Methods for assessing parenchymal tissue damage

Abstract

The present invention provides a method for assessing damage to organ tissue, in particular ischemic damage / injury to organ tissue, comprising the step of measuring the concentration of at least one marker molecule in the perfusate of the organ tissue, wherein the measured concentration of the at least one marker molecule in the perfusate is used in at least one computer-based predictive algorithm for generating at least one success score, the success score being predefined based on at least one parameter value of at least one predefined parameter, the at least one parameter value being determined after transplantation of the organ tissue, and wherein based on the at least one success score, at least one signal and / or at least one dataset is generated to assist in determining whether the organ tissue is suitable for transplantation.

Description

[Technical field]

[0001] The present invention relates to a method for assessing organ tissue damage, in particular ischemic damage / injury to organ tissue, and to a spectroscopic analysis unit for use in such a method. [Background technology]

[0002] explanation Since the early days of liver transplantation (LT), graft quality has always been considered an important confounding factor for prognosis. However, even now, half a century later, the assessment of liver quality before transplantation and the decision to accept or reject the graft still largely depend on “intuition” in combination with donor attributes and past medical history.

[0003] Ex situ machine perfusion is a method of increasing interest to assess organ function and the repair potential of damaged organs. Moreover, machine perfusion before transplantation can provide an objective assessment of the metabolic state of an organ.

[0004] Machine perfusion techniques are used to preserve human organs (e.g., ex vivo) by perfusing the organ / tissue with a perfusion solution for a certain period of time. Depending on the perfusion temperature (hypothermia 0-12 °C, parathermia 12-35 °C, normothermia 35-37 °C) and the organ used, the perfusion solution can be whole blood, saline, perfusion solutions based on artificial oxygen carriers, modified blood (e.g., leukocyte-reduced) or custom-made solutions.

[0005] Clinical applications of machine perfusion for human organs can be performed both within the patient (in vivo, e.g., normothermic regional perfusion (NRP)) and outside the patient (ex vivo). For hypothermic perfusion (e.g., 4 °C), perfusion can be performed with or without an active oxygenator (artificial lung) that is part of the perfusion loop (tubing set). Thus, perfusion techniques under hypothermic conditions can be oxygenated or non-oxygenated. When the perfusion temperature is raised to normothermia, an artificial lung becomes part of the perfusion loop and also requires at least one oxygen carrier (e.g., red blood cells) in the perfusion solution. Machine perfusion techniques can be generally applied to all human organs (e.g., liver) and human tissues. In 2012, a new machine perfusion approach, hypothermic oxygenated perfusion (HOPE), was introduced into routine clinical practice to improve human liver grafts donated after pre-transplant circulatory death (DCD).

[0006] Only a few parameters related to metabolic and cellular damage have been quantified in machine liver perfusates, most notably perfusate lactate, transaminases, pH, bile production and bile quality. In contrast, two recent studies have demonstrated the predictive value of deeper analysis of cryoflushing out at the end stage of cryopreservation by metabolomics and glyconomics. These findings suggest superior clinical relevance of specific metabolic pathways in the liver compared to quantification of released cytosolic compounds.

[0007] However, in the field of liver transplantation, there are no studies regarding reliable prediction of graft function during machine perfusion. Summary of the Invention [Problem to be solved by the invention]

[0008] It is therefore an object of the present invention to identify and test the predictive value of organ tissue perfusate markers analysed during machine perfusion, in particular HOPE perfusion, for organ tissue function after transplantation, e.g. liver function after liver transplantation. [Means for solving the problem]

[0009] This object is solved by providing a method for assessing organ tissue damage, in particular ischemic damage / injury to organ tissue, and a spectroscopic analysis unit for use in such a method.

[0010] According to the present invention, there is provided a method for assessing damage / injury to an organ tissue, in particular ischemic damage / injury to an organ tissue, comprising the steps of: - measuring the concentration of at least one marker molecule (as a predicted target molecule) in the perfusate of an organ tissue (i.e. any fluid that flows through a tissue or an organ); - the measured concentration of the at least one marker molecule in the perfusate is used in at least one computer-based predictive algorithm to generate at least one success score; - the success score is predefined based on at least one parameter value of at least one predefined parameter; - the at least one parameter value is determined after transplantation of the organ tissue, and - Based on said at least one success score, at least one signal and / or at least one data set is generated to assist in determining whether said organ tissue is suitable for transplantation or not. Effect of the Invention

[0011] Thus, a method is provided that allows predicting the likelihood or probability of organ transplant success prior to organ transplantation. The method allows assessing the quality of the organ tissue, since the determined success score reflects the degree of organ tissue damage. The signal or data set provided by the method allows a yes / no indication that the organ transplant will be successful or not. [Brief description of the drawings]

[0012] The method is explained in more detail below with reference to the figures. [Figure 1] General scheme of the spectroscopic analysis unit; [Diagram 2] A spectroscopic analysis unit describing a first embodiment for the realization of spectroscopic fluorescence and absorption measurements in perfusate based on LEDs and at least one full-spectrum light source; [Diagram 3] A spectroscopic analysis unit illustrating a second embodiment for the realization of spectroscopic fluorescence and absorption measurements in perfusate based on a full-spectrum light source; and [Figure 4] General scheme of an embodiment of the method according to the invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] In an embodiment of the method, the at least one success score corresponds to a concentration of at least one marker molecule (predicted target molecule) in the perfusate, where a pre-determined threshold value of the at least one marker molecule is used to generate at least one signal and / or at least one set of data to assist in organ tissue transplantation decision before organ tissue transplantation. For example, if the concentration of the marker molecule is below a certain threshold, the organ may be used for transplantation, but if the concentration of the marker molecule is above a certain threshold, the organ may not be used for transplantation.

