METHOD OF DETERMINING WHETHER A HOST HAS AN INFECTION

Abstract

The invention relates to a method of determining whether a host has an infection. The method comprises obtaining measured values comprising a number of leukocytes and a number of a first type of leukocytes in a blood sample from the host, and comparing the measured values with one or more stored threshold value(s) to determine whether the host has said infection.

Claims

1. A method of determining whether a host has an infection, the method comprising: obtaining measured values comprising a number of leukocytes and a number of a first type of leukocytes in a blood sample from the host, and comparing the measured values with one or more stored threshold value(s) to determine whether the host has said infection.

2. A method according to claim 1, wherein the infection is infection by the virus SARS-CoV2.

3. A method according to claim 1, wherein comparing the measured values with one or more stored threshold value(s) to determine whether the host has the infection comprises performing at least one calculation on the measured values and comparing the result of the calculation(s) with a respective one of the threshold value(s).

4. A method according to claim 1, wherein the first type of leukocytes is neutrophils.

5. A method according to claim 3, wherein one of the at least one calculation(s) is λ, λ being the number of leukocytes, L, minus the number of neutrophils, N.

6. A method according to claim 1, wherein the measured values further comprise a number of eosinophils.

7. A method according to claim 3, wherein one of the at least one calculation(s) is κ, κ being the number of leukocytes, L, minus the number of eosinophils, E.

8. A method according to claim 1, wherein the measured values further comprise a number of lymphocytes.

9. A method according to claim 1, wherein comparing measured values with one or more stored threshold values(s) comprises comparing each of λ, κ, the number of leukocytes, the number of neutrophils, the number of eosinophils and the number of lymphocytes with a respective threshold.

10. A method according to claim 9, wherein the respective threshold for the calculation λ is ν (nu)=2.02 10{circumflex over ( )}9/L, wherein when the result of λ is less than ν (nu) then a result of a comparison for λ is positive, and/or wherein the respective threshold for the calculation κ is μ (mu)=7.00 10{circumflex over ( )}9/L, wherein when the result of κ is greater than μ (mu) then a result of a comparison for κ is positive, and/or wherein the respective threshold for the number of eosinophils is ξ (xi)=0.09 10{circumflex over ( )}9/L, wherein when the number of eosinophils is less than ξ (xi) then a result of a comparison for the number of eosinophils is positive, and/or wherein the respective threshold for the number of lymphocytes is o (omicron)=1.64 10{circumflex over ( )}9/L, wherein when the number of lymphocytes is less than o (omicron) then a result of a comparison for the number of lymphocytes is positive, and/or wherein the respective threshold for the number of neutrophils is ρ (ro)=5.07 10{circumflex over ( )}9/L, wherein when the number of neutrophils is greater than ρ (ro) then a result of a comparison for the number of neutrophils is positive, and/or wherein the respective threshold for the number of leukocytes is σ (sigma)=8.00 10{circumflex over ( )}9/L, wherein when the number of neutrophils is greater than σ (sigma) then a result of a comparison for the number of leukocytes is positive.

11. A method according to claim 1 and wherein the measured values further comprise a number of lymphocytes, wherein comparing measured values with one or more stored threshold values(s) comprises comparing each of λ, the number of eosinophils and the number of lymphocytes with a respective threshold.

12. A method according to claim 11, wherein the respective threshold for the calculation λ is υ (upsilon)=1.50 10{circumflex over ( )}9/L, wherein when the result of λ is less than υ (upsilon) then a result of a comparison for λ is positive, and/or wherein the respective threshold for the number of eosinophils is ϕ (phi)=0.03 10{circumflex over ( )}9/L and when the number of eosinophils is less than ϕ (phi) then a result of a comparison for number of eosinophils is positive, and/or wherein the respective threshold for the number of lymphocytes is χ (chi)=1.10 10{circumflex over ( )}9/L and when the number of lymphocytes is less than χ (chi) then a result of a comparison for the number of lymphocytes is positive.

13. A method according to claim 10, wherein when the result of at least one of the comparisons is positive, comparing the measured values with one or more stored threshold value(s) determines that the host has said infection.

14. A method of calculating a threshold value for use in determining whether a host has an infection, the method comprising: obtaining measured values comprising a number of leukocytes and a number of a first type of leukocytes in a blood sample from each of a plurality of subjects, obtaining an infection status for each of the subjects, analysing the measured values and infection statuses to determine the threshold value.

