Wearable Biosensors for Glucose Monitoring in Sweat: A Patent Analysis ^†

Abstract

Metabolic diseases are increasing in relevance both in health and the economy in most countries. In this direction, if gold-standard technologies are based on blood analysis, non-invasive glucose monitoring is a relevant and great challenge that has not yet been fully resolved. Sweat represents a more suitable medium for the non-invasive sensing and monitoring of glucose than other bodily fluids, such as saliva, tears, or urine. However, the measurement of glucose levels requires the use of highly precise and sensitive sensors, given the low glucose concentration in sweat. This paper provides an overview of the patent landscape related to wearable biosensors for the monitoring of glucose levels in sweat.

1. Introduction

Metabolic diseases are increasing in relevance in health and the economic expenditure in most countries. Among them, diabetes has been identified as a prevalent and enduring chronic condition in humans that is also associated with a variety of health complications, including but not limited to heart disease, stroke, kidney disease, nerve damage, and vision impairment. These complications may stem from fluctuating blood glucose levels. Thus, it is imperative to engage people in regular monitoring of blood glucose levels to ensure they remain within the optimal range. This practice is essential for effectively managing diabetes and facilitating the progression toward optimal disease control. It is noteworthy that a correlation exists between diabetes and obesity. Also, early diagnosis represents a key action requiring an observation period with regular monitoring before taking the most appropriate therapeutic decisions.

Current technologies used for the monitoring of glucose levels can be classified into two distinct categories: invasive (requiring the analysis of blood drops) and non-invasive (by analyzing other biological fluids).

Conventionally, blood analysis in medical facilities entails a multifaceted process involving the manual extraction and examination of blood samples. This method is both time-consuming and produces a certain discomfort for the patient, in addition to the need for the proper management of contaminated biological waste [1].

Recent technological advancements have led to the development of methods for the continuous and real-time monitoring of glucose levels in different fluids (sweat, tears, saliva, and interstitial fluids) other than blood. This development addresses the requirement of mitigating the aforementioned drawbacks [2]. Non-invasive technologies are showing to have considerable potential, in particular sweat. Sweat represents the most interesting and suitable medium for the non-invasive sensing and monitoring of glucose instead of other fluids, such as saliva, tears, or urine, because it is easy to collect and available in many areas of the body (also without relevant privacy issues that have an impact on compliance and adherence to clinical procedures). However, the measurement of glucose levels requires the use of highly precise and sensitive sensors, given the low glucose concentration in sweat. Indeed, the normal range of glucose levels in blood is between 3.9 and 6.1 millimoles per liter (mM) [3], while in sweat, the concentration is considerably lower, ranging from 0.02 to 0.6 mM [4]. Therefore, it is imperative that sensors exhibit superior sensitivity.

Due to the social and economic relevance regarding the possible exploitation of sweat biosensors for this purpose, a patent landscape of the field is important to consider. A patent landscape, also known as patent mapping or state-of-the-art analysis, is an in-depth analysis of patents within a specific technology domain to systematically review and extract useful insights from patent search results. It assists researchers, investors, and policymakers in comprehending innovation trends and research and development (R&D) opportunities. This paper provides an overview of the patent landscape related to wearable biosensors for the monitoring of glucose levels in sweat.

2. Materials and Methods

Espacenet [5] was selected as the most appropriate database for the purposes of this research. The comprehensive coverage and free availability of the database are key factors in this decision. The bibliographic coverage is extensive, encompassing over 150 million documents, while the full text comprises more than 127 million documents.

In order to identify all relevant applications, we conducted a search on the Espacenet database using the following parameters: the Title Abstract Claims (TACs), the full-text fields, and some specific Cooperative Patent Classification (CPC) codes (G01N27/327, referred to biochemical electrodes and A61B5/1486 referred to devices for measuring bodily fluids using enzyme electrodes).

The final query, reported in Appendix A, may be copied and pasted into the Smart search field of Espacenet in order to retrieve the results obtained.

The data retrieved from Espacenet were imported into the Orbit Patent Intelligence platform v.2.0.0 [6] and subsequently analyzed. Orbit is equipped with a more extensive array of analytical tools for the examination of patent data in comparison to Espacenet.

The inclusion criteria set for patent selection were for time and typology of application:

• Patent applications or those granted using sweat as the biological fluid dedicated to the measurement of glucose levels.
• Patent applications or those granted filed in a 20-year period, starting from January 2005 and before 16 December 2024.
• Patents or patent applications pertaining to a wearable device designed for the continuous monitoring of glucose levels in sweat.

