MARTIN LISA (US)
DING LILI (US)
What Is Claimed Is: 1. A method of assessing risk of chronic post-surgical pain (CPSP) in a subject, the method comprising: (i) obtaining a biological sample from a subject; (ii) analyzing the biological sample to determine a genetic profile of the subject, wherein the genetic profile comprises a combination of single nucleotide polymorphisms (SNPs) in one or more of the genes selected from the group consisting of ATXN1, PRKCA, CACNG2, CTSG, DRD2, KCND2, KCNJ3, KCNJ6, and KCNK3; (iii) calculating a polygenic risk score (PRS) based on the genetic profile determined in step (ii); and (iv) assessing a risk of developing CPSP in the subject based on the PRS. 2. The method of claim 1, further comprising applying to the subject a pain management approach based on the subject’s risk of developing CPSP. 3. The method of claim 1 or claim 2, wherein the combination of SNPs comprises SNPs selected from the following: (a) rs61131185, rs12665284, rs493352, rs12198202, rs1150635, rs3812204, rs4716060, rs531089, and/or rs6459476 in ATXN1; (b) rs9754467 in CACNG2; (c) rs1957523 in CTSG; (d) rs7125415 in DRD2; (e) rs202146909 and/or rs1992701 in KCNJ3; (f) rs2850125 in KCNJ6; (g) rs2891519 in KCNK3; (h) rs7809109 in KCND2; and (i) rs62069959, rs11079653, rs9914723, rs7220480, rs200369418, rs227912, and/or rs744214 in PRKCA. 4. The method of claim 3, wherein the combination of SNPs comprises: (a) rs61131185, rs12665284, rs493352, rs12198202, rs1150635, rs3812204, rs4716060, rs531089, and rs6459476 in ATXN1; (b) rs9754467 in CACNG2; (c) rs1957523 in CTSG; (d) rs7125415 in DRD2; (e) rs202146909 and rs1992701 in KCNJ3; (f) rs2850125 in KCNJ6; (g) rs2891519 in KCNK3; (h) rs7809109 in KCND2; and (i) rs62069959, rs11079653, rs9914723, rs7220480, rs200369418, rs227912, and rs744214 in PRKCA. 5. The method of claim 3, wherein the combination of SNPs consists of: (a) rs61131185, rs12665284, rs493352, rs12198202, rs1150635, rs3812204, rs4716060, rs531089, and rs6459476 in ATXN1; (b) rs9754467 in CACNG2; (c) rs1957523 in CTSG; (d) rs7125415 in DRD2; (e) rs202146909 and rs1992701 in KCNJ3; (f) rs2850125 in KCNJ6; (g) rs2891519 in KCNK3; (h) rs7809109 in KCND2; and (i) rs62069959, rs11079653, rs9914723, rs7220480, rs200369418, rs227912, and rs744214 in PRKCA. 6. The method of claim 1 or claim 2, wherein the one or more genes are selected from the group consisting of CACNG2, DRD2, KCND2, KCNJ3, KCNJ6, and PRKCA. 7. The method of claim 6, wherein the combination of SNPs comprises SNPs selected from the following: (a) rs9754467 and/or rs713952 in CACNG2; (b) rs7125415 in DRD2; (c) rs17376373, rs10488301, rs7809109, rs67881942, and/or rs17142908 in KCND2; (d) rs202146909, rs77929576, and/or rs1992701 in KCNJ3; (e) rs7281804, rs2836019, rs2850125, rs1892681, rs11088404, rs858018, rs13048511, rs7280905, and/or rs2032090 in KCNJ6; and (f) rs4791072, rs11079653, rs62069959, rs744214, rs7225452, rs6504424, rs973753, rs61687889, rs56399646, rs9635753, rs9914792, rs7220480, rs9914723, rs9903921, rs35105653, rs200369418, rs227912, rs58162388, rs9890911, rs11079656, rs12945884, and/or rs7218425 in PRKCA. 8. The method of claim 7, wherein the combination of SNPs comprises: (a) rs9754467 and rs713952 in CACNG2; (b) rs7125415 in DRD2; (c) rs17376373, rs10488301, rs7809109, rs67881942, and rs17142908 in KCND2; (d) rs202146909, rs77929576, and rs1992701 in KCNJ3; (e) rs7281804, rs2836019, rs2850125, rs1892681, rs11088404, rs858018, rs13048511, rs7280905, and rs2032090 in KCNJ6; and (f) rs4791072, rs11079653, rs62069959, rs744214, rs7225452, rs6504424, rs973753, rs61687889, rs56399646, rs9635753, rs9914792, rs7220480, rs9914723, rs9903921, rs35105653, rs200369418, rs227912, rs58162388, rs9890911, rs11079656, rs12945884, and rs7218425 in PRKCA. 9. The method of claim 7, wherein the combination of SNPs consists of: (a) rs9754467 and rs713952 in CACNG2; (b) rs7125415 in DRD2; (c) rs17376373, rs10488301, rs7809109, rs67881942, and rs17142908 in KCND2; (d) rs202146909, rs77929576, and rs1992701 in KCNJ3; (e) rs7281804, rs2836019, rs2850125, rs1892681, rs11088404, rs858018, rs13048511, rs7280905, and rs2032090 in KCNJ6; and (f) rs4791072, rs11079653, rs62069959, rs744214, rs7225452, rs6504424, rs973753, rs61687889, rs56399646, rs9635753, rs9914792, rs7220480, rs9914723, rs9903921, rs35105653, rs200369418, rs227912, rs58162388, rs9890911, rs11079656, rs12945884, and rs7218425 in PRKCA. 10. The method of any one of claims 1-9, wherein the subject is a human patient who is scheduled for or has undergone a surgery. 11. The method of claim 10, wherein the human patient is a child or an adolescent. 12. The method of claim 10 or claim 11, wherein the surgery is a spine and/or pectus surgery. 13. The method of any one of claims 1-12, further comprising determining one or more non-genetic factors of the subject, wherein the one or more non-genetic factors are selected from the group consisting of childhood anxiety sensitivity index (CASI), pain catastrophizing (PCS-C) score, pediatric pain sensitivity score, surgical duration, and area under curve (AUC) of a pain score at days 1 and 2 post-surgery (Pain_AUC_POD12). 14. The method of any one of claims 1-13, wherein the PRS is weighted by regression coefficients of the SNPs in the SNP combination to produce a weighted PRS. 15. The method of claim 14, wherein the risk of CPSP assessed in step (iv) is based on the weighted PRS in combination with one or more of the non-genetic factors by a regression model. 16. The method of claim 15, wherein the risk of CPSP assessed in step (iv) is based on: (a) CASI, surgical duration, Pain_AUC_POD12, and weighted PRS; or (b) CASI and weighted PRS. 17. The method of any one of claims 2-16, wherein the pain management approach comprises administration of an analgesic to the subject before and/or after the surgery. 18. The method of claim 17, wherein the analgesic is locally or systemically administered. 19. The method of claim 17 or claim 18, wherein the analgesic comprises bupivacaine, ropivacaine, prilocaine, mepivacaine, chloroprocaine, lidocaine, gabapentin, morphine, fentanyl, oxycodone, hydrocodone, codeine, clonidine, or a combination thereof. 20. The method of any one of claims 2-19, wherein the pain management approach comprises a psychosocial therapy. 21. A method for determining a genetic profile, the method comprising: (a) obtaining a biological sample from a subject in need thereof, (b) isolating nucleic acids from the biological sample, (c) detecting a combination of single nucleotide polymorphisms (SNPs) in one or more of the genes selected from the group consisting of ATXN1, PRKCA, CACNG2, CTSG, DRD2, KCND2, KCNJ3, KCNJ6, and KCNK3; and (d) determining a genetic profile of the subject based on the SNP combination detected in step (c). 22. The method of claim 21, wherein the combination of SNPs comprises SNPs selected from the following: (a) rs61131185, rs12665284, rs493352, rs12198202, rs1150635, rs3812204, rs4716060, rs531089, and/or rs6459476 in ATXN1; (b) rs9754467 in CACNG2; (c) rs1957523 in CTSG; (d) rs7125415 in DRD2; (e) rs202146909 and/or rs1992701 in KCNJ3; (f) rs2850125 in KCNJ6; (g) rs2891519 in KCNK3; (h) rs7809109 in KCND2; and (i) rs62069959, rs11079653, rs9914723, rs7220480, rs200369418, rs227912, and/or rs744214 in PRKCA. 23. The method of claim 22, wherein the combination of SNPs comprises: (a) rs61131185, rs12665284, rs493352, rs12198202, rs1150635, rs3812204, rs4716060, rs531089, and rs6459476 in ATXN1; (b) rs9754467 in CACNG2; (c) rs1957523 in CTSG; (d) rs7125415 in DRD2; (e) rs202146909 and rs1992701 in KCNJ3; (f) rs2850125 in KCNJ6; (g) rs2891519 in KCNK3; (h) rs7809109 in KCND2; and (i) rs62069959, rs11079653, rs9914723, rs7220480, rs200369418, rs227912, and rs744214 in PRKCA. 