In vitro method for diagnosing or assessing progression of condition or response to treatment by circulating disease load parameters

Abstract

The present invention relates to an in vitro method for diagnosing or assessing the progression of a condition in an individual or assessing the responsiveness / response of an individual to a treatment, comprising: counting the number of cells of interest and the number of cell debris of interest; -determining a circulating disease load (CDL) parameter which is a function of the number of target cells and the number of target cell debris; and-correlating the circulating disease load (CDL) parameter with the presence or progression of a condition or responsiveness / response to treatment in the individual.

Description

[0001] Cross-reference to related applications

[0002] This patent application claims priority to Italian Patent Application No. 102023000019227, filed on September 19, 2023, the entire disclosure of which is incorporated herein by reference. Technical Field

[0003] This invention relates to an in vitro method for diagnosing or assessing the progression of an individual's disease or assessing an individual's response to treatment, or assessing the presence of therapeutic targets in an organic body fluid sample, and determining circulating disease burden (CDL) parameters, which are functions of the amount of target cells, target cell debris, circulating proteins, RNA, and DNA, the latter three collectively referred to as circulating biomarkers. Background Technology

[0004] Cancer development begins with a primary tumor in the body. These tumors grow locally and, after a period of time, begin to spread to more distant locations. During this evolution, cancer cells either acquire or lose genes that allow them to migrate and grow in distant organs. Most patients die from metastasis, either because the metastasis suppresses vital functions or because the cancer's growth throughout the body leads to organ failure. At each stage of this process, different issues need to be addressed to make the best treatment decisions.

[0005] First, cancer must be diagnosed, then a treatment plan should be selected, and the success of the chosen plan should be evaluated. In selecting a treatment plan, the risk of disease progression should be considered whether it is high enough to justify treatments with severe side effects. Furthermore, the potential effectiveness of targeted therapy should be assessed.

[0006] Information from circulating tumor cells (CTCs) can be extremely helpful in this decision-making process. However, CTCs are very rare cells, and even in patients with advanced disease where metastasis is already underway, their counts can be zero. In this regard, tumor-derived extracellular vesicles (tdEVs) have been found to play an important role in homeostasis and oncogenesis, and can provide information related to the presence, progression, and prognosis of tumors.

[0007] In principle, CTC counts are considered more reliable due to their very low background counts, but tdEV can be used as a supplementary evaluation when CTC counts are low.

[0008] As a real-world example, most patients diagnosed with prostate cancer or with metastatic disease initially respond to androgen deprivation but eventually progress to castration-resistant prostate cancer (CRPC) and death. In recent years, numerous new drugs have been introduced or are under development for patients with mCRPC. A major challenge in evaluating the effectiveness of these new drugs is adequately assessing treatment response, as many patients have disease confined to the bone, making them unsuitable for evaluation by RECIST and bone scans, and PSA changes are unreliable before 12–16 weeks of treatment. In multivariate analyses, CTC measurement has shown to be an excellent biomarker for assessing prognosis and disease response in mCRPC. CTC load is highly correlated with overall survival. To simplify the process using the FDA-approved CellSearch CTC assay (Menarini Silicon Biosystems), mCRPC patients were divided into those with 5 or more CTCs (poor) or fewer than 5 CTCs per 7.5 ml of blood. Response to therapy was assessed after 3–6 weeks of treatment by a change from poor to good. Several methods have been investigated for optimally assessing treatment response based on changes in cytokine cytokines (CTCs). A barrier to accurately assessing all patients with mCRPC is the low frequency of CTCs. In the initial prospective CTC study, 38% of patients had fewer than 5 CTCs at all blood draws, making response assessment impossible. It has been observed that CTCs represent only a small fraction of cytokeratin+, DAPI+, CD45- objects in images generated by CellSearch, and objects not assigned as CTCs showed similar prognostic value, suggesting the potential to increase the proportion of patients who can be assessed for response. The open-source image analysis program ACCEPT was used to identify and classify cytokeratin+, DAPI+, CD45- objects, and within these objects, the population was defined as tumor-derived extracellular vesicles (tdEVs). These tdEVs occurred 10-fold more frequently than CTCs and showed a similar relationship with overall survival. Furthermore, among mCRPC patients with 5 or fewer CTCs, those with elevated tdEVs had a 4.7-fold higher risk of death compared to those without elevated tdEVs.

[0009] To date, CTC counts and tdEV counts have been studied separately and have been associated with disease and patient status, respectively.

[0010] In addition, plasma from collected blood can be used to detect other circulating markers, including small tdEVs, circulating tumor DNA, circulating tumor RNA, and circulating proteins, all of which may be related to tumor metastatic activity.

[0011] Therefore, there is a need to find ways to combine information from circulating endothelial cells (CTCs) with information about tdEVs to support clinicians in making treatment decisions, particularly by assessing the presence and development of tumors in patients. This combination can be used not only in oncology but also in other medical fields. For example, the detection of circulating endothelial cells (CECs) and their debris (hereinafter also referred to as endothelial-derived extracellular vesicles, edEVs) can indicate cardiovascular disease. Summary of the Invention

[0012] Therefore, the object of this invention is to provide an improved method to support clinicians in treatment decision-making.

[0013] This objective is achieved by the method defined in claim 1. Attached Figure Description

[0014] Figure 1 A flowchart of a method according to an embodiment of the present invention is shown.

[0015] Figure 2 A graph depicting the combination of CTC count and tdEV count in the BTL parameters is shown.

[0016] Figure 3 A block diagram is shown according to a specific embodiment of the present invention.

[0017] Figure 4 A block diagram is shown according to a specific embodiment of the present invention.

[0018] Figure 5 A block diagram is shown according to a specific embodiment of the present invention.

[0019] Figure 6 A graph showing the overall survival hazard ratio assessed using BTL as a continuous variable is presented, obtained using three different BTL equations.

[0020] Figure 7 A graph showing the continuous risk ratio and the risk ratio (HR) of the operator CTC for the best form of different BTL equations.

