In vitro method and uses thereof for the differential diagnosis of CRC (colorectal cancer) and / or colorectal adenomas

Abstract

The present invention refers to an in vitro method for identifying patients at risk of suffering from colorectal cancer and / or colorectal adenomas, based on measuring the expression profile of some specific scores identified throughout the present invention.

Description

[0001] In vitro method and uses thereof for the differential diagnosis of CRC (colorectal cancer) and / or colorectal adenomas.

[0002] TECHNICAL FIELD OF THE INVENTION

[0003] The present invention refers to an in vitro method for identifying patients at risk of suffering from colorectal cancer and / or colorectal adenomas based on measuring the expression profile or level of some specific markers taught in the present invention which are altered in patients suffering from said diseases.

[0004] BACKGROUND OF THE INVENTION

[0005] Colorectal cancer (also known as colon cancer, rectal cancer, or bowel cancer) is the development of cancer in the colon or rectum (parts of the large intestine). The vast majority of colorectal cancers are adenocarcinomas. This is because the colon has numerous glands within the tissue. When these glands undergo a number of changes at the genetic level, they proceed in a predictable manner as they move from benign to an invasive, malignant colon cancer. The adenomas of the colon, also called adenomatous polyps, are a benign version of the malignant adenocarcinomas but still with malignant potential if not removed (they are usually removed because of their tendency to become malignant and to lead to colon cancer).

[0006] Screening is an effective way for preventing and decreasing deaths from colorectal cancer and is recommended starting from the age of 50 to 75. The best known and most frequently used screening test for colorectal cancer is called Fecal Immunochemical Test (FIT). FIT detects blood in the stool samples which can be a sign of pre-cancer or cancer. If abnormal results are obtained, usually a colonoscopy is recommended which allows the physician to look at the inside of the colon and rectum to make a diagnosis. During colonoscopy, small polyps may be removed if found. If a large polyp or tumor is found, a biopsy may be performed to check if it is cancerous. The gastroenterologist uses a colonoscopy to find and remove these adenomas and polyps to prevent them from continuing to acquire genetic changes that will lead to an invasive adenocarcinoma.Although, as explained above, FIT is nowadays used for screening colorectal cancer, it is important to note that FIT offers a low sensitivity for adenomas (around 20-30% depending on literature) which means that most of said kind of patients can be wrongly classified as not having the disease. Consequently, FIT is not able to identify adenomas due to its low sensitivity. Moreover, since FIT uses stool samples, it offers a low compliance (less than 50%). On the other hand, the colonoscopy is an invasive technique wherein the most severe complication generally is the gastrointestinal perforation (1 % of the cases). Moreover, colonoscopy is nowadays a procedure involving anesthesia, and the laxatives which are usually administered during the bowel preparation for colonoscopy are associated with several digestive problems.

[0007] The present invention offers a clear solution to the problems cited above because it is focused on an in vitro method performed from a blood sample for identifying or screening human subjects at risk of suffering from colorectal cancer or colorectal adenomas.

[0008] BRIEF DESCRIPTION OF THE FIGURES

[0009] Fig. 1. This figure shows how different combinations of biomarkers of the present invention are capable of differentiating healthy donors from patients with colon cancer with a sensitivity between 85.71% - 76.19% and a specificity between 89.47% and 78.95%.

[0010] Fig. 2. This figure shows how different combinations of biomarkers of the present invention are capable of differentiating healthy donors from patients with colon adenoma with a sensitivity of 93.75% and a specificity of 78.95%.

[0011] Fig. 3. This figure shows how different combinations of biomarkers of the present invention are capable of differentiating patients with colon adenoma from patients with colon cancer with a sensitivity of 100% and a specificity of 100%.

[0012] Fig. 4. This figure shows how the different populations of cells defined in the following invention are identified based on the expression of different markers. The definition of all these populations is well established in the state of the art as shown in (Park, et al., Cytometry Part A, 2020, doi: 10.1002 / cyto.a.24213, Figure 1). Briefly, from live (DEAD negative) mononuclear immunological cells, four populations are defined based on theexpression of the CD3 and CD56 markers: CD3positive CD56negative (CD3+CD56-), CD3negative CD56negative (CD3-CD56-), CD3negative CD56positive (CD3-CD56+) and CD3positive CD56positive (CD3+CD56+). The CD3+CD56+ population is defined as NKT population. The CD3-CD56+ population is defined as natural killer cells (NK). Inside this NK cells population, two different populations based on two different levels of expression of the CD56 marker can be identified as explained in figure 5, defined as NK CD56dimand NK CD56bright. Inside the CD3+CD56-, two different populations based on the expression of the CD4 and CD8 markers can be identified, the CD4negative CD8positive (CD3+CD8+) is defined as CD8 lymphocytes, and the CD4positive CD8negative (identified as CD4 in the figure). Inside this CD4 population, based on the expression of the CD127 and CD25 markers, the CD127positive CD25negative population is defined as CD4 lymphocytes (CD3+CD4+).

[0013] Fig. 5. This figure illustrates how either "bright levels of CD56" or "dim / intermediate levels of CD56" are defined. From the total population of mononuclear immunological cells, inside the CD3negative CD56positive population, two different populations based on the expression level of CD56 can be defined, one expressing "dim / intermediate levels of CD56" and another one expression "bright levels of CD56". "dim / intermediate levels of CD56" refers to the population that express CD56 at a level over the expression observed in cells in the absence of antibody against CD56, but below the population showing the brightest levels. "Bright levels of CD56" refers to the population that express the highest CD56 expression, being then defined as the population with bright levels of CD56. These two populations based on two different levels of CD56 expression, inside the CD3 negative population are well-established in the state of the art as shown in (Esteso, et al., Frontiers in Immunology, 2021, doi:10.3389 / fimmu.2021.622995, Figure 1) or (Merkt, et al., Arthritis Research and Therapy, 2016, doi: 10.1186 / sl3075-016-1098-7, Figure 3).

[0014] DETAILED DESCRIPTION OF THE INVENTION

[0015] The present invention refers to an in vitro method for identifying patients at risk of suffering from colorectal cancer and / or colorectal adenomas, preferably advancedcolorectal adenomas, based on measuring the expression profile or level, in cells obtained from a blood sample of a human subject in need thereof, of some specific markers taught in the present invention which are altered in patients suffering from said diseases.

[0016] In particular, in the present invention we show different combinations of biomarkers capable of differentiating:

[0017] healthy donors from patients with colon cancer with a sensitivity between 85.71% - 76.19% and a specificity between 89.47% and 78.95%;

[0018] healthy donors from patients with colon adenoma with a sensitivity of 93.75% and a specificity of 78.95%; and

[0019] patients with colon adenoma from patients with colon cancer with a sensitivity of 100% and a specificity of 100%.

[0020] More particularly, these markers can be obtained by determining the CD3 (in particular CD3s), CD8 (in particular CD8a), CD56, CD127, CD25, CD4, and TIM-3 expression in an isolated biological sample, such as blood, preferably peripheral whole blood, comprising mononuclear immunological cells and polyclonally activated mononuclear immunological cells, preferably concanavalin A (ConA) activated mononuclear immunological cells, as measured with flow cytometry, and then calculating one or more scores based on the percentage of expression of each of these markers, or combinations thereof. Some of these obtained markers, of particular relevance for the present invention, can be selected from any one of the following list:

[0021] a. Percentage of NK CD56dimcells

[0022] b. Percentage ofTIM3 cells in NK CD56bright- FOLD

[0023] c. Percentage of NKT (CD56+ CD3+) - FOLD

[0024] d. Percentage of activated NK CD56dimcells

[0025] e. Percentage of TIM3 in activated CD8 lymphocytes (CD3+CD8+) f. Percentage of CD4 lymphocytes (CD3+CD4+) - FOLD

[0026] g. Percentage of TIM3 in CD8 lymphocytes (CD3+CD8+) - FOLDh. Percentage of TIM3 in CD4 lymphocytes (CD3+CD4+) - FOLD

[0027] It is noted that as indicated in figure 4, four populations are defined based on the expression of the CD3 and CD56 markers: CD3positive CD56negative (CD3+CD56-), CD3negative CD56negative (CD3-CD56-), CD3negative CD56positive (CD3-CD56+) and CD3positive CD56positive (CD3+CD56+). The CD3+CD56+ population is defined as NKT population. The CD3-CD56+ population is defined as natural killer cells (NK). Inside this NK cells population, two different populations based on two different levels of expression of the CD56 marker can be identified as explained in figure 5, defined as NK CD56dimand NK CD56bright. Inside the CD3+CD56-, two different populations based on the expression of the CD4 and CD8 markers can be identified, the CD4negative CD8positive (CD3+CD8+) is defined as CD8 lymphocytes, and the CD4positive CD8negative (identified as CD4 in the figure). Inside this CD4 population, based on the expression of the CD127 and CD25 markers, the CD127positive CD25negative population is defined as CD4 lymphocytes (CD3+CD4+).

[0028] As used herein the term "expression" shall be understood as the quantification of the presence of each marker indicated thereof on the surface of mononuclear immune cells based on the fluorescence captured by flow cytometry due to the binding of specific antibodies to these markers. To be considered as "expression", this quantification will be over the quantification detected in cells lacking those specific antibodies.

[0029] As used herein the term "mononuclear immunological cells" shall be understood as those cellular components of the blood that have a single, round nucleus, such as lymphocytes, natural killer cells and monocytes.

[0030] As used herein "CD3 or CD3s" shall be understood as a lymphocytic marker identified by flow cytometry based on the binding of an anti-CD3s antibody. It is noted that CD3s can be also named as "T3E; TCRE; IMD18".

[0031] As used herein "CD8 or CD8a" shall be understood as a lymphocytic marker identified by flow cytometry based on the binding of an anti-CD8a antibody. It is noted that CD8a can be also named as Leu2a; MAL; T8; p32.As used herein "CD127" shall be understood as a lymphocytic marker identified by flow cytometry based on the binding of an anti-CD127 antibody. It is noted that CD127 can be also named as ILRA; IL7RA; CDW127; IMD104; slL-7R; lnc-IL7R; I L7Ralpha; IL-7Ralpha; I L-7R-alpha.

[0032] As used herein "CD25" shall be understood as a lymphocytic marker identified by flow cytometry based on the binding of an anti-CD25 antibody. It is noted that CD25 can be also named as p55; IL2R; IMD41; TCGFR; IDDM10.

[0033] As used herein "CD4" shall be understood as a lymphocytic marker identified by flow cytometry based on the binding of an anti-CD4 antibody. It is noted that CD4 can be also named asT4; IMD79; Leu-3; OKT4D; CD4mut.

[0034] As used herein "CD56" shall be understood as a lymphocytic marker identified by flow cytometry based on the binding of an anti-CD56 antibody. It is noted that CD56 can be also named as NCAM; MSK39.

[0035] As used herein "TIM-3" shall be understood as a lymphocytic marker identified by flow cytometry based on the binding of an anti- TIM-3 antibody. It is noted that TIM-3 can be also named as CD366; KIM-3; SPTCL; TIMD3; Tim-3; TIMD-3; HAVcr-2.

[0036] As used herein "polyclonally activated mononuclear immunological cells" shall be understood as the polyclonal activation of the mononuclear immune cells. It is noted that such activation is preferably carried out by using concanavalin A, by exposing plated mononuclear immunological cells to such polyclonal activation for about 6 hours.

[0037] As used herein "positive population" shall be understood as the group of mononuclear immunological cells where the expression of each of the markers indicated thereof is superior to the one found in the absence of the specific antibodies used to identify those markers.

[0038] As used herein "negative population" shall be understood as the group of mononuclear immunological cells where the expression of each of the markers indicated thereof is comparable or equivalent to the one found in the absence of the specific antibodies used to identify those markers.As used herein "dim / intermediate levels of CD56" shall be understood as the expression of the CD56 marker in mononuclear immune cells at levels ranging over the CD56 negative population and the CD56 bright population. See the brief description of figures 4 and 5 for the specific detail regarding the determination of this term.

[0039] As used herein "bright levels of CD56" shall be understood as the expression of the CD56 marker in mononuclear immune cells at the highest levels, meaning over the levels detected in the population expressing dim / intermediate levels of CD56. See figure 4 and 5 for clarification.

[0040] As used herein "CD56 dim / intermediate population" shall be understood as the population of mononuclear immune cells expressing the CD56 marker at levels ranging over the CD56 negative population and the CD56 bright population. See figure 4 and 5 for clarification.

[0041] As used herein "CD56 bright population" shall be understood as the population of mononuclear immune cells expressing the CD56 marker at the highest levels, meaning over the levels detected in the population expressing dim / intermediate levels of CD56. See figure 4 and 5 for clarification.

[0042] As used herein "NKT population" shall be understood as the population of mononuclear immune cells expressing the CD56 marker and the CD3 marker at the same time. See figure 4 for clarification.

[0043] As used herein the term "score" shall be understood as the parameter generated after subjecting the values of the expression of the markers indicated thereof to a predictive model such as the wald backwards logistic regression model.

[0044] As used herein "concanavalin A (ConA)" shall be understood as the chemical compound utilized in this invention to polyclonally activate mononuclear immunological cells, consisting in a purified glycoprotein isolated from the jack beam whose scientific name is Canavalia ensiformis.

