315-1321. doi:10.1007/s40265-023-01928-yBradshaw, J. M., McFarland, J. M., Paavilainen, V. O., Bisconte, A., Tam, D., Phan, V. T., … Taunton, J. (2015). Prolonged and tunable residence time using reversible covalent kinase inhibitors. Nat Chem Biol, 11 (7), 525-531. doi:10.1038/nchembio.1817Campuzano, I. D., San Miguel, T., Rowe, T., Onea, D., Cee, V. J., Arvedson, T., & McCarter, J. D. (2016). High-Throughput Mass Spectrometric Analysis of Covalent Protein-Inhibitor Adducts for the Discovery of Irreversible Inhibitors: A Complete Workflow. J Biomol Screen, 21 (2), 136-144. doi:10.1177/1087057115621288Canon, J., Rex, K., Saiki, A. Y., Mohr, C., Cooke, K., Bagal, D., … Lipford, J. R. (2019). The clinical KRAS(G12C) inhibitor AMG 510 drives anti-tumour immunity. Nature, 575 (7781), 217-223. doi:10.1038/s41586-019-1694-1Copeland, R. A. (2013). Evaluation of enzyme inhibitors in drug discovery: a guide for medicinal chemists and pharmacologists (2nd edn). Hoboken, N.J.: Wiley.Copeland, R. A., Basavapathruni, A., Moyer, M., & Scott, M. P. (2011). Impact of enzyme concentration and residence time on apparent activity recovery in jump dilution analysis. Anal Biochem, 416 (2), 206-210. doi:10.1016/j.ab.2011.05.029Curran, M. P., & McKeage, K. (2009). Bortezomib: a review of its use in patients with multiple myeloma. Drugs, 69 (7), 859-888. doi:10.2165/00003495-200969070-00006Daryaee, F., & Tonge, P. J. (2019). Pharmacokinetic-pharmacodynamic models that incorporate drug-target binding kinetics. Curr Opin Chem Biol, 50 , 120-127. doi:10.1016/j.cbpa.2019.03.008Daryaee, F., Zhang, Z., Gogarty, K. R., Li, Y., Merino, J., Fisher, S. L., & Tonge, P. J. (2017). A quantitative mechanistic PK/PD model directly connects Btk target engagement and in vivo efficacy. Chem Sci, 8 (5), 3434-3443. doi:10.1039/c6sc03306gDe Cesco, S., Kurian, J., Dufresne, C., Mittermaier, A. K., & Moitessier, N. (2017). Covalent inhibitors design and discovery. Eur J Med Chem, 138 , 96-114. doi:10.1016/j.ejmech.2017.06.019De Vita, E. (2021). 10 years into the resurgence of covalent drugs. Future Med Chem, 13 (2), 193-210. doi:10.4155/fmc-2020-0236Faridoon, Zheng, J., Zhang, G., & Li, J. J. (2025). Key advances in the development of reversible covalent inhibitors. Future Med Chem, 17 (4), 389-392. doi:10.1080/17568919.2025.2453407Fell, J. B., Fischer, J. P., Baer, B. R., Blake, J. F., Bouhana, K., Briere, D. M., … Marx, M. A. (2020). Identification of the Clinical Development Candidate MRTX849, a Covalent KRAS(G12C) Inhibitor for the Treatment of Cancer. J Med Chem, 63 (13), 6679-6693. doi:10.1021/acs.jmedchem.9b02052Fulcher, L. J., Sobajima, T., Batley, C., Gibbs-Seymour, I., & Barr, F. A. (2025). MDM2 functions as a timer reporting the length of mitosis. Nat Cell Biol, 27 (2), 262-272. doi:10.1038/s41556-024-01592-8Greig, M. J., Niessen, S., Weinrich, S. L., Feng, J. L., Shi, M., & Johnson, T. O. (2015). Effects of Activating Mutations on EGFR Cellular Protein Turnover and Amino Acid Recycling Determined Using SILAC Mass Spectrometry. Int J Cell Biol, 2015 , 798936. doi:10.1155/2015/798936Hallenbeck, K. K., Davies, J. L., Merron, C., Ogden, P., Sijbesma, E., Ottmann, C., … Arkin, M. R. (2018). A Liquid Chromatography/Mass Spectrometry Method for Screening Disulfide Tethering Fragments. SLAS Discov, 23 (2), 183-192. doi:10.1177/2472555217732072Hansen, R., Peters, U., Babbar, A., Chen, Y., Feng, J., Janes, M. R., … Zarrinkar, P. P. (2018). The reactivity-driven biochemical mechanism of covalent KRAS(G12C) inhibitors. Nat Struct Mol Biol doi:10.1038/s41594-018-0061-5Harris, C. M., Foley, S. E., Goedken, E. R., Michalak, M., Murdock, S., & Wilson, N. S. (2018). Merits and Pitfalls in the Characterization of Covalent Inhibitors of Bruton’s