15. A method according to claim 14, wherein analysing the measured values and infection statuses to determine a threshold value comprises separating the measured values into an infected set and an uninfected set, wherein the infected set contains measured values for subjects with an infection status of infected and the uninfected set contains values measured values for subjects with an infection status of uninfected.

16. A method according to claim 15, wherein the analysis further comprises performing a calculation on the measured values of each subject and comparing results of the calculation for measured values in the infected set with results of the calculation for measured values in the uninfected set.

17. A method according to claim 16, wherein comparing results in the infected set with results in the uninfected set comprises calculating a mean of the results for each set and determining a threshold value comprises selecting a value between the mean of the results for the infected set and the mean of the results in the uninfected set.

18. A method according to claim 14, wherein the first type of leukocytes is neutrophils.

19. A method according to claim 18, when dependent on claim 16 wherein the calculation is the number of leukocytes minus the number of neutrophils L−N=λ.

20. A method according to claim 14, wherein the first type of leukocytes is eosinophils.

21. A method according to claim 20, wherein the calculation is the number of leukocytes minus the number of eosinophils L−E=κ.

22. A method of adjusting the threshold value determined using a method of claim 14, the method comprising: obtaining the threshold value and an allowed range for the threshold value and modifying the threshold value in response to a received sensitivity parameter, within the allowed range.

23. A method of determining whether a host has an infection according to claim 1, wherein the one or more stored threshold value(s) are calculated using a method of calculating a threshold value for use in determining whether a host has an infection.

24. A system configured to carry out a method according to claim 1, wherein the system comprises: a user device and a server, wherein the user device is configured to obtain the measured values and compare the measured values with a respective one of the stored threshold value(s) to determine whether the host has said infection, wherein the server is configured to store the stored threshold value(s) and wherein the user device is configured to obtain the stored threshold value(s) from the server.

Description

BRIEF DESCRIPTION OF DRAWINGS

[0077] Embodiments of the present disclosure will now be described by way of example only, with reference to the following diagrams, in which:—

[0078] FIG. 1 is a flow chart outlining a method of determining threshold values for a specific infection;

[0079] FIG. 2 is a flow chart outlining a method of using the threshold value determined in the flow chart of FIG. 1 to determine if a host has the infection.

DETAILED DESCRIPTION

[0080] In general terms embodiments of the invention relate to the use of measured values comprising a number of leukocytes and a number of a type of leukocytes in a blood sample of a host to determine whether the host has a particular infection. The examples discussed below concern an implementation of such a method to determine whether a host has an infection with the virus SARS-CoV2 which causes the illness Covid-19, however, it will be understood that the invention is not limited to this infection.

[0081] The method described below for determining threshold values for a specific infection and using those threshold values to make a diagnosis of that infection in a patient beneficially allows a diagnosis of a specific infection to be made earlier in the infection's lifecycle than the current available methods for detecting infections.

[0082] To place embodiments of the invention in a suitable context reference will now be made to FIG. 1 which shows a flow chart outlining a method of determining threshold values for a specific infection, in this case, infection by the virus SARS-CoV2. In Step 10, a full blood count test is conducted on blood samples obtained from a sample of the population. The sample contains at least some people who are likely to have been exposed to SARS-CoV2. Further subjects may be added to the sample over time as more subjects are tested, thereby increasing the size of the data set to be analysed. Increasing the sample size may allow for more accurate threshold values to be determined.

[0083] Step 11 shows that there is a period of time between blood samples being taken from the subjects and tests for the infection being performed on the subjects in Step 12. This is important as it allows time for the infection to develop to detectable levels in the subjects. As discussed above, one of the advantages of the methods described herein is that the infection can be detected before currently available infection tests. So, the waiting time signified by Step 11 allows comparison of the present method at an early stage of infection with currently available tests at a time when the infection can be detected by the available tests.

[0084] In some embodiments, more than one infection test may be performed at different times to each other, to ensure that any infection that develops in a subject is detected by an infection test. The blood samples for the CBC may be collected, for example 7 days before a first infection test and 14 days before a second infection test to allow sufficient time for the infection to become detectable by at least one of the diagnostic tests.