[0045] As used herein "as measured with flow cytometry" shall be understood as the evaluation of the expression of each of the markers indicated thereof in a flow cytometer device.As used herein "activated conditions" shall be understood as the status achieved by mononuclear immunological cells following polyclonal activation, preferably due to the exposition for about 6 hours to concanavalin-a.

[0046] As used herein "not activated conditions" shall be understood as the status achieved by mononuclear immunological cells following their culture for 6 hours without any further stimulation.

[0047] As used herein "frequency of cells expressing ..." shall be understood as the percentage of mononuclear immunological cells showing expression of any of the markers indicated thereof. Meaning the percentage of cells whose expression of each specific marker is detected over the level detected in the absence of the specific antibodies used to identify those markers.

[0048] As used herein "Percentage of NK CD56dimcells" shall be understood as the frequency of cells expressing dim / intermediate levels of CD56, inside the CD56 positive and CD3 negative population, upon not activated conditions. Thus, inside the CD56 positive CD3 negative population (see figure 4), upon not activated conditions:

[0049] "Percentage of NK CD56dimcells" = 100 - Frequency of cells expressing bright levels of CD56

[0050] As used herein "Percentage of TIM3 cells in NK CD56bright- FOLD" shall be understood as the value generated from the division of the frequency of cells expressing positive levels of TIM3 in the cellular population expressing bright levels of CD56, between the activated and not activated conditions. Thus, inside the CD56 positive CD3 negative population expressing bright levels of CD56 (see figure 4):

[0051] "Percentage of TIM3 cells in NK CD56bright- FOLD" = Frequency of cells expressing TIM3 upon activated conditions / Frequency of cells expressing TIM3 upon not activated conditions

[0052] As used herein "Percentage of NKT (CD56+ CD3+) - FOLD" shall be understood as the value generated from the division of the frequency of cells expressing positive levels of both CD56 and CD3 simultaneously, between the activated and not activated conditions. Thus:"Percentage of NKT (CD56+ CD3+) - FOLD" = Frequency of cells expressing both CD56 and CD3 upon activated conditions / Frequency of cells expressing both CD56 and CD3 upon not activated conditions

[0053] As used herein "Percentage of activated NK CD56dimcells" shall be understood as the frequency of cells expressing dim / intermediate levels of CD56, inside the CD56 positive and CD3 negative population, upon activated conditions. Thus, inside the CD56 positive CD3 negative population (see figure above), upon activated conditions:

[0054] "Percentage of NK CD56dimcells" = 100 - Frequency of cells expressing bright levels of CD56

[0055] As used herein "Percentage of TIM3 in activated CD8 lymphocytes (CD3+CD8+)" shall be understood as the frequency of cells expressing positive levels of TIM3 inside the cellular population expressing CD8, upon activated conditions. Thus, inside the CD3 positive CD8 positive population, upon activated conditions:

[0056] "Percentage of TIM3 in activated CD8 lymphocytes (CD3+CD8+)" = Frequency of cells expressing TIM 3

[0057] As used herein "Percentage of CD4 lymphocytes (CD3+CD4+) - FOLD" shall be understood as the value generated from the division of the frequency of cells expressing CD3 but not CD56 multiplied by the frequency of CD4 positive, CD127 positive, CD25 negative cells, divided between 100, between the activated and not activated conditions. Thus:

[0058] "Percentage of CD4 lymphocytes (CD3+CD4+) - FOLD" = [(Frequency of cells expressing CD3 x frequency of cells expressing CD4 and CD127 but not expressing CD25 / 100) upon activated conditions] / [(Frequency of cells expressing CD3 x frequency of cells expressing CD4 and CD127 but not expressing CD25 / 100) upon not activated conditions)]

[0059] As used herein "Percentage of TIM3 in CD8 lymphocytes (CD3+CD8+) - FOLD" shall be understood as the value generated from the division of the frequency of cells expressing positive levels of TIM3 inside the cellular population expressing CD3 and CD8simultaneously, between the activated and not activated conditions. Thus, inside the CD3 positive CD8 positive population:

[0060] "Percentage of TIM3 in CD8 lymphocytes (CD3+CD8+) - FOLD" = Frequency of cells expressing TIM3 upon activated conditions / Frequency of cells expressing TIM3 upon not activated conditions.

[0061] As used herein "Percentage of TIM3 in lymphocytes (CD3+CD4+) - FOLD" shall be understood as the value generated from the division of the frequency of cells expressing positive levels of TIM3 inside the cellular population expressing CD3, CD4 and CD127 simultaneously but not expressing CD25, between the activated and not activated conditions. Thus, inside the CD3 positive, CD4 positive, CD127 positive, CD25 negative population:

[0062] "Percentage of TIM3 in lymphocytes (CD3+CD4+) - FOLD" = Frequency of cells expressing TIM3 upon activated conditions / Frequency of cells expressing TIM3 upon not activated conditions.

[0063] Preferably, particularly useful markers useful in the present invention are the following combination:

[0064] a. Percentage of NK CD56dimcells

[0065] b. Percentage of TIM3 cells in NK CD56bright- FOLD

[0066] c. Percentage of NKT (CD56+ CD3+) - FOLD

[0067] d. Percentage of TIM3 in activated CD8 lymphocytes (CD3+CD8+) e. Percentage of TIM3 in CD8 lymphocytes (CD3+CD8+) - FOLD f. Percentage of CD4 lymphocytes (CD3+CD4+) - FOLD

[0068] as the combination of these markers, as shown in figure 1, is capable of differentiating healthy donors from patients with colorectal cancer.

[0069] Another particularly useful set of markers is the following combination:

[0070] a) Percentage of NK CD56dimcells

[0071] b) Percentage ofTIM3 cells in NK CD56bright- FOLD

[0072] c) Percentage of NKT (CD56+ CD3+) - FOLD

[0073] d) Percentage of activated NK CD56dimcellsas the combination of these scores, as shown in figure 2, is capable of differentiating healthy donors from patients with colon adenoma. In particular, figure 2 shows a discriminatory capacity with a sensitivity of 93.75% and a specificity of 78.95%.

[0074] A yet another particularly useful set of markers is the following combination:

[0075] a. Percentage of activated NK CD56dimcells

[0076] b. Percentage of TIM3 cells in CD8 lymphocytes (CD3+CD8+) - FOLD

[0077] c. Percentage of lymphocytes CD4 (CD3+CD4+) - FOLD

[0078] d. Percentage of TIM3 cells in CD4 lymphocytes (CD3+CD4+) - FOLD

[0079] as the combination of these scores, as shown in figure 3, is capable of differentiating patients with colon adenoma from patients with colon cancer. In particular, figure 3 shows a discriminatory capacity with a sensitivity of 100% and a specificity of 100%. Therefore, in a first aspect, the present invention refers to an in vitro method for determining CD3 (in particular CD3s), CD8 (in particular CD8a), CD56, CD127, CD25, CD4, and TIM-3expression in one or more isolated biological samples of mononuclear immunological cells and / or polyclonally activated mononuclear immunological cells, preferably concanavalin A (ConA) activated mononuclear immunological cells, as measured with flow cytometry.

[0080] In a preferred embodiment, the method of the first aspect further comprises determining at least one or more of the following markers:

[0081] a. Percentage of NK CD56dimcells

[0082] b. Percentage of TIM3 cells in NK CD56bright- FOLD

[0083] c. Percentage of NKT (CD56+ CD3+) - FOLD

[0084] d. Percentage of activated NK CD56dim

[0085] e. Percentage of TIM3 in activated CD8 lymphocytes (CD3+CD8+) f. Percentage of CD4 lymphocytes (CD3+CD4+) - FOLD

[0086] g. Percentage of TIM3 in CD8 lymphocytes (CD3+CD8+) - FOLD h. Percentage of TIM3 in CD4 lymphocytes (CD3+CD4+) - FOLD or any combination therefrom.In another preferred embodiment, the method of the first aspect comprises determining at least the following combination of markers:

[0087] a. Percentage of NK CD56dimcells

[0088] b. Percentage of TIM3 cells in NK CD56bright- FOLD

[0089] c. Percentage of NKT (CD56+ CD3+) - FOLD

[0090] d. Percentage of TIM3 in activated CD8 lymphocytes (CD3+CD8+) e. Percentage of TIM3 in CD8 lymphocytes (CD3+CD8+) - FOLD

[0091] f. Percentage of CD4 lymphocytes (CD3+CD4+) - FOLD

[0092] In another preferred embodiment, the method of the first aspect comprises determining at least the following combination of markers:

[0093] a. Percentage of NK CD56dimcells

[0094] b. Percentage of TIM3 cells in NK CD56bright- FOLD

[0095] c. Percentage of NKT (CD56+ CD3+) - FOLD

[0096] d. Percentage of activated NK CD56dimcells

[0097] In another preferred embodiment, the method of the first aspect comprises determining at least the following combination of scores:

[0098] a. Percentage of activated NK CD56dimcells

[0099] b. Percentage of TIM3 cells in CD8 lymphocytes (CD3+CD8+) - FOLD c. Percentage of CD4 lymphocytes (CD3+CD4+) - FOLD

[0100] d. Percentage of TIM3 cells in CD4 lymphocytes (CD3+CD4+) - FOLD A second aspect of the invention refers to an in vitro method for screening or identifying subjects at risk of suffering from colorectal cancer and / or colorectal adenomas, preferably advanced colorectal adenomas, based on measuring the expression profile or level of CD3 (in particular CD3s), CD8 (in particular CD8a), CD56, CD127, CD25, CD4, and TIM-3 expression in one or more isolated biological samples comprising mononuclear immunological cells and / or polyclonally activated mononuclear immunological cells, preferably concanavalin A (ConA) activated mononuclear immunological cells, as measured with flow cytometry, taken from the subjects. The present invention also refers to an in vitro method for obtaining useful data for the diagnosis of colorectalcancer and / or colorectal adenomas, preferably advanced colorectal adenomas, in a subject, preferably in a human subject.

[0101] In particular, the present invention is based on the discovery that the expression profile or level of CD3 (in particular CD3s), CD8 (in particular CD8a), CD56, CD127, CD25, CD4, and TIM-3 in an isolated biological sample taken from the subjects comprising or consisting of mononuclear immunological cells and / or polyclonally activated mononuclear immunological cells, preferably concanavalin A (ConA) activated mononuclear immunological cells, as preferably measured with flow cytometry, is associated with the risk of suffering from colorectal cancer and / or colorectal adenomas. Moreover, remarkably as shown in figures 1 to 3, the results provided herein show that specific scores related to the expression profile or level of CD3 (in particular CD3s), CD8 (in particular CD8a), CD56, CD127, CD25, CD4, and TIM-3 in an isolated biological sample comprising mononuclear cells and / or preferably concanavalin A (ConA) activated mononuclear cells based on the following markers:

[0102] a. Percentage of NK CD56dimcells

[0103] b. Percentage ofTIM3 cells in NK CD56bright- FOLD

[0104] c. Percentage of NKT (CD56+ CD3+) - FOLD

[0105] d. Percentage of activated NK CD56dimcells

[0106] e. Percentage of TIM3 in activated CD8 lymphocytes (CD3+CD8+) f. Percentage of CD4 lymphocytes (CD3+CD4+) - FOLD

[0107] g. Percentage of TIM3 in CD8 lymphocytes (CD3+CD8+) - FOLD h. Percentage of TIM3 in CD4 lymphocytes (CD3+CD4+) - FOLD or any combination therefrom, are significantly associated with the presence of colorectal adenoma and / or colorectal cancer. In fact, the results obtained with different combinations of these scores are deemed useful for screening for the presence of colorectal adenoma and / or colorectal cancer or for obtaining useful data for the diagnosis of advanced colorectal adenoma and / or colorectal cancer in a subject, preferably in a human subject.

[0108] Consequently, a second aspect of the present invention refers to an in vitro method for screening for subjects, preferably human subjects, at risk of developing colorectaladenomas and / or colorectal cancer comprising: (a) determining at least one, two, three, four or more of the following markers:

[0109] a. Percentage of NK CD56dimcells

[0110] b. Percentage of TIM3 cells in NK CD56bright- FOLD

[0111] c. Percentage of NKT (CD56+ CD3+) - FOLD

[0112] d. Percentage of activated NK CD56dimcells

[0113] e. Percentage of TIM3 in activated CD8 lymphocytes (CD3+CD8+) f. Percentage of CD4 lymphocytes (CD3+CD4+) - FOLD

[0114] g. Percentage of TIM3 in CD8 lymphocytes (CD3+CD8+) - FOLD h. Percentage of TIM3 in CD4 lymphocytes (CD3+CD4+) - FOLD or any combination therefrom, in a, preferably minimally-invasive, isolated biological sample comprising or consisting of mononuclear cells and / or polyclonally activated mononuclear immunological cells, preferably concanavalin A (ConA) activated mononuclear immunological cells of the subjects, preferably human subjects, to be screened and (b) comparing said markers of the subjects, preferably human subjects, to be screened with an already established expression pattern of markers, wherein an statistical difference is indicative of colorectal adenomas and / or colorectal cancer. In a preferred embodiment, the, preferably minimally-invasive, biological sample used in step (a) can be selected from the group consisting of: blood sample, plasma sample or serum sample, preferably whole blood or peripheral blood. More preferably, the, preferably minimally-invasive, biological sample used in step (a) comprises the cellular component of the blood.

