Research progress in electrochemical determination of Janus kinase inhibitors and proposals for amplifying clinical applicability

Abstract

Janus kinase inhibitors (JAKis) represent a class of drugs that treat inflammatory and autoimmune diseases by blocking Janus kinase (JAK) enzymes. Monitoring of precise drug concentration ensures therapeutic effect while minimizing the risk of toxicity. In parallel to conventional methods based on high performance liquid chromatography, liquid chromatography − mass spectrometric or spectrometric methods, sensitive electrochemical methods for the detection of JAKis have been developed in very recent years. The procedures utilize conventional bare glassy carbon electrode or boron-doped diamond electrode and, particularly, chemically modified electrodes incorporating nanomaterials and their composites as powerful catalysts as well as imprinted polymers. The linear concentration ranges and limits of detection achieve very low 10^− 9 M to 10^− 12 M (µg to ng/mL) values, matching clinically relevant drug levels and are applied to analysis of biological matrices and pharmaceutical products. In this study, the concentration ranges obtained for individual JAKis are presented and compared with those of conventional methods. The manuscript covers the years 2022 to 2025 highlighting the JAKis electrochemical detection as a new topic. The paper aims to address both current trends and future potential in the development of novel sensors and procedures for the JAK inhibitors detection directed to a real-time point-of-care analysis enabling personalized therapeutic drug monitoring. Advantages and disadvantages of electrochemical approaches for the JAKis assay in clinical settings are critically evaluated. To facilitate the development of more reliable, robust and clinically applicable electrochemical methods, a few recommendations that future studies should follow are proposed.

Graphical abstract

Supplementary Information

The online version contains supplementary material available at 10.1007/s00604-026-07854-6.

Introduction

Janus kinases (JAKs) are a subclass of non-receptor cytosolic tyrosine kinases (subtypes JAK1, JAK2, JAK3, and tyrosine-protein kinase TYK2) that play a crucial role in immune system signalling, particularly in the activation of signal transducers and activators of transcription (STAT) proteins [1]. Upon activation by JAK proteins, the STAT proteins translocate to the cell nucleus, where they stimulate the transcription of target genes. The JAK2/STAT3 is frequently detected in varieties of tumours and involved in oncogenesis, angiogenesis, and metastasis of many cancer diseases that are usually refractory to the standard chemotherapy. The JAK3/STAT3 pathway emerged as a new site for the development of novel anti-tumour agents [2]. In comparison, tyrosine kinases represent a broad functional group of enzymes that catalyse the phosphorylation of tyrosine residues on proteins, i.e., not all tyrosine kinases are JAKs. Janus kinases were initially named “just another kinase” 1 and 2 (since they were just two of many discoveries in a PCR-based screen of kinases) but were ultimately published as “Janus kinase”. The name is taken from the two-faced god of beginnings, endings and duality in Roman religion, Janus, because the JAKs possess two nearly identical phosphate-transferring domains. One domain exhibits the kinase activity, while the other negatively regulates the kinase activity of the first.

JAK inhibitors (JAKis) are small molecules representing a class of drugs approved for the treatment of various chronic inflammatory and immune-mediated conditions, including rheumatoid arthritis, psoriatic arthritis, juvenile idiopathic arthritis, atopic dermatitis (eczema), alopecia areata, axial spondyloarthritis, ulcerative colitis and Crohn’s disease, and myeloproliferative disorders like myelofibrosis. One crucial aspect of their therapeutic efficacy is the disruption of JAK enzymatic function. JAK inhibitors can be divided into two generations. The first-generation includes small molecules such as baricitinib and tofacitinib, which act as non-selective inhibitors of JAKs. On the other hand, second-generation drugs such as filgotinib and upadacitinib have selective inhibitory activity against JAKs [3]. JAKis may also be classified based on their binding mode and the type of interactions with the amino acids in JAKs into reversible (competitive) and irreversible (covalent) inhibitors [1]. Partial list of JAK inhibitors approved by US Food and Drug Administration (FDA) and/or European Medicines Agency (EMA) for various autoimmune diseases include (in alphabetic order): Abrocitinib, Baricitinib, Delgocitinib, Deucravacitinib, Fedratinib, Filgotinib, Oclacitinib, Pacritinib, Peficitinib, Ritlecitinib, Ruxolitinib, Tofacitinib, and Upadacitinib. More, momelotinib, golidocitinib, and deuruxolitinib have been also indicated as JAKis [4–8]. Chemical structure of these drugs together with indication of their use in medical treatment and side effects are summarized in Table 1. The chemical structure is responsible for a distinctive mode of binding within the catalytic site of the target JAK and gives rise to distinct pharmacological characteristics. The available agents have differing selectivity for JAK isoforms as well as off-target effects against non-JAKs [13].

Table 1.

Summarization of basic characteristics of the JAKi drugs, JAK selectivity, years of approval, disease indications, and side effects [9, 10]

^a[11], ^b[12]

Various analytical methods have been developed for the detection of the JAK inhibitors. Among them, examples of high-performance liquid chromatography (HPLC) method can be named for baricitinib [14–16], filgotinib [17], oclacitinib [18], ruxolitinib [19, 20], tofacitinib [21–24]. Liquid chromatography − mass spectrometric tandem techniques (LC − MS, LC–MS/MS, and UPLC-MS/MS) are represented by several recent works [25–36]. Fluorescence spectrometry [37–39], and UV-Vis spectrometry [37, 40–42] were utilized for the detection as well. While these methods are effective and sensitive, they are not environmentally friendly due to the extensive use of organic solvents and often involve high costs what restricts their use in many clinical laboratories. Additionally, they require long analysis time and expertise in handling analytical devices.

In contrast, electrochemical methods preferred for analysing electroactive compounds represent a more environmentally friendly and powerful approach as they typically require low sample volume, short analysis time, low costs, and simple operation. For the JAKis detection, the application of electrochemical methods is rather novel and mostly relies on modern efficient working electrodes. Individual JAKis possess well-shaped anodic responses considering the 2e⁻ and 2 H⁺ electrochemical oxidation at the pyrrolopyrimidine moiety or 1e⁻ and 1 H⁺ oxidation out of this moiety (Fig. 1). This paper covers the progress in the field of electrochemical detection of individual JAKis within the years 2022 to 2025, gathering the pioneering reports on the topic, and highlights trends in both, the fabrication of sensors as well as development, validation and application of analytical procedures. To our best knowledge there is no general report in literature which covers the novelty in electroanalytical sensors and procedures for JAKis detection directed to analysis of clinically relevant samples. Therefore, advantages and disadvantages of electrochemical approaches for the assay of JAKis in clinical settings are critically evaluated and some recommendations for future studies are proposed.

Fig. 1.

