Systems and methods for treating, diagnosing and predicting the response to therapy of breast cancer

1. A method for administering a course of treatment to a subject with a cancer type comprising the steps of:

A. measuring, using an immunofluorescence imaging device having a connection to a computing device configured to execute code in a processor, at least a plurality of protein expression levels of the subject by contacting a formalin fixed paraffin-embedded tissue sample of the subject to antibodies of Her2, Her2-ECD, cytokeratin and p95HER2;

B. capturing an image of the tissue sample with the immunofluorescence imaging device and analyzing the image with an image analysis device to determine plurality of protein expression levels present in the sample;

C. automatically storing the plurality of protein expression levels measured by the image analysis device in a subject dataset in a memory of the computing device wherein the subject dataset includes clinical and biographic data relating to the subject;

D. calculating, using the computing device, a protein expression level of p95HER2 for the subject according to a function (1);

i. where function (1)=p95HER2=(Her2+pHer2)−Her2−ECD;

E. providing a machine learning application with a population dataset from a data storage device, wherein the machine learning application is configured to generate an effective treatment model using a linear discriminant analysis on the population dataset, wherein the population dataset includes data relating to each member of the population obtained at least two different points in time, wherein each member is diagnosed with the same cancer type and similar clinical features as the subject and the data obtained at the at least two different points in time relating to each member includes:

i. at least one data value corresponding to treatment resistance status,

ii. at least one data value corresponding to treatment status;

iii. at least one data value corresponding to the health outcome,

iv. and the protein expression levels for each individual as obtained by:

1. contacting a formalin fixed paraffin-embedded tissue sample of each member of the population with antibodies to Her2, Her2-ECD, cytokeratin, p95HER2 and p95HER2 for each member of the population dataset, and

2. deriving the protein expression level of p95HER2 of each member of the population dataset according to function (1);

F. evaluating the subject dataset with the effective treatment model, wherein the evaluation yields a value related to a probability that the subject will have a resistance to at least one treatment for the cancer type;

G. generating, using a treatment module configured as instructions for the computer, a output data object that includes at least information indicating a course of treatment for the subject that does not include any treatment where the probability that the subject will have resistance is above a given threshold.

2. The computer implemented method of claim 1 , wherein the antibodies of step (A) are selected from AE1/AE3 (cytokeratin), TAB250 (Her2-ECD), c-erb B2 (Her2) and an antibody to phosphorylated Her 2 (pHer2).

3. The computer implemented method of claim 1 , wherein the discriminant analysis of step (E) is an Eigengene-based Linear Discriminant analysis implemented to determine the relationship between protein expression values found in the population dataset and treatment resistance status.

4. The computer implemented method of claim 1 , wherein the subject and population datasets include morphometric measurement data obtained from an image analysis appliance.

5. The method of claim 4 , wherein the morphometric measurement data are obtained by the steps of:

1) accessing with an image access module configured as code executing in a processor, a first set of tissue images associated with the subject of the subject dataset of step (C);

2) accessing with an image access module configured as code executing in a processor, a second set of tissue images associated with the each member of the population dataset of step (E); and

3) obtaining morphometric measurements for the first and second sets of tissue images using a morphometric analysis module configured as code executing in the processor.

6. The method of claim 1 , wherein the course of treatment of step (G) comprises administering lapatinib.

7. The method of claim 1 , wherein the course of treatment of step (G) comprises administering immunotherapy.

8. The method of claim 7 , wherein the treatment comprises at least one of trastuzumab and bevacizumab.

9. The method of claim 1 , wherein the generating a course of treatment of step (G) further comprises:

assigning the subject, based on the evaluation of the population dataset, to at least one of a plurality of survivability categories, and assigning the subject, based on the survivability category, to one of a plurality of treatment categories, wherein assigning the subject to at least one of a plurality of treatment categories comprises determining, based on the amount of pHER2 in the tissue sample from the subject, at least one of (a) trastuzumab sensitivity and (b) trastuzumab resistance.