[0014] The method uses marker molecules (such as FMN) as indicators of damage, among others, to characterize damage in tissues and organs. For example, the amount of FMN is used as an indicator of ischemic tissue damage and / or reperfusion injury. In particular, the amount of FMN present in blood after ischemic heart injury shows a positive correlation with the degree of ischemic tissue damage. Thus, a large amount of FMN reveals more ischemic damage compared to a small amount of FMN.

[0015] Thus, the concentration of at least one marker molecule in the perfusate can be used as a predictor of organ tissue quality prior to transplantation of the organ tissue and for classifying the organ tissue into risk groups.

[0016] In a further embodiment of the method, the computer-based predictive algorithm is a regression algorithm or a classification algorithm, where the predefined parameters used by at least one predictive algorithm are pre-transplant information and post-transplant parameters. The predictive algorithm uses information stored in a database, where the database includes pre-transplant parameters and post-transplant parameters.

[0017] The computer model may include a corresponding set of predictive algorithms, each of which may be trained on a unique information pool of previously available pre-transplant information along with a corresponding set of post-transplant parameters of meaningful previous transplants.

[0018] The computer model further combines a set of predicted post-transplant parameters and associates them with a success score, preferably weighting each parameter to emphasize certain post-transplant parameters over others, which is used to aid in the decision process of whether or not to transplant the tissue.

[0019] In the most preferred embodiment of the method, machine learning and artificial intelligence are applied to characterize the state of the organ tissue.

[0020] In this approach, many clinical parameters can be used as predefined parameters in the prediction algorithm. The available information used to decide whether to transplant solid organ tissue to a recipient or to predict other post-transplant parameters consists of serial and / or single point measurements of any or a combination of marker molecules (or predictive target molecules), in particular spectroscopic measurements as described in detail below.

[0021] At least one marker molecule (predicted target molecule) may be selected from the group including FMN, FAD, NADH, alanine aminotransferase (ALT), aspartate aminotransferase (AST), glucose, lactate, where FMN, NADH and FAD are the most preferred marker molecules.

[0022] This information is passed to a prediction algorithm and is hereafter referred to as pre-transplantation information. In a further embodiment of the present invention, the pre-transplantation information comprises information about the donor of the solid organ tissue, such as gender, age, cause of death, ethnicity, medical records, medical status, height, body mass index, ischemia time of the solid organ tissue, and also comprises information about the potential recipient of the solid organ tissue, such as gender, age, cause of death, ethnicity, medical records, medical status, height, body mass index, etc. The pre-transplantation information further comprises information about both the donor and the recipient of the solid organ tissue. In a further embodiment of the present invention, the pre-transplantation information consists of data about the solid organ tissue and / or about its donor and can be utilized to find the optimal recipient, for example by maximizing the probability of a high success score. For example, in the case of liver transplantation, pre-transplant information consists of measurements of FMN, FAD and NADH during 30 minutes of hypothermic parenchymal perfusion (HOPE) using, for example, Belzer MPS™ UW Machine Perfusion Solution (Bridge To Life) as the perfusion fluid at 9-11°C, and information about the liver donor, and a success score consisting of a weighted sum of EAD and MELD (Model for End-Stage Liver Disease) defined by Olthoff et al. (Validation of a current definition of early allograft dysfunction in liver transplant recipients and analysis of risk factors. Liver Transplant. 2010;16:943-949) can be used in a computer model to match optimal recipients.

[0023] Typical post-transplant parameters are often clinical parameters that characterize the transplant outcome, the most important parameters including the survival of the recipient or the major non-function of the solid organ transplanted into the recipient. Other important clinical parameters include the concentration of lactate in the blood at different time points, INR at different time points, transcription factors, inflammatory markers, tumor necrosis factor, creatinine, and others. Furthermore, post-transplant parameters can be associated with economic values, such as post-transplant costs incurred by patient care, reimbursements, intensive care unit costs, etc.

[0024] The predictive algorithms used in the present invention can include any classification or correlation algorithm, such as (boosted and gradient) random forest algorithms, decision tree algorithms, logistic regression algorithms, neural network algorithms or genetic algorithms. In general, to train or otherwise improve the algorithm, a set of pre-transplant information previously available, together with post-transplant meaningful parameters corresponding to past transplant data, is compared with the actual tissue for which the prediction is sought. With this pool of information, the predictive algorithm can learn to predict, or improve its prediction of, the meaningful parameters. And in this way, the predictive algorithm can be used for predicting post-transplant parameters. In the case of classification, the post-transplant parameter can be, for example, patient survival or solid organ graft survival after a period of 4 months, 1 year, 5 years. In this scenario, the predictive algorithm could result in a binary yes / no prediction or it could determine the probability of a successful outcome for patient survival or solid organ graft survival. In the case of regression algorithms, the post-transplant parameter can be factor V or peak AST after a series of periods, for example, 48 hours post-transplant or 7 days post-transplant. In this scenario, the predictive algorithm produces a numerical value for the predicted post-transplant parameter. Preferably, the prediction algorithm will be updated as more real-world data on transplant outcomes becomes available, e.g., 1-10 8 The algorithm is periodically refined, with n in the range of 0 to 1, and the algorithm is periodically trained for every n additional data points.