[0021] Figure 8This graph shows the proportion and mean intensity of HER2 marker expression (a combination of CTCs and tdEV) in samples from 98 breast cancer patients. The top left graph shows the proportion of positive (>25 higher than control staining) or weakly positive (>1 higher than control staining) CTCs. Samples were sorted by positive proportion; if two or fewer CTCs were present, no results were available. The top right graph shows the TME parameters determined using CTCs and tdEV. Samples were sorted by positive proportion; if two or fewer CTCs and / or tdEV were present, no results were available. The bottom graph shows the mean intensity of the same samples. Detailed Implementation

[0022] The method according to the invention is used to diagnose or assess the progression of an individual's condition or to assess an individual's (patient's) responsiveness / response to treatment.

[0023] Specifically, the following description will focus on the application of this method in circulating tumor cells (CTCs) and circulating cell debris (hereinafter also referred to as tumor-derived extracellular vesicles, tdEVs), as well as other prognostic biomarkers in the patient's organic fluid samples. CTCs and tdEVs indicate metastatic tumors, particularly carcinoma or melanoma. However, this method can also be applied to detect other types of target cells.

[0024] The term "target cells" is intended to indicate the presence and quantity of cells in a patient's organic fluid sample that may indicate the presence of a disease in the patient. In a preferred embodiment, target cells are rare cells that have a low density in the patient's organic fluid. The density ranges for different cell types are listed below.

[0025] For example, the method according to the invention can also be applied to the detection of circulating endothelial cells (CECs) and circulating cell debris (hereinafter also referred to as endothelial-derived extracellular vesicles, edEVs) to indicate cardiovascular disease, infectious diseases, or cancer. The density of CECs in blood ranges from 0 to 200 / mL whole blood, particularly from 0 to 50 / mL whole blood, and even more particularly about 5 / mL whole blood.

[0026] This method can also be applied to the detection of circulating multiple myeloma cells (CMMCs) and circulating cell debris (hereinafter also referred to as extracellular vesicles of multiple myeloma origin, mmEVs). The presence and quantity of CMMCs in a patient's blood can indicate their presence in the bone marrow at the time of diagnosis. Furthermore, their quantity can be monitored to assess myeloma and the efficacy of different therapies against the tumor. The density of CMMCs in the blood ranges from 0 to 20,000 / mL whole blood, particularly from 0 to 2,000 / mL whole blood. Typically, the median before treatment is about 1,000, decreasing to about 5 in response to effective treatment, and increasing back to about 100 upon relapse.

[0027] In some diseases, counting more than one type of target cell and target cell debris is particularly beneficial, such as in the cases of CTC, CEC, and tdEV, edEV for colorectal cancer. Other circulating markers may be present, which can be derived from the same sample with independent prognostic value, including circulating tumor DNA (ctDNA), various types of RNA and protein markers.

[0028] Other types of rare cells to which this method is applicable include fetal cells, respiratory virus cells, and cells collected from cerebrospinal fluid via puncture.

[0029] The term "target cell debris" is intended to indicate phospholipid membrane-bound structures, particularly cellular particles such as exosomes, microvesicles, and apoptotic bodies, whose presence and quantity in a patient's organic fluid sample can indicate the presence of disease in the patient.

[0030] The term “target component” is intended to commonly and generally refer to one or more of “target cell”, “target cell fragment”, and optionally, also to a target circulating biomarker derived from a pathological condition (symptom), such as circulating tumor DNA, circulating RNA, or circulating protein.

[0031] The term "responsiveness" is used to indicate an individual's eligibility for a particular treatment. The term "response" is used to indicate the actual effect of the treatment on an individual.

[0032] The method according to the present invention includes the following steps.

[0033] First, the number of target cells and the number of target cell fragments are counted in an individual's bodily fluid sample. Optionally, the number of target cell clusters may also be counted. For example, occasionally, CTCs in whole blood samples are found not to be isolated, but rather aggregated into clusters. Counting these clusters, and combining this information with relevant data on the number of target cells and the number of target cell fragments, can be additionally helpful in assessing disease or disease progression. Optionally, the presence of circulating tumor DNA (ctDNA), disease-specific RNA, and disease-specific proteins in plasma may also be analyzed.

[0034] As a next step, circulating disease burden (CDL) parameters are determined based on the number of target cells and target cell fragments, and optionally the number of target cell clusters, as well as the concentrations of target ctDNA, RNA, and / or proteins.

[0035] Subsequently, the cyclic disease burden (CDL) parameter was correlated with the presence or progression of an individual’s condition or response to treatment.

[0036] The CDL parameters in patients are functions of the background number of target cells, target cell fragments, and optional target cell clusters relative to the number of target cells, target cell fragments, and optional target cell clusters in normal individuals.

[0037] Figure 1 A flowchart of a method according to an embodiment of the present invention is shown.

[0038] In step S01, the patient's organic body fluid sample, particularly a blood sample, is provided.

[0039] Then, in step S02, the organic body fluid sample is processed to enable selective imaging of the sample and detection of the presence of target cells and target cell fragments (and optionally, target cell clusters) in the sample.

[0040] In detail, in step S02, the organic body fluid sample is enriched for the target component COI (e.g., CTC and tdEV) and labeled.

[0041] For example, to detect CTCs and tdEVs, blood samples can be enriched with immunomagnetic beads to target epithelial cell adhesion molecules (EpCAM).

[0042] The enriched samples of the organic body fluid are labeled with at least one marker that is specific to rare cells and their fragments (e.g., CTCs and tdEVs), and / or optionally, with at least one marker that is specific to cells and / or their fragments other than the rare cells and / or their fragments to be detected.

[0043] The markers can be identified through selective imaging.

[0044] According to a preferred embodiment, the marker is a fluorescent marker that can be imaged by fluorescence imaging. For example, to detect CTCs and their fragments, the enriched sample can be stained with one or more fluorescent markers, particularly DAPI, CD45-APC, and CK-PE.

[0045] However, depending on the specific application, sample processing can vary to allow for the detection of the presence or absence of the target component, for example, by using scattering markers or time-dependent behavior in response to stimuli, or any other known method.

