1. A method for the examination of a sample (11), the method comprising the steps of:

a) generation of an image (I) of the sample (11);

b) analyzing the image with respect to at least one sample parameter;

c) selection of a region of interest in said image (I), said region being called "image-ROI" (R)_I）；

d) From the sample (11) corresponding to the image-ROI (R)_I) And is called "sample-ROI" (R)_S) Isolation of the region of interest;

e) performing a molecular assay using the sample-ROI;

f) linking assay data to the sample parameters for the corresponding sample region.

2. An apparatus (1000) for examination of a sample (11), comprising:

a) an image generation unit (200) for generating an image (I) of the sample (11);

b) an image analyser for analysing the image with respect to at least one sample parameter;

c) an image selection unit (300) for selecting a region of interest in the image (I), said region being called "image-ROI" (R)_I）；

d) For isolating a region from the sample (11) corresponding to the image-ROI (R)_I) And is called "sample-ROI" (R)_S) A sample isolation unit (400) of the region of interest;

e) for performing the utilization of the sample-ROI (R)_S) A molecular test unit (500) for measuring the molecule of (1);

f) an evaluation unit (301, 302, 303, 304) for linking assay data to the sample parameter of a corresponding sample region.

3. The method of claim 1, wherein the first and second light sources are selected from the group consisting of,

characterized in that the sample (11) is stained before the generation of the image (I) and/or after the performance of the molecular assay.

4. The method of claim 1, wherein the first and second light sources are selected from the group consisting of,

characterized in that the sample (11) is covered by a coverslip during the generation of the image (I).

5. The device (1000) according to claim 2,

characterized in that the image selection unit (300) comprises an automatic image analysis module and/or a user interface (301, 304).

6. The method according to claim 1 or the apparatus (1000) according to claim 2,

characterized in that the sample parameter is indicative of a local amount of a specific cell type or tissue type.

7. The method according to claim 1 or the apparatus (1000) according to claim 2,

characterized in that the sample parameter is based at least in part on a staining assay performed with the sample.

8. The method according to claim 1 or the apparatus (1000) according to claim 2,

characterized in that said image-ROI (R)_I) Is based at least in part on the sample parameter.

9. The method according to claim 1 or the apparatus (1000) according to claim 2,

characterised by passing the possible sample-ROI (R)_S) Set requirements to adjust the image-ROI (R)_I) The shape and/or size of (a).

10. The device (1000) according to claim 2,

characterized in that the sample isolation unit (400) comprises a laser microdissection device and/or a printing device.

11. The method according to claim 1 or the apparatus (1000) according to claim 2,

characterized in that the sample (11) is arranged on a carrier (10) comprising at least one marker (M).

12. The method according to claim 1 or the apparatus (1000) according to claim 2,

characterized in that the performed assay comprises a PCR step, a sequencing step and/or microarray hybridization or another molecular diagnostic technique.

13. The method of claim 1, wherein the first and second light sources are selected from the group consisting of,

characterized in that said image selection comprises performing a manual rough selection of a region in said image and performing an algorithm to automatically adjust refine said region, thereby obtaining a more precise definition of said image-ROI.

14. One method for isolating a so-called "sample-ROI" (R) from a sample (11)_S) In particular for the device (1000) according to claim 2,

characterized in that the sample isolation unit (400) comprises:

-a light source (401) for generating a light beam (L);

-for directing said light beam (L) at said sample-ROI（R_S) Out of position and/or in the sample-ROI (R)_S) The position within is such that the sample therein is altered to become a light directing system (402) that is separable from the rest.

15. Use of the device (1000) according to claim 2 or the sample isolation unit (400) according to claim 14 for the preparation of a kit for molecular pathology, in particular for oncology applications.

Background

US6091842 discloses an analyzer in which an image of a biological specimen is generated and automatically analyzed for a region containing cytological material. These regions are then automatically presented to the human operator in a temporally optimized sequence.

US6159681 discloses a method for performing regional analysis of biological material. The method employs a photoresist layer built on the biomaterial. The region of interest is selected and illuminated to expose a particular region of the biological material. The exposed biological material can then be selectively analyzed using any of a variety of analytical methods.

The development of argon Laser chromosome microdissection counts in combination with Polymerase Chain Reaction (PCR) protocols for the direct amplification of microdissected chromosomes is disclosed in Hadano S et al, "Laser microdissection and single unique primer PCR aggregate generation of a regional chromosomal DNA clones from a single human chromosome," Genomics, Academic Press, san Diego, USA, Vol.11, No. 2, 1991, p.10 and 373.

US2011/177518 discloses an apparatus and a method for microscopic isolation of biological cell material. Microscopic isolation apparatus and methods utilize a photomask that protects a region of interest from DNA damaging illumination.

WO01/33190 discloses a system and method for automated laser capture microdissection providing high throughput microdissection by using cell acquisition and pre-selected multiple imaging tools for cells of interest.

US6194157 discloses a method for separating biological substances by using a photoresist. After fixing the biological sample using the photoresist, in order to obtain a specific biological substance, such as a certain cell or biopolymer, from the biological sample embedded in the photoresist, the properties of the photoresist are changed by exposing the portion of the photoresist covering the biological substance to light L having an appropriate wavelength. The biological species embedded in the altered photoresist are collected.

US2010/081923 discloses systems, methods, and other modalities for generating or otherwise processing images or other data indicative of (a) extraction of chemically treated tissue frozen in a living organism, (b) treatment of a tissue sample in a cavity extending into the tissue of an organ, and/or (c) cells within a living organism to which an optical enhancement material has been applied.

Disclosure of Invention

It is an object of the present invention to provide a means for more versatile examination of a sample. It is desirable that the inspection has a high accuracy and/or efficiency.

According to a first aspect thereof, the present invention solves the above-mentioned concerns by a method for the examination of a sample, in particular a sample of biological origin (such as a tissue mass) that should be examined for the presence of e.g. tumor cells. The method comprises at least one of the following steps or a combination thereof, wherein a step or a sequence of steps may optionally be repeated one or more times:

a) generation of an image of the sample.

b) The analysis of the image with respect to at least one given parameter, which can be determined from the image and which will be referred to below for reference purposes as "sample parameter". Typically, the sample parameters will be determined in a spatially resolved manner, i.e. from image regions or image pixels. Reference to "sample parameter" will hereinafter refer to reference to at least one sample parameter, and preferably all sample parameters if more than one type of sample parameter has been determined.

c) Selection of a region of interest (ROI) in the mentioned image, wherein said region of interest will be referred to as "image-ROI" in the following.

d) Isolation of the region from the sample corresponding to the image-ROI described above, wherein this region of the sample will be referred to as "sample-ROI" hereinafter. By "isolated" is meant that the sample material belonging to the sample-ROI is physically separated from the rest of the sample.

e) Performing at least one molecular assay using the sample-ROI.

f) Link of the assay data to the sample parameters of the corresponding sample region.

