IN-VITRO METHOD FOR DETERMINING A CELL TYPE OF A WHITE BLOOD CELL WITHOUT LABELING

Abstract

The invention relates to an in-vitro method for determining a cell type of a white blood cell in a biological sample without labeling, wherein a microscopy apparatus images the cell and physical parameters of the cell are ascertained from the image of the cell by means of an automated image analysis, wherein the cell type of the white blood cell is determined on the basis of the physical parameters and on the basis of principal component analysis parameters (PCA parameters), wherein the principal component analysis parameters comprise linear combinations of at least some of the physical parameters.

Claims

1. An in-vitro method for determining a cell type of a white blood cell in a biological sample without labeling, the method comprising: imaging the white blood cell using microscopy apparatus; ascertaining physical parameters of the white blood cell from an image of the white blood cell by an automated image analysis; and determining the cell type of the white blood cell based on the physical parameters and principal component analysis parameters, wherein the principal component analysis parameters comprise linear combinations of at least some of the physical parameters.

2. The method as claimed in claim 1, wherein a determined cell type of the white blood cell is a principle type of white blood cell selected from the group consisting of monocytes, neutrophils, basophils, eosinophils, and lymphocytes.

3. The method as claimed in claim 2, wherein the determined cell type is one of the principal types of white blood cells and a sub-type of white blood cells selected from the group consisting of myelocytes, metamyelocytes, promyelocytes, blasts, megakaryocytes, plasma cells, atypical lymphocytes, and Sézary cells.

4. The method as claimed in claim 1, wherein determination of a respective cell type of a multiplicity of white blood cells in a sample is used to characterize respective cell populations, and present cell populations are used to determine whether acute myeloid leukemia (AML), acute lymphoblastic leukemia (ALL), chronic myeloid leukemia (CML) or chronic lymphocytic leukemia (CLL) is present a patient from whom the biological sample originated.

5. The method as claimed in claim 1, wherein the physical parameters of the cell comprise parameters selected from the group comprising area covered by the cell, perimeter of the cell, width of the cell, height of the cell, ratio of the width to the height of the cell, similarity of the geometric shape of the cell to a circle, mean radius of the cell, variance of the radius of the cell, degree of coverage of the cell, equivalent diameter corresponding to the covered area of the cell, optical volume of the cell, maximum optical height of the cell, minimum optical height of the cell, mean optical height of the cell, variance of the optical height of the cell, biconcavity of the cell, sphericity of the cell, shift of the center of mass of the cell, contrast of the cell, dissimilarity of the cell, homogeneity of the cell, energy of the cell, and entropy of the cell.

6. The method as claimed in claim 1, wherein imaging the white blood cell comprises overlaying a reference wave on an object wave, and recording a resultant interferogram or computer-implemented mathematical reconstruction.

7. The method as claimed in claim 1, wherein the microscopy apparatus is a microscope for performing digital holographic microscopy (DHM), interference phase microscopy, or quantitative phase microscopy.

8. The method as claimed in claim 1, wherein the white blood cells are imaged by the microscopy apparatus in a flow cell.

9. The method as claimed in claim 8, wherein the white blood cells are focused by laminar sheath flows.

10. The method as claimed in claim 1, further comprising removing red blood cells from the biological sample using selective lysis before the imaging of the white blood cell using the microscopy apparatus.

11. The method as claimed in claim 1, wherein the method is carried out using a non-stained or non-desiccated white blood cell.

12. The method as claimed in claim 1, further comprising: phenotyping the white blood cell with a label; or expressing a predetermined receptor for further assigning of the white blood cell.

13. The method as claimed in claim 1, wherein the method is performed using a whole blood sample or the biological sample contains blood cells, monocytes, neutrophils, basophils, eosinophils or lymphocytes.

14. A cell analysis device configured to carry out a method for determining a cell type of a white blood cell in a biological sample without labeling, the cell analysis device comprising a microcontroller configured to: cause microscopy apparatus to image the white blood cell; ascertain physical parameters of the white blood cell from an image of the white blood cell; and determine the cell type of the white blood cell based on the physical parameters and principal component analysis parameters, wherein the principal component analysis parameters comprise linear combinations of at least some of the physical parameters.

15. A microscopy apparatus for determining a cell type of a cell of a biological sample without labeling, comprising a cell analysis device as claimed in claim 14 and at least one of a light microscope, a scanning electron microscope, a phase contrast microscope, a digital holographic microscope, or a microscope with an ultrasonic sensor.

16. The method as claimed in claim 8, wherein the flow cell comprises a channel with a rectangular or square cross-section.