24. The method of claim 22, wherein the combination of SNPs consists of: (a) rs61131185, rs12665284, rs493352, rs12198202, rs1150635, rs3812204, rs4716060, rs531089, and rs6459476 in ATXN1; (b) rs9754467 in CACNG2; (c) rs1957523 in CTSG; (d) rs7125415 in DRD2; (e) rs202146909 and rs1992701 in KCNJ3; (f) rs2850125 in KCNJ6; (g) rs2891519 in KCNK3; (h) rs7809109 in KCND2; and (i) rs62069959, rs11079653, rs9914723, rs7220480, rs200369418, rs227912, and rs744214 in PRKCA. 25. The method of claim 21, wherein the one or more genes are selected from the group consisting of CACNG2, DRD2, KCND2, KCNJ3, KCNJ6, and PRKCA. 26. The method of claim 25, wherein the combination of SNPs comprises SNPs selected from the following: (a) rs9754467 and/or rs713952 in CACNG2; (b) rs7125415 in DRD2; (c) rs17376373, rs10488301, rs7809109, rs67881942, and/or rs17142908 in KCND2; (d) rs202146909, rs77929576, and/or rs1992701 in KCNJ3; (e) rs7281804, rs2836019, rs2850125, rs1892681, rs11088404, rs858018, rs13048511, rs7280905, and/or rs2032090 in KCNJ6; and (f) rs4791072, rs11079653, rs62069959, rs744214, rs7225452, rs6504424, rs973753, rs61687889, rs56399646, rs9635753, rs9914792, rs7220480, rs9914723, rs9903921, rs35105653, rs200369418, rs227912, rs58162388, rs9890911, rs11079656, rs12945884, and/or rs7218425 in PRKCA. 27. The method of claim 26, wherein the combination of SNPs comprises: (a) rs9754467 and rs713952 in CACNG2; (b) rs7125415 in DRD2; (c) rs17376373, rs10488301, rs7809109, rs67881942, and rs17142908 in KCND2; (d) rs202146909, rs77929576, and rs1992701 in KCNJ3; (e) rs7281804, rs2836019, rs2850125, rs1892681, rs11088404, rs858018, rs13048511, rs7280905, and rs2032090 in KCNJ6; and (f) rs4791072, rs11079653, rs62069959, rs744214, rs7225452, rs6504424, rs973753, rs61687889, rs56399646, rs9635753, rs9914792, rs7220480, rs9914723, rs9903921, rs35105653, rs200369418, rs227912, rs58162388, rs9890911, rs11079656, rs12945884, and rs7218425 in PRKCA. 28. The method of claim 26, wherein the combination of SNPs consists of: (a) rs9754467 and rs713952 in CACNG2; (b) rs7125415 in DRD2; (c) rs17376373, rs10488301, rs7809109, rs67881942, and rs17142908 in KCND2; (d) rs202146909, rs77929576, and rs1992701 in KCNJ3; (e) rs7281804, rs2836019, rs2850125, rs1892681, rs11088404, rs858018, rs13048511, rs7280905, and rs2032090 in KCNJ6; and (f) rs4791072, rs11079653, rs62069959, rs744214, rs7225452, rs6504424, rs973753, rs61687889, rs56399646, rs9635753, rs9914792, rs7220480, rs9914723, rs9903921, rs35105653, rs200369418, rs227912, rs58162388, rs9890911, rs11079656, rs12945884, and rs7218425 in PRKCA. 29. The method of any one of claims 21-28, wherein the subject in need thereof is a human patient who is scheduled for or has undergone a surgery. 30. The method of claim 29, wherein the human patient is a child or an adolescent. 31. The method of claim 29 or claim 30, wherein the surgery is a spine and/or pectus surgery. 32. The method of any one of claims 21-31, wherein the biological sample from a subject in need thereof is blood, plasma, salvia, urine, sweat, feces, a buccal smear, or a tissue. 33. A kit for determining a genetic portfolio of a subject, the kit comprising: (i) means for determining a combination of single nucleotide polymorphisms (SNPs) in one or more of the genes selected from the group consisting of ATXN1, PRKCA, CACNG2, CTSG, DRD2, KCND2, KCNJ3, KCNJ6, and KCNK3. 34. The kit of claim 33, wherein the combination of SNPs comprises SNPs selected from the following: (a) rs61131185, rs12665284, rs493352, rs12198202, rs1150635, rs3812204, rs4716060, rs531089, and/or rs6459476 in ATXN1; (b) rs9754467 in CACNG2; (c) rs1957523 in CTSG; (d) rs7125415 in DRD2; (e) rs202146909 and/or rs1992701 in KCNJ3; (f) rs2850125 in KCNJ6; (g) rs2891519 in KCNK3; (h) rs7809109 in KCND2; and (i) rs62069959, rs11079653, rs9914723, rs7220480, rs200369418, rs227912, and/or rs744214 in PRKCA. 35. The kit of claim 34, wherein the combination of SNPs comprises: (a) rs61131185, rs12665284, rs493352, rs12198202, rs1150635, rs3812204, rs4716060, rs531089, and rs6459476 in ATXN1; (b) rs9754467 in CACNG2; (c) rs1957523 in CTSG; (d) rs7125415 in DRD2; (e) rs202146909 and rs1992701 in KCNJ3; (f) rs2850125 in KCNJ6; (g) rs2891519 in KCNK3; (h) rs7809109 in KCND2; and (i) rs62069959, rs11079653, rs9914723, rs7220480, rs200369418, rs227912, and rs744214 in PRKCA. 36. The kit of claim 34, wherein the combination of SNPs consists of: (a) rs61131185, rs12665284, rs493352, rs12198202, rs1150635, rs3812204, rs4716060, rs531089, and rs6459476 in ATXN1; (b) rs9754467 in CACNG2; (c) rs1957523 in CTSG; (d) rs7125415 in DRD2; (e) rs202146909 and rs1992701 in KCNJ3; (f) rs2850125 in KCNJ6; (g) rs2891519 in KCNK3; (h) rs7809109 in KCND2; and (i) rs62069959, rs11079653, rs9914723, rs7220480, rs200369418, rs227912, and rs744214 in PRKCA. 37. The kit of claim 33, wherein the one or more genes are selected from the group consisting of CACNG2, DRD2, KCND2, KCNJ3, KCNJ6, and PRKCA. 38. The kit of claim 37, wherein the combination of SNPs comprises SNPs selected from the following: (a) rs9754467 and/or rs713952 in CACNG2; (b) rs7125415 in DRD2; (c) rs17376373, rs10488301, rs7809109, rs67881942, and/or rs17142908 in KCND2; (d) rs202146909, rs77929576, and/or rs1992701 in KCNJ3; (e) rs7281804, rs2836019, rs2850125, rs1892681, rs11088404, rs858018, rs13048511, rs7280905, and/or rs2032090 in KCNJ6; and (f) rs4791072, rs11079653, rs62069959, rs744214, rs7225452, rs6504424, rs973753, rs61687889, rs56399646, rs9635753, rs9914792, rs7220480, rs9914723, rs9903921, rs35105653, rs200369418, rs227912, rs58162388, rs9890911, rs11079656, rs12945884, and/or rs7218425 in PRKCA. 39. The kit of claim 38, wherein the combination of SNPs comprises: (a) rs9754467 and rs713952 in CACNG2; (b) rs7125415 in DRD2; (c) rs17376373, rs10488301, rs7809109, rs67881942, and rs17142908 in KCND2; (d) rs202146909, rs77929576, and rs1992701 in KCNJ3; (e) rs7281804, rs2836019, rs2850125, rs1892681, rs11088404, rs858018, rs13048511, rs7280905, and rs2032090 in KCNJ6; and (f) rs4791072, rs11079653, rs62069959, rs744214, rs7225452, rs6504424, rs973753, rs61687889, rs56399646, rs9635753, rs9914792, rs7220480, rs9914723, rs9903921, rs35105653, rs200369418, rs227912, rs58162388, rs9890911, rs11079656, rs12945884, and rs7218425 in PRKCA. 40. The kit of claim 38, wherein the combination of SNPs consists of: (a) rs9754467 and rs713952 in CACNG2; (b) rs7125415 in DRD2; (c) rs17376373, rs10488301, rs7809109, rs67881942, and rs17142908 in KCND2; (d) rs202146909, rs77929576, and rs1992701 in KCNJ3; (e) rs7281804, rs2836019, rs2850125, rs1892681, rs11088404, rs858018, rs13048511, rs7280905, and rs2032090 in KCNJ6; and (f) rs4791072, rs11079653, rs62069959, rs744214, rs7225452, rs6504424, rs973753, rs61687889, rs56399646, rs9635753, rs9914792, rs7220480, rs9914723, rs9903921, rs35105653, rs200369418, rs227912, rs58162388, rs9890911, rs11079656, rs12945884, and rs7218425 in PRKCA. 41. The kit of any one of claims 33-40, wherein the means for determining the combination of SNPs comprise a set of primer pairs, each of which is for amplifying a fragment encompassing one of the SNPs in the combination and wherein the set of primer pairs collectively amply fragments encompassing the SNPs in the combination. 42. The kit of any one of claims 33-41, wherein the means for determining the combination of SNPs comprise a set of oligonucleotides, each of which is for detecting one of the SNPs in the combination, and wherein the set of oligonucleotides collectively detects the SNPs in the combination. 