The above-mentioned image-ROI may be a single connected slice of the image, or it may comprise a plurality of disconnected slices. The decision whether a point of the image belongs to the image-ROI depends on the intended examination of the sample. In oncology applications, the image-ROI may, for example, contain those image portions that are suspected of showing tumor cells.

The correspondence between image-ROI and sample-ROI is generally such that

a) Each point of the image belonging to the image-ROI shows the position of the sample belonging to the sample-ROI and/or

b) Each point of the sample belonging to the sample-ROI is represented by a point of the image belonging to the image-ROI.

Typically, the two relations a) and b) are valid simultaneously (bijective function), but it may also be encompassed by the invention of only a) or only b) valid (bijective function).

Furthermore, the term "molecular assay" should be understood in a broad sense, including any examination, test or experiment by which one or more parameters of a sample-ROI may be determined that depend on its chemical composition. As an example, a molecular assay may include the (qualitative or quantitative) detection of a particular protein or nucleic acid sequence (e.g., a tumor marker).

The linking in step f) means that the assay data determined in the molecular assay(s) for the sample-ROI correlate with the values of the sample parameters determined for the corresponding sample region (i.e. image-ROI). This may be done, for example, by storing them together at a given location in memory, a database, or a table, or by referencing them with a pointer. Due to this link, the assay data may then be evaluated automatically and/or by the user considering sample parameters belonging to the same part of the sample (or vice versa).

According to a second aspect, the invention relates to a device for the examination of a sample, said device comprising at least one of the following components or a combination thereof:

a) an image generation unit for generating an image of the sample.

b) An image analyzer for analyzing the image with respect to at least one sample parameter.

c) An image selection unit for selecting a region of interest (ROI) in the above-mentioned image, said region being referred to as "image-ROI" in the following.

d) A sample isolation unit for isolating a region of interest from the sample corresponding to the above-mentioned image-ROI in the image and hereinafter referred to as "sample-ROI".

e) A molecular examination unit for performing a molecular assay using the sample-ROI.

f) An evaluation unit for linking the assay data to the sample parameters of the corresponding sample area. The functions of the image analyser, the image selection unit and/or the evaluation unit may optionally be provided by the same device, for example by a general purpose computer.

The method and apparatus according to the first and second aspects of the invention comprise the following same innovative concepts: first a region of interest in the sample is selected from an image of the sample and then actually extracted from the physical sample and subjected to molecular assays, wherein corresponding image data and assay data are linked to each other. In particular, the method may be performed with the described apparatus. The explanations and definitions provided for the methods are therefore valid also for the apparatus and vice versa.

The method and the device have the following advantages: they allow a more informative examination of the sample, since the specimen can be subjected to both visual inspection and molecular diagnostics, wherein the results are linked to each other in a spatially resolved manner. Furthermore, a high accuracy of the molecular assay can be achieved, because the inclusion of non-relevant (interfering) sample material is minimized by the targeted definition of the sample-ROI and because the visual inspection and the molecular assay involve exactly the same material (and for example do not involve different thin sections of the sample). A further advantage is that many or even all process steps can be automated, thus achieving a high efficiency of the overall inspection process.

Analysis of sample materials for molecular diagnostic ("MDx") analysis may yield parameters that are critical to the quality of the MDx analysis, or provide additional information that enables better resolution of MDx results. For example, when the selection is based on individual cells, a secure identification of those cells is obtained. This can be, for example, a tumor cell, which is characterized by the following parameters: such as a certain ratio between nucleus and cytoplasm, a certain shape irregularity of the nucleus, a specific immunostaining of proteins in the cell membrane (e.g. HER 2) or cytoplasm (cytokeratins), nuclear receptors (e.g. ER), or genomic parameters such as the number of copies of certain genes (e.g. HER 2) or the number of copies of mRNA, or combinations thereof. Certain complementary staining agents may be used to exclude false calls (false calls) by identifying cells with a tumor-like appearance using immunological staining agents specific for epithelial cells, stromal cells, or immune cells, etc.

Another example is the use of image parameters (sample parameters) for the interpretation of MDx results. MDx tests performed using methods such as q-PCR, microarray, sanger sequencing and next generation sequencing have certain sensitivity and specificity limitations. Furthermore, when looking at gene expression patterns, the specific profile will depend on the cell type being analyzed. Using information from the analysis of the image and/or image-ROI may enable more accurate interpretation of the expression profile. Expression can be scaled, for example, to fractions of tumor cells and corrected for contribution from nominal expression profiles of other cell types (e.g., non-malignant). Sequencing data can be corrected for reference genomic contributions originating from non-tumor cells, e.g., according to the hierarchical presence in the ROI. In some cases, when the results of the MDx test are ambiguous, the role of sample selection may be considered and another, more stringent, selection may be suggested. The sample may be revisited to identify a replacement ROI based on the conclusion from the first MDx test.

Hereinafter, preferred embodiments of the present invention that can be implemented using both the method and the apparatus will be described.

The sample to be examined may in particular be a thin section of body tissue. Furthermore, the sample may be stained prior to generating the image in order to (better) visualize a particular feature of interest. Accordingly, the method of the invention may optionally comprise as a first step the generation of a thin section of body tissue and/or the staining of the sample. In the device of the invention, a sample preparation unit may optionally be included, in which these steps may be performed.

Staining may additionally or alternatively be done at other times during the entire examination of the sample. In some embodiments of the invention, subsequent staining assays on the sample sections may be selected, for example, using the results of the MDx test. The staining may optionally be performed on the remaining sample sections that have been analyzed and partially removed. In this case, previous results (e.g., on an individual cell basis) may be included, with new results obtained from subsequent staining. As an example, MDx assays can provide insight into gene mutations and/or the activity of certain signaling pathways in tumor cells. Specific staining for the presence and/or activation of proteins that play a role in this pathway can provide important information about the relationship between genetic information and tumor development and/or susceptibility to certain therapeutic regimens.

The generated image of the sample is preferably a microscopic image, i.e. it reveals details that are not visible to the naked eye. Additionally or alternatively, it is preferably a digital image, thus allowing the application of a general digital image processing flow. Furthermore, the image may be generated by scanning, i.e. by sequential generation of sub-images of a smaller portion of the sample. The device may accordingly comprise a digital microscope, in particular a digital scanning microscope, to allow for embodiments of the above-described features. Furthermore, the generated microscopic image may be a bright field or a fluorescent image, or a combination of different images.

