WO2011091363A2

WO2011091363A2 - Accuracy improving desiccants

Info

Publication number: WO2011091363A2
Application number: PCT/US2011/022258
Authority: WO
Inventors: Amy H. Chu; Mary Ellen Warchal-Windham
Original assignee: Bayer Healthcare Llc
Priority date: 2010-01-22
Filing date: 2011-01-24
Publication date: 2011-07-28
Also published as: JP6751067B2; WO2011091363A3; CN105603045B; RU2569753C2; EP2526417A4; JP2018031788A; CN102713608A; KR101783067B1; JP2013518255A; CN105603045A; CN102713608B; CA2786154A1; RU2012136132A; BR112012017876A2; MX2012008427A; KR20120118046A; JP6464067B2; JP2016026296A; CA2786154C; EP2526417A2

Abstract

A biosensor system for determining the concentration of an analyte in a sample includes a plurality of test sensors, and includes a container including a desiccant. When the plurality of test sensors is sealed in the container for two weeks at a temperature of 50°C and then removed from the container, and each test sensor is subsequently connected through the at least two conductors to a measurement device and then contacted with one of a plurality of samples including an analyte, where the plurality of samples have an analyte concentration that spans the range of 10 mg/dL - 600 mg/dL, and the analyte concentration in each sample is measured by the test sensor and the measuring device, the bias of each measured analyte concentration is within ±10 mg/dL or ±10%.

Description

ACCURACY IMPROVING DESICCANTS

REFERENCE TO RELATED APPLICATIONS

[001] This application claims the benefit of U.S. Provisional Application No.

61/297,515 entitled "Accuracy Improving Desiccants" filed January 22, 2010, which is incorporated by reference in its entirety.

BACKGROUND

[002] Biosensors provide an analysis of a biological fluid, such as whole blood, serum, plasma, urine, saliva, interstitial, or intracellular fluid. Typically, biosensors have a measurement device that analyzes a sample residing in a test sensor. The sample usually is in liquid form and may be a biological fluid or a derivative of a biological fluid, such as an extract, a dilution, a filtrate, or a reconstituted precipitate. The analysis performed by the biosensor determines the presence and/or concentration of one or more analytes in the biological fluid. Examples of analytes include alcohol, glucose, uric acid, lactate, cholesterol, bilirubin, free fatty acids, triglycerides, proteins, ketones, phenylalanine, or enzymes. The analysis may be useful in the diagnosis and treatment of physiological abnormalities. For example, a diabetic individual may use a biosensor to determine the glucose level in whole blood, and this information may be used in adjusting the individual's diet and/or medication.

[003] Biosensors may be designed to analyze one or more analytes and may use different sample volumes. Some biosensors may analyze a single drop of whole blood, such as from 0.25-15 microliters (JJL) in volume. Biosensors may be implemented using bench-top, portable, and like measurement devices. Portable measurement devices may be hand-held and allow for the identification and/or quantification of one or more analytes in a sample. Examples of portable measurement devices include the BREEZE® and CONTOUR® meters of Bayer HealthCare in Tarrytown, New York, while examples of bench-top measurement devices include the Electrochemical Workstation available from CH Instruments in Austin, Texas. [004] In electrochemical biosensors, the analyte concentration is determined from an electrical signal generated by an oxidation/reduction or redox reaction of the analyte, or of a species responsive to the analyte, when an input signal is appl ied to the sample. The input signal may be appl ied as a single electrical pu lse or in multiple pu lses, sequences, or cycles. A redox substance, such as a med iator, an enzyme or simi lar species, may be added to the sample to enhance the electron transfer from a first species to a second species during the redox reaction. The redox substance(s) may react with a single analyte, thus providing specificity to a portion of the generated output signal .

[005] Electrochemical biosensors usual ly incl ude a measurement device having electrical contacts that connect with electrical conductors in the test sensor. The test sensor may be adapted for use outside, inside, or partial ly inside a l iving organ ism. When used outside a l iving organism, a sample of the biological fl u id is introduced into a sample reservoir in the test sensor. The test sensor may be placed in the measurement device before, after, or during the introduction of the sample for analysis. When inside or partial ly inside a l iving organism, the test sensor may be continual ly immersed in the sample, or the sample may be introduced interm ittently to the test sensor. The test sensor may incl ude a reservoir that partial ly isolates a vol ume of the sample, or the test sensor may be open to the sample. Simi larly, the sample may continuously flow through the test sensor or be interrupted for analysis.

[006] The test sensor may be formed by d isposing or printing electrodes on an insulating substrate by d isposing one or more reagent compositions on one or more of the conductors. More than one of the conductors may be coated by the same reagent composition, such as when the working and counter electrodes are coated by the same composition. Multiple techniques known to those of ord inary ski l l in the art may be used to dispose the reagent composition on the test sensor. The reagent composition may be d isposed on the conductors as a reagent fl u id and then dried. When the sample is introduced to the test sensor, the reagent composition begins to rehydrate. [007] The reagent compositions disposed on each conductor may be the same or different. Thus, the reagent composition of the working electrode may contain an enzyme, a mediator, and a binder while the reagent composition of the counter electrode may contain only a mediator, which could be the same as or different from the mediator of the working electrode, and a binder. The reagent composition may include an ionizing agent for facilitating the oxidation or reduction of the analyte, such as an oxidoreductase enzyme, as well as any mediators or other substances that assist in transferring electrons between the analyte and the working electrode.

[008] One or more components of a reagent composition may undergo a chemical transformation prior to use of the test sensor. In particular, it is believed that the oxidation state of the mediator may change over time under certain conditions.

Mediators such as ferricyanide and organic quinones and hydroquinones may undergo reduction in the presence of water. The presence of reduced mediator in the reagent composition can cause an increase in background current of the sensor, leading to inaccurate assay results, particularly for samples with low analyte concentration.

[009] Typically, undesirable and/or premature chemical transformations in the reagent composition are inhibited by storing the test sensor in proximity to a desiccant. Desiccants typically are used in test sensor primary packaging, such as bottles or foil pouches, to prevent degradation of the reagent composition so as to maintain the desired shelf life of the test sensor. Conventional desiccants for test sensor storage systems can quickly adsorb moisture that may leak into the package containing the test sensor. Examples of desiccants used to protect test sensors include molecular sieves, which quickly adsorb moisture even in low humidity environments.

[0010] A drawback to the protection of test sensors with a desiccant is that one or more components of the reagent composition may require a threshold level of moisture to retain their function in the composition. For example, the FAD dependent Glucose Dehydrogenase enzyme (FAD-GDH) is believed to require some residual moisture to maintain its native active configuration. Depletion of moisture from the reagent composition below the threshold level could lead to enzyme conformational change and inactivation.