[0025] It can be supplied with metrics used in the learning process of the predictive algorithm. If a regression algorithm is used, the metrics may consist of mean squared error, root mean squared error, mean absolute error, or any other user-provided function that produces a scalar output. If a classification algorithm is used, these may include classification accuracy, sensitivity (at a user-provided cutoff value), specificity (at a user-provided cutoff value), prevalence, detection rate, detection prevalence, balanced accuracy, precision, recall, F1 score, area under the receiver operating characteristic curve (AUROC), lift curve, or any other user-provided function that produces a numerical output.

[0026] The predictive algorithms may be used to generate a computer model that reflects the likelihood of a successful transplant outcome characterized by a set of post-transplant parameters. Preferably, the computer model produces predictions for a set of different clinical parameters. In such a scenario, the computer model may include a corresponding set of predictive algorithms, each of which may be trained on a unique pool of previously available pre-transplant information along with a set of corresponding post-transplant parameters of meaningful previous transplants.

[0027] The computer model can further combine a set of predicted post-implantation parameters, for example by weighting each parameter to emphasize some over others, into a success score that can be used to aid in the decision process of whether or not to transplant the tissue.

[0028] One aspect of the present invention is that a database can be constructed by accumulating data. The database includes data on the donor of the solid organ tissue, data on the stored solid organ tissue measured by any or a combination of the methods (predictive target molecules) described above. If the solid organ tissue is transplanted, the database includes data on the recipient and the corresponding post-transplant outcome parameters. If the solid organ tissue is not transplanted, it is possible to provide the database with the appropriate reasons why the transplant was denied. Both pre-transplant information and post-transplant outcome data can be provided to the database and are available immediately as well as subsequently.

[0029] The database can then be used to improve the prediction algorithm, for example when the data of the five new transplant outcomes is provided to the database. The resulting more pre-transplant information and corresponding post-transplant parameters can be queried from the database and fed to the prediction algorithm. By having the prediction algorithm learn more pre-transplant information, it can improve its success score and better predict the post-transplant parameters of subsequent transplants.

[0030] The database may be on a local data carrier of the abovementioned analysis unit or the database may be on a remote data carrier. In the latter case, a (secure) connection to the database can be established in order to query from / add data to the database via a communication protocol, e.g. wireless technology such as Bluetooth, network protocols, etc. If necessary, the communication between the analysis unit and the remote database carrier can be encrypted. For example, the communication between the analysis unit and the remote database carrier can occur via a secure shell protocol.

[0031] Embodiments of the invention can include a visual computer interface and printable reports that monitor current parenchymal organ tissue statistics and compare them to past success scores in real-time (i.e., during parenchymal organ tissue evaluation) or on a historical success score basis. Additionally, additional measurement data or donor specific data can optionally be displayed and visualized on the interface.

[0032] The computer model and / or database may be implemented in hardware or software or a combination of the two. The computer model is preferably implemented in a computer program executed on a programmable computer, each of which includes a processor, a processor-readable storage medium (including volatile and non-volatile memory and / or storage elements) and suitable input / output devices. The program code is applied to data entered using the input device to perform the described functions and generate output information. The output information is applied to one or more output devices. Furthermore, each program is preferably implemented in a high-level procedural or object-oriented programming language to communicate with the analysis unit and the database. If necessary, the programs can also be implemented in either assembly or machine language. In any case, it may be a compiled or interpreted language. Each such computer program is preferably stored on a general-purpose or special-purpose programmable computer-readable storage medium or device (e.g., hard disk, magnetic diskette, CD-ROM) for configuring and operating the computer when the storage medium or device is read by the computer and executes the described procedures. Furthermore, the system may be implemented as a computer-readable storage medium configured with a computer program, the storage medium so configured causing the computer to operate in a specific and predefined manner.

[0033] The method according to the invention combines several aspects: · Algorithms that relate clinical markers (e.g. FMN values) to patient outcomes in organ transplants such as liver transplants and other clinical diagnostic procedures; · The cut-off value of marker molecules such as FMN (8800 AU, corresponding concentration) can determine the feasibility of transplantation; · Perfusate / organ combinations; Extension to other organs, tissues, or damage as a whole; and -Data exist on measurements in dialysis fluid, cardiac and hepatic perfusion.

[0034] The method is performed using a spectroscopic analysis unit including at least one spectrometer and at least one computer processor for predicting at least one success score and implementing at least one predictive algorithm, said at least one spectrometer using UV / VIS spectroscopy and / or fluorescence spectroscopy, and the spectroscopic analysis unit is coupled to at least one perfusion machine and / or at least one perfusion loop, as described in more detail below.

[0035] The amount of marker molecules (predictive target molecules), especially FMN in the perfusate, i.e., the fluid used to wash and / or perfuse parenchymal organ tissues during explantation, harvesting, in vivo, or ex vivo storage, positively correlates with the degree of damage or injury in the case of parenchymal organs, e.g., liver, kidney, heart, lung, pancreas, uterus, limbs, intestine, and reproductive organs. In the case of organ washing or machine perfusion of organs (in vivo or ex vivo), FMN and other meaningful marker molecules are actively washed out of the tissue into the perfusate. Thus, information about the degree of organ damage and injury is also present in the perfusate, based on the amount of such marker molecules (e.g., FMN and / or FAD, and / or NADH) present therein. Thus, a high amount of FMN as a marker molecule in the perfusate reveals that there is more ischemic damage compared to when the amount of FMN is low. Ischemic injury of organs is one of the main risks for recipients in the case of transplantation. Since FMN correlates with various clinically important parameters and ultimately with transplant outcome, the amount of FMN measured in the perfusate as a marker molecule can be used as a criterion for deciding whether a donor organ should be transplanted or discarded.