The sample is preferably covered by a coverslip during the generation of the image. This is desirable, for example, when using a digital microscope (particularly a whole slide scanner). Image analysis from an entire slide scanner requires high image quality, as the pathologist needs to be able to extract all the information he/she would otherwise have gained from manipulation with the microscope. High image quality requires that the sample be embedded covered (standard procedure in pathology laboratories, known as coverslipping). Commercial instruments are available for this step. Because with digital pathology, digital files can be stored and archived instead of stained slides, slides can be used for sample retrieval for MDx. This has the advantage of a 100% match between the tissue and cells analyzed and the selected material for MDx, while otherwise projection and interpolation are required to correlate the image with the next slice from the sample (paraffin block) from which the ROI for MDx was dissected.

However, having the possibility to obtain samples from the same fragments (coupe, from which the digital file has been recorded) requires additional sample preparation steps. For removal of the ROI, the coverslip is an obstacle. Therefore, it is preferably removed prior to isolation of the sample-ROI. The slide can be physically aligned with the same reference (marker) used for digital scanning. No new images are required to make the physical sample selection. The virtual image may be used to guide a laser beam or mechanical or other device that acquires the sample-ROI and removes it from the rest of the sample. Alternatively, a low quality image may be obtained on the selection device in this state and mapped to a high quality stored image to aid alignment and selection.

Once the coverslip has been removed, additional staining may be performed on the same slide before and/or after removal of the sample-ROI for the reasons described above. After the cover slip, a high quality image can be obtained, which is analyzed and resolved along with the results of all previous tests (the most important MDx tests).

The selection of the image-ROI may be done automatically by a suitable image processing routine, by manual input by the user, or by a mixture of both. Accordingly, the apparatus may preferably comprise an image analysis module, such as a digital microprocessor with associated software for analysis of the digital image. Additionally or alternatively, it may comprise a user interface comprising an input device through which a user may enter data relating to the selection of the image-ROI. Typically, the user interface will also include an output device, such as a display (monitor), on which an image of the sample can be displayed, optionally together with a representation of the currently defined image-ROI. The output means may preferably allow the representation of the sample image with an adjustable zoom factor.

The image analyser will typically be a digital data processing unit with suitable image processing software by which sample parameters can be automatically determined.

The sample parameter may generally be any type of parameter that can be determined from an image of the sample, such as a local concentration of a given chemical substance (e.g., revealed via the color of the substance). In a preferred embodiment, the sample parameter is indicative of a local amount of a particular cell type or tissue type. The sample reference may for example express the absolute or relative number of tumor cells in a given region. In particular, it may be the number and/or fraction of tumor cells in the image-ROI. Knowing this number for the image-ROI can provide important clues for proper interpretation of the measured data relating to this region.

A further advantageous embodiment of the invention is achieved if the sample parameters are based (at least in part) on staining assays performed with the sample. Possible staining assays include, for example, H & E (hematoxylin-eosin), IHC (immunohistochemistry), FISH (fluorescence in situ hybridization), PLA (proximity ligation assay from olin, sweden), PPA (padlock probe assay from olin, sweden), rolling circle amplification, RCA (olin, sweden), branched strand DNA signal amplification, and combinations of all these techniques or other assays to obtain specific biological information. Staining may be particularly helpful in identifying particular cell types or tissue types, and/or molecules indicative of particular properties or functions or abnormalities of the cells.

The selection of the image-ROI may be based at least in part on the determined sample parameters. If the sample parameter is indicative of, for example, a local amount of tumor cells, the image-ROI may be selected to include those regions where the parameter is above a given threshold.

While it is generally possible to generate image-ROIs of nearly arbitrary shape and size, this is not generally true for sample-ROIs, as this must be achieved with actual physical samples. In a preferred embodiment of the invention, the size and/or shape of the image-ROI is thus adjusted according to the requirements set by the possible sample-ROI. For example, the size of the image-ROI and/or the curvature of its boundary may be limited to be larger than a given minimum value or smaller than a given maximum value, respectively. This adjustment has the following advantages: only such image-ROIs are generated that can actually be transferred into the physical sample-ROI, thus avoiding a mismatch between the desired and actual selection of samples for molecular assays. This adjustment may preferably be done automatically by a suitable (digital) image processing routine, e.g. based on a given user selection.

Obtaining a pure fraction of the cells of interest can be time consuming and requires high resolution with removal. The foregoing scheme allows relaxing the requirements on selection and balancing, with additional parameters that take into account how easy or reliable slices can be removed and control the overall size of the selection. For larger routine selections, serial slices are preferred. Image analysis may provide tabulated parameters such as the number and fraction of each identified cell type in a region of potential interest (e.g., total surface area). The shape of the region may be limited by design criteria including parameters such as total area, allowable curvature, and connectivity. Based on the algorithm, given a certain selection algorithm specific for each MDx assay, the optimum condition can be determined. Instead of providing a homogenous sample, a well characterized sample is obtained in this way, which fulfils the requirements regarding MDx testing (like fractions of tumor cells) and the requirements for easy isolation (like geometrical parameters of the selected ROI).

There are several possibilities how the sample-ROI can be isolated from the sample. One possibility is the use of Laser microdissection known from the literature (cf. Falko Fend, Mark Raffeld: "Laser capture microdissection in pathology", J. Clin. Pathol. 2000, 53: 666-. Another possibility is the use of a printing device with which the indication of the sample-ROI can be printed on the sample itself, thus allowing a human operator (or machine) to separate the sample-ROI from the rest of the sample. According to a further embodiment, the image-ROI may be represented in a (e.g. digital) microscope such that it is visible to a human operator or machine together with the original sample.

The sample may preferably be provided on some carrier to allow easy handling. For example, a thin section of body tissue will typically be provided on a microscope slide as a carrier. Typically, the material of the carrier (or substrate) on which the sample may be provided comprises glass, transparent plastic and/or a composite of glass and plastic, optionally with a surface layer for the desired interaction with the biological sample. Furthermore, the carrier may have the shape of a cartridge, e.g. a cartridge having an open cavity, a closed cavity or a cavity connected to other cavities by fluid connection channels.

The aforementioned support may preferably comprise at least one marker, i.e. an element on the support that can easily be physically located and localized in the image of the sample. The marker thus provides a reference which allows mapping of the "image coordinate system" (in the image plane) to the "sample coordinate system" (relating to the actual sample on the carrier). This coordinate mapping is important for proper isolation of the sample-ROI, which must be done in physical space based on the image coordinates.

