Calibration of analyte amount in mass spectrometry

Abstract

A method of determining the concentration or quantity of an analyte of interest is disclosed. The method comprises obtaining a calibration relationship that relates the concentration or quantity of th

Description

CROSS-REFERENCE TO RELATED APPLICATION This application claims priority from and the benefit of United Kingdom patent application No. 2408112.7 filed on 7 June 2024. The entire contents of this application are incorporated herein by reference. FIELD OF THE INVENTION The present invention relates generally to mass spectrometry, and in particular to techniques for quantifying the amount of an analyte that is present in a sample. BACKGROUND It is known to mass analyse samples using mass spectrometry in order to determine the quantity or concentration of an analyte in the sample. In order to determine the quantity or concentration of the analyte in the sample it is necessary to use a calibration relationship that relates the ion signal for the analyte that is detected by the mass spectrometer to the quantity or concentration of the analyte in the sample. In order to obtain the calibration relationship it is typical to mass analyse various calibration samples. Although this approach is effective, the calibration samples take additional time to mass analyse. Also, when samples are being analysed in batches of a restricted size, the calibration samples take up space in the batch that cannot then be used to analyse analytical samples. This is particularly problematic for batches that are restricted to having relatively few samples therein. SUMMARY From a first aspect the present invention provides a method of determining the concentration or quantity of an analyte of interest in an analytical sample comprising: obtaining a first calibration relationship that relates the concentration or quantity of an analyte of interest to an ion signal detected by a mass spectrometer for the analyte of interest; using a mass spectrometer to mass analyse a first set of samples that includes analytical samples containing the analyte of interest, wherein a first plurality of calibration checkers having different known concentration levels are provided in one or more of the samples in said first set of samples, wherein said mass analysing obtains an ion signal for an analyte of interest in each of the analytical samples and an ion signal for each of the known concentration levels of said calibration checkers; and performing a comparison in which values that are representative of the ion signals and known concentration levels of at least some of the calibration checkers are compared to the first calibration relationship to determine if the first calibration relationship is still valid. As the present invention performs said comparison, e.g. at various intervals during the analysis of samples, it is able to check whether the calibration relationship remains valid prior to performing a recalibration. This avoids unnecessary recalibration that would otherwise be performed, which could result in errors being introduced into the analytical results. Also, checking if the calibration relationship is still valid, rather than immediately determining a new calibration relationship, requires fewer calibration checkers than the number of calibrators that would be desired for a recalibration. The plurality of calibration checkers having different known concentration levels may be added to a single one of the samples. Alternatively, different ones of the plurality of calibration checkers could be added to different respective ones of the samples. The method may obtain the first calibration relationship by mass analysing a plurality of calibrators having different known concentration levels in one or more samples, so as to obtain an ion signal each of the known concentration levels of said calibrators; and using the ion signals and known concentration levels of the calibrators to generate the first calibration relationship. For example, the plurality of calibrators at different known concentration levels could be added to a single sample, which is then mass analysed so as to determine the first calibration relationship. Alternatively, different ones of the plurality of calibrators could be added to different respective samples, which are then mass analysed so as to determine the first calibration relationship. The sample(s) containing the calibrators may be mass analysed prior to said first set of samples. The calibration checkers may be provided in an analytical sample and / or in a quality control sample. For example, the plurality of calibration checkers may be provided in a single analytical sample or in a single quality control sample. Alternatively, the plurality of calibration checkers may be distributed amongst multiple analytical samples, or distributed amongst multiple quality control samples. Alternatively, some of the calibration checkers may be provided in one or more analytical sample and others calibration checkers may be provided in one or of the quality control samples. Each analytical sample may contain an unknown quantity or concentration of the analyte of interest. In contrast, any quality control samples that are present in the first set of samples include a concentration of the analyte of interest that is known prior to it being mass analysed. The samples in the first set of samples may be mass analysed by the same mass spectrometer. The first set of samples is preferably formed from consecutively mass analysed samples. The analytical samples may be mass analysed consecutively without mass analysing any other types of sample in between them. Alternatively, one or more other type of sample, such as a quality control sample, may be mass analysed between the consecutively mass analysed analytical samples. Optionally, only one sample is said first set of samples includes all of said plurality of calibration checkers. The remaining samples preferably do not contain at least one of said calibration checkers. For example, the remaining samples may have none of said calibration checkers therein. Alternatively, the remaining samples may have only one of said calibration checkers therein, which is at the same concentration in all of said remaining samples. This calibration checker in the remaining samples may be used as an internal standard. For example, when each of the remaining samples is mass analysed, a scaling factor may be obtained based on the ion signal (e.g. peak area or height) for the calibration checker in that sample and the ion signal (e.g. peak area or height) for the analyte of interest in that sample, and the ion signal (e.g. peak area or height) for the analyte of interest may then be multiplied by the scaling factor. As all of the plurality of calibration checkers are only provided in one, or some, of the samples, only a relatively small amount of calibration checker material is needed to perform the method. Embodiments are contemplated in which all of the different known concentrations levels are provided in all of the analytical samples and / or quality control samples, but this is less preferable as it requires a relatively large amount of calibrator material. It also requires each concentration level to be analysed in each of these samples, thus requiring a relatively large amount of analysis per sample. The method may comprise determining the concentration or quantity of the analyte of interest in at least one of the analytical samples in the first set of samples, using the ion signal detected for the analyte of interest in that analytical sample and the first calibration relationship. Said comparison may comprise: determining that the first calibration relationship is still valid if said values fit the first calibration relationship to within a preselected tolerance; and determining that the first calibration relationship is no longer valid if said values do not fit the first calibration relationship to within a preselected tolerance. Said comparison may initially use the ion signals and known concentration levels for only a sub-set of said calibration checkers in determining if their values fit the first calibration relationship to within the preselected tolerance, wherein if they do then the method determines that the first calibration relationship is still valid. If the values for the sub-set of calibration checkers do not fit the first calibration relationship to within a preselected tolerance then the method may use the ion signals and known concentration levels for all of the calibration checkers in determining if their values fit the first calibration relationship to within the preselected tolerance, and if they do then the method may determine that the first calibration relationship is still valid, whereas if they do not then the method determines that the first calibration relationship is no longer valid. When said comparison determines that the first calibration relationship is no longer valid, the method may use the ion signals and known concentration levels of the calibration checkers to generate a second calibration relationship. The method may comprise using the ion signals and known concentration levels for only a sub-set of the calibration checkers in said comparison step, and using the ion signals and known concentration levels for all of the calibration checkers to generate the second calibration relationship. Alternatively, the ion signals and known concentration levels of all of the calibration checkers may be used in both the comparison step and to generate the second calibration relationship. The method may comprise mass analysing a second set of samples that includes analytical samples containing the analyte of interest after said one or more sample containing the calibration checkers has been mass analysed; and determining the concentration or quantity of the analyte of interest in at least one of the analytical samples in the second set of samples, using the ion signal detected for the analyte of interest in that analytical sample and the second calibration relationship. The method may comprise determining the concentration or quantity of the analyte of interest in at least one of the analytical samples in the first set of samples, using the ion signal detected for the analyte of interest in that analytical sample and the second calibration relationship. The method may comprise: controlling apparatus to generate an output that indicates a determined concentration or quantity of the analyte of interest in an analytical sample; using the first calibration relationship to determine the concentration or quantity of the analyte of interest in an analytical sample in the first set of samples; determining from said comparison step that the first calibration relationship is no longer valid, and in response to this either: i) controlling the apparatus so as not to generate an output that indicates the determined concentration or quantity of the analyte of interest in the analytical sample in the first set of samples; or ii) generating an output that indicates the determined concentration or quantity of the analyte of interest in the analytical sample in the first set of samples along with an indication that the determined concentration or quantity is not valid. For example, according to option i) the method may control the apparatus so that a printer or display does not print or display the concentration or quantity of the analyte of interest for the analytical sample. Alternatively, according to option ii) the method may control the apparatus so that a printer or display print or displays the concentration or quantity of the analyte of interest for the analytical sample along with the indication that the determined concentration or quantity is not valid. In response to said comparison determining that the first calibration relationship is no longer valid, the method may add a plurality of calibration checkers at different known concentration levels to one or more further samples, mass analyse those one or more further samples, and determine a new calibration relationship from the resulting ion signals. For example, this plurality of calibration checkers could be added to a single further sample, which is then mass analysed so as to determine the new calibration relationship. Alternatively, different ones of the plurality of calibration checkers could be added to different respective further samples, which are then mass analysed so as to determine the new calibration relationship. The method may use automated machinery to perform the steps herein, e.g. to automatically determine that the first calibration relationship is no longer valid and to add the plurality of calibration checkers to the one or more further