Tyrosine Kinase. SLAS Discov, 23 (10), 1040-1050. doi:10.1177/2472555218787445Heinzlmeir, S., & Muller, S. (2022). Selectivity aspects of activity-based (chemical) probes. Drug Discov Today, 27 (2), 519-528. doi:10.1016/j.drudis.2021.10.021Hennecker, C., Venegas, F., Wang, G., Stille, J., Milaczewska, A., Moitessier, N., & Mittermaier, A. (2025). Mechanistic Characterization of Covalent Enzyme Inhibition by Isothermal Titration Calorimetry Kinetic Competition (ITC-KC). Anal Chem, 97 (12), 6368-6381. doi:10.1021/acs.analchem.4c04003Hong, D. S., Fakih, M. G., Strickler, J. H., Desai, J., Durm, G. A., Shapiro, G. I., … Li, B. T. (2020). KRAS(G12C) Inhibition with Sotorasib in Advanced Solid Tumors. N Engl J Med, 383 (13), 1207-1217. doi:10.1056/NEJMoa1917239Janne, P. A., Riely, G. J., Gadgeel, S. M., Heist, R. S., Ou, S. I., Pacheco, J. M., … Spira, A. I. (2022). Adagrasib in Non-Small-Cell Lung Cancer Harboring a KRAS(G12C) Mutation. N Engl J Med, 387 (2), 120-131. doi:10.1056/NEJMoa2204619Jeon, J., Kholodar, Tran, B. H., Mallinger, K. E., Erlanson, D. A., & Everley, R. A. (2025). A Practical Method for Determining the Rate of Covalent Modification of Fragments and Leads. Nat Commun in press. Kalgutkar, A. S., & Dalvie, D. K. (2012). Drug discovery for a new generation of covalent drugs. Expert Opin Drug Discov, 7 (7), 561-581. doi:10.1517/17460441.2012.688744Kantae, V., Polanski, R., Lewis, H. J., Haider, A., Barratt, D., & Srinivasan, B. (2022). Accelerating the Validation of Endogenous On-Target Engagement and In Cellulo Kinetic Assessment for Covalent Inhibitors of KRAS(G12C) in Early Drug Discovery. ACS Chem Biol doi:10.1021/acschembio.2c00483Kubicki, T., Bednarek, K., Kostrzewska-Poczekaj, M., Luczak, M., Lewandowski, K., Gil, L., … Dytfeld, D. (2022). Bortezomib- and carfilzomib-resistant myeloma cells show increased activity of all three arms of the unfolded protein response. Am J Cancer Res, 12 (7), 3280-3293. Langrish, C. L., Bradshaw, J. M., Francesco, M. R., Owens, T. D., Xing, Y., Shu, J., … Nunn, P. A. (2021). Preclinical Efficacy and Anti-Inflammatory Mechanisms of Action of the Bruton Tyrosine Kinase Inhibitor Rilzabrutinib for Immune-Mediated Disease. J Immunol, 206 (7), 1454-1468. doi:10.4049/jimmunol.2001130Li, K. S., Quinn, J. G., Saabye, M. J., Guerrero, J. F. S., Nonomiya, J., Lian, Q., … Mulvihill, M. M. (2022). High-Throughput Kinetic Characterization of Irreversible Covalent Inhibitors of KRAS(G12C) by Intact Protein MS and Targeted MRM. Anal Chem, 94 (2), 1230-1239. doi:10.1021/acs.analchem.1c04463Liclican, A., Serafini, L., Xing, W., Czerwieniec, G., Steiner, B., Wang, T., … Feng, J. Y. (2020). Biochemical characterization of tirabrutinib and other irreversible inhibitors of Bruton’s tyrosine kinase reveals differences in on - and off - target inhibition. Biochim Biophys Acta Gen Subj , 129531. doi:10.1016/j.bbagen.2020.129531Lin, H., Yang, B., Ding, L., Yang, Y. Y., Holt, M. V., Jung, S. Y., … Wang, J. (2025). COOKIE-Pro: covalent inhibitor binding kinetics profiling on the proteome scale. Nat Commun, 16 (1), 8373. doi:10.1038/s41467-025-63491-2Mader, L. K., Borean, J. E., & Keillor, J. W. (2024). A practical guide for the assay-dependent characterisation of irreversible inhibitors. RSC Med Chem, 16 (1), 63-76. doi:10.1039/d4md00707gMader, L. K., & Keillor, J. W. (2024). Fitting of k(inact) and K(I) Values from Endpoint Pre-incubation IC(50) Data. ACS Med Chem Lett, 15 (5), 731-738. doi:10.1021/acsmedchemlett.4c00054Mader, L. K., & Keillor, J. W. (2025). Methods for kinetic evaluation of reversible covalent inhibitors from time-dependent IC(50) data. RSC Med Chem, 16 (6), 2517-2531. doi:10.1039/d5md00050eMartin, D. A., Telliez, J. B., Pleasic-Williams, S., Zhang, Y., Tierney, B., Blatnik, M., … Karanam, A. (2024). Target