When the sample is provided on a carrier, the sample-ROI is preferably transferred from the carrier to a separate holder (container, cartridge, tube, etc.) after or during isolation of the sample-ROI from the rest of the sample. The individual holders may then be further transferred to a molecular examination device for performing a desired molecular assay of the sample-ROI. The rest of the sample may instead rest on the carrier and be stored or discarded, for example.

According to a further development of the invention, images of the sample-ROI and/or of the remaining sample are generated after isolation of the sample-ROI. The image may be generated using the same image generation unit that also generates an image of the entire sample, or using a separate device. The image of the sample-ROI (or of the rest of the sample) can be compared with the image of the entire sample and in particular with the selected image-ROI, thus allowing to verify whether the actual sample-ROI corresponds to the desired region of interest.

It has been mentioned that molecular assays may comprise one or more of a wide variety of different tests. In particular, the molecular assay may comprise PCR (e.g.q-PCR, qRT-PCR, RT-PCR, qRT-PCR or digital PCR), sequencing (in particular next generation sequencing), or microarray hybridization, or another molecular assay technique or a combination of these.

The image data showing the entire sample (and optionally also the selected image-ROI) is preferably combined with the assay data generated during the molecular assay(s) such that they are simultaneously accessible to the user. The specific tissue staining data and the molecular assay data can be parsed to provide additional information about the individual cellular function or characteristic. Accordingly, the apparatus preferably comprises a user interface at which the image data and the assay data are accessible. The user interface may for example comprise a memory of data and a display (monitor) on which the data can be displayed simultaneously, preferably so that the assay data is displayed at the image location from which they originate. Thus, the collected information can be presented in a user-friendly and intuitive manner, facilitating their evaluation.

According to a further development of the invention, a plurality of sample-ROIs (based on a corresponding plurality of image-ROIs) is isolated, wherein different sample-ROIs are subjected to respectively different determinations. Thus, it will be possible to investigate different regions in the same sample with respect to different questions, e.g. to search for a first marker in one region and another marker in another region.

The aforementioned definition and isolation of the image-ROI and the sample-ROI can occur in parallel (simultaneously) and/or sequentially. The selection of the (second) image-ROI and the isolation of the corresponding (second) sample-ROI may for example be done based on the combined results of the previous examinations of the (first) image-ROI and the corresponding (first) sample-ROI. Such an inspection may be subsequently repeated even more than once.

According to a third aspect, the present invention relates to a sample isolation unit for isolating a sample-ROI in a sample of biological origin (e.g. a tissue mass which should be examined for the presence of e.g. tumor cells), wherein the sample isolation unit comprises the following components:

-a light source for generating a light beam.

-a light guiding system for guiding the aforementioned light beam to locations outside the sample-ROI such that the sample material at these locations is altered to become separable from the rest of the sample.

The sample isolation unit may be used in particular in a device of the kind described above. It has the following advantages: which allows a simple and versatile isolation of a region of interest in a sample by appropriately treating parts of the sample not belonging to the region of interest with a light beam. Due to the flexibility and precision with which the beam can be controlled, it is possible to isolate sample-ROIs of almost any arbitrary shape.

According to a fourth aspect, the present invention relates to a method for isolating a sample-ROI from a biological sample, said method comprising the steps of:

-generating a light beam.

-directing the aforementioned beam to a location outside the sample-ROI such that the sample material therein is altered to become separable from the rest.

The sample isolation unit and method described above are based on the same innovative concept of treating the area of the sample outside the sample-ROI by means of a light beam. The explanations and definitions provided for the sample isolation unit are therefore also valid for the method and vice versa. Hereinafter, various preferred embodiments of the present invention will be described in relation to both the sample isolation unit and the method.

Preferably, all locations outside the sample-ROI are treated in the described manner. In other words, the entire complement of sample-ROI was treated.

In general, the present invention encompasses any reaction of a sample material to a light beam by which the material becomes separable from the sample-ROI (i.e., the sample material that has never been disposed). For example, sample material may be fixed ("burned in") to a carrier on which the sample is provided, thus allowing later selective removal of untreated sample-ROIs from the carrier.

In a preferred embodiment of the invention, the modification of the sample material outside the sample-ROI by the light beam comprises ablation of the sample material. Thus, simply remove the undesired sample material, leaving the desired sample-ROI. An advantage of this solution is that ablation of the sample material is possible in particular for almost all kinds of biological tissue, provided that sufficient optical energy is applied to the respective region. Moreover, transfer of the sample-ROI to another container or holder is not necessary, which further simplifies the workflow.

According to a further development of the aforementioned solution, a waste receptacle can be provided for collecting ablated sample material. The waste reservoir may preferably be replaceable so that it can be renewed from time to time, for example after each isolation of a sample-ROI. Providing a waste reservoir avoids the problem of uncontrolled build-up of ablated sample material, which may for example lead to contamination of the sample-ROI.

To enable and/or support the modification of the sample material by the light beam, the sample material may comprise a photosensitive reagent that has been added prior to the light treatment. The reagent may for example be a stain for microscopic investigation and to which molecules or chemical agents are coupled, and which may be activated by light to destroy stained areas.

The modified beam for sample material outside the sample-ROI may in particular comprise a high power LED beam or a laser beam, which has the advantage of allowing a spatially well-localized application of high intensity.

Typically, the power density of the applied beam is higher than about 0.1mW/μm²Preferably greater than about 1.0mW/μm². When applying a modulated light beam (e.g. a pulsed laser), the aforementioned values refer to the average power density determined over the entire period of time of light application.

In general, the light of the light beam will advantageously have a spectrum comprising wavelengths that are well absorbed by the biological tissue. Such wavelengths may generally include UV (ultraviolet) light (about 100nm to 380 nm) and/or IR (infrared) light (about 800nm to about 1 mm), but also visible light. IR light is particularly suitable for ablation of (biological) sample material.

In one embodiment, the light beam may be adapted to illuminate the entire sample under consideration simultaneously. In another embodiment, the light directing system may include a scanning element (e.g., a movable mirror and/or a movable sample holder) for scanning the light beam across the sample or portion thereof. In this case, the beam reaches only a small limited portion of the sample at a time, wherein these portions are sequentially and repeatedly moved over the entire sample or portions thereof.

In each of the aforementioned embodiments, care must be taken that the beam does not (or at least does not with an intensity exceeding a given threshold) reach a position in the sample-ROI. This can be achieved if the light guiding system comprises light controlling elements for (fully or partially) suppressing the generation and/or propagation of the light beam if it is to reach a position within the sample-ROI.