Schemes of possible electrochemical oxidation of (A) ruxolitinib, (B) tofacitinib, and (C) ritlecitinib. Reprinted with permission from (A) [43], (B) [44], and (C) [45]

Recent advances in arrangement of electrochemical experimental setup

Sensors

Working electrode material plays crucial role in analytical performance of the electrochemical methods. Conventional bare electrodes such as glassy carbon electrode (GCE), boron-doped diamond electrode (BDDE), carbon paste electrode (CPE), and screen-printed carbon electrodes as well as chemically modified electrodes are used as the voltammetric electrodes [46, 47]. Regarding low clinically relevant JAKis concentrations, the electrocatalytic behaviour of the electrochemical sensor surface is of particular interest to reach necessary sensitivity and limits of the detection. Here, effective chemical modification of an electrode surface represents particularly an application of various metal-based or carbon-based nanomaterials (NMs), some of them prepared by a green synthesis using plants or plant extracts. Layers of nanomaterials are formed as single deposits (metal and metal oxide nanoparticles (NPs) and nanowires, transition metal carbides, nitrides or carbonitrides called MXenes, carbon nanotubes (CNTs), and graphene-based materials) or binary or ternary composites of nanomaterials [48, 49]. A nanocomposite is characterized as a multiphase solid material in which one of the phases has one, two or three dimensions of less than 100 nm or structures having nanoscale repeat distances between the different phases that make up the material. The synergistic interaction of different components results in the superior properties of a nanocomposite, which are beyond the properties of each individual component, and thereby significantly promoting the development [50].

At advanced approaches the nanocomposites typically combine two metal-oxides, metal oxides and carbon-based NMs, metal complex compounds, or magnetic nanoparticles, and metal–organic frameworks (MOFs) [51]. Triple and multiple nanocomposites are formed, for instance, by graphene nanosheets, graphene oxide (GO), CNTs and others with the metal oxides or fullerenes [52]. Regarding the morphology, the building NMs are considered as 0D (NPs, quantum dots, carbon black), 1D (wires, tubes), 2D (metal-based nanosheets: transition metal dichalcogenides MoS₂, transition metal–carbides/nitrides (MXenes)), and 3D structures. The 3D interconnected hierarchical structures of the graphene nanocomposites with metal or metal oxide NPs, and polymer hydrogel networks can facilitate the diffusion of analytes and other molecules [53]. Progress continues by the construction of multi-layer sensor interfaces. Carbon nanotubes, particularly as functionalized multi-walled carbon nanotubes (f.MWCNTs) find new applications as electro-conductive materials with large specific surface area and effective adsorption sites. Metal–organic frameworks (MOFs) attract wide attention having not only high specific surface area, but also rich pore structure and active metal sites.

Another advanced way of ensuring high sensor selectivity and sensitivity is the use of molecularly imprinted polymers (MIPs) as artificial recognition materials. They represent synthetic engineered materials with a predetermined affinity for an analyte via cavities formed during the synthesis of MIP in the presence of a chosen small template molecule. After removing the template, the cavities remain available for binding of the target analyte on a lock-and-key principle. Consequently, the MIP-based electrochemical sensors should be highly effective tools even in complex matrices of biological samples. As MIPs exhibit low electrical conductivity, nanomaterials such as metalic NPs, CNTs, graphene or graphene oxide are employed to enhance their conductivity at electrochemical sensors [54]. Examples of novel constructions within a synergistic strategy include a combination MOF-on-MOF such as the MOF-supported Au NPs-based molecular imprinted electrochemical sensor [55]. Nevertheless, the synthesized MIPs generally have also some problems such as few and uneven imprinting sites, high density of non-selective binding sites, incomplete removal of template molecules, slow mass transfer rate, and consumption of organic solvents for the analyte elution from the cavity. Recently, the application of chemometrics tools in optimisation problems during the development and analytical assessment of electrochemical MIP- based sensors have been reviewed [56]. Multi-template molecular imprinting consists of the simultaneous use of two or more templates of the target to be determined to generate multiple types of recognition sites with a great potential for the simultaneous recognition, enrichment, and analysis of numerous target components, thus overcoming the shortcomings of single-template MIP that recognizes only one or one type of target component [57].

For the typical detection of JAKis, Fig. 2 the MIP function relies on the occupation of its specific cavities with JAK inhibitor molecules during incubation. During the voltammetric measurement with the MIP-based sensors, a decrease in the peak current of the redox indicator [Fe(CN)₆]^3–/4– present in solution phase is typically evaluated instead of direct electrochemical measurements of the drug signal. Thus, the sensor response to JAKi depends on blocking the transport of [Fe(CN)₆]^3–/4– moieties to the electrode surface modified by MIP (Fig. 3b). As a result, the response is referred to as the differences of the redox indicator currents between the analyte (i.e., JAK inhibitor) extraction process and rebinding process. JAKi removal time and rebinding time should be experimentally optimized, for instance, to 10 min and 20 min, respectively [58]. Advanced assays in (bio)chemically modified electrodes represented by the biosensors should be mentioned as well, e.g. recently published enzyme biosensors for the pyruvate kinase inhibitors including ruxolitinib [59].

Fig. 2.

Scheme of molecularly imprinted polymer function at the electrode surface with corresponding voltammetric and impedimetric response records

Fig. 3.

Examples of electrochemical response calibration. (a) DPV records of Ni-Co-MW/GCE with additions of RITL in 0.1 M BR pH 6.0. Reprinted with permission from [45], (b) DPV records of GCE/MIP@PHEMA-ThyM with standard addition of RUX and concentration relatioship at GCE/MIP@PHEMA-ThyM and GCE/NIP@PHEMA-ThyM in spiked serum samples. Reprinted with permission from [60], and (c) DPV records of Co_6.8Se₈@NPC modified GCE with additions of UPA in 0.1 M PBS pH 2.0. Reprinted with permission from [61]

Prior to an application, the developed electrochemical sensors use to be widely characterized by numerous techniques of morphological analysis, X-ray diffraction analysis, Fourier-transform infrared spectroscopy (FTIR) analysis and the electrochemical methods of cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) mostly performed in 5.0 × 10^{− 3} M [Fe(CN)₆]^3−/4− redox indicator solution. Basic studies cover the investigation of electrochemical reaction (redox response potential) and optimal experimental conditions including choice of proper medium. Sensitive voltammetric techniques such as differential pulse voltammetry (DPV), square wave voltammetry (SWV) and adsorptive striping voltammetry (AdsSV) represent the electroanalytical methods typically used for the JAKis detection and quantification. Recent treatment of individual types of the electrodes and electroanalytical techniques by the International Union of Pure and Applied Chemistry (IUPAC) can serve for general information [47, 62].

Electrochemical measuring systems

Typically, an electrochemical experiment is performed in a commercial voltammetric cell using a conventional three-electrode arrangement with working, counter, and reference electrodes. Flow-through analytical systems and microsystems with electrochemical detection (ED) cover low-cost and effective electrochemical platforms and procedures such as injection analysis, capillary electrophoresis, lab-on-a-chip and other devices that have emerged as a simple and robust alternative to conventional electrochemical tests [47]. The electrochemical detection often follows a pre-separation step using HPLC or UHPLC and represents a highly selective and extremely sensitive mode. Here, the sample is first introduced and separated on a chromatographic columnconnected to an ED cell wherethe electrical current resulting from oxidation or reduction of the analyte is measured. Flow cells and electrodes are already miniaturized, making this type of detection suitable for standard- to nano-HPLC. In combination with the proper electronics, ED is able of an enormous linear range (LR) of more than 7 orders of magnitude. This means that the concentrations can be measured linearly from 10 pM to 100 µM (e.g., 3 × 10^{− 4}−3 × 10⁴ ng JAKi/mL) values with the same instrument.