[0115] It is herein noted that the term "already established expression pattern of markers" shall be understood as the expression pattern or level of a normal subject or normal population of subjects, wherein the normal subject or normal population is a healthy subject or population not suffering from colorectal adenoma or colorectal cancer.

[0116] A preferred embodiment of the present invention refers to an in vitro method for screening for subjects, preferably human subjects, at risk of developing colorectal cancer comprising: (a) determining at least one, two, three, four or more of the following markers:a. Percentage of NK CD56dimcells

[0117] b. Percentage of TIM3 cells in NK CD56bright- FOLD

[0118] c. Percentage of NKT (CD56+ CD3+) - FOLD

[0119] d. Percentage of TIM3 in activated CD8 lymphocytes (CD3+CD8+) e. Percentage of TIM3 in CD8 lymphocytes (CD3+CD8+) - FOLD f. Percentage of CD4 lymphocytes (CD3+CD4+) - FOLD.

[0120] or any combination therefrom, in a, preferably minimally-invasive, isolated biological sample comprising or consisting of mononuclear cells and / or polyclonally activated mononuclear immunological cells, preferably concanavalin A (ConA) activated mononuclear immunological cells, of the subjects, preferably human subjects, to be screened and (b) comparing said markers of the subjects, preferably human subjects, to be screened with the expression pattern or level of a normal subject or with an already established expression pattern of markers, wherein the normal subject is a healthy subject not suffering from colorectal cancer, and wherein an statistical difference is indicative of colorectal cancer. In a preferred embodiment, the minimally-invasive biological sample used in step (a) can be selected from the group consisting of: blood sample, plasma sample or serum sample, preferably whole blood or peripheral blood. More preferably, the, preferably minimally-invasive, biological sample used in step (a) comprises the cellular component of the blood.

[0121] A preferred embodiment of the present invention refers to an in vitro method for the diagnosis of colorectal adenoma and / or colorectal cancer in a subject, preferably in a human subject, comprising: (a) determining at least one, two, three, four or more of the following markers:

[0122] a. Percentage of NK CD56dimcells

[0123] b. Percentage ofTIM3 cells in NK CD56bright- FOLD

[0124] c. Percentage of NKT (CD56+ CD3+) - FOLD

[0125] d. Percentage of activated NK CD56dimcells

[0126] e. Percentage of TIM3 in activated CD8 lymphocytes (CD3+CD8+) f. Percentage of CD4 lymphocytes (CD3+CD4+) - FOLD

[0127] g. Percentage of TIM3 in CD8 lymphocytes (CD3+CD8+) - FOLD h. Percentage of TIM3 in CD4 lymphocytes (CD3+CD4+) - FOLDor any combination therefrom, in a, preferably minimally-invasive, isolated biological sample comprising or consisting of mononuclear cells and / or polyclonally activated mononuclear immunological cells, preferably concanavalin A (ConA) activated mononuclear immunological cells of the subjects, preferably human subjects, suspected of suffering from colorectal adenoma and / or colorectal cancer and (b) comparing said markers of the human subjects to be screened with the expression pattern or level of a normal subject or with an established expression pattern, wherein the normal subject is a healthy subject not suffering from colorectal adenoma and / or colorectal cancer, or between subjects suspected of suffering from either colorectal adenoma and / or colorectal cancer, wherein an statistical difference is indicative of colorectal adenomas and / or colorectal cancer. In a preferred embodiment, the, preferably minimally-invasive, biological sample obtained in the step (a) comprises: blood sample, plasma sample or serum sample. More preferably, the, preferably minimally-invasive, biological sample obtained in step (a) comprises the cellular component of the blood. In another preferred embodiment, the method comprises optionally (c) confirming the diagnosis by means of the examination of the bowel by any means, preferably using colonoscopy.

[0128] Another preferred embodiment of the present invention refers to an in vitro method for the diagnosis of colorectal cancer in a subject, preferably in a human subject, comprising: (a) determining at least one, two, three, four or more of the following markers:

[0129] a. Percentage of NK CD56dimcells

[0130] b. Percentage of TIM3 cells in NK CD56bright- FOLD

[0131] c. Percentage of NKT (CD56+ CD3+) - FOLD

[0132] d. Percentage of TIM3 in activated CD8 lymphocytes (CD3+CD8+) e. Percentage of TIM3 in CD8 lymphocytes (CD3+CD8+) - FOLD f. Percentage of CD4 lymphocytes (CD3+CD4+) - FOLD.

[0133] or any combination therefrom, obtained from a, preferably minimally-invasive, isolated biological sample comprising or consisting of mononuclear immunological cells and / or polyclonally activated mononuclear immunological cells, preferably concanavalin A (ConA) activated mononuclear immunological cells of the subjects, preferably human subjects, suspected of suffering from colorectal cancer and (b) comparing markers of the human subjects to be screened with the expression pattern or level of a normal subjector with an already established expression pattern of markers, wherein the normal subject is a healthy subject not suffering from colorectal cancer, and wherein an statistical difference is indicative of colorectal cancer. In a preferred embodiment, the minimally-invasive biological sample obtained in the step (a) comprises: blood sample, plasma sample or serum sample. More preferably, the minimally-invasive biological sample obtained in step (a) comprises the cellular component of the blood. In another preferred embodiment, the method comprises optionally (c) confirming the diagnosis by means of the examination of the bowel by any means, preferably using colonoscopy. Another preferred embodiment of the invention refers to an in vitro method for screening for subjects, preferably human subjects, at risk of developing colorectal adenomas comprising: (a) determining at least one, two, three, four or more of the following markers:

[0134] a) Percentage of NK CD56dimcells

[0135] b) Percentage of TIM3 cells in NK CD56bright- FOLD

[0136] c) Percentage of NKT (CD56+ CD3+) - FOLD

[0137] d) Percentage of activated NK CD56dimcells

[0138] or any combination therefrom, in a, preferably minimally-invasive, isolated biological sample comprising or consisting of mononuclear cells and / or polyclonally activated mononuclear immunological cells, preferably concanavalin A (ConA) activated mononuclear immunological cells, of the human subjects to be screened and (b) comparing said markers of the human subjects to be screened with the expression pattern or level of a normal subject or with an already established expression pattern of markers, wherein the normal subject is a healthy subject not suffering from colorectal adenoma, and wherein an statistical difference is indicative of colorectal adenomas. In a preferred embodiment, the, preferably minimally-invasive, biological sample obtained in the step (a) comprises: blood sample, plasma sample or serum sample. More preferably, the, preferably minimally-invasive, biological sample obtained in step (a) comprises the cellular component of the blood.A preferred embodiment of the present invention refers to an in vitro method for the diagnosis of colorectal adenoma in a subject, preferably in a human subject, comprising: (a) determining at least one, two, three, four or more of the following markers:

[0139] a) Percentage of NK CD56dimcells

[0140] b) Percentage of TIM3 cells in NK CD56bright- FOLD

[0141] c) Percentage of NKT (CD56+ CD3+) - FOLD

[0142] d) Percentage of activated NK CD56dimcells

[0143] or any combination therefrom, in a, preferably minimally-invasive, isolated biological sample comprising or consisting of mononuclear cells and / or concanavalin A (ConA) activated mononuclear cells of the subjects, preferably human subjects, suspected of suffering from colorectal adenoma and (b) comparing said markers of the human subjects to be screened with the expression pattern or level of a normal subject, wherein the normal subject is a healthy subject not suffering from colorectal adenoma, wherein an statistical difference is indicative of colorectal adenomas. In a preferred embodiment, the minimally-invasive biological sample obtained in the step (a) comprises: blood sample, plasma sample or serum sample. More preferably, the minimally-invasive biological sample obtained in step (a) comprises the cellular component of the blood. Another preferred embodiment of the invention refers to an in vitro method for differentiating patients with colon adenoma from patients with colorectal cancer, preferably human subjects, comprising: (a) determining at least one, two, three, four or more of the following markers:

[0144] a. Percentage of activated NK CD56dimcells

[0145] b. Percentage of TIM3+ cells in CD8 lymphocytes (CD3+CD8+) - FOLD c. Percentage of lymphocytes CD4 (CD3+CD4+) - FOLD

[0146] d. Percentage of TIM3+ cells in CD4 lymphocytes (CD3+CD4+) - FOLD or any combination therefrom, in a, preferably minimally-invasive, isolated biological sample comprising or consisting of mononuclear cells and / or concanavalin A (ConA) activated mononuclear cells of the human subjects to be screened and (b) comparing markers of the human subjects to be screened with the expression pattern or level of subjects suspected of suffering from colorectal cancer, wherein an statistical differenceis indicative of colorectal adenomas. In a preferred embodiment, the preferably minimally-invasive, biological sample obtained in the step (a) comprises: blood sample, plasma sample or serum sample. More preferably, the minimally-invasive biological sample obtained in step (a) comprises the cellular component of the blood.

[0147] A third aspect of the present invention refers to an in vitro method for screening for subjects, preferably human subjects, at risk of developing colorectal adenomas and / or colorectal cancer comprising: (a) determining at least one, two, three, four or more of the following markers:

[0148] a. Percentage of NK CD56dimcells

[0149] b. Percentage of TIM3 cells in NK CD56bright- FOLD

[0150] c. Percentage of NKT (CD56+ CD3+) - FOLD

[0151] d. Percentage of activated NK CD56dimcells

[0152] e. Percentage of TIM3 in activated CD8 lymphocytes (CD3+CD8+) f. Percentage of CD4 lymphocytes (CD3+CD4+) - FOLD

[0153] g. Percentage of TIM3 in CD8 lymphocytes (CD3+CD8+) - FOLD h. Percentage of TIM3 in CD4 lymphocytes (CD3+CD4+) - FOLD or any combination therefrom, in an, preferably minimally-invasive, isolated biological sample comprising or consisting of mononuclear cells and / or polyclonally activated mononuclear immunological cells, preferably concanavalin A (ConA) activated mononuclear immunological cells, of the human subjects to be screened and (b) identifying the subject as a subject at risk of developing colorectal adenomas and / or colorectal cancer by a predictive model which correlates at least one or more of the markers identified in step (a) with representative values or scores of the same from samples obtained or isolated from subjects previously identified as suffering from colorectal adenomas and / or colorectal cancer, said predictive model having been generated by training a computer with a plurality of marker profiles (profiles comprising one or more, preferably two or more, preferably all, of the markers identified in step (a)) from previously identified subjects having colorectal adenomas and / or colorectal cancer, by machine learning on said plurality of marker profiles so as to obtain representative score profiles associated with colorectal adenomas and / or colorectal cancer.In an embodiment, the third step of the second aspect of the invention is performed by a machine learning method selected from a regression method, a classification method or a combination thereof.

[0154] It will be appreciated that the term "machine learning" generally refers to algorithms that give a computer the ability to learn without being explicitly programmed, including algorithms that learn from and make predictions about data. Machine learning algorithms employed by the embodiments disclosed herein may include, but are not limited to, random forest ("RF"), least absolute shrinkage and selection operator ("LASSO") logistic regression, regularized logistic regression, XGBoost, decision tree learning, artificial neural networks ("ANN"), deep neural networks ("DNN"), support vector machines, rule-based machine learning, and / or others.

[0155] For clarity, algorithms such as linear regression or logistic regression can be used as part of a machine learning process. However, it will be understood that using linear regression or another algorithm as part of a machine learning process is distinct from performing a statistical analysis such as regression with a spreadsheet program. Whereas statistical modeling relies on finding relationships between variables (e.g., mathematical equations) to predict an outcome, a machine learning process may continually update model parameters and adjust a classifier as new data becomes available, without relying on explicit or rules-based programming.

[0156] In a particular embodiment, the second step of the third aspect of the invention is performed by a classification method, which results in identifying the subject as a subject at risk of suffering from colorectal adenomas and / or colorectal cancer.

[0157] In one embodiment, step (b) is carried out by a classification method; preferably selected from logistic regression, random forest, gradient boosting (GB), adaptive boosting (AB), extreme Gradient Boosting (XGB) k-nearest neighbors (kNN), artificial neural network (ANN), support vector machine (SVM), and combinations thereof.

[0158] In an embodiment, the predictive model is generated by training the computer with a plurality of score profiles, as indicated in the third aspect of the invention (please refer to steps a) to h)) from previously identified samples from subjects having colorectal adenomas and / or colorectal cancer by machine learning on said plurality of scoreprofiles so as to obtain representative multivariable data sets associated with this group of patients; wherein the training comprises the following steps:

[0159] (i) training data, from a plurality of score profiles, is randomly stratified into:

[0160] - a calibration dataset (for example in a percentage of about 75%), and - a validation dataset (for example in a percentage of about 25%);

[0161] (ii) the predictive model is seeded on the calibration dataset (particularly is developed by applying a machine learning method selected from a regression method, a classification method or a combination thereof on the calibration dataset);

[0162] (iii) the predictive model is optimized by an internal cross validation; preferably by a k-fold cross validation, wherein each of the k cases of the k-fold cross validation is used for testing only once and one at a time; and

[0163] (iv) the predictive model is further validated by predicting new samples using the validation dataset.