Given the above-described scanning of the light beam, the light control element may control the scanning element of the light guiding system such that it only guides the light beam to a desired position. Alternatively, the scanning element may be designed to scan the entire area of the sample, and the light control element inhibits the generation of the light beam (e.g., by turning off the light source) or the propagation of the light beam (e.g., by closing the shutter) if the light beam is to be directed to a location within the sample-ROI.

According to a further embodiment of the present invention, the aforementioned light control element may be adapted to receive position data defining a sample-ROI. This allows for flexible application of the sample isolation unit, as only appropriate data has to be transmitted in order for it to isolate a sample-ROI of any desired shape. In particular, it thus becomes possible to isolate different, individual sample-ROIs from each sample.

Other embodiments of the invention include at least one of the following features:

-generating an image of the sample-ROI and/or the remaining sample after isolation of the sample-ROI.

-transferring the sample-ROI from the carrier to a separate holder.

Performing a molecular assay using the isolated sample-ROI, said assay preferably comprising a PCR step, a sequencing step and/or a microarray hybridization.

The sample ROI corresponds to the "image-ROI" (region of interest) that has been selected from the image of the sample.

It has been mentioned that the sample isolation unit and the corresponding method may be applied in particular in the method according to the first aspect as well as in the device according to the second aspect of the invention. Additional information about the sample isolation unit can thus be obtained from the description of the method and apparatus.

The invention further relates to the use of the above-described device or sample isolation unit for molecular diagnostics, molecular pathology, in particular for oncology applications, biological sample analysis, chemical sample analysis, food analysis and/or forensic analysis. Molecular diagnostics can be done, for example, with the help of magnetic beads or fluorescent particles that are directly or indirectly attached to target molecules.

Drawings

These and other aspects of the invention will be apparent from and elucidated with reference to the embodiments described hereinafter.

In the figure:

figure 1 schematically shows the examination of a sample according to the invention;

FIG. 2 schematically shows a sample isolation unit according to the present invention;

fig. 3 and 4 show photographs of the sample-ROI after ablation of the undesired material.

In the drawings, like reference numerals designate identical or similar parts.

Detailed Description

The present application also relates to the following embodiments:

1. a method for the examination of a sample (11), the method comprising the steps of:

a) generation of an image (I) of the sample (11);

b) analyzing the image with respect to at least one sample parameter;

c) selection of a region of interest in said image (I), said region being called "image-ROI" (R)_I）；

d) From the sample (11) corresponding to the image-ROI (R)_I) And is called "sample-ROI" (R)_S) Isolation of the region of interest;

e) performing a molecular assay using the sample-ROI;

f) linking assay data to the sample parameters for the corresponding sample region.

2. An apparatus (1000) for examination of a sample (11), comprising:

a) an image generation unit (200) for generating an image (I) of the sample (11);

b) an image analyser for analysing the image with respect to at least one sample parameter;

c) image selection for selecting a region of interest in the image (I)Selecting a cell (300), said region being called "image-ROI" (R)_I）；

d) For isolating a region from the sample (11) corresponding to the image-ROI (R)_I) And is called "sample-ROI" (R)_S) A sample isolation unit (400) of the region of interest;

e) for performing the utilization of the sample-ROI (R)_S) A molecular test unit (500) for measuring the molecule of (1);

f) an evaluation unit (301, 302, 303, 304) for linking assay data to the sample parameter of a corresponding sample region.

3. According to the method of the embodiment 1,

characterized in that the sample (11) is stained before the generation of the image (I) and/or after the performance of the molecular assay.

4. According to the method of the embodiment 1,

characterized in that the sample (11) is covered by a coverslip during the generation of the image (I).

5. The apparatus (1000) according to embodiment 2,

characterized in that the image selection unit (300) comprises an automatic image analysis module and/or a user interface (301, 304).

6. The method according to embodiment 1 or the apparatus (1000) according to embodiment 2,

characterized in that the sample parameter is indicative of a local amount of a specific cell type or tissue type.

7. The method according to embodiment 1 or the apparatus (1000) according to embodiment 2,

characterized in that the sample parameter is based at least in part on a staining assay performed with the sample.

8. The method according to embodiment 1 or the apparatus (1000) according to embodiment 2,

characterized in that said image-ROI (R)_I) Is based at least in part on the sample parameter.

9. The method according to embodiment 1 or the apparatus (1000) according to embodiment 2,

characterised by passing the possible sample-ROI (R)_S) Set requirements to adjust the image-ROI (R)_I) The shape and/or size of (a).

10. The apparatus (1000) according to embodiment 2,

characterized in that the sample isolation unit (400) comprises a laser microdissection device and/or a printing device.

11. The method according to embodiment 1 or the apparatus (1000) according to embodiment 2,

characterized in that the sample (11) is arranged on a carrier (10) comprising at least one marker (M).

12. The method according to embodiment 1 or the apparatus (1000) according to embodiment 2,

characterized in that the performed assay comprises a PCR step, a sequencing step and/or microarray hybridization or another molecular diagnostic technique.

13. According to the method of the embodiment 1,

characterized in that said image selection comprises performing a manual rough selection of a region in said image and performing an algorithm to automatically adjust refine said region, thereby obtaining a more precise definition of said image-ROI.

14. One type for samples (11) is called "sample-ROI" (R)_S) The isolated sample isolation unit (400) of a region of interest, in particular for use in the device (1000) according to embodiment 2,

characterized in that the sample isolation unit (400) comprises:

-a light source (401) for generating a light beam (L);

-for directing said light beam (L) to a region in said sample-ROI (R)_S) Out of position and/or in the sample-ROI (R)_S) The position within is such that the sample therein is altered to become a light directing system (402) that is separable from the rest.

15. Use of the device (1000) according to embodiment 2 or the sample isolation unit (400) according to embodiment 14 for molecular pathology, in particular for oncology applications.

Pathological diagnostic investigations of patient material (tissues and cells) are the basis for many therapeutic decisions, particularly in oncology. Standard, thin sections from biopsies are presented on microscope slides and stained according to some protocol to visualize the morphology of the tissue. Recently, in situ staining for disease-specific biomarkers is being developed for concomitant diagnosis of targeted drugs. Assessment can be accomplished using bright field microscopy. Slides need to be stored for long periods of time after the investigation as a backup, if the diagnosis needs to be re-assessed.

Digital pathology is a new technology in which a microscope is replaced by a digital scanner that automatically scans stained tissue sections and stores the images in a digital format with sufficient resolution and color rendering so that the pathologist can make the same diagnosis from the digital images as he/she would have made directly under the microscope. The latter means that the digital image may replace a physical slide. The digital image is stored instead of the slide. The original biopsy sample is of course always stored.