Application advances in the electrochemical detection of JAK inhibitors

Electroanalytical procedures for the JAKis detection/determination are typically tested for sensitivity, linearity of the concentration range, limits of detection and quantification, interferences, and JAKis recovery in spiked samples as indicated in Table 2.

Table 2.

Parameters for published electroanalytical determination of JAK inhibitors

JAK inhibitor	Working electrode	Method	Peak potential vs. Ag/AgCl	Sensitivity, (µA/M)	Linear range, M	Limit of detection, M	Recovery, %	Sample	Reference
abrocitinib	4-ABA/ZnNFs@GO/MIP-GCE	DPV	0. 20 V with 5.0 × 10− 3 M [Fe(CN)6]3–/4– redox indicator	4.0 × 1013	1.0 × 10–13 – 1.0 × 10–12	2.2 × 10− 14 in standard solution, 2.4 × 10− 14 in serum	101.0–101.6	commercial human serum	[63]
baricitinib	ZnO/α-Fe2O3-NPs/CPE	SWV	1.10 V (04 M BRB pH 7.00)	5.0 × 106	8.0 × 10− 8 – 1.0 × 10− 6	1.8 × 10− 8	98.8–100.4 in tablet, 99.2–103.6 in plasma	pharmaceutical preparation and human plasma	[64]
	(Ni-NPs/f.MWCNTs-CPE	SWV	1.18 V (0.04 M BRB pH 4.0)	2.1 × 105	4.0 × 10− 8 – 5.6 × 10− 5	9.7 × 10− 9	99.1–100.4 in tablet	spiked human plasma and pharmaceutical tablet preparation	[65]
	poly(Py-co-2-TBA)/BAR@MIP/GCE	DPV	0.19 V with 5.0 × 10− 3 M [Fe(CN)6]3–/4– redox indicator	1.5 × 1012	5.0 × 10− 12 – 2.5 × 10− 11	2.2 × 10− 13	98.3–100.2 in tablet	tablet dosage form	[66]
ritlecitinib	Ni-CO-MW-GCE	DPV	0.75 V (pH 6.0 BRB)	1.6 × 105	9.0 × 10− 8 – 6.8 × 10− 6	2.8 × 10− 9	97,2–102.2 in tablet, 98.4–102.0 in plasma, 99.0–102.9 in urine	pharmaceutical formulations, human serum and human urine	[45]
ruxolitinib	GCE	DPV	0.60 V and 1.18 V (pH 4.7 AB)	3.0 × 105 in standard solution, 3.7 × 104 in serum	4.0 × 10− 6 − 8.0 × 10− 4 in standard solution, 4.0 × 10− 6 − 2.0 × 10− 4 in serum	5.2 × 10− 7 in standard solution, 1.1 × 10− 6 in serum	104.4 in tablet, 103.1 in serum	pharmaceutical tablet forms, synthetic human serum	[43]
	BDDE		1.24 V (pH 4.7 AB)	1.2 × 105 in standard solution, 6.8 × 104 in serum	1.0 × 10− 6 − 8.0 × 10− 4 in standard, 4.0 × 10− 6 − 2.0 × 10− 4 in serum	1.9 × 10− 7 in standard solution, 1.2 × 10− 6 in serum	100.3 in tablet, 100.1 in serum
	poly(taurine)-CPE	SWV	1.15 V (pH 4.7 AB)	4.7 × 106	1.0 × 10− 8 – 1.0 × 10− 6	5.0 × 10− 9	97.5 ± 0.8	pharmaceutical samples (10 mg of RXL)	[67]
	MIP@PHE MA-ThyM-GCE	DPV	0.25 V with 5.0 × 10− 3 M [Fe(CN)6]3–/4– redox indicator	2.7 × 1014 in standard solution, 5.4 × 1014 in serum	1.0 × 10–14 – 1.0 × 10–13	1.9 × 10− 15 in standard solution, 2.8 × 10− 15 in serum	102.4 in tablet, 102.3 in serum	tablets dosage (20 mg of RUX), synthetic serum	[60]
	SC-Co3O4-GCE	AdSDPV	1.25 V	4.4 × 105 in standard solution, 7.1 × 105 in serum	8.0 × 10− 8 – 2.0 × 10− 5 in standard solution, 6.0 × 10− 7 – 2.0 × 10− 5 in serum	6.7 × 10− 9 in standard solution, 2.6 × 10− 8 in serum	101.4 in tablet, 100.7 in serum	tablet dosage forms and commercial serum samples	[68]
tofacitinib	GCE	DPV, EIS	0.65 V (pH 4.7 AB), 0.55 V (pH 8.0 PB)	5.5 × 104 (pH 4.7 AB), 4.6 × 104 (pH 8.0 PB)	2.0 × 10− 6 − 1.0 × 10− 4 (pH 4.7 AB), 4.0 × 10− 6 − 1.0 × 10− 4 (pH 8.0 PB)	2.8 × 10− 7 (pH 4.7 AB), 1.3 × 10− 7 (pH 8.0 PB)	100.7 in tablet, 100.4 in serum (pH 4.7 AB)	tablet dosage forms and commercial human serum	[44]
	BDDE		0.74 V (pH 4.7 AB), 0.66 V (pH 8.0 PB)	3.3 × 104 (pH 4.7 AB), 3.5 × 104 (pH 8.0 PB)	1.0 × 10− 6 − 1.0 × 10− 4 (pH 4.7 AB), 2.0 × 10− 6 − 1.0 × 10− 4 (pH 8.0 PB)	5.2 × 10− 8 (pH 4.7 AB), 1.7 × 10− 7 (pH 8.0 PB)	100.3 in tablet, 99.6 in serum (pH 4.7 AB)
	ACR@MIP/GCE	DPV, EIS	0.15 V with 5.0 × 10− 3 M [Fe(CN)6]3–/4– redox indicator	DPV:∗ 4.5 × 1011 uA/M cm− 2 in standard solution, 1.0 × 1012 in serum EIS:∗ 9.7 × 1012 uA/M cm− 2 in standard solution, 3.3 × 1013 in serum ∗ the surface area after drug rebinding of 0.023 cm2	1.0 × 10–11 – 1.0 × 10–10 in both, standard solution and serum	DPV: 3.5 × 10–13 in standard solution, 1.1 × 10–12 in serum, EIS: 2.8 × 10− 12 in standard solution, 1.4 × 10− 13 in serum	DPV: 100.5% in tablet, 99.5% in serum. EIS: 98.0% in tablet, 98.9% % in serum	spiked commercial human serum and tablets	[58]
	ND@CuAl2O4@Fe3O4	DPV	0.82 V (pH 5 BRB)	6.1 × 104	5.0 × 10–8 – 1.3 × 10–5	7.8 × 10− 9	99.9–100.7 in tablet, 99.0–101.9 in urine, 100.3–101.3 in plasma	tablets, spiked human urine and plasma	[69]
upadacitinib	GCE	DPV	1.17 V (pH 2 BRB) and 1.009 V (pH 4.7 AB)	9.1 × 104 in standard solution, 3.8 × 104 in serum	8.0 × 10− 7 – 6.0 × 10− 5 (pH 4.7 AB)	1.9 × 10− 8 in standard solution, 3.6 × 10− 8 in serum	100.8	spiked commercial human serum	[70]
	BDDE		1.25 V (pH 2 BRB) and 1.12 V (pH 4.7 AB)	4.7 × 104 in standard solution, 3.7 × 104 in serum	4.0 × 10− 6 – 6.0 × 10− 5	3.7 × 10− 8 in standard solution, 4.7 × 10− 8 in serum	103.0
	UPA-ZnO/3-APBA@MIP-GCE	DPV	0.15 V with 5.0 × 10− 3 M [Fe(CN)6]3–/4– redox indicator	7.6 × 1013 in standard solution, 9.6 × 1013 in serum	1.0 × 10–13 – 1.0 × 10–12	1.4 × 10− 14 in standard solution, 2.2 × 10− 14 in serum	99.6–100.3 in serum	commercial human serum	[71]
	Co6.8Se8@NPC-GCE	DPV	1.12 V (pH 2.0 PBS)	1.1 × 105	5.0 × 10–7 – 1.4 × 10–5	1.1 × 10− 8	99.1–101.9 in tablet, 100.1–101.2 in urine	pharmaceutical tablets and spiked urine	[61]