[0036] In particular, the amount of FMN in the perfusate significantly correlates with clinical parameters in liver transplantation, as shown in Table 1 below:

[0037] [Table 1]

[0038] FMN in the perfusate (also in combination with other parameters and / or other predictive target molecules, e.g. FAD, NADH) can be used as a predictor to determine the quality of the organ before transplantation, and organs can be successively grouped into two (low, high) to three (low, medium, high) risk groups according to FMN release at the desired time point (FMN cut-off value). These risk groups characterize the recipient's risk when receiving the organ. Moreover, FMN concentration in the perfusate of ex vivo perfused human organs strongly correlates with graft function, early graft loss and patient survival. Thus, with regard to such predictive target molecules for transplantation outcome, there clearly exists a predictive value of machine perfusate analysis (e.g. FMN concentration and / or FAD concentration and / or NADH concentration in the perfusate). FMN is expected to be released from the mitochondria of cells linked to the level of mitochondrial complex I damage. For donated organs (e.g. livers) that are themselves marginal due to DCD (donation after cardiac death) or extended criteria DBD (donation after brain death), this evaluation technique allows for sparing use of such organs, thus increasing the donor pool and saving patients on the waiting list. Also, for seemingly perfect organs, this evaluation step is useful because it implements an additional safety feature that checks the quality of the organ before transplantation.

[0039] Furthermore, apart from FMN as a predictive target molecule for quantifying organ damage / injury and thus predicting the success rate of transplantation, there are other molecules that have proven their predictive value. Other predictive target molecules (predictive parameters) for tissue / organ (ischemic) damage / injury in the perfusate are flavin adenine dinucleotide (FAD), nicotinamide adenine dinucleotide (NAD) and lactate. Furthermore, ALT, glucose and AST in the perfusate can also be considered. The amount of these predictive target molecules (substances) in the perfusate (e.g., during machine perfusion) also strongly correlates with organ damage / injury, the quality of the respective organ and ultimately the outcome of the transplantation. For example, a combination of high values ​​of FMN (e.g., >9000 AU) and high lactate (e.g., >4 mmol / L) clearly indicates an organ of poor quality.

[0040] The amount of these predicted target molecules in the perfusate can be measured at the desired time points by taking perfusate samples from the perfusion loop. On the other hand, in situ (online) continuous measurements are possible during organ machine perfusion. Furthermore, these predicted target molecules can be measured in the perfusate after organ washing, even in the preservation solution in which the organ is stored, for example during back-table or static cryopreservation. The concentration of e.g. FAD, FMN, NADH, AST, ALT, glucose, lactate, etc. (predicted target molecules) in the perfusate can be measured at separate time points with standard independent clinical laboratory measuring devices / methods (e.g. Radiometer ABL90, Piccolo Xpress from Abaxis, UV / VIS spectroscopy, MRI analysis, chromatography) or continuously and in situ (online) in the perfusate during machine perfusion. Online measurements of glucose and lactate can be performed continuously and in situ with these two systems (CITSens Bio and MeMo from C-CIT Sensor AG).

[0041] Continuous or discontinuous measurements of FMN, FAD and NADH in the perfusate are performed by detecting the fluorescence and absorption spectra by the spectroscopic analysis unit. Discontinuous FMN and FAD measurements are performed by sampling the perfusate, for example, in a cuvette or Eppendorf tube. Then, this sample (e.g., cuvette or Eppendorf tube) is placed in the sample holder of the spectroscopic analysis unit for spectroscopic analysis. On the other hand, continuous measurements during organ perfusion are performed using a flow cell (flow-through cell, flow-through cuvette) integrated in the spectroscopic analysis unit. In such a spectroscopic analysis flow cell, the perfusate from the machine perfusion loop can flow constantly through this flow cell (sensor) for spectroscopic analysis to obtain real-time and time-resolved data. Therefore, the flow cells of the spectroscopic analysis unit must each be connected via a tube, for example, a LuerLock connector to the perfusion loop assembly of the perfusion device. After passing through the flow cell, the perfusate can be returned to the machine perfusion loop or discarded. To perform reliable measurements, a permanent and stable flow must be established in the flow cell. This can be achieved by an appropriate pressure difference between the inlet and outlet of the flow cell, given by different pressure levels in the perfusion loop. Alternatively, a pump (such as a roller pump) can be used to pump fluid through the flow cell.

[0042] Additionally, the spectroscopic analysis units can measure directly through special tubing in the perfusion loop itself. In this case, no sample holder or flow cell is required, since each spectroscopic analysis unit is directly connected to the desired outer tubing of the perfusion loop of the perfusion device. In this case, the spectroscopic sensor unit can be clamped on or threaded through.

[0043] In embodiments where the perfusate is blood-based (e.g. whole blood or perfusate with red blood cells), the perfusate can be directly analyzed in the spectroscopic analysis unit. Alternatively, a portion of the perfusate first passes through a hemodialysis filter that separates (blood) plasma from blood cells (e.g. red blood cells), and then the plasma portion flows through a flow cell. The plasma may or may not then be recirculated back into the perfusion loop. Alternatively, a continuous separation of blood into blood cells and plasma can also be achieved by a microfluidic device (microreactor). In case of sample-based analysis, the plasma is spectroscopically analyzed after filtration or centrifugation of the blood cells.