Following the above format of analysis, tissue and cell biopsies can also be investigated using molecular methods such as q-PCR and sequencing (referred to as "molecular diagnostics" or "MDX" for short). This so-called molecular pathology is increasingly important with the advent of new molecular biomarkers. Typically, a pathologist decides to run molecular tests based on morphological information to identify the biological properties of the cancer tissue for proper treatment selection. Since many molecular biomarkers cannot be quantified in situ on tissue, or at least cannot be quantified with the required accuracy, individual molecular diagnostic tests (such as PCR or sequencing) are performed on samples obtained from biopsies, usually from fragments that have been obtained from biopsies. The tissue sections are treated by cell lysis prior to measurement of the DNA or mRNA markers. Therefore, spatial information is lost.

Tumor tissue is usually composed of many different cell types (not just cancer cells), and even cancer cells can differ greatly in molecular composition in different regions of the tumor. The results of the molecular analysis will therefore depend on the precise composition of the tissue section of the sample used for the molecular test. The more diluted the cancer cells, the less sensitive and uncertain the test results will be. Furthermore, heterogeneity within cancer cell populations will also cause noise in MDX assays, reducing sensitivity and specificity and reproducibility.

In the near future, it will also become important to select other cell types, such as particular types of immune cells, from slides with thin sections of cancerous tissue; but also from tissue slides from diseases other than cancer, the selection of particular cell types will be required to perform molecular diagnostics.

With digital pathology, there is currently no possibility to run molecular tests on the sub-sections of the biopsy fragment. Due to the dilution of target cells with benign cells of different origin (e.g., endothelium, fibroblasts, and immune cells) and cancer cell heterogeneity, tests on all segments have suboptimal accuracy and sensitivity. With conventional pathology it is not possible to accurately label tissue sections for molecular analysis, because the samples for molecular testing must come from new slides, since the stained and examined samples need to be stored.

While molecular diagnostic information is of increasing importance for the correct diagnosis of cancer (and other diseases), it is not actually used frequently by pathologists. As explained above, the main problem is to define a representative, well-defined sample from the tissue biopsy used for MDX testing. Manual selection is imprecise due to the heterogeneity of tumor tissue (or other diseased tissue) and the imprecise location of tissue slices relative to the tumor (or other diseased) location in general. Manual selection may establish contamination, especially for PCR that amplifies even very low contamination concentrations. Manual selection does not allow for good annotation of the organization selected for MDX. Computer-assisted tissue selection suffers from the problem of loss of reference between successive slides because the selection cannot be made from the exact same tissue section that is used for in situ staining (which is the basis for selection).

Therefore, there is a need for accurate selection of sample materials for molecular testing based on pathological images.

To address this need, a new scheme is proposed according to which a region of interest (ROI) in the sample is first selected from an image of the sample, and then the region of interest (ROI) in the sample is actually extracted from the physical sample, and subjected to molecular determination.

In an exemplary embodiment of the protocol, the tissue slides are stained according to some clinical indication, for example using HER2 immunohistochemistry or immunofluorescence staining (IHC) or a combination of staining assays. The slide is then scanned by a digital scanner and the resulting image may be analyzed by a computer program to identify and indicate regions of common features. Those areas may be presented to a pathologist for confirmation and adjustment, if necessary. From those regions, regions of interest (referred to as "sample-ROIs") can be defined automatically or semi-automatically by a software program that represents the portions of the sample selected for MDX testing. Typical parameters were annotated to the sample-ROI, such as the mean expression of HER2 in the given example, as well as the statistical distribution of expression over the cells and cellular components in the selection. The coordinates of this selected sample-ROI may be communicated to a sample isolation unit, which is responsible for the physical selection of the sample for MDX. The selected "sample-ROI" can be transferred to a transfer device or directly to a disposable device used for MDX testing. MDX testing may include sample preparation and molecular analysis, such as qRT-PCR, sequencing or next generation sequencing, or microarray hybridization, or a combination of these. The results of this analysis can ultimately be correlated with information from the tissue selection algorithm and optionally parsed and presented to a pathologist along with a digital image of the tissue (where the sample-ROI selected for MDX is also indicated).

Fig. 1 schematically illustrates an apparatus 1000 according to the invention suitable for examination of a sample 11 of body tissue according to the described procedure.

The examination starts at a sample preparation unit 100, in which a thin section 11 of body tissue is prepared on a microscope slide 10 serving as a carrier. Typically, tissue samples are obtained by cutting thin sections of approximately 4-8 microns from paraffin embedded biopsies. The so-called sections were placed on a microscope glass slide 10 on a water film to stress relax from the microtome and then allowed to dry.

Moreover, the sample 11 may optionally be stained by a suitable stain 12, for example by hematoxylin & eosin (H & E) or IHC. There are several staining protocols available for different applications. The staining protocol may be performed manually on the bench by soaking the slide with the fragments in different solutions containing reagents, but may also be performed in an automated manner.

One or more markers M may be printed on the microscope slide 10 or engraved in the microscope slide 10, which may later be used as reference points for correlating image coordinates with actual coordinates on the slide 10. Furthermore, the slide 10 with the sample 11 is preferably covered with a cover slip (not shown) in order to allow for a high image quality during later scanning.

After the preparation step, the slide 10 with the sample 11 is transferred to an image generation unit 200, which image generation unit 200 may in particular be a digital scanning microscope.

A digital image I of the sample is generated by the microscope 200 and communicated to the sample selection unit 300, where the sample selection unit 300 is implemented by a workstation 302 having a display (monitor) 301, a memory 303 and input devices 304, such as a keyboard and a mouse. The image I of the sample may be displayed on a monitor 301 to allow visual inspection by a pathologist. The pathologist may identify a region of interest R in the image_IAnd marks it accordingly.

This identification of the image-ROI RI may preferably be assisted by (or accomplished by) an automatic image analysis routine. As a first non-limiting example, the software tool may identify individual cells and calculate a score for the biomarker under consideration based on a digital image of the IHC stained slide (optionally overlaid with information from the H & E scan). The tool may then calculate a similar or identical scored region with corresponding statistics of cell numbers, an average and histogram of the scores of the cells in that region(s). The region may be optimized for overall size, continuity, and scoring statistics, optionally taking into account constraints resulting from tissue selection techniques, as requested by a pathologist. As a result of the calculation, an image-ROI RI is determined, which can be visualized on the screen 301 and/or in the microscope 200 (while viewing the sample slide) together with the corresponding statistics. The image-ROI RI can then be manually adjusted and selected by the pathologist with the help of a cursor. As a second non-limiting example, the identification of the image-ROI RI can be defined in a first step by a pathologist making a rough selection of regions inside or outside the image-ROI RI. The region may be identified, for example, by a display of a boundary line including points representing a certain number of clicks the user makes with respect to quickly defining the region. In a second step, the region may be used in an algorithm (e.g. the one just described in the previous paragraph) to refine the position of the boundary line. In this effect, the algorithm can significantly provide a rough segmentation of the region to allow identification of individual nuclei or cells. For each individual cell in the field of view, a characteristic (e.g., stain uptake, cell type, cell proliferation, cell size, cell morphology, etc.) can be calculated. As is well known in the art, the adjustment to achieve a more accurately identified region of the image-ROI RI may then be performed by searching for neighboring cells or nuclei having similar characteristics. Separation techniques such as k-nearest neighbor or online machine learning may be used.