To our best knowledge, no electrochemical sensors and procedures for the detection of delgocitinib, deucravacitinib, fedratinib, filgotinib, oclacitinib, pacritinib, peficitinib as well as for momelotinib, golidocitinib, and deuruxolitinib were published till now. The reported electrochemical detection of individual JAKis is characterized by features as follows:

Abrocitinib

(ABR) is the JAK 1 inhibitor. The first electrochemical sensor designed for its sensitive and selective determination incorporates the MIP prepared using 4-aminobenzoic acid as a functional monomer and ABR as a template molecule deposited on zinc nanoflower (ZnNFs)-graphene oxide (GO) conjugate (ZnNFs@GO) at GCE. The sensor response to ABR was evaluated using the redox indicator [Fe(CN)₆]^3–/4– redox indicator (Table 2). Additionally, the repeatability of response was characterized by the relative standard deviation (RSD) of 1.0% and 1.1%, and reproducibility by RSD 1.9% and 1.8% in standard solution and in serum sample, respectively. In the samples of commercial human serum after separation of proteins with the ABR concentration 2.5 × 10^{− 13} M and spike 5.0 × 10^{− 13} M, the average recovery was 101.0% with RSD of recovery 1.9%. The interference study with 1:10 ratio of inorganic ions, dopamine, uric acid, ascorbic acid, and paracetamol obtained the ABR recovery values between 99.1% and 102.5%. The designed sensor showed a performance of 97.4% in the first 3 days of storage and 90.7% in 7 days of storage. Selectivity of the MIP-based sensor was verified by imprinting factor calculations using ibrutinib, ruxolitinib, tofacitinib, zonisamide, and acetazolamide [63].

Baricitinib

(BCT, BARI) is a reversible Janus kinase inhibitor that targets the JAK1 and JAK2 subtypes. An electrochemical platform comprising zinc oxide and iron oxide nanoparticles modified carbon paste electrode (5% ZnO/α-Fe₂O₃-NPs/CPE) was reported which exhibited superior sensitivity and remarkable peak current amplification compared to other tested electrodes: bare CPE, 2% α-Fe₂O₃-NPs/CPE, 5% α-Fe₂O₃-NPs/CPE, 8% α-Fe₂O₃-NPs/CPE, and 8% ZnO/α-Fe₂O₃-NPs/CPE. The diffusion-controlled process and oxidation mechanism consisting of 2e⁻ and 2 H⁺ transfer were confirmed as a basis for sensitive square wave voltammetric determination. Repeatability was characterized by RSD of 0.8%. Robustness of the proposed method is demonstrated by the fact that adjustments of equilibrium time (5 ± 2 s) and pH variation (7.0 ± 0.1) had no impact on the peak current or peak potential of BCT. During the interference study, the presence of commonly used pharmaceutical excipients such as mannitol, magnesium stearate, cellulose, iron oxide red, croscarmellose sodium, and titanium dioxide (0.1 mM of each) had negligeable impact on the measured signal and the RSD values remained below 2.0%. The absence of interference was also demonstrated commonly prescribed COVID-19 medications, including paracetamol, ascorbic acid, and molnupiravir [64]. A unique sensor was developed by modifying the carbon paste with nickel nanoparticles (Ni-NPs) incorporated within f.MWCNTs, resulting in nanohybrids tailored for the highly sensitive BARI detection. For a comparison, the [Fe(CN)₆]^3–/4– response obtained at Ni-NPs/f.MWCNTs-CPE was 2.37 times greater and 1.55 times greater than that at bare CPE and f.MWCNT/CPE, respectively. The BARI response was 5.78 µA, 9.56 µA, and 13.75 µA at bare CPE, f.MWCNT and Ni-NPs/f.MWCNTs-CPE, respectively. The primarily diffusion controlled electrochemical process was governed by the peak-current vs. scan rate study and the 2e⁻ and 2 H⁺ process was proposed (Fig. 1A). Analytical repeatability was given RSD 0.7%. The interference of 60-fold concentrations of inorganic ions such as Ca²⁺, Zn²⁺, Na⁺, Fe³⁺, K⁺, Mg²⁺, NO₃⁻, SO₄²⁻, and Cl⁻, as well as substances like magnesium stearate, mannitol, microcrystalline cellulose, ferric oxide, titanium dioxide, poly(vinyl alcohol), lecithin, polyethylene glycol, and talc shown the tolerance threshold less than 5% [65].

A highly sensitive and selective molecularly imprinted polymer (MIP)-based electrochemical sensor was fabricated using the electropolymerization on GCE, utilizing 2-phenylboronic acid (2-TBA) as the functional monomer and pyrrole (Py) to provide both conductivity and stability to the polymeric structure. The (poly(Py-co-2-TBA)/BAR@MIP/GCE) sensor has excellent selectivity for BAR, even in the presence of structurally similar compounds and is recognized as a reliable tool for BAR monitoring, supporting personalized dosing and improved therapeutic outcomes [66].

Ritlecitinib

(RITL) is a selective inhibitor of JAK3 and tyrosine kinase expressed in hepatocellular carcinoma (TEC) kinase family, exhibits promising potential as an effective therapeutic modality for addressing alopecia areata and vitiligo [72]. An innovative electrochemical sensor utilizes GCE modified with a novel nanocomposite of nickel-cobalt multilayer nanowires (Ni-CO-MW). Electrochemical CV and EIS analyses revealed its excellent conductivity, enhanced electrocatalytic performance, and efficient electron transfer behaviour. RITL in BR buffer pH 6.0 exhibited the DPV anodic peak with the current of 2.84 µA at bare GCE and 4.6 µA with a lower oxidation potential at the Ni–Co–MW-modified GCE. The electrochemical oxidation on the modified electrode is a diffusion-controlled process. In contrast to other JAKis studies, the authors proposed 1e⁻ and 1 H⁺ transfer with alternative oxidation mechanism involving the formation of a radical on the nitrogen linker between pyrrolopyrimidine and piperidine rings (Fig. 1C). The results of sensitive DPV measurements at the modified GCE are shown on Fig. 3a. The electrode exhibited outstanding selectivity, reproducibility, and repeatability. Potassium chloride, L-methionine, sodium nitrate, D-glucose, L-arginine, sodium sulfate, uric acid, ascorbic acid, caffeine, dopamine, and paracetamol did not significantly affect the RITL response. Sensor stability characterizes its storage under ambient conditions for 15 days when 95% of the initial peak current response was maintained. Density Function Theory calculations and molecular docking calculations were also carried out to predict non-covalent interactions between the inhibitor molecule RITL and the active sites of the JAK3 enzyme system [45].