[0044] Conductivity and capacitance measurements of the perfusate / body fluids can be additionally implemented in the spectroscopic analysis unit. Conductivity and / or capacitance measurements of the perfusate / body fluids can also be useful to quantify the quality of the organ or to detect diseases in the patient.

[0045] In general, the spectroscopic analysis unit can be incorporated into an independent in vitro diagnostic device, where a sample of perfusate taken (e.g., in a cuvette or Eppendorf tube) can be spectroscopically analyzed (e.g., by absorption and fluorescence) to detect a molecule of interest (predicted target molecule), e.g., FMN. Furthermore, several body fluids other than perfusate can be spectroscopically analyzed to measure the amount of a particular molecule, thereby detecting / diagnosing a patient's disease, functional disorder, genetic abnormality, cancer, and infectious disease.

[0046] On the other hand, the spectroscopic analysis unit can be fully integrated (part of) all state-of-the-art machine perfusion techniques, for several organs, types of perfusate and all temperature ranges (hypothermic, subnormothermic and normothermic), in which case, for example, FMN can be measured directly (in real time and continuously) in the perfusion loop by a flow cell (e.g. disposable) and also directly from the perfusion loop tubing itself.

[0047] Furthermore, the spectroscopic analysis unit can be used as an add-on product to all state-of-the-art organ perfusion technologies. Therefore, the spectroscopic analysis unit must be connected to the desired organ perfusion loop via tubing and, for example, a Luer lock connector. After passing through the flow cell (of the spectroscopic analysis unit), the perfusate can be returned to the machine perfusion loop or can be discarded.

[0048] Moreover, the spectroscopic analysis unit can be used as an add-on product to all state-of-the-art dialysis machines. The spent dialysate leaving the dialysis filter is fed to the flow cell of the spectroscopic analysis unit for spectroscopic analysis to detect the amount of molecules (predictive target molecules) that are meaningful for detecting / diagnosing diseases, functional disorders, genetic abnormalities, cancer, and infectious diseases in patients. In general, the spectroscopic analysis unit can also be fully integrated into all state-of-the-art dialysis machines. In general, flavins (e.g., FMN and FAD as predictive target molecules) have two absorption maxima at about 360-390 nm and about 440-470 nm. The two most common biological forms of flavins are flavin mononucleotide (FMN, riboflavin 5' phosphate) and flavin adenine dinucleotide (FAD), both of which emit fluorescence. Free FMN has an absorption maximum at 373 nm and 445 nm (ε = 10,400 and 12,500 M, respectively). -1 cm -1 ), whereas free FAD has absorption maxima at 375 nm and 450 nm (ε = 9300 and 11300 M, respectively). -1 cm -1 ) The fluorescence emission maxima of FAD and FMN are at 525 nm.

[0049] The applied spectrometer of the spectroscopic analysis unit usually operates in the ultraviolet / visible (UV / VIS) region, but can also cover a wider range (100-3000 nm). The spectroscopic analysis unit is preferably used to measure the fluorescence of flavin mononucleotide (FMN) and / or FAD as fragments of mitochondrial complex I. For example, during ex vivo machine perfusion, detection of FMN and / or FAD in the perfusate can detect the degree of (ischemic mitochondrial) damage / injury to parenchymal organ grafts and tissues before transplantation. In particular, the method is applicable to all parenchymal organs and tissues in ex vivo perfusion systems. FMN and / or FAD signal intensities extracted from the fluorescence spectra (linked to their concentrations in the perfusate) can be used to assess the degree of (reperfusion) damage / injury of the organ. Real-time or non-real-time measurement of FMN and / or FAD in the perfusate aids in clinical decision making (whether to transplant an organ or not, minimizing transplant risk to the recipient) by optimizing the graft-recipient matching process, thereby improving the recipient's survival and quality of life after transplantation. In addition to liver transplantation, this method is applicable to any ex vivo perfused tissue, preferably parenchymal organ tissue such as liver, heart, lung, kidney, pancreas, uterus, limbs, reproductive organs or intestine.

[0050] FMN is detected by fluorescence spectroscopy. In detail, at least one optical probe is connected to at least one light source (LED and / or full range (100-3000 nm, e.g. halogen) combined with bandpass filters) that emits nearly monochromatic light at a wavelength of about 445 nm (and / or about 373 nm) by excitation of the perfusate flow or the perfusate sample. At least one receiving probe connected to at least one spectrometer (preferably placed at 90° to the optical probe) with a sufficiently high resolution (e.g. 4.6 nm) and sensitivity was used to quantify the proportion of fluorescence emitted by FMN molecules. The fluorescence emission maximum of FMN was measured between 475 and 600 nm, more precisely at 525 nm.

[0051] FAD is detected by fluorescence spectroscopy. Specifically, at least one optical probe is connected to at least one light source (LED and / or full range (100-3000 nm, e.g. halogen) combined with bandpass filters) that emits nearly monochromatic light at a wavelength of about 445 nm (and / or about 373 nm) by excitation of the perfusate flow or the perfusate sample. At least one receiving probe connected to at least one spectrometer (preferably placed at 90° to the optical probe) with a sufficiently high resolution (e.g. 4.6 nm) and sensitivity was used to quantify the proportion of fluorescence emitted by the FAD molecule. The fluorescence emission maximum of FAD was measured between 475 and 600 nm, more precisely at 525 nm.