43. The kit of claim 42, wherein the set of oligonucleotides is attached to a microarray chip. 44. The kit of any one claims 33-43, wherein the kit further comprises: (ii) a tool for collecting a biological sample from a subject; (iii) a container for placing the biological sample; and/or (iv) one or more reagents for extracting nucleic acids from the biological sample. 45. A method of assessing risk of chronic post-surgical pain (CPSP) in a subject, the method comprising: (i) determining one or more non-genetic factors of a subject, wherein the one or more non-genetic factors are selected from the group consisting of childhood anxiety sensitivity index (CASI), pain catastrophizing (PCS-C) score, pediatric pain sensitivity score, surgical duration, and area under curve (AUC) of a pain score at days 1 and 2 post-surgery (Pain_AUC_POD12); and (ii) assessing risk of CPSP of the subject based on the one or more non-genetic factors. 46. The method of claim 45, further comprising applying to the subject a pain management approach based on the subject’s risk of developing CPSP. 47. The method of claim 45 or claim 46, wherein the subject is a human patient who is scheduled for or has undergone a surgery. 48. The method of claim 47, wherein the human patient is a child or an adolescent. 49. The method of claim 47 or claim 48, wherein the surgery is a spine and/or pectus surgery. 50. The method of any one of claims 47-49, wherein the pain management approach comprises administration of an analgesic to the subject before and/or after the surgery. 51. The method of claim 50, wherein the analgesic is locally or systemically administered. 52. The method of claim 50 or claim 51, wherein the analgesic comprises bupivacaine, ropivacaine, prilocaine, mepivacaine, chloroprocaine, lidocaine, gabapentin, morphine, fentanyl, oxycodone, hydrocodone, codeine, clonidine, or a combination thereof. 53. The method of any one of claims 47-52, wherein the pain management approach comprises a psychosocial therapy. |
remainder of the disclosure in any way whatsoever. All publications cited herein are incorporated by reference for the purposes or subject matter referenced herein. EXAMPLES While the present disclosure has been described with reference to the specific embodiments thereof, it should be understood by those skilled in the art that various changes may be made and equivalents may be substituted without departing from the true spirit and scope of the disclosure. In addition, many modifications may be made to adapt a particular situation, material, composition of matter, process, process step or steps, to the objective, spirit, and scope of the present disclosure. All such modifications are intended to be within the scope of the disclosure. Example 1. Recruitment and Selection for Observational Prospective Cohort. An observational prospective cohort study was conducted in 171 adolescents with idiopathic scoliosis, pectus excavatum, and/or kyphosis undergoing posterior spine fusion using standard surgical techniques, anesthetic and pain protocols. The studies were registered with ClinicalTrials.gov (Identifier: NCT01839461, NCT01731873), the disclosures of which are incorporated herein in their entirety. Regarding inclusion criteria, healthy non-obese children aged 10-18 years of American Society of Anesthesiologists (ASA) physical status less than or equal to two (mild systemic disease) with a diagnosis of idiopathic scoliosis and/or kyphosis, scheduled to undergo elective spinal fusion were selected for the study. The ASA Physical Status Classification System that was used for inclusion criteria is provided in Table 1 below. Table 1: Physical Status (PS) Classification Levels, Definitions and ASA-Approved Examples 21
Excluded from the study were females who were pregnant or breastfeeding, subjects with a diagnosis of chronic pain or opioid use in the past six months, hepatic/renal disease and/or developmental delays. The rationale for using spine pediatric cohorts in the study was that idiopathic scoliosis (increased spine curvature), kyphosis (outward spine curvature), and pectus excavatum (caved-in chest) are common, musculoskeletal chest wall deformities. Spine fusion for scoliosis/ kyphosis and endoscopic Nuss procedure for pectus were known to be the most painful corrective surgeries healthy children/adolescents undergo. Further, unlike
adults, it was known that children/adolescents have minimal/no pain before surgery and surgery was usually the child’s first exposure to severe pain and opioids. A 37-40% incidence of CPSP was observed at 2-3 months and 1 year after spine fusion, which aligned with postsurgical outcomes data from international pediatric spine surgery registries. Example 2. Assessment of Psychosocial and Perioperative Factors as Predictors of CPSP in a Scoliosis Cohort A cohort of 171 children undergoing spinal fusion under standard anesthesia/pain protocols was selected according to the protocol in Example 1. Preoperative data on demographics (sex, age, race, etc.), weight, home medications, and scoliosis curve from X- ray reports, were collected from the cohort. The numerical rating scale (NRS) for self-report of pain intensity in children and adolescents – a validated measure of pain in children aged ~7 to 17 years – was determined in a manner similar to that described in von Baeyer, (2009), European Journal of Pain, 13: 1005-1007, the disclosure of which is incorporated herein in its entirety. Because socioeconomic status can influence pain responses, preoperative data also includes socioeconomic status (SES) data on education level and financial condition of the family. Questionnaires to assess pain catastrophizing (Pain catastrophizing scale/PCS), functional disability (FDI), and anxiety sensitivity (childhood anxiety sensitivity index (CASI)), were administered preoperatively in a manner as described in Crombez et al., (2003) Pain, 104(3):639-646, Walker et al, (1991) Journal of Pediatric Psychology 16(1):39-58, and Silverman et al., (1991) J Clin Child Psychol 20:162-168, respectively, the disclosures of which are incorporated herein in their entirety. In brief, the data for psychosocial variables (e.g., PCS, CASI, FDI), were collected from answers obtained from child and parent questionnaires, the details of which are provided in Table 2 below. Table 2: Validated Questionnaires for Psychosocial Variables Used in Study - 23 -
Additionally, psychophysical test data, as described in Table 3, was collected preoperatively. Since the order of the tests may make a difference to results and to prevent bias, the less dynamic short tests - TPD and pressure pain (PPT) order based on ID (odd/even), are randomized and then TS is performed. CPM (a more dynamic test) is performed last since there is a risk of carryover with CPM. Table 3: Psychophysical Tests Performed and Rationale for Performance in Study Somatosensory assessment is also performed preoperatively based on parameters developed by the German Neuropathic Pain Network as described in Lim et al., (2010) Lancet 380(9859): 2224-60 and King et al., (2004) Journal of Pain 5(7):377-84, the disclosures of which are incorporated herein in their entirety. Somatosensory assessments are performed done prior to surgery by the same examiner at each testing site, to minimize observer bias. Participants are comfortably positioned. To familiarize participants with the test, a standardized set of instructions is read, and practice trials demonstrated on the non- testing site of the participants’ bodies, preferably without the parent/legal guardians present to eliminate parental influence. Each mechanical procedure is conducted over sets of trials to