[0046] In step S03, an image of the labeled sample is captured in a manner known per se. Specifically, in this embodiment, the image is a fluorescence image.

[0047] For example, labeled samples can be arranged inside a box, and digital images of the surface of each fluorophore can be recorded.

[0048] According to a preferred embodiment, one or more of steps S02 and S03, particularly both steps S02 and S03, can be performed by the CellSearch System®. Step S03 can be performed by, for example, a DEPArray. TM Other imaging systems can be used.

[0049] Then, at step S04, the captured images are processed by the control unit to determine the presence of target cells and target cell fragments in the images and thus in the patient's bodily fluid sample. The control unit is configured to detect and enumerate the target components in the received images.

[0050] Specifically, the control unit can be configured to detect and enumerate target cells and target cell fragments by using one or more machine learning algorithms trained to identify whether an object stained with markers in an image is a target cell or a target cell fragment.

[0051] exist Figure 1 In the illustrated embodiment, the method further includes step S05, where the control unit provides multiple counters, each counter used for each target component COI that has been detected and enumerated in step S04. For example, the counters could be a CTC counter CTC-CNT and a tdEV counter tdEV-CNT.

[0052] For example, various counters can be provided to users, such as through visual signals on a graphical user interface.

[0053] In step S06, rare counters are combined to establish the value of the Circulating Disease Burden (CDL), and the patient's diagnosis is assessed as a function of the CDL value (step S07). See below for further details. Figure 3 Describe steps S06 and S07 in detail.

[0054] Optionally, according to Figure 1 In an implementation method, after detecting and enumerating the target components in step S04, the method may further include:

[0055] - Calculate (step S08) the biomarker expression for each target component detected in step S04. Biomarker expression for each COI can be calculated in a manner known per se. For example, by calculating the average biomarker expression for CTCs and tdEVs separately, or by determining the number of negative, weakly positive, and strongly positive CTCs and tdEVs. The threshold for the transition from negative to weakly positive to strongly positive is biomarker- and disease-dependent. For example, Her2 expression in breast cancer is negative when at the same level as control cells, weakly positive when 20 more than control cells, and strongly positive when above 20; and

[0056] - Calculate (step S09) the tumor marker expression (TME) value, which is a function of the expression of all individual markers calculated for each COI, along with counters of CTCs and tdEVs. For example, the average marker values ​​of CTCs and tdEVs are weighted together in a manner dependent on the relative frequencies of CTCs and tdEVs, or the proportions of negative CTCs and tdEVs, weakly stained CTCs and tdEVs, and strongly positive CTCs and tdEVs are weighted in a manner dependent on the relative frequencies of CTCs and tdEVs.

[0057] The following text is for reference only. Figure 4 The implementation details provide further details regarding steps S08 and S09.

[0058] TME can be used to assess whether a patient is eligible for targeted therapy.

[0059] According to embodiments, the method of the present invention may include additional steps for analyzing ctDNA, RNA, and / or proteins. For example, in step S50, plasma may be separated from the cellular portion of the body fluid obtained in step S01. However, in addition to the tube from which blood is drawn for CTC analysis, another tube of blood may be collected.

[0060] At step S51, ctDNA, RNA, and / or proteins are analyzed using an assay optimized for each target. For example, plasma samples can be centrifuged at high centrifugation (20,000 xg) to remove any cellular material, and next-generation sequencing can be applied to the supernatant from this process to identify circulating DNA with typical tumor mutations. The extracted concentrations of circulating proteins, DNA, and / or RNA can then be used in step S06 in combination with a cell counter that provides CDL.

[0061] The method according to the present invention includes:

[0062] - The sample is labeled with at least one marker specific to target cells and target cell debris and / or optionally with at least one marker specific to cells and / or cell debris other than target cells and / or target cell debris, said marker being identifiable by selective imaging.

[0063] - Obtain at least one image of a labeled sample, wherein the image includes an object stained with at least one marker to be classified;

[0064] - Classify the objects stained with the markers; and

[0065] - Based on the classification results of the objects stained by the marker, determine whether the objects stained by the marker are target cells or target cell fragments.

[0066] According to this method, counting the number of target cells and / or the number of target cell fragments and / or the number of target cell clusters in a sample includes:

[0067] In response to each marker-stained object identified as a target cell, increment the target cell counter (CTC-CNT), and / or

[0068] In response to each marker-stained object identified as target cell debris, increment the target cell debris counter (tdEV-CNT), and / or

[0069] Optionally, in response to each marker-stained object identified as a target cell cluster, the target cell cluster counter (CTCcl-CNT) is incremented.

[0070] The classification step can be performed by running one or more classification-based machine learning algorithms, which are trained to identify whether the stained objects in the image contain target cells or target cell fragments.

[0071] Preferably, the CDL parameter is a function of the difference between the values ​​of the CTC counters CTC-CNT and their relative background levels in normal individuals.

[0072] In a preferred embodiment, the target cells are CTCs, the fragments are tdEVs, and the clusters are CTC clusters. In this case, these elements can be labeled using cytokeratin and CD45 antibodies, as well as DAPI staining. CTCs and clusters will be positive for cytokeratin and DAPI, but negative for CD45. tdEVs will be positive for cytokeratin, but negative for CD45.

[0073] Target cells, debris, and clusters can be counted, for example, using the Cellsearch system (Menarini Silicon Biosystems), which detects, enriches, and enumerates circulating tumor cells or other rare cell types from whole blood in a standardized and reproducible workflow. Target cell and debris counting can also be performed using other imaging systems such as DEPArray. TM (Menarini Silicon Biosystems) is used for this purpose.

[0074] The value of the cyclic disease burden parameter CDL can be a linear combination of the values ​​of the CTC counter CTC-CNT and the tdEV counter tdEV-CNT.

[0075] The circulatory disease burden parameter CDL is a function of the difference between the value of the CTC counter CTC-CNT and its relative background level in healthy patients, as shown below:

[0076] CDL=f(CTC - CTC BG tdEV - tdEV BG ),in

[0077] CTC is the number of target cells in an individual.