[0052] Nicotinamide adenine dinucleotide (NAD) exists in two forms: oxidized and reduced, abbreviated as NAD+ and NADH, respectively. Similar to FMN, the amount of NADH in the perfusate can also quantify the tissue (ischemic) damage / injury and mitochondrial function of each organ. Therefore, NADH (NAD+, NAD, respectively) is also a predictive target molecule to predict the success rate of each transplant outcome. NAD, NAD+, and NADH are detected by fluorescence spectroscopy in the perfusate. NADH absorbs light at 320-380 nm (maximum at 340 nm) and emits fluorescence in the range of 420-480 nm (maximum at 463 nm).

[0053] Also, UV / VIS absorption spectroscopy techniques can be implemented in the perfusion loop using a spectroscopic analysis unit to perform optical absorbance measurements of the perfusate, for example to quantify the amount of molecules significant for organ quality (predictive target molecules) and therefore to predict the outcome of the transplant. For absorbance measurements, the optical probe and the receiving probe are aligned with each other.

[0054] Further methods that can be applied to detect putative target molecules in perfusates and / or body fluids are UV-VIS spectroscopy, fluorescence spectroscopy, Raman spectroscopy, mass spectroscopy, circular dichroism spectroscopy, (near) infrared spectroscopy.

[0055] In general, a spectroscopic analysis unit is used to detect the amount of predicted target molecules (e.g., FMN and / or FAD) in the perfusate via fluorescence and / or absorbance spectroscopy. The spectroscopic analysis unit can measure only fluorescence spectra (fluorescence spectroscopy), only absorption spectra (absorbance spectroscopy), or both methods are applied.

[0056] The parts of the spectroscopic analysis unit include a light source, a spectroscopic measurement unit with a sample holder, a computer processor connected to a storage medium, and a user interface (see Figure 1). Further parts of the spectroscopic analysis unit are a data acquisition, processing and storage unit, as well as a control unit for switching, synchronizing and controlling the light source, bandpass filters, optical filters, shutters and spectrometers. Furthermore, several modern data and connection interfaces are part of the system.

[0057] Moreover, the spectroscopic analysis unit can be combined with any state-of-the-art perfusion machine. Thus, the sensor unit can perform spectroscopic measurements (fluorescence and / or absorbance spectroscopy) directly through (specialized) tubing of the perfusion loop of the perfusion machine itself. In this case, no sample holder or flow cell is needed, since the spectroscopic analysis unit is directly connected to the desired outer tubing of the perfusion loop of the perfusion machine. In this case, a clamp-on or thread-through method of the spectroscopic sensor unit is conceivable. Alternatively, the sensor unit is connected to a special flow-through device that is part of the perfusion loop. This special flow-through device is preferably a disposable device and is integrated by the manufacturer of the tubing set for the desired perfusion machine.

[0058] Multiple spectrometers with variable wavelength operating ranges (e.g., 200-3000 nm), different resolutions and sensitivities can be part of a spectroscopic analysis unit to detect and screen different predicted target molecules with fluorescence and / or absorbance spectroscopy.

[0059] Generally, all light sources can be operated in pulsed and non-pulsed (continuous) mode. For absorption spectroscopy (e.g., optical probe 3), light sources in the full wavelength range (e.g., 200-3000 nm) are generally used. Suitable here are, for example, white light LEDs, halogen lamps, deuterium tungsten halogen lamps, xenon lamps, etc.

[0060] In fluorescence spectroscopy, the perfusate is excited with (nearly) monochromatic light and / or a narrow wavelength band (depending on the type of light source) at a wavelength where the desired predicted target molecule has an absorption maximum. The response to such excitation is then detected by a spectrometer as fluorescence at the characteristic emission maximum of the desired predicted target molecule. In fluorescence spectroscopy, different LEDs of different wavelengths are used, coupled to the desired optical probes (e.g., optical probes 1 and 2), to excite the perfusate at different wavelengths, allowing screening of predicted target molecules that fluoresce differently.

[0061] FIG. 2 represents an embodiment of a spectroscopic analysis unit for realizing the measurement of fluorescence and / or absorbance in the perfusate. In a preferred embodiment, the spectroscopic analysis unit comprises at least one LED light source (106) emitting approximately monochromatic light at a desired wavelength for excitation of the perfusate to at least one fluorescent optical probe (e.g., 101 or 102). The optical probes (101, 102) are part of a flow cell / sample holder / sensor unit (100). Furthermore, a full wavelength light source (107) emitting light in a full wavelength range is connected to the flow cell / sample holder / sensor unit (100) via an absorbance optical probe (103). For fluorescence spectroscopy, the optical probes (101, 102) are positioned at 90° to the receiving probe (104). For absorbance spectrum measurement, the optical probe (103) is aligned (180°) to the receiving probe (104). The receiving probe (104) is part of the flow cell / sample holder / sensor unit (100). The various possible LEDs (106) emit light at different wavelengths, and preferably only one LED alternately emits light towards one optical probe (101, 102). At the same time, the receiving probe (104) detects the fluorescence emission of the perfusate and sends this light information via a fiber optic cable (105) to a spectrometer (108) for detection and analysis. Fluorescence and absorbance measurements of the perfusate are preferably performed alternately, and a spectrum is constantly detected via the receiving probe (104). When an absorbance measurement is performed, the full-spectrum light source (107) sends light via the fiber optic cable (105) to the absorbance optical probe (103). All detected spectra are processed by a data acquisition / processing / storage / transfer-unit (109). Control of the light sources (106, 107), shutter and spectrometer (108) is achieved via a control unit (109).

