GIRARD ROMUALD (US)
| WHAT IS CLAIMED: 1. A method for treating cerebral cavernous malformations (CCMs) in a subject, the method comprising administering a therapy to a subject determined to have differential expression of one or more biomarkers selected from the biomarkers listed in Tables S1-S8 and S13 in a biological sample from the subject. 2. A method for treating cerebral cavernous malformations (CCMs) in a subject, the method comprising administering a therapy to a subject determined to have differential expression of one or more biomarkers selected from let-7e-5p, miR-93-5p, miR-20b-5p, miR- 128-3p, miR-9-5p, IL-10, Tsp-2, IL-2, and TNFR1. 3. The method of claim 2, wherein the subject has been determined to have to have differential expression of one or more of let-7e-5p, miR-93-5p, miR-20b-5p, miR-128-3p, IL- 10, and Tsp-2. 4. The method of claim 3, wherein the subject has been determined to have to have differential expression of let-7e-5p, miR-93-5p, miR-20b-5p, miR-128-3p, IL-10, and Tsp-2. 5. The method of claim 4, wherein the subject has been determined to have a canonical value of greater than 0, wherein the canonical value is equal to -114.9*[let-7e-5p]r + 359.5*[miR-93-5p]r – 118.2*[miR-20b-5p]r+ 88.9*[miR-128-3p]r + 199.5*[IL-10] – 19.7*[Tsp-2] – 463.2, wherein [X]r denotes the relative quantification value for each miRNA biomarker. 6. The method of claim 2, wherein the subject has been determined to have to have differential expression of one or more of miR-9-5p, miR-93-5p, IL-2, TNFR1, and Tsp-2. 7. The method of claim 6, wherein the subject has been determined to have to have differential expression of miR-9-5p, miR-93-5p, IL-2, TNFR1, and Tsp-2. 8. The method of claim 7, wherein the subject has been determined to have a canonical value of less than 15, wherein the canonical value is equal to 2.1*[miR-9-5p]r – 3.0*[miR-93- 5p]r – 0.01*[IL-2] + 0.1*[TNFR1] + 0.5*[Tsp-2], wherein [X]r denotes the relative quantification value for each miRNA biomarker. 9. The method of any one of claims 1-8, wherein the biomarker is increased relative to a control. 10. The method of any one of claims 1-8, wherein the biomarker is decreased relative to a control. 11. The method of claim 1, wherein the subject has been diagnosed with CCM. 12. The method of claim 1, wherein the subject has not been diagnosed with CCM. 13. The method of any one of claims 1-12, wherein CCM is further defined as familial CCM. 14. The method of claim 13, wherein the familial CCM is further define as familial-CCM1. 15. The method of claim 13, wherein the familial CCM is further define as familial-CCM3. 16. The method of any one of claims 1-16, wherein the subject is being treated with a therapy for CCM. 17. The method of any one of claims 1-16, wherein the sample from the subject comprises a tissue sample, a blood sample, a whole blood sample, a fractionated sample, a plasma sample, a fecal sample, or a urine sample. 18. The method of any one of claims 1-17, wherein the sample from the subject comprises a plasma sample from the subject. 19. The method of any one of claims 1-18, wherein at least the expression level of miR-9- 5p was determined. 20. The method of any one of claims 1-19, wherein at least the expression level of miR-93- 5p was determined. 21. The method of any one of claims 1-20, wherein at least the expression level of IL-2 was determined. 22. The method of any one of claims 1-21, wherein at least the expression level of TNFR1 was determined. 23. The method of any one of claims 1-22, wherein at least the expression level of Tsp-2 was determined. 24. The method of any one of claims 1-23, wherein at least the expression level of let-7e- 5p was determined. 25. The method of any one of claims 1-24, wherein at least the expression level of miR- 20b-5p was determined. 26. The method of any one of claims 1-25, wherein at least the expression level of miR- 128-3p was determined. 27. The method of any one of claims 1-26, wherein at least the expression level of IL-10 was determined. 28. The method of any of claims 1-27, wherein the expression level of no other biomarker in the sample was determined. 29. The method of any one of claims 1-28, wherein the determined biomarker comprises one or more of let-7e-5p, miR-93-5p, miR-20b-5p, miR-128-3p, and miR-9-5p, and wherein the expression level(s) of the biomarker(s) were determined by evaluating the RNA level of the biomarker. 30. The method of any one of claims 1-29, wherein the biomarker comprises one or more of IL-10, Tsp-2, IL-2, and TNFR1, and wherein the expression level(s) of the biomarker(s) were determined by evaluating the protein level of the biomarker. 31. The method of any one of claims 9-30, wherein the control comprises the expression or protein level of the biomarker in samples from subjects identified as low risk or in subjects identified as not having CCM. 32. The method of any one of claims 9-30, wherein the control comprises the expression or protein level of the biomarker in samples from subjects identified as high risk or in subjects identified as having CCM. 33. The method of any one of claims 1-32, wherein the therapy comprises a gene therapy, a B-cell immunomodulation therapy, surgical excision of a CCM lesion, or combinations thereof. 34. A method for evaluating a subject comprising measuring the level of expression of one or more biomarkers selected from the biomarkers listed in Tables S1-S8 and S13 in a biological sample from the subject. 35. A method for evaluating a subject comprising measuring the level of expression of one or more of let-7e-5p, miR-93-5p, miR-20b-5p, miR-128-3p, miR-9-5p, IL-10, Tsp-2, IL-2, and TNFR1 in a sample from the subject. 36. The method of claim 35, wherein the expression levels of one or more of let-7e-5p, miR-93-5p, miR-20b-5p, miR-128-3p, IL-10, and Tsp-2 are evaluated in the sample. 37. The method of claim 36, wherein the expression levels of let-7e-5p, miR-93-5p, miR- 20b-5p, miR-128-3p, IL-10, and Tsp-2 are evaluated in the sample. 38. The method of claim 37, wherein the method further comprises determining a canonical value; wherein the canonical value was determined to be greater than 0, wherein the canonical value is equal to -114.9*[let-7e-5p]r + 359.5*[miR-93-5p]r – 118.2*[miR-20b-5p]r + 88.9*[miR-128-3p]r + 199.5*[IL-10] – 19.7*[Tsp-2] – 463.2, wherein [X]r denotes the relative quantification value for each miRNA biomarker. 39. The method of claim 35, wherein the expression levels of one or more of miR-9-5p, miR-93-5p, IL-2, TNFR1, and Tsp-2 are evaluated in the sample. 40. The method of claim 39, wherein the expression levels of miR-9-5p, miR-93-5p, IL-2, TNFR1, and Tsp-2 are evaluated in the sample. 41. The method of claim 40, wherein the method further comprises determining a canonical value; wherein the canonical value was determined to be less than 15, wherein the canonical value is equal to 2.1*[miR-9-5p]r – 3.0*[miR-93-5p]r – 0.01*[IL-2] + 0.1*[TNFR1] + 0.5*[Tsp-2], wherein [X]r denotes the relative quantification value for each miRNA biomarker. 42. The method of any one of claims 34-41, wherein the biomarker is increased relative to a control. 43. The method of any one of claims 34-41, wherein the biomarker is decreased relative to a control. 44. The method of any one of claims 34-44, wherein the subject has been diagnosed with CCM. 45. The method of any one of claims 34-44, wherein the subject has not been diagnosed with CCM. 46. The method of any one of claims 34-45, wherein CCM is further defined as familial CCM. 47. The method of claim 46, wherein the familial CCM is further define as familial-CCM1. 48. The method of claim 46, wherein the familial CCM is further define as familial-CCM3. 49. The method of any one of claims 34-48, wherein the subject is being treated with a therapy for CCM. 50. The method of any one of claims 34-49, wherein the sample from the subject comprises a tissue sample, a blood sample, a whole blood sample, a fractionated sample, a plasma sample, a fecal sample, or a urine sample. 51. The method of any one of claims 34-50, wherein the sample from the subject comprises a plasma sample from the subject. 52. The method of any one of claims 34-51, wherein at least the expression level of miR- 9-5p is evaluated. 53. The method of any one of claims 34-52, wherein at least the expression level of miR- 93-5p is evaluated. 54. The method of any one of claims 34-53, wherein at least the expression level of IL-2 is evaluated. 55. The method of any one of claims 34-54, wherein at least the expression level of TNFR1 is evaluated. 56. The method of any one of claims 34-55, wherein at least the expression level of Tsp-2 is evaluated. 57. The method of any one of claims 34-56, wherein at least the expression level of let-7e- 5p is evaluated. 58. The method of any one of claims 34-57, wherein at least the expression level of miR- 20b-5p is evaluated. 59. The method of any one of claims 34-58, wherein at least the expression level of miR- 128-3p is evaluated. 60. The method of any one of claims 34-59, wherein at least the expression level of IL-10 is evaluated. 61. The method of any of claims 34-60, wherein the expression level of no other biomarker in the biological sample are evaluated. 62. The method of any one of claims 34-61, wherein the biomarker comprises one or more of let-7e-5p, miR-93-5p, miR-20b-5p, miR-128-3p, and miR-9-5p, and wherein the expression level(s) of the biomarker(s) were measured by evaluating the RNA level of the biomarker. 63. The method of any one of claims 34-62, wherein the biomarker comprises one or more of IL-10, Tsp-2, IL-2, and TNFR1, and wherein the expression level(s) of the biomarker(s) were measured by evaluating the protein level of the biomarker. 64. The method of any one of claims 42-63, wherein the control comprises the expression or protein level of the biomarker in samples from subjects identified as low risk or in subjects identified as not having CCM. 65. The method of any one of claims 42-63, wherein the control comprises the expression or protein level of the biomarker in samples from subjects identified as high risk or in subjects identified as having CCM. 66. The method of any one of claims 34-65, wherein the method further comprises treating the subject with a CCM therapy. 67. The method of claim 66, wherein the therapy comprises a gene therapy, a B-cell immunomodulation therapy, surgical excision of a CCM lesion, or combinations thereof. 68. A method of prognosing, diagnosing, and/or determining the efficacy of a therapy in a subject, the method comprising: a) measuring the level of expression of one or more biomarkers listed in Tables S1-S8 and S13 in a sample from the subject; b) i)comparing the level(s) of expression of the one or more biomarkers to a control sample(s) or control level(s) of expression; or ii) calculating a canonical value; and, c) prognosing, diagnosing, and/or determining the efficacy of a therapy based on the levels of measured expression and/or the canonical value. 69. A method of prognosing, diagnosing, and/or determining the efficacy of a therapy in a subject, the method comprising: a) measuring the level of expression of one or more biomarkers selected from let- 7e-5p, miR-93-5p, miR-20b-5p, miR-128-3p, miR-9-5p, IL-10, Tsp-2, IL-2, and TNFR1 in a sample from the subject; b) i)comparing the level(s) of expression of the one or more biomarkers to a control sample(s) or control level(s) of expression; and/or ii) calculating a canonical value; and, c) prognosing, diagnosing, and/or determining the efficacy of a therapy based on the levels of measured expression and/or the canonical value. 70. The method of claim 69, wherein the expression levels of one or more of let-7e-5p, miR-93-5p, miR-20b-5p, miR-128-3p, IL-10, and Tsp-2 are measured in the sample. 71. The method of claim 70, wherein the expression levels of let-7e-5p, miR-93-5p, miR- 20b-5p, miR-128-3p, IL-10, and Tsp-2 are measured in the sample. 72. The method of claim 71, wherein the method comprises determining a canonical value; wherein the canonical value is equal to -114.9*[let-7e-5p]r + 359.5*[miR-93-5p]r – 118.2*[miR-20b-5p]r + 88.9*[miR-128-3p]r + 199.5*[IL-10] – 19.7*[Tsp-2] – 463.2, wherein [X]r denotes the relative quantification value for each miRNA biomarker. 