[0078] CTC BG It is the number of target cells in a normal individual.

[0079] tdEV is the number of target cell fragments in an individual.

[0080] tdEV BG It is the number of target cell fragments in a normal individual.

[0081] Choose any location

[0082] CDL=f(CTC - CTC BG tdEV - tdEV BG CTC clust - CTC clust,BG ),in

[0083] CTC clust It refers to the number of target cell clusters in an individual.

[0084] CTC clust,BG It is the number of target cell clusters in a normal individual.

[0085] In a preferred embodiment, the cyclic disease burden (CDL) parameter ranges from a minimum (e.g., 0) to a maximum (e.g., 1), as a function of the values ​​of the CTC counter CTC-CNT and / or the tdEV counter tdEV-CNT, wherein the risk of disease increases monotonically with the value of the CDL.

[0086] Figure 2 A graph illustrating this possible calculation of CDL is shown, taking into account the combination of CTC count and tdEV count. Therefore, the CDL parameter is renamed "Hematologic Tumor Burden" (BTL) and ranges from 0 (low disease risk) to 1 (high-risk disease).

[0087] Low concentrations of CTCs limit the reliability of CTC-based diagnosis, making treatment selection and efficacy determination impossible in patient groups where CTCs are not detected prior to treatment. Metastatic patients almost certainly have CTCs, but the concentrations may be too low to be detected in a single blood tube. However, most CTCs will not extravasate and may therefore disintegrate at metastatic sites or in the circulation, resulting in tdEV. Enhancing CTC-based diagnosis with tdEV can expand the number of patients who can be diagnosed, but two measures providing the same procedural means of outcome will be more difficult to interpret. The aim of BTL is to integrate CTCs and tdEV (and other possible indicators) into a single measure above a larger proportion of patients in the background.

[0088] BTL can be applied, for example, to a) assess the risk or probability of a patient having metastases, to b) assist in the selection of targeted therapies, and to c) monitor the effect of treatment on metastatic activity.

[0089] The table below summarizes clinical problems, current standard methods, and potential approaches to improve diagnosis using CTC and tdEV.

[0090]

[0091] Referring to the above applications, in order to provide the probability that a patient's CDL is higher than the background CDL (application a), Figure 3 This requires modeling the background BTL or the background CTC and tdEV. This means representing the background BTL as a random variable, or representing the CTC and tdEV as random variables.

[0092] In detail, according to Figure 3 The implementation method is collected in Figure 1 The multiple counters determined at step S05 (steps S10_1, ..., S10_N) are, for example, determined by the processing unit. Each counter indicates the number of corresponding COIs.

[0093] Then, in step S11, the CDL value is determined based on the collected counters. The CDL value can be calculated according to any equation reported below. In this case, the CDL value can be referred to as the cyclic disease burden (CDL).

[0094] Then, in step S12, using an applicable statistical distribution, such as a combined Poisson distribution or a normal distribution, the probability that the thus obtained CDL value appears in a normal or benign patient is calculated from reference values ​​for benign and / or normal donors. This probability is supplemented by the probability that the patient has circulating tumor-derived material and therefore metastatic tumors. For example, a probability higher than 0.8 would require additional diagnostic methods, such as more extensive imaging, tissue biopsy, or similar procedures.

[0095] At step S13, a signal indicating the calculated probability is provided at the output. For example, the signal can be provided to the user through a graphical user interface or any other known interface.

[0096] To determine the marker expression (apply b), such as Figure 4 In addition to CDL parameters, tumor marker expression (TME) parameters can also be used.

[0097] In this specific embodiment, the method according to the present invention further includes the following steps:

[0098] - Evaluate the expression of at least one biomarker, said biomarker being a target for treating a disease in target cells and / or target cell debris;

[0099] - Determine tumor marker expression (TME) parameters based on the assessed biomarker expression;

[0100] - Correlate TME parameters with an individual's response to treatment.

[0101] In detail, according to Figure 4 Implementation method, retrieval Figure 1 The expression of multiple markers in step S08 (steps S14_1, ..., S_14N).

[0102] Then, in step S15, tumor marker expression is determined based on the expression of all markers collected from the COI. Tumor marker expression can be calculated using the equations described below for TME.

[0103] In step S16, a signal indicating tumor marker expression is provided upon output. For example, the signal can be provided to the user via a graphical user interface or any other known interface.

[0104] To determine the treatment response (using c), Figure 5(Reaction, disease stability, disease progression) requires modeling previous CTC and tdEV counts as random variables and calculating the probability of extracting posterior counts from the same distribution.

[0105] Specifically, according to Figure 5 In this implementation method, samples of organic bodily fluids from an individual are processed at different times (e.g., Figure 1 (Steps S01-S04). For example, the first sample is processed at time T1 (step S17) and the second sample is processed at time T2, which is greater than T1 (step S20), to assess the patient's progression.

[0106] In response to the processing of both the first and second samples, a COI counter (e.g., a counter for CTC and tdEV, and an optional CTC cluster) is provided (steps S18_1, ... S18_N are for the first sample, and steps S21_1, ... S21_N are for the second sample).

[0107] In step S19, the circulating disease burden (CDL) of the first sample is determined according to the corresponding COI counter [1], such as Figure 3 The steps discussed in step S11.

[0108] In step S22, the circulating disease burden (CDL) of the second sample is determined according to the corresponding COI counter [2], such as Figure 3 The steps discussed in step S11.

[0109] Then, in step S23, the first probability P is calculated. 减少 This reflects the probability that the second measurement result is not derived from the same distribution but from a downwardly shifted distribution, i.e., the probability that the true posterior CDL[2] is smaller relative to the prior CDL[1]. For example, assuming that the CDL follows a joint Poisson distribution and using the prior CDL as a parameter of that distribution, the probability of finding the posterior CDL or any possible lower value of the CDL can be calculated.

[0110] Then, in step S24, the first probability P is... 减少 Compare with a threshold (e.g., 0.8). The threshold can be selected based on the specific application.