[0062] Apart from LEDs, full wavelength light sources (e.g., 200-3000 nm) can also be applied in fluorescence spectroscopy. Thus, for example, a light beam emitted by a halogen lamp hits a bundle of switchable and controllable optical filters (e.g., bandpass filters) to excite the perfusate in the desired narrow wavelength band (e.g., 440-460 nm) that passes through the bandpass filters. The various bandpass filters can be individually controlled to allow excitation in the desired wavelength range. In a preferred embodiment, each single and circular filter disk has an empty position where no filtering takes place. For each filter disk, there are several individual bandpass filters. By rotating the filter disk around the central axis, the wavelengths that pass through the filter to excite the perfusate can be changed. Every switchable and controllable bandpass filter bundle can be composed of several individual filter disks to allow fluorescence screening over a wide wavelength range. Furthermore, filter disks that do not filter but only reduce the light intensity can also be part of the system.

[0063] FIG. 3 represents a spectroscopic analysis unit that uses a full-spectrum light source (107) that emits light that passes through a switchable and controllable bandpass filter bundle (110). Different bandpass filters within the bundle can be individually controlled to allow excitation at desired wavelength ranges. Besides the release of FMN, FAD, and NADH in the perfusate, various other predicted target molecules, namely xanthine, hypoxanthine, succinate, xanthosine, nicotinic acid, nicotinamide adenine dinucleotide (NAD / NADH), flavin adenine dinucleotide (FAD / FADH), inosine, inosine-5'-monophosphate, 8-hydroxyguanosine, uric acid, biliverdin, protoporphyrin, purines, riboflavin, uracil, uridine, 8-aminopropyl phosphate, 1,2-di ... -Hydroxyguanosine, adenosine triphosphate, adenosine diphosphate, malonic acid, pyruvate, aconitic acid, fumaric acid, malic acid, aspartic acid, citrate, aconitic acid, adenine, propionylcarnitine, choline, lactate, proline, leucine, tryptophan, phenylalanine, tetramethylrhodamine, adenosine diphosphate (ADP), adenosine triphosphate (ATP), creatine, N-acetyl-L-glutamic acid can be released and detected. These molecules are measured and monitored by spectroscopy, e.g., nuclear magnetic resonance spectroscopy analysis of the perfusate at different time points during perfusion. This data, together with FMN measurements and functional tests (bile production, coagulation), further improves the evaluation of previous organ / tissue (e.g., liver) grafts by improving the recipient's post-transplant survival.

[0064] The combination of information available in any of the above embodiments can be used in a predictive algorithm.

[0065] For example, in the case of liver transplantation, the amount of FMN is measured during hypothermic oxygenated perfusion (HOPE) using Belzer MPS™ UW machine perfusion solution (Bridge To Life, Inc.), and the perfusion is operated at 4-12°C, preferably 9-11°C. The peak value of FMN, for example the peak value after 30 minutes of perfusion, is used as pre-transplant information. Using the AUROC method, a computer model of the success criterion "Early Allograft Dysfunction (EAD) as defined by Olthoff et al." in this simple example consists of a decision rule, where EAD is predicted if the amount of FMN is higher than a threshold (any 8800 units, e.g., a concentration amount of FMN in weight per volume of fluid, e.g., mg / mL, which can be matched by a device-specific calibration curve), and EAD is not predicted if the amount of FMN is lower than the threshold. In this context, a prediction algorithm can assist in the decision to transplant or discard the liver.

[0066] As a second example, the success score includes a combination of the INR at 24 hours, the peak of the recipient's AST at 7 days after transplantation, and the number of days the recipient must spend in the ICU, with the first two parameters being weighted more than the latter. Thus, the economic contribution (expressed through the ICU stay) is weighted less than the contribution characterizing the health status of the recipient.

[0067] The scheme in Figure 4 summarizes the steps taken by the method: The spectrometer provides a signal corresponding to the concentration of the marker molecule as one piece of pre-transplant information. This pre-transplant information is complemented with donor and / or recipient data. The (combined) data is passed through a predictive algorithm running on a computer processor to return a prediction of a success score. A computer model is created that reflects the post-transplant parameter predictions. The computer model is stored in a database.

[0068] Definitions, technical terms: Perfusion fluid: a liquid used to wash and / or perfuse parenchymal organ tissue in vivo or ex vivo. Also, solutions applied to human organs for preservation purposes during static cryopreservation on ice are called perfusion fluids in this context. Moreover, body fluid and perfusion fluid have the same meaning in the framework of this patent. Perfusion fluids can be whole blood, blood-based or non-blood-based perfusion fluids, saline solutions, perfusion fluids based on artificial oxygen carriers, modified blood (e.g. leukocyte-reduced) or custom solutions, such as: University of Wisconsin machine perfusion solution (Belzer-MPS™); Institute-George Lopez-1 (IGL-1) solution; Ringerfundin™, B. Braun; Plasma-Lyte A, Baxter).

[0069] Body fluids: blood, plasma, urine, bile, saliva, sweat, lymphatic fluid, gastric juice, pancreatic secretions, breast milk, vaginal secretions, tears, nasal mucus, sperm, menstrual fluid, surfactants Marker molecules or predictive target molecules: molecules / substances that are measured in the perfusate in real time (on-line, continuous) or non-real time (sample-based). The amount (concentration) of these predictive target molecules present in the perfusate quantifies (ischemic) damage / injury of human organs / tissues and thus predicts the success rate and outcome of transplantation. Predictive target molecules / marker molecules are FMN, FAD, NADH, lactate and others.