derive an average mean measure and/or pain (numerical pain rating, NRS). Each device is wiped and cleaned with 70% alcohol after being used on each participant. All patients received total intravenous anesthesia (propofol and remifentanil) and midazolam in the intraoperative period followed by standardized doses of patient controlled analgesia (morphine or hydromorphone) postoperatively. Pertinent surgical details (duration, number of vertebral levels fused) and anesthetic data (propofol and remifentanil doses) were collected. In the immediate Postoperative, pain scores (every four hours), doses of morphine equivalents administered over on postoperative days (POD) one and two were recorded. After hospital discharge, at 6-12 months, patients were asked to rate their average pain score (NRS) over the previous week and open-ended questions about the nature and site of pain, use of medications/alternative therapies/physician consults for pain, and functional disability (FDI). Briefly, questions (pain score, anxiety and medication dose/use over 24 hours) are sent at 6 PM every day until no opioid use is reported for 3 consecutive days. These follow up questionnaires are administered through Twilio messaging links via REDCap, electronic diary or phone by a trained research coordinator. Typically, opioid use decreases by about 2-3 weeks. Documentation about opioid use after the texting period was obtained through questionnaires administered at 2-3 months and 4-6 months after surgery. The questionnaires are administered in a standard fashion, without prompting answers, giving subjects time to think. Patients were asked to rate their current pain score (NRS) at rest and with daily activity, maximum pain score over the previous week, nature and site of pain, medications/alternative therapies/physician consults for pain. At least 3 phone call attempts at all provided contact numbers to ensure contact, use of monthly reminders, and incentives to facilitate retention and future follow-up are performed. Rate of prescription refills and electronic medication monitoring are alternative strategies to obtain objective opioid use data when contact with patient post-surgery is lost. Outcomes evaluated included a a) binary outcome: CPSP which was determined based on a cut-off of pain score>3/10 on an 11-point Numerical rating scale (range 0-10) at 6- 12 months after surgery. This cut-off was used as NRS pain scores>3 (moderate/severe pain) at three months was a known predictor for persistence of pain, associated with functional disability. Maladaptive coping strategies in children and negative parent responses affect chronic pain conditions in children. Three coping sub-scales from Pain Coping Scale (approach, behavioral distraction, emotion-focused avoidance) and 3 parenting factors (Protection, Minimizing, Encouragement) from Adult Response to Child’s Symptoms are considered for association with CPSP. PROMIS domains have shown construct validity and responsiveness to change in children (ages 8-18) with chronic pain; as suc, pain Interference and Depression is also assessed. Descriptive statistics (mean and standard deviation for continuous, and frequency and percentage for categorical variables) were calculated for all study variables. Additionally, univariate association between independent variables and outcomes (primary and secondary) were examined using 2-sample t-tests or Wilcoxon rank-sum test, ANOVA or Kruskal Wallis, Spearman or Pearson correlation coefficient, and chi-square or Fisher’s exact tests, as appropriate. For each outcome variable, linear (for continuous outcome) or logistic (for binary CPSP outcome) regression models with one primary independent variable of interest at a time were conducted, adjusting for covariates. To compare QST, data are calculated for each individual QST variable or by z transformation as described in Baron et al., (2017) Pain, 158(2):261-72, the disclosure of which is incorporated herein in its entirety. Z scores of zero represent a value corresponding to the mean of the non-CPSP cohort for that measure; z scores above “0” indicate a gain of function when the patient was more sensitive to the test stimuli compared with controls, z scores below “0” indicate loss of function referring to a lower sensitivity of the patient. For each outcome variable, linear regression models with one primary independent variable of interest at a time are conducted, adjusting for covariates. Increase in coefficient of determination (R 2 ) due the additional primary independent variable of interest were reported. To build a multivariable model for each outcome, factors associated at p < 0.10 in univariate analysis are subject to testing of multicollinearity (and removed if collinearity exists) and then entered into multivariable regression models. Stepwise selection was used to derive a final model for each outcome where only variables with a p <0.05 are retained. R 2 is reported for the final models. A random effect is included in all regression models to account for the cluster effect of study center. As a secondary analysis, pairwise correlation between continuous and categorical outcomes are examined using either Spearman or Pearson correlation coefficient as appropriate. Cronbach's alpha is used to assess internal consistency of the questionnaires. In general, an alpha of >0.7 is acceptable. Demographics and summary of the variables examined for the prospective cohort are given in Table 4 where the following abbreviations were used: CASI: Childhood anxiety sensitivity index; PCS: Pain catastrophizing scale; AUC: Area under curve of pain scores over postoperative days (POD) 1 and 2; CPSP: Chronic post-surgical pain; FDI: Functional disability index. Table 4: Baseline and pain follow-up characteristics of the surgical cohort
CPSP outcome was determined for 131 of the 171 patients (loss to follow up of about 23%). The characteristics of the cohort that was lost to follow up and those followed for 6-12 months were examined for all pertinent measures included in the models and did not find any significant differences in terms of age (p=0.390), sex (p=0.361), race (0.906), CASI (p=0.364), surgical duration (p=0.322) and preoperative pain (p=0.879). Incidence of CPSP was found to be 53/131 (40.4%). Although 83% of the cohort had no preoperative pain, there was a 37.8% incidence of post-surgical pain at 2-3 months, and 41.8% at one year after spine fusion. FDI scores were higher (p=0.001) and PedsQL scores were lower (P=0.001) in patients with CPSP vs. those without. CPSP was defined as numerical rating scale (NRS) pain scores >3/10 at 6-12 months. AUC of repeat pain scores over 6 and 12 months mesured cumulative pain experience over time after surgery and was well correlated with CPSP (p<0.0001). Figure 1. Pain trajectories measured for over six years after spine surgery were evaluated in the spine cohort. Of the 66% of patients who developed CPSP, they reported high pain scores throughout the 6 years follow-up (h-h) and those who did not had low scores throughout (l-l). However, 33% of those who developed CPSP had unexpected trajectories. Figure 2. Higher anxiety sensitivity causes fear of pain and avoidance behavior leading to chronic pain and disability. Accordingly, PPST, (composite assessment of pain catastrophizing, fear of pain, anxiety, and depressive symptoms) was evaluated in the cohort. On univariate analysis, PPST was positively associated with CPSP (p<0.002; β:0.95, SE 0.26) functional disability (p<0.0001; β2.92, SE 0.43), and negatively associated with QOL scores (p<0.0001; β-20.72, SE 3.64). Child anxiety and pain catastrophizing, parent anxiety and catastrophizing