Preferably, the pathologist can mark the position at any magnification. This labeling will generally be done based on tissue morphology, which is the basis for decisions about malignancy of the lesion or the different cell types present. An unlimited number of images can be selected and marked. When done, the software can provide the pathologist with an overview of the selected image-ROI at appropriate magnification and adjust this region to the necessary resolution that can be processed later by the sample isolation unit 400, which sample isolation unit 400 is responsible for the physical selection/isolation of the sample for molecular testing. A file is created containing the actual (image and/or sample) coordinates of the image-ROI RI (positive and negative selection) boundaries and the necessary references that can be resolved by the sample isolation unit 400.

The aforementioned document can be used as an input to a device that can transfer this information to the slide, for example by printing techniques to indicate only areas that can then be removed manually or by another device, or as an input to a sample isolation unit that can directly remove the indicated sections and transfer them to a sample holder that can be introduced into a molecular testing instrument (or sample preparation instrument).

In the embodiment shown, the microscope slide 10 with the sample 11 is then transferred to a sample isolation unit 400, where it corresponds to the selected image-ROI R_IROI R of the sample_SIsolated (e.g., by positive or negative selection) and separated from the rest of the sample. Preferably, the sample-ROI R_SAre transferred to a separate holder (e.g., test tube 20). Furthermore, the sample isolation unit may preferably be capable of successively removing several regions and submitting those to separate molecular tests. If the sample 11 on slide 10 is covered with a cover slip, this will be removed before isolation of the sample-ROI occurs.

The actual physical selection and transfer of the organization may be performed in several ways. One of them is by Laser Microdissection (LMD). LMD is a technique that can be used to select individual cells or tissue portions from a tissue slide with the aid of laser-induced transfer of cells to tape or into a container. This technique allows for precise transfer of tissue. The LMD laser may be moved on the slide and, alternatively, the slide may be moved under a stationary laser beam. In the latter case, all samples will be collected at the same spot, so that the collection device can be very compact.

In digital pathology scanning, the instrument moves the slide under consideration under the objective lens. Tools can be used to find the area where the sample is present in order to optimize the scan time. Current applications of such scanning procedures require having physical reference coordinates. Due to tolerances in slide size, it is preferable to have an indicator on the slide, such as the aforementioned indicia M, that can be simultaneously read by a digital scanner with image scanning. An alternative is to use mechanical stops and push the slide against the stops for reproducible positioning. Another alternative is to include a scanning function in the sample isolation unit responsible for selection and use a software tool to overlay the newly scanned image with the original image using the surface delineated by feature recognition.

In sample-ROI R_SAfter the isolation, a new image can be scanned from the tissue slide 10 to confirm and control the selection. The image may be archived on workstation 302 along with the raw image and the results of the MDX analysis.

With sample-ROI R_SIs transferred in a final step to the molecular examination unit 500, wherein the determination of interest is performed using the sample-ROI. The microscope slide 10 with the rest of the sample may optionally be stored in the storage unit 600 for subsequent access and verification, or it may simply be discarded. Later processing steps with the slide 10 may particularly include a subsequent examination including, for example, at least one new (particularly different) staining and/or at least one new (particularly different) molecular assay with another region of interest.

In the molecular examination unit 500, several molecular techniques may be available for analysis of the selected sample-ROI, such as PCR (under this term including several techniques such as q-PCR, RT-PCR, qrt-PCR, digital PCR, etc.), for detecting single point gene mutations or any other DNA mutations, DNA deletions, DNA insertions, DNA rearrangements or amplification of the copy number or other structural changes in cancer cells, and/or determining the extent of RNA expression of genes or other transcribed DNA sequences in cancer cells, or RNA or DNA sequencing (next generation sequencing), for determining a broader spectrum of genetic variations in cancer cells, e.g. over the whole genome or exome, or targeted to smaller regions of the genome, targeted to exosomes, or transcriptomes, and performed at various depths. The results are interpreted in terms of the genetic mutations of the cancer cells and their corresponding RNA expression profiles, which correlate with prognosis or alternatively sensitivity to a certain treatment, e.g. by targeted drugs, or actually assessing the effect of a treatment that has already started or ended (therapy monitoring). Moreover, the results of such molecular analysis can be coupled to image-based analysis of the same sample slice, and reported together and also interpreted in combination.

Fig. 2 schematically shows a preferred sample isolation unit 400 according to the present invention. The sample isolation unit 400 generally allows for a new way of selecting tissue sample material from thin sections after histopathological investigation. A necessary step of this protocol is the removal of all material that should not be part of the sample-ROI, optionally followed by the transfer of all remaining material into the tube 20 for further analysis.

The removal of the undesired material may be based in particular on laser ablation. As explained above, the regions to be removed can be selected by the software tool based on the pathology examination following histopathological staining. Images of the remaining sample may be generated after removal of the undesired material for accurate documentation and characterization (e.g., quantification) of the input material for MDX analysis.

The special sample isolation unit 400 shown in fig. 2 comprises a light source 401, by means of which light source 401 a light beam L, for example a (laser) light beam, is generated. The light beam L is directed onto the sample 11 provided on the slide 10 using a light directing system, for example comprising a scanning element (such as a movable mirror 402 and/or a shutter). The light-guiding system further comprises a control unit 403 which receives data defining a desired sample-ROI on the slide 10 and which may switch the light source 401 on and off and/or control the movement of the scanning element 402. Thus, the light source and/or the light guiding system may be controlled such that only locations outside the sample-ROI are illuminated by the light beam L.

In practice, it may be possible to have a concentration of about 3mW/μm²The pulsed laser is used for the pulse of the power density of (1). Typical wavelengths of the laser light are, for example, 355nm and 405 nm. With shorter wavelengths, the absorption of the tissue increases, but the absorption of the substrate also increases, if it is desired to operate in a closed compartment. The absorption of biological material has a minimum in the green color unless the tissue is stained with a marker for absorption within this range. If an IR laser is used, the absorption of water increases with wavelength. This works because of the equipment and assays involvedThe tissue may be dry or soaked with water.