[0085] In Step 12, diagnostic tests for the infection being targeted are carried out on the same sample of the population as Step 10. The diagnostic infection test performed may be the current‘gold standard’ diagnostic test for the specific infection being targeted. For example, if the targeted infection is the virus SARS-CoV2 which causes the illness Covid-19, the relevant ‘gold standard’ diagnostic test may be a PCR test.

[0086] In Step 13, the full blood count test results are separated into an infected and uninfected set based on whether the diagnostic infection tests returned an infected or uninfected result for the subject.

[0087] In Step 14, the infected set is compared with the uninfected set to determine a calculation and infection threshold value. The calculation and associated threshold value is chosen to best separate the infected and uninfected sets, so that the calculation and threshold can be used later to determine whether a host has an infection from their blood test results alone, without the need for an infection test.

[0088] Calculations are performed on the full blood count test results to determine values, such as ratios and differences, between the types of white blood cell. The method analyses only a selection of the results given by the full blood count. In this example, from the full blood count results, the number of leukocytes, number of monocytes and number of eosinophils were analysed for each subject. A plurality of calculations may be performed on each of the subjects' results, and the calculation which best separates the infected and uninfected sets may be selected.

[0089] Step 13 of grouping the test results into an infected set and an uninfected set may be done before or after the calculations are performed.

[0090] In this example, two different calculations are performed using only the number of monocytes and leukocytes from the full blood count tests. The first calculation is to determine the ratio (a, alpha) between the number of monocytes and leukocytes (monocytes/leukocytes). The second calculation is to determine the difference (P, beta) between the number of leukocytes and the number of monocytes (leukocytes−monocytes).

[0091] Two further calculations are performed using the number of monocytes, leukocytes and eosinophils from the full blood count tests. The first calculation is to determine the ratio (γ, gamma) between the number of monocytes and the number of leukocytes plus the number of eosinophils (monocytes/(leukocytes+eosinophils)). The second calculation is to determine the net difference (δ, delta) between the number of leukocytes plus eosinophils and the number of monocytes (leukocytes+eosinophils−monocytes).

[0092] The above calculations all show statistically significant correlations between the calculation and the infection status (infected or uninfected). From these correlations, a threshold value for each calculation (δ, γ, β, α) can be determined.

[0093] Statistical measures may also be determined to indicate the proportion of false positives and false negatives that may be obtained from the threshold values. The threshold values may be adjusted according to how the test results will be used. For example, the specificity and sensitivity of the test can be altered to take into account the balance of health, social and economic consequences of false positives and false negatives.

[0094] A threshold value (ζ, zeta) which distinguishes between subjects who will later go on to test positive for the infection being targeted using the full set of diagnostic tests is determined for the β calculation. The threshold value (ζ, zeta) may be determined by standardizing the data for β to have a mean of 0 and a standard deviation of 1. The infection threshold ζ (zeta) in this example is −1. The threshold value (ζ, zeta) may be adjusted to alter the proportion of false positives and negatives in the full blood count test results. The threshold value (ζ, zeta) may be derived by calculating the mean and standard deviation for each of the infected set, β+ and uninfected set, β−. The threshold value (ζ, zeta) lies between the mean value of the β+set and the β− set. Adjusting the threshold value (ζ, zeta) between the β+ set and the β− set may alter the proportion of false positives and negatives when using the threshold to determine infection.

[0095] A threshold value (θ, theta) which distinguishes patients who will later go on to test positive for the specific infection being targeted using the full set of diagnostic tests may be determined for the δ calculation. The threshold value (θ, theta) may be determined by standardizing the data for δ to have a mean of 0 and a standard deviation of 1. The infection threshold (θ, theta) in this example is −1. Adjusting the threshold value (θ, theta) may alter the proportion of false positives and negatives in the full blood count test results. The threshold value (θ, theta) may be derived by calculating the mean and standard deviation for the two δ groups (patients who have gone on to test positive for the targeted infection and those who have gone on to test negative), creating two new groups δ+ and δ. The threshold value (θ, theta) lies between the mean value of the δ+ set and the δ− set. Adjusting the threshold value (θ, theta) between the δ+ set and the δ− set may alter the proportion of false positives and negatives when using the threshold to determine infection.