[0070] Reperfusion injury: Reperfusion injury is tissue damage caused by the re-supply of blood to tissues after a state of ischemia or oxygen deficiency (anoxia or hypoxia). Ischemia: Ischemia, ischemic (or blood loss) is the restriction of blood supply to a tissue, causing a deficiency of oxygen, nutrients, and glucose necessary for cellular metabolism to maintain tissue survival. Ex vivo: Outside the body; In vivo Perfusion Loop: The perfusion loop must be connected to the human organ by suitable means such as a cannula. Furthermore, the perfusion loop is related to the durability of the perfusion machine. The perfusion loop is technically realized through a tubing set. Tubing set: The tubing set is the technical realization of a perfusion loop, including all tubing, connectors, sensors, valves, ports, oxygenators, filters, dialyzers, pump heads, clamps, etc. to allow for the desired organ / tissue machine perfusion.

[0071] Abbreviation: ADP: adenosine diphosphate Al:Artificial intelligence ALT: Alanine aminotransferase AST: Aspartate aminotransferase ATP: Adenosine triphosphate AUROC: Area under the receiver operating characteristic curve DBD: Brain death specimen DCD: specimens after circulatory death D-HOPE: Dual Hypothermic Oxygenation Mask Perfusion EAD: Early graft dysfunction FAD: flavin adenine dinucleotide FMN: flavin mononucleotide HOPE: Hypothermic Oxygen Perfusion (HOPE) INR: International Normalized Ratio LED: Light Emitting Diode LT:Liver transplant MELD: A model for end-stage liver disease NAD, NADH: Nicotinamide adenine dinucleotide NRP: normal temperature local perfusion ROS: reactive oxygen species [Explanation of symbols]

[0072] 100 Flow cell, sample holder, sensor unit 101 Fluorescent Probes 102 Fluorescent Probes 103 Absorption Probe 104 Light receiving probe 105 Fiber optic cable 106 LED 107 Full wavelength range light source 108 Spectrometer 109 Data acquisition, processing, storage and transmission units, control units 110 Switchable and controllable bandpass filter bundle

Claims

1. 1. A method for assessing organ tissue damage, comprising: - characterized by measuring the concentration of at least one marker molecule in the perfusate of the organ tissue, - the measured concentration of said at least one marker molecule in said perfusate is used in at least one computer-based predictive algorithm to generate at least one success score; - the success score reflects the degree of organ tissue damage, - said success score is predefined based on at least one parameter value of at least one predefined parameter; - said computer-based predictive algorithm is a regression algorithm or a classification algorithm; - the predefined parameters used by said at least one predictive algorithm are pre-transplant information determined before transplantation of said organ tissue and post-transplant parameters determined after transplantation of a previous organ tissue, and - based on said at least one success score, at least one signal and / or at least one set of data is generated to aid in deciding whether said organ tissue is suitable for transplantation or not.

2. The method described in claim 1, characterized in that the damage to the organ tissue is ischemic damage / injury to the organ tissue.

3. 10. The method of claim 1, wherein machine learning and artificial intelligence are applied to characterize the state of the organ tissue.

4. The method according to any one of claims 1 to 3, characterized in that the at least one success score corresponds to a concentration of the at least one marker molecule in the perfusate, and a predetermined threshold value of the at least one marker molecule is used to generate the at least one signal and / or at least one set of data for assisting organ tissue transplantation decisions prior to organ tissue transfer.

5. The method according to any one of claims 1 to 4, characterized in that the computer model comprises a set of corresponding predictive algorithms, each algorithm potentially having been trained on a unique information pool of previously available pre-transplant information together with a corresponding set of post-transplant parameters meaningful for the previous transplant.

6. 6. The method of any one of claims 1 to 5, wherein the computer model further combines a set of predicted post-transplant parameters and associates them with a success score, said success score being used to aid in the decision process of whether to transplant the tissue.

7. A method according to any one of claims 1 to 6, wherein the computer model further combines a set of predicted post-transplant parameters and associates them with a success score, weighting each parameter to emphasize certain post-transplant parameters over others, and the success score is used to assist in the decision process of whether or not to transplant tissue.

8. The method according to any one of claims 1 to 7, characterized in that the prediction algorithm uses information stored in a database, said database comprising pre-transplant and post-transplant parameters.

9. 9. The method according to any one of claims 1 to 8, characterized in that at least one marker molecule is selected from the group consisting of FMN, lactate, FAD, NADH, alanine aminotransferase (ALT), aspartate aminotransferase (AST), glucose, with FMN and FAD being the most preferred marker molecules.

10. 10. The method according to any one of claims 1 to 9, characterized in that the concentration of at least one marker molecule in the perfusate is measured in real time (online, continuous) or non-real time (sample-based).

11. 11. The method according to any one of claims 1 to 10, characterized in that the pre-transplant parameters are at least one of the following: information relating to the donor of the solid organ tissue; and / or information relating to the potential recipient of said solid organ tissue, in addition to the concentration of at least one marker molecule.

12. A method according to any one of claims 1 to 11, characterized in that the information relating to the donor of the solid organ tissue is information regarding gender, age, cause of death, ethnicity, medical records, medical status, height, body mass index, and ischemia time of the solid organ tissue.

13. A method according to any one of claims 1 to 12, characterized in that the information relating to the potential recipient of the solid organ tissue is at least one of the following information: gender, age, cause of death, ethnicity, medical records, medical status, height, and body mass index.

14. The method according to any one of claims 1 to 13, characterized in that the post-transplant parameters comprise the survival of the recipient or the major non-functioning of the transplanted solid organ in the recipient, the concentration of lactate in the blood at different time points, the INR, transcription factors, inflammatory markers, tumor necrosis factor, creatinine at different time points.

Images