[0111] If the first probability is higher than a threshold, a (partial) response is provided at the output of the patient assessment evaluation method (step S25). For example, the (partial) response can be provided to the user via a graphical user interface or any other known interface. The partial response output indicates that the CDL is decreasing and can be interpreted as a therapeutic effect, thereby supporting the maintenance of this treatment where appropriate.

[0112] If the first probability P 减少If the value is not higher than the threshold, then calculate the second probability P. 增长 (Step S26) reflects the probability that the second measurement result is derived from an upwardly shifted distribution rather than the same distribution, i.e., the probability that the true posterior CDL2 is increased relative to the prior CDL1. For example, assuming that CDL follows a joint Poisson distribution, and using the prior CDL as a parameter of that distribution, the probability of finding the posterior CDL or any possible higher value of CDL can be calculated.

[0113] The second probability P 增长 The comparison is made with a threshold (e.g., 0.8) (step S27). The threshold can be selected based on the specific application.

[0114] If the second probability P 增长 If the value exceeds a threshold, progress is provided at the output of the patient assessment evaluation method (step S28). The progress output indicates that the patient's disease has worsened between the time T1 when the first sample was collected and the time T2 when the second sample was collected. This progress can be interpreted as an indication that the treatment has not yet produced any effect, and depending on the treatment, it can be interpreted as supporting a modification to the current treatment.

[0115] If the second probability is not higher than a threshold, disease stability is provided at the output of the patient assessment evaluation method (step S28). Disease stability can be a known type of signal provided to the user via a graphical user interface or any other known interface. Disease stability indicates that the patient's disease has not changed significantly between the time T1 when the first sample was collected and the time T2 when the second sample was collected. Depending on the current treatment goals, this can be interpreted as treatment success or failure, as interpreted by the treating oncologist.

[0116] All of the above applications rely on the most general concept that BTL is a measure of tumor metastatic burden, which, as a result of treatment, should preferably be reduced or stabilized. BTL is a function of an enumeration of CTCs, tdEVs, and CTC clusters found in a blood sample relative to their relative background levels in normal / benign donors (BG), and other characteristics.

[0117] BTL=f(CTC,CTC BG tdEV, tdEV BG CTC cluster, CTC cluster BG )

[0118] BTL has many applications. Some of them will be described below, but are not intended to limit the invention in any way.

[0119] The first application of the BTL concept was to assess the expression of biomarkers for tumor antigens, which are potential therapeutic targets. Some examples of targets are listed in the table below.

[0120]

[0121] The table above is based on the American Society of Clinical Oncology (ASCO, August 2023), where (+) indicates cancer with higher than normal target expression, (-) indicates normal expression, ER: endogenous receptor, PR: progesterone receptor, EGFR: epidermal growth factor receptor, PARP: poly-ADP-ribose polymerase, and gene mutations are indicated by -gm after the gene name or abnormal description. In multiple myeloma, CD38, CD47, CD138, SLAMF7, GPRC5D, FcRH5, and BCMA are highly expressed by malignant plasma cells.

[0122] Tumor antigen expression can be measured on both CTCs and tdEVs. However, false positive identification is most likely to be antigen-negative, thus artificially downplaying the determined expression. Due to the more favorable false positive rate of CTCs compared to tdEVs, and because some of the aforementioned therapeutic targets can be expressed at very low levels, detection of therapeutic targets on CTCs should be considered more reliable than detection on tdEVs. However, expression can be detected on both CTCs and tdEVs and combined to obtain a joint value of the fraction of circulating tumor-derived material expressing the target biomarker. The equation required to combine CTC and tdEV expression into biomarker expression is similar in form to the BTL equation. As with BTL, diagnostic information from CTCs is combined with information from tdEVs whenever information from CTCs alone is insufficient to provide adequate information. This biomarker expression is a standard of tumor susceptibility to a specific treatment. The accuracy of tumor biomarker expression is a function of the number of events assessed and can be estimated using a random noise model. This random noise model can also be used to estimate the blood volume that needs to be assessed to achieve a certain desired accuracy.

[0123] The second application is using blood-to-cell volume (BTL) to determine the amount of CTCs needed for molecular characterization. Molecular characterization of CTCs assesses the molecular distribution and intracellular heterogeneity of cancer, and this characterization is closely related to the number of CTCs from which a molecular distribution can be obtained. Based on morphological features and nuclear staining, the likelihood of obtaining a molecular distribution of CTCs can be provided. The success rate of single-cell isolation procedures can be used to estimate the number of CTCs from which a molecular distribution can be obtained and determine if this is sufficient for the desired distribution. For example, based on a molecular distribution obtained from CTCs, cellular abnormalities and the presence of therapeutic targets can be identified, but this is clearly insufficient to assess tumor heterogeneity or determine if a patient is a candidate for a particular therapy. Conversely, the reliability of the assessment is quite high when 10 molecular CTC distributions with all similar distributions are obtained, but when they are all abnormal, 10 different distributions may not be enough to obtain a clear picture of the tumor. Because tdEV is associated with CTCs, it is possible to estimate how much blood needs to be studied to obtain a sufficient number of CTCs for molecular characterization. Depending on the required blood volume, this can be achieved through regular blood draws or diagnostic leukoablation (DLA). This extrapolation method requires estimating the true CTC concentration using confidence intervals. When CTC counts are low (0, 1, 2, etc.), the relationship between CTC and tdEV concentrations for estimating the true CTC concentration can be derived from the BTL. For molecular characterization, cell isolation and nucleic acid amplification are required; the better the cells look, the higher the chance of successful isolation and characterization. Such features can be extracted from the derived images and morphological characteristics, making it possible to provide the probability of success. The distribution of cells that look very good versus those that look very poor may differ between cancer and treatment. Knowledge of this distribution can be incorporated into the equations and can provide an estimate of how much blood is needed for successful molecular characterization.

[0124] Another application is the identification and counting of blood myeloma burden (BML) based on CMMC and mmEV.