were evaluated as predictors of CPSP. In multivariate regression models, CASI was identified as significant psychosocial predictor of CPSP. Table 5. To assess further assess the diagnostic ability of psychological and perioperative factors as predictors phenotypes of risk for CPSP, a multiple regression model with psychosocial factors was developed which predicted CPSP with 75% predictive accuracy as determined by receiver operating characteristic (ROC) curves. Figure 3. Table 5: Psychosocial Predictors of CPSP Next, data from the psychosocial tests were subjected to Hierarchical cluster analysis (HCA) to group patients with similar structure into clusters of phenotypes. Briefly, HCA was used to group patients with similar structure into phenotypes based on 3 factors: 1) pain character (AUC and nature – inflammatory or neuropathic based on PainDETECT score (1- 12: nociceptive pain, and 19-38: neuropathic pain is likely)); 2) psychophysical characteristics; and 3) opioid use (dose and days of use over weeks after surgery). For cluster characteristics, continuous values of the measures or dichotomize by high tertile and low tertile for discordant phenotypes was used. Before clustering, a Principle Component Analysis (PCA) was performed to reduce dimensionality of data. A screen plot was analyzed to detect an “elbow” that suggested the number of PCs for entering hierarchical clustering. A wide variety of hierarchical clustering techniques were applied to achieve robust and meaningful data segregation, including divisive and agglomerative algorithm, Ward’s and other methods for the linkage criterion, Euclidean, Manhattan and correlation-bases distances for similarities. To detect an optimal number of clusters, heatmaps and dendrograms were visually inspected and R software (package NbClust) that identifies optimal number of clusters using multiple indexes was employed. These clusters were evaluated in the context of prior knowledge to identify the most parsimonious clustering. HCA and PCA were performed in R software (package factoextra v1.0.3). Cophenetic correlation, which was the Pearson correlation between actual and predicted distances based on clustering approach as calculated. A value of 0.75 or above was needed for goodness of cluster fit. Hierarchical cluster analysis identified five phenotype clusters based on high and low risk for acute postoperative pain, CPSP and CASI. Figure 4. PRS were able to differentiate the phenotype clusters on co-clustering, thus indicating that unique genotypes determine phenotype sub-groups. Figure 5. Example 3. Assessment of Psychosocial and Perioperative Factors as Predictors of CPSP in a Pectus Cohort A cohort of 7 children undergoing Pectus surgery under standard anesthesia/pain protocols was selected according to the protocol in Example 1. Preoperative data on demographics (sex, age, race), weight, home medications, and Haller’s Index (ratio of the measure of the transverse diameter of the chest, divided by the sagittal measure of the distance from the sternum to the vertebral body, reported on routine preoperative CT/MRI chest – in the pectus cohort), were collected from the cohort. NRS and the psychosocial variables PCS, FDI, CASI were assessed in the pectus cohort in the same manner as described in Example 2 for the scoliosis cohort. Similarly, psychophysical sensory test data was collected in same manner as described in Example 2 for the scoliosis cohort in 7 patients before pectus surgery. Briefly, two-point discrimination (TPD), pressure pain threshold testing (PPT), conditioned pain modulation (CPM), temporal summation (TS), and pain scores were collected and mapped over 2-6 months postoperatively. Pain intensity after hand insertion at 20 seconds and withdrawal time noted. Figures 6A-6E shows the preoperative sensory testing differences observed in each patient. Pain scores were collected and mapped over 2-6 months postoperatively through REDCap data collection using Twilio messaging and phone calls/emails. AUC under pain trajectories were calculated (except for patient ID NO.5 who withdrew from study). Figure 7. Data from the preoperative sensory tests for each patient were subjected to quantitative sensory testing (QST) to refine an endophenotype characterization. The resulting sensory testing profiles of each patient along with AUC are presented in Figure 8. Sensory profiles of patient ID NO.3 and patient ID NO.4, (lowest and highest AUC respectively) were almost mirror images. Direction of the responses was as expected. A closer look at the association of CPM response with AUC showed impaired CPM response (less negative pain response after cold modulation) was associated with higher AUC (p=0.02) indicating decreased descending inhibition of nociception, predicting a high pain phenotype (enhanced TS). Clusters of phenotypes were identified using HCA (in the manner described in Example 2) after decreasing dimensionality via principal component analysis. Figure 9. Cluster 1 (patient ID NOs.3, 7, 2) and cluster 2 (patient ID NOs.1, 4, 6) differed mainly in CPM responses. These data demonstrate feasibility of preoperative sensory testing in this population and risk clustering predicted by sensory testing risk profiles. Example 4. Identification of a Polygenic Risk Score (PRS) predictive for CPSP risk Genetic factors influence individual differences in pain perception but the effect these factors have on pain perception is small and causal variants are difficult to detect. It is possible that it is not just one genetic factor contributing to the risk of CPSP, but instead the combined effects of a large number of susceptibility loci may become large enough to be useful for targeted risk prediction and prevention. A polygenic risk score (PRS) can be used to model these weak contributions, where an individual's genetic risk is the sum of all their risk alleles weighted by significance of the corresponding allele in genome wide association studies. To identify polygenic risk scores for CPSP, an approach outlined in Figure 14 was followed. (i) Identification of “Training” and “Candidate” Genes for Psychosocial and/or Perioperative Phenotypes. To generate the “training set” of genes for each psychosocial and/or perioperative phenotype observed in either the scoliosis cohort or pectus cohort, a comprehensive literature search limited to human studies using electronic databases including PubMed and MEDLINE (according to the Preferred Reporting Items of Systematic Reviews and Meta-Analyses (PRISMA) guidelines), was conducted from January 2001 to December 2017, using the following search terms: ("postoperative pain" OR "postsurgical pain" OR "post-operative pain" OR "post-surgical pain" OR "postoperative analgesia" OR "postoperative opioid" OR “CPSP”) AND (genetic association OR polymorphism OR variant OR "genotype" OR “Genome wide association” OR “SNP”). Searches were limited to English language articles and human studies (including clinical study, clinical trial, multicenter study, observational study and twin study) using filters. The search was conducted from 2001/01/01 to 2017/12/31. Detailed results of this review including description of studies, genes, variants and outcomes studied are detailed in Chidambaran et al., (2019) J Pain, May 23. pii: S1526- 5900(19)30079-3, the disclosure of which is incorporated herein in its entirety. Ultimately, the literature search identified 31 “training” genes associated with postoperative pain and CPSP in humans. Those “training” gene were: COMT (rs6269, rs4633); GCH1 (rs3783641, rs8007267); COMT rs4680; ABCB1 C3435T; 5HTR2A rs6311; IFNG1 (rs2069727, rs2069718); IL1R1 rs3917332; IL1R2 rs11674595; IL4 rs2243248; IL10 (rs3024498, rs1878672, rs3024491); IL13 (rs1881457, rs1800925, rs1295686, rs20541); NFKB1 rs4648141; HLA-DRB1*4 and DQB1/03:02; PRKCA rs887797; CDH18 rs4866176; TG rs1133076; ATXN1 rs179997; DRD2 (rs4648317, rs12364283); NFKB1A rs8904; GCH1 rs4411417; CHRNA6 rs7828365; KCND2 (rs17376373, rs702414, rs802340, rs12706292); KCNJ3 (rs6435329, rs11895478, rs3106653, rs3111006, rs12471193, rs7574878, rs12995382); KCNJ6 rs2835925; KCNK3 (rs1662988, rs7584568); KCNK9 rs2014712; CACNG2 (rs4820242, rs2284015, rs2284017, rs2284018, rs1883988); COMT (rs4680, rs6269); P2X7R (rs208294, rs208296, rs7958311); KCNS1 (rs734784, rs13043825); TNF alpha rs1800629; and, GCH1 rs8007627. Next, the ToppFun application of the Transcriptome Ontology Pathway PubMed based prioritization of Genes (ToppGene) Suite for candidate gene prioritization was used similar to that described in Chen et al., (2009) Nucleic Acids Research 37(Web Server issue):W305-311, the disclosure of which is incorporated herein in its entirety. 