[0011] Loss of enzyme activity due to excessive desiccation of the test sensor typically is addressed either by including excess amounts of enzyme in the reagent composition or by adding a substance that is believed to stabilize the enzyme to the reagent composition. Examples of substances that may stabilize the enzyme in a test sensor reagent composition include sugars such as trehalose or sucrose, and sugar alcohols such as mannitol, maltitol or sorbitol. These substances may be used in a lyophilization process to preserve enzyme activity. See, for example EP 1 785 483 A1 . High loadings of the enzyme or of other solids such as stabilizers can present other difficulties, however. Since the enzyme component typically is expensive, it is not desirable to increase the enzyme loading beyond the level needed for the assay. In addition, the enzyme or other solids can slow down the rehydration of the reagent composition by the sample, resulting in longer assay times, especially at lower temperatures. Excess enzyme in the test sensor, beyond that required for interaction with the analyte and/or other ingredients in the reagent composition such as the mediator, also may reduce the accuracy of the sensor.

[0012] Accordingly, there is an ongoing need for improved biosensor systems, especially those that may provide increasingly accurate and/or precise determination of the concentration of the analyte in the sample, and/or that may provide increasingly shorter analysis times. Moreover, there is a need for improved biosensor systems that have an increased shelf life over a wider range of storage conditions, while supplying the desired accuracy, precision and/or analysis time. The systems, devices, and methods of the present invention overcome at least one of the disadvantages associated with conventional biosensor systems.

SUMMARY

[0013] In one aspect, the invention provides a biosensor system for determining the concentration of an analyte in a sample that includes a plurality of test sensors. Each test sensor includes at least two conductors, where one of the conductors is a working electrode, and further includes a reagent composition disposed on or near the working electrode. The biosensor system further includes a container including a desiccant. When the plurality of test sensors is sealed in the container for two weeks at a temperature of 50°C and then removed from the container, and each test sensor is subsequently connected through the at least two conductors to a measurement device and then contacted with one of a plurality of samples including an analyte, where the plurality of samples have an analyte concentration that spans the range of 10 mg/dL - 600 mg/dL, and the analyte concentration in each sample is measured by the test sensor and the measuring device, the bias of each measured analyte concentration is within ±10 mg/dL for samples having an analyte concentration less than 100 mg/dL and within ±10% for samples having an analyte concentration of at least 100 mg/dL.

[0014] In another aspect, the invention provides a biosensor system for determining the concentration of an analyte in a sample that includes a plurality of test sensors. Each test sensor includes at least two conductors, where one of the conductors is a working electrode, and further includes a reagent composition disposed on or near the working electrode, where the reagent composition includes a redox enzyme. The biosensor system further includes a container including a desiccant. When the plurality of test sensors is sealed in the container for two weeks at a temperature of 50°C and then removed from the container, the reagent composition of each test sensor retains at least 75% of the activity of the redox enzyme.

[0015] The scope of the present invention is defined solely by the appended claims and is not affected by the statements within this summary.

BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The invention can be better understood with reference to the following drawings and description. The components in the figures are not necessarily to scale, emphasis instead being placed upon illustrating the principles of the invention. [0017] FIGs. 1 A - 1 C represent the output signals from test sensors for whole blood samples having glucose concentrations of 400 mil l igrams per decil iter (mg/dL). The test sensors were sealed with molecular sieve desiccant (1 A), silica gel desiccant (1 B) or no desiccant (1 C).

[0018] FIGs. 2A and 2B represent graphs of assay bias for glucose assays of whole blood samples having glucose concentrations of 50, 100, 400 or 600 mg/dL.

[0019] FIGs. 3A and 3B represent graphs of background current for glucose assays of whole blood samples containing no glucose, for test sensors sealed in containers having varying types and levels of desiccant.

[0020] FIG. 4 represents a graph of in-sensor enzyme activity for test sensors stored for two weeks either at -20°C, at 50°C or at room temperature, in containers having varying types and levels of desiccant.

[0021] FIG. 5 represents graphs of in-sensor enzyme activity ("% enzyme recovery") for test sensors sealed for two weeks either at 50°C with varying types of desiccant, and for reagent compositions with and without an enzyme stabil izer.

[0022] FIG. 6 represents graphs of the variation of the R5/4 ratio parameter for test sensors stored for two weeks at 50°C, relative to the R5/4 ratio parameter for test sensors stored for two weeks at -20°C, where the test sensors had varying levels of enzyme density above the working electrode of the test sensors.

[0023] FIG. 7 depicts a schematic representation of a biosensor that determines an analyte concentration in a sample of a biological fluid using a test sensor.

[0024] FIG. 8 depicts a sealed container containing a desiccant and a plural ity of test sensors. DETAILED DESCRIPTION

[0025] A biosensor system includes test sensors sealed in a container having a desiccant that retains a residual moisture level in the container. In low humidity environments, the desiccant does not quickly absorb moisture, which can allow the reagent composition of the test sensors to maintain a moisture level conducive to maintaining an enzyme in its active configuration. Test sensors stored in a container that includes such a desiccant can provide measurements of analyte concentration that are more accurate and/or precise than those of comparable test sensors stored in a container that includes a conventional desiccant or no desiccant. Thus, the test sensors having consistently accurate assays with fast assay times, even when the test sensors are stored for long periods of time under non-optimal conditions.

[0026] A biosensor system includes a plurality of test sensors, each test sensor including at least two conductors, where one of the conductors is a working electrode, and a reagent composition disposed on or near the working electrode. The biosensor system further includes a container including a desiccant. The plurality of test sensors is sealed in the container.

[0027] The desiccant in the container preferably adsorbs at most 1 5% of its weight in water when in contact with an environment of 10% - 20% relative humidity (RH) at 40°C. More preferably the desiccant adsorbs at most 10% of its weight in water when in contact with an environment of 10% - 20% RH at 40°C. More preferably the desiccant absorbs from 5% - 10% of its weight in water when in contact with an environment of 10% - 20% RH at 40°C.

[0028] An example of a desiccant that absorbs from 5% - 10% of its weight in water when in contact with an environment of 10% - 20% RH at 40°C includes silica gel. Silica gels can adsorb moisture at a level roughly proportional to the relative humidity of the surrounding environment for RH values of 0% to approximately 60%. In contrast, the molecular sieve desiccants conventionally used in test sensor containers can adsorb large amounts of moisture quickly from environments having 10% - 20% RH. Molecular sieves can adsorb 15% to 20% of their weight in water when in contact with an environment of 5% RH at 40°C, and then may adsorb minimal amounts of additional moisture as the relative humidity increases.

[0029] An example of a desiccant that can absorb at most 15% of its weight in water when in contact with an environment of 10% - 20% RH at 40°C includes a composition of polymer-blended molecular sieves. The efficacy of a desiccant can be lowered by blending the desiccant with a polymer. As the desiccant in the polymer is only partially exposed to the environment, moisture adsorption can occur at a rate that is slower than the adsorption rate of the pure desiccant. Another example of a desiccant that can absorb at most 15% of its weight in water when in contact with an environment of 10% - 20% RH at 40°C includes a blend of molecular sieves with silica gel. The selection of the types and relative amounts of molecular sieves and silica gel in the blend may allow for tailoring of the total moisture adsorbed by the blended composition at low relative humidity.