73. The method of claim 72, wherein the subject is diagnosed as having CCM or prognosed as high risk when the canonical value is greater than 0. 74. The method of claim 72 or 73, wherein the efficacy of the therapy is determined to be ineffective when the canonical value is greater than 0. 75. The method of claim 73 or 74, wherein the method further comprises treating the subject with a CCM therapy. 76. The method of claim 75, wherein the therapy comprises a gene therapy, a B-cell immunomodulation therapy, surgical excision of a CCM lesion, or combinations thereof. 77. The method of claim 69, wherein the expression levels of one or more of miR-9-5p, miR-93-5p, IL-2, TNFR1, and Tsp-2 are measured in the sample. 78. The method of claim 77, wherein the expression levels of miR-9-5p, miR-93-5p, IL-2, TNFR1, and Tsp-2 are evaluated in the sample. 79. The method of claim 78, wherein the method further comprises determining a canonical value; wherein the canonical value is equal to 2.1*[miR-9-5p]r – 3.0*[miR-93-5p]r – 0.01*[IL- 2] + 0.1*[TNFR1] + 0.5*[Tsp-2], wherein [X]r denotes the relative quantification value for each miRNA biomarker. 80. The method of claim 79, wherein the subject is diagnosed as having CCM or prognosed as high risk when the canonical value is less than 15. 81. The method of claim 79 or 80, wherein the efficacy of the therapy is determined to be ineffective when the canonical value is less than 15. 82. The method of claim 80 or 81, wherein the method further comprises treating the subject with a CCM therapy. 83. The method of claim 82, wherein the therapy comprises a gene therapy, a B-cell immunomodulation therapy, surgical excision of a CCM lesion, or combinations thereof. 84. The method of any one of claims 68-83, wherein the subject has been diagnosed with CCM. 85. The method of any one of claims 68-83, wherein the subject has not been diagnosed with CCM. 86. The method of any one of claims 68-85, wherein CCM is further defined as familial CCM. 87. The method of claim 86, wherein the familial CCM is further define as familial-CCM1. 88. The method of claim 86, wherein the familial CCM is further define as familial-CCM3. 89. The method of any one of claims 68-88, wherein the subject is being treated with a therapy for CCM. 90. The method of any one of claims 68-89, wherein the sample from the subject comprises a tissue sample, a blood sample, a whole blood sample, a fractionated sample, a plasma sample, a fecal sample, or a urine sample. 91. The method of any one of claims 68-90, wherein the sample from the subject comprises a plasma sample from the subject. 92. The method of any one of claims 68-91, wherein at least the expression level of miR- 9-5p is measured. 93. The method of any one of claims 68-92, wherein at least the expression level of miR- 93-5p is measured. 94. The method of any one of claims 68-93, wherein at least the expression level of IL-2 is measured. 95. The method of any one of claims 68-94, wherein at least the expression level of TNFR1 is measured. 96. The method of any one of claims 68-95, wherein at least the expression level of Tsp-2 is measured. 97. The method of any one of claims 68-96, wherein at least the expression level of let-7e- 5p is measured. 98. The method of any one of claims 68-97, wherein at least the expression level of miR- 20b-5p is measured. 99. The method of any one of claims 68-98, wherein at least the expression level of miR- 128-3p is measured. 100. The method of any one of claims 68-99, wherein at least the expression level of IL-10 is measured. 101. The method of any of claims 68-100, wherein the expression level of no other biomarker in the biological sample are measured. 102. The method of any one of claims 68-101, wherein the biomarker is increased relative to a control. 103. The method of any one of claims 68-102, wherein the biomarker is decreased relative to a control. 104. The method of any one of claims 68-103, wherein the biomarker comprises one or more of let-7e-5p, miR-93-5p, miR-20b-5p, miR-128-3p, and miR-9-5p, and wherein the expression level(s) of the biomarker(s) were measured by evaluating the RNA level of the biomarker. 105. The method of any one of claims 68-104, wherein the biomarker comprises one or more of IL-10, Tsp-2, IL-2, and TNFR1, and wherein the expression level(s) of the biomarker(s) were measured by evaluating the protein level of the biomarker. 106. The method of any one of claims 68-105, wherein the control comprises the expression or protein level of the biomarker in samples from subjects identified as low risk or in subjects identified as not having CCM. 107. The method of any one of claims 68-105, wherein the control comprises the expression or protein level of the biomarker in samples from subjects identified as high risk or in subjects identified as having CCM. 108. A kit comprising 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, or 15 detection agents for determining expression levels of biomarkers for CCM, wherein the biomarkers comprise one or more listed in Tables S1-S8 and S13. 109. A kit comprising 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, or 15 detection agents for determining expression levels of biomarkers for CCM, wherein the biomarkers comprise one or more of let-7e-5p, miR-93-5p, miR-20b-5p, miR-128-3p, miR-9-5p, IL-10, Tsp-2, IL-2, and TNFR1. 110. The kit of claim 108 or 109, wherein the kit further comprises one or more negative or positive control samples and/or control detection agents. 111. The kit of any one of claims 108-110, wherein the kit further comprises instructions for use. 112. The kit of any one of claims 108-110, wherein the CCM biomarkers comprise one or more of let-7e-5p, miR-93-5p, miR-20b-5p, miR-128-3p, IL-10, and Tsp-2. 113. The kit of claim 112, wherein the CCM biomarkers consist of let-7e-5p, miR-93-5p, miR-20b-5p, miR-128-3p, IL-10, and Tsp-2. 114. The kit of claim 112, wherein the CCM biomarkers comprise or consist of let-7e-5p, miR-93-5p, miR-20b-5p, miR-128-3p, IL-10, Tsp-2 and/or one or more controls. 115. The kit of any one of claims 111-114, wherein the instructions comprises evaluating a canonical value, wherein the canonical value is equal to -114.9*[let-7e-5p]r + 359.5*[miR-93- 5p]r – 118.2*[miR-20b-5p]r + 88.9*[miR-128-3p]r + 199.5*[IL-10] – 19.7*[Tsp-2] – 463.2, wherein [X]r denotes the relative quantification value for each miRNA biomarker. 116. The kit of any one of claims 108-109, wherein the CCM biomarkers comprise one or more of miR-9-5p, miR-93-5p, IL-2, TNFR1, and Tsp-2. 117. The kit of claim 116, wherein the CCM biomarkers consist of miR-9-5p, miR-93-5p, IL-2, TNFR1, and Tsp-2. 118. The kit of claim 116, wherein the CCM biomarkers comprise or consist of miR-9-5p, miR-93-5p, IL-2, TNFR1, Tsp-2.and/or one or more controls. 119. The kit of any one of claims 111-118, wherein the instructions comprises evaluating a canonical value, wherein the canonical value is equal to 2.1*[miR-9-5p]r – 3.0*[miR-93-5p]r – 0.01*[IL-2] + 0.1*[TNFR1] + 0.5*[Tsp-2], wherein [X]r denotes the relative quantification value for each miRNA biomarker. 120. The kit of any one of claims 109-119, wherein the kit comprises reagents for isolation and/or amplification of biomarkers from a biological sample. 121. The kit of claim 120, wherein the biological sample comprises a tissue sample, a blood sample, a whole blood sample, a fractionated sample, a plasma sample, a fecal sample, or a urine sample. 122. The kit of claim 121, wherein the biological sample comprises a plasma sample. 123. The kit of any one of claims 120-122, wherein the biological sample is from a subject. 124. The kit of claim 123, wherein the subject has been diagnosed with CCM. 125. The kit of claim 123, wherein the subject has not been diagnosed with CCM. 126. The kit of claim 124 or 125, wherein CCM is further defined as familial CCM. 127. The kit of claim 126, wherein the familial CCM is further define as familial-CCM1. 128. The kit of claim 126, wherein the familial CCM is further define as familial-CCM3. 129. The kit of any one of claims 123-128, wherein the subject is being treated with a therapy for CCM. 130. The kit of claim 129, wherein the therapy comprises a gene therapy, a B-cell immunomodulation therapy, surgical excision of a CCM lesion, or combinations thereof. 131. A method for treating cerebral cavernous malformations (CCMs) in a subject, the method comprising administering one or more miRNA modulators, wherein the miRNA comprises let-7e-5p, miR-93-5p, miR-20b-5p, miR-128-3p, and/or miR-9-5p. 132. The method of claim 131, wherein the modulator is an agomer of the miRNA. 133. The method of claim 131, wherein the modulator is an antagomir of the miRNA. 134. The method of any one of claims 131-133, wherein the modulator is linked to a targeting moiety. 135. The method of claim 134, wherein the targeting moiety is an aptamer. 136. The method of claim 134 or 135, wherein the targeting moiety delivers the modulator to a specific cell type or tissue. 137. The method of any one of claims 131-136, wherein the modulator comprise a modified oligonucleotide. 138. The method of claim 137, wherein the modified oligonucleotide comprises a locked nucleotide, ethylene bridged nucleotide, a peptide nucleic acid, or a 5’(E)-vinyl-phosphonate (VP) modification. 139. The method of any one of claims 131-138, wherein the modulator is administered by systemic, intravenous, intramuscular, or parenteral administration. 140. The method of any one of claims 131-139, wherein the subject has been diagnosed with CCM. 141. The method of any one of claims 131-139, wherein the subject has not been diagnosed with CCM. 142. The method of any one of claims 131-141, wherein CCM is further defined as familial CCM. 143. The method of claim 142, wherein the familial CCM is further define as familial- CCM1. 144. The method of claim 13, wherein the familial CCM is further define as familial-CCM3. 145. The method of any one of claims 131-144, wherein the subject is being treated with a therapy for CCM. 146. The method of claim 145, wherein the therapy comprises a gene therapy, a B-cell immunomodulation therapy, surgical excision of a CCM lesion, or combinations thereof. 147. The method of any one of claims 131-146, wherein the subject has been determined to have differential expression of one or more biomarkers selected from the biomarkers listed in Tables S1-S8 and S13 in a biological sample from the subject. 148. The method of any one of claims 131-147, wherein the subject has been determined to have differential expression of one or more biomarkers selected from let-7e-5p, miR-93-5p, miR-20b-5p, miR-128-3p, miR-9-5p, IL-10, Tsp-2, IL-2, and TNFR1. |
Table S2: Differentially Expressed Plasma miRNAs in Ccm1 Heterozygous vs. Wild Type Mice (p<0.1, FDR-corrected)
Table S3: Differentially Expressed Plasma miRNAs in Ccm3 Homozygous vs. Wild Type Mice (p<0.1, FDR-corrected) Table S4: Differentially Expressed Plasma miRNAs in Ccm3 Heterozygous vs. Wild Type Mice (p<0.1, FDR-corrected) Table S5: Differentially Expressed Plasma miRNAs in Ccm1 Homozygous vs. Ccm3 Homozygous Mice (p<0.1, FDR-corrected)
Table S6: Differentially Expressed Plasma miRNAs in CCM1 Patients vs. Healthy non-CCM Controls (p<0.05, non FDR-corrected) Table S7: Differentially Expressed Plasma miRNAs in CCM3 Patients vs. Healthy non-CCM Controls (p<0.05, non FDR-corrected) Table S8: Differentially Expressed Plasma miRNAs in CCM1 vs. CCM3 Patients (p<0.05, non FDR-corrected) Table S13: Differentially Expressed Plasma miRNAs in New Lesion vs. No New Lesion Patients (p<0.05, non FDR-corrected)
