Methods for predicting an individual's clinical treatment outcome from sampling a group of patient's biological profiles

• US 20040193019A1
• Filed: 03/24/2003
• Published: 09/30/2004
• Est. Priority Date: 03/24/2003
• Status: Active Grant

• Current Assignee: Nien Wei
• Original Assignee: Nien Wei
• Inventors: Wei, Nien
• Granted Patent: US 9,342,657 B2
• Time in Patent Office: Days
• US Class Current: 600/300
• CPC Class Codes: A61K 31/337 having four-membered rings, e.g. taxol
• CPC Class Codes: A61K 31/4745 condensed with ring systems having nitrogen as a ring hetero atom, e.g. phenantrolines yohimbine derivatives, vinblastine A61K31/475; ergoline derivatives A61K31/48
• CPC Class Codes: A61K 31/513 having oxo groups directly attached to the heterocyclic ring, e.g. cytosine
• CPC Class Codes: A61K 31/519 ortho- or peri-condensed with heterocyclic rings
• CPC Class Codes: A61K 31/555 containing heavy metals, e.g. hemin, hematin, melarsoprol
• CPC Class Codes: A61K 31/664 Amides of phosphorus acids
• CPC Class Codes: A61K 33/243 Platinum; Compounds thereof
• CPC Class Codes: G16B 20/00 ICT specially adapted for functional genomics or proteomics, e.g. genotype-phenotype associations
• CPC Class Codes: G16B 20/20 Allele or variant detection, e.g. single nucleotide polymorphism [SNP] detection
• CPC Class Codes: G16B 25/00 ICT specially adapted for hybridisation; ICT specially adapted for gene or protein expression
• CPC Class Codes: G16B 25/10 Gene or protein expression profiling; Expression-ratio estimation or normalisation
• CPC Class Codes: G16B 40/00 ICT specially adapted for biostatistics; ICT specially adapted for bioinformatics-related machine learning or data mining, e.g. knowledge discovery or pattern finding
• CPC Class Codes: G16B 40/20 Supervised data analysis
• CPC Class Codes: G16B 5/00 ICT specially adapted for modelling or simulations in systems biology, e.g. gene-regulatory networks, protein interaction networks or metabolic networks
• CPC Class Codes: G16H 50/50 for simulation or modelling of medical disorders
• CPC Class Codes: G16H 50/70 for mining of medical data, e.g. analysing previous cases of other patients

Abstract

Methods, systems, and computer program products that predict an individual'"'"'s treatment outcome from a sampling of a group of patients'"'"' biological profiles. Biological profile information is received from patients who had a medical condition and who received a treatment. Treatment outcome information regarding the patients who had the medical condition and who received the treatment is also received. A discriminant analysis-based pattern recognition process is then performed on the biological profile information and the treatment outcome information, thereby generating a model that correlates between the biological profile information and the treatment outcome information. The model can be used for, among other things, predicting treatment outcome for the new patient for the treatment.

Claims

1. A method of predicting a medical treatment outcome for a medical condition, comprising:

(1) receiving biological profile information from patients who had the medical condition and who received a treatment;

(2) receiving treatment outcome information regarding the patients who had the medical condition and who received the treatment; and

(3) performing a discriminant analysis-based pattern recognition process on the biological profile information and the treatment outcome information, thereby generating a model that correlates between the biological profile information and the treatment outcome information.

2. The method according to claim 1, further comprising:

(4) providing biological profile information from a new patient to the model, whereby the model outputs a predicted treatment outcome for the new patient for the treatment.

3. The method according to claim 2, wherein the model outputs the predicted treatment outcome as a positive or negative indication.

4. The method according to claim 2, wherein the model outputs the predicted treatment outcome on a scale.

5. The method according to claim 1, further comprising repeating steps (1) through (3) for one or more additional treatments for the medical condition.

6. The method according to claim 5, further comprising:

(4) providing biological profile information from a new patient to the models, whereby the models output predicted treatment outcomes for the patient for the treatments.