1310 “candidate” genes enriched for CPSP were identified and prioritized using Toppgene suite based on functional enrichment using several gene ontology annotations, and curated gene sets associated with mouse phenotype–knockout studies. Training and candidate gene sets together formed the case set of genes whose variants were used for association analyses as described herein. (ii) DNA collection and Genotyping. Blood was drawn upon intravenous line placement for genotyping from 171 adolescents undergoing spine fusion (the Scoliosis Cohort described in Examples 1 and 2). DNA degradation or modification caused by environmental factors and by pathogens present in the sample will be prevented by using DNA/RNA Shield which preserves the genetic integrity of samples at ambient temperatures (cold-free) for >2 years. Deoxy ribonucleic acid (DNA) was isolated and purified using standard procedures on the same day, frozen at -20 ° C. Concentration and purity of genomic DNA was determined using a Thermo Scientific NanoDrop spectrophotometer, to insure a minimum of 400 ng of genomic DNA free of contaminants was obtained to be used in genotype assays. Genotyping was done using the Illumina Human Omni5 v41-0 array (85 patients), Human Omni5Exome v41-1 (33 patients) and Infinium Omni5-4-v1 (53 patients). Arrays were changed due to availability of new array which had more SNPs and functional ones. (iii) Genetic Association Analyses. Analyses were conducted using SAS 9.4 (SAS, Cary, NC) and R. Prior to genetic analyses, 121,301 SNPs from the sex chromosome, chromosome zero, mitochondrial, indels and other were excluded from analysis. Quality of SNP calls from the chip were also evaluated. SNPs were assessed for Hardy–Weinberg equilibrium (HWE) by means of goodness of fit χ2 test. SNPs deviating HWE (p ≤ 0.0001), whose MAF is less than 90%, or whose call rates fall below 95% were excluded. Thresholds for quality control for call rates at individual and SNP levels were 99% and 90%, respectively. Low-frequency variants were also excluded, threshold for minor allele frequencies was 10%. SNPs in high linkage disequilibrium (LD) (80%) were pruned out in PLINK using the command --indep-pairwise 5050.8. SNPs from autosomal chromosomes only were selected for analysis and were annotated using ANNOVAR software in a manner similar to that described in Wang et al., (2010) Nucleic Acids Research 38(16):e164-e164, the disclosure of which is incorporated herein in its entirety. SNPs located in intergenic regions and not associated with a specific gene according to ANNOVAR annotation were also excluded prior to analysis. Prior to genetic analyses, cryptic relatedness was checked using Graphical Representation of Relationship (GRR), in a similar manner as described in Abecasis et al., (2001) Bioinformatics Applications Note 2001;17(8):742-743, the disclosure of which is incorporated herein in its entirety. Principal component analysis was employed to confirm European and African continental ancestry using 482 validated ancestry informative markers. Concordance with self-reported race was > 95%. To identify significant SNPs, we used logistic and linear models for association of each SNP with CPSP and the pain score at 6-12 months as outcomes. In all association tests we used an additive genetic model in which major homozygotes were coded as 0, heterozygotes as 1, and minor homozygotes as 2. Surgical duration, CASI and acute postoperative pain over 48 hours (area under curve of pain scores over time) were used as covariates, based on our previous multiple regression model which did not include genetics. PLINK v.07 was used for association tests. There were 4,186,587 variants on the exome chip initially and 542,313 variants remained after exclusion. After quality control, of the 542,313 variants that were analyzable, 61,348 variants were annotated to training/candidate genes of interest. Pruning was then done based on Linkage disequilibrium (r 2 = 0.5) to remove variants that were highly linked and minimize correlation issues. Thus, a set of 33,104 variants were finally included for association analyses in our cohort. (iv) Gene Enrichment Analyses: Case gene variants for each outcome (training and prioritized candidate genes) were analyzed as sequence of cumulative sums of ranked variant sets with 10% increment. The first addend in each sequence was the training gene variant set. For each cumulative sum we compared the number of associations in our case sets that met the p<.05 threshold to the number of associations meeting the same criteria in over 10,000 matched runs of our control set of genes. SNPs from the control set were selected in the same ratio for MAF as it was observed in the case set. Specifically, we used MAF bands as follows: 10-15%:15-20%:20- 30%:30-50%. Empirical p-values of resampling tests were computed as follows: we calculated how many samples out of 10,000 had the number of significant SNPs equal to or greater than the number of significant SNPs from the case set and divided this number by 10,000. Training and ranked genes that formed the earliest cumulative group where a number of significant SNPs were greater than in the matched control group were considered as a minimal set of variants enriched for associations with corresponding outcomes. As shown in Figure 10, the box plots represent the number of significant single nucleotide polymorphisms (SNPs) in the 10,000 runs of control gene SNPs. The upper and lower bounds of the box represents the 75 th and 25 th percentile, respectively, and error bars represent the 5 th and 95 th percentiles. Vertical axis represents the number of SNPs and the horizontal axis is the centiles of the ranked case genes as described in methods using ToppGene. The Red dot represents the number of positive SNPs in the case set of genes. When the red dot was above the 95 th centile of the box plot whisker on the figure, this indicated that more case genes were associated with behavioral outcomes than expected, indicating the case genes were enriched compared to control-gene SNPs. At the 10 th centile and above, there were more case-gene SNPs significantly associated with behavioral outcomes than in the 10,000 runs of the control-gene SNPs, indicating gene enrichment. Because Figure 10 demonstrates main effects controlling for the group, these findings indicated that a complex network of genes/polymorphisms was associated with increased risk of CSPS. Case genes for each outcome were analyzed as sequence of cumulative sums of ranked genes with 10% increment. The first addend in each sequence was a number of training genes. For each cumulative sum we compared the number of associations in our case sets that met the p<.05 threshold to the number of associations meeting the same criteria in over 10,000 matched runs of our control set of genes. After adjusting for covariates, compared to control sets, there was enrichment of SNP associations in training set for CPSP. Training and ranked genes that formed the earliest cumulative group where a number of significant SNPs were greater than in the matched control group were considered as a minimal set of variants enriched for associations with corresponding outcomes. The number of genes and SNPs that were included in these significant case sets were 12 genes (80 SNPs) for CPSP ((ATXN1 (29); CACNG2 (2); CTSG (2); DRD2 (1); HLA-DQB1 (3); IL10 (1); KCNA1 (1); KC ND2 (5); KCNJ3 (3); KCNJ6 (9); KCNK3 (2); PRKCA (22)). The complete list of the 80 significant SNPs is provided in Table 6. Table 6: Cumulative Group of Significant SNPs