3. Results and Discussion

The search extracted a total of 115 records that were identified in Espacenet.

Following the elimination of 10 duplicate patents, a total of 105 patents were selected for title and abstract screening. After the selection, the abstract analysis of each patent allowed the exclusion of 18 patents concerning non-relevant applications like “solar cell,” “nanotube,” and/or “rGO”, originally included in the wide search string. Finally, following the screening process, a total of 87 records were selected for inclusion in this study (see Figure 1).

The first interesting outcome is in relation to a geographical analysis of the selected IP category. For this analysis, Priority Country (Figure 2a) and Protection Country (Figure 2b) databases are the relevant sources of information. China has the highest number of priority patent applications, with 50 out of 87 patents, followed by the USA with 28. Other countries turned out to be marginal. Therefore, it is interesting to note that if the patenting procedure widely adopted by assignees to extend priority filings is the PCT, in this case, the most frequently utilized system is the single country application, while the PCT strategy is the third used option with 26 entries (Figure 2a). Concerning the geographical distribution of the selected IP category, the top markets are China, the USA, and Europe. In contrast, Japan and South Korea exhibit a lower level of patent protection (Figure 2b). This could be somehow coherent with two factors: China and the US are the world-leading technology developers, while the USA and Europe are the most relevant market targets for the presence of metabolic diseases.

Another important analysis is related to the chronology of related patent applications: the graphs in Figure 3a,b illustrate the filing trend over the past two decades. The majority of applications were submitted subsequent to 2020, with a peak in filings observed in 2021. However, this trend underwent a decline in subsequent years. This new decreasing trend is probably due to the wide market penetration of the new CGM sensor generations released since 2014, based on the analysis of the interstitial fluids (today we are at the third generation). In a few years, they have become the reference devices thanks to their non-invasiveness and simplicity of use. After a 6-year period of being widely adopted by clinicians, the IP interest has been focusing on these devices, so this could be the most probable reason for the decrease in patent activity for sweat sensors and also for other invasive techniques after 2021.

The patents that were granted in 2021 and 2022 were predominantly Chinese, with two utility granted models among them.

Concerning the legal status of filings (Figure 4), a substantial proportion of the applications that are retrieved are found to be active—we found that only 18 out of 87 patents (20,69%) were abandoned; 11.5% have lapsed due to the failure to pay the annual fees, and 9.2% have been revoked by the Patent Office for failing to meet the patentability requirements or have been withdrawn by the applicant. Therefore, currently, in total, only 69 patents are still active. Among them, 31 (35.6%) patent applications have been granted, while 38 (43.7%) are currently under examination.

As illustrated in

Table 1. List of the top CPC codes.

CPC Codes	%	Definition
A61B5/145	57.47%	Measuring the concentration of an analyte in a body fluid
A61B5/14532	47.13%	For measuring glucose
A61B5/14517	36.78%	For sweat
G01N27/327	33.33%	Biochemical electrodes
A61B5/14546	26.44%	For measuring analytes not otherwise provided
A61B5/00	22.99%	Measuring for diagnostic purposes
A61B5/1477	22.99%	Non-invasive
A61B5/6833	18.39%	Adhesive patches
A61B5/1486	17.24%	Using enzyme electrodes (e.g., with immobilized oxidase)
A61B5/01	14.94%	Measuring temperature of body parts
A61B5/681	13.79%	Wristwatch-type devices

, the CPC codes for the 87-patent data are documented. The classification system is a tool that provides insight into the evolution of a technical field. The most frequently utilized patent classification systems are the International Patent Classification (IPC) [7] and the Cooperative Patent Classification (CPC) [8]. The IPC is a hierarchical classification system. The highest level comprises eight sections (A–H), which are further divided into 80,000 subdivisions. The updated version is 2025.01. The CPC represents an enhanced version of the IPC, featuring a greater number of subdivisions, a feature that is constantly augmented as novel technical domains emerge.

The majority of the leading patent assignees are Chinese or American universities and public research institutions, as illustrated in Figure 5.

Only 13% of active patents and patent applications have valuable geographic protection, defined as a patent family comprising a minimum of three members. The preferred validation countries or regions are China, the USA, Europe, and Japan. According to this, we tried to provide a potential assessment of the economic value of the 69 alive patents. We adopted the following two criteria:

Consider the most frequently cited applications: we have made the hypothesis that if a document is frequently cited in other patents, this means that it is a reference technology or application that is setting the state of the art to be improved.