[0030] FIGs. 1 A through 1 C show the output signals from test sensors for whole blood samples having glucose concentrations of 400 milligrams per deciliter (mg/dL) and having hematocrit contents of 40%. The test sensors were sealed in a container having either 22.5 mg per test sensor of the conventional desiccant "molecular sieve 1 3x" (FIG. 1 A), 30 mg per test sensor of silica gel (FIG. 1 B), or having no desiccant (FIG. 1 C). For each type of container, half of the containers were stored at 50°C for two weeks, and half were stored at -20°C for two weeks. The heat stress environment of two weeks at 50°C is an accelerated stress condition typically used to assess the performance of a biosensor at the end of its shelf-life. After the storage period, the test sensors were used to perform an electrochemical assay of a whole blood sample.

[0031] The signal input to the test sensors by the measurement device was a gated amperometric pulse sequence and one or more output current values were correlated with the analyte concentration of the sample, such as described in U.S. Patent Pub. 2008/01 73552, and in U.S. Patent Pub. 2009/0145779. The disclosures of these patent applications regarding gated amperometric pulse sequences and the correlation of output current values with analyte concentrations are herein incorporated by reference. The pulses used to generate the graphs of FIGS 1 A-1 C included eight excitations separated by seven relaxations. The second through eighth excitations were about 0.4 second in duration, and the second through seventh relaxations were about 1 second in duration. Three output current values were recorded during the second through eighth excitations.

[0032] A correlation of one or more output current values with the analyte

concentration of the sample may be prepared by plotting the output current at a particular time in the analysis against a known concentration of the analyte in a series of stock solutions containing the analyte. To correlate the output current values from the input signal with the analyte concentration of the sample, the initial current value from the excitation is preferably greater than those that follow in the decay. Preferably, the output current value or values correlated with the analyte concentration of the sample are taken from a decay including current data reflecting the maximum kinetic performance of the test sensor. The kinetics of the redox reaction underlying the output currents are affected by multiple factors. These factors may include the rate at which the reagent composition rehydrates, the rate at which the enzyme system reacts with the analyte, the rate at which the enzyme system transfers electrons to the mediator, and the rate at which the mediator transfers electrons to the electrode.

[0033] The maximum kinetic performance of the test sensor may be reached during an excitation of a gated amperometric pulse sequence when the initial current value of an excitation having decaying current values is greatest for the multiple excitations. Preferably, the maximum kinetic performance of a test sensor is reached when the last in time current value obtained for an excitation having decaying current values is the greatest last in time current value obtained for the multiple excitations. More

preferably, the maximum kinetic performance of a test sensor is reached when the initial current value of an excitation having decaying current values is greatest for the multiple excitations and the last in time current value obtained for the same excitation is the greatest last in time current value obtained for the multiple excitations. The maximum kinetic performance may be reached at the first excitation having decaying current values, or it may be reached at a subsequent excitation, such as the second, third or later excitation having decaying current values.

[0034] The maximum kinetic performance can be described in terms of the parameter "peak time", which is the time at which an electrochemical test sensor obtains its maximum output current value after a sample containing an analyte contacts the test sensor. The maximum output current value is preferably used for correlation with the analyte concentration of the sample. Preferably the peak time for a test sensor is less than about 7 seconds, and more preferably less than about 5 seconds, of introducing the sample to the test sensor. Preferably, the peak time is within about 0.4 to about 7 seconds, more preferably within about 0.6 to about 6.4 seconds, more preferably within about 1 to about 5 seconds, more preferably within about 1 .1 to about 3.5 seconds of introducing the sample to the test sensor.

[0035] Referring to FIG. 1 A, the test sensor that had been sealed in a container having the conventional desiccant had a longer peak time after being stored at 50°C for two weeks than after being stored at -20°C for two weeks. In contrast, the sensors sealed either with sil ica desiccant (FIG. 1 B) or with no desiccant (FIG. 1 C) had no increase in their peak times when stored at 50°C for two weeks relative their storage at - 20°C for two weeks.

[0036] Any change in the current profile of a test sensor can lead to inconsistent glucose assay results, as test sensor glucose results typical ly are derived from the measured current at a fixed time point. This increased inaccuracy is especial ly evident for assays performed at shorter times such as 10 seconds or less. For the test sensors examined for FIGs. 1 A - 1 C, the change in the current profile for the test sensors sealed with the conventional desiccant resulted in an undesirable increase in bias of the biosensor.

[0037] The measurement performance of a biosensor is defined in terms of its accuracy and/or precision. Increases in accuracy and/or precision provide for an improvement in measurement performance of the biosensor. Accuracy may be expressed in terms of bias of the biosensor's analyte reading in comparison to a reference analyte reading, with larger bias values representing less accuracy. Precision may be expressed in terms of the spread or variance of the bias among multiple analyte readings in relation to a mean. Bias is the difference between one or more values determined from the biosensor and one or more accepted reference values for the analyte concentration in the biological fluid. Thus, one or more errors in the measured analysis results in the bias of the determined analyte concentration of a biosensor system. Bias may be expressed in terms of "absolute bias" or "percent bias", depending on the analyte concentration in the sample. Absolute bias may be expressed in the units of the measurement, such as mg/dL, and may be used for analyte concentrations less than 1 00 mg/dL. Percent bias may be expressed as a percentage of the absolute bias value over the reference value, and may be used for analyte concentrations of at least 100 mg/dL. Accepted reference values may be obtained with a reference instrument, such as the YSI 2300 STAT PLUS™ glucose analyzer available from YSI Inc., Yel low Springs, Ohio.

[0038] FIGs. 2A and 2B depict graphs of bias for glucose assays of whole blood samples having hematocrit contents of 40% and having glucose concentrations of 50, 1 00, 400 or 600 mg/dL. The test sensors used in the analysis were sealed in containers having from 0 to 22.5 mg per test sensor of the conventional desiccant molecular sieve 1 3x (FIG. 2A), or containing from 0 to 30 mg per test sensor of sil ica gel, and were stored at 50°C for two weeks.

[0039] Without desiccant (0 mg desiccant / test sensor), the blood glucose assays after the test sensor heat stress had a positive bias of 1 5 mg/dL for samples containing low glucose (50 mg/dL), 7-10% bias for samples having glucose concentrations of 1 00 mg/dL and 400 mg/dL, and had almost no bias for samples containing high glucose (600 mg/dL). Seal ing the test sensors with the conventional molecular sieve desiccant (FIG. 2A) corrected the positive bias for samples with low and normal glucose; however, the assay bias for samples with 600mg/dL glucose increased to -10% and -1 5% as the desiccant level increased. In contrast, the assay bias for sensors stored with 30 mg / sensor silica gel was within 5 mg/dL for samples having less than 100 mg/dL glucose, and was within ±5% for samples having 100 mg/dL to 600 mg/dL glucose (FIG. 2B).