Ruxolitinib

(RUX, RXL) is a potent and selective inhibitor of both JAK1 and JAK2, and it received the repeated approval. Currently, numerous novel therapeutic applications of RXL are under investigation which includes COVID-19 and dermatological autoimmune diseases. For the RXL determination, bare GCE and BDDE substrates have shown pH dependent position of CV and DPV responses with shifts to less positive values with increasing the pH revealing the proton-dependent nature of RXL two-electron electrochemical oxidations. The electrooxidation procedure of RXL was a diffusion-controlled process on both electrode materials. DPV methods were carried out by quantifying RXL without any pretreatment steps in the pharmaceutical dosage form and in spiked synthetic human serum after the precipitation of protein residues [43]. An electrochemical sensor with photopolymerized molecular imprinted polymer based on thymine methacrylate (ThyM) as a functional monomer deposited at a glassy carbon electrode (GCE/MIP@PHEMA-ThyM) possessed well developed peak of the [Fe(CN)₆]^3−/4− redox indicator round 0.35 V vs. Ag/AgCl using with very low values of LR concentration and LOD (Fig. 3b). An interference effect of dopamine, uric acid, paracetamol, and ascorbic acid was found at the ratio of 1:10 [60].

A waste sponges (SC) derived carbon materials applied for the first time as electrode modification agents together with magnetic Co₃O₄ nanoparticles was also investigated. The CV study showed that RXL depends on the diffusion-adsorption mixed process with the number of electrons exchanged in the electrochemical oxidation process of number of electrons equal to 1.46. Comparing to bare electrode, the anodic peak potential is shifted to a less positive value of 1.250 V vs. Ag/AgCl depending on medium pH [68]. A poly(taurine)-modified carbon paste electrode was prepared where the CV oxidation of the RXL molecule on this sensor was completely irreversible and diffusion-controlled under the experimental conditions studied (Fig. 4). The logarithmic I_p vs. scan rate equation shows that there is also an adsorptive effect in addition to the diffusion. For 4 × 10^{− 7} M RXL, the excellent response reproducibility (RSD of 1.9%, n = 3) and three-weeks stability of the RXL peak potential (RSD of 2.0%) were obtained. No significant change in the peak potential of RXL was observed upon the addition of inorganic ions Ca²⁺, Cu²⁺, Zn²⁺, Ag⁺, Na⁺, NO₃⁻, and Cl⁻, as well as dopamine, ascorbic acid, and uric acid, possessing anodic peaks in the region of 0.35 to 0. 48 V. The sensor also demonstrated excellent selectivity, accuracy, precision, and repeatability. It is easy to manufacture and use, cost-effective, reliable, and precise [67].

Fig. 4.

Schematic process of fabrication of poly(taurine)-modified CPE. Reprinted with permission from [67]

Tofacitinib

(TOF) is the JAK1, JAK2, and JAK3 inhibitor. The first report on its electrochemical determination using GCE and BDDE was by Budak et al. and indicated a diffusion-controlled and irreversible 2-electron oxidation of TOF at both electrodes. Applied to commercial serum samples, BDDE showed a wide linear range compared to GCE and the lower LOD value. For the tablet dosage a good average recovery of 100.71% and 100.26% with RSD of recovery of 0.5% and 1.0%, and for serum 100.39% and 99.62% with RSD of 0.7% and 0.7% were obtained in pH 4.7 AB on GCE and BDDE, respectively, which demonstrate the interference-free performance of the sensors [44]. A highly selective MIP-based sensor was prepared by photopolymerization of acrylamide and hydroxyethyl methacrylate with crosslinking ethylene glycol dimethacrylate on the GCE. Analytical parameters of the sensor were evaluated after rebinding various concentrations of TOF by measuring the redox indicator response. Selectivity of the sensor was confirmed with imprinting factor and interference studies, and the sensor performance was controlled using non-imprinted polymer for comparison at every step. Very close values of linear range, LOD, LOQ, sensitivity, and correlation coefficient using the DPV and EIS methods show the applicability of this sensor with both methods. However, better repeatability and reproducibility values were obtained with the DPV comparing to EIS measurements. For spiked serum, the repeatability and reproducibility are characterized byRSD equal to 0.3% and 0.8%, respectively. The ACR@MIP/GCE sensor indicated sufficient stability after 1 week preserving 90% of its initial value. The linear range, LOD and applicability is comparable to UHPLC-MS/MS method [58].

Another electrochemical sensor for the efficient TOF detection utilizes a composite of nanodiamond (ND), copper aluminate spinel oxide (CuAl₂O₄), and iron (II, III) oxide (Fe₃O₄) as modifiers on the GCE surface. The synergistic effect of these nanomaterials provides significant advancements. Moreover, this study conducted a thorough investigation using Density Functional Theory for the geometry optimization of TOF and the TOF-ND complex. Molecular docking studies using JAK1 (PDB ID: 3EYG) and JAK3 (PDB ID: 3LXK) indicated higher interaction of the TOF-ND conjugate with the JAKis. The sensor exhibited exceptional selectivity for TOF as evidenced by RSD of 4.4% observed at maximum concentrations of matrix components such as ascorbic acid, uric acid, D-glucose, L-arginine, L-methionine, potassium chloride, sodium sulphate, and sodium nitrate even in the presence of a 1000-fold excess of these components. Analysis of 1 µM TOF solution possessed the repeatability with RSD of 2.0% and the reproducibility RSD of 2.5% [69].

The interaction between tofacitinib and calf thymus double-stranded deoxyribonucleic acid has been reported very recently for the first time using differential pulse voltammetry in two distinct ways - biosensor with a surface attached DNA and biosensing in solution. Binding between TFC and DNA has been confirmed that allows quantitative electroanalytical determination of TFC with high sensitivity also in commercial serum sample with recovery values of 100.3–101.1% [73].

Upadacitinib

(UPA) is a highly selective JAK1 inhibitor. Its oxidation at two different solid electrodes was found to be diffusion-controlled for GCE and a mixture of adsorption/diffusion for BDDE. However, further experimental and theoretical studies have been indicated as necessary to gain deeper insights into the oxidation mechanism of heteroaromatic compounds and drugs, which correlates with enzymatic oxidation. The GCE and BDDE used in standard solution possessed the UPA response repeatability of 1.0% and 0.4% and reproducibility of 1.5% and 1.1%, respectively. The method has been applied to spiked serum samples, where for 2.0 × 10^{− 5} M UPA the average recovery of 100.8% and 102.9% with RSD of recovery 1.3% and 2.2% were found for GCE and BDDE, respectively. The interference studies performed in the presence of naturally occurring or exogenous compounds such as Na⁺, SO₄^2–, K⁺, NO₃^–, Mg²⁺, and Cl⁻ at 10- and 100-fold excess, and dopamine, paracetamol, ascorbic acid, and uric acid at 1:1 concentration showed no detrimental effect to the current response of UPA. This study demonstrates affordable and sensitive electrochemical option for the UPA determination using reusable GCE and BDDE electrodes [70].