Consider those that have a valuable geographic scope of protection intended as market relevance (the USA, Europe, and Asia, specifically China and Japan), together with a patent family comprising a minimum of three members.

According to these inclusion criteria, only 19% of active patents seemed to have a potential economic value. This is a general evaluation, and no market exploitation analysis was carried out.

3.1. Academic and Industrial Research

From the perspective of academic research or industrial R&D, the main trend in the field is the development of glucose sensors that are enzyme-based and functionalized with carbon or metal nanoparticles.

Non-invasive glucose monitoring techniques can be classified into four categories: (1) optical (photoplethysmography and surface-enhanced Raman scattering), (2) biochemical, (3) biomechanical (triboelectric and piezoelectric sensors), and (4) thermal (thermosensitive sensors and wearable thermoelectric generators) [9].

Biochemical sensing can be achieved through the implementation of two distinct approaches: the enzymatic approach, exemplified by first-to-third-generation glucose sensors, and the non-enzymatic approach, typified by fourth-generation glucose sensors.

In the case of an enzyme-based sweat sensor, the glucose oxidase enzyme catalyzes the oxidation of glucose, resulting in the formation of gluconic acid. Glucose can be oxidized by oxygen (first type), resulting in the production of hydrogen peroxide, by a redox mediator, such as Prussian blue or ferrocene (second type), or by engineered enzymes (third type). For instance, electrodes treated with carbon nanotubes (CNTs) can be coupled to glucose oxidase (GOx), enabling direct electron transfer from flavin adenine dinucleotide (FAD) to the electrodes [10,11,12,13].

In a non-enzymatic sweat sensor (fourth type), glucose reacts directly with nanomaterials, including metals, alloys, and metal oxides. These nanocatalysts function as a replacement for GOx, facilitating the direct oxidation of glucose on the electrode [14,15].

An analysis of patent data reveals a concerted effort in research and development aimed at the development of glucose sensors of the third and fourth types.

Graphene represents the most prevalent carbon material utilized in the electrode, followed by rGO and carbon nanotubes. The employment of MXenes and MOF is comparatively limited.

For example, in CN118830835A [16], the glucose sensor is made by depositing gold nanoparticles and titanium carbide multilayer nano-sheets on a screen-printed working electrode using cyclic voltammetry. Prussian blue is then deposited using cyclic voltammetry, and glucose oxidase (GOx) is dripped on to form a glucose sensor.

In CN106923842B [17], the glucose sensor is a graphene flexible patch sensor.

In CN118191059A [18], a polydimethylsiloxane/carbon nanotube/glucose oxidase sensor is claimed.

A patent of particular interest (No. CN115950942B) [19] pertains to the methodology of fabricating an organic electrochemical transistor, with the innovative technique being based on laser-induced graphene.

The technical scheme involves the utilization of dodecylbenzene sulfonic acid (DBSA) as a surfactant, which facilitates the effective combination of poly (3, 4-ethylenedioxythiophene) (PEDOT: PSS) with a porous laser-induced graphene (LIG) electrode. The sensing capability is achieved through the modification of platinum nanoparticles (PtNPs) and glucose oxidase (GOx) on the surface of a porous LIG gate.

The flexible, wearable, self-powered sensor system claimed in CN117269261A [20] comprises an anode and a cathode and is characterized by the following features: At the anode, the LIG nano-enzyme Au NPs with simulated glucose oxidase activity catalyze glucose oxidation in sweat, thereby generating electrons that reach the cathode through an external circuit. At the cathode, the LIG Pt NPs nano-enzyme with laccase-simulating activity obtains electrons and catalyzes oxygen reduction, thus forming a loop.

The glucose electrode described in CN112697857A [21] is constituted by a matrix electrode, and a polyaniline film and glucose oxidase film, which are sequentially coated on the surface of the aforementioned matrix electrode.

In US10722160B2 [22], the redox mediator of the non-invasive epidermal electrochemical sensor device includes glucose oxidase (GOx) or glucose dehydrogenase (GDH).

The smart wristband described in US20180263539A1 [23] comprises one or more sensors comprising glucose oxidase (GOx) and/or lactate oxidase (LOx) immobilized within a chitosan film.