7. The method according to claim 1, wherein the medical condition is a disease.

8. The method according to claim 7, wherein the medical condition is a human disease.

9. The method according to claim 8, wherein the medical condition is a cancer.

10. The method according to claim 1, wherein the medical condition is a non-disease.

11. The method according to claim 10, wherein the medical condition is a fertility analysis.

12. The method according to claim 1, wherein the biological profile information comprises genetic profile information.

13. The method according to claim 1, wherein the biological profile information comprises DNA profile information.

14. The method according to claim 1, wherein the biological profile information comprises RNA profile information.

15. The method according to claim 1, wherein the biological profile information comprises protein profile information.

16. The method according to claim 1, wherein the biological profile information comprises one or more of the following:

genetic profile information;

DNA profile information;

RNA profile information; and

protein profile information.

17. The method according to claim 1, wherein the biological profile information is embedded on a microarray.

18. The method according to claim 17, wherein the biological profile information is generated from a genomic analysis.

19. The method according to claim 1, wherein the information regarding patients who had the medical condition is obtained from a set of fewer than 100 patients.

20. The method according to claim 1, wherein the information regarding patients who had the medical condition is obtained from a set of fewer than 30 patients.

21. The method according to claim 1, wherein the information regarding patients who had the medical condition is obtained from a set of fewer than 20 patients.

22. The method according to claim 1, wherein the information regarding patients who had the medical condition is obtained from a set of fewer than 10 patients.

23. The method according to claim 1, wherein the information regarding patients who had the medical condition is obtained from a set of greater than 100 patients.

24. The method according to claim 12, wherein the genetic profile information includes genetic information regarding approximately 10,000 genes.

25. The method according to claim 12, wherein the genetic profile information includes genetic information regarding at least 10 genes.

26. The method according to claim 12, wherein the genetic profile information includes genetic information regarding at least 100 genes.

27. The method according to claim 12, wherein the genetic profile information includes genetic information regarding at least 1000 genes.

28. The method according to claim 12, wherein the genetic profile information includes genetic information regarding at least 1 gene.

29. The method according to claim 1, wherein the biological profile information comprises genetic profile information, and wherein step (2) comprises reducing the genetic profile information to a sub-set of genes that account for a substantial portion of diversity.

30. The method according to claim 29, further comprising embedding a representation of the sub-set of genes on a microarray chip as part of the model.

31. The method according to claim 29, wherein step (4) comprises providing genetic profile information from the new patient regarding only the sub-set of genes.

32. The method according to claim 30, wherein step (4) comprises providing genetic profile information from the new patient regarding only the sub-set of genes.

33. The method according to claim 1, wherein the model is implemented in a computer program.

34. The method according to claim 1, wherein the treatment comprises a drug-based treatment.

35. The method according to claim 1, wherein the treatment comprises a non drug-based treatment.

36. The method according to claim 1, wherein the treatment comprises a combination of drug-based treatment and non drug-based treatment.

37. The method according to claim 36, wherein the treatment comprises a combination of chemotherapy treatment and radiation treatment.

38. The method according to claim 1, further comprising using the model to predict a first line treatment for a cancer patient.

39. The method according to claim 36, further comprising using the model to predict a first line treatment for a cancer patient.

40. The method according to claim 1, further comprising using the model to predict efficacy of a single drug treatment.

41. The method according to claim 1, further comprising using the model to predict efficacy of a multiple drug treatment.

42. The method according to claim 1, further comprising using the model to predict efficacy of a first line single drug treatment.

43. The method according to claim 1, further comprising using the model to predict efficacy of a first line multiple drug treatment.

44. The method according to claim 1, further comprising using the model to predict efficacy of a later-line drug treatment.

45. The method according to claim 1, further comprising using the model to predict short term efficacy of the treatment.

46. The method according to claim 1, further comprising using the model to predict long term efficacy of the treatment.

47. The method according to claim 46, further comprising using the model to predict long term drug resistance.

48. The method according to claim 46, further comprising using the model to predict time before relapse.

49. The method according to claim 1, further comprising using the model to predict efficacy of the treatment for another medical condition.

50. The method according to claim 49, wherein the biological profile information and the treatment outcome information pertain to colon cancer patients, further comprising using the model to predict efficacy of the treatment for an ovarian cancer patient.

51. The method according to claim 49, wherein the biological profile information and the treatment outcome information pertain to a first type of tumor, further comprising using the model to predict efficacy of the treatment for a second type of tumor related to the first type of tumor.

52. The method according to claim 51, wherein the first type of tumor is primary tumor and the second type of tumor is a metastatic tumor.

53. The method according to claim 1, wherein the biological profile information comprises microarray expression profile information.