Occupancy and Functional Inhibition of JAK3 and TEC Family Kinases by Ritlecitinib in Healthy Adults: An Open-Label, Phase 1 Study. J Clin Pharmacol, 64 (1), 67-79. doi:10.1002/jcph.2347Mathieson, T., Franken, H., Kosinski, J., Kurzawa, N., Zinn, N., Sweetman, G., … Savitski, M. M. (2018). Systematic analysis of protein turnover in primary cells. Nat Commun, 9 (1), 689. doi:10.1038/s41467-018-03106-1Mons, E., Roet, S., Kim, R. Q., & Mulder, M. P. C. (2022). A Comprehensive Guide for Assessing Covalent Inhibition in Enzymatic Assays Illustrated with Kinetic Simulations. Curr Protoc, 2 (6), e419. doi:10.1002/cpz1.419Morrison, J. F., & Walsh, C. T. (1988). The behavior and significance of slow-binding enzyme inhibitors. Adv Enzymol Relat Areas Mol Biol, 61 , 201-301. doi:10.1002/9780470123072.ch5Nicola, G., Tomberg, J., Pratt, R. F., Nicholas, R. A., & Davies, C. (2010). Crystal structures of covalent complexes of beta-lactam antibiotics with Escherichia coli penicillin-binding protein 5: toward an understanding of antibiotic specificity. Biochemistry, 49 (37), 8094-8104. doi:10.1021/bi100879mOuerdani, A., Valenzuela, B., Treijtel, N., Haddish-Berhane, N., Desphande, S., Srinivasan, S., … Perez Ruixo, J. J. (2025). Evaluation of Bruton’s Tyrosine Kinase (BTK) inhibition with alternative doses of ibrutinib in subjects with Chronic Lymphocytic Leukemia (CLL). Cancer Chemother Pharmacol, 95 (1), 38. doi:10.1007/s00280-025-04753-0Patel, D., Huma, Z. E., & Duncan, D. (2024). Reversible Covalent Inhibition─Desired Covalent Adduct Formation by Mass Action. ACS Chem Biol, 19 (4), 824-838. doi:10.1021/acschembio.3c00805Perloff, E. S., Mason, A. K., Dehal, S. S., Blanchard, A. P., Morgan, L., Ho, T., … Stresser, D. M. (2009). Validation of cytochrome P450 time-dependent inhibition assays: a two-time point IC50 shift approach facilitates kinact assay design. Xenobiotica, 39 (2), 99-112. doi:10.1080/00498250802638155Podoll, T., Pearson, P. G., Evarts, J., Ingallinera, T., Bibikova, E., Sun, H., … Slatter, J. G. (2019). Bioavailability, Biotransformation, and Excretion of the Covalent Bruton Tyrosine Kinase Inhibitor Acalabrutinib in Rats, Dogs, and Humans. Drug Metab Dispos, 47 (2), 145-154. doi:10.1124/dmd.118.084459Potashman, M. H., & Duggan, M. E. (2009). Covalent modifiers: an orthogonal approach to drug design. J Med Chem, 52 (5), 1231-1246. doi:10.1021/jm8008597Siewert, E., Muller-Esterl, W., Starr, R., Heinrich, P. C., & Schaper, F. (1999). Different protein turnover of interleukin-6-type cytokine signalling components. Eur J Biochem, 265 (1), 251-257. doi:10.1046/j.1432-1327.1999.00719.xSingh, J., Petter, R. C., Baillie, T. A., & Whitty, A. (2011). The resurgence of covalent drugs. Nat Rev Drug Discov, 10 (4), 307-317. doi:10.1038/nrd3410Smith, P. F., Krishnarajah, J., Nunn, P. A., Hill, R. J., Karr, D., Tam, D., … Gourlay, S. G. (2017). A phase I trial of PRN1008, a novel reversible covalent inhibitor of Bruton’s tyrosine kinase, in healthy volunteers. Br J Clin Pharmacol, 83 (11), 2367-2376. doi:10.1111/bcp.13351Spagnuolo, L. A., Eltschkner, S., Yu, W., Daryaee, F., Davoodi, S., Knudson, S. E., … Tonge, P. J. (2017). Evaluating the contribution of transition-state destabilization to changes in the residence time of triazole-based InhA inhibitors. J Am Chem Soc, 139 (9), 3417-3429. doi:10.1021/jacs.6b11148Srinivasan, B. (2025). Diffusion Limit and the Reactivity/Affinity Conundrum: Implications for Optimization and Hit Finding for Irreversible Modulators. J Med Chem, 68 (13), 13137-13147. doi:10.1021/acs.jmedchem.4c02863St John-Campbell, S., & Bhalay, G. (2025). Target Engagement Assays in Early Drug Discovery. J Med Chem, 68 (12), 