In the year 2024, Chongqing Medical University patented a flexible non-enzymatic electrochemical sensor [24,25]. The sensor was synthesized using hydrothermal and one-pot preparation methods, incorporating gold nanoparticles (AuNPs) functionalized onto aminated multi-walled carbon nanotubes (AMWCNTs) as an efficient catalyst and crosslinked with carboxylated styrene butadiene rubber (XSBR) and PEDOT:PSS. Subsequently, the sensors were integrated onto screen-printed electrodes (SPEs) to create flexible glucose sensors (XSBR-PEDOT:PSS-AMWCNTs/AuNPs/SPE).

Cyclic voltammetry is the most frequently adopted electrochemical sensing technique, followed by chronoamperometry and potentiometry, as depicted in Figure 6.

In contrast to biochemical techniques, a variety of other sensing methodologies are employed, including piezoelectric, optical, thermosensitive, and acoustic techniques. However, these alternative methods are not as frequently utilized as the aforementioned biochemical approach.

From the viewpoint of the preferred typology of device, the development of sensors is predominantly focused on their integration within patches.

The operational stability of sensors was presented by only one patent. CN111624244 B Glucose oxidase nano capsule sensor and preparation and application thereof) ([26], page 11) is the sole patent that reports a stability study. The reference is located in paragraph [0080]: “Operating stability from cycle 1 to cycle 500 was measured by continuous cyclic voltammetry at a scan rate of 100 mV s⁻¹ in 0.1 m phosphate buffer saturated with nitrogen (pH 7.01). As shown in Figure 10 of ref. [26], nGOx/N-CNTs-Chi/GCE biosensor cycles 300 and 500 maintained 97% and 95.33% of the initial peak current, respectively. This strong operational stability of the nGOx/N-CNTs-Chi/GCE biosensor can be attributed to the good loading of nGOx on the modified electrode, indicating that the nGOx molecules on the enzyme electrode are stable in long-term operation and do not leach from the modified electrode”.

In a second patent—in WO2022/054044 A1 (Oxygen-insensitive amperometric biosensors)—inventors asserted that the strong interactions of DCNQ, TeGDH, and polydopamine with the $\pi -\pi$ of MWCNTs result in enhanced stability of the biosensor.

Finally, a technical insight can be related to the lifetime improvement of sensors. In WO2021/148952 A1 (Nonenzymatic electrochemical sensors) [27], it is stated that the utilization of multiple electrodes of a similar nature has been demonstrated to enhance the longevity of an ESS. Consequently, each working electrode is utilized for a predetermined duration (e.g., until the electrode ceases to respond), at which point the system transitions to the utilization of a subsequent working electrode, thereby prolonging the lifespan of the sensor.

3.2. IP Exploitation and Patented Products on the Market

Regarding the patent exploitation, we can affirm that there is a lack of commercialized non-invasive glucose monitors despite an interesting patent activity. In our opinion, most of the IP outcomes failed to achieve a successful technological transfer and market penetration because of the occurrence of several factors at the same time:

• The minor reliability of the technologies based on sweat with respect to the blood sampling techniques: Even if the performances are quite good, the lack of selectivity in sweat is still shown and the sensitivity is still to be improved considering low glucose concentration, ranging from 0.02 to 0.6 mM [4], to reach the same accuracy.
• The occurrence of new long-term sensors deployed in the market: Continuous Glucose Monitoring through a small patch able to measure glucose levels in the interstitial fluid for up to 14 days is today the gold-standard reference of the technology. With this system, blood sampling is no longer required, and the mobile app can directly read the glucose levels with a simple gesture. Its reliability and simplicity of use are far from being achieved by sweat analysis devices. These systems were first introduced to the market in fall 2014. Its clinical adoption was gradual but rapidly grew for its revolution in the users’ lifestyles. With the new CGM sensor generations released to the market (today we are at the third generation), after a few years, they have become the reference devices; this is the most probable reason for the decrease in patent activity for sweat sensors after 2021.
• Medical resilience is another barrier: Doctors are quite stable in their device adoption because they have also developed efficient diabetes management strategies for categories of patients (with the possibility of customization programs) with large numbers of subjects. The availability of a new technology implies a long period of knowledge acquisition and testing with several patients before reaching the same level of awareness and efficacy in the development of customizable clinical programs.
• The regulatory constraints are the final remark to be considered: The new MDR 645/2017 regulation—entered into force in 2021 in Europe—requires more investments and activity to fulfill its obligations.