samples. The method may comprise mass analysing a further set of samples that includes analytical samples containing the analyte of interest after said one or more samples containing the plurality of calibration checkers have been mass analysed; and determining the concentration or quantity of the analyte of interest in at least one of the analytical samples in the further set of samples, using the new calibration relationship. When said comparison determines that the first calibration relationship is still valid, the method may comprise determining the concentration or quantity of the analyte of interest in at least one analytical sample that is mass analysed, using the ion signal detected for the analyte of interest in that analytical sample and the first calibration relationship. The method may comprise mass analysing a second set of samples that includes analytical samples containing the analyte of interest after said sample containing the first plurality of calibration checkers has been mass analysed; and determining the concentration or quantity of the analyte of interest in at least one of the analytical samples in the second set of samples, using the ion signal detected for the analyte of interest in that analytical sample and the first calibration relationship. The method may comprise determining the concentration or quantity of the analyte of interest in at least one of the analytical samples in the first set of samples, using the ion signal detected for the analyte of interest in that analytical sample and the first calibration relationship; and wherein when the first calibration relationship is determined to still be valid and is, or has been, used to determine the concentration or quantity of the analyte of interest in an analytical sample, the method controls apparatus to generate an output that indicates the determined concentration or quantity of the analyte of interest. For example, the method may control a printer or display to print or display the concentration or quantity of the analyte of interest for the analytical sample. After the any of the sets of samples described above has been mass analysed, the method may check whether the calibration relationship that is in use is still valid or not. Accordingly, the method may further comprise: using the mass spectrometer to mass analyse one or more further sample containing a third plurality of calibration checkers at different known concentration levels, after mass analysing said second set samples; wherein said mass analysing obtains an ion signal for each of the known concentration levels of said third plurality of calibration checkers; and wherein a comparison is performed in which values that are representative of the ion signals and known concentration levels of at least some of the third plurality of calibration checkers are compared to the first, second or new calibration relationship to determine if that calibration relationship is still valid. The method may comprise determining that the first calibration relationship is still valid based on said values fitting the first calibration relationship to within a preselected tolerance; determining a difference between said values and the first calibration relationship, and predicting when the first calibration relationship will turn from being valid to no longer being valid based on said difference. The method may comprise: a) repeatedly performing a cycle, wherein each cycle comprises: i) mass analysing one or more samples that contain a plurality of calibration checkers at different known concentration levels so as to obtain an ion signal for each of the known concentration levels of said calibration checkers, ii) performing a comparison in which values that are representative of the ion signals and known concentration levels of at least some of these calibration checkers are compared to the first calibration relationship, and iii) determining that the first calibration relationship is still valid based on said values fitting the first calibration relationship to within a preselected tolerance, and determining a difference between said values and the first calibration relationship; and b) determining a trend of how the difference determined in step iii) changes between the cycles; and c) predicting when the first calibration relationship will turn from being valid to no longer being valid based on said trend. The method may predict the time, or maximum number of samples that can be mass analysed, until the first calibration relationship will turn from being valid to no longer being valid. The method may comprise, in response to said predicting step, calculating a new calibration relationship at a point before the first calibration relationship has been predicted to turn from being valid to no longer being valid. The new calibration relationship may be determined using the ion signals and known concentrations from one of the sets of calibration checkers. The new calibration relationship may then be used to determine the concentration or quantity of the analyte of interest in analytical samples before the first calibration relationship has been predicted to turn from being valid to no longer being valid. In each set of samples described above, the final sample in the set may be the sample that includes the calibration checkers. Each set of samples described herein may contain >3, >4, >5, >6, >8, >10, >20, >30, >40, >50, >60, >70, >80, >90, or >100 samples. However, it may be preferred that the set of samples contains relatively few samples. Accordingly, each set of samples may be <70, <60, <50, <40, <30, <20, <10, or <5 samples. The method may comprise estimating a range of concentrations that the analyte of interest is expected to have in the analytical samples, wherein the different known concentration levels of the calibration checkers include a first concentration level above said range and / or a second concentration level below said range. Alternatively, or additionally, the different known concentration levels of the calibration checkers may include at least one concentration level within said range. Each of the calibration checkers may be a compound that produces an ion signal when mass analysed that varies according to its concentration in substantially the same way that the ion signal for the analyte of interest varies according to its concentration. Each of said calibration checkers may be a compound that is the same as the analyte of interest, except that one of more of the atoms in its chemical structure has been substituted for a different, stable isotope of that atom. Such calibration checker compounds are referred to herein as stable isotope labelled calibration checker (SILCC) compounds. However, other compounds may be used as the calibration checkers instead, such as a compound that is different to the analyte of interest but is in the same family of compounds as the analyte of interest. The calibration checkers in each plurality of calibration checkers may have different molecular masses. The mass spectrometer may be operated to always mass analyse the calibration checkers if they are present, or it may be operated to not mass analyse the calibration checkers in some of the samples. For example, the mass spectrometer may be programmed not to mass analyse the calibration checkers for samples that the calibration checkers have not been added to. The number of samples or analytical samples in each set of samples may be the same as the number of calibration checkers having different known concentration levels. The method may comprise providing a plurality of batches of samples and performing any of the methods described herein on each of the batches. Each batch may have a predetermined maximum number of samples therein. For example, the batch may have a number of samples that is <100, <90, <80, <70, <60, <50, <40, <30, <20, or< 10. Each batch of samples may be provided in a sample plate having a predetermined number of sample wells therein. However, it is contemplated that the samples may not be mass analysed in batches and may instead be mass analysed as a stream of samples arriving at the mass analyser. The samples may be mass analysed using liquid chromatography mass spectrometry. Optionally, each of the analytical samples is mass analysed using tandem mass spectrometry by monitoring an MRM transition for said analyte of interest, and at least the samples containing the calibration checkers are mass analysed using tandem mass spectrometry by monitoring an MRM transition for each of the calibration checkers. The analytical sample may be a biological sample, such as from a patient. For example, the sample may be, or comprise, blood, blood plasma, blood serum, urine, stool, saliva or cerebrospinal fluid. The analyte of interest may be a steroid hormone, a vitamin, a drug of abuse (e.g. an opiate), a therapeutic drug such as an immunosuppressant, an anti-epileptic drug, an anti-infective drug, a steroid, an anti-fungal drug, or an amino acids etc. The number of calibration checkers having different known concentration levels may be between four and seven calibration checkers. However, it is contemplated that fewer or a greater number of calibration checkers having different known concentration levels may be used. For example, at least two, at least three, or at least seven, calibrators having different known concentration levels may be used. Although the method has been described as determining the concentration or quantity of a single analyte of interest, the method described herein may be used to determine the concentration or quantity of multiple different analytes of interest in each analytical sample. Calibrators and calibration checkers may be provided for each of these analytes of interest. The method may be performed by automated machinery that: (i) receives samples; (ii) adds said plurality of different known concentration levels of the calibration checkers to one or more of the samples in said set of samples; and (iii) transfers the set of samples to one or more mass spectrometer so as to then perform any of the methods described above. Each of the analytical samples that is received at the automated machinery may have an identifier thereon that represents the type of sample that it is and / or a test that is required to be performed on that sample. The automated machinery may read the identifier and select analytical samples of the same type and / or that are to undergo the same test and includes these analytical samples in said set of samples. The automated machinery may group the selected analytical samples into one or more batches of analytical samples. The identifier may be a barcode and the automated machinery may be a barcode reader. However, it will be appreciated that may other types of identifier and automated reader may be used. Although embodiments have been described in which the samples are mass analysed, it is contemplated that the samples may be analysed using other techniques. Accordingly, from a second aspect the present invention provides a method of determining the concentration or quantity of an analyte of interest in an analytical sample comprising: obtaining a first calibration relationship that relates the concentration or quantity of an analyte of interest to a signal detected by an analyser for the analyte of interest; using an analyser to analyse a first set of samples that includes analytical samples containing the analyte of interest, wherein a first plurality of calibration checkers having different known concentration levels are provided in one or more of the samples in said first set of samples, wherein said analysing obtains a signal for an analyte of interest in each of the analytical samples and a signal for each of the known concentration levels of said calibration checkers; and performing a comparison in which values that are representative of the signals and known concentration levels of at least some of the calibration checkers are compared to the first calibration relationship to determine if the first calibration relationship is still valid. The second aspect of the present invention may have any of the features described in relation to the first aspect of the present invention, except that the mass analysis steps described need not be mass analysis steps and may instead be performed by other types of analysis. For example, in the second aspect of the invention each said analyser may be an ion mobility analyser that performs ion mobility analysis. In such ion mobility analysis each sample is ionised and the resulting ions, or ions derived therefrom, are separated according to their ion mobility. The separated ions are then detected at an ion detector such that ions having different ion mobilities generate ion