[0164] In some embodiments, the second step is performed by a classification method wherein the patients are assigned a probability of belonging to given category such as patients having colorectal adenomas and / or colorectal cancer. In some embodiments, the classification method is carried out by a method selected from gradient boosting, support vector machine (SVM), decision trees, K nearest neighbors, Naive Bayes or neural networks. In a preferred embodiment, the classification method is carried out by a Gradient Boosting. As used herein, Gradient Boosting is a machine learning algorithm that uses a gradient boosting framework. Gradient Boosting trees, a decision-tree-based ensemble model, differ fundamentally from conventional statistical techniques that aim to fit a single model using the entire dataset. Such ensemble approach improves performance by combining strengths of models that learn the data by recursive binary splits, such as trees, and of "boosting", an adaptive method for combining several simple (base) models. At each iteration of the gradient boosting algorithm, a subsample of the training data is selected at random (without replacement) from the entire training data set, and then a simple base learner is fitted on each subsample. The final boosted treesmodel is an additive tree model, constructed by sequentially fiting such base learners on different subsamples. This procedure incorporates randomization, which is known to substantially improve the predictor accuracy and also increase robustness. Additionally, boosted trees can fit complex nonlinear relationships, and automatically handle interaction effects between predictors as addition to other advantages of tree-based methods, such as handling features of different types and accommodating missing data. Hence, in many cases their predictive performance is superior to most traditional modelling methods. In a particular embodiment, the second step is performed by a regression method; preferably selected from multiple linear regression (MLR), principal component regression (PCR), partial least squares regression (PLSR), artificial neural network (ANN), support vector machine (SVM), random forest (RF), lassor regression, ridge regression and combinations thereof.

[0165] In a particular embodiment, the second step is performed by a classification method, more in particular, by gradient boosting, which includes the value of one or more variables of the gene expression profile collected in step (i) and which contribute to the identification of the patients having colorectal adenomas and / or colorectal cancer. In a particular embodiment, the second step is performed by a regression method which includes the value of one or more variables of the score profile collected in step (i) and which contribute to the identification of the patients having colorectal adenomas and / or colorectal cancer.

[0166] Therefore, a preferred embodiment of the third aspect of the present invention refers to an in vitro method for screening for subjects, preferably human subjects, at risk of developing colorectal cancer comprising: (a) determining at least one, two, three, four or more of the following markers:

[0167] a. Percentage of NK CD56dimcells

[0168] b. Percentage of TIM3 cells in NK CD56bright- FOLD

[0169] c. Percentage of NKT (CD56+ CD3+) - FOLD

[0170] d. Percentage of TIM3 in activated CD8 lymphocytes (CD3+CD8+) e. Percentage of TIM3 in CD8 lymphocytes (CD3+CD8+) - FOLD f. Percentage of CD4 lymphocytes (CD3+CD4+) - FOLD.or any combination therefrom, in an, preferably minimally-invasive, isolated biological sample comprising or consisting of mononuclear cells and / or polyclonally activated mononuclear immunological cells, preferably concanavalin A (ConA) activated mononuclear immunological cells, of the human subjects to be screened and (b) identifying the subject as a subject at risk of developing colorectal cancer by a predictive model which correlates at least one or more of the markers identified in step (a) with representative values or scores of those identified in step (a) from samples obtained from subjects previously identified as suffering from colorectal cancer, said predictive model having been generated by training a computer with a plurality of marker profiles from at least one or more of those marker identified in step (a) from previously identified subjects having colorectal cancer, by machine learning on said plurality of marker profiles so as to obtain representative marker profiles associated with colorectal cancer. Please note that the combination of all of these scores, as shown in figure 1, is capable of differentiating healthy donors from patients with colorectal cancer.

[0171] It is noted that any of the above methods for screening can be carried for diagnostic or prognostic purposes, in such case, optionally, the method can be subsequently followed by step (c) for confirming the diagnosis by means of the examination of the bowel by any means, preferably using colonoscopy.

[0172] Another preferred embodiment of the invention refers to an in vitro method for screening for subjects, preferably human subjects, at risk of developing colorectal adenomas comprising: (a) determining at least one two, three, four or more of the following markers:

[0173] a) Percentage of NK CD56dimcells

[0174] b) Percentage of TIM3 cells in NK CD56bright- FOLD

[0175] c) Percentage of NKT (CD56+ CD3+) - FOLD

[0176] d) Percentage of activated NK CD56dimcells

[0177] or any combination therefrom, in an, preferably minimally-invasive, isolated biological sample comprising or consisting of mononuclear cells and / or polyclonally activated mononuclear immunological cells, preferably concanavalin A (ConA) activated mononuclear immunological cells, of the human subjects to be screened and (b)identifying the subject as a subject at risk of developing colorectal adenomas by a predictive model which correlates at least one or more of the markers identified in step (a) with representative values or scores of the same taken from samples obtained from subjects previously identified as suffering from colorectal adenomas, said predictive model having been generated by training a computer with a plurality of marker profiles from the markers identified in step (a) from previously identified subjects having colorectal adenomas, by machine learning on said plurality of marker profiles so as to obtain representative marker profiles associated with colorectal adenomas. It is noted that the combination of all of these scores, as shown in figure 2, is capable of differentiating healthy donors from patients with colon adenoma. In particular, figure 2 shows a discriminatory capacity with a sensitivity of 93.75% and a specificity of 78.95%. Therefore, preferably the score profiles shall comprise the combination of all of these scores as shown in step a).

[0178] Another preferred embodiment of the invention refers to an in vitro method for differentiating patients with colon adenoma from patients with colorectal cancer, preferably human subjects, comprising: (a) determining at least one, two, three, four or more of the following markers:

[0179] a. Percentage of activated NK CD56dimcells

[0180] b. Percentage of TIM3 cells in CD8 lymphocytes (CD3+CD8+) - FOLD c. Percentage of lymphocytes CD4 (CD3+CD4+) - FOLD

[0181] d. Percentage of TIM3 cells in CD4 lymphocytes (CD3+CD4+) - FOLD or any combination therefrom, in an, preferably minimally-invasive, isolated biological sample comprising or consisting of mononuclear cells and / or concanavalin A (ConA) activated mononuclear cells, of the human subjects to be screened and (b) identifying the subject as a subject at risk of developing colorectal adenomas and / or colorectal cancer by a predictive model which correlates at least one or more of the scores identified in step (a) with representative markers of the same taken from samples obtained from subjects previously identified as suffering from colorectal adenomas and / or colorectal cancer, said predictive model having been generated by training a computer with a plurality of marker profiles from the markers identified in step (a) from previously identified subjects having colorectal adenomas and / or colorectal cancer, bymachine learning on said plurality of marker profiles so as to obtain representative marker profiles associated with colorectal adenomas and / or colorectal cancer. It is noted that the combination of these markers, as shown in figure 3, is capable of differentiating patients with colon adenoma from patients with colon cancer. In particular, figure 3 shows a discriminatory capacity with a sensitivity of 100% and a specificity of 100%. Therefore, preferably the score profiles shall comprise the combination of all of these scores as shown in step a).

[0182] It is herein noted that the results obtained with the different combinations of markers are deemed indicated in the screening methods of the third aspect of the invention, for the presence of colorectal adenoma and / or colorectal cancer, can be further used for obtaining useful data for the diagnosis of advanced colorectal adenoma and / or colorectal cancer in a subject, preferably in a human subject. Therefore, a fourth aspect of the present invention refers to an in vitro method for obtaining useful data for the diagnosis of colorectal adenoma and / or colorectal cancer in subjects, preferably human subjects, comprising implementing the steps of any of the methods of the second or third aspect of the invention.

[0183] A fifth aspect of the present invention refers to the use of a kit comprising biomarker detecting reagents for determining the expression level of the markers of the first aspect of the invention, for diagnosing in vitro the risk for colorectal adenoma and / or colorectal cancer. In a preferred embodiment, colorectal adenoma is advanced colorectal adenoma. In a preferred embodiment, the step which comprises measuring the expression pattern or level of any of the markers, is carried out by using a detectably labelled probe.

[0184] A sixth aspect of the present invention refers to an in vitro method for classifying subjects, preferably human subjects, as healthy subjects or as subjects suffering from colorectal adenoma and / or colorectal cancer comprising implementing any of the methods of the second or third aspect of the invention.

[0185] A seventh aspect of the present invention refers to a method for treating subjects, preferably human subjects, suffering from colorectal adenoma and / or colorectal cancer comprising implementing any of the methods of the second, third or fourth aspect ofthe invention and (c) treating the patient at risk or diagnosed with colorectal adenoma and / or colorectal cancer. In a preferred embodiment, the minimally-invasive biological sample obtained in the step (a) comprises: blood sample, plasma sample or serum sample. In a preferred embodiment the method comprises confirming the diagnosis by means of the examination of the bowel by any means, preferably using colonoscopy. Since colorectal adenoma can be seen as a precursor of colorectal cancer, because of the acknowledged adenoma-carcinoma sequence, and the notion that advanced colorectal adenomas are more likely to transition to cancer, it is well established that colorectal adenomas, and preferably colorectal advanced adenomas, should be treated, preferably, by being removed through colonoscopy (subsequent surveillance could be performed). Treatment of colorectal cancer depends on the stage at which cancer was discovered. Early-stage colorectal cancer is best treated with surgery. Approximately 95% of Stage I and 65-80% of Stage II colorectal cancers are curable with surgery. Rectal cancer, however, may require additional radiation therapy to minimize the risk of recurrence. Advanced stage (Stage III and Stage IV) treatment often comprises combination of therapies, including: surgery, chemotherapy, treatment with antibodies, therapies anti-VEGF / R and radiation.

[0186] An eighth embodiment of the present invention refers to an in vitro method for assessing or monitoring the response to a therapy in a subject suffering from colorectal adenoma and / or colorectal cancer comprising implementing any of the methods of the second aspect of the invention in the subjects, preferably human subjects, to be monitored. On another note, a ninth aspect of the invention refers to a method for screening or identifying a subject at risk of having colorectal cancer and / or colorectal adenomas, the method comprising:

[0187] a. Determining or receiving at least two, three, four or more of the following plurality of measurable parameters or markers:

[0188] i. Percentage of NK CD56dimcells;

[0189] ii. Percentage of TIM3 cells in NK CD56bright- FOLD;

[0190] iii. Percentage of NKT (CD56+ CD3+) - FOLD;1

[0191] iv. Percentage of TIM3 in activated CD8 lymphocytes (CD3+CD8+); v. Percentage of TIM3 in CD8 lymphocytes (CD3+CD8+) - FOLD; vi. Percentage of CD4 lymphocytes (CD3+CD4+) - FOLD;

[0192] vii. Percentage of activated NK CD56dimcells; and

[0193] viii. Percentage of TIM3 in CD4 lymphocytes (CD3+CD4+) - FOLD; from an isolated blood sample of the subject comprising mononuclear immunological cells and / or polyclonally activated mononuclear immunological cells, preferably concanavalin A (ConA) activated mononuclear immunological cells, as measured with flow cytometry;

[0194] b. obtaining a composite score, the composite score having been generated from the plurality of measurable parameters or markers determined or received in step (a) by applying a classification model configured to estimate a weighted contribution of each parameter, wherein the classification model is selected from: - logistic regression, - a supervised machine-learning model selected from at least a neural network or a random forest, or- a linear or non-linear statistical method selected from at least partial least squares regression or linear discriminant analysis, and providing that composite score for subsequent comparison;

[0195] c. comparing the composite score to one or more predefined thresholds, each threshold being derived from either (i) a reference population or (ii) a reference value that permits classification of individuals as being at risk of having the disease versus not at risk, wherein a composite score above a given threshold is classified as "higher risk", and a composite score below that threshold is classified as "lower risk"; and

[0196] d. categorizing the subject into one of at least two discrete risk categories based on the comparison in step (c), thereby providing a risk stratification output indicative of whether the subject's risk of having colorectal cancer and / or colorectal adenomas is increased or decreased relative to the reference.In certain embodiments, a composite score is obtained by first determining or receiving the plurality of measurable parameters or markers, or a combination of four or more of the plurality of measurable parameters, described above from each subject under evaluation. To combine these parameters into a single predictive measure, a classification model is applied. The model can be a logistic regression, a supervised machine-learning algorithm (for example, a neural network or a random forest), or a linear or non-linear statistical method (such as partial least squares regression or linear discriminant analysis). Each of these approaches is well known in the statistical and machine-learning fields and may be readily implemented using existing software libraries or standard analytical methods.

[0197] Before application to new data, the classification model is typically trained or calibrated on a set of reference samples where the disease status is known. During this training phase, the model learns how to weight the contribution of each measurable parameter. For instance, in a logistic regression model, coefficients are computed through maximum likelihood estimation. In machine-learning approaches like neural networks, the model iteratively adjusts its internal parameters (weights) to minimize prediction error, while in random forests, multiple decision trees are constructed and combined for robust classification. For partial least squares regression, latent variables capturing major trends in the data are extracted; and in linear discriminant analysis, parameters are estimated by analyzing the covariance structures of data belonging to different categories. Through these established techniques, each parameter is assigned a relative weighting that reflects its importance in predicting disease risk.