From an investor’s perspective, investing in wearable devices for glucose monitoring through sweat is a risky investment. This is due to the fact that the filing trend is in decline and because the products on the market do not measure glucose but rather other parameters, such as lactate, sodium, and potassium. This is a change in perspective and destination that could offer different exploitation pathways. Some examples follow.

The Smart Patch [28], distributed by Epicore, is a microfluidic wearable device that captures sweat and analyzes fluid and electrolyte losses during exercise.

The OnaVital wearable device [29], manufactured by Onalabs, is a certified piece of medical equipment that facilitates the non-invasive, continuous, and real-time monitoring of vital signs, including blood pressure, oxygen saturation, heart rate, and skin temperature.

The technology developed by Limina^TM [30] enables the measurement of hypoxanthine metabolite levels in an athlete’s sweat.

The NIX hydration biosensor [31], when used in conjunction with the NIX Solo App, has been demonstrated to quantitatively measure fluid and electrolyte losses in real time. The app provides precise information regarding the timing, quantity, and type of fluid intake necessary to maintain optimal hydration levels.

The HDROP wearable sensor [32] has been developed to track more than just sweat rate and sweat loss. It has also been designed to monitor the loss of potassium and sodium, as well as the body’s temperature.

3.3. Patent Strategies and Freedom-to-Operate Analysis

From the perspective of technology transfer, the patent strategy is of paramount importance for the exploitation of the Intellectual Property. This strategy review can be conducted jointly with the Freedom-to-Operate (FTO) analysis that follows the following general principles: (a) The subject matter of a rejected patent application is freely usable. (b) A patent is a territorial right, meaning that it is valid only in the country in which it is filed and granted—if a national patent is not extended abroad, it cannot be enforced. For instance, if a patent is valid only in China and not extended abroad, the subject matter of that patent could be commercialized in all countries except China. In this case, there would be no infringement. A brief but interesting FTO analysis in accordance with the previous geographical analysis reveals that the FTO is available in most countries for the CN patent applications that are not extended with an international filing as per PCT procedures or with a regional filing (European procedure); therefore, this opens a large free market outside of Chinese borders. This is probably due to their priority for industrial interests in production and licensing or selling abroad. A specific analysis of interesting cases is reported below.

The Technion R&D Foundation has implemented an effective patent strategy, having initiated two US provisional applications in 2020 pertaining to an amperometric enzyme-based biosensor. These provisional applications were subsequently consolidated into a unified PCT extension in 2021 [33], followed by validation in both the US and EP applications. The invention pertains to a carbon-based electrode for utilization in a biosensor, wherein the matrix material is polydopamine or agarose. The carbon allotrope is selected from a range of structures, including CNTs, fullerenes, and carbon nanobuds. The redox charge mediator is selected from substituted naphthoquinones. Glucose dehydrogenase (GDH) is the enzyme used.

An illustration of a suboptimal patent strategy is provided by a patent application filed by Yangzhou University on a flexible glucose electrochemical sensor [34]. A scholarly article on the topic was published on 23 November 2021 [35]. The earliest priority date of the application in question is 16 March 2022. Consequently, the patent examiner determined that the application lacked novelty, as evidenced by the disclosure. The invention is currently in the public domain, and as such, it is available for use by anyone.

An illustration of inadequate geographic protection scope is provided by the patent issued for an organic electrochemical transistor based on laser-induced graphene, which was filed by the Nanjing University of Technology. The patent was filed in 2022 and granted in China [19] but not extended abroad. A paper was published on 16 March 2023 on this topic [36]. This has resulted in the fact that the implementation of this technology in a wearable device is a viable commercialization strategy, with the exception of within the Chinese market.

This recommendation is directed towards researchers and entrepreneurs, emphasizing the crucial nature of conducting a Freedom-to-Operate (FTO) search prior to the commercialization of novel technologies.

4. Conclusions

China has the highest number of patent filings, characterized by a narrow scope of geographic protection.

A decline in filings was observed in 2023, following a period of significant growth that peaked in 2021.

A significant proportion of the retrieved patents and patent applications, constituting 80% of the total, are still active and remain subject to consideration for an FTO search.

A review of the most frequently cited applications, in conjunction with those that have a minimum geographic scope of protection, reveals that only approximately 19% of active patents appear to possess a potential economic value.

A review of the current market reveals that only five products have been granted a patent (though none of these products are designed for glucose measurement).

Glucose sensors that are enzyme-based represent the most prevalent type of glucose sensor in current patent filings.