signals at different times. Therefore, different ion signals may be obtained for the analyte of interest and for the calibration checkers. Accordingly, the first calibration relationship may relate the concentration or quantity of the analyte of interest to an ion signal detected by an ion mobility analyser for the analyte of interest. An ion mobility analyser maybe used to ion mobility analyse the first set of samples, wherein said ion mobility analysing obtains an ion signal for the analyte of interest in each of the analytical samples and an ion signal for each of the known concentration levels of the calibration checkers. Alternatively, the samples may be analysed by techniques other than ion mobility analysis, such as optical analysis (e.g. spectroscopy). The present invention also provides automated apparatus having control circuitry and being configured to perform any of the methods described herein. BRIEF DESCRIPTION OF THE DRAWINGS Various embodiments of the present invention will now be described, by way of example only, and with reference to the accompanying drawings in which: Fig. 1 shows a layout of a typical batch of samples on a 96-well sample plate; Fig. 2 illustrates the concentrations of calibrator and analyte in different samples using a conventional approach; Fig. 3 shows two 96-well sample plates according to an embodiment of the present invention; Fig. 4 illustrates the concentrations of calibration checkers and analyte of interest in different samples according to the embodiment of Fig. 3; Figs. 5A-5D show example ion-chromatograms for an analyte of interest and three different calibration checkers; Fig. 6 shows a calibration relationship that is calculated based on the peak areas measured from the ion chromatograms shown in Figs. 5B-5D; Fig. 7 shows two 96-well sample plates according to another embodiment of the present invention; Fig. 8 illustrates the concentrations of calibration checkers and analyte of interest in different samples according to the embodiment of Fig. 7; and Fig. 9 shows an embodiment of automated machinery used in performing the methods described herein. DETAILED DESCRIPTION Liquid chromatography mass spectrometry (LC-MS) is regularly used within a clinical laboratory setting to determine concentrations of various clinically relevant molecules in a biological patient sample, such as blood, blood plasma, blood serum, urine, stool, saliva or cerebrospinal fluid (CSF). For example, it may be desired to quantify the amount of one or more certain analytes in the sample, such as steroid hormones, vitamins, drugs of abuse (for example opiates), and therapeutic drugs such as immunosuppressants, antiepileptics, antipsychotics, antifungals, and anti-infective drugs. There is also a growing interest in peptide and protein analyses as potential indicators of disease or disorder. LC-MS is often undertaken in a batch-based workflow, i.e. multiple samples are processed simultaneously,, with each sample in the batch undergoing the same preparation and analysis processes. For example, a 96-well sample plate is typically loaded with multiple samples in multiple respective sample wells, and then these samples are analysed in the same way. For instance, the samples may be analysed so as to test for a panel (i.e. plurality) of analytes of interest or for a single analyte of interest. It is important to standardise the results of the analyses and to ensure matrix effects are calibrated. Matrix effects occur when one or more compound in the matrix results in the ion signal detected for an analyte of interest being suppressed or enhanced relative to what it would be if that one or more compound was not present. In order to standardise the results of the analyses and to ensure matrix effects are calibrated, good laboratory practice (for example, CLSI C62A or equivalent guidelines) suggests that the following types of samples should be included in the batch being analysed:- an analytical sample, a solvent blank sample, a matrix blank sample, a matrix double blank sample, quality control samples, a blinded or challenge quality control sample, and calibrator samples. These types of sample are described in more detail below. It will be appreciated that the above list of sample types is not exhaustive and there may also be more specialist sample types that are used. It should also be appreciated that, subject to the proper considerations and validation, not all of the above sample types need be used in all cases. An analytical sample is a sample, generally from a patient, which contains an unknown quantity of one or more analyte of interest that is desired to be measured. Typically, each analytical sample will be mixed with a known quantity of an internal standard. An internal standard is a compound that is added to an analytical sample at a known concentration. The internal standard and analyte of interest may then both be measured and the response of the analyte measurement adjusted based on the measurement response of the internal standard. This may be considered to be a single point temporary calibration adjustment of the analyte measurement. It is required that the internal standard is distinguishable from the analyte of interest by the mass spectrometer and also that it behaves, chemically and physically, in as similar manner as possible to the analyte of interest during storage, measurement and analysis. As such, the internal standard is often a compound that is the same as the analyte of interest, except that one or more of its atoms has been substituted for a different, stable isotope of that atom. Such an internal standard is frequently known as a stable isotope labelled internal standard (SILIS). In other words, the internal standard is often an analogue of the analyte of interest, labelled with a stable isotope to produce a compound that is substantially chemically identical but which has a different molecular mass. As such, any change in the measurement response of the internal standard due to the processing and analysing conditions would also be expected to occur for the analyte of interest. The internal standard may also be used as a form of risk control for the analysis, e.g. to detect where a sample might have been omitted during a manual sample preparation, as no internal standard would be detected. A solvent blank is a sample of the solvent that is used to carry the analytical sample. The analysis of the solvent blank sample is useful for determining if there is contamination in the analytical system. Typically, the solvent blank sample is analysed first when analysing the batch of samples. A matrix blank sample is a sample that is representative of the matrix present in the analytical sample (e.g. whole blood, blood plasma, blood serum, urine etc.) and which has the requisite internal standard added to it. In each batch that is analysed there may be multiple matrix blank samples. The analysis of a matrix blank sample is used to check for unintentional contamination of the batch with analyte of interest from a source other than the analytical sample. The matrix blank sample is sometimes referred to as a blank quality control sample (or QC-0). A double matrix blank sample is a sample that is representative of the matrix present in the analytical sample, but to which no internal standards or analyte of interest have been added. The analysis of the double matrix blank sample is used to check whether the source matrices used to run the analysis are contaminated with either analyte or internal standards. A quality control sample is a sample that contains an amount of analyte of interest at a known concentration. These samples are used to check that the calibration of the instrument performing the analysis is still valid and to determine the accuracy of the quantity or concentration of analyte of interest determined by the system. In other words, the analyte in the quality control sample is analysed and the resulting signal converted into a quantity or concentration of the analyte using the calibration. If this concentration matches the known concentration, within a predetermined tolerance, then the calibration is determined to be valid. On the other hand, if the determined concentration does not match the known concentration, and is outside of the predetermined tolerance, then the calibration is determined not to be valid. Typically, a quality control sample is measured both before and after the analytical samples in the batch are measured, so as to determine that there has been no change in response of the system in the meantime, e.g. due to contamination during the analysis. As mentioned above, the determined quantity or concentration of the analyte must typically be within the stated error of the measurement, although one measurement outside the measurement error window may be allowable. Quality control samples are frequently provided having low, high and medium known concentrations of the analyte. A blinded or challenge quality control sample is a sample that is submitted externally by an auditing body (either internal or external to the laboratory). This type of sample is submitted for analysis in order to check the laboratory practice and the quality of results from the analysis. Typically, the laboratory or operator is unaware that a sample is a blinded or challenge quality control sample and it is handled identically to an analytical sample. Calibrator samples are a plurality of samples that are used to obtain the calibration relationship between a detected analyte signal and the quantity or concentration of that analyte. Each of these calibrator samples is representative of the matrix present in the analytical samples and include the internal standard added to it at a known concentration. Each of the calibrator samples also has a known concentration of the analyte in it, wherein different calibrator samples have different known concentrations of the analyte in it. The different calibrator samples are typically contained within separate sample wells. The calibrator samples are analysed and, as the concentrations of the analyte in each of the calibrator samples is known, the measured signal from each of these samples is able to be correlated with the concentration of analyte giving rise to it. This allows inference of the quantity of analyte present in a sample based of the response of the system. It is often important that calibrator samples contain the same matrix as the analytical samples so as to ensure that any matrix suppression or enhancement effects are accounted for in the measurement. Fig. 1 shows a layout of a typical LC-MS batch of samples on a 96-well sample plate. The sample plate has 8 rows (labelled A-H) of sample wells, where each row has 12 sample wells (labelled 1-12). In the example shown in Fig. 1, the sample plate includes a solvent blank sample at sample well A1, a double matrix blank sample at sample well A2, a matrix blank sample at sample well A3, and six calibrator samples at sample wells A4-A9. The six calibrator samples have six different respective known concentrations of the analyte, which are labelled as calibrator levels 1-6 in the diagram. The sample plate also includes a matrix blank sample at sample well A10, a quality control sample having a first relatively low known concentration of the analyte at sample well A11, a quality control sample having a second higher known concentration of the analyte at sample well A12, and a quality control sample having a third still higher known concentration of the analyte at sample well B1. The sample plate also contains a quality control sample having the third (highest) known concentration of the analyte at each of sample wells E3 and H12. The sample plate also contains a quality control sample having the first (lowest) known concentration of the analyte at each of sample wells C7 and H10. The sample plate also contains a quality control sample having the second (middle) known concentration of the analyte at each of sample wells F12 and H11. The sample plate also contains a matrix blank sample at sample well H9. All of the remaining sample wells in the sample plate are filled with analytical samples. In use the samples may be analysed in sequence, starting with the first row and then moving to the next row until all samples on the plate have been analysed. For example, the method may first sequentially analyse consecutive samples in the first row (i.e. samples A1 to A12), and then sequentially analyse consecutive samples in the second row (i.e. samples B1 to B12) etc. Accordingly, the blank samples in sample wells A1-A3 are analysed sequentially and then the calibrator samples in sample wells A4-A9 are analysed sequentially. The signals detected from the calibrator samples, and their