[0198] Once the model has been trained, it is used to generate a composite score for each subject. The composite score is a single numerical value (or a similarly combined index) that encapsulates the weighted contribution of all relevant parameters. This numeric result is then compared against one or more predefined thresholds. These thresholds may be derived from statistical analysis of a reference population (e.g., the distribution of scores in healthy versus diseased individuals) or from clinically established cutoffs (e.g., a consensus value indicating heightened risk). Thresholds can be tailored to specific disease profiles, clinical objectives, or desired sensitivity and specificity performance.By comparing the composite score to the relevant threshold(s), the subject is categorized into at least two discrete risk categories, typically labelled as "higher risk" or "lower risk." For example, a composite score exceeding a selected threshold might classify the individual as "higher risk," whereas a score below that threshold would indicate "lower risk." Additional tiered thresholds can be incorporated if multiple stratification levels are clinically useful, allowing for further subclassification (e.g., moderate risk, very high risk, etc.).

[0199] The final output of this process is a risk stratification measure, which provides an indication of whether the subject's likelihood of having the disease, or of developing the disease in the future, is increased or decreased relative to the reference. By integrating diverse parameters into a single predictive framework and comparing the resulting composite score to established thresholds, clinicians or researchers can make informed decisions regarding diagnosis, monitoring, or treatment pathways. A skilled person would be able to replicate this process using conventional training data sets, standard software implementations of classification models, and suitable thresholds derived from empirical or published reference data, all without undue experimentation.

[0200] In a preferred embodiment of the ninth aspect of the invention, at least one of the predefined thresholds of step (c) is determined by maximizing an index that accounts for the sensitivity and specificity of the classification, preferably the Youden Index defined as Sensitivity + Specificity - 1, such that the threshold permits classification of individuals as being at risk of having the disease or condition versus those not having the disease or condition. Therefore, in certain embodiments, at least one of the predefined thresholds used to classify disease risk is determined by maximizing an index that takes into account both sensitivity and specificity of the classification. Sensitivity (also referred to as the true-positive rate) measures the proportion of actual positives that are correctly identified, whereas specificity (the true-negative rate) measures the proportion of actual negatives that are correctly identified. By simultaneously considering these two metrics, the threshold selection process ensures that the classification model optimallydistinguishes between individuals who are at risk of having the disease or condition and those who are not.

[0201] A particularly preferred metric for guiding threshold determination is the Youden Index, defined as Youden lndex=(Sensitivity+Specificity)-l. This index reaches its maximum value of 1 when both sensitivity and specificity are perfect (i.e., each equals 1, indicating zero misclassifications). Conversely, the index approaches 0 when the model's performance is no better than random chance. To find the optimal threshold, sensitivity and specificity are calculated across a range of possible threshold values, and the threshold that yields the highest Youden Index is selected. Such a procedure is common in receiver operating characteristic (ROC) curve analysis and can be implemented using widely available statistical or data-analytics tools.

[0202] By maximizing the Youden Index, the selected threshold provides a robust balance between correctly identifying diseased individuals (maximizing sensitivity) and correctly excluding non-diseased individuals (maximizing specificity). Consequently, this threshold offers a high level of diagnostic accuracy for classifying individuals as either at risk of having the disease or condition, or not at risk. A skilled person, with access to standard epidemiological or statistical software and relevant validation data, can readily apply this process without undue experimentation to achieve reliable risk stratification in various clinical or research setings.

[0203] In another preferred embodiment of the ninth aspect of the invention, optionally in combination with any of the previous or subsequent embodiments of the invention, the method is for screening for subjects, preferably human subjects, at risk of having colorectal adenomas (from hereinafter "method for screening for subjects at risk of having colorectal adenomas of the invention"), wherein step a) determines or receives at least all of the following plurality of measurable parameters or markers:

[0204] i. Percentage of NK CD56dimcells;

[0205] ii. Percentage of TIM3 cells in NK CD56bright- FOLD;

[0206] iii. Percentage of NKT (CD56+ CD3+) - FOLD; andiv. Percentage of activated NK CD56dimcells;

[0207] and wherein step d) of the method categorizes the subject into one of at least two discrete risk categories based on the comparison in step (c), thereby providing a risk stratification output indicative of whether the subject' risk of having or developing colorectal adenomas is increased ("higher risk"), the subject is at risk of having colorectal adenomas, or decreased "lower risk", the subject is at low or no risk of having colorectal adenomas, relative to the reference of step c).

[0208] In another preferred embodiment of the ninth aspect of the invention, optionally in combination with any of the previous or subsequent embodiments of the invention, the method is for screening for subjects, preferably human subjects, at risk of having colorectal cancer (from hereinafter "method for screening for subjects at risk of having colorectal cancer of the invention"), and step a) determines or receives at least all of the following plurality of measurable parameters or markers:

[0209] i. Percentage of NK CD56dimcells;

[0210] ii. Percentage of TIM3 cells in NK CD56bright- FOLD;

[0211] iii. Percentage of NKT (CD56+ CD3+) - FOLD;

[0212] iv. Percentage of TIM3 in activated CD8 lymphocytes (CD3+CD8+); v. Percentage of TIM3 in CD8 lymphocytes (CD3+CD8+) - FOLD; and vi. Percentage of CD4 lymphocytes (CD3+CD4+) - FOLD;

[0213] and wherein step d) of the method categorizes the subject into one of at least two discrete risk categories based on the comparison in step (c), thereby providing a risk stratification output indicative of whether the subject's risk of having colorectal cancer is increased ("higher risk"), the subject is at risk of having colorectal cancer, or decreased "lower risk", the subject is at low or no risk of having colorectal cancer, relative to the reference of step c).

[0214] In another preferred embodiment of the ninth aspect of the invention, optionally in combination with any of the previous or subsequent embodiments of the invention, the method further comprises carrying out a confirmatory method (from hereinafter "confirmatory method of the invention") for classifying subjects classified as at risk ("higher risk") of having colorectal adenomas according to the method for screening forsubjects at risk of having colorectal adenomas of the invention, as individuals at risk of having colorectal cancer or at risk of having colorectal adenomas, the method comprising:

[0215] a. Determining or receiving at least four or more of the following plurality of measurable parameters or markers:

[0216] i. Percentage of CD4 lymphocytes (CD3+CD4+) - FOLD

[0217] ii. Percentage of activated NK CD56dimcells

[0218] iii. Percentage of TIM3 in lymphocytes (CD3+CD8+) - FOLD iv. Percentage of TIM3 in lymphocytes (CD3+CD4+) - FOLD from an isolated blood sample of the subject, classified as at risk ("higher risk") of having colorectal adenomas according to the method for screening for subjects at risk of having colorectal adenomas of the invention, comprising mononuclear immunological cells and / or polyclonally activated mononuclear immunological cells, preferably concanavalin A (ConA) activated mononuclear immunological cells, as measured with flow cytometry;

[0219] b. obtaining a composite score, the composite score having been generated from the plurality of measurable parameters determined or received in step (a) by applying a classification model configured to estimate a weighted contribution of each parameter, wherein the classification model is selected from: - logistic regression, - a supervised machinelearning model selected from at least a neural network or a random forest, or - a linear or non-linear statistical method selected from at least partial least squares regression or linear discriminant analysis, and providing that composite score for subsequent comparison; c. comparing the composite score to one or more predefined thresholds, each threshold being derived from a reference value that permits classification of individuals as being at risk of having colorectal cancer versus colorectal adenomas, wherein a composite score above a given threshold is classified as "higher risk" of having colorectal cancer, and a composite score below that threshold is classified as "lower risk" ofhaving colorectal cancer thereby confirming the higher risk of the subject having colorectal adenomas; and

[0220] d. categorizing the individual into one of at least two discrete risk categories based on the comparison in step (c), thereby providing a risk stratification output indicative of whether the individual's risk of having colorectal cancer is increased or decreased relative to the reference.

[0221] In another preferred embodiment of the ninth aspect of the invention, optionally in combination with any of the previous or subsequent embodiments of the invention, the method further comprises carrying out a confirmatory method (from hereinafter "second confirmatory method of the invention") for classifying subjects classified as at risk ("higher risk") of having colorectal cancer according to the method for screening for subjects at risk of having colorectal cancer of the invention, as individuals at risk of having colorectal cancer or at risk of having colorectal adenomas, the method comprising:

[0222] a. Determining or receiving at least four or more of the following plurality of measurable parameters:

[0223] i. Percentage of CD4 lymphocytes (CD3+CD4+) - FOLD

[0224] ii. Percentage of activated NK CD56dimcells

[0225] iii. Percentage of TIM3 in lymphocytes (CD3+CD8+) - FOLD iv. Percentage of TIM3 in lymphocytes (CD3+CD4+) - FOLD from an isolated blood sample of the subject, classified as at risk of having colorectal cancer according to the method for screening for subjects at risk of having colorectal cancer of the invention, comprising mononuclear immunological cells and / or polyclonally activated mononuclear immunological cells, preferably concanavalin A (ConA) activated mononuclear immunological cells, as measured with flow cytometry;

[0226] b. obtaining a composite score, the composite score having been generated from the plurality of measurable parameters determined or received in step(a) by applying a classification model configured to estimate a weighted contribution of each parameter, wherein the classification model is selected from: - logistic regression, - a supervised machine-learning model selected from at least a neural network or a random forest, or - a linear or non-linear statistical method selected from at least partial least squares regression or linear discriminant analysis, and providing that composite score for subsequent comparison;

[0227] c. comparing the composite score to one or more predefined thresholds, each threshold being derived from a reference value that permits classification of individuals as being at risk of having colorectal cancer versus colorectal adenomas, wherein a composite score above a given threshold is classified as "higher risk" of having colorectal cancer thereby confirming the risk of having colorectal cancer according to the method for screening for subjects at risk of having colorectal cancer of the invention, and a composite score below that threshold is classified as "lower risk" of having colorectal cancer thereby indicating a higher risk that the subject has colorectal adenomas; and d. categorizing the subject into one of at least two discrete risk categories based on the comparison in step (c), thereby providing a risk stratification output indicative of whether the subject's risk of having colorectal cancer is increased or decreased relative to the reference.

[0228] In another preferred embodiment of the ninth aspect of the invention, optionally in combination with any of the previous or subsequent embodiments of the invention, the classification model is logistic regression, and the one or more weighting factors are obtained by applying a Backwards Wald stepwise selection procedure to the logistic regression coefficients.

[0229] In another preferred embodiment of the ninth aspect of the invention, optionally in combination with any of the previous or subsequent embodiments of the invention, the classification model is a supervised machine-learning model, and the one or more weighting factors are learned by a machine-learning algorithm selected from at least oneof: a random forest model, a support vector machine, a neural network, or a gradient boosting model.

[0230] In another preferred embodiment of the ninth aspect of the invention, optionally in combination with any of the previous or subsequent embodiments of the ninth aspect of the invention, the method for screening or identifying a subject at risk of having colorectal cancer and / or colorectal adenomas, comprises the following steps:

[0231] a. Step a) of the method determines or receives the plurality of measurable parameters or markers as defined in anyone of the embodiments of the ninth aspect of the invention;

[0232] b. Step b) calculates, via a logistic regression model, a combined risk score by multiplying each measurable parameter by a corresponding coefficient (B factor) and summing the products;

[0233] c. Step c) compares the combined risk score to at least one predefined threshold to categorize the individual into a higher-risk group or a lower-risk group according to any one of the embodiments of the ninth aspect of the invention; and

[0234] d. Step d) provides a risk stratification output indicating whether the individual is at increased or decreased risk of the disease or condition based on said comparison.

[0235] From hereinafter the method above will be referred to as "specific method of the invention".

[0236] In a preferred embodiment of the specific method of the invention, at least one of the predefined thresholds of step (c) is determined by maximizing an index that accounts for the sensitivity and specificity of the classification, preferably the Youden Index defined as Sensitivity + Specificity - 1, such that the threshold permits classification of individuals as being at risk of having the disease or condition versus those not having the disease or condition.In another preferred embodiment of the specific method of the invention, optionally in combination with any previous specific embodiment of the specific method of the invention, this method is for screening for subjects, preferably human subjects, at risk of having colorectal adenomas, wherein the plurality of measurable parameters of step a) included in this method and their corresponding B factors of step b), are the following:

[0237]

[0238] and wherein the value of the Youden index as the predefined thresholds is 21.07.

[0239] In another preferred embodiment of the specific method of the invention, optionally in combination with any of the previous specific embodiments of the specific method of the invention, this method is for screening for subjects, preferably human subjects, at risk of having colorectal cancer, wherein the plurality of measurable parameters of step a) included in this method and their corresponding B factors of step b), are the following:

[0240]

[0241] and wherein the value of the Youden index as the predefined thresholds is 45.80.In another preferred embodiment of the ninth aspect of the invention, the confirmatory method of the invention or the second confirmatory method of the invention comprises the following steps:

[0242] a. Step a) of the method determines or receives the plurality of measurable parameters as defined in the confirmatory method of the invention or the second confirmatory method of the invention;

[0243] b. Step b) calculates, via a logistic regression model, a combined risk score by multiplying each measurable parameter by a corresponding coefficient (B factor) and summing the products;

[0244] c. Step c) compares the combined risk score to at least one predefined threshold to categorize the individual into a higher-risk group or a lower-risk group according to the confirmatory method of the invention or the second confirmatory method of the invention; and

[0245] d. Step d) provides a risk stratification output indicating whether the individual is at increased or decreased risk of the disease or condition based on said comparison;

[0246] wherein, preferably, at least one of the predefined thresholds of step (c) is determined by maximizing an index that accounts for the sensitivity and specificity of the classification, preferably the Youden Index defined as Sensitivity + Specificity - 1, such that the threshold permits classification of individuals as defined in the confirmatory method of the invention or the second confirmatory method of the invention; and

[0247] wherein, also preferably, the plurality of measurable parameters included in this method and their corresponding B factors, are the following:

[0248]

[0249] and wherein the value of the Youden index as the predefined thresholds is 260.