XI. Kits [0125] The present invention also concern kits containing compositions of the invention or compositions to implement methods of the invention. Kits can be used to evaluate one or more biomarkers. A kit may contain, contains at least or contains at most 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 100, 500, 1,000 or more probes, primers or primer sets, synthetic molecules, antibodies, or inhibitors, or any value or range and combination derivable therein. Also provided are kits for evaluating biomarker activity or level in a cell. [0126] Kits may comprise components, which may be individually packaged or placed in a container, such as a tube, bottle, vial, syringe, or other suitable container means. [0127] Individual components may also be provided in a kit in concentrated amounts; a component may be provided individually in the same concentration as it would be in a solution with other components. Concentrations of components may be provided as 1x, 2x, 5x, 10x, or 20x or more. [0128] Kits for using probes, antibodies, synthetic nucleic acids, nonsynthetic nucleic acids, and/or inhibitors of the disclosure for prognostic or diagnostic applications are included as part of the disclosure. Specifically contemplated are any such molecules corresponding to any biomarker identified herein, which includes antibodies that bind to such biomarkers as well as nucleic acid primers/primer sets and probes that are identical to or complementary to all or part of a biomarker, which may include noncoding sequences of the biomarker, as well as coding sequences of the biomarker. [0129] Negative and/or positive control nucleic acids, antibodies, probes, and inhibitors may be included. In addition, a kit may include a sample that is a negative or positive control for methylation of one or more biomarkers. [0130] It is contemplated that any method or composition described herein can be implemented with respect to any other method or composition described herein and that different embodiments may be combined. The claims originally filed are contemplated to cover claims that are multiply dependent on any filed claim or combination of filed claims. XII. Examples [0131] The following examples are included to demonstrate preferred embodiments of the invention. It should be appreciated by those of skill in the art that the techniques disclosed in the examples which follow represent techniques discovered by the inventor to function well in the practice of the invention, and thus can be considered to constitute preferred modes for its practice. However, those of skill in the art should, in light of the present disclosure, appreciate that many changes can be made in the specific embodiments which are disclosed and still obtain a like or similar result without departing from the spirit and scope of the invention. Example 1: Circulating plasma miRNA homologs in mice and humans reflect familial cerebral cavernous malformation disease [0132] Patients with familial cerebral cavernous malformation (CCM) inherit germline loss of function mutations and are susceptible to progressive development of brain lesions and neurological sequelae during their lifetime. To date, no homologous circulating molecules have been identified that can reflect the presence of a CCM mutation, both in animal models and patients. The inventors hypothesize that homologous differentially expressed (DE) plasma miRNAs can reflect the CCM germline mutation in preclinical murine models and patients and can be applied to develop translatable circulating biomarkers. Herein, several homologous DE plasma miRNAs with mechanistic putative gene targets within the transcriptome of preclinical and human CCM lesions were identified. Several of these gene targets were additionally found to be associated with CCM-enriched pathways identified using the Kyoto Encyclopedia of Genes and Genomes. DE miRNAs were also identified in familial CCM patients who developed new brain lesions within the year following blood sample collection. The miRNome results were then validated in an independent cohort of human subjects with real-time-qPCR quantification, a technique facilitating plasma assays. Finally, a Bayesian-informed machine learning approach showed that a combination of plasma levels of miRNAs and circulating proteins improves their association with familial-CCM disease in human subjects to 95% accuracy. These findings act as an important proof of concept for developing translatable circulating biomarkers of CCM disease in mice and humans. These can be applied in preclinical studies and human trials aimed at monitoring and restoring gene function in CCM and other diseases. [0133] Cerebral cavernous malformations (CCMs), also known as cavernous angiomas, are enlarged blood-filled vascular caverns prone to hemorrhage due to dysfunctional vessel wall angioarchitecture affecting up to 0.5% of the population [1, 2]. Approximately 30% of CCM patients manifest familial disease, with an autosomal dominant inheritance of a heterozygous germline loss of function mutation in one of three genes, CCM1/KRIT1, CCM2/Malcavernin, or CCM3/PDCD10 [1]. These familial cases develop new hemorrhagic brain lesions throughout their life, predisposing them to a risk of symptomatic hemorrhage, seizures, and/or focal neurological deficits [1]. The CCM3/PDCD10 disease manifests exceptional aggressiveness with greater lesion burden and earlier disease manifestations than other genotypes [3]. Surgical excision of symptomatic CCM lesions is the only current therapeutic option, with serious morbidity and costs, and obvious limitations with multiple lesions [1]. Thus, there is an intense effort at developing novel therapies [4, 5], including gene restoration therapies with viral vectors for familial disease [6]. [0134] There is an ongoing search for facile biomarkers that can accurately reflect disease status and lesional activity of CCMs, to guide the selection of aggressive cases for clinical trials and to monitor the impact of novel therapies [7-10]. Peripheral blood plasma provides a unique window into greater systemic biological processes, offering a promising opportunity for biomarker development [7, 11, 12]. In fact, dysregulated lesional pathways identified in CCM lesions may be reflected within the plasma [13]. Recent studies in preclinical murine models and human CCM patients have reported a differential plasma proteome in CCM disease, including proteins which have previously been implicated in CCM mechanisms [14-20], and reflect the transcriptome of micro-dissected human and murine CCM lesions [13, 21]. However, biomarker discovery in CCM has not systematically examined circulating molecules other than plasma proteins. [0135] Overall, it is beneficial to identify circulating homologous miRNAs with postulated mechanistic links, which can efficiently be cross validated in preclinical models of disease and translated to human subjects [27]. This is especially advantageous where experimental therapeutics could be tested, and biomarker efficacy monitored using the same biomarker in preclinical studies, in human trials and ultimately in clinical practice. To date, homologous plasma miRNAs which reflect disease states in murine models and humans have not yet been identified. The inventors hypothesize that homologous DE plasma miRNAs can reflect the heterozygous inherited CCM germline mutation in preclinical murine models and patients, can distinguish cases who would develop new lesions, and can be assayed in peripheral blood and thus developed as translatable circulating biomarkers in specific clinical contexts of use. A. Methods 1. Mouse sample collection [0136] Blood was collected via the submandibular vein route from 3 to 5 month-old mice including 7 Ccm1 +/- mice, 7 Ccm3 +/- mice and 7 sex matched wild type mice. These mice were bred at the University of Chicago (Table A). The pre-clinical heterozygous mice have been used to more closely mimic the physiopathology of the human familial disease [3]. Table A: Summary of Mouse Experimental Information used for the Plasma miRNome Sequencing [0137] Seven Ccm1 flox/flox mice and 5 Ccm3 flox/flox mice with a Pdgfb-Cre + driver, along with 8 sex matched wild type mice were bred at the University of California San Diego (Table A). These mice were injected with 50µg of 4-hydroxy-tamoxifen intraperitoneally at P1 to induce Cre activity and gene loss, bypassing embryonic lethality of the homozygous state [28]. These mice and their controls were sacrificed uniformly at 7-10 weeks of age through cardiac puncture and blood was collected simultaneously. [0138] After collection, plasma was separated using Z-gel tubes (Sarstedt, Nümbrecht, Germany) through centrifugation (AllegraX-30R, Beckman Coulter). One hundred ul of plasma was then aliquoted into 1.7-ml microcentrifuge tubes and stored at –80°C until RNA extraction. The sample size for this study was estimated based on a previous study that identified DE miRNAs in the plasma of only 3 Ccm3 +/- mice when compared to 3 wild type mice [21]. 2. Human CCM cohort characteristics [0139] For this study, the plasma miRNome of 23 familial-CCM patients (13 CCM1 and 10 CCM3), enrolled between July 2014 and July 2019, was sequenced (Table 1). The diagnosis of CCM was established via magnetic resonance imaging (MRI) at a single referral center (uchicagomedicine.org/ccm) by a senior neurosurgeon (IAA) with more than 30 years of experience in CCM disease management. The CCM genotype was defined as either CCM1 or CCM3, through genetic testing performed by PreventionGenetics (Marshfield, WI, USA), utilizing Sanger and NextGen sequencing followed by deletion/duplication analysis [2]. Patients with partial or complete resection of CCM or any prior brain irradiation were excluded. Due to low numbers of CCM2 cases, which only make up 20 % of familial cases, this genotype was unable to be analyzed separately [29]. Among the 23 familial-CCM patients enrolled, 10 (5 CCM1 and 5 CCM3) harbored at least 1 new CCM on their routine clinical follow-up SWI MRI in the year following their blood sample collection. SWI imaging is recommended for the diagnosis of CCM and assessment of lesion burden as this sequence is more sensitive in detecting perilesional hemosiderin and capturing smaller lesions than traditional T1 and T2 MRI sequences [2, 30]. Table 1: Cohort Demographics of CCM Patients and Healthy Controls for miRNome Sequencing.
[0140] An independent cohort including 24 familial-CCM patients (12 CCM1 and 12 CCM3) and 7 healthy non-CCM subjects was also enrolled between July 2014 and July 2019 for validation of the miRNA sequencing using real time quantitative PCR (RT-qPCR) (Table 2). Demographic analyses showed that a greater proportion of white/Caucasian was enrolled in both cohorts of CCM patients [p<0.05]. No other statistically significant demographic difference was observed between any of the human CCM cohorts, including age and sex. Previous statistical simulations and clinical studies have reported several DE plasma miRNAs in CCM patients using similar sample sizes [7, 26]. Table 2. Cohort Demographics of CCM Patients and Healthy Controls for RT-qPCR Validation.