[0040] The increase in assay peak time and in assay bias for test sensors sealed at 50°C for two weeks in the presence of a conventional desiccant is surprising when compared to the results for similarly treated test sensors sealed with no desiccant or with the weaker desiccant of sil ica gel. Typical ly, desiccants have been used to prevent transformations of components of the reagent layer, including the mediator, prior to use of the test sensor. Thus, it would be unexpected that storage of a test sensor with a conventional desiccant would impair the test sensor's accuracy and/or its shelf-l ife, relative to that of a comparable test sensor stored with no desiccant or with a less aggressive desiccant, especial ly when analyzing samples having high glucose

concentrations.

[0041] For a biosensor system that includes a plural ity of test sensors sealed in a container having a desiccant, the system may be evaluated by using the test sensors to measure the analyte content of samples having known concentrations of the analyte that span a certain range of concentrations, and then calculating the bias of the

measurements with regard to the actual concentrations. In one example, a plural ity of test sensors is sealed in a container including a desiccant for two weeks at a temperature of 50°C, where each test sensor includes at least two conductors, one of which is a working electrode, and a reagent composition disposed on or near the working electrode. The test sensors are then removed from the container, and each test sensor is connected through the at least two conductors to a measurement device. Once connected, each test sensor is contacted with one of the samples and used to measure the analyte concentration in the sample. In this example, for samples having an analyte concentration that spans the range of 10 mg/dL - 600 mg/dL, the bias of each measured analyte concentration preferably is within ±1 0 mg/dL for samples having an analyte concentration less than 100 mg/dL and within ±10% for samples having an analyte concentration of at least 100 mg/dL. The phrase "an analyte concentration that spans the range of 10 mg/dL - 600 mg/dL" means that at least one of the samples has an analyte concentration of 1 0 mg/dL, and at least one of the other samples has an analyte concentration of 600 mg/dL. The remaining samples, if any, may have analyte concentrations between 10 mg/dL and 600 mg/dL.

[0042] In the above example, the bias of each measured analyte concentration preferably is within ±7 mg/dL for samples having an analyte concentration less than 100 mg/dL and within ±7% for samples having an analyte concentration of at least 1 00 mg/dL. More preferably the bias of each measured analyte concentration is within ±5 mg/dL for samples having an analyte concentration less than 100 mg/dL and within ±5% for samples having an analyte concentration of at least 1 00 mg/dL. Preferably, in this example the number of test sensors in the plural ity is at least 1 0, and preferably is at least 25, at least 50, or at least 100. Preferably, in this example the samples have an analyte concentration that spans the range of 50 mg/dL - 600 mg/dL.

[0043] For a biosensor system that includes a plural ity of test sensors sealed in a container having a desiccant, the system may be evaluated by using the test sensors to measure the analyte content of samples having a known concentration of the analyte, and then calculating the coefficient of variance (%CV) of the measurements. In the above example, the %CV for each measured analyte concentration is at most 2.5%. More preferably, in this example the %CV for each measured analyte concentration is at most 2%.

[0044] Table 1 l ists the %CV for glucose assays of whole blood samples having hematocrit contents of 42% and having glucose concentrations of 50, 1 00, 400 or 600 mg/dL. The test sensors used in the analysis were sealed in containers having from 0 to 22.5 mg per test sensor of the conventional desiccant molecular sieve 1 3x, or containing from 0 to 30 mg per test sensor of sil ica gel, and were stored at 50°C for two weeks. Each result listed is based on 10 test sensors. Table 1 - Assay Precision for Test Sensors Heat Stressed at 50°C for 2 Weeks

Desiccant %CV (n = 10) for glucose concentrations:

Type Amount 50 mg/dL 100 mg/dL 400 mg/dL 600 mg/dL

(mg/sensor)

None 0 1.9 1.8 2.4 1.3

7.5 2.4 4.9 1.5 2.8

Molecular

Sieves 22.5 2.9 2.4 2.1 2.0

10.0 3.3 1.6 2.5 1.4

Silica gel

30.0 1.5 1.1 1.3 1.1

[0045] Table 2 lists the %CV for glucose assays as described for Table 1, but where the test sensors were stored at -20°C for two weeks. Each result listed is based on 10 test sensors.

Table 2 - Assay Precision for Test Sensors Stored at -20°C for 2 Weeks

Desiccant %CV (n = 10) for glucose concentrations:

Type Amount 50 mg/dL 100 mg/dL 400 mg/dL 600 mg/dL

(mg/sensor)

None 0 2.9 2.9 1.6 1.0

7.5 1.3 1.5 3.4 1.3

Molecular

Sieves 22.5 5.8 4.2 1.9 1.5

10.0 1.8 3.3 1.6 1.3

Silica gel

30.0 1.8 2.1 2.0 1.3

[0046] Without desiccant (0 mg desiccant / test sensor), the blood glucose assays after the test sensor heat stress (2 weeks at 50°C) had coefficients of variance of 1.3 - 2.4% for samples having analyte concentrations that spanned the range of 50 mg/dL - 600 mg/dL. Sealing the test sensors with the conventional molecular sieve desiccant (7.5 or 22.5 mg / test sensor) or with 10.0 mg / test sensor silica gel did not reduce the upper l imit of %CV for the blood glucose assays. Seal ing the test sensors with 30.0 mg / test sensor silica gel, however, did reduce the upper l imit of %CV for the blood glucose assays to 1 .5% . A similar trend was also measured for blood glucose assays after the test sensors had been sealed at -20°C for 2 weeks. For both sets of storage conditions, blood glucose assays performed using test sensors sealed with 30.0 mg / test sensor sil ica gel had %CV values below 2.1 % for samples having analyte concentrations that spanned the range of 50 mg/dL - 600 mg/dL.

[0047] FIGs. 3A and 3B depict graphs of background current for glucose assays of whole blood samples containing no glucose. The test sensors used in the analysis were sealed in a container containing from 0 to 22.5 mg per test sensor of the conventional desiccant molecular sieve 1 3x (FIG. 3A), or containing from 0 to 30 mg per test sensor of silica gel (FIG. 3 B), and were stored for two weeks at -20°C, at room temperature ("RT", 25°C) or at 50°C. As the samples contained no glucose, the measured

background current is due to the presence of substances in reduced oxidation states, such as reduced mediator.