[0096] A threshold value (ε, epsilon) which distinguishes patients who will later go on to test positive for the specific infection being targeted using the full set of diagnostic tests may be determined for the α calculations. The threshold value (ε, epsilon) may be determined by standardizing the data for α to have a mean of 0 and a standard deviation of 1. Adjusting the threshold value (ε, epsilon) may alter the proportion of false positives and negatives in the full blood count test results. The threshold value (ε, epsilon) may be derived by calculating the mean and standard deviation for the two α groups (patients who have gone on to test positive for the targeted infection and those who have gone on to test negative), creating two new groups α+ and α−. The threshold value (ε, epsilon) lies between the mean value of the α+ set and the α− set. Adjusting the threshold value (ε, epsilon) between the α+ set and the α− set may alter the proportion of false positives and negatives when using the threshold to determine infection.

[0097] A threshold value (η, eta) which distinguishes patients who will later go on to test positive for the specific infection being targeted using the full set of diagnostic tests may be determined for the γ calculations. The threshold value (η, eta) may be determined by standardizing the data for α to have a mean of 0 and a standard deviation of 1. Adjusting the threshold value (η, eta) may alter the proportion of false positives and negatives in the full blood count test results. The threshold value (η, eta) may be derived by calculating the mean and standard deviation for the two α groups (patients who have gone on to test positive for the targeted infection and those who have gone on to test negative), creating two new groups γ+ and γ−. The threshold value (η, eta) lies between the mean value of the γ+ set and the γ− set. Adjusting the threshold value (η, eta) between the means of the γ+ set and the γ− set may alter the proportion of false positives and negatives when using the threshold to determine infection.

[0098] FIG. 2 shows a method of determining if a host has an infection using the determined threshold(s) along with results of a CBC test on a blood sample from the host.

[0099] In Step 18, a full blood count test is performed on a blood sample obtained from a host (in this case, a patient who may be infected with SARS-CoV2). Measured values including number of leukocytes, number of monocytes and number of eosinophils are extracted from the CBC test results. In other embodiments, another type of blood test including these measured values may be used in place of a CBC test.

[0100] The four calculations (δ, γ, β, α) are made on the measured values and compared to the threshold values θ, η, ζ, ε. In other embodiments, less than four, for example only one calculation may be made and compared with the calculation's associated threshold.

[0101] In Step 20, a determination of diagnosis is made based on the comparison of the result of the calculation on the patient's measured values with the associated threshold value. If the result of the calculation on the patient's measured values is on the infected sied of the threshold, then the patient is determined to be infected, but if the result of the calculation on the patient's measured values is on the uninfected side of the threshold, then the patient is determined to be uninfected. For example, where the mean of the uninfected set is greater than the threshold, the patient will be determined to be uninfected if the result of the calculation on the patient's measured values is also greater than the threshold.

[0102] A second example will now be described and is the same as the example described above in relation to FIGS. 1 and 2 with the changes described below.

[0103] In the second example, different measured values are taken from the full blood count results and different set of calculations are performed. At step 14, from the full blood count results, the number of leukocytes, number of neutrophils, number of eosinophils and number of lymphocytes were analysed for each subject. Two calculations were performed on each of the subjects' results. The first calculation is the number of leukocytes minus the number of eosinophils (κ=L−E) and the second is the number of leukocytes minus the number of neutrophils (λ=L−N).

[0104] Threshold values which distinguish patients who will later go on to test positive for the specific infection being targeted using the full set of diagnostic tests is determined for each calculation and for each of the number of leukocytes, number of eosinophils, number of neutrophils and number of lymphocytes. The threshold values may be determined by standardizing the data for each calculation to have a mean of 0 and a standard deviation of 1. The threshold values may be derived by calculating the mean and standard deviation for the two groups (patients who have gone on to test positive for the targeted infection and those who have gone on to test negative), creating two new groups of infected and uninfected results. The threshold value lies between the mean value of the infected set and the uninfected set. Adjusting the threshold values between the means of the infected set and the uninfected set may alter the proportion of false positives and negatives when using the threshold to determine infection.

[0105] In Step 18, a full blood count test is performed on a blood sample obtained from a host (in this case, a patient who may be infected with SARS-CoV2). Measured values including number of leukocytes, number of neutrophils, number of lymphocytes and number of eosinophils are extracted from the CBC test results. In other embodiments, another type of blood test including these measured values may be used in place of a CBC test.