[0141] Healthy non-CCM subjects were excluded if they had (a) any medical or neurologic condition requiring ongoing follow-up or medical treatment in the preceding year, (b) a history of concussion or brain trauma in the preceding year, (c) a history of prior brain irradiation at any time, (d) been pregnant or lactating in the preceding year, (e) used recreational, psychoactive, or neuroleptic drugs in the prior year. The healthy non-CCM subjects had an MRI performed to ensure that they did not have any unknown neurological conditions that could confound the study. 3. miRNA sequencing and differential expression analyses [0142] Total RNAs from the serum of both heterozygous and homozygous mice (200µl), and from the plasma of familial-CCM patients (up to100µl) were extracted using the miRNeasy Serum/Plasma Kit (Qiagen, Hilden, Germany) following the manufacturer recommendation [21, 26]. cDNA libraries were then generated with commercially available Illumina small RNA-Seq kits (Clontech, Mountain View, CA, USA) and sequenced with the Illumina HiSeq 4000 platform (Illumina, San Diego, CA, USA) using single-end 50-bp reads, at the University of Chicago Genomics Core for sequencing. The sequencing depth median for the plasma miRNomes was 10 million and 12 million reads per sample for mouse and human analyses, respectively. The sequencing depth used herein remains within reasonable range to identify known miRNAs that are differentially expressed between groups [31]. [0143] The differential expression analyses were first performed in the plasma of the preclinical mouse models between (1) Ccm1 -/- , (2) Ccm1 +/- , (3) Ccm3 -/- , (4) Ccm3 +/- and their respective wild type controls [p<0.1, false discovery rate (FDR) corrected]. Additional differential expression analyses were done between (1) Ccm1 -/- and Ccm3 -/- , and (2) Ccm1 +/- and Ccm3 +/- (Fig.1). [0144] Differential expression analyses were performed in human between healthy non- CCM controls and (1) CCM1 patients, (2) CCM3 patients, and then (3) between CCM1 and CCM3 patients (Fig. 1). An additional analysis was performed between the CCM patients showing New Lesion Formation and No New Lesion Formation. Though these results were not FDR corrected, additional stringency was added as only homologous DE miRNA with (1) gene targets within CCM transcriptomes and (2) mechanistic links reported in CCM disease were considered most relevant. All differential expression analyses were conducted using R bioconductor package DESeq2. Additionally, a partial least-squares discriminant analysis (PLS-DA) was used to validate the capacity of DE miRNAs (p<0.05, non-FDR corrected) to distinguish between CCM1, CCM3, and healthy non-CCM [32-34]. Using the mixOmics package in R [35], PLS-DA components were sorted based on the variance explained, and the AUC and p-value were reported for the top 2 components of each genotype comparison. 4. Putative targets of homologous mouse-human DE miRNAs [0145] The human homologs of DE miRNAs identified in preclinical murine models were queried using the MirGeneDB database (found online at mirgenedb.org/) (Fig. 1). Putative gene targets of homologous mouse-human DE miRNAs were identified using miRWalk 3.0 [36, 37] (Fig. 1). For each miRNA, gene targets were identified for the 3 different gene locations (3′ untranslated region (UTR), 5′ UTR, and coding sequence (CDS)) using a random forest tree algorithm with a bonding prediction probability higher than 95%. All known gene sequences are scanned and any matches to known miRNA sequences are flagged. These matches are then compared and validated using 8 other established miRNA-target prediction programs [38]. MiRWalk also consolidates data from several databases, including TargetScan, miRDB, and miRTarBase, and reports known miRNA-gene interactions from the literature to provide both predicted and validated gene target binding sites of miRNAs [38]. Only genes that appeared in at least 2 of the 3 databases on miRWalk were considered to be putative gene targets of the homologous DE miRNA, ensuring these interactions were validated by more than one source [13, 21, 26]. [0146] The putative targets of each DE miRNA identified in the plasma of mouse Ccm1 -/- and Ccm1 +/- were queried within the transcriptome of in vitro Ccm1 -/- mouse brain microvascular endothelial cells (BMECs). While the putative targets of the DE miRNAs identified in the plasma of Ccm3 -/- and Ccm3 +/- mice were queried in the transcriptome of (1) in vitro Ccm3 -/- mouse BMECs, and (2) laser micro-dissected lesional neurovascular units (NVUs) of Ccm3 +/- mice. Gene targets of DE miRNAs identified in the plasma of CCM1 and CCM3 patients were queried within the transcriptome of laser micro-dissected NVUs of human surgically resected CCM lesions. 5. CCM-enriched Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses [0147] The enriched KEGG pathways of each transcriptome were derived independently using the differentially expressed genes (DEGs) [p<0.05, FDR corrected; |FC|>1.5] identified in (1) Ccm1 -/- BMECs, (2) Ccm3 -/- BMECs, as well as micro-dissected lesional NVUs of (3) Ccm3 +/- mice, and (4) human surgically resected CCMs. Enriched KEGG pathways [p<0.05, FDR corrected] were generated using a database and knowledge extraction engine [39] with a Bayes factor cutoff of 3. The putative target DEGs of each homologous DE miRNA were then associated with the list of CCM-enriched KEGG pathways within their respective transcriptomes. Finally, KEGG pathways and putative target DEGs were categorized into 6 biological processes related to CCM disease identified after a comprehensive literature search [21]. These biological processes include: (1) Neuron, Glia, Pericyte Function, (2) Permeability/Adhesion, (3) Apoptosis and Oxidative Stress, (4) Inflammation/Immune Response, (5) Vascular Processes, (6) Cellular Proliferation. 6. Selection of miRNAs for validation using RT-qPCR [0148] RT-qPCR was not only used to validate the miRNA sequencing results but also serve as proof of feasibility for direct measurement of miRNAs in plasma for further biomarker use. Validation of potential biomarker candidates is needed for eventual biomarker development [40]. Among the homologous DE plasma miRNA with gene targets within CCM transcriptomes, 4 were chosen for RT-qPCR validation as (1) they have at least 10 DEGs [p<0.05, FDR corrected] as putative targets within the human CCM lesional transcriptome, and (2) had putative gene targets previously implicated in CCM disease. In addition, miR-375-5p was assessed because it was dysregulated across both CCM1 and CCM3 genotypes in the plasma of both preclinical murine models and CCM patients. Finally, not DE in mice, let-7e- 5p was selected since it was the only miRNA dysregulated in CCM1 patients when compared to healthy controls as well as CCM3 patients, making it a potential specific marker for the CCM1 genotype. 7. RT-qPCR assay of putative DE miRNAs [0149] Relative and absolute quantifications of a panel of 6 pre-selected DE miRNAs were assessed using RT-qPCR. The total RNA was first extracted from plasma using the MagMAX mirVana Total RNA Isolation Kit (Thermo Fisher, Waltman, MA, USA) An exogenous spike- in control, miR-cel-39-3p (1.5x10 10 copies of cel-miR-39-3p in 5µl) (IDT, Coralville, IA, USA), was added to all plasma samples prior to extraction and then measured to correct for extraction efficiency, as well as to ensure the inter-plate reproducibility of the RT-qPCR [41, 42]. RT-qPCR was performed using the TaqMan Advanced miRNA Assays Kit (Thermo Fisher). A miR-Amp reaction was performed to amplify miRNAs prior to real-time PCR using the QuantStudio3 (Thermo Fisher). The absolute quantification (number of miRNAs strands/µl) of each miRNA was estimated using a standard curve comprised of serial dilutions of 6 known concentrations of the miRNAs of interest (10 5 to 10 10 copies/ µl). [0150] For the relative quantification, the Cq values of each selected plasma miRNA as well as of two human endogenous controls, miR423-5p and miR16-5p, and of exogenous control cel-miR39-3p were measured. MiR-423-5p was eventually used as the endogenous control as it is expected to be expressed at equal levels across tissue types and samples [43]. This was validated using the NormFinder software, as miR-423-5p was shown to have the most stable expression across the controls tested (found online at moma.dk/normfinder-software) [44]. The Cq values of the endogenous control were then used to calculate the relative plasma expression [45]: ∆Cq = Cq miRNA of interest - Cq endogenous control [0151] The comparison of the relative quantification of each pre-defined miRNA was performed between (1) familial-CCM1 patients, (2) familial-CCM3 patients, and (3) healthy non-CCM controls. For each miRNA, a relative quantification value greater than ±3 standard deviations away from the mean was defined as an outlier [46, 47]. A non-parametric Mann- Whitney was performed between the ∆Cq for each comparison to estimate the ∆∆Cq using Prism (GraphPad, San Diego, California) [48]. 8. Combined plasma miRNA and protein biomarker development [0152] A three-step Bayesian-informed machine learning (ML) approach was implemented to assess if a weighted combination of plasma levels of miRNAs (i.e., assessed by RT-qPCR), and proteins with a documented role in CCM disease (i.e., assessed by ELISA) shows better performance to distinguish familial-CCM. This Bayesian paradigm relies on prior information, such as the biological relevance of candidate molecules, to inform biomarker development [49]. Additionally, ML approaches allow for the combination of the best performing plasma molecules, given that individual molecules may have imperfect sensitivity/specificity [7]. A thorough literature search was conducted to identify 21 plasma proteins related to angiogenesis, endothelial permeability, inflammation, immune response, and extracellular matrix remodeling, with reported associations to CCM or neurovascular diseases [7, 26, 50, 51]. As a proof of concept, the inventors assessed familial disease reflecting germline heterozygosity of any CCM gene, since the sample sizes of specific CCM genotypes were insufficient for separate statistical analyses (Fig.10-11). Two different approaches were developed using 10-fold and leave-one out cross validation to first select the best individual proteins and miRNAs which reflected familial-CCM disease (Fig.6). These cross validations reduce the overfitting effects on a limited dataset, while excluding any molecules which would not individually contribute to the model’s diagnostic association [52]. Selected miRNAs or proteins were then used to create the best models including (1) only plasma levels of miRNA, (2) only plasma levels of proteins, and finally (3) combining miRNAs and proteins to diagnose familial-CCM (Fig.6). In each case, the best model was identified as having the lowest Akaike information criterion (AIC) score. Feature selection minimized the dependency among features under the assumption of logistic regression, while selecting the model with AIC scores reduced the overall complexity of the model to prevent overfitting. Finally, all three models were tested on the same cohort of patients to compare performance through sensitivity and specificity (Fig. 6). Refer to supplemental material for additional Methods. B. Results 1. DE miRNAs in peripheral blood plasma of Ccm1 and Ccm3 murine models [0153] Differential expression analyses identified 17 DE miRNAs in the plasma of Ccm1 -/- mice (Table S1), and 44 in Ccm1 +/- mice [all: p<0.1, FDR corrected] (Table S2) when compared to their respective wild type mice. In the Ccm3 models, 142 plasma miRNAs were DE in the homozygous mice (Table S3), and 9 in the heterozygous mice when compared to their respective wild type mice [all: p<0.1, FDR corrected] (Table S4). Finally, 99 DE plasma miRNAs were identified between Ccm1 -/- and Ccm3 -/- mice [all: p<0.1, FDR corrected] (Table S5). No plasma miRNAs were DE between Ccm1 +/- and Ccm3 +/- mice. Non-sensitized Ccm1 and Ccm3 heterozygous mice have been shown to harbor a negligible lesion burden. Thus, these DE miRNAs were discovered to reflect disease genotype, rather than disease severity. 2. Homologous mouse-human DE miRNAs in peripheral blood plasma of patients with CCM1 and CCM3 disease [0154] Differential analyses identified 16 DE plasma miRNAs between familial-CCM1 patients and healthy non-CCM controls [p<0.05] (Table S6). The PLS-DA results showed that the DE miRNAs (p<0.05) identified through RNA-seq are able to distinguish CCM1 from healthy non-CCM subjects (top 2 components, both: AUC 100%, p=7.5x10 -4 ). Two of these 16 miRNAs, miR-20b-5p [fold change (FC)=25.9] and miR-375-3p [FC=3.5] had their murine homolog also DE in the plasma of Ccm1 +/- mice [FC=2.6, 5.2 respectively; both: p<0.1, FDR corrected] (Fig.2). No plasma miRNAs were commonly DE between familial-CCM1 patients and Ccm1 -/- mice. [0155] In the familial-CCM3 patients, 12 DE plasma miRNAs were identified compared to healthy non-CCM controls [p<0.05] (Table S7). PLS-DA results confirmed that DE miRNAs (p<0.05) can differentiate CCM3 from healthy non-CCM subjects (top 2 components, both: AUC 100%, p=6.4x10 -4 ). Five of these 12 DE plasma miRNAs, miR-93-5p [FC=0.6], miR- 370-3p [FC=4.3], miR-487b-3p [FC=13.3], miR-369-5p [FC=19.2], and miR-375-3p [FC=2.8] had their murine homolog also DE in the plasma of Ccm3 -/- mice [FC= 0.5, 2.7, 6.7, 2.7, 3.5 respectively; all: p<0.1, FDR corrected] (Fig. 2). No plasma miRNAs were commonly DE between familial-CCM3 patients and Ccm3 +/- mice. It is unclear why homologous DE miRNAs associated with germline CCM1 and CCM3 heterozygosity in humans were shared with heterozygous Ccm1 +/- mice and homozygous Ccm3 -/- mice, respectively. The number of plasma DE miRNAs in heterozygous Ccm3 +/- and postnatally induced Ccm1 -/- mice was low, which may have been due to a low read depth during sequencing. In addition, postnatally induced gene loss in mouse models probably reflects an acute phase of the pathogenesis of CCMs, and not the chronic aspect seen in the human disease; these differences are exaggerated in the more aggressive Ccm3 models. [0156] The inventors further identified 24 plasma DE miRNAs between CCM1 and CCM3 patients [p<0.05] (Table S8). DE miRNAs (p<0.05) identified through RNA-seq were shown to distinguish CCM1 from CCM3 through an additional PLS-DA (component 1, AUC 98.15%, p=2.2x10 -4 ; component 2, AUC 100%, p=1.2x10 -4 ). Of these, miR-128-3p [FC=0.6], miR-410- 3p [FC=2.0], miR-9-5p [FC=0.07], miR-323b-3p [FC=13.7], and miR-93-5p [FC=0.7] had their murine homolog also DE between Ccm1 -/- and Ccm3 -/- murine models [FC=1.8, 1.8, 28.2, 6.6, and 0.6 respectively; all: p<0.1, FDR corrected] (Fig.2). 