known concentrations of analyte, are then used to determine a calibration curve that correlates a measured analyte signal to the concentration of that analyte. The samples in the remaining sample wells are then mass analysed. The calibration curve is used to convert the response measured for the analyte in any given analytical sample to a quantity or concentration value of the analyte in that sample. The quantity or concentration value for the analyte in any given analytical sample may be adjusted using the measured response of the internal standard in that analytical sample. The quality control samples are analysed and used for their purpose as has been described above. Fig. 2 illustrates the known concentrations of the analyte in the calibrator samples relative to the expected range of unknown concentrations of the analyte in some of the analytical samples. This illustration depicts the quantity (amount) of analyte in a sample for different samples to be analysed. In this example seven calibrator samples having different known concentrations of the analyte are provided in seven different respective sample wells, which are represented by the values at times 1-7 on the x-axis. Analytical samples having unknown concentrations of the analyte are provided in other sample wells, which are represented by the values at times 8-12 on the x-axis. The vertically elongated rectangle at each of times 8-12 represents the range of concentrations of analyte in an analytical sample that might be expected. The calibrator samples are selected to have concentrations that extend from a concentration below the minimum expected concentration to a concentration above the maximum expected concentration, with concentrations evenly spaced therebetween. The above described calibration process is undertaken for each batch in conventional LC-MS workflows. This is because current LC-MS workflows are highly flexible, with the same LC-MS system open to be used in many different ways. For example, the same system may be used to detect multiple different analytes or different panels of analytes, it may be used in different methods such as to analyse samples having different matrices or different sample preparation steps, it may use different chromatography columns, different chromatography solvents, different chromatography washes, and different chromatographic gradients etc. Due to the variety of ways that the system may be used, good laboratory practice recommends the acquisition of fresh calibration samples when the way in which the system is being used changes, e.g. at the start of each batch of samples to be analysed. For example, when batches of samples are being analysed that are to undergo different experimental conditions, there may be many permutations of possible analyses to consider, e.g. analysis A and then analysis B, analysis B and then analysis C, analysis D and then analysis A, etc. It is good practice to perform a recalibration each time that the conditions are changed. As laboratories and their supplies move towards more standardised conditions for analyses, in many cases there will be a low risk of the analysis changing substantially within batches or between batches. However, the avoidance of an incorrect result, and a safe and effective diagnosis, must always be of primary importance. Accordingly, recalibration is performed relatively frequently. There are several disadvantages in recalibrating the analysis frequently. For example, even if an existing calibration remains valid, the conventional approach of frequently recalibrating is still undertaken and is therefore wasteful. For instance, differences between calibration relationships that are generated due to small acceptable variations in the analyses may lead to small differences in the analytical results that would be within the stated error for the test anyway. Worse still, the conventional approach of frequently recalibrating can introduce errors into the analytical results. Also, according to the conventional approach, even if there is a significant difference between the last calibration relationship and the recalibrated calibration relationship, the last calibration relationship is still used for the preceding analytical samples. If there has been an absolute change in assay performance, the conventional technique may lead to an incorrect result for the preceding analytical samples. The conventional calibration process therefore may not account well for short term changes that affect calibration. Fig. 3 shows two 96-well sample plates according to an embodiment of the present invention. The first sample plate is filled with samples in the same manner as that shown in Fig. 1, except that calibration checkers have been added to the analytical samples at sample wells E4 and H8. In other words, the calibration checkers have been added approximately mid-way through the batch of samples on the first sample plate, and also in the final analytical sample in that batch. As such, the sample plate includes a solvent blank sample at sample well A1, a double matrix blank sample at sample well A2, a matrix blank sample at sample well A3, and six calibrator samples at sample wells A4-A9. The six calibrator samples have six different respective known concentrations of the analyte of interest, which are labelled as calibrator levels 1-6 in Fig. 3. The sample plate also includes a matrix blank sample at sample well A10, a quality control sample having a first relatively low known concentration of the analyte at sample well A11, a quality control sample having a second higher known concentration of the analyte at sample well A12, and a quality control sample having a third still higher known concentration of the analyte at sample well B1. The sample plate also contains a quality control sample having the third (highest) known concentration of the analyte at each of sample wells E3 and H12. The sample plate also contains a quality control sample having the first (lowest) known concentration of the analyte at each of sample wells C7 and H10. The sample plate also contains a quality control sample having the second (middle) known concentration of the analyte at each of sample wells F12 and H11. The sample plate also contains a matrix blank sample at sample well H9. As mentioned above, calibration checkers have been added to the analytical samples at sample wells E4 and H8. All of the remaining sample wells in the sample plate are filled with analytical samples that do not include the calibration checkers, i.e. sample wells B2-C6, C8-E2, E5-F11, and G1-H7. In the exemplary embodiment described below, three calibration checkers having three different respective known concentrations are provided in each of sample wells E4 and H8, although it will be appreciated that fewer or a greater number of calibration checkers may be used in each of these samples. The calibration checkers should ideally behave, chemically, in as similar manner as possible to the analyte of interest. As such, each of the calibration checkers may be a compound that is the same as the analyte of interest, except that one of more of the atoms in its chemical structure has been substituted for a different, stable isotope of that atom. Such a calibration checker will be referred to herein as a stable isotope labelled calibration checkers (SILCC). As such, in the embodiment having multiple calibration checkers of different known concentrations in the same analytical sample, these multiple calibration checkers will be multiple different SILCCs having multiple different respective molecular masses. In the following embodiments the calibration checkers will be described as SILCCs, although it will be appreciated that other types of calibration checker may be used instead, such as compounds that are different to the analyte of interest but in the same family of compounds as the analyte of interest. An internal standard may also be added to each of the analytical samples in the batch, where the internal standard has the same concentration in all of the analytical samples. The internal standard should also ideally behave chemically in as similar manner as possible to the analyte of interest. As such, the internal standard may be a compound that is the same as the analyte of interest, except that one of more of the atoms in its chemical structure has been substituted for a different, stable isotope of that atom. Such an internal standard compound will be referred to as a stable isotope labelled internal standard (SILIS) compound. In the following embodiments the internal standard compound will be described as a SILIS compound, although other types of internal standard could be used, such as a compound that is different to the analyte of interest but in the same family of compounds as the analyte of interest. The internal standard compound and the calibration checkers may be different, at least in the sense that they are distinguishable by the mass spectrometer. For example, the SILIS compound and the SILCCs may be the same chemical compound, but wherein one or more atoms in each of the SILCCs is a different isotope to in the SILIS. It is alternatively contemplated that the same concentration of one of the SILCCs may be added to all of the analytical standards so as to be used as the internal standard. Referring again to Fig. 3, the second sample plate is filled with samples in the same manner as the first sample plate, except that it does not include the calibrator samples and subsequent blank sample that are located in samples wells A4-A10 of the first sample plate. Instead, the second sample plate includes the calibration checkers in the first analytical sample, i.e. at sample well A7. As such, the second sample plate includes a solvent blank sample at sample well A1, a double matrix blank sample at sample well A2, a matrix blank sample at sample well A3, a quality control sample having the first relatively low known concentration of the analyte at sample well A4, a quality control sample having the second higher known concentration of the analyte at sample well A5, and a quality control sample having the third still higher known concentration of the analyte at sample well A6. The sample plate also contains a quality control sample having the third (highest) known concentration of the analyte at each of sample wells E3 and H12. The sample plate also contains a quality control sample having the first (lowest) known concentration of the analyte at each of sample wells C7 and H10. The sample plate also contains a quality control sample having the second (middle) known concentration of the analyte at each of sample wells F12 and H11. The sample plate also contains a matrix blank sample at sample well H9. The calibration checkers have been added to the analytical samples at each of sample wells A7, E1 and H8. All of the remaining sample wells in the sample plate are filled with analytical samples that do not include the calibration checkers, i.e. sample wells A8-C6, C8-D12, E2, E4-F11, and G1-H7. In use, the samples on the first sample plate are analysed one at a time and consecutively by LC-MS in the sequence described above in relation to Fig. 1, i.e. from left to right in each row and one row at a time, starting with row A and then proceeding to the next row consecutively from row B to row H. As will be appreciated, in LC-MS a sample is introduced into a liquid chromatograph device that separates components in the sample such that they elute from the liquid chromatograph device over different time periods. As the components elute they are ionised by an ionisation source and the resulting ions are mass analysed in the mass spectrometer. In this example the mass spectrometer is a tandem (MS / MS) mass spectrometer that monitors certain multiple reaction monitoring (MRM) transitions and produces an ion chromatogram for each of these, i.e. records the detected ion signal as a function of time. The method starts by using LC-MS to sequentially analyse the blank samples in sample wells A1-A3 of the first sample plate, and then the calibrator samples in sample wells A4-A9 are analysed sequentially in the conventional manner. For example, the mass spectrometer monitors an MRM transition for the analyte of interest when mass analysing the calibrator samples. The ion signals detected from the calibrator samples, and their known concentrations of the analyte of interest, are then used to determine a first calibration relationship that correlates an ion signal measured for the analyte of interest to the concentration of that analyte. The quality control samples in sample wells A10-B1 are then analysed sequentially in the conventional manner. The method then proceeds to sequentially analyse the samples in the subsequent