[0106] The two calculations (κ=L−E and λ=L−N) are made on the measured values and the calculation results and the measured values are compared to the threshold values.

[0107] The infection threshold μ (mu) for the κ calculation is 7.00 10{circumflex over ( )}9/L. If the value is greater than μ (mu) then the result of the comparison is positive. The infection threshold ν (nu) for the λ calculation is 2.02 10{circumflex over ( )}9/L. If the value is less than that then the result of the comparison is positive. The infection threshold ξ (xi) for number of eosinophils is 0.09 10{circumflex over ( )}9/L. If the value is less than that then the result of the comparison is positive. The infection threshold o (omicron) for number of lymphocytes is 1.64 10{circumflex over ( )}9/L. If the value is less than that then the result of the comparison is positive. The infection threshold ρ (ro) for number of neutrophils is 5.07 10{circumflex over ( )}9/L. If the value is greater than that then the result of the comparison is positive. The infection threshold σ (sigma) for number of leukocytes is 8.00 10{circumflex over ( )}9/L. If the value is greater than that then the result of the comparison is positive.

[0108] When one or more of the comparisons give a positive result, the determination that the host is infected is made. Using the thresholds μ, ν, ξ, o, ρ and σ, the sensitivity is 97.29% and the specificity may be 67.95%.

[0109] In Step 20, a determination of diagnosis is made based on the comparisons of the calculation results and measured values with the respective threshold values. If any of the comparison results determine that the patient is infected, then the patient may be determined to be infected.

[0110] A third example will now be described and is the same as the example described above in relation to FIGS. 1 and 2 with the changes described below.

[0111] In the third example, different measured values are taken from the full blood count results and different set of calculations are performed. At step 14, from the full blood count results, the number of leukocytes, number of neutrophils, number of eosinophils and number of lymphocytes were analysed for each subject. One calculation was performed on each of the subjects' results. The calculation is the number of leukocytes minus the number of neutrophils (λ=L−N).

[0112] Threshold values which distinguish patients who will later go on to test positive for the specific infection being targeted using the full set of diagnostic tests is determined for the calculation and for the number of eosinophils and number of lymphocytes. The threshold values may be derived by calculating the mean and standard deviation for the two groups (patients who have gone on to test positive for the targeted infection and those who have gone on to test negative), creating two new groups of infected and uninfected results. The threshold value lies between the mean value of the infected set and the uninfected set. Adjusting the threshold values between the means of the infected set and the uninfected set may alter the proportion of false positives and negatives when using the threshold to determine infection.

[0113] Alternatively, the threshold values may be determined by standardizing the data for each calculation to have a mean of 0 and a standard deviation of 1.

[0114] In Step 18, a full blood count test is performed on a blood sample obtained from a host (in this case, a patient who may be infected with SARS-CoV2). Measured values including number of leukocytes, number of neutrophils, number of lymphocytes and number of eosinophils are extracted from the CBC test results. In other embodiments, another type of blood test including these measured values may be used in place of a CBC test.

[0115] The calculation (λ=L−N) is made on the measured values and the calculation results and the measured values are compared to the threshold values.

[0116] The infection threshold υ (upsilon) for the λ calculation is 1.50 10{circumflex over ( )}9/L. If the value is less than that then the result of the comparison is positive. The infection threshold φ (phi) for number of eosinophils is 0.03 10{circumflex over ( )}9/L. If the value is less than that then the result of the comparison is positive. The infection threshold χ (chi) for number of lymphocytes is 1.10 10{circumflex over ( )}9/L. If the value is less than that then the result of the comparison is positive.

[0117] When one or more of the comparisons give a positive result, the determination that the host is infected may be made. When the thresholds υ, φ and χ are used, the sensitivity of the determination of whether the host is infected is 84.90% and the specificity is 96.40%.

[0118] In Step 20, a determination of diagnosis is made based on the comparisons of the calculation results and measured values with the respective threshold values. If any of the comparison results determine that the patient is infected, then the patient may be determined to be infected.

[0119] Although particular embodiments of the disclosure have been disclosed herein in detail, this has been done by way of example and for the purposes of illustration only. The aforementioned embodiments are not intended to be limiting with respect to the scope of the statements of invention and appended claims.

[0120] It is contemplated by the inventors that various substitutions, alterations, and modifications may be made to the invention without departing from the scope of the invention as defined by the statements of invention and claims.