3. Gene targets of homologous mouse-human DE plasma miRNAs [0157] Of the 10 homologous DE miRNAs, 8 were DE in the same direction in familial- CCM patients and preclinical mouse models (all genotypes) (Fig. 2). Seven homologous DE miRNAs, targeting a range of 0.1%-2.4% of the total human genome, had DEG targets within the published transcriptome of laser micro-dissected NVUs of human surgically resected CCMs (Fig. 3). In addition, 5 of those 7 miRNAs also had DEG targets within either the transcriptome of (1) in vitro Ccm1 -/- or Ccm3 -/- BMECs, or (2) laser micro-dissected lesional NVUs of Ccm3 +/- mice. To further classify these gene interactions, the KEGG was used to generate enriched KEGG pathways [p<0.05, FDR corrected; Bayes factor≥3] for each CCM transcriptome (data not shown). [0158] DE miR-20b-5p, which was increased in the plasma of familial-CCM1 patients, and miR-93-5p, which was decreased in familial-CCM3 patients compared to healthy non-CCM controls, respectively had 8.9% and 9.1% of their gene targets within the transcriptome of human micro-dissected CCM lesional NVUs. CCM-enriched KEGG pathway analyses showed that these DEGs were associated with PI3K-Akt signaling, focal adhesion, HIF-1, Rap1 signaling, and other pathways involved in CCM mechanisms [all: p<0.05, FDR corrected] (data not shown). Specifically, both miR-20b-5p and miR-93-5p target VEGFA and ADAMTS5. Additionally, miR-93-5p targets ROCK2 and MAP3K14, which are associated with CCM pathogenesis. miR-93-5p provides a unique translational opportunity for biomarker development as it also targets Vegfa in mice, which is a DEG in the transcriptome of laser micro-dissected lesional NVUs of Ccm3 +/- mice. [0159] DE miR-370-3p and miR-487b-3p, both found to be increased in the plasma of familial-CCM3 patients, were also found to target several DEGs in the transcriptome of human micro-dissected CCM lesional NVUs. Specifically, 9.4% of miR-370-3p’s gene targets, and 33.3% of miR-487b-3p’s targets were within the human CCM lesional transcriptome. miR-370- 3p targets NFASC and FGF7, which were both found to be associated with PI3K-Akt signaling, Rap1 signaling, and cell adhesion molecules [all: p<0.05, FDR corrected] (data not shown). It also targets Klhl21, a DEG in the transcriptome of micro-dissected lesional NVUs of Ccm3 +/- mice. In addition, miR-487b-3p was found to target NRARP, which is DE in the human CCM lesional transcriptome. [0160] Finally, DE miR-128-3p and miR-9-5p, which were decreased, and miR-410-3p, which was increased (i.e., all between familial-CCM1 compared to familial-CCM3 patients) also target DEGs within the human CCM lesional transcriptome. miR-128-3p, miR-9-5p, and miR-410-3p had 10.8%, 13.5%, and 9.0% of their gene targets in the human CCM lesional transcriptome, respectively. Among these, IGF1 and NRXN1 are targeted by mir-128-3p. Of interest, IGF1 is associated with enriched-KEGG pathways such as PI3K-Akt signaling, HIF- 1 signaling, and Rap1 signaling, while NRXN1 is related to cell adhesion molecules [all: p<0.05, FDR corrected] (data not shown). Also, miR-128-3p targets Foxq1 in mice, which is DE in the transcriptome of Ccm1 -/- mouse BMECs. Furthermore, miR-9-5p targets TNC, VAV3, and VCAN, which are associated with focal adhesion and cell adhesion KEGG pathways [all: p<0.05, FDR corrected] (data not shown). Additionally, miR-9-5p targets Rap1b which is DE in the transcriptome of Ccm1 -/- mouse BMECs. Finally, miR-410-3p targets two DEGs within the human CCM lesional transcriptome, both of which are part of the solute carrier group of membrane transport proteins. Interestingly, one of these genes, SLC8A1, is also targeted by miR-410-3p in mice and is a DEG in the transcriptome of laser micro-dissected lesional NVUs of Ccm3 +/- mice. 4. Plasma miRNA as biomarkers of patients with new lesion formation [0161] Ninety-seven plasma miRNAs [p<0.05, non-FDR corrected] were DE in familial- CCM patients that developed a new lesion within a year after their blood draw, as compared to those with stable lesion burden (Table S13). Fifteen of these 97 were also DE in familial- CCM3 patients, when compared to healthy non-CCM subjects or familial-CCM1 patients. Of these 15 plasma miRNAs, 9 (miR-141-3p, miR-409-3p, miR-431-5p, miR-485-3p, miR-9-5p, miR-128-3p, miR-379-5p, miR-205-5p, and miR-1271-5p) were found to target DEGs within the transcriptome of laser micro-dissected NVUs of human surgically resected CCMs. These DEGs include ROCK2, VEGFA, ADAMTS5, VCAN, and NRXN1, which are associated with several CCM-enriched KEGG pathways such as PI3K-Akt signaling, HIF-1 signaling, Rap1 signaling, focal adhesion, and cell adhesion molecules [all: p<0.05, FDR corrected] (data not shown). 5. Peripheral blood assays of DE miRNAs as biomarkers of CCM genotype [0162] A validation of the miRNA sequencing differential analyses was performed by assessing the plasma levels of a selected panel of six DE miRNAs using RT-qPCR within an independent cohort of 12 familial-CCM3, 12 familial-CCM1 patients, and 13 healthy non-CCM controls (Fig.4, Fig. 6). Relative quantification levels assessed through RT-qPCR supported the directionality and dysregulation observed in the miRNome sequencing data. [0163] Lower RT-qPCR relative quantification values of miR-20b-5p in the plasma of familial-CCM1 were observed compared to healthy non-CCM controls [p=0.04] (Fig. 4a). A trend toward lower relative quantification values of miR-20b-5p was also observed in the plasma of familial-CCM1 patients compared to familial-CCM3 patients [p=0.07] (Fig.4a). No difference was observed between familial-CCM3 patients and healthy non-CCM subjects. [0164] Familial-CCM1 patients also showed greater RT-qPCR relative quantification values of plasma miR-93-5p compared to healthy non-CCM subjects [p=0.007] (Fig. 4b). In addition, a trend toward higher relative quantification values was observed between familial- CCM3 compared to healthy non-CCM subjects [p=0.06] (Fig. 4b). No difference between familial-CCM3 and familial-CCM1 patients was observed. [0165] The relative plasma quantification RT-qPCR values of miR-9-5p were higher in the plasma of familial-CCM1 patients [p=0.01] as well as in healthy non-CCM subjects compared to familial-CCM3 patients [p=0.01] (Fig. 4c). The relative quantification values of miR-375- 3p were also higher in the plasma of familial-CCM1 [p=0.05] and familial-CCM3 patients [p=0.02] compared to healthy non-CCM subjects (Fig.4d). [0166] Finally, the relative quantification plasma values for miR-128-3p did not show differences in any of the comparisons (Fig. 4e), while relative quantification values of the miRNA let-7e-5p were higher in the plasma of familial-CCM1 compared to healthy non-CCM subjects [p=0.01] (Fig.4f). [0167] Further linear Pearson correlation analyses showed no correlation between relative quantification of any of these miRNAs and the lesion burden. These results suggest that these DE plasma miRNAs reflect disease genotype, rather than severity. 6. Integration of plasma levels of DE miRNAs with proteins enhances their association with familial-CCM [0168] The relative quantification plasma levels of the six miRNAs were used to create the most optimal weighted combination (i.e., achieving the lowest AIC score) able to diagnose familial-CCM patients in comparison to non-CCM controls (n=24). This model’s accuracy was defined as fair [53] in distinguishing familial-CCM patients with 84% sensitivity and 67% specificity [area under the ROC curve (AUC)=75.4%, CI = 52.4% - 98.5%] (Fig.8). [0169] A similar approach was performed using plasma levels of proteins in familial-CCM patients (n=46). Using two independent ML approaches, the best weighted combination of plasma proteins identified familial-CCM patients from healthy non-CCM with sensitivity up to 89.5% and specificity up to 100% [optimal AUC =96.5%, CI = 90.1% - 100%] (Fig.9). [0170] Given the imperfect diagnostic performance of miRNAs or proteins alone, combining the two may offer a more holistic model reflecting various aspects of the disease. Thus, an integrated biomarker including proteins and miRNAs was developed using familial- CCM patients (n=19) with both plasma levels of protein and relative quantification of miRNAs. Using a 10-fold cross validation ML approach, the weighted combination of proteins and relative quantification levels of miRNAs diagnosed familial-CCM patients with an accuracy of 95%, with a sensitivity of 94.7% and sensitivity of 100%. (Fig.5a and b): [0171] Canonical Value = -114.88*[let-7e-5p] r + 359.47*[miR-93-5p] r - 118.16*[miR-20b- 5p]r + 88.93*[miR-128-3p]r + 199.47*[IL-10] - 19.7*[Tsp-2] - 463.18 [0172] with [X] r denoting relative quantification of miRNAs. This model estimated higher canonical values for familial-CCM patients compared to non-CCM healthy controls [p<0.0001] (Fig.5a). [0173] In order to validate these results, another ML approach, a leave-one-out cross validation, was also applied. This approach yielded a combination with the same performance of 94.7% sensitivity and 100% specificity (Fig. 5c and d) in diagnosing familial-CCM, however it included less compounds making it potentially more feasible for future clinical applicability: [0174] Canonical Value = 2.05*[miR-9-5p]r - 3.02*[miR-93-5p]r - 0.01*[IL-2] + 0.1*[TNFRI] + 0.46*[Tsp-2] [0175] This additional weighted combination calculated lower canonical scores for familial- CCM patients compared to non-CCM healthy controls [p<0.0001] (Fig. 5c). Refer to supplemental material for additional results. C. Supplementary Methods [0176] Murine and Human CCM Transcriptomes : The transcriptomes of in vitro Ccm1 -/- and Ccm3 -/- BMECs, in vivo Ccm3 +/- murine lesional CCMs, and neurovascular units (NVUs) from human lesional tissue were all previously published and were used here for putative target and KEGG pathway correlation of DE miRNAs. Methods for collection of tissue and construction of transcriptomic data is clearly delineated within these previously published articles. [1-3] [0177] Processing of plasma from familial-CCM patients: For each patient, the blood sample was collected at the time of their clinic visit. Standard clinical 10-ml heparinized Vacutainer tubes (BD Vacutainer, Becton, Dickinson and Company) were used to draw blood samples. The plasma was isolated by centrifugation at 500 g at 4°C for 10 minutes (AllegraX- 30R, Beckman Coulter). Subsequently, 300μl of plasma was aliquoted into 1.7-ml microcentrifuge tubes and stored at –80°C [4-6]. [0178] Plasma miRNome sequencing: Following RNA extraction using the miRNeasy Serum/Plasma Kit (Qiagen), cDNA libraries were generated with commercially available Illumina small RNA-Seq kits and sequenced using the Illumina HiSeq 4000 platform (Illumina) with single-end 50-bp reads. Fast QC (v0.11.5) was used to assess raw sequencing quality (found on the world wide web at bioinformatics.babraham.ac.uk/projects/fastqc/). [0179] The small RNA adapter sequences were trimmed from small RNA sequencing data with cutadapt (found online at cutadapt.readthedocs.io). The trimmed reads were then mapped and quantified either to the human mature miRNA database for human samples, or the mouse mature miRNA database for mouse samples (miRbase, found on the world wide web at mirbase.org/) using sRNAbench library (found online at bioinfo2.ugr.es/ceUGR/srnabench/) mapping