sample wells, wherein for each of the analytical samples the mass spectrometer monitors different MRM transitions in order to detect an ion signal for the analyte of interest (and the SILIS, if one has been added to the analytical samples). The first calibration relationship is used to convert the ion signal measured for the analyte of interest in the analytical samples to a concentration or quantity value of the analyte of interest in that sample. For example, the first calibration relationship may be used to calibrate the concentration or quantity of the analyte of interest in each of the analytical samples in sample wells B2-C6 and C8-E2. The concentration or quantity value for the analyte in any given analytical sample may be adjusted in the known manner using the measured response of the SILIS in that analytical sample, if one is present. The quality control samples at sample wells C7 and E3 are mass analysed in the known manner and used for the known purpose, as has already been described above. When the mass spectrometer mass analyses the analytical samples that contains the SILCCs, such as the analytical sample in sample well E4, the mass spectrometer monitors an MRM transition in order to detect an ion signal for the analyte of interest, and also monitors three different MRM transitions in order to detect an ion signal for each of the three SILCCs therein. If a separate SILIS has also been added to the analytical samples then the mass spectrometer also monitors an MRM transition in order to detect an ion signal for the SILIS. As one skilled in the art will appreciate, in order to monitor any given MRM transition the mass spectrometer controls a first mass filter to have a mass transmission window such that it is only capable of transmitting precursor ions having a specific mass to charge ratio. The precursor ions that are transmitted by the first mass filter are guided into a fragmentation or reaction device, in which the precursor ions are fragmented or reacted so as to form fragment or product ion species. The fragment or product ions are then transmitted to a second mass filter, which is operated so as to have a mass transmission window such that it is only capable of transmitting ions having a specific mass to charge ratio to an ion detector. The mass to charge ratio values of the mass transmission windows for the first and second mass filters are selected so that if ions are detected by the ion detector then it is determined that ions of interest are present in the sample being analysed. The ion signal is also used to determine the quantity of ions in the sample. For example, for the MRM transition for the analyte of interest, the mass to charge ratio values of the mass transmission windows for the first and second mass filters are selected so that only ions having a mass to charge ratio corresponding to a specific fragment ion species of the analyte of interest are able to be detected by the ion detector. The ion signal for these ions may therefore be used to determine the quantity or concentration of the analyte of interest in the sample. A different MRM transition for the analyte of interest may also be monitored, wherein the mass to charge ratio values of the mass transmission windows for the first and second mass filters are selected so that only ions having a mass to charge ratio corresponding to a different fragment ion species of the analyte of interest are able to be detected by the ion detector. The mass spectrometer may monitor this MRM transition so that if ions are detected for this MRM transition then it is confirmed that it is actually the analyte of interest that is being detected, since it is highly unlikely that a precursor ion other than the analyte of interest would give rise to ion signals for both MRM transitions. Similarly, for each SILCC, the mass to charge ratio values of the mass transmission windows for the first and second mass filters are selected so that only ions having a mass to charge ratio corresponding to a specific fragment ion species of the SILCC are able to be detected by the ion detector. The ion signal for these ions may therefore be used to determine the quantity or concentration of the SILCC in the sample. A different MRM transition for the SILCC may also be monitored, wherein the mass to charge ratio values of the mass transmission windows for the first and second mass filters are selected so that only ions having a mass to charge ratio corresponding to a different fragment ion species of the SILCC are able to be detected by the ion detector. The mass spectrometer may monitor this MRM transition so that if ions are detected for this MRM transition then it is confirmed that it is actually the SILCC that is being detected, since it is highly unlikely that a precursor ion other than the SILCC would give rise to ion signals for both MRM transitions. For the SILIS, the mass to charge ratio values of the mass transmission windows for the first and second mass filters are selected so that only ions having a mass to charge ratio corresponding to a specific fragment ion species of the SILIS are able to be detected by the ion detector. The ion signal for these ions may therefore be used to determine the quantity or concentration of the SILIS in the sample. A different MRM transition for the SILIS may also be monitored, wherein the mass to charge ratio values of the mass transmission windows for the first and second mass filters are selected so that only ions having a mass to charge ratio corresponding to a different fragment ion species of the SILIS are able to be detected by the ion detector. The mass spectrometer may monitor this MRM transition so that if ions are detected for this MRM transition then it is confirmed that it is actually the SILIS that is being detected, since it is highly unlikely that a precursor ion other than the SILIS would give rise to ion signals for both MRM transitions. Preferably, the MRM transition(s) for each of the analyte of interest, the SILIS, and the SILCCs are selected such that ions will be detected at the detector if these undergo the same manner of fragmentation, i.e. these compounds are preferably fragmented in the same place(s) of the chemical structure. The MRM transition(s) for each of the analyte of interest, the SILIS, and the SILCCs will however require the mass filters to have slightly different mass transmission windows, as these compounds have atoms of different isotopes in them. This is how the mass spectrometer is able to distinguish them from each other when they are present in the same sample. For the analysis of each of the samples, the mass spectrometer may cycle through the above-described MRM transitions as sample elutes from the liquid chromatograph device, so as to obtain ion chromatograms for each of the MRM transitions that correspond to the analyte of interest, the SILIS, and the SILCCs. The mass spectrometer may be programmed not to monitor the MRM transitions for the SILCCs when mass analysing samples that are known not to contain the SILCCs. The mass spectrometer determines the abundance of ions in the ion peak for each ion chromatogram, e.g. by determining the area under each peak. The mass spectrometer may have slightly different responses when detecting the analyte of interest, the SILIS, and the SILCCs and as such a relative response factor may be calculated for each of the analyte of interest and the SILCCs (relative to the SILIS compound). For example, the abundance of the analyte of interest may be divided by the abundance of the SILIS so as to obtain a relative response factor for the analyte of interest, and similarly the abundance of each SILCC may be divided by the abundance of the SILIS so as to obtain a relative response factor for the SILCC. The relative response factor for each SILCC may then be used to calculate an apparent concentration for the SILCC, e.g. by multiplying the known concentration of that SILCC by its relative response factor. As such, the mass spectrometer determines a data pair consisting of the ion signal detected (ion abundance) and the apparent quantity or concentration of the SILCC that gave rise to this signal. This process is performed on each SILCC in sample well E4 so as to obtain a data pair for each of the three SILCCs. These three data pairs are then used to check if the first calibration relationship that was obtained by analysing the samples at sample wells A4-A9 is still valid for use in calibrating the analyte of interest in analytical samples, or if a new calibration relationship should be used to do this. For example, the spectrometer may determine if at least one, or at least some, of the data pairs fall within a preselected tolerance of the first calibration relationship, and if they do then the first calibration relationship may be determined to still be valid. If the first calibration relationship is determined to still be valid then it is used to calibrate the concentration or quantity of the analyte of interest in at least some of the analytical samples that are mass analysed after the sample containing the SILCCs, i.e. in at least some of sample wells E5-H7. In contrast, if the spectrometer determines that at least one, or at least some, of the data pairs fall outside of the preselected tolerance of the first calibration relationship, that first calibration relationship may be determined to no longer be valid. In this event, the spectrometer uses a new calibration relationship to calibrate the concentration or quantity of analyte in at least some of the analytical samples that are mass analysed after the sample containing the SILCCs, i.e. in sample wells E5-H7. The new calibration relationship may be calculated based at least in part on the data pairs for the SILCCs in sample well E4. The new calibration relationship may be calculated based only on the data pairs for these SILCCs. In other words, the new calibration relationship may be calculated based on the ion signal that is detected for each SILCC and the known quantity or concentration of the SILCC giving rise to the detected ion signal, e.g. by least squares fitting of the instrument response to the known concentrations or quantities. Alternatively, the first calibration relationship may be updated by using data that was used to obtain the first calibration relationship and also the new data pairs for the SILCCs. In the event that the mass spectrometer determined that the first calibration relationship is no longer valid, it may also prevent the concentration or quantity of the analyte of interest in at least some of the analytical samples immediately prior to the sample containing the SILCCs (e.g. a predetermined number of them) from being reported / displayed. Alternatively, the spectrometer may report / display these values, but it may flag them as being inaccurate. It may also indicate the level of inaccuracy, e.g. based on the difference between the first calibration relationship and the data pairs for the SILCCs. Alternatively, the spectrometer may calibrate the concentration or quantity of the analyte of interest in at least some of the analytical samples immediately prior to the sample containing the SILCCs (e.g. a predetermined number of them) using the new calibration relationship. When the analytical sample in sample well H8 is mass analysed, a data pair for each of the three SILCCs therein is obtained, in a corresponding manner to that described above for the sample in sample well E4. These three data pairs are then used to check if the calibration relationship that is being used at that time is still valid for use in calibrating the analyte of interest in analytical samples, or if a new calibration relationship should be used to do this. For example, if the data obtained from the SILCCs in sample well E4 indicated that the first calibration relationship was still valid then the SILCCs in sample well H8 are used to check again if the first calibration relationship is still valid. This is done in a corresponding manner to that described above. The spectrometer may determine that at least one, or at least some, of the data pairs for the SILCCs in sample well H8 fall within a preselected tolerance of the first calibration relationship, in which case that first calibration relationship is determined to still be valid. If the first calibration relationship is determined to still be valid then it is used to calibrate the concentration or quantity of analyte of interest in the analytical samples that are mass analysed before the SILCCs in sample well H8, i.e. in sample wells E5-H7. In the event that the data obtained from the SILCCs in sample well H8 indicates that the first calibration relationship is no longer valid, the spectrometer may prevent the concentration or