10. The method of claim 1 , wherein the course of treatment generated in step (G) is based on the probability of at least one of (a) trastuzumab sensitivity and (b) trastuzumab resistance is above a given threshold value.

11. The method of claim 1 , further comprising generating a plurality of treatment options wherein each of the plurality of treatment options correspond to a treatment where the probability that the subject will have resistance to that treatment is below a given threshold.

12. The method of claim 1 , wherein the population dataset of step (E) includes data obtained from one or more individuals who have been exposed to the particular course of treatment.

13. The method of claim 1 , wherein the population dataset of step (E) includes data obtained from one or more individuals who have not been exposed to the particular course of treatment.

14. The method of claim 1 , wherein the population dataset of step (E) comprises at least one of:

(a) one or more individuals who are responsive to chemotherapy in breast cancer;

(b) one or more individuals who have a higher disease free survival rate from breast cancer;

(c) one or more individuals who have a higher overall survival rate from breast cancer;

(d) one or more subjects who are not at risk for developing a recurrence of breast cancer; and

(e) one or more subjects who are at a low risk for developing a recurrence of breast cancer.

15. A method for administering a course of treatment to a subject with a cancer type comprising the steps of:

A. measuring, using an immunoassay, at least a plurality of protein expression levels of the subject by contacting a formalin fixed paraffin-embedded tissue sample of the subject to antibodies of Her2, Her2-ECD, cytokeratin and pHER2;

B. storing the plurality of protein expression levels as data values in a subject dataset stored in a data storage device;

C. calculating, using the computer configured to access the subject dataset, a protein expression level of p95Her2 for the subject according to a function (1);

i. where function (1)=p95Her2=(Her2+pHer2)−Her2−ECD;

D. providing a machine learning application configured, to generate an effective treatment model using a linear discriminant analysis on a population dataset, wherein the population dataset includes at least data relating to individuals diagnosed with the same cancer type as the subject, at least one treatment resistance status for each individual, and the protein expression levels for each individual as obtained by:

i. contacting a formalin fixed paraffin-embedded tissue sample of each member of the population dataset with antibodies to Her2, Her2-ECD, cytokeratin, p95Her2 and pHER2, and

ii. deriving the protein expression level of p95Her2 of each member of the population dataset according to function (1);

E. evaluating the subject dataset with the effective treatment model, wherein the evaluation yields a value related to a probability that the subject will have a resistance to at least one treatment for the cancer type; wherein the evaluation includes identifying at least one hyperplane that separates at least a first class of disease resistance from at least a second class of disease resistance using the population dataset and locating the subject at a distance from the at least one hyperplane, the probability that the subject will have a resistance to at least one treatment corresponding to the distance of the subject from the at least one hyperplane; and

F. generating, using a treatment module configured as instructions for the computer, a course of treatment for the subject that does not include any treatment where the probability that the subject will have resistance is above a given threshold.

16. The computer-implemented method of claim 15 further comprising the steps of:

G. updating the subject dataset of step (B) and the population dataset of step (D) over a plurality of time periods;

H. generating a revised effective treatment model based on the comparison of the updated subject dataset of step (B) and population dataset of step (D) and

I. generating a revised course of treatment based on the comparison of the updated first and second datasets.

17. The computer-implemented method of claim 16 , wherein the revised treatment effective treatment model comprises generating at least a revised course of treatment regimen for administering to the subject a given amount of a medication for a period of time.

18. The method of claim 16 , wherein updating the subject database over time comprises determining, based on one or more changes, that a cancer is progressive, wherein the one or more changes comprise one or more changes in an amount of cytokeratin and Her2 as determined by contacting antibodies to cytokeratin, Her2-ECD, Her2 and phosphorylated Her 2 with a plurality of samples from the subject over time relative to the population dataset of step (D).