[0250] In another preferred embodiment of the ninth aspect of the invention, optionally in combination with any of the previous or subsequent embodiments of the invention, the risk stratification output includes a recommended clinical intervention selected from at least one of:

[0251] a. additional diagnostic testing,

[0252] b. initiation or adjustment of a therapeutic regimen, or c. an increase in monitoring frequency.

[0253] A tenth aspect of the invention refers to a system comprising:

[0254] a. Optionally an analysis module configured to perform step (a) of the ninth aspect of the invention, or of any of its preferred embodiments, by obtaining the plurality of measurable parameters;

[0255] b. a processing module configured to perform step (b) of the ninth aspect of the invention, or of any of its preferred embodiments, by calculating a composite score derived from the measurable parameters using one or more weighting factors determined from a classification model selected from at least one of a logistic regression, a machine-learning algorithm, or any other statistical method capable of identifying parameter contributions; and

[0256] c. perform step (c) of the ninth aspect of the invention, orof any of its preferred embodiments, by comparing the composite score to one or more decision thresholds to categorize the individual into at least a higher-risk group or a lower-risk group; andd. an output interface configured to perform step (d) of the ninth aspect of the invention, or of any of its preferred embodiments, by providing a risk stratification output indicating whether the individual is at increased or decreased risk of the disease or condition based on the categorization.

[0257] In a preferred embodiment of the tenth aspect of the invention, the analysis module comprises a flow cytometer for obtaining flow cytometry measurements as part of the plurality of measurable parameters, thereby performing step (a) of the ninth aspect of the invention, or of any of its preferred embodiments, for such measurements.

[0258] In another preferred embodiment of the tenth aspect of the invention, the processing module is further configured to generate a ROC curve for the composite score from a training or validation dataset, determine an optimal threshold by maximizing at least one performance metric selected from sensitivity, specificity, or the Youden index, and store said optimal threshold for subsequent comparisons of new individuals' composite scores.

[0259] In another preferred embodiment of the tenth aspect of the invention, the system further comprises a user interface that displays a visual risk index indicating one of a plurality of risk categories after performing steps (b) and (c) of the ninth aspect of the invention, or of any of its preferred embodiments, and wherein the visual risk index corresponds to step (d) of the ninth aspect of the invention, or of any of its preferred embodiments.

[0260] An eleventh aspect of the invention refers to a non-transitory computer-readable medium storing instructions that, when executed by one or more processors, cause the one or more processors to perform steps (a )— (d) of the ninth aspect of the invention, or of any of its preferred embodiments,, the instructions comprising:

[0261] a. instructions to perform step (a) of the ninth aspect of the invention, or of any of its preferred embodiments, by obtaining a plurality of measurable parameters;b. instructions to perform step (b) of the ninth aspect of the invention, or of any of its preferred embodiments, by calculating a composite score derived from the measurable parameters using one or more weighting factors determined from a classification model selected from at least one of a logistic regression, a machine-learning algorithm, or any other statistical method capable of identifying parameter contributions;

[0262] c. instructions to perform step (c) of the ninth aspect of the invention, or of any of its preferred embodiments, by comparing the composite score to one or more decision thresholds to categorize the individual into at least a higher- risk group or a lower-risk group; and

[0263] d. instructions to perform step (d) of the ninth aspect of the invention, or of any of its preferred embodiments, by providing a risk stratification output indicating whether the individual is at increased or decreased risk of the disease or condition based on the categorization.

[0264] In a preferred embodiment of the eleventh aspect of the invention, the instructions further cause the one or more processors to update the classification model by incorporating new data from previously classified individuals, thereby dynamically recalibrating the weighting factors over time to improve the performance of steps (b) and (c) of the ninth aspect of the invention, or of any of its preferred embodiments,.

[0265] In another preferred embodiment of the eleventh aspect of the invention, the instructions further cause the one or more processors to:

[0266] a. generate at least one ROC curve based on historical composite scores and associated clinical outcomes,

[0267] b. calculate an optimal threshold by applying a performance metric selected from the Youden index, the Fl score, or precision-recall analysis, c. and store the optimal threshold for subsequent execution of step (c) of the ninth aspect of the invention, or of any of its preferred embodiments when classifying additional individuals.On the other hand, it is worth mentioning that the present invention is preferably carried out in the cellular component of the blood obtained from the patients. On the other hand, it is important to emphasize that obtaining the mononuclear immunological cells requires their isolation from the blood by means of gradient centrifugation, separating them from the red blood cells, polymorphonuclear cells and plasma. Once the mononuclear immunological cells are isolated, they required to be plated and cultured in complete culture medium, to be next polyclonally stimulated (activated conditions) or kept in culture (not activated conditions), followed by their incubation with the cocktail of antibodies used to detect the markers indicated thereof, before fixing the sample. These stained and fixed samples are run on a flow cytometer device to detect the indicated populations based on the expression of the markers indicated thereof.

[0268] For the purpose of the present invention, the following definitions are included below:

[0269] • The term "screening" is understood as the examination or testing of a group of individuals pertaining to the general population, at risk of suffering from colorectal cancer or colorectal adenoma, with the objective of discriminating healthy individuals from those who are suffering from an undiagnosed colorectal cancer or colorectal adenoma or who are at high risk of suffering from said indications.

[0270] • The term "colorectal cancer" is a medical condition characterized by cancer of cells of the intestinal tract below the small intestine (i.e., the large intestine (colon), including the cecum, ascending colon, transverse colon, descending colon, sigmoid colon, and rectum).

[0271] • The expression "colorectal adenoma" refers to adenomas of the colon, also called adenomatous polyps, which is a benign and pre-cancerous stage of the colorectal cancer but still with high risk of progression to colorectal cancer.

[0272] • The expression "advanced colorectal adenoma" refers to adenomas having a size of at least 10 mm or histologically having high grade dysplasia or a villous component higher than 20%.

[0273] • The expression "minimally-invasive biological sample" refers to any sample which is taken from the body of the patient without the need of using harmful instruments, other than fine needles used for taking the blood from the patient, and consequently withoutbeing harmfully for the patient. Specifically, minimally-invasive biological sample refers in the present invention to: blood samples.

[0274] • The term "threshold value" or "cutoff value", when referring to the expression levels of the scores described in the present invention, refers to a reference expression level indicative that a subject is likely to suffer from colorectal cancer or colorectal adenoma with a given sensitivity and specificity if the expression levels of the patient are above said threshold or cut-off or reference levels.

[0275] • The term "comprising" it is meant including, but not limited to, whatever follows the word "comprising". Thus, use of the term "comprising" indicates that the listed elements are required or mandatory, but that other elements are optional and may or may not be present.

[0276] By "consisting of" is meant including, and limited to, whatever follows the phrase "consisting of". Thus, the phrase "consisting of" indicates that the listed elements are required or mandatory, and that no other elements may be present.

[0277] It is also noted that the term "kit" as used herein is not limited to any specific device and includes any device suitable for working the invention such as but not limited to microarrays, bioarrays, biochips or biochip arrays. A variety of statistical and mathematical methods for establishing the threshold or cutoff level of expression are known in the prior art. A threshold or cutoff expression level for a particular biomarker may be selected, for example, based on data from Receiver Operating Characteristic (ROC) plots, as described in the Examples and Figures of the present invention. One of skill in the art will appreciate that these threshold or cutoff expression levels can be varied, for example, by moving along the ROC plot for a particular biomarker or combinations thereof, to obtain different values for sensitivity or specificity thereby affecting overall assay performance. For example, if the objective is to have a robust diagnostic method from a clinical point of view, we should try to have a high sensitivity. However, if the goal is to have a cost-effective method we should try to get a high specificity. The best cutoff refers to the value obtained from the ROC plot for a particular biomarker that produces the best sensitivity and specificity. Sensitivity and specificity values are calculated over the range of thresholds (cutoffs). Thus, the threshold or cutoffvalues can be selected such that the sensitivity and / or specificity are at least about 70 %, and can be, for example, at least 75 %, at least 80 %, at least 85 %, at least 90 %, at least 95 %, at least 96 %, at least 97 %, at least 98 %, at least 99 % or at least 100% in at least 60 % of the patient population assayed, or in at least 65 %, 70 %, 75 % or 80 % of the patient population assayed.

[0278] The inventions illustratively described herein may suitably be practiced in the absence of any element or elements, limitation or limitations, not specifically disclosed herein. Thus, for example, the terms "comprising", "including", "containing", etc. shall be read expansively and without limitation. Additionally, the terms and expressions employed herein have been used as terms of description and not of limitation, and there is no intention in the use of such terms and expressions of excluding any equivalents of the features shown and described or portions thereof, but it is recognized that various modifications are possible within the scope of the invention claimed. Thus, it should be understood that although the present invention has been specifically disclosed by preferred embodiments and optional features, modification and variation of the inventions embodied therein herein disclosed may be resorted to by those skilled in the art, and that such modifications and variations are considered to be within the scope of this invention.

[0279] The invention has been described broadly and generically herein. Each of the narrower species and sub-generic groupings falling within the generic disclosure also form part of the invention. This includes the generic description of the invention with a proviso or negative limitation removing any subject matter from the genus, regardless of whether or not the excised material is specifically recited herein.

[0280] Further aspects of the present invention relate to computer implemented methods. In particular to the following embodiments:

[0281] A computer implemented method for screening for subjects, preferably human subjects, at risk of developing colorectal adenomas and / or colorectal cancer comprising: (a) receiving at least one, two, three, four or more of the following markers:

[0282] a. Percentage of NK CD56dimcells

[0283] b. Percentage of TIM3 cells in NK CD56bright- FOLDc. Percentage of NKT (CD56+ CD3+) - FOLD

[0284] d. Percentage of activated NK CD56dimcells

[0285] e. Percentage of TIM3 in activated CD8 lymphocytes (CD3+CD8+) f. Percentage of CD4 lymphocytes (CD3+CD4+) - FOLD

[0286] g. Percentage of TIM3 in CD8 lymphocytes (CD3+CD8+) - FOLD h. Percentage of TIM3 in CD4 lymphocytes (CD3+CD4+) - FOLD or any combination therefrom, from an isolated biological sample comprising or consisting of mononuclear cells and / or polyclonally activated mononuclear immunological cells, preferably concanavalin A (ConA) activated mononuclear immunological cells, of the human subjects to be screened and (b) identifying the subject as a subject at risk of developing colorectal adenomas and / or colorectal cancer by a predictive model which correlates at least one or more of the markers identified in step (a) with representative values of the same from samples obtained or isolated from subjects previously identified as suffering from colorectal adenomas and / or colorectal cancer, said predictive model having been generated by training a computer with a plurality of marker profiles (profiles comprising one or more, preferably two or more, preferably all, of the markers identified in step (a)) from previously identified subjects having colorectal adenomas and / or colorectal cancer, by machine learning on said plurality of marker profiles so as to obtain representative marker profiles associated with colorectal adenomas and / or colorectal cancer.

[0287] A computer implemented method for screening for subjects, preferably human subjects, at risk of developing colorectal cancer comprising: (a) receiving at least the following markers:

[0288] a. Percentage of NK CD56dimcells

[0289] b. Percentage of TIM3 cells in NK CD56bright- FOLD

[0290] c. Percentage of NKT (CD56+ CD3+) - FOLD

[0291] d. Percentage of TIM3 in activated CD8 lymphocytes (CD3+CD8+) e. Percentage of TIM3 in CD8 lymphocytes (CD3+CD8+) - FOLD f. Percentage of CD4 lymphocytes (CD3+CD4+) - FOLD.from an isolated biological sample comprising or consisting of mononuclear cells and / or polyclonally activated mononuclear immunological cells, preferably concanavalin A (ConA) activated mononuclear immunological cells, of the human subjects to be screened and (b) identifying the subject as a subject at risk of developing colorectal cancer by a predictive model which correlates at least one or more of the markers identified in step (a) with representative values of those identified in step (a) from samples obtained from subjects previously identified as suffering from colorectal cancer, said predictive model having been generated by training a computer with a plurality of marker profiles from those scores identified in step (a) from previously identified subjects having colorectal cancer, by machine learning on said plurality of maker profiles so as to obtain representative maker profiles associated with colorectal cancer. Please note that the combination of all of these scores, as shown in figure 1, is capable of differentiating healthy donors from patients with colorectal cancer.