quantity of the analyte of interest in at least some of the analytical samples immediately prior to the sample containing the SILCCs in sample well H8 (e.g. a predetermined number of them) from being reported / displayed. Alternatively, the spectrometer may report / display these values, but it may flag them as being inaccurate. It may also indicate the level of inaccuracy, e.g. based on the difference between the first calibration relationship and the data pairs for the SILCCs in sample well H8. Alternatively, the spectrometer may calibrate the concentration or quantity of the analyte of interest in at least some of the analytical samples immediately prior to the sample containing the SILCCs (e.g. a predetermined number of them) using a new calibration relationship generated from the SILCCs in sample well H8. This new calibration relationship may be calculated in a corresponding manner to that described above, except based on the ion signal that is detected for each SILCC in sample well H8 and the known quantity or concentration of the SILCC giving rise to the detected ion signal. On the other hand, if the data obtained from the SILCCs in sample well E4 indicated that the first calibration relationship was not valid and a new, second calibration relationship was obtained using the SILCCs in sample well E4, then the SILCCs in sample well H8 are used to check if the second calibration relationship is still valid. This check is done in a corresponding manner to that described above in relation to checking the first calibration relationship. For example, the spectrometer may determine that at least one, or at least some, of the data pairs for the SILCCs in sample well H8 fall within a preselected tolerance of the second calibration relationship, in which case that second calibration relationship is determined to still be valid. If the second calibration relationship is determined to still be valid then it is used to calibrate the concentration or quantity of analyte of interest in at least some of the analytical samples that are mass analysed before the SILCCs in sample well H8, i.e. in sample wells E5-H7. However, if the data obtained from the SILCCs in sample well H8 indicate that the second calibration relationship is no longer valid, the spectrometer may prevent the concentration or quantity of the analyte of interest in at least some of the analytical samples immediately prior to the sample containing the SILCCs in sample well H8 (e.g. a predetermined number of them) from being reported / displayed. Alternatively, the spectrometer may report / display these values, but it may flag them as being inaccurate. It may also indicate the level of inaccuracy, e.g. based on the difference between the second calibration relationship and the data pairs for the SILCCs in sample well H8. Alternatively, the spectrometer may calibrate the concentration or quantity of the analyte of interest in at least some of the analytical samples immediately prior to the sample containing the SILCCs (e.g. a predetermined number of them) using a new calibration relationship generated from the SILCCs in sample well H8. This new calibration relationship may be calculated in a corresponding manner to that described above, except based on the ion signal that is detected for each SILCC in sample well H8 and the known quantity or concentration of the SILCC giving rise to the detected ion signal. Once the first batch of samples on the first sample plate has been analysed, the method then proceeds to analyse the samples in the second batch on the second sample plate. As mentioned above, the second batch of samples include the same various blank samples and quality control samples that were included in the first batch, since it is good laboratory practice to analyse these samples. However, as also described above, the calibrator samples in sample wells A4-A9 of the first batch need not be included in the second batch (and hence also the blank sample in sample well A10 of the first batch need not be included). Rather, the first analytical sample in sample well A7 includes the SILCCs and when this sample is mass analysed a new calibration relationship may be calculated based on the data pairs for the SILCCs in this sample well. The new calibration relationship is calculated based on the ion signal that is detected for each SILCC and the known quantity or concentration of the SILCC giving rise to the detected ion signal, e.g. by least squares fitting of the instrument response to the known concentrations (or quantities). The subsequent samples in the second batch are then sequentially mass analysed, and the SILCCs in sample wells E1 and H8 are used to check the validity of the calibration relationship being used at their time of analysis, in a corresponding way to that which has been described above in relation to the SILCCs in sample wells E4 and H8 of the first batch. Less preferably, rather than the SILCCs in sample well A7 of the second batch being used to form a new calibration relationship, these SILCCs could be used to check if the calibration relationship used at the end of the previous batch is still valid or not. If it is determined to be valid then that calibration relationship could be applied to the analytical samples in the second batch, whereas if is not then a new calibration relationship could be determined from the SILCCs in sample well A7 of the second batch and that new calibration relationship could be applied to the analytical samples in the second batch. The second sample plate in Fig. 3 has been described as having SILCCs in sample well A7, rather than calibrator samples as has been described in relation to sample wells A4-A9 of the first sample plate. However, it is contemplated that the types of samples in all of the sample wells of the second sample plate may instead be the same as in the first sample plate, i.e. so that the second sample plate also includes the calibrator samples. Although only two sample plates are shown in Fig. 3, it will be appreciated that the method may analyse further sample plates having samples types that are arranged in the same or different manners to those which have been described in relation to the first or second sample plates. For example, further sample plates may be provided in which the samples are prepared and analysed in the same manner as the second sample plate, and these samples mass analysed. If a significant change in a calibration relationship is detected when checking if it is still valid, e.g. greater than a threshold amount, then when a subsequent sample plate is prepared the calibrator samples may be included in that sample plate. Fig. 4 illustrates the known concentrations of the SILCCs relative to the expected range of concentrations of the analyte of interest in some of the samples on the first sample plate of Fig. 3. More specifically, concentrations are shown for sample numbers 52-55, which correspond to sample wells E4-E7 on the sample plate. The vertically elongated rectangles represent the range of concentrations of the analyte of interest in the analytical samples that might be expected. It can be seen that the known concentrations of the SILCCs are selected so as to extend over a range from a concentration that is below the minimum expected concentration of the analyte of interest to a concentration that is above the maximum expected concentration of the analyte of interest, with the remaining known concentration of the SILCCs therebetween. The technique described above may also be used to determine that a calibration relationship is valid, but predict when a recalibration will be required at a future time. For example, the method may check the validity of a calibration relationship multiple times in the manner described above (i.e. using the SILCCs in multiple respective sample wells), wherein during each of those checks the calibration relationship is determined to be valid. However, the spectrometer may determine how the data pairs obtained from the SILCCs differ from the calibration relationship at each of the multiple validity checks, and determine a trend of how this difference changes with time. The spectrometer may use the trend to predict a future time when the data pairs obtained from the SILCCs will differ from the calibration relationship by more than the pre-selected tolerance. The spectrometer may then use the data from SILCCs that are mass analysed prior to said future time in order to generate a new calibration relationship that is used to calibrate the concentration or quantity of the analyte of interest in analytical samples that are subsequently mass analysed. An example will now be described that models an embodiment in which the analyte of interest having an unknown concentration (e.g. in a patient sample) is testosterone and in which there are three SILCCs having different concentrations present in the analytical sample. In this example, the SILCCs are di-, tri-, and penta-deuterated testosterone. During the analyses described below, the analyte of interest of unknown concentration is monitored using an MRM transition of 289.25 >96.9. The di-deuterated testosterone is monitored using an MRM transition of 291.25 >98.9. The tri-deuterated testosterone is monitored using an MRM transition of 292.25 >96.9. The penta-deuterated testosterone is monitored using an MRM transition of 294.25 >99.9. It will therefore be appreciated that when the analytical sample containing the SILCCs is analysed the mass spectrometer will monitor the MRM transition for the analyte of interest and also the MRM transitions for each of di-, tri- and penta-deuterated testosterone, and produce ion chromatograms for these MRM transitions. Figs. 5A-5D show example ion-chromatograms for the above described MRM transitions. More specifically, Fig. 5A shows an ion-chromatogram for the analysis of the analyte of interest having an unknown concentration, which was obtained by monitoring the MRM transition of 289.25 >96.9. Fig. 5B shows an ion-chromatogram for the analysis of 0.2 ng / mL di-deuterated testosterone, which was obtained by monitoring the MRM transition of 291.25 >98.9. Fig. 5C shows an ion-chromatogram for the analysis of 4.0 ng / mL tri-deuterated testosterone that was obtained by monitoring the MRM transition of 292.25 >96.9. Fig. 5D shows an ion-chromatogram for the analysis of 10.0 ng / mL pentadeuterated testosterone that was obtained by monitoring the MRM transition of 294.25 >99.9. The integrated peak area is determined for the peak in each ion chromatogram, so as to obtain a data pair consisting of the peak area and known concentration for each of the SILCCs. These data pairs for the SILCCs may then be used to form a calibration relationship that relates a peak area measured in an ion chromatogram to a concentration of analyte of interest. Fig. 6 shows a calibration relationship that is calculated based on the peak areas measured from the ion chromatograms shown in Figs. 5B-5D and the known concentrations of the SILCCs that gave rise to these peaks. The peak area for the analyte of interest that is shown in the ion chromatogram of Fig. 5A was determined and, using the calibration relationship shown in Fig. 6, it was determined that the analyte of interest has a concentration of 2.85 ng / mL. As mentioned above, although a set of three SILCCs having three different concentrations has been described as being added to each of the analytical samples that contains SILCCs, the number of SILCCs having different concentrations that are added to each of these analytical samples may be different to this. For example, a greater number of SILCCs may be provided in each of the analytical samples in which they are provided. This increases the accuracy with which the validity of the existing calibration relationship can be checked, and it also provides a better calibration relationship if the SILCCs are used to form a new calibration relationship. Fig. 7 shows two 96-well sample plates according to another embodiment of the present invention. This embodiment is the same as that shown in Fig. 3, except that the first sample plate in Fig. 7 does not include the six calibrator samples and the subsequent blank sample that are located in samples wells A4-A10 of the first sample plate of Fig. 3. Instead, the first sample plate in Fig. 7 includes six calibrators in the first analytical sample, i.e. at sample well A7. As such, the first sample plate in the embodiment of Fig. 7 includes a solvent blank sample at sample well A1, a double matrix blank sample at sample well A2, a matrix blank sample at sample well A3, a quality control sample having the first relatively low known concentration of the analyte at sample well A4, a quality control sample having the second