strategy [7] and wrapped bowtie alignment (with alignment type = –n, seed length for alignment = 20, minimum read-count = 2, allowed number of mismatch = 0, minimum read count = 2, and maximum number of multiple mappings = 20). MiRNAs with low expression were removed before further downstream analyses. Finally, the miRNA expression values were normalized following trimmed mean of M-values normalization methods with library size correction [2, 6]. [0180] Plasma proteins with a documented role in CCM disease: A panel of 21 plasma proteins which had previously elucidated roles in CCM (Table S2) were measured using the enzyme-linked immunosorbent assay (ng/ml) and Multiplex (pg/ml) assay in accordance with manufacturer protocols in 46 CCM patients and 28 non-CCM controls. [0181] The plasma levels of thrombomodulin (TM), thrombospondin-1 (Tsp-1), thrombospondin-2 (Tsp-2) were assessed using commercially available ELISA assays kits (R&D Systems, Minneapolis, Minnesota, USA). All plates were washed with a BioTek 405TS plate washer (BioTek Instruments, Winooski, VT, USA), and absorbances measured using Bio- Rad iMark plate reader (Bio-Rad, Hercules, CA, USA). The assays were performed in the Neurovascular Research team laboratory at the University of Chicago. In each plate, the plasma samples were loaded in parallel duplicate wells, and then averaged. Fifty beads per region were collected for each well, and a 5-parameter logistic regression analysis was performed to estimate the sample concentration. [0182] The plasma levels of tumor necrosis factor alpha (TNFα), tumor necrosis factor receptor I (TNFRI), matrix metalloproteinase-2 (MMP2) and -9 (MMP9), chemokine ligand 2 (CCL2/MCP1), soluble endoglin/CD105 (sENG), soluble vascular cell adhesion protein 1 (sVCAM1), soluble intercellular adhesion molecule 1 (sICAM1/CD54), interleukin-1 beta (IL- 1β), IL-2, IL-6, IL-8/CXCL-8, IL-10, , vascular endothelial growth factor (VEGF), soluble roundabout guidance receptor 4 (sROBO4), interferon gamma (IFNγ), soluble cluster of differentiation 14 (sCD14), and C-reactive protein (CRP) were assessed using 5 customized magnetic bead-based multiplex Luminex screening immunoassay kits (R&D Systems, Minneapolis, Minnesota, USA) [8, 9]. The measurements were performed with a BioRad BioPlex-100 analyzer (Bio-Rad Laboratories, Hercules, California, USA) running the BioPlex Manager Software version 5·0, or the Luminex 200 System (Luminex Corporation, Austin, Texas, USA) running with xPONENT Software. The assessment was performed at the Flow Cytometry Core Facility at the University of Chicago. In each plate, the plasma samples were loaded in parallel duplicate wells, and then averaged. Plasma values greater ±3 standard deviations from the mean were excluded as outliers. [0183] Univariate Statistics and Batch correction for protein level assessment: Differences in relative and absolute levels of miRNAs measured by RT-qPCR, were analyzed using a Mann-Whitney test. Canonical values of the best weighted combinations developed during the three-step integrative approach were calculated for each patient and compared between familial-CCM patients and healthy-non-CCM subjects using an unpaired two sample Student’s t-test, assessed with pooled standard deviation, or Mann-Whitney test according to the equality of the variance. All statistical analyses were performed on Prism (GraphPad, San Diego, California, USA). All p-values were considered statistically significant at p<0.05. [0184] Variation due to non-biological experimental variations is common when utilizing multiplex or microarray analysis [10]. A principal component analysis revealed significant variation between batches in the plasma levels of the molecules. Thus, batch effects were corrected using the Combat function from the R package SVA [11]. [0185] Integrated plasma miRNA and protein biomarker development: A three-step integrative approach was developed to define and compare the best weighted models of circulating plasma compounds (Fig. 6). The three models generated were a weighted combination of (1) only relative quantification of plasma miRNAs (Familial-CCM patients, n=24; Healthy non-CCM, n=13), (2) only plasma levels of proteins with a documented role in CCM disease (Familial-CCM patients, n=46; Healthy non-CCM, n=28), (3) both relative quantification of plasma miRNAs and proteins (Familial-CCM patients, n=19; Healthy non- CCM, n=6). [0186] The selection of plasma compounds with higher individual predictive ability, and lower degree of redundancy was performed using machine learning cross-validation (CV). Two CV approaches, (1) 10-fold CV and (2) leave-one-out CV, were independently performed, creating two different sets of candidate compound biomarkers of (1) miRNAs alone, (2) proteins alone, and (3) a combined integrative model. The features selected at least once during the CV process were included in the initial set of candidate compound biomarkers to be used for model selection. The overall accuracy, specificity, sensitivity, receiver operating characteristic (ROC) curve were assessed using the Youden index [12]. [0187] A machine learning framework was used, consisting of a correlation-based feature selection, a logistic linear regression, and a tree-based model selection. A 10-fold CV was enforced throughout the framework to reduce the potential overfitting effects on the limited dataset. The logistic regression model with ridge penalty was applied as the classifier. This was given as a binary classification on n samples with m features. Probability for the positive class for any sample i was estimated using: where is the feature vector for sample i and β is the vector of the m coefficients. The negative multinomial log-likelihood was minimized according to the following: where γ i is the label for sample i and ridge parameter=1.0x10 -8 . [0188] After the feature selection, a breadth-first tree search was performed with all candidate compound biomarkers in order to assess all the possible combinations. The most optimal combination was selected as minimizing the Akaike information criterion (AIC) score calculated using the mean absolute error from the logistic regression classifier as the predictive power, and the number of features k as the complexity of the model. The final logistic regression coefficients for the best model were then defined on all the patients with both miRNAs and proteins. [0189] The leave-one-out CV was performed with multivariable logistic regression using feature selection through the glmnet library implemented in R program (Miami, Florida, USA) [13]. During each CV cycle, one subject was left as the independent testing sample, while the remaining subjects were enrolled as training samples for multivariate modeling. The best model with the lowest AIC score was selected similarly to the 10-fold CV. [0190] Literature review for CCM-relevant biological processes: In order to identify a comprehensive list of CCM-related publications, a systematic electronic review was performed querying the PubMed database of peer-reviewed articles published between January 2011 and April 2021 using the following key terms: (cerebral cavernous malformation [Title/Abstract] OR cerebral cavernous malformations [Title/Abstract] OR Cavernous Malformations [Title/Abstract] OR cerebral vascular malformation [Title/Abstract] OR cerebrovascular malformation [Title/Abstract] OR cavernous angioma [Title/Abstract] OR CCM1 [Title/Abstract] OR CCM2 [Title/Abstract] OR CCM3 [Title/Abstract] OR Krit1 [Title/Abstract] OR PDCD10 [Title/Abstract] OR MGC4607 [Title/Abstract]) AND (english [Language]) NOT (case report [Publication Type] NOT liver [Title/Abstract] NOT surgery [Title/Abstract] NOT management [Title/Abstract]) NOT treatment [Title/Abstract]). [0191] This search yielded a total of 1074 abstract, of which, 127 reported mechanistic pathways associated to the physiopathogenesis of CA disease. The list of KEGG pathways enriched within the human CMM lesional transcriptome and related to differently expressed genes (DEGs) targeted by DE miRNAs of interest were queried within these 127 abstracts for biological functions. Based on this search, KEGG pathways were classified into 6 previously published CCM-related biological processes [1, 2], including: cellular proliferation, inflammation and immune response, permeability and adhesion, neuron, glia and pericyte functions, apoptosis and oxidative stress, and vascular processes. Target DEGs of DE miRNAs were also linked to these 6 CCM processes based on the KEGG pathways they were found to be associated with. D. Supplementary Results 1. CCM-enriched KEGG pathways [0192] To further classify gene interactions, the Kyoto Encyclopedia of Genes and Genomes (KEGG) was used to independently generate enriched-KEGG pathways (p<0.05, FDR corrected; Bayes factor≥3) using the DEGs (p<0.05, FDR corrected) within the transcriptome of (1) laser micro-dissected NVUs of human surgically resected CCMs, (2) in- vitro Ccm1 -/- , (3) Ccm3 -/- mouse brain microvascular endothelial cells (BMECs), or (4) laser micro-dissected lesional NVUs of Ccm3 +/- mice. Eighty-five enriched-KEGG pathways were identified in the transcriptome of laser micro-dissected NVUs of human surgically resected CCMs (data not shown); 130 in the transcriptome of mouse in vitro Ccm1 -/- BMECs (data not shown); 23 in the transcriptome of mouse in vitro Ccm3 -/- BMEC (data not shown), and 5 in the transcriptome of mouse in vivo Ccm3 +/- lesional NVUs (data not shown). 2. Diagnostic biomarkers using plasma miRNAs and proteins independently [0193] Both the 10-fold and leave-one-out CV did not exclude any of the 6 miRNAs tested for individual diagnostic contribution as they all had non-zero coefficients (i.e., CV coefficients>0), and thus all were considered as possible variables during model selection. However, the relative quantification was selected over absolute quantification as it showed higher sensitivity and specificity when tested for diagnostic association with familial-CCM. [0194] The 10-fold CV selected 7 plasma proteins to be used as candidate compound biomarkers, including CRP, IL-10, MMP-9, TNFRI, VEGF, ROBO4, and Tsp-2 (all: CV coefficients>0). While the leave-one-out CV selected 9 proteins: CCL2, sCD14, IL-2, IL-6, MMP-9, TNFRI, VEGF, ICAM1, Tsp-2. [0195] The best weighted combination using only the plasma relative quantification of miRNAs assessed by the 10-fold cross-validation was able to distinguish familial-CCM patients with 68% accuracy, performing with 84% sensitivity and 67% specificity when tested on the full dataset (Fig.8a and b): Canonical Value = -0.72*[let-7e-5p]r + 1.07*[miR-93-5p]r - 3.94 with [X] r denoting relative quantification of miRNAs [0196] Canonical values estimated with this model showed a higher trend in familial-CCM patients compared to healthy non-CCM subjects (p=0.06; Fig.8a). [0197] The leave one out cross-validation approach defined a similar weighted combination able to distinguish familial-CCM patients with 84% sensitivity and 67% specificity (Fig.8c & d), and included: Canonical Value = 0.82*[let-7e-5p]r - 1.21*[miR-93-5p]r [0198] The canonical values calculated with this weighted combination also showed a lower trend (p=0.06) in familial-CCM patients compared to healthy non-CCM subjects (Fig.7c). [0199] The same approach was implemented to compute a weighted combination using plasma levels of proteins only assessed by 10-fold cross-validation. The model generated was able to diagnose familial-CCM patients with 79% sensitivity and 67% specificity on the full dataset, with an accuracy of 68% during CV (Fig.9a & b), and included: Canonical Value = 0.59*[IL-10] - 0.10*[VEGF] - 0.15*[Tsp-2] - 5.97 [0200] Canonical values of familial-CCM patients were higher (p=0.05) compared to healthy non-CCM subjects (Fig.9a). [0201] The leave-one-out cross validation pipeline selected a weighted combination of proteins able to diagnose familial-CCM patients with 89.5% sensitivity and 100% specificity (Fig.9c and d), and included: Canonical Value = 0.01*[TNFRI] - 0.01*[IL-2] - 0.29*[Tsp-2] [0202] Finally, the canonical values were lower (p<0.0001) in familial-CCM patients compared to healthy non-CCM subjects (Fig.9c). [0203] Diagnostic biomarkers of specific genotype using combined plasma miRNAs and proteins [0204] A similar three-step integrative approach was also applied to create the best diagnostic biomarkers for specific familial-CCM1 and familial-CCM3 genotypes using both plasma miRNAs and proteins. [0205] 10-Fold Cross Validation: The best weighted combination using plasma miRNAs and proteins was able to distinguish familial-CCM1 patients from non-CCM healthy controls with 71% accuracy when tested using CV, and performed with 100% sensitivity and 83% specificity on the full dataset (Fig.10a and b): Canonical Value = -2.18*[let-7e-5p]r + 2.87*[miR-93-5p]r + 3.67*[TM] – 23.04 [0206] Higher canonical values (p=0.01) were calculated in familial-CCM1 patients compared to healthy non-CCM subjects (Fig.10a). [0207] The same approach led to weighted combination able to distinguish familial-CCM3 patients from non-CCM healthy controls with 100% sensitivity and 100% specificity on the full dataset, and 88% accuracy during CV (Fig.10c and d): Canonical Value = 38.88*[miR-93-5p]r – 13.18*[miR-128-3p]r – 22.54*[miR-375-3p]r + 0.33*[CCL2] – 0.22*[TNFRI] – 6.51*[Tsp-2] + 186.92 [0208] Again, higher canonical values were higher (p=0.0002) in familial-CCM3 patients compared to healthy non-CCM subjects (Fig.10c). [0209] Finally, the weighted combination using plasma miRNAs and proteins was able to distinguish familial-CCM1 from familial-CCM3 patients with 88% sensitivity and 91% specificity when tested on the full dataset, and 79% accuracy during CV (Fig.10e and f): Canonical Value = -3.17*[let-7e-5p]r + 2.63*[miR-9-5p]r – 1.80*[IL-6] – 5.55 [0210] Canonical values estimated with this model were higher (p=0.0018) in familial- CCM1 patients compared to CCM3 patients (Fig.10e). [0211] Leave-one-out Cross Validation: The best weighted combination using plasma miRNAs and proteins assessed by leave-one-out cross-validation was able to distinguish familial-CCM1 patients from non-CCM healthy controls with 100% sensitivity and 75% specificity (Fig.11a and b), and included: Canonical Value = 0.89*[miR-128-3p]r – 2.22*[IL-6] [0212] Familial-CCM1 patients had higher canonical values (p=0.01) than healthy non- CCM subjects (Fig.11a). [0213] A similar approach using plasma miRNAs and proteins computed a weighted combination able to distinguish familial-CCM3 patients from non-CCM healthy controls with 83% sensitivity and 91% specificity (Fig.11c and d): Canonical Value = -0.29*[miR-20b-5p]r – 0.01*[TNFRI] [0214] Canonical values estimated with this model were higher in familial-CCM3 patients (p=0.0048) compared to healthy non-CCM subjects (Fig.11c). [0215] Finally, the best weighted combination able to distinguish familial-CCM1 from familial-CCM3 patients was achieved with only proteins and reached a 100% sensitivity and 82% specificity (Fig.11e and f): Canonical Value = -2.33*[CRP] + 1.64*[IL-6] – 5.01*[Tsp-2] [0216] Lower canonical values (p=0.0018) in familial-CCM1 compared to familial-CCM3 patients (Fig.11e). E. Discussion [0217] Herein, the inventors identify several homologous DE plasma miRNAs in preclinical murine models and CCM patients, with links to genes dysregulated in CCM disease. Additionally, the inventors show that the integration of readily measurable plasma proteins and plasma levels of miRNAs into a combined biomarker can diagnose familial-CCM disease with up to 95% accuracy. To the inventors’ knowledge, such a translational approach to biomarker discovery in mouse models and humans, with mechanistic links to the disease transcriptome, validation in an independent cohort of human subjects, and the combination of miRNA and protein levels to optimize biomarker performance have not been previously described in this or any other disease. [0218] This study is the first to show a weighted combination of levels of plasma miRNAs and proteins associated with genotype in a Mendelian disease, based on ML approaches and Bayesian concepts. Two cross-validation methods, which reduced overfitting effects, confirmed that integrating plasma miRNAs and proteins improves the diagnostic association with familial-CCM compared to each circulating compound (e.g., plasma miRNA and protein) alone. This is an important proof of concept for integration of several types of molecules in plasma biomarker development [7]. A homologous biomarker in mice and humans, which can differentiate familial-CCM with germline loss of function mutations can be immediately applied in preclinical studies and early trials, monitoring effectiveness of gene restoration therapeutics, which are currently of great interest. As patients undergo gene therapy, one could monitor their progress and the duration of therapeutic effect as biomarker canonical values shift in comparison to a healthy non-CCM profile. This approach to combined biomarker development is also applicable in other clinical contexts of use, including the prediction of new lesion formation. [0219] In summary, the inventors have shown for the first time in any disease, that homologous DE plasma miRNAs can reflect the germline loss of function mutation in preclinical murine models and patients. They have additionally shown that plasma miRNAs can also distinguish cases who would develop new lesions, and plasma levels of these miRNAs can be assessed for facile diagnostic assays. The inventors describe a novel approach of circulating biomarker development which can be applied in specific clinical contexts of use in familial-CCM and other Mendelian diseases. * * * [0220] All of the methods disclosed and claimed herein can be made and executed without undue experimentation in light of the present disclosure. While the compositions and methods of this invention have been described in terms of preferred embodiments, it will be apparent to those of skill in the art that variations may be applied to the methods and in the steps or in the sequence of steps of the method described herein without departing from the concept, spirit and scope of the invention. More specifically, it will be apparent that certain agents which are both chemically and physiologically related may be substituted for the agents described herein while the same or similar results would be achieved. All such similar substitutes and modifications apparent to those skilled in the art are deemed to be within the spirit, scope and concept of the invention as defined by the appended claims. 1. Awad IA, Polster SP. Cavernous angiomas: deconstructing a neurosurgical disease. J Neurosurg.2019;131:1-13. https://doi.org/10.3171/2019.3.JNS181724. 2. Akers A, Al-Shahi Salman R, I AA, Dahlem K, Flemming K, Hart B, et al. Synopsis of Guidelines for the Clinical Management of Cerebral Cavernous Malformations: Consensus Recommendations Based on Systematic Literature Review by the Angioma Alliance Scientific Advisory Board Clinical Experts Panel. Neurosurgery. 2017;80:665-80. https://doi.org/10.1093/neuros/nyx091. 3. Shenkar R, Shi C, Rebeiz T, Stockton RA, McDonald DA, Mikati AG, et al. Exceptional aggressiveness of cerebral cavernous malformation disease associated with PDCD10 mutations. Genet Med.2015;17:188-96. https://doi.org/10.1038/gim.2014.97. 4. Polster SP, Cao Y, Carroll T, Flemming K, Girard R, Hanley D, et al. Trial Readiness in Cavernous Angiomas With Symptomatic Hemorrhage (CASH). Neurosurgery. 2019;84:954- 64. https://doi.org/10.1093/neuros/nyy108. 5. Snellings DA, Hong CC, Ren AA, Lopez-Ramirez MA, Girard R, Srinath A, et al. Cerebral Cavernous Malformation: From Mechanism to Therapy. Circ Res. 2021;129:195-215. https://doi.org/10.1161/CIRCRESAHA.121.318174. 6. Zhu W, Shen F, Mao L, Zhan L, Kang S, Sun Z, et al. Soluble FLT1 Gene Therapy Alleviates Brain Arteriovenous Malformation Severity. Stroke. 2017;48:1420-3. https://doi.org/10.1161/STROKEAHA.116.015713. 7. Girard R, Li Y, Stadnik A, Shenkar R, Hobson N, Romanos S, et al. A Roadmap for Developing Plasma Diagnostic and Prognostic Biomarkers of Cerebral Cavernous Angioma With Symptomatic Hemorrhage (CASH). Neurosurgery. 2021;88:686-97. https://doi.org/10.1093/neuros/nyaa478. 8. Mikati AG, Khanna O, Zhang L, Girard R, Shenkar R, Guo X, et al. Vascular permeability in cerebral cavernous malformations. J Cereb Blood Flow Metab. 2015;35:1632-9. https://doi.org/10.1038/jcbfm.2015.98. 9. Zeineddine HA, Girard R, Cao Y, Hobson N, Fam MD, Stadnik A, et al. Quantitative susceptibility mapping as a monitoring biomarker in cerebral cavernous malformations with recent hemorrhage. J Magn Reson Imaging. 2018;47:1133-8. https://doi.org/10.1002/jmri.25831. 10. Girard R, Fam MD, Zeineddine HA, Tan H, Mikati AG, Shi C, et al. Vascular permeability and iron deposition biomarkers in longitudinal follow-up of cerebral cavernous malformations. J Neurosurg.2017;127:102-10. https://doi.org/10.3171/2016.5.JNS16687. 11. Huang Z, Ma L, Huang C, Li Q, Nice EC. Proteomic profiling of human plasma for cancer biomarker discovery. Proteomics.2017;17. https://doi.org/10.1002/pmic.201600240. 12. Loke SY, Lee ASG. The future of blood-based biomarkers for the early detection of breast cancer. Eur J Cancer.2018;92:54-68. https://doi.org/10.1016/j.ejca.2017.12.025. 13. Koskimaki J, Girard R, Li Y, Saadat L, Zeineddine HA, Lightle R, et al. Comprehensive transcriptome analysis of cerebral cavernous malformation across multiple species and genotypes. JCI Insight.2019;4. https://doi.org/10.1172/jci.insight.126167. 14. Koskimaki J, Polster SP, Li Y, Romanos S, Srinath A, Zhang D, et al. Common transcriptome, plasma molecules, and imaging signatures in the aging brain and a Mendelian neurovascular disease, cerebral cavernous malformation. Geroscience. 2020;42:1351-63. https://doi.org/10.1007/s11357-020-00201-4. 15. Lopez-Ramirez MA, Pham A, Girard R, Wyseure T, Hale P, Yamashita A, et al. Cerebral cavernous malformations form an anticoagulant vascular domain in humans and mice. Blood. 2019;133:193-204. https://doi.org/10.1182/blood-2018-06-856062. 16. Lopez-Ramirez MA, Fonseca G, Zeineddine HA, Girard R, Moore T, Pham A, et al. Thrombospondin1 (TSP1) replacement prevents cerebral cavernous malformations. J Exp Med.2017;214:3331-46. https://doi.org/10.1084/jem.20171178. 17. Jenny Zhou H, Qin L, Zhang H, Tang W, Ji W, He Y, et al. Endothelial exocytosis of angiopoietin-2 resulting from CCM3 deficiency contributes to cerebral cavernous malformation. Nat Med.2016;22:1033-42. https://doi.org/10.1038/nm.4169. 18. Wustehube J, Bartol A, Liebler SS, Brutsch R, Zhu Y, Felbor U, et al. Cerebral cavernous malformation protein CCM1 inhibits sprouting angiogenesis by activating DELTA-NOTCH signaling. Proc Natl Acad Sci U S A. 2010;107:12640-5. https://doi.org/10.1073/pnas.1000132107. 19. Stockton RA, Shenkar R, Awad IA, Ginsberg MH. Cerebral cavernous malformations proteins inhibit Rho kinase to stabilize vascular integrity. J Exp Med. 2010;207:881-96. https://doi.org/10.1084/jem.20091258. 20. Wetzel-Strong SE, Weinsheimer S, Nelson J, Pawlikowska L, Clark D, Starr MD, et al. Pilot investigation of circulating angiogenic and inflammatory biomarkers associated with vascular malformations. Orphanet J Rare Dis. 2021;16:372. https://doi.org/10.1186/s13023- 021-02009-7. 21. Koskimaki J, Zhang D, Li Y, Saadat L, Moore T, Lightle R, et al. Transcriptome clarifies mechanisms of lesion genesis versus progression in models of Ccm3 cerebral cavernous malformations. Acta Neuropathol Commun.2019;7:132. https://doi.org/10.1186/s40478-019- 0789-0. 22. Correia de Sousa M, Gjorgjieva M, Dolicka D, Sobolewski C, Foti M. Deciphering miRNAs' Action through miRNA Editing. Int J Mol Sci. 2019;20. https://doi.org/10.3390/ijms20246249. 23. Backes C, Meese E, Keller A. Specific miRNA Disease Biomarkers in Blood, Serum and Plasma: Challenges and Prospects. Mol Diagn Ther. 2016;20:509-18. https://doi.org/10.1007/s40291-016-0221-4. 24. Lugli G, Cohen AM, Bennett DA, Shah RC, Fields CJ, Hernandez AG, et al. Plasma Exosomal miRNAs in Persons with and without Alzheimer Disease: Altered Expression and Prospects for Biomarkers. PLoS One. 2015;10:e0139233. https://doi.org/10.1371/journal.pone.0139233. 25. Kar S, Bali KK, Baisantry A, Geffers R, Samii A, Bertalanffy H. Genome-Wide Sequencing Reveals MicroRNAs Downregulated in Cerebral Cavernous Malformations. J Mol Neurosci.2017;61:178-88. https://doi.org/10.1007/s12031-017-0880-6. 