17. The method as claimed in claim 9, wherein the white blood cells are focused by four laminar sheath flows.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0057] The invention is once again explained in more detail by specific exemplary embodiments on the basis of the attached drawings. The shown examples represent preferred embodiments of the invention. In detail:

[0058] FIG. 1 shows an overview for an AML sample,

[0059] FIG. 2 shows an overview for a CML sample,

[0060] FIG. 3 shows an overview for an ALL sample,

[0061] FIG. 4 shows an overview for a CLL sample,

[0062] FIG. 5 shows an overview for an OMF sample,

[0063] FIG. 6 shows an overview for a healthy sample, with PCA parameters being plotted in each case.

[0064] Further:

[0065] FIG. 7 shows characteristic scattering patterns for myeloblast and lymphoblast gates of an AML sample,

[0066] FIG. 8 shows characteristic scattering patterns for myeloblast and lymphoblast gates of a CML sample,

[0067] FIG. 9 shows characteristic scattering patterns for myeloblast and lymphoblast gates of an ALL sample,

[0068] FIG. 10 shows characteristic scattering patterns for myeloblast and lymphoblast gates of a CLL sample,

[0069] FIG. 11 shows characteristic scattering patterns for myeloblast and lymphoblast gates of an OMF sample,

[0070] FIG. 12 shows characteristic scattering patterns for myeloblast and lymphoblast gates of a healthy sample,

[0071] Further:

[0072] FIG. 13 shows distinguishing monocytes,

[0073] FIG. 14 shows distinguishing neutrophils and eosinophils,

[0074] FIG. 15 shows distinguishing basophils and lymphocytes,

[0075] FIG. 16 shows distinguishing eosinophils, promyelocytes, blasts, and megakaryocytes,

[0076] FIG. 17 shows distinguishing metamyelocytes

[0077] on the basis of characteristic scattering patterns in defined regions (“gates”). PCA parameters (Val-X) or a combination of physical and PCA parameters are plotted in each case.

DETAILED DESCRIPTION OF THE INVENTION

[0078] With the aid of the in-vitro method according to the invention, there can be a determination of a cell type of a white blood cell in a biological sample without labeling. In this case, without labeling means that the cells need not be labeled, for example, by fluorescent dyes or radioactive particles. Here, a biological sample may comprise a sample of animal or human blood cells, for example. Preferably, this is a whole blood sample, which, for example, comprises blood cells 10, e.g., leukocytes, eosinophils, or basophils.

[0079] In a first experimental example, AML, CML, ALL, CLL and OMF are distinguished on the basis of the respective characteristic scattering patterns.

[0080] FIG. 1 to FIG. 6 show characteristic scattering patterns of AML (FIG. 1), CML (FIG. 2), AL (FIG. 3), CLL (FIG. 4), OMF (FIG. 5), and a healthy sample (FIG. 6). Here, “OMF” represents osteomyelofibrosis, a sub-type of the group of myeloproliferative neoplasms (MPNs). The PCA parameters Val4 and Val5 (see above for the corresponding linear combination) are used for the distinction.

[0081] The distinguishing criteria for the samples of patients with one of the different types of leukemia or of healthy subjects are as follows:

[0082] Sample of a healthy subject: three clearly separates data regions: [0083] 1.sup.st region: Val4=400-650, Val5=200-500, [0084] 2.sup.nd region: Val4=600-750, Val5=400-600, [0085] 3.sup.rd region: Val4=700-850, Val5=250-400.

[0086] Sample of a patient with AML: only one continuous data region: [0087] Val4=600-900, Val5=300-600.

[0088] Sample of a patient with CML: contiguous data region which can be subdivided into 3 regions: [0089] 1.sup.st region: small population in the region of Val4=400-550, Val5=200-500, [0090] 2.sup.nd region: dominant main population in the region of Val4=500-700, Val5=200-500, [0091] 3.sup.rd region: small population in the region of Val4=700-850, Val5=250-400.

[0092] Sample of a patient with ALL: large main population with 2 small secondary populations: [0093] 1.sup.st region: dominant main population in the region of Val4=700-850, Val5=300-500, [0094] 2.sup.nd region: small population in the region of Val4=550-700, Val5=300-450, [0095] 3.sup.rd region: small population in the region of Val4=500-700, Val5=450-600.

[0096] Sample of a patient with CLL: large main population with a small secondary population: [0097] 1.sup.st region: dominant main population in the region of Val4=650-900, Val5=300-500, [0098] 2.sup.nd region: small population in the region of Val4=400-600, Val5=300-500.

[0099] Sample of a patient with OMF: contiguous data region which can be subdivided into 4 regions: [0100] 1.sup.st region: Val4=700-900, Val5=200-400, [0101] 2.sup.nd region: Val4=600-750, Val5=400-550, [0102] 3.sup.rd region: Val4=500-650, Val5=250-450, [0103] 4.sup.th region: small population in the region of Val4=400-650, Val5=450-700.

[0104] In a second experimental example, AML, CML, ALL, CLL and OMF are distinguished on the basis of the respective characteristic scattering patterns.