[0048] Test sensors stored without desiccant in the container showed a large increase in biosensor background current after the heat stress. This was consistent with the conventional theory that a desiccant is important to maintain low background current in the test sensor, l ikely by preventing auto-reduction of the mediator. An increase in sensor background current may have contributed to the positive assay bias for samples with low glucose represented in FIGs. 2A and 2B. Test sensors stored in the presence of the conventional molecular sieve desiccant (FIG. 3A) required less of the desiccant to maintain a low background current than did test sensors stored in the presence of sil ica gel (FIG. 3B). Thus, the conventional desiccant appeared to accompl ish its intended function of inhibiting premature reduction of the mediator.

[0049] The mediator in the reagent compositions of the test sensors used in FIGs. 1 through 6 was the two electron transfer mediator 3-(2',5'-disulfophenyl imino)-3H- phenothiazine bis-sodium salt. The observed effects of moisture during storage of test sensors are believed to apply to other two electron transfer mediators, such as other organic quinones and hydroquinones. Examples of such mediators include phenathroline quinone; phenothiazine and phenoxazine derivatives, such as

3-phenylimino-3H-phenothiazines (PIPT) and 3-phenylimino-3H-phenoxazines (PIPO); 3-(phenylamino)-3H-phenoxazines; phenothiazines; and 7-hydroxy-9,9-dimethyl-9H- acridin-2-one and its derivatives. The observed effects of moisture during storage of test sensors also are believed to apply to one electron transfer mediators such as 1,1'- dimethyl ferrocene, ferrocyanide and ferricyanide, ruthenium(lll) and ruthenium(ll) hexaamine.

[0050] One possible explanation for the surprising results regarding peak time, bias and/or precision is that a less aggressive desiccant can protect the enzyme at a level that is unexpectedly high. A less aggressive desiccant, such as silica gel, appeared to be more compatible with the FAD-GDH enzyme than the conventional desiccants, yet still provided sufficient protection for the mediator. The impact of the loss of enzyme activity on the assay bias may have been underestimated previously, particularly for high glucose samples.

[0051] FIG.4 depicts a graph of in-sensor FAD-GDH enzyme activity for test sensors sealed for two weeks either at -20°C (diamond symbols), at 50°C (triangle symbols) or at room temperature (square symbols), in containers having varying types and levels of desiccant. The solid symbols correspond to conventional molecular sieve desiccant, and the open symbols correspond to silica gel desiccant. Neither desiccant appeared to allow loss of enzyme activity at -20°C. There was an approximately 10% loss in sensor enzyme activity after storing the sensors at 50°C for two weeks for sensors packaged without a desiccant (0 mg desiccant / sensor). The enzyme activity decreased to approximately 60% for sensors packaged with the molecular sieve (solid triangle symbols), even at relatively low levels of 7 mg desiccant per sensor. In contrast, the enzyme activity for sensors packaged with silica gel was higher by approximately 25%, maintaining enzyme activities of 75-80% (open triangle symbols). Even at room temperature, test sensors stored with the molecular sieve (solid square symbols) showed enzyme activities that were approximately 5% lower than test sensors stored with silica gel (open square symbols).

[0052] The results of FIG. 4, combined with the results of FIGs. 1 through 3, are consistent with an analysis that the FAD-GDH enzyme requires a threshold level of moisture to maintain its native structure and activity. The increase in negative bias with increasing molecular sieve desiccant for 600 mg/dL glucose (FIG. 2A) correlated with an approximately 40% loss in FAD-GDH enzyme activity for test sensors stored with molecular sieve desiccant (FIG. 4). In contrast, the relatively constant and near-zero bias with increasing silica gel desiccant for 600 mg/dL glucose (FIG. 2B) correlated with only a 20-25% loss in FAD-GDH enzyme activity for test sensors stored with silica gel desiccant (FIG. 4).

[0053] FIG. 5 depicts graphs of in-sensor FAD-GDH enzyme activity ("% enzyme recovery") for test sensors sealed for two weeks either at 50°C, for containers having varying types of desiccant, and for reagent compositions with and without the enzyme stabilizer sorbitol. The desiccants used were silica gel (SG), molecular sieve 1 3x (MS- 1 3x), a bottle sleeve containing molecular sieve 4A (Bottle-MS), and two different polymer-blended desiccants - a polypropylene film coated with molecular sieves (SLF/MS), and a polypropylene film coated with silica gel (SLF/SG). The polymer- blended desiccants were obtained from Multisorb Technologies (Buffalo, NY).

[0054] The reagent compositions for test sensors labeled "PD18-control" and "PD1 6- control" were formed by deposition and drying of a reagent fluid that included water, 80 millimolar (mM) 3-(2',5'-disulfophenylimino)-3H-phenothiazine bis-sodium salt mediator, 3.75 enzyme units FAD-GDH per microliter, 0.2% (w/w) hydroxyethylene cellulose (HEC) binder having a weight average molecular weight (Mw) of 300,000, 0.362% (w/w) HEC binder having a Mw of 90,000, 1 12.5 mM Na₂HP0₄ buffer salt, 0.225% (w/w) N-octanoyl-N-methyl-D-glucamine (MEGA-8), and 0.01 % (w/w) sodium methyl cocoyl taurate (Geropon TC-42). The reagent composition for test sensors labeled "PD18 plus 0.4% sorbitol" was formed as for the sensors labeled "PD18- control," except that the reagent fluid also included 0.4% (w/w) sorbitol. [0055] Test sensors stored with pure molecular sieve desiccant (MS-1 3x) or with bottle desiccant sleeve (Bottle-MS) had an approximately 30% decrease in enzyme activity, while test sensors stored with silica gel desiccant (SG) had only a 15% decrease. Stabilization of the enzyme with 0.4% sorbitol diminished the loss of enzyme activity; however, the test sensors stored with molecular sieve desiccants again allowed for twice the amount of enzyme inactivation. The differences in enzyme recovery between PD18-control test sensors and PD1 6-control test sensors stored with pure molecular sieve desiccant or with silica gel desiccant are believed to be within experimental error.

[0056] Blending of molecular sieve desiccant with polypropylene (SLF/MS) provided for retention of enzyme activity comparable with that provided by the silica gel desiccant. Thus, inhibiting the desiccating ability of the molecular sieves allowed the enzyme to retain its activity during the heat stress. The desiccating ability of the silica gel also was inhibited. Decreases in assay accuracy may be related to a lack of protection of the other reagent composition ingredients from moisture during the heat stress.

[0057] For a biosensor system that includes a plurality of test sensors sealed in a container having a desiccant, the system may be evaluated by measuring the activity of the redox enzyme in the reagent composition of the test sensors that is retained after the test sensors are stored in various conditions. In one example, a plurality of test sensors is sealed in a container including a desiccant for two weeks at a temperature of 50°C, where each test sensor includes at least two conductors, one of which is a working electrode, and a reagent composition including a redox enzyme disposed on or near the working electrode. The test sensors are then removed from the container, and the activity of the redox enzyme in the reagent composition of each test sensor is measured. In this example, the reagent composition of each test sensor preferably retains at least 75% of the activity of the redox enzyme. More preferably, in this example the reagent composition of each test sensor preferably retains at least 80% of the activity of the redox enzyme, and more preferably retains at least 85% of the activity of the redox enzyme. Preferably, in this example the number of test sensors in the plurality is at least 10, and preferably is at least 25, at least 50, or at least 100.