[0125] In healthy individuals, epithelial cells do not circulate in the blood. The presence of epithelial cells in the blood indicates a pathological condition, and in most cases, these epithelial cells originate from a tumor (cancer) that sheds tumor cells into the blood, which can lead to extravasation and distant metastasis. In contrast, plasma cells are present in the blood, albeit in smaller numbers, making the detection of malignant plasma cells (CMMCs) more difficult. Differences in antigen expression between normal plasma cells and CMMCs, such as the lack of CD27 and CD19 expression and the monoclonal nature (λ or κ) of CMMCs, can be used to help distinguish between normal and malignant plasma cells. Morphological differences can also help differentiate between the two. The latter can be utilized using methods described for CTCs and tdEVs. After experts classify CMMCs and train a deep learning convolutional neural network (DLCN) to identify CMMCs, a second round of classification can be performed by presenting DLCN objects detected in normal individuals or patients with infectious diseases (where a greater number of normal plasma cells are present) and objects detected in CMMCs. Furthermore, the difference in EV numbers derived from normal plasma cells and myeloma cells (mmEVs) can help increase the sensitivity and specificity of CMMC detection. Additionally, similar to BTL, BML can be derived from both CMMC and mmEV counts.

[0126] Another application is blood endothelial load (BEL) based on the identification and counting of CECs and edEVs. Endothelial cells and extracellular vesicles derived from endothelial cells can be detected in the blood of healthy individuals, and are often detected in higher quantities in patients with cardiovascular disease (heart attack), infectious diseases, and cancer. Endothelial cells line the entire vascular system and can be released into the circulation from all locations as CECs and edEVs. The origin and cause of CEC and edEV release can vary greatly in cancer, cardiovascular disease, infectious diseases, and healthy donors. The morphology of CECs is highly heterogeneous, but there is no precise classification of the origin of CECs and the cause of their release. Although a DLCN can be trained on objects classified as CECs and edEVs by experts, the DLCN does not identify their origin. If CECs and edEVs are identified in the first round of DLCN in healthy donors, patients with cardiovascular disease, infectious diseases, and cancer, then a second round of classification can now produce a classification that identifies the origin of CECs and edEVs. As in other instances, the resulting values ​​can be reduced to a blood endothelial load from which the severity of cancer, cardiovascular disease, and / or infection can be deduced.

[0127] Example

[0128] Example 1 - BTL Equations

[0129] Several equations were tested. Three of them are explained below.

[0130] The main considerations based on the equation are as follows.

[0131] 1) Although CTC and tdEV are associated with the same pathology, CTC is more important than tdEV.

[0132] 2) CTC and tdEV are correlated, but tdEV is relatively more accurate at low CTC concentrations, while it is relatively less accurate at high CTC concentrations.

[0133] 3) The estimation of the confidence interval for CTC concentration is the basis for any extrapolation method for the required sample size for the desired molecular analysis. BTL allows the use of tdEV to estimate such a confidence interval.

[0134] A simpler form of BTL is a linear combination of the CTC and tdEV counts, as shown in Equation 1. Here, both counts are logarithmically transformed to reduce the impact of large outliers, and the logarithm of 0 is set to -1. Additionally, the coefficient "a" balances the contribution of tdEV relative to CTC.

[0135] (Equation 1)

[0136] A more advanced form of the BTL equation is based on the observation that CTC and tdEV are highly correlated, therefore a high tdEV count implies a higher probability that the CTC count is underestimated. This is represented in Equation 2, where the sigmoid function in parentheses can be simplified if necessary.

[0137] (Equation 2)

[0138] When the CTC count is high, another advanced form of the equation explicitly reduces the effect of the tdEV count.

[0139] (Equation 3)

[0140] Another advanced form of the BTL equation utilizes a log-log function optimized in previous studies, where the shape and displacement parameters of CTC and tdEV are obtained separately by fitting distributions, as described in [Coumans, FA, et al., Challenges in the enumeration and phenotyping of CTC. Clin Cancer Res, 2012.18(20): p. 5711-8.].

[0141] (Equation 4)

[0142] in,

[0143] in,

[0144] For all these equations, the CTC / tdEV count can be an absolute count, or an absolute count minus a certain background count level.

[0145] As can be seen, the CDL parameters can be functions of the number of target cells and the number of target cell fragments, including transformations of the number of target cells and / or the number of target cell fragments. This transformation can be any nonlinear function suitable for a right-skewed distribution, including square root, cube root, logarithmic, or power transformations, or combinations thereof.

[0146] Other variations of equations 1-3 include, but are not limited to:

[0147]

[0148]

[0149]

[0150]

[0151]

[0152] It can be noted that some of the equations above also consider other circulating biomarkers related to pathological conditions, such as circulating tumor DNA mutations (ctDNA). 突变 ), circulating RNA (exRNA) pCA3 ) and protein (PSA).

[0153] There are many variations of Equations 1-3, but Equations 1-3 will be used here to illustrate the potential of BTL.

[0154] Any of these functions can be scaled and centered to achieve a range that is most easily interpreted by clinicians, such as from 0 to 1, or from 0 to 100.

[0155] Example 2 - Evaluation of the BTL Equation in Cancer Research

[0156] BTL was applied to study archives from metastatic prostate cancer (IMMC38), metastatic colorectal cancer (CAIRO-2), and metastatic and non-metastatic breast cancer (IMMC01 and IMMC26, respectively). Background was not subtracted from the tdEV count in this assessment. Values ​​of “α” were assessed from 0.1 to 10, indicating the weight of tdEV in the BTL from too low to too high. BTL was used as a continuous variable to assess the hazard ratio for overall survival, but the hazard ratio was scaled to the interquartile range to account for scaling differences. Results were presented in... Figure 6 The data shows the danger ratio relative to operator / manual CTC.

[0157] The following should be noted.

[0158] - For all equations, there exists a range of coefficient "a" values, for which the prognostic value of BTL is higher than that of operator CTC (OP-CTC).

[0159] - When tdEV contributes too much, the BTL prognosis decreases. This is to be expected from the assumption that CTC is more important than tdEV.

[0160] Studies with overall lower CTC concentrations (non-metastatic breast and metastatic colorectal) showed greater benefits from BTL.