54. The method according to claim 53, wherein the microarray expression profile data is obtained from one or more of the following sources:

nucleic acid expression profiles;

protein expression profiles; and

single nucleotide polymorphisms.

55. The method according to claim 1, further comprising:

(4) using the model for clinical diagnostic, prognostic, and deterministic treatment administration tools for personalized medical treatment.

56. The method according to claim 1, further comprising:

(4) using the model for data mining of markers in human diseases.

57. The method according to claim 1, further comprising:

(4) using the model for identification of markers in human diseases.

58. The method according to claim 1, further comprising:

(4) using the model for analysis of a combination of genomics and clinical data for correlation in human diseases.

59. The method according to claim 1, wherein the treatment outcome information comprises medically measurable features.

60. The method according to claim 1, wherein the treatment outcome information comprises objective features.

61. The method according to claim 60, wherein the treatment outcome information comprises objective short-term features.

62. The method according to claim 60, wherein the treatment outcome information comprises objective long-term features.

63. The method according to claim 1, wherein the treatment outcome information comprises side-effects information.

64. The method according to claim 1, wherein the treatment outcome information comprises subjective features.

65. The method according to claim 64, wherein the treatment outcome information comprises quality-of life features.

66. The method according to claim 64, wherein the treatment outcome information comprises short-term subjective features.

67. The method according to claim 64, wherein the treatment outcome information comprises long-term subjective features.

68. The method according to claim 1, wherein the biological profile information and the treatment outcome information is received from patients who benefitted from a clinical trial of the treatment.

69. The method according to claim 68, further comprising:

(4) providing biological profile information from a new patient to the model, whereby the model outputs a predicted treatment outcome for the new patient for the treatment.

70. The method according to claim 68, further comprising submitting an application for approval of the treatment for use with new patients for whom the model outputs a favorable predicted treatment outcome.

71. The method according to claim 1, wherein step (3) comprises:

(i) dividing the biological profile information and the treatment outcome information into a training sub-set and a testing sub-set;

(ii) performing the discriminant analysis-based pattern recognition process on the training sub-set; and

(iii) testing the resultant model with the testing sub-set.

72. The method according to claim 1, wherein step (3) comprises:

(i) dividing the biological profile information and the treatment outcome information into a training sub-set of biological profile information and treatment outcome information and a testing sub-set biological profile information and treatment outcome information;

(ii) performing a plurality of discriminant analysis-based pattern recognition processes on the training sub-set of biological profile information and treatment outcome information, thereby generating a plurality of models that correlate between the training set of biological profile information and the training set of treatment outcome information;

(iii) testing the plurality of models with the testing sub-set of biological profile information and treatment outcome information;

(iv) selecting the model that most accurately predicts the testing sub-set of treatment outcome information from the testing sub-set of biological profile information.

73. The method according to claim 72, further comprising:

(v) selecting new testing sub-sets of biological profile information and treatment outcome information;

(vi) repeating step(3)(ii) for the new sub-sets of information; and

(vii) averaging the results for each of the plurality of models;

wherein step (3)(iv) comprises selecting the model that most accurately predicts, on average, the testing sub-set of treatment outcome information from the testing sub-set of biological profile information.

74. The method according to claim 1, wherein the biological profile information is obtained from targeted tissue samples.

75. The method according to claim 1, wherein the biological profile information is obtained from non-targeted tissue samples.

76. The method according to claim 2, further comprising selecting a mode of treatment based on the predicted treatment outcome.

77. The method according to claim 2, further comprising avoiding a mode of treatment based on the predicted treatment outcome.

78. The method according to claim 1, wherein step (3) is performed without pathway information.

80. The method according to claim 2, wherein the biological profile information is embedded on a microarray.

81. The method according to claim 80, wherein the model and computer program are embedded on a microarray to output the predication.

79. A computer program product including a computer usable medium having computer program logic stored therein to enable a computer system to generate a model that correlates between biological profile information and treatment outcome information, wherein the computer program logic comprises:

a first function that causes the computer system to receive biological profile information from patients who had the medical condition and who received a treatment;

a first function that causes the computer system to receive treatment outcome information regarding the patients who had the medical condition and who received the treatment; and

a first function that causes the computer system to perform a discriminant analysis-based pattern recognition process on the biological profile information and the treatment outcome information, thereby generating a model that correlates between the biological profile information and the treatment outcome information.