[0058] The correlation of one or more output current values, such as the output current values depicted in FIGs. 1 A - 1 C, with the analyte concentration of the sample may be adjusted to account for errors in the measurement. One approach to correct errors associated with a biosensor analysis is to adjust the correlation for determining analyte concentrations in a sample from output current values with index functions extracted from intermediate current values of the output current values. Index functions can compensate the correlation for determining analyte concentrations from the output current values for one or more errors in the analyses that could result in bias of the determined analyte concentrations. Index functions correspond to the %-bias in the correlation between the analyte concentrations and the output current values due to one or more errors in the analysis.

[0059] The glucose assay %-bias may be represented by one or more AS values obtained from one or more error parameters. The AS values represent slope deviations of the correlation between analyte concentrations and output current values determined from one or more error parameters. The slope of the correlation corresponds to the change in output current for a given change in sample glucose concentration. Index functions corresponding to the slope or change in slope may be normalized to reduce the statistical effect of changes in the output current values, improve the differentiation in variations of the output current values, standardize the measurements of the output current values, a combination thereof, or the like. The adjusted correlation may be used to determine analyte concentrations in biological samples from the output current values and may have improved accuracy and/or precision in comparison to

conventional biosensors. Error correction using index functions and AS values is described, for example, in U.S. Patent Pub. 2009/01 77406, and in International Patent Application No. PCT/US2009/0671 50, filed December 8, 2009, entitled "Complex Index Functions". The disclosures of these patent applications regarding error correction using index functions and AS values are herein incorporated by reference. [0060] Thus, an output current value responsive to sample glucose concentration may be converted into a corrected glucose concentration of the sample using an index function representing AS/S. Alternatively, a corrected glucose concentration value may be determined from an uncorrected glucose concentration value using an index function and an equation such as Gcorr = G_raw/(1 + f(lndex)), where Gcorr is the corrected glucose concentration of the sample, G_raw is the determined analyte concentration of the sample without compensation, and f(lndex) is an index function.

[0061] Index functions may include ratios extracted from an output signal, such as the output signals depicted in FIGs. 1 A - 1 C. For example, the output signal values may be compared within an individual pulse-signal decay cycle, such as ratio R3 = 3,3 / 3,i , where 3,3 denotes the third current value recorded for the third signal decay, and 3,i denotes the first current value recorded for the third signal decay. In another example, the output signal values may be compared between separate pulse-signal decay cycles, such as ratio R4/3 = 4,3 / 3,3 , where 4,3 denotes the third current value recorded for the fourth signal decay. Index functions may include combinations of ratios extracted from the output signal. In one example, an index function may include a simple ratio of ratios, such as Ratio3/2 = R3/R2. In another example, an index function may include a more complicated combination of simpler index functions. For example, an index function lndex-1 may be represented as lndex-1 = R4/3 - Ratio3/2. In another example, an index function lndex-2 may be represented at lndex-2 = (R4/3)^p - (Ratio3/2)^q , where p and q independently are positive numbers.

[0062] Preferably an index function corrects errors associated with variations in hematocrit content. For example, conventional biosensor systems may be configured to report glucose concentrations presuming a 40% (v/v) hematocrit content for a whole blood sample, regardless of the actual hematocrit content of the sample. In these systems, any glucose measurement performed on a blood sample containing less or more than 40% hematocrit will include error and thus have bias attributable to the hematocrit effect. [0063] Calculation of an index function that corrects errors associated with variations in hematocrit content can be facilitated by using a test sensor that produces an output signal that varies with hematocrit content. For some biosensors, the R5/4 ratio parameter has served as an indicator of hematocrit in a sample, and has been used to adjust the measured analyte concentration to account for the hematocrit content of the sample. The R5/4 ratio parameter represents the relationship between the currents generated by the analyte in response to the 4^th and 5^th pulses of a gated amperometry pulse sequence of FIGs. 1 A - 1 C.

[0064] FIG. 6 depicts graphs of the variation of the R5/4 ratio parameter for test sensors stored for two weeks at 50°C, relative to the R5/4 ratio parameter for test sensors stored for two weeks at -20°C, where the test sensors had varying levels of enzyme density above the working electrode of the test sensors. The two types of data points represent the two different anionic surfactants Phospholan CS1 31 (nonylphenol ethoxylate phosphate) and Geropon TC-42.

[0065] At higher concentrations of enzyme, the difference between the R5/4 ratio parameters for test sensors stored at 50°C and for test sensors stored at -20°C was smal ler. This trend was evident for both types of anionic surfactants used in the reagent compositions. Since the R5/4 ratio parameter can be used as a variable in an index function for correcting analyte measurements, a lower variation of the parameter due to environmental factors is desirable. Thus, the increased retention of enzyme activity provided by the less aggressive desiccants can provide an added benefit of decreasing the variabil ity of correction factors.

[0066] The enzyme in the reagent compositions of the test sensors used in FIGs. 1 through 6 was the FAD-GDH enzyme. The observed effects of residual moisture during storage of test sensors are bel ieved to apply to other enzymes, such as alcohol dehydrogenase, lactate dehydrogenase, β-hydroxybutyrate dehydrogenase, glucose-6- phosphate dehydrogenase, glucose oxidase (GOx), glucose dehydrogenase,

formaldehyde dehydrogenase, malate dehydrogenase, and 3-hydroxysteroid

dehydrogenase. [0067] Preferable enzyme systems are oxygen independent, thus not substantially oxidized by oxygen. One such oxygen independent enzyme family is glucose dehydrogenase (GDH). Using different co-enzymes or co-factors, GDH may be mediated in a different manner by different mediators. Depending on their association with GDH, a co-factor, such as flavin adenine dinucleotide (FAD), can be tightly held by the host enzyme, such as in the case of FAD-GDH; or a co-factor, such as

Pyrroloquinolinequinone (PQQ), may be covalently linked to the host enzyme, such as with PQQ-GDH. The co-factor in each of these enzyme systems may either be permanently held by the host enzyme or the co-enzyme and the apo-enzyme may be reconstituted before the enzyme system is added to the reagent fluid. The co-enzyme also may be independently added to the host enzyme moiety in the reagent fluid to assist in the catalytic function of the host enzyme, such as in the cases of nicotinamide adenine dinucleotide NAD/NADH ⁺ or nicotinamide adenine dinucleotide phosphate NADP/NADPH⁺ in combination with NAD-dependent glucose dehydrogenase (NAD- GDH).