12331-12368. doi:10.1021/acs.jmedchem.4c03115Strelow, J. M. (2017). A Perspective on the Kinetics of Covalent and Irreversible Inhibition. SLAS Discov, 22 (1), 3-20. doi:10.1177/1087057116671509Telliez, J. B., Dowty, M. E., Wang, L., Jussif, J., Lin, T., Li, L., … Thorarensen, A. (2016). Discovery of a JAK3-Selective Inhibitor: Functional Differentiation of JAK3-Selective Inhibition over pan-JAK or JAK1-Selective Inhibition. ACS Chem Biol, 11 (12), 3442-3451. doi:10.1021/acschembio.6b00677Thorarensen, A., Balbo, P., Banker, M. E., Czerwinski, R. M., Kuhn, M., Maurer, T. S., … Wittwer, A. J. (2021). The advantages of describing covalent inhibitor in vitro potencies by IC(50) at a fixed time point. IC(50) determination of covalent inhibitors provides meaningful data to medicinal chemistry for SAR optimization. Bioorg Med Chem, 29 , 115865. doi:10.1016/j.bmc.2020.115865Thorarensen, A., Dowty, M. E., Banker, M. E., Juba, B., Jussif, J., Lin, T., … Telliez, J. B. (2017). Design of a Janus Kinase 3 (JAK3) Specific Inhibitor 1-((2S,5R)-5-((7H-Pyrrolo[2,3-d]pyrimidin-4-yl)amino)-2-methylpiperidin-1-yl)prop-2-en-1-one (PF-06651600) Allowing for the Interrogation of JAK3 Signaling in Humans. J Med Chem, 60 (5), 1971-1993. doi:10.1021/acs.jmedchem.6b01694Tonge, P. J. (2018). Drug-target kinetics in drug discovery. ACS Chem Neurosci, 9 (1), 29-39. doi:10.1021/acschemneuro.7b00185Tonge, P. J. (2019). Quantifying the Interactions between Biomolecules: Guidelines for Assay Design and Data Analysis. ACS Infect Dis, 5 (6), 796-808. doi:10.1021/acsinfecdis.9b00012Vane, J. R. (1971). Inhibition of prostaglandin synthesis as a mechanism of action for aspirin-like drugs. Nat New Biol, 231 (25), 232-235. doi:10.1038/newbio231232a0Yang, J., Tallman, K. A., Porter, N. A., & Liebler, D. C. (2015). Quantitative chemoproteomics for site-specific analysis of protein alkylation by 4-hydroxy-2-nonenal in cells. Anal Chem, 87 (5), 2535-2541. doi:10.1021/ac504685yZhang, J. D., Parrott, N. J., Ebeling, M., Eigenmann, M., Umehara, K., & Fischer, H. (2025). Leveraging Protein Turnover in Drug Discovery: A Practical Guide. In 2025 Medicinal Chemistry Reviews , (pp. 429-467).Zhao, Q., Haga, R., Tamura, S., Shimada, I., & Nishida, N. (2023). Real-time monitoring of the reaction of KRAS G12C mutant specific covalent inhibitor by in vitro and in-cell NMR spectroscopy. Sci Rep, 13 (1), 19253. doi:10.1038/s41598-023-46623-wZhou, W., Ye, X., Pandey, S., & Du, X. (2025). Application and Cross-Validation of a High-Throughput SPR Method for Characterizing Covalent Binding Ligands. ACS Omega, 10 (27), 29637-29646. doi:10.1021/acsomega.5c03239 Supplementary Material File (image1.emf) Download 94.37 KB File (image2.emf) Download 275.71 KB File (image3.emf) Download 522.23 KB Information & Authors Information Version history V1 Version 1 10 December 2025 Peer review timeline Published British Journal of Pharmacology Version of Record 12 Feb 2026 Published Copyright This work is licensed under a Non Exclusive No Reuse License. Collection British Journal of Pharmacology Authors Affiliations Madeeha I. Ali Stony Brook University View all articles by this author Peter J. Tonge [email protected] Stony Brook University View all articles by this author Metrics & Citations Metrics Article Usage 879 views 229 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Madeeha I. Ali, Peter J. Tonge. Only Time Will Tell: Modeling the Kinetics of Covalent Inhibitors. Authorea . 10 December 2025. DOI: https://doi.org/10.22541/au.176537376.66820079/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.176537376.66820079/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'9ffda1a6bcaf300f',t:'MTc3OTQ3MTg5Mw=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();