[0105] FIGS. 7 to 12 show characteristic scattering patterns of AML (FIG. 7), CML (FIG. 8), ALL (FIG. 9), CLL (FIG. 10), OMF (FIG. 11), and a healthy sample (FIG. 12). The PCA parameters Val3 and Val4 (see above for the corresponding linear combination) are used for the distinction. Two large regions are defined: [0106] 1.sup.st region: Val4=235-920, Val5=576-816 (=“myeloblasts” gate (2)), [0107] 2.sup.nd region: Val4=235-920, Val5=816-1015 (=“lymphoblasts” gate (1)).

[0108] The distinguishing criteria for the samples of patients with one of the different types of leukemia or of healthy subjects are as follows:

[0109] Sample of healthy subject: zero or only a very small number of data points in both gates.

[0110] Sample of a patient with AML: a significant population in the 1.sup.st region: [0111] >90% of all data points in the 1.sup.st region, [0112] <10% of all data points in the 2.sup.nd region.

[0113] Sample of a patient with CML: a small number of data points in the 1.sup.st region: [0114] >90% of all data points in the 1.sup.st region, [0115] <10% of all data points in the 2.sup.nd region.

[0116] Sample of a patient with ALL: a significant population extends over the 1.sup.st and 2.sup.nd region: [0117] <90% of all data points in the 1.sup.st region, [0118] >10% of all data points in the 2.sup.nd region.

[0119] Sample of a patient with CLL: a significant population in the 2.sup.nd region: [0120] <10% of all data points in the 1.sup.st region, [0121] >90% of all data points in the 2.sup.nd region.

[0122] Sample of a patient with OMF: a small number of data points in the 1.sup.st and 2.sup.nd region: [0123] >80% of all data points in the 1.sup.st region, [0124] <20% of all data points in the 2.sup.nd region.

[0125] In a third experimental example, there is the distinction between the 5 principal types of white blood cells and further sub-types without labeling.

[0126] FIGS. 13 to 17 show the distinction between the 5 principal types of white blood cells and further sub-types without labeling, on the basis of characteristic scattering patterns in defined regions (“gates”). The abbreviation “E & N & IG” (12) represents eosinophils, neutrophils, and immature granulocytes (see FIG. 13). In this case, the immature granulocytes are metamyelocytes and myelocytes. The following regions are defined for distinguishing between the cell types (specified in each case are the use parameters and the X/Y-coordinates of the defined regions (gates)):

[0127] 1.sup.st region=“monocytes” gate (3): The monocyte cell type is distinguished as follows on account of the values of Val4 and Val5.

TABLE-US-00002 Val4 Val5 606 483 636 522 664 538 697 527 727 496 746 435 650 361 622 438

[0128] 2.sup.nd region=“neutros” gate (4): The neutrophil cell type is distinguished as follows on account of the values of Val4 and Val5.

TABLE-US-00003 Val4 Val5 501 620 494 486 477 487 456 486 416 487 321 487 322 564 323 607 322 659 322 703 480 696

[0129] 3.sup.rd region=“eos” gate (5): The eosinophil cell type is distinguished as follows on account of the values of Val4 and entropy.

TABLE-US-00004 Val4 Entropy 320 703 406 831 564 833 580 831 654 828 739 827 838 826 838 671 840 488 724 488 625 486 494 487 501 627 479 697

[0130] 4.sup.th region=“basos” gate (6): The basophil cell type is distinguished as follows on account of the values of Val4 and Val5.

TABLE-US-00005 Val4 Val5 732 422 717 347 699 222 643 254 645 386 683 415

[0131] 5.sup.th region=“lymphos” gate (7): The lymphocyte cell type is distinguished as follows on account of the values of Val4 and Val5.

TABLE-US-00006 Val4 Val5 781 438 826 436 936 426 937 223 861 223 801 218 698 212 717 347 731 424

[0132] 6.sup.th region=“megakaryocytes” gate (8): The megakaryocyte cell type is distinguished as follows on account of the values of Val3 and entropy.

TABLE-US-00007 Val3 Entropy 33 233 332 805

[0133] 7.sup.th region=“promyelocytes” gate (9): The promyelocyte cell type is distinguished as follows on account of the values of Val3 and entropy.

TABLE-US-00008 Val3 Entropy 233 846 642 805

[0134] 8.sup.th region=“myelocytes” gate (10): The myelocyte cell type is distinguished as follows on account of the values of equivalentDiameter and homogeneity.

TABLE-US-00009 EquivalentDiameter Homogeneity 333 534 304 534 304 493 304 436 304 272 367 273 429 274 427 408 427 470 418 524 400 532 377 534 356 534

[0135] 9.sup.th region=“metamyelocytes” gate (11): The metamyelocyte cell type is distinguished as follows on account of the values of equivalentDiameter and homogeneity.

TABLE-US-00010 EquivalentDiameter Homogeneity 304 534 301 907 680 900 684 273 428 273 426 372 426 468 420 524 402 532 379 534