[0161] Figure 7 The optimal form of Equations 1-3 is shown for the continuous hazard ratio and the operator's CTC hazard ratio (HR). Vertical markers represent HRs, and horizontal bars represent the 95% confidence interval (CI) for HRs. It should be noted that the 95% CI is influenced by study data (individual patient survival and review) and BTL characteristics. It is reasonable to expect study data to dominate the confidence interval.

[0162] The HR is the change in risk due to an increase of 1 in the predictor variable. Therefore, even if a simple scaling change does not alter the prognostic value of the CTC count, the equation BTL = 0.5 × CTC will have a higher HR. To avoid this misconception, the HR shown here is a continuous hazard ratio, scaled to the interquartile range for metastatic studies and to the 10–90 percentile for non-metastatic studies. With this scaling, the scaled equation above has the same HR as the unscaled version.

[0163] The main impact of BTL is to broaden the coverage of patients with low CTC counts. When determining hazard ratios on dichotomous data (i.e., splitting at values ​​close to the median), BTL is expected to have slightly lower hazard ratios because the noise in BTL will be higher than that in individual CTCs.

[0164] The above results demonstrate the potential of CDL parameters for application in tumors.

[0165] Example 3 - TME Equation

[0166] Tumor marker expression (TME) can indicate the tumor marker expression score (TMEF), which is calculated as follows:

[0167]

[0168] Among them, MEF⁺ RC It is the fraction of target cells (RC) expressing positive biomarkers, MEF⁺ FR is the fraction of target cell fragments (FRs) expressing positive markers, and a is a coefficient that balances the contribution of target cell fragments relative to the target cell.

[0169] For example, referring to Her2+ biomarkers, the tumor marker expression score TMEF could be:

[0170]

[0171]

[0172] in, Frac Her2+ is the CTC / tdEV score of expression with an expression level higher than the threshold applicable to Her2 marker expression, and ctDNA. TOP2A The copy number variation of TOP2A was determined from circulating tumor DNA.

[0173] Tumor marker expression (TME) can indicate tumor marker expression scores, or serve as a substitute for them, and can also indicate tumor marker expression levels (TMEL). Tumor marker expression levels can be calculated as follows:

[0174]

[0175] Among them, MEL RC It is the level of marker expression intensity on target cells (RC), MEL FR It is the level of marker expression intensity on target cell debris (FR), N RC It is the number of target cells, N FR is the number of target cell fragments, and 'a' is a coefficient that balances the contribution of target cell fragments relative to the target cell. The expression for coefficient 'a' is merely an example and may differ from the value shown.

[0176] In other words, TMEL can be considered a weighted average expression of biomarkers.

[0177] For example, referencing Her2+ biomarkers, the tumor marker expression level TMEL could be:

[0178]

[0179] .

[0180] Combining information from multiple sources can fill information gaps. Tumor marker expression is increasingly used to guide treatment. However, in most cases, the only available tissue material for determining expression is collected at diagnosis, sometimes more than 10 years ago. The tumor has progressed and no longer expresses the same markers as before. Fluid biopsies in the form of circulating material can provide a current assessment of marker expression. Circulating tumor cells (CTCs) are of particular interest because they allow for the identification of proteins, DNA, and RNA. However, this assay is significantly hampered if only a small number of CTCs are found in a single tube of blood. As with CDL, combining information from multiple sources into a single parameter can fill gaps when needed. This will be demonstrated here in this article regarding HER2 expression on CTCs and tdEVs. Other targets and other circulating markers are also possible. Figure 8 The scores and mean intensities of HER2 marker expression, targeting only CTC and tumor marker expression (TME) parameters, are shown in samples from 98 breast cancer patients.

[0181] Most patients with CTCs who are HER2-positive are sensitive to trastuzumab, and most patients with CTCs who are at least weakly positive are sensitive to other HER2-targeted therapies. Figure 8 The left-middle figure shows that if only CTCs are assessed, HER2 biomarker expression (both score and mean intensity) is unavailable in 40 of the 98 patient samples because fewer than 3 CTCs were present (i.e., not enough CTCs for reliable measurement). On the other hand, if HER2 expression measurements of CTCs and tdEVs are combined into a single TME parameter, HER2 biomarker expression cannot be provided in only 11 of the 98 samples, where fewer than 3 combined CTCs and / or tdEVs were present. Figure 8 As shown in the right figure.

[0182] The method of the present invention may further include a sorting step that allows the separation of cells from an organic body fluid sample that have been identified as target cells (e.g., rare cells, particularly CTCs or fetal cells) or target cell debris (e.g., circulating cell debris, particularly tdEVs) for downstream processing, including molecular analysis of the target cells (e.g., whole genome amplification, low-pass sequencing, next-generation sequencing, copy number variation, FISH, RT-PCR, qPCR, dPCR, STR analysis, mutation detection, gene expression analysis, etc.) or for the culture and proliferation of the target cells. The sorting step may be performed by a DEPArray. TM Alternatively, it can be performed using other known image-based cell sorters.

Claims

1. An in vitro method for diagnosing or assessing the progression of an individual's condition or evaluating an individual's response to treatment, comprising: - The organismal fluid sample of the individual is labeled with at least one marker specific to target cells and target cell debris, the marker being identifiable by selective imaging. - Obtain at least one image of a labeled sample, wherein the image includes an object stained with at least one marker to be classified; - Classify the objects stained with the markers; and - Based on the classification results of the objects stained with the markers, determine whether the objects stained with the markers are target cells or target cell debris. - Count the number of target cells and the number of target cell fragments, wherein the target cells are circulating tumor cells (CTCs) and the target cell fragments are tumor-derived extracellular vesicles (tdEVs); or, the target cells are circulating endothelial cells (CECs) and the target cell fragments are endothelial-derived extracellular vesicles (edEVs); or, the target cells are circulating multiple myeloma cells (CMMCs) and the target cell fragments are multiple myeloma extracellular vesicles (mmEVs). The counting of the number of target cells and the number of target cell fragments includes: - In response to each object identified as a marker stained for the target cell, increment the target cell counter (CI-CNT), and / or - In response to each object stained with a marker identified as a target cell fragment, increment the target cell fragment counter (CFI-CNT); - Determine the circulating disease burden (CDL) parameter, which is a function of the difference between the values ​​of the target cell counter (CI-CNT) and / or the target cell fragmentation counter (CFI-CNT) and their relative background levels in normal individuals, wherein the function is: CDL = f(CTC - CTC BG , tdEV - tdEV BG ),in, CTC is the number of target cells in the individual. CTC BG It is the number of target cells in the normal individual. tdEV is the number of target cell fragments in the individual. tdEV BG It is the number of target cell fragments in the normal individual; and - Correlate the cyclic disease burden (CDL) parameter with the presence or progression of the individual's condition or the individual's responsiveness / response to treatment.