higher known concentration of the analyte at sample well A5, a quality control sample having the third still higher known concentration of the analyte at sample well A6, and six calibrators in the first analytical sample at sample well A7. The sample plate also contains a quality control sample having the third (highest) known concentration of the analyte at each of sample wells E3 and H12. The sample plate also contains a quality control sample having the first (lowest) known concentration of the analyte at each of sample wells C7 and H10. The sample plate also contains a quality control sample having the second (middle) known concentration of the analyte at each of sample wells F12 and H11. The sample plate also contains a matrix blank sample at sample well H9. The SILCCs have been added to the analytical samples at each of sample wells E1 and H8. All of the remaining sample wells in the sample plate are filled with analytical samples that do not include the calibrators or SILCCs, i.e. sample wells A8-C6, C8-D12, E2, E4-F11, andG1-H7. The second sample plate in Fig. 7 is the same as the second sample plate in Fig. 3. The calibrators in sample well A7 of the first sample plate should ideally behave, chemically, in as similar manner as possible to the analyte of interest. As such, each of the calibrators may be a compound that is the same as the analyte of interest, except that one of more of the atoms in its chemical structure has been substituted for a different, stable isotope of that atom. Such a calibrator will be referred to herein as a stable isotope labelled calibrator (SILC). As such, in the embodiment having six calibrators of different known concentrations in the same analytical sample, these six calibrators will be six different SILCs having six different respective molecular masses. It will be appreciated that these SILCs are detected by monitoring MRM transitions for them. In the following embodiments the calibrators will be described as SILCs, although it will be appreciated that other types of calibrator may be used instead, such as compounds are different to the analyte of interest but in the same family of compounds as the analyte of interest. It will be appreciated that as the SILCs are not provided in the same analytical samples as the SILCCs, and hence need not be distinguished from each other in any given analytical sample, the materials used for the SILCs may be the same materials that are used for the SILCCs. Alternatively, the number of SILCCs may be fewer than the number of SILCs. For example, if n SILCs are used, then n-1 SILCCs may be provided that are the same materials as used for n-1 of the SILCs. The material for the remaining SILC may be provided in the same concentration in all of the analytical samples, so that it can be used as an internal standard (i.e. as the SILIS described herein). After the analytical sample at sample well A7 of the first sample plate has been mass analysed, the mass spectrometer is able to determine a first calibration relationship based on the known concentration levels of SILCs 1-6, in a corresponding manner to that described in the above embodiments. The method then proceeds to sequentially analyse the further samples in other sample wells of the sample plate. For example, the method will next sequentially analyse the samples from sample wells A8-D12 in a corresponding manner to that described above. The method may use the calibration relationship determined using the SILCs in sample well A7 to calibrate the concentration or quantity of the analyte of interest in some or all of the samples in sample wells A8-D12. The mass spectrometer then mass analyses the sample in sample well E1, including the SILCCs therein, and checks the validity of the first calibration relationship in a corresponding manner to as has been discussed in the previous embodiment. The embodiment of Fig. 7 may then proceed as has been described above in relation to Fig. 3. Fig. 8 illustrates the known concentrations of the SILCs and SILCCs relative to the expected range of concentrations of the analyte of interest in some of the samples on the first sample plate of Fig. 7. In this embodiment, the same materials are used for both the SILCs and the SILCCs, i.e. six analogues of the analyte of interest having six different concentrations and molecular masses mi-me are used for the SILCs and also the SILCCs. Concentrations are shown for sample numbers 7-8, 48-50 and 91-92, which correspond to sample wells A7-A8, D12-E2 and H7-H8 respectively. The vertically elongated rectangles represent the range of concentrations of the analyte of interest in the analytical samples that might be expected. It can be seen that the known concentrations of the SILCs and SI LCCs are selected so as to extend over a range from a concentration that is below the minimum expected concentration of the analyte of interest to a concentration that is above the maximum expected concentration of the analyte of interest, with the remaining known concentration of the SILCs and SILCCs therebetween. The methods described herein may be performed by automated machinery that receives samples to be tested and selects some of them to be tested together, e.g. as shown in Fig. 9. Fig. 9 shows an embodiment of automated machinery that receives samples 2, e.g. which are deposited on a conveyer 3. The samples 2 have identifiers thereon, such as barcodes, that identify the type of sample that they are and / or the type(s) of test that they should be subjected to. The samples are conveyed to a reading device 4 that reads the information on the identifier. The automated machinery moves the samples from the conveyor 3 to a storage region 5, where samples of the same type and / or that are to be subjected to the same test(s) are grouped together in the storage region. Samples of different types and / or that are to be subjected to different tests (or different sets of tests) may be placed in the storage region in different groups. The automated machinery is configured to select a group of samples from the storage region 5 and prepare a batch of analytical samples from them, e.g. by transferring samples to sample wells on a sample plate 6 using a robotic arm 7 or other automated means. The automated machinery then performs sample preparation steps required in order to conduct the test(s) on the analytical samples, e.g. by adding one or more compound 8 to, or removing one or more compound from, the analytical samples. These steps are known to those skilled in the art and will not be described, save to say that the sample preparation steps involve adding the different known concentration levels of the calibrators and SILCCs to the analytical samples. The automated machinery then outputs the batch of prepared samples for mass analysis by the mass spectrometer. In the illustrated embodiment this is performed by moving the sample plate 6 to a conveyor 9, although other mechanism could be used. The automated machinery may include an automated sampler that automatically subjects the analytical samples to LC-MS as has been described above. Although the present invention has been described with reference to various embodiments, it will be understood by those skilled in the art that various changes in form and detail may be made without departing from the scope of the invention as set forth in the accompanying claims. For example, although embodiments have been described in which all of the SILCCs in an analytical sample are used to check the validity of the current calibration relationship, it is contemplated that only a sub-set of the SILCCs may be used to perform this check. If this check results in the determination that the calibration relationship is no longer valid and a new calibration relationship is required, then all of the SILCCs may be used to form the new calibration relationship. Various layouts of sample types on each sample plate have been described. However, other layouts are contemplated, such as wherein the SILCCs are provided in every mth sample well on the sample plate, where m is an integer. Alternatively, the SILCCs may be provided in all analytical samples. Although the SILCCs have been described as being provided in analytical samples, they may instead, or additionally, be provided in other types of samples, such as in the quality control samples. Although a plurality of SILCCs have been described as being in the same sample, they may alternatively be distributed amongst a plurality of the samples. Although embodiments have been described in which a batch of 96 samples are provided on a sample plate, it will be appreciated that fewer or more samples may be provided in each batch and / or that they need not be provided on a sample plate. It is contemplated that the samples may not even be provided on sample plates or in batches, e.g. they may be provided as a stream of samples that are consecutively analysed. The samples on the sample plate have been described as being analysed consecutively, from left to right in each row and one row at a time, starting with row A and then proceeding to the next row. However, the samples may be analysed in any order. For example, the samples in each row may be analysed consecutively, from right to left. Alternatively, the samples may be analysed consecutively, from top to bottom in each column and one column at a time. The LC-MS analysis described herein may use an Ultra Performance Liquid Chromatography (UPLC) system, e.g. such as the Waters l-Class system. However, alternative approaches and chromatography pressure regimes may be used. The LC-MS analysis may use a single chromatography column and the samples may be injected into the column sequentially. Alternatively, different samples may be analysed during at least partially overlapping time periods by injecting them into different chromatography columns. The chromatography may be carried out according to any of the known techniques, such as multichannel chromatography, two-dimensional chromatography, flow injection analysis, or trap-and-elute chromatography etc.). Embodiments have been described in which the chromatographically separated samples are analysed by a tandem mass spectrometer, such as a triple quadrupole mass spectrometer, operating in a MRM mode. However, other mass analysis techniques and / or mass spectrometer may be used to obtain ion peaks for the analyte of interest and calibrators in any given sample. For example, a triple quadrupole mass spectrometer operating in a mode other than an MRM mode may be used. Alternatively, a single quadrupole mass spectrometer, time of flight mass spectrometer, Orbitrap mass spectrometer, FT-ICR mass spectrometer, or other mass spectrometer may be used. Although the samples have been described as being separated by liquid chromatography, the liquid samples need not be separated prior to analysis, or they may be separated by alternative techniques. For example, the samples may be subjected to capillary electrophoresis, infusion techniques or direct analysis prior to mass analysis. Although the samples have been described as being mass analysed, it is contemplated that the samples may additionally, or alternatively, be analysed using ion mobility analysis. Alternatively, the samples may be analysed by other techniques, such as optical analysis (e.g. spectroscopy). Various sample types and analytes of interest have been described herein. However, the invention is not limited to these. For example, the analyte of interest may be a steroid hormone, a vitamin, a drug of abuse (for example an opiate), or a therapeutic drug such as an immunosuppressant drug, an anti-epileptic drug, an anti-psychotic drug, an anti-infective drug, a steroid, or an anti-fungal drug etc. The analyte of interest may be an amino acid, a peptide or a protein. The methods described herein determine the concentration of an analyte of interest in the samples. However, it will be appreciated that the invention extends to determining the concentration of one or more additional analyte of interest in the analytical samples in a corresponding manner to that described. For example, each analyte of interest may have its own respective calibration checkers. The multiple analytes of interest may be in the same samples or may be in different respective samples. For instance, a given batch of samples to be analysed may contain a first set of analytical samples, and their associated quality control samples and / or blank samples, that are to be subjected to a first test (e.g. for an immunosuppressant drug panel) and a second different set of analytical samples, and their associated quality control samples and / or blank samples, that are to be subjected to a different test (e.g. for a steroid hormone panel).