[0068] Ingredients of reagent compositions for test sensors, and of reagent fluids for forming the reagent compositions, are described, for example, in U.S. Patent Pub.

2009/01 78936, and in International Patent Application No. PCT/US2009/066963, filed December 7, 2009, entitled "Low Total Salt Reagent Compositions And Systems For Biosensors". The disclosures of these patent applications regarding reagent composition ingredients and fluids for forming reagent compositions are herein incorporated by reference.

[0069] Both the enzyme activity in test sensors and the assay performance of the test sensors appear to be affected by the type of desiccant used in the container for the sensors. A desiccant that adsorbs at most 15% of its weight in water, or that preferably adsorbs at most 10% or from 5% - 10% of its weight in water, when in contact with an environment of 10% - 20% RH at 40°C may provide a residual moisture level in the regent composition that allows the enzyme to be retained in its active state. In contrast, excessive drying of the reagent composition by an aggressive desiccant, such as molecular sieve, may lead to enzyme inactivation. The less aggressive desiccants may balance the opposite moisture requirements for the mediator and the enzyme in containers for test sensors by adsorbing water from the atmosphere only when humidity level in the package exceeds 20% RH. Thus, the less aggressive desiccants may protect the mediator from high moisture without adversely affecting the enzyme activity.

[0070] FIG. 7 depicts a schematic representation of a biosensor 700 that determines an analyte concentration in a sample of a biological fluid using a test sensor. The biosensor 700 includes a measurement device 702 and a test sensor 704, which may be implemented in any analytical instrument, including a bench-top device, a portable or hand-held device, or the like. The biosensor 700 may be utilized to determine analyte concentrations, including those of glucose, uric acid, lactate, cholesterol, bilirubin, and the like. While a particular configuration is shown, the biosensor 700 may have other configurations, including those with additional components.

[0071] The test sensor 704 has a base 706 forming a reservoir 708 and a channel 710 with an opening 712. The reservoir 708 and the channel 710 may be covered by a lid with a vent. The reservoir 708 defines a partially-enclosed volume. The reservoir 708 may contain a composition that assists in retaining a liquid sample such as water- swellable polymers or porous polymer matrices. Reagents may be deposited in the reservoir 708 and/or channel 710. The reagent composition at the working electrode 707 includes a low total salt reagent composition and may include one or more enzyme system, mediator, and like species. The counter electrode 705 may be formed using the same or a different reagent composition, preferably one lacking an enzyme system. The test sensor 704 also may have a sample interface 714 disposed adjacent to the reservoir 708. The sample interface 714 may partially or completely surround the reservoir 708. The test sensor 704 may have other configurations.

[0072] The sample interface 714 has conductors 709 connected to the working electrode 707 and the counter electrode 705. The electrodes may be substantially in the same plane or in more than one plane. The electrodes 704, 705 may be disposed on a surface of the base 706 that forms the reservoir 708. The electrodes 704, 705 may extend or project into the reservoir 708. A dielectric layer may partially cover the conductors 709 and/or the electrodes 704, 705. The sample interface 714 may have other electrodes and conductors.

[0073] The measurement device 702 includes electrical circuitry 716 connected to a sensor interface 718 and a display 720. The electrical circuitry 716 includes a processor 722 connected to a signal generator 724, an optional temperature sensor 726, and a storage medium 728.

[0074] The signal generator 724 provides an electrical input signal to the sensor interface 718 in response to the processor 722. The electrical input signal may be transmitted by the sensor interface 718 to the sample interface 714 to apply the electrical input signal to the sample of the biological fluid. The electrical input signal may be a potential or current and may be applied in multiple pulses, sequences, or cycles. The signal generator 724 also may record an output signal from the sensor interface as a generator-recorder.

[0075] The optional temperature sensor 726 determines the temperature of the sample in the reservoir of the test sensor 704. The temperature of the sample may be measured, calculated from the output signal, or assumed to be the same or similar to a measurement of the ambient temperature or the temperature of a device implementing the biosensor system. The temperature may be measured using a thermister,

thermometer, infrared sensor, thermopile or other temperature sensing device. Other techniques may be used to determine the sample temperature.

[0076] The storage medium 728 may be a magnetic, optical, or semiconductor memory, another storage device, or the like. The storage medium 728 may be a fixed memory device, a removable memory device, such as a memory card, remotely accessed, or the like.

[0077] The processor 722 implements the analyte analysis and data treatment using computer readable software code and data stored in the storage medium 728. The processor 722 may start the analyte analysis in response to the presence of the test sensor 704 at the sensor interface 718, the application of a sample to the test sensor 704, in response to user input, or the like. The processor 722 directs the signal generator 724 to provide the electrical input signal to the sensor interface 718. The processor 722 may receive the sample temperature from the optional temperature sensor 726. The processor 722 receives the output signal from the sensor interface 718. The output signal is generated in response to the redox reaction of the analyte in the reservoir 708.

[0078] The processor 722 preferably measures the output signal to obtain a current value from an excitation where the initial current value is greater than those that follow in the decay and within less than about 3 seconds of introducing the sample to the test sensor 704. More preferably, the processor 722 measures the output signal to obtain a current value within less than about 3 seconds of introducing the sample to the test sensor in 704 and obtains the first current value recorded from an excitation where the current values that follow the first current value continuously decrease. Even more preferably, the processor 722 measures the output signal to obtain a current value within less than about 3 seconds of introducing the sample to the test sensor in 704, to obtain the first current value recorded from an excitation where the current values that follow the first current value continuously decrease, and to obtain a current value during the maximum kinetic performance of the test sensor.

[0079] The one or more obtained current value is correlated with the analyte concentration of the sample using one or more correlation equations in the processor 722. The results of the analyte analysis may be output to the display 720 and may be stored in the storage medium 728. Preferably, the results of the analyte analysis are output to the display 720 within five seconds or less of introducing the sample to the test sensor, more preferably the results are output to the display 720 within three seconds or less of introducing the sample to the test sensor.

[0080] The correlation equations relating analyte concentrations and output current values may be represented graphically, mathematically, a combination thereof, or the like. The correlation equations may be represented by a program number (PNA) table, another look-up table, or the like that is stored in the storage medium 728. Instructions regarding implementation of the analyte analysis may be provided by the computer readable software code stored in the storage medium 728. The code may be object code or any other code describing or controlling the functionality described herein. The data from the analyte analysis may be subjected to one or more data treatments, including the determination of decay rates, K constants, ratios, and the like in the processor 722.

[0081] The sensor interface 718 has contacts that connect or electrically

communicate with the conductors 709 in the sample interface 714 of the test sensor 704. The sensor interface 718 transmits the electrical input signal from the signal generator 724 through the contacts to the conductors 709 in the sample interface 714. The sensor interface 718 also transmits the output signal from the sample through the contacts to the processor 722 and/or signal generator 724.