26. Lyne SB, Girard R, Koskimaki J, Zeineddine HA, Zhang D, Cao Y, et al. Biomarkers of cavernous angioma with symptomatic hemorrhage. JCI Insight. 2019;4. https://doi.org/10.1172/jci.insight.128577. 27. Wendler A, Wehling M. The translatability of animal models for clinical development: biomarkers and disease models. Curr Opin Pharmacol. 2010;10:601-6. https://doi.org/10.1016/j.coph.2010.05.009. 28. Tang AT, Choi JP, Kotzin JJ, Yang Y, Hong CC, Hobson N, et al. Endothelial TLR4 and the microbiome drive cerebral cavernous malformations. Nature. 2017;545:305-10. https://doi.org/10.1038/nature22075. 29. Morrison L, Akers A. Cerebral Cavernous Malformation, Familial. In: Adam MP, Ardinger HH, Pagon RA, Wallace SE, Bean LJH, Gripp KW, et al., editors. GeneReviews((R)). Seattle (WA)1993. 30. de Champfleur NM, Langlois C, Ankenbrandt WJ, Le Bars E, Leroy MA, Duffau H, et al. Magnetic resonance imaging evaluation of cerebral cavernous malformations with susceptibility-weighted imaging. Neurosurgery. 2011;68:641-7; discussion 7-8. https://doi.org/10.1227/NEU.0b013e31820773cf. 31. Blondal T, Brunetto MR, Cavallone D, Mikkelsen M, Thorsen M, Mang Y, et al. Genome- Wide Comparison of Next-Generation Sequencing and qPCR Platforms for microRNA Profiling in Serum. Methods Mol Biol.2017;1580:21-44. https://doi.org/10.1007/978-1-4939- 6866-4_3. 32. Decker JT, Hall MS, Blaisdell RB, Schwark K, Jeruss JS, Shea LD. Dynamic microRNA activity identifies therapeutic targets in trastuzumab-resistant HER2(+) breast cancer. Biotechnol Bioeng.2018;115:2613-23. https://doi.org/10.1002/bit.26791. 33. Kim SH, MacIntyre DA, Binkhamis R, Cook J, Sykes L, Bennett PR, et al. Maternal plasma miRNAs as potential biomarkers for detecting risk of small-for-gestational-age births. EBioMedicine.2020;62:103145. https://doi.org/10.1016/j.ebiom.2020.103145. 34. Johnson JJ, Loeffert AC, Stokes J, Olympia RP, Bramley H, Hicks SD. Association of Salivary MicroRNA Changes With Prolonged Concussion Symptoms. JAMA Pediatr. 2018;172:65-73. https://doi.org/10.1001/jamapediatrics.2017.3884. 35. Rohart F, Gautier B, Singh A, Le Cao KA. mixOmics: An R package for 'omics feature selection and multiple data integration. PLoS Comput Biol. 2017;13:e1005752. https://doi.org/10.1371/journal.pcbi.1005752. 36. Dweep H, Gretz N, Sticht C. miRWalk database for miRNA-target interactions. Methods Mol Biol.2014;1182:289-305. https://doi.org/10.1007/978-1-4939-1062-5_25. 37. Sticht C, De La Torre C, Parveen A, Gretz N. miRWalk: An online resource for prediction of microRNA binding sites. PLoS One. 2018;13:e0206239. https://doi.org/10.1371/journal.pone.0206239. 38. Dweep H, Sticht C, Pandey P, Gretz N. miRWalk--database: prediction of possible miRNA binding sites by "walking" the genes of three genomes. J Biomed Inform. 2011;44:839-47. https://doi.org/10.1016/j.jbi.2011.05.002. 39. Sulakhe D, Balasubramanian S, Xie B, Feng B, Taylor A, Wang S, et al. Lynx: a database and knowledge extraction engine for integrative medicine. Nucleic Acids Res.2014;42:D1007- 12. https://doi.org/10.1093/nar/gkt1166. 40. Amur S, LaVange L, Zineh I, Buckman-Garner S, Woodcock J. Biomarker Qualification: Toward a Multiple Stakeholder Framework for Biomarker Development, Regulatory Acceptance, and Utilization. Clin Pharmacol Ther. 2015;98:34-46. https://doi.org/10.1002/cpt.136. 41. Iguchi T, Niino N, Tamai S, Sakurai K, Mori K. Absolute Quantification of Plasma MicroRNA Levels in Cynomolgus Monkeys, Using Quantitative Real-time Reverse Transcription PCR. J Vis Exp.2018. https://doi.org/10.3791/56850. 42. Johnston S, Gallaher Z, Czaja K. Exogenous reference gene normalization for real-time reverse transcription-polymerase chain reaction analysis under dynamic endogenous transcription. Neural Regen Res. 2012;7:1064-72. https://doi.org/10.3969/j.issn.1673- 5374.2012.14.004. 43. Dakterzada F, Targa A, Benitez ID, Romero-ElKhayat L, de Gonzalo-Calvo D, Torres G, et al. Identification and validation of endogenous control miRNAs in plasma samples for normalization of qPCR data for Alzheimer's disease. Alzheimers Res Ther. 2020;12:163. https://doi.org/10.1186/s13195-020-00735-x. 44. Andersen CL, Jensen JL, Orntoft TF. Normalization of real-time quantitative reverse transcription-PCR data: a model-based variance estimation approach to identify genes suited for normalization, applied to bladder and colon cancer data sets. Cancer Res. 2004;64:5245- 50. https://doi.org/10.1158/0008-5472.CAN-04-0496. 45. Pfaffl MW. A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res.2001;29:e45. https://doi.org/10.1093/nar/29.9.e45. 46. Selst MV, Jolicoeur P. A solution to the effect of sample size on outlier elimination. The quarterly journal of experimental psychology.1994;47:631-50. 47. Davies L, Gather U. The identification of multiple outliers. Journal of the American Statistical Association.1993;88:782-92. 48. Yuan JS, Reed A, Chen F, Stewart CN, Jr. Statistical analysis of real-time PCR data. BMC Bioinformatics.2006;7:85. https://doi.org/10.1186/1471-2105-7-85. 49. Dridi N, Giremus A, Giovannelli JF, Truntzer C, Hadzagic M, Charrier JP, et al. Bayesian inference for biomarker discovery in proteomics: an analytic solution. EURASIP J Bioinform Syst Biol.2017;2017:9. https://doi.org/10.1186/s13637-017-0062-4. 50. Girard R, Zeineddine HA, Koskimaki J, Fam MD, Cao Y, Shi C, et al. Plasma Biomarkers of Inflammation and Angiogenesis Predict Cerebral Cavernous Malformation Symptomatic Hemorrhage or Lesional Growth. Circ Res. 2018;122:1716-21. https://doi.org/10.1161/CIRCRESAHA.118.312680. 51. Girard R, Zeineddine HA, Fam MD, Mayampurath A, Cao Y, Shi C, et al. Plasma Biomarkers of Inflammation Reflect Seizures and Hemorrhagic Activity of Cerebral Cavernous Malformations. Transl Stroke Res. 2018;9:34-43. https://doi.org/10.1007/s12975- 017-0561-3. 52. Arlot S, Celisse A. A survey of cross-validation procedures for model selection. Statistics Surveys.2010;4:40-79, 40. 53. Xia J, Broadhurst DI, Wilson M, Wishart DS. Translational biomarker discovery in clinical metabolomics: an introductory tutorial. Metabolomics. 2013;9:280-99. https://doi.org/10.1007/s11306-012-0482-9. 54. Lien EC, Dibble CC, Toker A. PI3K signaling in cancer: beyond AKT. Curr Opin Cell Biol.2017;45:62-71. https://doi.org/10.1016/j.ceb.2017.02.007. 55. Alzahrani AS. PI3K/Akt/mTOR inhibitors in cancer: At the bench and bedside. Semin Cancer Biol.2019;59:125-32. https://doi.org/10.1016/j.semcancer.2019.07.009. 56. Chen H, Zhou L, Wu X, Li R, Wen J, Sha J, et al. The PI3K/AKT pathway in the pathogenesis of prostate cancer. Front Biosci (Landmark Ed). 2016;21:1084-91. https://doi.org/10.2741/4443. 57. Kar S, Samii A, Bertalanffy H. PTEN/PI3K/Akt/VEGF signaling and the cross talk to KRIT1, CCM2, and PDCD10 proteins in cerebral cavernous malformations. Neurosurg Rev. 2015;38:229-36; discussion 36-7. https://doi.org/10.1007/s10143-014-0597-8. 58. Martinez-Lopez A, Salvador-Rodriguez L, Montero-Vilchez T, Molina-Leyva A, Tercedor-Sanchez J, Arias-Santiago S. Vascular malformations syndromes: an update. Curr Opin Pediatr.2019;31:747-53. https://doi.org/10.1097/MOP.0000000000000812. 59. Ren AA, Snellings DA, Su YS, Hong CC, Castro M, Tang AT, et al. PIK3CA and CCM mutations fuel cavernomas through a cancer-like mechanism. Nature. 2021. https://doi.org/10.1038/s41586-021-03562-8. 60. Hong T, Xiao X, Ren J, Cui B, Zong Y, Zou J, et al. Somatic MAP3K3 and PIK3CA mutations in sporadic cerebral and spinal cord cavernous malformations. Brain. 2021. https://doi.org/10.1093/brain/awab117. 61. Weng J, Yang Y, Song D, Huo R, Li H, Chen Y, et al. Somatic MAP3K3 mutation defines a subclass of cerebral cavernous malformation. Am J Hum Genet. 2021;108:942-50. https://doi.org/10.1016/j.ajhg.2021.04.005. 62. Lisowska J, Rodel CJ, Manet S, Miroshnikova YA, Boyault C, Planus E, et al. The CCM1- CCM2 complex controls complementary functions of ROCK1 and ROCK2 that are required for endothelial integrity. J Cell Sci.2018;131. https://doi.org/10.1242/jcs.216093. 63. McKerracher L, Shenkar R, Abbinanti M, Cao Y, Peiper A, Liao JK, et al. A Brain-Targeted Orally Available ROCK2 Inhibitor Benefits Mild and Aggressive Cavernous Angioma Disease. Transl Stroke Res.2020;11:365-76. https://doi.org/10.1007/s12975-019-00725-8. 64. Wei S, Li Y, Polster SP, Weber CR, Awad IA, Shen L. Cerebral Cavernous Malformation Proteins in Barrier Maintenance and Regulation. Int J Mol Sci. 2020;21. https://doi.org/10.3390/ijms21020675. 65. Polster SP, Stadnik A, Akers AL, Cao Y, Christoforidis GA, Fam MD, et al. Atorvastatin Treatment of Cavernous Angiomas with Symptomatic Hemorrhage Exploratory Proof of Concept (AT CASH EPOC) Trial. Neurosurgery. 2019;85:843-53. https://doi.org/10.1093/neuros/nyy539. 66. Zhou Z, Tang AT, Wong WY, Bamezai S, Goddard LM, Shenkar R, et al. Cerebral cavernous malformations arise from endothelial gain of MEKK3-KLF2/4 signalling. Nature. 2016;532:122-6. https://doi.org/10.1038/nature17178. 67. Lopez-Ramirez MA, Lai CC, Soliman SI, Hale P, Pham A, Estrada EJ, et al. Astrocytes propel neurovascular dysfunction during cerebral cavernous malformation lesion formation. J Clin Invest.2021;131. https://doi.org/10.1172/JCI139570. 68. Park JH, Lee JY, Shin DH, Jang KS, Kim HJ, Kong G. Loss of Mel-18 induces tumor angiogenesis through enhancing the activity and expression of HIF-1alpha mediated by the PTEN/PI3K/Akt pathway. Oncogene. 2011;30:4578-89. https://doi.org/10.1038/onc.2011.174. 69. Hong CC, Tang AT, Detter MR, Choi JP, Wang R, Yang X, et al. Cerebral cavernous malformations are driven by ADAMTS5 proteolysis of versican. J Exp Med. 2020;217. https://doi.org/10.1084/jem.20200140. 70. Liu M, Banerjee R, Rossa C, Jr., D'Silva NJ. RAP1-RAC1 Signaling Has an Important Role in Adhesion and Migration in HNSCC. J Dent Res. 2020;99:959-68. https://doi.org/10.1177/0022034520917058. 71. Rho SS, Ando K, Fukuhara S. Dynamic Regulation of Vascular Permeability by Vascular Endothelial Cadherin-Mediated Endothelial Cell-Cell Junctions. J Nippon Med Sch. 2017;84:148-59. https://doi.org/10.1272/jnms.84.148. 72. Li X, Zhang R, Draheim KM, Liu W, Calderwood DA, Boggon TJ. Structural basis for small G protein effector interaction of Ras-related protein 1 (Rap1) and adaptor protein Krev interaction trapped 1 (KRIT1). J Biol Chem. 2012;287:22317-27. https://doi.org/10.1074/jbc.M112.361295. 73. Wu M, Wang G, Tian W, Deng Y, Xu Y. MiRNA-based Therapeutics for Lung Cancer. Curr Pharm Des.2018;23:5989-96. https://doi.org/10.2174/1381612823666170714151715. 74. Rupaimoole R, Slack FJ. MicroRNA therapeutics: towards a new era for the management of cancer and other diseases. Nat Rev Drug Discov. 2017;16:203-22. https://doi.org/10.1038/nrd.2016.246. 75. Burns TC, Li MD, Mehta S, Awad AJ, Morgan AA. Mouse models rarely mimic the transcriptome of human neurodegenerative diseases: A systematic bioinformatics-based critique of preclinical models. Eur J Pharmacol. 2015;759:101-17. https://doi.org/10.1016/j.ejphar.2015.03.021. 76. Mullane K, Williams M. Preclinical Models of Alzheimer's Disease: Relevance and Translational Validity. Curr Protoc Pharmacol.2019;84:e57. https://doi.org/10.1002/cpph.57. 77. Shenkar R, Venkatasubramanian PN, Wyrwicz AM, Zhao JC, Shi C, Akers A, et al. Advanced magnetic resonance imaging of cerebral cavernous malformations: part II. Imaging of lesions in murine models. Neurosurgery. 2008;63:790-7; discussion 7-8. https://doi.org/10.1227/01.NEU.0000315862.24920.49. 78. Zeineddine HA, Girard R, Saadat L, Shen L, Lightle R, Moore T, et al. Phenotypic characterization of murine models of cerebral cavernous malformations. Lab Invest. 2019;99:319-30. https://doi.org/10.1038/s41374-018-0030-y. 79. Nagy ZB, Bartak BK, Kalmar A, Galamb O, Wichmann B, Dank M, et al. Comparison of Circulating miRNAs Expression Alterations in Matched Tissue and Plasma Samples During Colorectal Cancer Progression. Pathol Oncol Res. 2019;25:97-105. https://doi.org/10.1007/s12253-017-0308-1. 80. Malczewska A, Kidd M, Matar S, Kos-Kudla B, Modlin IM. A Comprehensive Assessment of the Role of miRNAs as Biomarkers in Gastroenteropancreatic Neuroendocrine Tumors. Neuroendocrinology.2018;107:73-90. https://doi.org/10.1159/000487326.