Claims

1. A method of determining the concentration or quantity of an analyte of interest in an analytical sample comprising:obtaining a first calibration relationship that relates the concentration or quantity of an analyte of interest to an ion signal detected by a mass spectrometer for the analyte of interest;using a mass spectrometer to mass analyse a first set of samples that includes analytical samples containing the analyte of interest, wherein a first plurality of calibration checkers having different known concentration levels are provided in one or more of the samples in said first set of samples, wherein said mass analysing obtains an ion signal for an analyte of interest in each of the analytical samples and an ion signal for each of the known concentration levels of said calibration checkers; andperforming a comparison in which values that are representative of the ion signals and known concentration levels of at least some of the calibration checkers are compared to the first calibration relationship to determine if the first calibration relationship is still valid.

2. The method of claim 1, wherein the calibration checkers are provided in an analytical sample and / or in a quality control sample.

3. The method of claim 1 or 2, wherein said comparison comprises: determining that the first calibration relationship is still valid if said values fit the first calibration relationship to within a preselected tolerance; and determining that the first calibration relationship is no longer valid if said values do not fit the first calibration relationship to within a preselected tolerance.

4. The method of claim 3, wherein said comparison comprises initially uses the ion signals and known concentration levels for only a sub-set of said calibration checkers in determining if their values fit the first calibration relationship to within the preselected tolerance, wherein if they do then the method determines that the first calibration relationship is still valid.

5. The method of claim 4, wherein if the values for the sub-set of calibration checkers do not fit the first calibration relationship to within a preselected tolerance then the method uses the ion signals and known concentration levels for all of the calibration checkers in determining if their values fit the first calibration relationship to within the preselected tolerance, wherein if they do then the method determines that the first calibration relationship is still valid, whereas if they do not then the method determines that the first calibration relationship is no longer valid.

6. The method of any preceding claim, wherein when said comparison determines that the first calibration relationship is no longer valid, the method comprises using the ion signals and known concentration levels of the calibration checkers to generate a second calibration relationship.

7. The method of claim 6, comprising using the ion signals and known concentration levels for only a sub-set of the calibration checkers in said comparison step, and using the ion signals and known concentration levels for all of the calibration checkers to generate the second calibration relationship.

8. The method of claim 6 or 7, comprising mass analysing a second set of samples that includes analytical samples containing the analyte of interest after said one or more sample containing the calibration checkers has been mass analysed; and determining the concentration or quantity of the analyte of interest in at least one of the analytical samples in the second set of samples, using the ion signal detected for the analyte of interest in that analytical sample and the second calibration relationship.

9. The method of claim 6, 7 or 8, comprising determining the concentration or quantity of the analyte of interest in at least one of the analytical samples in the first set of samples, using the ion signal detected for the analyte of interest in that analytical sample and the second calibration relationship.

10. The method of any preceding claim, wherein the method comprises: controlling apparatus to generate an output that indicates a determined concentration or quantity of the analyte of interest in an analytical sample;using the first calibration relationship to determine the concentration or quantity of the analyte of interest in an analytical sample in the first set of samples;determining from said comparison step that the first calibration relationship is no longer valid, and in response to this either:i) controlling the apparatus so as not to generate an output that indicates the determined concentration or quantity of the analyte of interest in the analytical sample in the first set of samples; orii) generating an output that indicates the determined concentration or quantity of the analyte of interest in the analytical sample in the first set of samples along with an indication that the determined concentration or quantity is not valid.

11. The method of any preceding claim, wherein in response to said comparison determining that the first calibration relationship is no longer valid, the method adds a plurality of calibration checkers at different known concentration levels to one or more further samples, mass analyses those one or more further samples, and determines a new calibration relationship from the resulting ion signals.

12. The method of claim 11, comprising mass analysing a further set of samples that includes analytical samples containing the analyte of interest after said one or more samples containing the plurality of calibration checkers have been mass analysed; and determining the concentration or quantity of the analyte of interest in at least one of the analytical samples in the further set of samples, using the new calibration relationship.

13. The method of any one of claims 1-5, wherein when said comparison determines that the first calibration relationship is still valid, the method comprises determining the concentration or quantity of the analyte of interest in at least one analytical sample that is mass analysed, using the ion signal detected for the analyte of interest in that analytical sample and the first calibration relationship.

14. The method of claim 13, comprising mass analysing a second set of samples that includes analytical samples containing the analyte of interest after said sample containing the first plurality of calibration checkers has been mass analysed; and determining the concentration or quantity of the analyte of interest in at least one of the analytical samples in the second set of samples, using the ion signal detected for the analyte of interest in that analytical sample and the first calibration relationship.

15. The method of claim 8, 12 or 14, further comprising:using the mass spectrometer to mass analyse one or more further sample containing a third plurality of calibration checkers at different known concentration levels, after mass analysing said second set samples;wherein said mass analysing obtains an ion signal for each of the known concentration levels of said third plurality of calibration checkers; andwherein a comparison is performed in which values that are representative of the ion signals and known concentration levels of at least some of the third plurality of calibration checkers are compared to the first, second or new calibration relationship to determine if that calibration relationship is still valid.

16. The method of any one of claims 1-5, 13 and 14, comprising determining that the first calibration relationship is still valid based on said values fitting the first calibration relationship to within a preselected tolerance; determining a difference between said values and the first calibration relationship, and predicting when the first calibration relationship will turn from being valid to no longer being valid based on said difference.

17. The method of claims 1-5, 13 and 14, comprising:a) repeatedly performing a cycle, wherein each cycle comprises:i) mass analysing one or more samples that contain a plurality of calibration checkers at different known concentration levels so as to obtain an ion signal for each of the known concentration levels of said calibration checkers,ii) performing a comparison in which values that are representative of the ion signals and known concentration levels of at least some of these calibration checkers are compared to the first calibration relationship, andiii) determining that the first calibration relationship is still valid based on said values fitting the first calibration relationship to within a preselected tolerance, and determining a difference between said values and the first calibration relationship; and b) determining a trend of how the difference determined in step iii) changes between the cycles; andc) predicting when the first calibration relationship will turn from being valid to no longer being valid based on said trend.

18. The method of claim 16 or 17, comprising, in response to said predicting step, calculating a new calibration relationship at a point before the first calibration relationship has been predicted to turn from being valid to no longer being valid.

19. The method of any preceding claim, wherein each of said calibration checkers is a compound that is the same as the analyte of interest, except that one of more of the atoms in its chemical structure has been substituted for a different, stable isotope of that atom.

20. The method of any preceding claim, comprising providing a plurality of batches of samples and performing the method of any preceding claim on each of the batches.

21. The method of any preceding claim, wherein the samples are mass analysed using liquid chromatography mass spectrometry; optionally wherein each of the analytical samples is mass analysed using tandem mass spectrometry by monitoring an MRM transition for said analyte of interest, and at least the samples containing the calibration checkers are mass analysed using tandem mass spectrometry by monitoring an MRM transition for each of the calibration checkers.

22. The method of any preceding claim, wherein the method is performed by automated machinery that:(i) receives samples;(ii) adds said plurality of different known concentration levels of the calibration checkers to one or more of the samples in said set of samples; and(iii) transfers the set of samples to one or more mass spectrometer so as to then perform the method of any preceding claim.

23. The method of claim 22, wherein each of the analytical samples that is received at the automated machinery has an identifier thereon that represents the type of sample that it is and / or a test that is required to be performed on that sample; andwherein the automated machinery reads the identifier and selects analytical samples of the same type and / or that are to undergo the same test and includes these analytical samples in said set of samples.

24. A method of determining the concentration or quantity of an analyte of interest in an analytical sample comprising:obtaining a first calibration relationship that relates the concentration or quantity of an analyte of interest to a signal detected by an analyser for the analyte of interest;using an analyser to analyse a first set of samples that includes analytical samples containing the analyte of interest, wherein a first plurality of calibration checkers having different known concentration levels are provided in one or more of the samples in said first set of samples, wherein said analysing obtains a signal for an analyte of interest in each of the analytical samples and a signal for each of the known concentration levels of said calibration checkers; andperforming a comparison in which values that are representative of the signals and known concentration levels of at least some of the calibration checkers are compared to the first calibration relationship to determine if the first calibration relationship is still valid.

25. Automated apparatus having control circuitry and being configured to perform the method of any preceding claim.-33-A

Images