[0292] A computer implemented method for screening for subjects, preferably human subjects, at risk of developing colorectal adenomas comprising: (a) receiving at least the following markers:

[0293] a) Percentage of NK CD56dimcells

[0294] b) Percentage of TIM3 cells in NK CD56bright- FOLD

[0295] c) Percentage of NKT (CD56+ CD3+) - FOLD

[0296] d) Percentage of activated NK CD56dimcells

[0297] from an isolated biological sample comprising or consisting of mononuclear cells and / or polyclonally activated mononuclear immunological cells, preferably concanavalin A (ConA) activated mononuclear immunological cells, of the human subjects to be screened and (b) identifying the subject as a subject at risk of developing colorectal adenomas by a predictive model which correlates at least the markers identified in step (a) with representative values of the same taken from samples obtained from subjects previously identified as suffering from colorectal adenomas, said predictive model having been generated by training a computer with a plurality of marker profiles from the markers identified in step (a) from previously identified subjects having colorectal adenomas, by machine learning on said plurality of marker profiles so as to obtain representative marker profiles associated with colorectal adenomas.A computer implemented method for differentiating patients with colon adenoma from patients with colorectal cancer, preferably human subjects, comprising: (a) receiving at least the following markers:

[0298] a. Percentage of activated NK CD56dimcells

[0299] b. Percentage of TIM3 cells in CD8 lymphocytes (CD3+CD8+) - FOLD c. Percentage of lymphocytes CD4 (CD3+CD4+) - FOLD

[0300] d. Percentage of TIM3 cells in CD4 lymphocytes (CD3+CD4+) - FOLD from an isolated biological sample comprising or consisting of mononuclear cells and / or concanavalin A (ConA) activated mononuclear cells, of the human subjects to be screened and (b) identifying the subject as a subject at risk of developing colorectal adenomas and / or colorectal cancer by a predictive model which correlates the markers identified in step (a) with representative values of the same taken from samples obtained from subjects previously identified as suffering from colorectal adenomas and / or colorectal cancer, said predictive model having been generated by training a computer with a plurality of maker profiles from the markers identified in step (a) from previously identified subjects having colorectal adenomas and / or colorectal cancer, by machine learning on said plurality of marker profiles so as to obtain representative marker profiles associated with colorectal adenomas and / or colorectal cancer.

[0301] Other embodiments are within the following claims and non-limiting examples. In addition, where features or aspects of the invention are described in terms of groups, those skilled in the art will recognize that the invention is also thereby described in terms of any individual member or subgroup of members of the group.

[0302] EXAMPLES

[0303] Example 1.

[0304] Materials and methods

[0305] Starting from a 9 - 10 ml of venous blood obtained in tubes using EDTA as anticoagulant, peripheral blood mononuclear cells (PBMCs) are isolated based on gradient centrifugation in Ficol at 400 g for 25 minutes at room temperature with accelerationand brake at the minimum available level provided by the centrifuge. PBMCs are isolated by aspiration of the resulting intermediate ring consisting of mononuclear immunological cells and transferred to a clean tube. These cells are immediately wash in PBS buffer to remove rests of Ficol, by centrifugation for 5 minutes at 300 g at room temperature. Pelleted cells are resuspended in complete RPMI medium including 10% of fetal calf serum. Isolated PBMCs are plated on flat-bottom p96 plates at a concentration of 200.000 cells per p96 well in 100 microliters of complete RPMI, plating three wells for non-activated conditions and three wells for activated conditions. Based on that, at the end of the protocol, up to 600.000 cells per condition will be available by pooling the three wells per condition. Cells are rested at 37^ in a humid incubator for an overnight.

[0306] The following day, cells are polyclonally stimulated with ConA for activated conditions by adding 50 microliters of complete RPMI to each well including ConA at a 4x concentration and incubating the cells for 2 hours in a humid incubator. Next, another 50 microliters of complete RPMI are added to each well including brefeldin-A at 4x concentration, leaving then the cells incubating for another 4 hours. For well corresponding to not activated conditions, the same procedures and volume of RPMI are added without including ConA and brefeldin-A. At the end of this 2 + 4 hours of incubation, plated cells are kept inside a 4^ fridge for an overnight.

[0307] The next day, the cells plated in the three wells of either activated or not activated conditions are pooled in one single round bottom vial per condition, having then two vials per individual, one corresponding to activated conditions and another corresponding to not activated conditions. This pooling is performed by centrifugation at 300 g for 5 minutes at room temperature. Pelleted cells are resuspended in 50 microliters of PBS buffer and stained in the dark for 10 minutes at room temperature with the viability dye. After this incubation, cells are washed with PBS buffer by centrifugation at 300 g for 5 minutes at room temperature.

[0308] Pelleted cells are resuspended in 50 microliters of PBS buffer and stained with the fluorescent label conjugated antibodies directed against the specific markers used in this invention.Markers used are the following:

[0309] - CD3

[0310] - CD4

[0311] - CD127

[0312] - CD25

[0313] - CD8

[0314] - CD56

[0315] - TIM-3

[0316] Cells are incubated in the dark for 15 minutes inside a 4- fridge. After this incubation, cells are washed with PBS buffer by centrifugation at 300 g for 5 minutes at room temperature.

[0317] Pelleted cells are resuspended in 100 microliters of a fixative solution, preferably 4% paraformaldehyde, an incubated for 30 minutes inside a 4- fridge. After this incubation, cells are washed twice with PBS buffer by centrifugation at 300 g for 5 minutes at room temperature.

[0318] Pelleted cells are resuspended in 200 microliters of PBS, being ready for their acquisition at the flow cytometry devise, evaluating them by analyzing the expression of the panel of markers indicated thereof. This analysis is based on the expression of the indicated markers in the two samples generated, both activated and not activated.

[0319] Attending to this scheme, three analytical conditions arise:

[0320] Not activated (denoted without specific indication)

[0321] Polyclonal stimulation (denoted as activated)

[0322] Ratio stimulated / not stimulated (denoted as FOLD)

[0323] Based on these analytical conditions and the gating strategy defined in figure 4, the variables generated by the combination of these markers, required for the identification of the pathologies under study are the following:

[0324] Percentage of NK CD56dimcells

[0325] - Percentage ofTIM3+ cells in NK CD56bright- FOLD

[0326] - Percentage of NKT (CD56+ CD3+) - FOLDPercentage of NK CD56dimcells - ConA

[0327] Percentage of TIM3 in lymphocytes (CD3+CD8+) - ConA

[0328] Percentage of lymphocytes CD4 (CD3+CD4+) - FOLD

[0329] Percentage of TIM3 in lymphocytes (CD3+CD8+) - FOLD

[0330] Percentage of TIM3 in lymphocytes (CD3+CD4+) - FOLD;

[0331] Results

[0332] The methods described above were applied to blood samples obtained from 19 healthy donors (volunteers) (HV), 16 patients suffering from colon adenoma and 21 patients suffering from colon cancer (CC). The results obtained after applying this methodology to this cohort of patients are shown in Figures 1 to 3. These figures show how the different combinations of variables included in any of the models and their corresponding B factors were capable of differentiating:

[0333] healthy donors (volunteers) (HV) from patients with colon cancer (CC) with a sensitivity between 85.71% - 76.19% and a specificity between 89.47% and 78.95% (Figure 1) when the variables and B factors indicated below:

[0334]

[0335] were applied (see example 2 for a full description);

[0336] healthy donors (volunteers) (HV) from patients with colon adenoma with a sensitivity of 93.75% and a specificity of 78.95% (Figure 2) when the variables and B factors indicated below:

[0337]

[0338] were applied (see example 2 for a full description); and

[0339] patients with colon adenoma from patients with colon cancer (CC) with a sensitivity of 100% and a specificity of 100% (Figure 3) when the variables and B factors indicated below:

[0340]

[0341] were applied (see example 2 for a full description).

[0342] Example 2

[0343] Three binary logistic regression models have been developed, just for illustrative purposes, to carry out the methods shown in the figures and in example 1:

[0344] 1. Evaluating healthy volunteers (HV) and colon cancer patients (CC).

[0345] The variables included in this model and their corresponding B factors, when 19 HV and 21 CC were evaluated, are the following:

[0346]

[0347] when applying this model for the identification of HV versus CC, the value of the Youden index as selection score is 45.80 (see figure 1).

[0348] 2. Evaluating healthy volunteers (HV) and adenoma patients (AD).

[0349] The variables included in this model and their corresponding B factors, when 19 HV and 16 AD were evaluated, are the following:

[0350]

[0351] when applying this model for the identification of HV versus AD, the value of the Youden index as selection score is 21.07 (figure 2).

[0352] 3. Evaluating adenoma (AD) and colon cancer patients (CC).

[0353] The variables included in this model and their corresponding B factors, when 16 AD and 21 CC were evaluated, are the following:

[0354]

[0355] when applying this model for the identification of AD versus CC, the value of the Youden index as selection score is 260.0 (figure 3).

[0356] To develop these models we generated and optimized a binary logistic regression model using the value of flow cytometry-generated variables as parameters. In particular, for such purpose, we used the backwards Wald stepwise procedure to iteratively remove and, if necessary, reintroduce flow cytometry parameters based on their statistical significance, yielding a final set of explanatory variables.Each retained flow cytometry parameter in the model was assigned a coefficient (B factor) indicating its contribution to class differentiation. A composite score was computed by summing the products of each retained parameter value and its corresponding B factor:

[0357] Score (paramet erix B factor*)

[0358]

[0359] The calculated scores were used to generate Receiver Operating Characteristic (ROC) curves, from which the Area Under the Curve (AUC) was derived. The AUC reflects the model's discrimination power. The final classification threshold was identified by maximizing the sum of sensitivity and specificity, known as the Youden index. This cutoff value separates the two classes with optimal performance.

[0360] Once we had the final model variables (for instance "% of activated NK CD56dim cells"), their corresponding B factors (logistic regression coefficients), and the cut-off value determined by the Youden index, the following steps can be followed to classify or stratify patients:

[0361] 1. Measure or Obtain Parameter Values

[0362] For each new patient, obtain the relevant flow cytometry measurements (the same variables that were used to build the model).

[0363] Calculate the Score

[0364] Multiply each parameter by its B factor and sum the products:

[0365] Score

[0366]

[0367] 2. Optionally Convert Score to Probability (Optional)

[0368] Although not strictly necessary for a simple binary classification, we can convert the Score to a predicted probability using the logistic function:

[0369] 1

[0370] P (

[0371] vevent

[0372] J) = - .

[0373] j _|_ g- Score3. Apply the Youden Index Cut-Off

[0374] Compare the patient's Score (or predicted probability) to the previously determined Youden index cut-off. If the Score / probability is above the cut-off, classify the patient as "positive" (e.g., higher risk or likely to have the condition). If it is below the cut-off, classify the patient as "negative" (e.g., lower risk or unlikely to have the condition).

[0375] 4. Stratify if Needed

[0376] For further clinical stratification (e.g., low-, moderate-, high-risk), additional thresholds may be defined around the cut-off. This can help tailor treatment pathways or monitoring schedules according to different risk groups.

[0377] Notwithstanding the above, a person of ordinary skill in the art would recognize that once the relevant measurable variables are determined, any classification techniques— be it logistic regression, a machine-learning model, or another statistical method— can be applied to combine those parameters into a single, clinically useful score. This is because:

[0378] Well-Established Mathematical Frameworks

[0379] Each method (e.g., logistic regression or machine-learning algorithms) is known to provide a systematic approach to weighting and combining input variables. The parameters end up contributing to a numeric score (often interpreted as a probability or risk level), which can then be thresholded or stratified.

[0380] Common Knowledge for Data Analysis

[0381] These methods are standard in the fields of statistics, data science, and clinical research, and are widely taught and referenced for risk modeling and classification. Hence, it would be apparent to a skilled person that any of these approaches can transform multiple input variables into a single, discriminative score.

[0382] Flexibility in Model ChoiceSince the core task— combining inputs to output a risk category— is essentially the same regardless of the mathematical technique, a skilled practitioner would know that multiple, interchangeable methods exist to accomplish the same end goal. Once the measurable parameters are identified, selection among such well-known methods typically depends on practical considerations (available data, software, computational resources) rather than conceptual differences in their ability to produce a useful score. Established Track Record in Clinical Risk Stratification

[0383] Logistic regression, tree-based models, and other machine-learning approaches have historically been used to derive risk scores in numerous medical contexts (e.g., cardiovascular risk, cancer prognostics). The success of these methods in related problems leads a skilled person to conclude that they will likewise be effective for generating a risk score in the scenario at hand.