- 38 -
To minimize risk of overfitting and identify the minimal number of SNPs for the PRS, we used penalized regression with least absolute shrinkage and selection operator (LASSO) in R software (package glmnet) in a manner similar to that described in Friedman et al., (2010) J Stat Softw 33(1):1-22, the disclosure of which is incorporated herein. A controlling penalty parameter lambda was selected via cross-validation approach. After least absolute shrinkage and selection operator (LASSO), the final set included 9 genes and 24 variants. Chromosomal location, genetic annotation, function, minor allele frequency, Odds ratios for CPSP and beta for NRS at 6-12 months with p-values for the LASSO selected variants are provided in Table 7 (where PRKCA (protein kinase C alpha); DRD2 (dopamine receptor D2); ATXN1 (ataxin 1); KCNJ3 (potassium voltage-gated channel subfamily J member 3); CACNG2 (calcium voltage-gated channel auxiliary subunit gamma 2); CTSG (cathepsin G); KCNJ6 (potassium voltage-gated channel subfamily J member 6); KCNK3 (potassium two pore domain channel subfamily K member 3); KCND2 (potassium voltage-gated channel subfamily D member 2); #Linear regression coefficients were used to calculate weighted polygenic risk scores; *Risk allele).
I h c e T C M H C 0 0 O W 9 3 0 0 7 - 3 3 1 8 4 6 - 4 4 1 3 0 1 : . o N t e k c o D y e n r o t t A 1 . 6 6 (v) Enriched genomic pathways: Toppfun application of Toppgene suite was used to identify top pathways enriched by the genes with significant associations with each phenotype. The pathways enriched by the 9 genes associated with CPSP, based on Bonferroni correction for multiple adjustment cut-offs (p<0.05) are presented in Table 8. Table 8. Genetic pathways enriched by the genes whose variants contributed to weighted polygenic risk. (vi) Polygenic risk scores and multiple regression model: Weighted genetic risk were calculated from the SNPs selected by LASSO. Briefly, SNPs with non-zero coefficients in the LASSO model were selected for PRS calculation. Variants included in the PRS are presented in Table 9. Table 9: SNPs included in PRS calculation. PRS was calculated as a weighted sum of products between number of risk alleles and their corresponding regression coefficients. The full model included PRS and non-genetic predictors. A stepwise approach was exploited for selecting covariates for a reduced model. Covariables associated at p<0.05 entered a final predictive model. For model performances we used the area under the receiver operating characteristics curve (AUC). AUCs with 95% confidence intervals for clinical and genetic models were used for model comparison in SAS 9.4 (SAS. Cary, NC). The polygenic risk scores (PRS) ranged from 12.1 to 35.7 (mean: 25.2; SD 4.4) and were normally distributed. The full multiple regression model inclusive of PRS is presented in Table 10. Two predictors were remained in the reduced final models after stepwise selection. Our final predictive model is presented in Table 10 (where CASI: Childhood anxiety sensitivity index; OR: Odds ratio; BS: Bootstrapping; AUC: Area under curve of pain scores over postoperative days 1 and 2 (POD12) after spine fusion; PRS: Polygenic risk score.) Table 10: Multiple regression models evaluated for prediction of chronic post-surgical pain (CPSP) and results of bootstrapping. The predicted probability (with 95% CI) of CPSP for a subject having a median (for the cohort) CASI = 28.16 using the regression model is plotted as a function of the PRS in Figure 11. The probability of CPSP is higher than 50% at a PRS>26. Comparison of performance of the predictive model with three clinical predictors (CASI, surgery duration, and acute pain) and performance of the predictive model with generic predictor (PRS and CASI) showed statistically significant higher performance of genetic model. C-statistics for genetic model was 0.97 (95% CI 0.93-0.99) compared to 0.77 (95% CI 0.66-0.87) for non- genetic model (p=0.0007). Figure 15. Further, PRS predicted continuous AUC and dichotomous CPSP outcomes (p<0.0001). PRS were included in a multiple regression model with CASI, surgical duration to predict pain scores at 6-12 months after surgery. It remained a significant predictor for CPSP after adjusting for the other covariates (p<0.0001). As such, data show that Inclusion of PRS improved predictive accuracy for CPSP to 92% and explained 50% variability. Figure 12. Example 5. Building a Predictive Model and Internal Validation The predictive model for CSPS was calculated for the Scoliosis Cohort (described in Examples 1 and 2) based on significant psychophysical predictors (identified in Example 2) and PRS (identified in Example 4) using a bootstrap method. Briefly, bootstrapping was used to build and internally validate the prediction model. At each iteration, a random bootstrap sample the same size as the original sample was drawn with replacement from the original sample. Stepwise selection was used to derive a final model for each outcome where the final model will either include only variables with a p <0.05 or be the one minimizing marginal Akaike Information Criterion (AIC) and/or Bayes Information Criterion (BIC). Both criteria penalized larger models, although BIC model heavily, therefore balance fit with model size and avoid overfitting. The procedure was repeated for at least 100 times. To assess predictive performance, the models developed in each bootstrap sample were evaluated in both the bootstrap sample (apparent performance) and the original sample (test performance) using area under the receiver operating characteristic (ROC) curve (or the c statistics) and calibration slope for binary outcomes, and mean square error (MSE) and marginal, conditional, and fitted R 2 for continuous outcomes. Optimism was estimated to equal average (bootstrap performance – test performance) as internally validated performance as well as the variance based on cross-validation (VEcv). Specifically, with bootstrapping the model parameters (regression coefficients) and model performance (AUC) were evaluated. At each iteration (n=1000), a random bootstrap sample the same size as the original sample was drawn with replacement from the original sample. Logistic models were generated for each bootstrap sample and bootstrapping results were compared with results from the original model. Bootstrapping was performed in R software with the package boot. Mean OR and AUC for the ROC after bootstrapping and 95% CI are given in Table 10. A decision tree was constructed from the top 80 most significant SNPs for CPSP (by chi-square) using C4.5 algorithm (J48 in R package RWeka) and partykit package for visualization. Generation of a decision tree can be used for predicting genetic