Another study by these authors with UPA used chemically modified electrodes. A nanoparticle-doped MIP-based electrochemical sensor was designed for the first time as a thin film layer using the photopolymerization on GCE with various nanomaterials such as multi-walled carbon nanotube, titanium dioxide, oxide, and zinc oxide in polymerization solution. The sensor utilized 2-hydroxyethyl methacrylate as the basic monomer, ethylene glycol dimethacrylate as the cross-linking agent, and 3-aminophenyl boronic acid as the functional monomer. N-type ZnO nanomaterials have semiconductor properties, are inexpensive, non-toxic, compatible with biological structures, and exhibit electrochemical behaviour. ZnO nanoparticles enhance the effective surface area and the number of regions, thereby improving the sensor’s overall performance. The sensor possessed the response repeatability RDS of 0.7% and 1.2% and reproducibility RDS of 1.9% and 1.1%, in standard UPA solution and serum, respectively. In commercial serum sample with the UPA concentration of 1.0 × 10^{− 11} M the average recovery was 100.3% with RSD of recovery of 1.0%. The sensor exhibited high selectivity for UPA in the presence of similar molecules ruxolitinib and tofacitinib as well as impurities at ratios of 1:1, 1:10, 1:100, and 1:1000 and in the presence of dopamine, ascorbic acid, paracetamol, and uric acid at ratios 1:1. The fabricated sensor retained stability after 5 days of storage with 95.3% performance [71].

GCE modified with a composite material comprising Co_6.8Se₈ uniformly dispersed within the porous carbon network (Co_6.8Se₈@NPC) was synthesized from metal organic framework (Fig. 5). The composite exhibited a notable catalytic effect towards the oxidation of UPA with an irreversible anodic peak at a reduced potential value and an enhancement in peak current: while 0.1 mM UPA in 0.1 M PBS pH 2.0 possessed on bare GCE an CV anodic peak with peak potential E_p.a. 1.23 V vs. Ag/AgCl accompanied by a peak current I_p.a. of 1.65 µA, at the modified electrode it was E_p.a. 1.12 V with I_p.a. 3.47 µA. A diffusion-controlled process was confirmed for an irreversible electrochemical reaction of UPA, and log I_p vs. log υ plot exhibited a linear curve with a value of 0.389 (less than 0.5) expected for a diffusion-controlled nature of the electrocatalytic oxidation. The very sensitive calibration of UPA measurements in the DPV mode is depicted on Fig. 3c. The RSD value of 2.4% for the response of 1.0 × 10^{− 6} M UPA indicates a consistent and repeatable sensor surface. The sensor retained 95% of its initial response after 15 days, demonstrating excellent stability. The selectivity of the sensor evaluated by measuring UPA in the presence of a 100-fold excess of interfering species including paracetamol, dopamine, L-arginine, ascorbic acid, D-glucose, uric acid, and L-methionine has shown no significant interference from these molecules, with an RSD below 1%. For 4.0 × 10^{− 6} M UPA spike to human serum possessed the recovery of 101.2% with RSD of 1.0% [61].

Fig. 5.

One pot synthesis of Co_6.8Se₈@NPC from ZIF-12 by controlled pyrolysis with selenium powder.. Reprinted with permission from [61]

Critical evaluation of JAKis detection by electroanalytical vs. other methods

Progress in drug analysis is driven by the demand for fast, cost-effective, and straightforward procedures capable of providing accurate and clinically interpretable results. Electroanalytical methods meet many of these requirements, offering the possibility of rapid measurements directly at the point of care without the need to transport biological samples to centralized laboratories. Based on the data in Table 2, the analytical performance of JAKis sensors relies heavily on effective electrocatalysis and/or molecular recognition, most notably achieved through the use of nanostructured materials as electrode modifiers. These strategies substantially enhance the sensitivity and selectivity of the JAKis detection.

For clinical applications, selectivity and sensitivity of the proposed electrochemical methods are crucial validation parameters. Thanks to the sophisticated electrode modifications, many of the electrochemical methods reported to date can quantify JAKi at clinically relevant concentrations in biological fluids (e.g., serum, urine). As shown in Table 3, in most cases the performance of these methods reaches clinically relevant concentration levels (0.5–200 ng/mL for abrocitinib, baricitinib and upadacitinib; 1–400 ng/mL for tofacitinib; 0.5–400 ng/mL for ruxolitinib, and 10–800 ng/mL for fedratinib [25]), indicated in bold, and in several cases is comparable to the results obtained with tandem LC–MS methods. The investigations have been performed both in standard solutions as well as in samples of serum and urine, although to date these matrices have been predominantly of commercial origin.

Table 3.

Comparison of linear range reported for analytical methods for the JAKis detection (reaching of clinically relevant values is indicated by bold)

JAK inhibitor	HPLC, ng/mL	LC-MS, LC-MS/MS, UPLC-MS/MS, ng/mL	Fluorescence spectrometry, ng/mL	UV-Vis spectrometry, ng/mL	Electrochemistry, ng/mL
abrocitinib		0.5–200 [25]			3 × 10− 5–3 × 10− 4 [63]
baricitinib	1 × 104–7.5 × 1013, 306 nm UV-Vis detection [14]	0.5–200 [25]	78–1.2 × 103 [37]	1 × 104–6 × 104 [37]	30–370 [64]
	1 × 103–3 × 104, 205 nm UV detection [15]	1–100 [26]	5 × 102–1 × 103 [38]	6.5 × 103–6.0 × 104 [40]	15–2.1 × 104 [65]
	1.2 × 1013–2.8 × 1013, 290 nm DAD detector [16]	1.0–100 [35]		1 × 104–1 × 105 [41]
ritlecitinib		5–100 [28]			25 − 1.9 × 103 [45]
ruxolitinib	2.4 × 104–1.4 × 105, photodiode array detector [19]		0.8–250 [71]		1.2 × 103−2.4 × 104 at GCE, 3.1 × 102−2.4 × 104 at BDDE [43]
	0.2–500, fluorometric detector [20]				30 − 300 [67]
					3 × 10− 6−3 × 10− 5 [60]
					250 − 6.0 × 103 [68]
tofacitinib	10 and 100 for rat plasma and urine, resp., 287 nm UV detection [21]	0.5–400 [29]			600–3.1 × 104 at GCE, 300–3.1 × 104 at BDDE [44]
	5 × 102–1 × 104, 287 nm UV detection [22]	0.05–100 [30]			3 × 10− 3−3 × 10− 2 [58]
	100–608, 220 nm UV detection [23]	0.2–100 [33]			16–4.1 × 103 [69]
	1.5 × 104–9.0 × 104, 254 nm UV detection [24]
	100–2 × 104, 285 nm UV detection [74]
upadacitinib		12.5–100 [34]			300–1.8 × 104 at GCE, 1.5 × 103–1.8 × 104 at BDDE [70]
		0.5–200 [25]			4 × 10− 5−4 × 10− 4 [71]
		0.5–200 [31]			190–5.2 × 103 [61]
		0.15–150 [32]