[0082] The display 720 may be analog or digital. The display may be a LCD, LED, OLED, TFT or other display adapted to display a numerical reading.

[0083] In use, a sample for analysis is transferred into the reservoir 708 by

introducing the sample to the opening 712. The sample flows through the channel 710, filling the reservoir 708 while expelling the previously contained air. The sample chemically reacts with the reagents deposited in the channel 710 and/or reservoir 708. Preferably, the sample is a fluid, more preferably, a liquid.

[0084] The test sensor 704 is disposed adjacent to the measurement device 702. Adjacent includes positions where the sample interface 714 is in electrical

communication with the sensor interface 718. Electrical communication includes wired or wireless transfer of input and/or output signals between contacts in the sensor interface 718 and conductors 709 in the sample interface 714.

[0085] FIG. 8 depicts a biosensor system 800 that includes a container 810 including a desiccant and a plurality of test sensors 830. The container 810 includes a closure 812 that can seal the test sensors 830 in the container 810. The container 810 may include desiccant 820 in a separate package in the container. The container 810 may include desiccant 822 in the closure 812. The container 810 may include desiccant 824 in a wall of the container. The container 810 may include desiccant 826 in the base of the container. The container 810 may be made of a variety of materials, including plastic, metal foil and/or glass. The amount and type of desiccant in the container 810 may be selected to provide a predetermined moisture level in the container.

[0086] While various embodiments of the invention have been described, it will be apparent to those of ordinary skill in the art that other embodiments and

implementations are possible within the scope of the invention. Accordingly, the invention is not to be restricted except in light of the attached claims and their equivalents.

Claims

WHAT IS CLAIMED IS:

1 . A biosensor system, for determining the concentration of an analyte in a sample, comprising:

a plurality of test sensors, each test sensor comprising

at least two conductors, where one of the conductors is a working electrode, and

a reagent composition disposed on or near the working electrode; and a container comprising a desiccant;

where, when the plurality of test sensors is sealed in the container for two weeks at a temperature of 50°C and then removed from the container, and each test sensor is subsequently connected through the at least two conductors to a measurement device and then contacted with one of a plurality of samples including an analyte, and the analyte concentration in each sample is measured by the test sensor and the measuring device, the plurality of samples having an analyte concentration that spans the range of 10 mg/dL - 600 mg/dL,

the bias of each measured analyte concentration is within ±10 mg/dL for samples having an analyte concentration less than 100 mg/dL and within ±10% for samples having an analyte concentration of at least 100 mg/dL.

2. The biosensor system of claim 1 , where the desiccant adsorbs at most 1 5% of its weight in water when in contact with an environment of 10% - 20% RH at 40°C.

3. The biosensor system of claim 2, where the desiccant comprises polymer- blended molecular sieves.

4. The biosensor system of claim 2, where the desiccant comprises a blend of molecular sieves and silica gel.

5. The biosensor system of claim 1 , where the desiccant adsorbs at most 1 0% of its weight in water when in contact with an environment of 1 0% - 20% RH at 40°C.

6. The biosensor system of claim 1 , where the desiccant absorbs from 5% - 10% of its weight in water when in contact with an environment of 1 0% - 20% RH at 40°C.

7. The biosensor system of claim 5 or 6, where the desiccant comprises sil ica gel.

8. The biosensor system of claim 7, where the container comprises at most 30 mg of the sil ica gel per test sensor.

9. The biosensor system of claim 7, where the container comprises at most 10 mg of the sil ica gel per test sensor.

1 0. The biosensor system of any one of claims 1 -9, where the plural ity of test sensors comprises at least 1 0 test sensors.

1 1 . The biosensor system of any one of claims 1 -9, where the plural ity of test sensors comprises at least 25 test sensors.

1 2. The biosensor system of any one of claims 1 -9, where the plural ity of test sensors comprises at least 50 test sensors.

1 3. The biosensor system of any one of claims 1 -9, where the plural ity of test sensors comprises at least 1 00 test sensors.

14. The biosensor system of any one of claims 1 -1 3, where the bias of each measured analyte concentration is within ±7 mg/dL for samples having an analyte concentration less than 100 mg/dL and within ±7% for samples having an analyte concentration of at least 100 mg/dL.

1 5. The biosensor system of any one of claims 1 -1 3, where the bias of each measured analyte concentration is within ±5 mg/dL for samples having an analyte concentration less than 100 mg/dL and within ±5% for samples having an analyte concentration of at least 100 mg/dL.

1 6. The biosensor system of any one of claims 1 -1 3, where the plural ity of samples has an analyte concentration that spans the range of 50 mg/dL - 600 mg/dL.

1 7. A biosensor system, for determining the concentration of an analyte in a sample, comprising:

a plural ity of test sensors, each test sensor comprising

a reagent composition disposed on or near the working electrode, the reagent composition comprising a redox enzyme; and

a container comprising a desiccant;

where, when the plural ity of test sensors is sealed in the container for two weeks at a temperature of 50°C and then removed from the container, the reagent composition of each test sensor retains at least 75% of the activity of the redox enzyme.

1 8. The biosensor system of claim 1 7, where the desiccant adsorbs at most 1 5% of its weight in water when in contact with an environment of 1 0% - 20% RH at 40°C.

1 9. The biosensor system of claim 1 8, where the desiccant comprises polymer- blended molecular sieves.

20. The biosensor system of claim 1 8, where the desiccant comprises a blend of molecular sieves and sil ica gel.

21 . The biosensor system of claim 1 7, where the desiccant adsorbs at most 10% of its weight in water when in contact with an environment of 1 0% - 20% RH at 40°C.

22. The biosensor system of claim 1 7, where the desiccant absorbs from 5% - 10% of its weight in water when in contact with an environment of 1 0% - 20% RH at 40°C.

23. The biosensor system of claim 21 or 22, where the desiccant comprises sil ica gel.

24. The biosensor system of claim 23, where the container comprises at most 30 mg of the sil ica gel per test sensor.

25. The biosensor system of claim 23, where the container comprises at most 1 0 mg of the sil ica gel per test sensor.

26. The biosensor system of any one of claims 1 7-25, where the plural ity of test sensors comprises at least 10 test sensors.

27. The biosensor system of any one of claims 1 7-25, where the plural ity of test sensors comprises at least 25 test sensors.

28. The biosensor system of any one of claims 1 7-25, where the plural ity of test sensors comprises at least 50 test sensors.

29. The biosensor system of any one of claims 1 7-25, where the plural ity of test sensors comprises at least 100 test sensors.

30. The biosensor system of any one of claims 1 7-29, where the reagent composition of each test sensor retains at least 80% of the activity of the redox enzyme.

31 . The biosensor system of any one of claims 1 7-29, where the reagent composition of each test sensor retains at least 85% of the activity of the redox enzyme.