signatures and to identify most important variant combinations that define risk strata. Here, the algorithm effectively classified subjects (correctly classified 87 % and misclassified 13 % subjects). Figure 13. Decision tree helped identify most informative SNP combinations (and the resulting strata), and to derive simple and easy to interpret logical rules, such as PRKCA rs9914723 = AG AND PRKCA rs62069959=GG THEN Risk=High. Figure 16. The training classification accuracy with the 3 SNPs in the figure (rs9914723, rs62069959 and rs493352) is 74% and can be further increased to 77% by adding two more SNPs (which are rs1150635 and rs1992701). In 5-fold cross validation the top three SNPs (and it especially true of the top rs9914723) are selected with a sufficient frequency to indicate that their selection as most discriminating features (also in combination) is quite robust. The actual confusion matrix for this tree is (low high, low high): 6420, 1433. As can be seen from the tree, just having at least one minor (risk) allele for rs9914723, increases the risk from 37% to about 54%. OTHER EMBODIMENTS All of the features disclosed in this specification may be combined in any combination. Each feature disclosed in this specification may be replaced by an alternative feature serving the same, equivalent, or similar purpose. Thus, unless expressly stated otherwise, each feature disclosed is only an example of a generic series of equivalent or similar features. From the above description, one skilled in the art can easily ascertain the essential characteristics of the present invention, and without departing from the spirit and scope thereof, can make various changes and modifications of the invention to adapt it to various usages and conditions. Thus, other embodiments are also within the claims. EQUIVALENTS While several inventive embodiments have been described and illustrated herein, those of ordinary skill in the art will readily envision a variety of other means and/or structures for performing the function and/or obtaining the results and/or one or more of the advantages described herein, and each of such variations and/or modifications is deemed to be within the scope of the inventive embodiments described herein. More generally, those skilled in the art will readily appreciate that all parameters, dimensions, materials, and configurations described herein are meant to be exemplary and that the actual parameters, dimensions, materials, and/or configurations will depend upon the specific application or applications for which the inventive teachings is/are used. Those skilled in the art will recognize, or be able to ascertain using no more than routine experimentation, many equivalents to the specific inventive embodiments described herein. It is, therefore, to be understood that the foregoing embodiments are presented by way of example only and that, within the scope of the appended claims and equivalents thereto, inventive embodiments may be practiced otherwise than as specifically described and claimed. Inventive embodiments of the present disclosure are directed to each individual feature, system, article, material, kit, and/or method described herein. In addition, any combination of two or more such features, systems, articles, materials, kits, and/or methods, if such features, systems, articles, materials, kits, and/or methods are not mutually inconsistent, is included within the inventive scope of the present disclosure. All definitions, as defined and used herein, should be understood to control over dictionary definitions, definitions in documents incorporated by reference, and/or ordinary meanings of the defined terms. All references, patents and patent applications disclosed herein are incorporated by reference with respect to the subject matter for which each is cited, which in some cases may encompass the entirety of the document. The indefinite articles “a” and “an,” as used herein in the specification and in the claims, unless clearly indicated to the contrary, should be understood to mean “at least one.” The phrase “and/or,” as used herein in the specification and in the claims, should be understood to mean “either or both” of the elements so conjoined, i.e., elements that are conjunctively present in some cases and disjunctively present in other cases. Multiple elements listed with “and/or” should be construed in the same fashion, i.e., “one or more” of the elements so conjoined. Other elements may optionally be present other than the elements specifically identified by the “and/or” clause, whether related or unrelated to those elements specifically identified. Thus, as a non-limiting example, a reference to “A and/or B”, when used in conjunction with open-ended language such as “comprising” can refer, in one embodiment, to A only (optionally including elements other than B); in another embodiment, to B only (optionally including elements other than A); in yet another embodiment, to both A and B (optionally including other elements); etc. As used herein in the specification and in the claims, “or” should be understood to have the same meaning as “and/or” as defined above. For example, when separating items in a list, “or” or “and/or” shall be interpreted as being inclusive, i.e., the inclusion of at least one, but also including more than one, of a number or list of elements, and, optionally, additional unlisted items. Only terms clearly indicated to the contrary, such as “only one of” or “exactly one of,” or, when used in the claims, “consisting of,” will refer to the inclusion of exactly one element of a number or list of elements. In general, the term “or” as used herein shall only be interpreted as indicating exclusive alternatives (i.e. “one or the other but not both”) when preceded by terms of exclusivity, such as “either,” “one of,” “only one of,” or “exactly one of.” “Consisting essentially of,” when used in the claims, shall have its ordinary meaning as used in the field of patent law. As used herein in the specification and in the claims, the phrase “at least one,” in reference to a list of one or more elements, should be understood to mean at least one element selected from any one or more of the elements in the list of elements, but not necessarily including at least one of each and every element specifically listed within the list of elements and not excluding any combinations of elements in the list of elements. This definition also allows that elements may optionally be present other than the elements specifically identified within the list of elements to which the phrase “at least one” refers, whether related or unrelated to those elements specifically identified. Thus, as a non-limiting example, “at least one of A and B” (or, equivalently, “at least one of A or B,” or, equivalently “at least one of A and/or B”) can refer, in one embodiment, to at least one, optionally including more than one, A, with no B present (and optionally including elements other than B); in another embodiment, to at least one, optionally including more than one, B, with no A present (and optionally including elements other than A); in yet another embodiment, to at least one, optionally including more than one, A, and at least one, optionally including more than one, B (and optionally including other elements); etc. It should also be understood that, unless clearly indicated to the contrary, in any methods claimed herein that include more than one step or act, the order of the steps or acts of the method is not necessarily limited to the order in which the steps or acts of the method are recited.
Next Patent: WIND TURBINE GENERATOR FOR LOW TO MODERATE WIND SPEEDS