However, to objectively assess the current state of electroanalytical approaches for the JAKis determination, it is necessary to compare them critically to alternative analytical methods to demonstrate their mutual competition (Table 4). Regarding the study of selectivity, that and the effect of interferences on the electrochemical determination of JAKis are generally performed using a standard set of inorganic ions and oxidizable substances typically present in biological samples, often in much higher concentrations than the JAKi of interest. However, limited attention has been given to evaluating selectivity in the presence of metabolites [1] that may exhibit similar electrochemical behaviour. As a result, the measured concentration may reflect the signal of the target analyte together with any metabolites sharing the same electrochemical properties, thus yielding incorrect results. This issue also applies to MIPs, where structurally related metabolites may bind to the imprinted cavities within the polymer matrix. Comparable challenges can arise in UV-Vis and fluorometric detection, where overlapping spectral signals may compromise selectivity. While some electrochemical papers went beyond the standard set of interferences, for example, the selectivity of MIP-based upadacitinib sensorwas tested against other JAKis (ruxolitinib and tofacitinib) with satisfactory results, such selectivity is of less clinical relevance because the guidelines [74] do not recommend the concurrent administration of multiple JAKis, therefore it is unlikely that a biological sample would contain several JAKis at the same time. With respect to the selective determination of JAKis in clinical samples, the highest level of selectivity is undoubtedly achieved using LC–MS approaches. These methods provide two orthogonal dimensions (chromatographic separation in time and mass-to-charge differentiation in the mass spectrometer) allowing effective resolution of the JAKis from structurally related metabolites and other endogenous and exogenous matrix components. Consequently, LC–MS methods remain the gold standard for quantification of JAKis in complex biological matrices, particularly in clinical settings. In a comparison of LC-MS/MS and electrochemistry approaches, very recently, Zouari et al. conclude that the use of both techniques is justified by their ability to meet distinct operational requirements, ranging from high-throughput centralized analysis to rapid, point-of-care monitoring, which is essential for therapeutic drug monitoring and pharmacokinetic studies. Some complementary strengths can be found such as the robustness and regulatory alignment of LC-MS/MS versus the portability, speed, and operational simplicity of the sensor [75].

Table 4.

Advantages and disadvantages of analytical approaches for the analysis of JAKi in clinical settings

Methods	Advantages	Disadvantages
HPLC-UVa HPLC-FLb	• Detection in complex matrices (after appropriate sample preparation) • Inherent fluorescence of some JAKi enhances sensitivity and selectivity • Greener method (compared to LC-MS) in terms of energy consumption • Widely available in routine clinical and QC laboratories	• Lower sensitivity and selectivity than LC–MS(/MS) • Need for derivatization (labeling) in case of inherently non-fluorescent JAKis • Need for standards (compared to LC-MS approach) for identification • Limited information on JAKis metabolites
LC-MS, LC-MS/MS, UPLC-MS/MS	• Highest sensitivity and selectivity • Detection in complex matrices at clinically relevant concentrations • Selective detection of JAKis and their metabolites • Highly suitable for therapeutic drug monitoring	• Expensive instrumentation • Required operator expertise • Higher requirements for sample pretreatment • More demanding validation • Limited to centralized laboratories
Electrochemistry	• Inexpensive • Relatively fast measurements • Possibility of miniaturization and portability • Mass production of disposable sensors • Potential for point-of-care testing • Easy electrochemical or physico-chemical (elution for MIP) regeneration of electrode • Low chemicals consumption and waste production	• Stability of electrode surface modification • Reproducibility of modification process • Electrode surface fouling/passivation in biological samples • Limited selectivity for JAKis metabolite discrimination • Limited data on the analysis of real clinical sample • Limited commercialized sources of chemically modified sensors

^{a, b}UV-VIS and Fluorescence spectrometry (not hyphenated): unsuitable for direct measurements in multicomponent (e.g., clinical) matrices due to sample matrix interferences, and limited discrimination of structurally related compounds (e.g., JAKis metabolites), suitable for fast, inexpensive batch analysis (multi-well plates) of pharmaceutical samples (e.g., JAKis in dosage forms); ^asensitivity issue typically prevents direct measurements at clinically relevant concentrations of JAKi

The above-mentioned unaddressed selectivity limitations also affect the applicability of electrochemical methods in the pharmaceutical quality control of JAKi, s whether as active pharmaceutical ingredients (APIs) or in dosage forms—often the first application highlighted in scientific publications (see also Table 2). Although direct quantification of JAKi by electrochemical methods may represent a suitable alternative to the commonly used titrations or liquid chromatography assays described in pharmacopoeial monographs (but none of the JAKi discussed in this review currently has an official USP or Ph. Eur. Monograph), the test for related substances using liquid chromatography will most likely always be necessary due to the selectivity constraints outlined above.

Another crucial validation parameter is the reproducibility of the measurements. Although many studies report more than satisfactory intra- and inter-day reproducibility of the sensor’s analytical response, considerably less attention is devoted to the reproducibility of the electrode surface modification itself. Each modification step introduces additional variability, thereby influencing the overall reproducibility of sensor preparation process and, ultimately, its performance. In this context, less than half of the publications [45, 61, 64, 67, 69] reported “inter-sensor” reproducibility, with the number of tested freshly prepared sensors ranged from 4 [67] to 10 [61, 69].

The preparation of a surface-modified electrochemical sensor has also other implications. Many publications describing electrochemical sensors with sophisticated surface modifications (including the ones mentioned in this review) assert that these systems enable faster analysis compared to other analytical methods, particularly chromatographic ones. However, in the case of the limited stability of such modified sensors, it is often necessary to prepare a “fresh” sensor at least every few days. The time required for sensor fabrication is not always considered in these comparisons. In fact, the total time needed to prepare the sensor prior to analysis is frequently not stated in the publications. In standardization of electrochemical method protocol, this factor is crucial. In this regard, electrochemical sensors with long term stability are desirable. If there are to be commercially available electrochemical sensors for the JAKis determination, the aim should be to achieve at least a 6-month stability (shelf-life) of the modified electrode. The stability studies of the surface-modified JAKi sensors discussed in this review were conducted only in short timespans, ranging from 4 to 30 days [45, 58, 61, 63–68, 71], while some published articles did not present any stability study.

Finally, the environmental sustainability of the proposed electrochemical platforms has, in two cases [64, 66], been evaluated using the Analytical Eco-Scale (a framework designed to systematically assess and quantify the environmental footprint of analytical procedures) and the AGREE methodology. Both approaches confirmed satisfactory greenness of the developed methods. However, it should be noted that the greenness of the developed analytical methods is not routinely assessed, and many published methods, whether electrochemical or alternative, do not report any evaluation of their environmental impact, making it impossible to compare them in this regard.

Future outlook

The current progress in the development of electrochemical methods for the JAKi determination demonstrates that appropriately modified electrodes can reach clinically relevant concentration levels and, in some cases, have been reported to rival LC–MS methods in terms of sensitivity. However, given the still limited number of published electrochemical sensor for the JAKis determination, it is currently difficult, or perhaps premature, to propose general recommendations regarding the optimal composition of nanocomposites or MIPs. Nevertheless, we propose that the use of nanomaterials and the synthesis of nanocomposites be evaluated systematically by comparing various nanomaterials within the same study to produce the sensor with the best overall performance. Additionally, we propose to continue the exploration of novel hybrid materials that might improve both sensitivity and especially antifouling properties.