Claims

CLAIMS1. A method for screening or identifying a subject at risk of having colorectal cancer and / or colorectal adenomas, the method comprising:a. Determining or receiving at least four or more of the following plurality of measurable parameters:i. Percentage of NK CD56dimcells;ii. Percentage of TIM3 cells in NK CD56bright- FOLD;iii. Percentage of NKT (CD56+ CD3+) - FOLD;iv. Percentage of TIM3 in activated CD8 lymphocytes (CD3+CD8+); v. Percentage of TIM3 in CD8 lymphocytes (CD3+CD8+) - FOLD; vi. Percentage of CD4 lymphocytes (CD3+CD4+) - FOLD;vii. Percentage of activated NK CD56dimcells; andviii. Percentage of TIM3 in CD4 lymphocytes (CD3+CD4+) - FOLD;from an isolated blood sample of the subject comprising mononuclear immunological cells and / or polyclonally activated mononuclear immunological cells, preferably concanavalin A (ConA) activated mononuclear immunological cells, as measured with flow cytometry;b. obtaining a composite score, the composite score having been generated from the plurality of measurable parameters determined or received in step (a) by applying a classification model configured to estimate a weighted contribution of each parameter, wherein the classification model is selected from: - logistic regression, - a supervised machinelearning model selected from at least a neural network or a random forest, or - a linear or non-linear statistical method selected from at least partial least squares regression or linear discriminant analysis, and providing that composite score for subsequent comparison; c. comparing the composite score to one or more predefined thresholds, each threshold being derived from either (i) a reference population or (ii) a reference value that permits classification of individuals as being at riskof having the disease versus not at risk, wherein a composite score above a given threshold is classified as "higher risk", and a composite score below that threshold is classified as "lower risk"; andd. categorizing the subject into one of at least two discrete risk categories based on the comparison in step (c), thereby providing a risk stratification output indicative of whether the subject's risk of having colorectal cancer and / or colorectal adenomas is increased or decreased relative to the reference.

2. The method of claim 1, wherein at least one of the predefined thresholds of step (c) is determined by maximizing an index that accounts for the sensitivity and specificity of the classification, preferably the Youden Index defined as Sensitivity + Specificity - 1, such that the threshold permits classification of individuals as being at risk of having the disease or condition versus those not having the disease or condition.

3. The method of claim 1 or 2, wherein the method is for screening for subjects, preferably human subjects, at risk of having colorectal adenomas, wherein step a) determines or receives at least all of the following plurality of measurable parameters:i. Percentage of NK CD56dimcells;ii. Percentage of TIM3 cells in NK CD56bright- FOLD;iii. Percentage of NKT (CD56+ CD3+) - FOLD; andiv. Percentage of activated NK CD56dimcells;and wherein step d) of the method categorizes the subject into one of at least two discrete risk categories based on the comparison in step (c), thereby providing a risk stratification output indicative of whether the subject' risk of having or developing colorectal adenomas is increased ("higher risk"), the subject is at risk of having colorectal adenomas, or decreased "lower risk", the subject is at low or no risk of having colorectal adenomas, relative to the reference of step c).

4. The method of claim 1 or 2, wherein the method is for screening for subjects, preferably human subjects, at risk of having colorectal cancer, and step a) determines or receives at least all of the following plurality of measurable parameters:i. Percentage of NK CD56dimcells;ii. Percentage of TIM3 cells in NK CD56bright- FOLD;iii. Percentage of NKT (CD56+ CD3+) - FOLD;iv. Percentage of TIM3 in activated CD8 lymphocytes (CD3+CD8+); v. Percentage of TIM3 in CD8 lymphocytes (CD3+CD8+) - FOLD; and vi. Percentage of CD4 lymphocytes (CD3+CD4+) - FOLD;and wherein step d) of the method categorizes the subject into one of at least two discrete risk categories based on the comparison in step (c), thereby providing a risk stratification output indicative of whether the subject's risk of having colorectal cancer is increased ("higher risk"), the subject is at risk of having colorectal cancer, or decreased "lower risk", the subject is at low or no risk of having colorectal cancer, relative to the reference of step c).

5. The method of claim 3, wherein the method further comprises carrying out a confirmatory method for classifying subjects classified as at risk of having colorectal adenomas according to the method of claim 3, as individuals at risk of having colorectal cancer or at risk of having colorectal adenomas, the method comprising:a. Determining or receiving at least four or more of the following plurality of measurable parameters:i. Percentage of CD4 lymphocytes (CD3+CD4+) - FOLDii. Percentage of activated NK CD56dimcellsiii. Percentage of TIM3 in lymphocytes (CD3+CD8+) - FOLD iv. Percentage of TIM3 in lymphocytes (CD3+CD4+) - FOLDfrom an isolated blood sample of the subject, classified as at risk of having colorectal adenomas according to the method of claim 3, comprising mononuclear immunological cells and / or polyclonally activated mononuclear immunological cells, preferably concanavalin A (ConA) activated mononuclear immunological cells, as measured with flow cytometry;b. obtaining a composite score, the composite score having been generated from the plurality of measurable parameters determined or received in step (a) by applying a classification model configured to estimate a weighted contribution of each parameter, wherein the classification model is selected from: - logistic regression, - a supervised machinelearning model selected from at least a neural network or a random forest, or - a linear or non-linear statistical method selected from at least partial least squares regression or linear discriminant analysis, and providing that composite score for subsequent comparison; c. comparing the composite score to one or more predefined thresholds, each threshold being derived from a reference value that permits classification of individuals as being at risk of having colorectal cancer versus colorectal adenomas, wherein a composite score above a given threshold is classified as "higher risk" of having colorectal cancer, and a composite score below that threshold is classified as "lower risk" of having colorectal cancer thereby confirming the higher risk of the subject having colorectal adenomas; andd. categorizing the individual into one of at least two discrete risk categories based on the comparison in step (c), thereby providing a risk stratification output indicative of whether the individual's risk of having colorectal cancer is increased or decreased relative to the reference.

6. The method of claim 4, wherein the method further comprises carrying out a confirmatory method for classifying subjects classified as at risk of having colorectal cancer according to the method of claim 4, as individuals at risk of having colorectal cancer or at risk of having colorectal adenomas, the method comprising:a. Determining or receiving at least four or more of the following plurality of measurable parameters:i. Percentage of CD4 lymphocytes (CD3+CD4+) - FOLDii. Percentage of activated NK CD56dimcellsiii. Percentage of TIM3 in lymphocytes (CD3+CD8+) - FOLD iv. Percentage of TIM3 in lymphocytes (CD3+CD4+) - FOLD from an isolated blood sample of the subject, classified as at risk of having colorectal cancer according to the method of claim 4, comprising mononuclear immunological cells and / or polyclonally activated mononuclear immunological cells, preferably concanavalin A (ConA) activated mononuclear immunological cells, as measured with flow cytometry;b. obtaining a composite score, the composite score having been generated from the plurality of measurable parameters determined or received in step (a) by applying a classification model configured to estimate a weighted contribution of each parameter, wherein the classification model is selected from: - logistic regression, - a supervised machinelearning model selected from at least a neural network or a random forest, or - a linear or non-linear statistical method selected from at least partial least squares regression or linear discriminant analysis, and providing that composite score for subsequent comparison; c. comparing the composite score to one or more predefined thresholds, each threshold being derived from a reference value that permits classification of individuals as being at risk of having colorectal cancer versus colorectal adenomas, wherein a composite score above a given threshold is classified as "higher risk" of having colorectal cancer thereby confirming the risk of claim 4, and a composite score below that threshold is classified as "lower risk" of having colorectal cancer thereby indicating a higher risk that the subject has colorectal adenomas; and d. categorizing the subject into one of at least two discrete risk categories based on the comparison in step (c), thereby providing a risk stratificationoutput indicative of whether the subject's risk of having colorectal cancer is increased or decreased relative to the reference.

7. The method according to any one of claims 1 to 6, wherein the classification model is logistic regression, and the one or more weighting factors are obtained by applying a Backwards Wald stepwise selection procedure to the logistic regression coefficients.

8. The method according to any one of claims 1 to 6, wherein the classification model is a supervised machine-learning model, and the one or more weighting factors are learned by a machine-learning algorithm selected from at least one of: a random forest model, a support vector machine, a neural network, or a gradient boosting model.

9. The method of any one of claims 1 to 4, wherein:a. Step a) of the method determines or receives the plurality of measurable parameters as defined in any one of claims 1 to 4;b. Step b) calculates, via a logistic regression model, a combined risk score by multiplying each measurable parameter by a corresponding coefficient (B factor) and summing the products;c. Step c) compares the combined risk score to at least one predefined threshold to categorize the individual into a higher-risk group or a lower- risk group according to any one of claims 1 to 4; andd. Step d) provides a risk stratification output indicating whether the individual is at increased or decreased risk of the disease or condition based on said comparison.

10. The method of claim 9, wherein at least one of the predefined thresholds of step (c) is determined by maximizing an index that accounts for the sensitivity and specificity of the classification, preferably the Youden Index defined as Sensitivity + Specificity - 1, such that the threshold permits classification of individuals asbeing at risk of having the disease or condition versus those not having the disease or condition.

11. The method according to any one of claims 9 to 10, wherein the method is for screening for subjects, preferably human subjects, at risk of having colorectal adenomas as defined in claim 3, wherein the plurality of measurable parameters of step a) included in this method and their corresponding B factors of step b), are the following:and wherein the value of the Youden index as the predefined thresholds is 21.07.

12. The method according to any one of claims 9 to 10, wherein the method is for screening for subjects, preferably human subjects, at risk of having colorectal cancer as defined in claim 4, wherein the plurality of measurable parameters of step a) included in this method and their corresponding B factors of step b), are the following:and wherein the value of the Youden index as the predefined thresholds is 45.80.

13. The method according to any one of claims 5 to 6, wherein the further confirmatory method comprises the following steps:a. Step a) of the method determines or receives the plurality of measurable parameters as defined in any one of claims 5 to 6;b. Step b) calculates, via a logistic regression model, a combined risk score by multiplying each measurable parameter by a corresponding coefficient (B factor) and summing the products;c. Step c) compares the combined risk score to at least one predefined threshold to categorize the individual into a higher-risk group or a lower- risk group according to any one of claims 5 to 6; andd. Step d) provides a risk stratification output indicating whether the individual is at increased or decreased risk of the disease or condition based on said comparison.

14. The method of claim 13, wherein at least one of the predefined thresholds of step (c) is determined by maximizing an index that accounts for the sensitivity and specificity of the classification, preferably the Youden Index defined as Sensitivity + Specificity - 1, such that the threshold permits classification of individuals as defined in any one of claims 5 to 6.

15. The method according to any one of claims 13 to 14, wherein the plurality of measurable parameters included in this method and their corresponding B factors, are the following:and wherein the value of the Youden index as the predefined thresholds is 260.

16. The method according to anyone of claims 1 to 15, wherein the risk stratification output includes a recommended clinical intervention selected from at least one of:a. additional diagnostic testing,b. initiation or adjustment of a therapeutic regimen, orc. an increase in monitoring frequency.

17. A system configured to implement the method of any one of claims 1-16, the system comprising:a. Optionally an analysis module configured to perform step (a) of any one of claims 1-16 by obtaining the plurality of measurable parameters; b. a processing module configured to perform step (b) of any one of claims 1-16 by calculating a composite score derived from the measurable parameters using one or more weighting factors determined from a classification model selected from at least one of a logistic regression, a machine-learning algorithm, or any other statistical method capable of identifying parameter contributions; andc. perform step (c) of any one of claims 1-16 by comparing the composite score to one or more decision thresholds to categorize the individual into at least a higher-risk group or a lower-risk group; andd. an output interface configured to perform step (d) of any one of claims 1- 16 by providing a risk stratification output indicating whether the individual is at increased or decreased risk of the disease or condition based on the categorization.

18. The system of claim 19, wherein the analysis module comprises a flow cytometer for obtaining flow cytometry measurements as part of the plurality ofmeasurable parameters, thereby performing step (a) of any one of claims 1-16 for such measurements.

19. The system of claim 17 or 18, wherein the processing module is further configured to generate a ROC curve for the composite score from a training or validation dataset, determine an optimal threshold by maximizing at least one performance metric selected from sensitivity, specificity, or the Youden index, and store said optimal threshold for subsequent comparisons of new individuals' composite scores.

20. The system of any one of claims 17 to 19, further comprising a user interface that displays a visual risk index indicating one of a plurality of risk categories after performing steps (b) and (c) of any one of claims 1-16, and wherein the visual risk index corresponds to step (d) of any one of claims 1-16.

21. A non-transitory computer-readable medium storing instructions that, when executed by one or more processors, cause the one or more processors to perform steps (a )— (d) of the method of any one of claims 1-16, the instructions comprising:a. instructions to perform step (a) of any one of claims 1-16 by obtaining a plurality of measurable parameters;b. instructions to perform step (b) of any one of claims 1-16 by calculating a composite score derived from the measurable parameters using one or more weighting factors determined from a classification model selected from at least one of a logistic regression, a machine-learning algorithm, or any other statistical method capable of identifying parameter contributions;c. instructions to perform step (c) of any one of claims 1-16 by comparing the composite score to one or more decision thresholds to categorize the individual into at least a higher-risk group or a lower-risk group; andd. instructions to perform step (d) of any one of claims 1-16 by providing a risk stratification output indicating whether the individual is at increased or decreased risk of the disease or condition based on the categorization.

22. The non-transitory computer-readable medium of claim 21, wherein the instructions further cause the one or more processors to update the classification model by incorporating new data from previously classified individuals, thereby dynamically recalibrating the weighting factors over time to improve the performance of steps (b) and (c) of any one of claims 1-16.

23. The non-transitory computer-readable medium of claim 21 or 22, wherein the instructions further cause the one or more processors to:a. generate at least one ROC curve based on historical composite scores and associated clinical outcomes,b. calculate an optimal threshold by applying a performance metric selected from the Youden index, the Fl score, or precision-recall analysis, c. and store the optimal threshold for subsequent execution of step (c) of any one of claims 1-18 when classifying additional individuals.

Images