2. The method according to claim 1, wherein: - The step of counting the number of target cells and the number of target cell fragments further includes: counting the number of target cell clusters, wherein, in response to each marker-stained object identified as a target cell cluster, a target cell cluster counter (CTCcl-CNT) is incremented; and - The Circulating Disease Burden (CDL) parameter is also a function of the number of the target cell clusters.

3. The method according to claim 1 or 2, wherein: - The step of counting the number of target cells and the number of target cell fragments further includes: counting at least one circulating biomarker associated with the disease; and - The circulatory disease burden (CDL) parameter is also a function of at least one circulatory biomarker associated with the condition.

4. The method according to any one of claims 1 to 3, wherein, The step of labeling the sample further includes labeling the sample with at least one marker that is specific to cells and / or cell fragments other than the target cells and / or the target cell fragments.

5. The method according to any one of the preceding claims, wherein, The cyclic disease burden (CDL) parameter ranges from a minimum (e.g., 0) to a maximum (e.g., 1), and is a monotonic function of the number of target cells (e.g., CTCs), the number of target cell fragments (e.g., tdEVs), and optionally the number of target cell clusters.

6. The method according to any one of the preceding claims, in, The CDL parameters depend on the transformation of the number of target cells and / or the transformation of the number of target cell fragments, and optionally, the transformation of the number of target cell clusters, wherein the transformation is, for example, a root function, a logarithmic function, a power function, or a combination thereof.

7. The method according to any one of claims 1 to 6, wherein, Where, N RC It is the number of target cells, N FR is the number of target cell fragments, and 'a' is a coefficient that balances the contribution of rare cell fragments relative to the target cell.

8. The method according to any one of claims 1 to 6, wherein, The CDL parameters are an S-shaped function depending on the number of rare cell fragments.

9. The method according to any one of claims 1 to 6, wherein, Where, N RC It is the number of target cells, N FR is the number of target cell fragments, and 'a' is a coefficient that balances the contribution of target cell fragments relative to the target cell.

10. The method according to any one of claims 1 to 6, wherein, Where, N RC It is the number of target cells, N FR is the number of target cell fragments, and 'a' is a coefficient that balances the contribution of target cell fragments relative to the target cell.

11. The method according to any one of claims 1 to 6, wherein, in, in, Where, N RC It is the number of target cells, N FR It is the number of target cell fragments, and llg is a log-logistic function having shape and displacement parameters of CTC and tdEV respectively obtained from fitting the corresponding distributions.

12. The method according to any one of claims 1 to 11, wherein, The CDL parameters depend on the number of target cell fragments scaled according to the number of target cells.

13. The method according to any one of the preceding claims, wherein, The condition described is a tumor.

14. The method according to any one of the preceding claims, wherein, - When the target cells are circulating tumor cells (CTCs) and the debris is tumor-derived extracellular vesicles (tdEVs), the condition is a tumor or melanoma; or - When the target cells are circulating multiple myeloma cells (CMMC) and the fragments are multiple myeloma extracellular vesicles (mmEVs), the condition is myeloma; or - When the target cells are circulating endothelial cells (CECs) and the fragments are endothelial-derived extracellular vesicles (edEVs), the condition is a cardiovascular disease.

15. The method according to any one of claims 1 to 13, wherein, The target cells are CTCs and CECs, and the fragments are tdEVs and edEVs, and the disease is colorectal cancer.

16. The method of claim 1, further comprising the step of: - Evaluate the expression of at least one biomarker in the target cells and / or fragments of the target cells, said biomarker being a target for treating said condition; - Determine tumor marker expression (TME) parameters based on the expression of the assessed biomarkers; - Correlate TME parameters with an individual's response to treatment.

17. The method according to claim 16, wherein, The biomarkers were selected from Her2, PSMA, MUC1, EGFR, ER, PR, KI67, PDL1, RAS, RAF, ROS1, KRAS, BRAF, MET, RET and PARP.

18. The method according to claim 16 or 17, wherein, The tumor marker expression (TME) indicates the tumor marker expression score (TMEF), and the tumor marker expression score is: Among them, MEF⁺ RC It is the fraction of target cells (RC) expressing positive biomarkers, MEF⁺ FR is the fraction of target cell fragments (FRs) expressing positive markers, and a is a coefficient that balances the contribution of target cell fragments relative to the target cell.

19. The method according to any one of claims 15 to 17, wherein, The tumor marker expression (TME) indicates the tumor marker expression level (TMEL), and the tumor marker expression level is: Among them, MEL RC It is the level of marker expression intensity on target cells (RC), MEL FR It is the level of marker expression intensity on target cell debris (FR), N RC It is the number of target cells, N FR is the number of target cell fragments, and 'a' is a coefficient that balances the contribution of target cell fragments relative to the target cell.

20. The method according to any one of claims 3 to 19, wherein, The circulating biomarker associated with the condition is at least one of circulating proteins, RNA, DNA, or combinations thereof.

21. The method according to any one of the preceding claims further includes sorting at least one of the cells identified as target cells or target cell fragments by an image-based cell sorter.

22. The method according to the preceding claim, wherein, The sorting step is used for downstream molecular analysis or for the culture and proliferation of the sorted cells.

Images