Despite the promising apparent performance of the discussed electrochemical procedures, several critical aspects remain in question: selectivity toward metabolites and other clinically relevant interferents, robustness of the preparation of electrode modification, long-term stability of the modified sensors, and application in analysis of real clinical samples. To facilitate the development of more reliable, robust and clinically applicable electrochemical methods, we propose below a few recommendations that future studies should follow:

1. Interference study with clinically relevant compounds

   The interference study should go beyond the standard panel of inorganic ions and simple electroactive species. The sensor response should be evaluated in the presence of relevant endogenous and exogenous interferents (exact list depending on the matrix) at appropriate concentrations.

2. Selectivity assessment using structural analogues

   If possible, selectivity of the sensors should be assessed using structurally related compounds, such as known JAKi metabolites. This is particularly important because metabolites (in many cases pharmacologically inactive) may exhibit comparable electrochemical behaviour or binding to MIPs, leading to overestimation of the JAKi concentration. The same as for the biodegradation, JAKis products also applies for other degradation JAKi products (electrochemical-, photo-, temperature-degradation, etc.). In the case of analytical signal overlap, such signal must be clearly declared as total/pool JAKis signal including the sum of responses of all examined structurally related JAKi analogues.

3. Reproducibility of electrode surface modification

   In addition to intra- and inter-day repeatability of the analytical signal, the reproducibility of the sensor fabrication should be systematically evaluated. Multiple independent modified electrodes (we propose at least 5) should be prepared and the variability of their response evaluated. Modified electrode surfaces should be thoroughly characterized using appropriate analytical techniques. This is essential for translation into commercial sphere.

4. Reporting of sensor Preparation time

   The total time required for sensor preparation (including all conditioning, activation, and deposition steps) should be explicitly reported. This would allow realistic comparison with other analytical methods (e.g., chromatographic) and would avoid otherwise unsupported claims of “rapid analysis” based solely on measurement time, without considering the time necessary for the fabrication of the sensor.

5. Short-term and long-term stability evaluation

   Stability studies should extend beyond a few days and try to aim toward time frames relevant for potential commercial sensors (e.g., at least 6 months of storage). The authors should assess the changes in signal during storage under defined conditions and provide clear performance criteria such as acceptable signal deviations.

6. Real-sample validation

   Validation solely in spiked model matrices is insufficient in regard to practical application. Whenever possible, method performance should be evaluated in authentic biological samples (e.g., patient serum/plasma, urine). To this end, authors should not be worried about sacrificing simplicity of their electrochemical method and perform appropriate sample pretreatment to achieve the goal.

7. Alignment with clinically relevant concentration ranges

   Methods should clearly state their intended application. When the aim is to determine JAKis in biological samples, the authors should relate their validation parameters (e.g., LR, LOD) to expected concentration ranges in a given matrix.

Besides electrochemical sensors, there are still unexplored areas regarding the JAKis determination in biological matrices. To the best of our knowledge, no method combining HPLC with electrochemical detection for JAKi has been reported to date. Such methods could benefit from the separation power of liquid chromatography and take advantage of the inherent sensitivity and simplicity of electrochemical detection, employing not only conventional amperometric detectors but also fast-scan cyclic voltammetry or square wave voltammetry to further enhance the performance. A hyphenated electrochemical detector provides electrochemical selectivity as an additional characteristic compared to a hyphenated spectroscopic and/or MS detector, so that the analysis can benefit from such a detection tandem. Therefore, any true progress in electrochemical method itself as well as its powerful combination/hyphenation is of high interest and importance for future research.

The trends in point-of-care and in-place analyses already stimulate and will control progress in the development of highly effective, sustainable, eco-friendly, and low-cost analytical devices with effective generation of analytical signal and user-friendly interfaces. Therefore, fabrication and commercialization of the specific miniaturized sensors and devices is expected. To achieve the goals, the optimization and standardization of the composition of sensors interface and analytical systems as well as the establishment of standardized testing protocols are of great importance.

Conclusion

Despite a critical appraisal of some problematic issues that need to be considered with respect to real clinical requirements and implementations, we fully recognize and appreciate the research carried out so far on electrochemical sensors for JAKis, as it already demonstrates an impressive degree of ingenuity and the use of state-of-the-art approaches in electrochemical sensor design. The detection of JAKis such as baricitinib, ruxolitinib, tofacitinib, and upadacitinib has been successfully addressed using thoughtfully engineered nanocomposites and molecularly imprinted polymers, often achieving extraordinarily low detection limits in pharmaceutical and biological matrices and, in some cases, performance approaching that of LC–MS-based assays. The collection of articles critically discussed in this review therefore represents a rapidly growing summary of concepts, materials, and validation strategies that future research can further refine and ultimately translate into robust point-of-care devices and/or hybrid platforms for routine pharmaceutical and biomedical analysis.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

AB

acetate buffer

ABR

abrocitinib

ACR

acrylamide

AdsSV

adsorptive striping voltammetry

APIs

active pharmaceutical ingredients

BCT

baricitinib

BARI

baricitinib

BDDE

boron-doped diamond electrode

CNTs

carbon nanotubes

CPE

carbon paste electrode

CV

cyclic voltammetry

DPV

differential pulse voltammetry

ED

electrochemical detection

EIS

electrochemical impedance spectroscopy

EMA

European Medicines Agency

FDA US

Food and Drug Administration

f.MWCNTs

functionalized multiwalled carbon nanotubes

FTIR

Fourier-transform infrared spectroscopy

GCE

glassy carbon electrode

GO

graphene oxide

HPLC

high-performance liquid chromatography

IUPAC

International union of pure and applied chemistry

JAK

Janus kinase

JAKi

Janus kinase inhibitor

LC−MS, LC–MS/MS,  UHPLC-MS/MS, and UPLC-MS/MS

liquid chromatography−mass spectrometric tandem techniques

LOD

limit of detection

LR

linear range

MIP

molecularly imprinted polymer

MOF

metal–organic framework

ND

nanodiamond

NIP

non-imprinted polymer

NM

nanomaterial

NP

nanoparticle

PBS

phosphate buffer saline solution

RITL

ritlecitinib

RSD

relative standard deviation

RUX

ruxolitinib

RXL

ruxolitinib

SC

waste sponges

STAT

signal transducers and activators of transcription

SWV

square wave voltammetry

TOF

tofacitinib citrate

TYK

tyrosine-protein kinase

UC

ulcerative colitis

UHPLC

ultra high-performance liquid chromatography

UPA

upadacitinib

Author contributions

Conceptualization (JL, PM), Formal analysis (MH, JL, PM, ZZ), Methodology (JL, PM), Visualization (MH, JL, PM, ZZ), Writing – original draft (JL, PM) Writing – review and editing (JL, PM, MH).

Funding

Open access funding provided by The Ministry of Education, Science, Research and Sport of the Slovak Republic in cooperation with Centre for Scientific and Technical Information of the Slovak Republic. This work was funded by the EU NextGenerationEU through the Recovery and Resilience Plan for Slovakia under the project No. 09I03-03-V04-00157 and by the project VEGA 1/0514/22.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Competing interests

The authors declare no competing interests.