Methods of treating diseases
The present invention relates to computer-implemented methods and system for analyzing a biomarker which cycles in a subject. In some other aspects, the present invention relates to analyzing a biomarker which at least initially increases or decreases in amount in a subject following a treatment for a disease. In further aspects, the present invention relates to computer-implemented methods and systems for determining a preferred time to administer a therapy to treat a disease in a subject. The present invention also relates to computer program product to implement the methods. Further, the present invention relates to methods of determining the timing of treating a disease in a subject in which the immune system is cycling.
1. A method of treating a disease in a subject in which the immune system is cycling, the method comprising;
administering a therapy to the subject at a preferred time in the future, wherein the preferred time in the future has been determined using a computer-implemented method for analysing a biomarker which cycles in the subject, the method comprising estimating a periodicity of the cycling of the biomarker based on measurements of the biomarker in samples from the subject by obtaining a best-fit curve of the measurements in accordance with a model of the cycling of the biomarker and determining from the estimated periodicity of the cycling a preferred time in the future to administer the therapy to treat the disease in the subject by projecting the obtained best-fit curve into the future, and wherein the disease is characterized by the production of regulator T cells and cycling immune system markers.
2. A method of treating a disease characterized by production of regulator cells and cycling immune system markers in a subject, the method comprising:
i) using a computer-implemented method for analysing a biomarker which cycles in the subject to estimate a periodicity of the cycling of the biomarker based on measurements of the biomarker in samples from the subject by obtaining a best-fit curve of the measurements in accordance with a model of the cycling of the biomarker;
ii) determining from the estimated periodicity of the cycling a preferred time in the future to administer the therapy to treat the disease in the subject by projecting the obtained best-fit curve into the future, and
ii) administering the therapy at the preferred time.
3. The method of claim 1 , wherein the best-fit curve is a fit to a harmonic model of the cycling of the biomarker.
4. The method of claim 3 , wherein obtaining the best-fit curve further comprises fitting the harmonic model to the measurements multiple times using different initial conditions for each fit, and generating a confidence measure by comparing a similarity in a fit result produced by each fitting.
5. The method of claim 4 , wherein the harmonic model has a heavy tail to allow for noisy nature of the measurements.
6. The method of claim 1 , wherein obtaining the best curve fit comprises fitting a plurality of different regression models to the measurements.
7. The method of claim 1 , wherein obtaining the best-fit curve further comprises imposing a box constraint on at least some parameters of the model to guide optimisation to biologically realistic regions.
8. The method of claim 1 which further comprises determining a variance-stabilizing transformation of the measurements prior to estimating the periodicity based on the transformed measurements.
9. The method of claim 1 , wherein determining the preferred time to administer the therapy comprises comparing the measurements of the subject with a database of measurements of the biomarker obtained from other subjects.
10. The method of claim 1 , wherein the measurements are for two or more different biomarkers which are cycling in the subject, and determining the preferred time to administer the therapy is based on the two or more sets of measurements.
11. The method of claim 1 , wherein the biomarker is an acute phase inflammatory marker.
12. The method of claim 1 , wherein the therapy is selected from the group consisting of an anti-cancer drug, a vaccine, radiation, dsRNA and an antibody which inhibit the production and/or activity of regulator T cells.
13. The method of claim 12 , wherein the anti-cancer drug is an anti-proliferative drug.
14. The method of claim 1 , wherein the disease is cancer.
15. The method of claim 2 , wherein the best-fit curve is a fit to a harmonic model of the cycling of the biomarker.
16. The method of claim 2 , wherein obtaining the best curve fit comprises fitting a plurality of different regression models to the measurements.
17. The method of claim 2 , wherein obtaining the best-fit curve further comprises imposing a box constraint on at least some parameters of the model to guide optimisation to biologically realistic regions.
18. The method of claim 2 which further comprises determining a variance-stabilizing transformation of the measurements prior to estimating the periodicity based on the transformed measurements.
19. The method of claim 2 , wherein determining the preferred time to administer the therapy comprises comparing the measurements of the subject with a database of measurements of the biomarker obtained from other subjects.
20. The method of claim 2 , wherein the measurements are for two or more different biomarkers which are cycling in the subject, and determining the preferred time to administer the therapy is based on the two or more sets of measurements.