The negative selection (i.e. removal of undesired material) may be very fast, since the possible deterioration of the material is insignificant. A scanning laser or a focused high power LED beam can be employed for ablation of materials based on thermal effects due to absorption of radiation. The ablated material may be concentrated or deposited on the waste reservoir or surface 404, which waste reservoir or surface 404 may optionally be replaced after each run.

The beam L of radiation can be controlled in the light guiding system 402 by actuators and shutters and can be patterned in any pattern (determined by the components in the optical path) with high resolution. The remaining sample-ROI can be transferred to the test tube by mechanical means or by buffer solution. The process may be performed in a closed or open system. Laser ablation may also be used to mark multiple regions that may then be individually selected manually.

The described solution is based on a scanning beam for removing material by thermal ablation. In contrast to laser microdissection, in which cells are transferred from a slide to a substrate or cup in a precisely controlled manner, all material that is not desired in the final sample is ablated here, so that only the selected sample material (sample-ROI) stays on the slide and the slide can be handled as if the entire section were to be subjected to MDX. This means that the downstream flow will be independent of whether a selection has taken place.

The light directing system 402 may be software controlled. The software control allows the use of a user-friendly visual interface that allows the delineation of the selected area on an interactive computer screen or equivalent tool, while zooming in or out, optionally in overlapping combination with images from a previous stain (such as IHC or ISH or any other method that supports the correct selection by the pathologist).

Depending on the light source used, there may be sufficient power to expand the laser beam in one or two dimensions. The spreading may be modulated according to the characteristics of the surface that needs to be removed. The gap may be skipped. Standard slides may be used, but specialized substrates are also possible to facilitate the ablation process or sample transfer to MDX or both. Moreover, the scanning procedure only requires a relative movement of the beam and the sample and may thus also comprise embodiments in which the sample holder 10 is moved (additionally or alternatively), e.g. via a movable stage.

In one example, a 1W UV laser at 355nm was used to ablate FFPE tissue samples from standard microscope slides. FIG. 3 shows a triangular shape Inv _ R_SA photograph of the sample 11 after ablation. It can be seen that very clean areas can be removed from the tissue section with a sharp interface. It can be concluded that the resolution of the removed tissue is better than 50 microns at the chosen conditions. The tissue in between the removed regions is stable and the same resolution can be obtained with the tissue left behind. The tissue used was breast tissue at a thickness of 4 microns. Utilizing H prior to laser ablation&E staining the tissue.

Fig. 4 shows the result of ablation of a larger area having a (negative) "miqi mouse" shape. The images demonstrate that large areas can also be removed easily and in any desired shape.

In summary, a method for selecting and annotating sample material for MDX analysis based on histopathological staining is described. Histochemical scoring on an intercellular basis is used as a selection criterion for determining regions of interest on slides that exhibit staining of samples for MDX analysis. The selection may be done by an automatic algorithm and/or optionally in combination with manual adjustment by the pathologist. Alternatively, the coarse selection may be done manually by a pathologist, in combination with automatic adjustment by an algorithm. Statistics can be built on cell type, number and score from this selection. This information is linked to the MDX analysis. MDX analysis uses only and precisely the tissue material of the described area. This is made possible by (i) using the same slides for (immuno) histochemistry and MDX sample selection, and (ii) having a digital image containing all the information so that the stained slides do not need to be kept intact and stored. Computer-assisted selection of regions of interest is combined with automated physical sample removal interfaced with MDX sample preparation and detection techniques. This method gives a well defined sample input for MDX. The input is considered with an evaluation of the MDX results. In this way, MDX results can be annotated to cell types and histochemical scores of the same cells, which makes pathological diagnosis more accurate and reproducible. By eliminating less relevant tissue, the accuracy and precision of these assays will be improved, resulting in improved diagnosis. This approach can be combined with a digital pathology scanner that can store relevant pathology images for diagnosis, making storage of the actual slides for later redundant reference. The overall flow may include one or more of the following steps:

1. staining of slides with samples (e.g. IHC).

2. Digital scanning of the slide.

3. Storage of the resulting image files (e.g. PACS, IMS).

4. Parsing of the image (e.g. by CAD).

5. Selection of a region for MDX ("image-ROI") (optionally using an optimization algorithm).

6. Annotation of regions for MDX (including scoring statistics).

7. Physical selection of regions for MDX ("sample-ROI").

8. Scanning and storing of images of the sample after selection.

9. The sample is transferred to the MDX holder(s) (cartridge, tube … …).

10. Storage of slides with residue.

11. MDX analysis of selected samples.

12. Annotation of MDX results using image scoring statistics (from 6).

13. Resolution of MDX (optionally using scoring statistics as input for the resolution algorithm).

14. Annotation of images with MDX results.

15. Dye and resolution of the combined results of MDX.

16. A possible new cycle of selection and MDX analysis.

According to another aspect of the invention, a method and apparatus are described that allow for accurate and efficient selection of sample material from tissue for further analysis (e.g., qPCR and/or sequencing). The method is based on destroying/removing all sample material except the part selected for analysis (in contrast to the tedious positive selection that must leave the selected material unaffected). This can be done very efficiently by laser irradiation (e.g. scanning IR laser) which vaporizes the material at the exposed area. The remaining samples can be easily transferred by hand or robotically into test tubes or cartridges for MDX analysis. The method can be applied directly to the investigated tissue sections on standard glass slides. The inspection after selection can be easily done because the selected material is not changed and not displaced. By controlling the actuated focused beam with software, high resolution is possible. Depending on the laser power, high speeds can be achieved. The system may be linked to a digital pathology scanner and image analysis software to perform the selection of regions of interest and characterization of input materials.

An important advantage of the proposed approach is a single tissue section, which implies 100% accurate mapping of the selected image ROI to sample-ROI and 100% complete annotation of MDX samples. Moreover, due to the quantitative input and reduced heterogeneity of MDX samples, more accurate and reproducible MDX results can be achieved, allowing for better resolution. The procedure also requires less handling (no manual steps) and less organization (single fragments). It generates a digital file of integrated staining and selection and MDX results and finally resolves histopathological scores comprising the selected MDX samples.

The invention can be applied in molecular pathology, in particular for oncology applications based on patient stratification for the identification of molecular changes in cancer cells, but also for diagnosis/monitoring of other diseases.

While the invention has been illustrated and described in detail in the drawings and foregoing description, such illustration and description are to be considered illustrative or exemplary and not restrictive; the invention is not limited to the disclosed embodiments. Other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the appended claims. In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite article "a" or "an" does not exclude a plurality. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage. Any reference signs in the claims shall not be construed as limiting the scope.