Techniques for resolving discordant HER2 status in pathology images for diagnosing breast cancer

1 . A computer-implemented pathway analysis method, the computer-implemented method comprising:

obtaining, via an electronic network using a pre-processing application layer of a pathway analysis computing system, human epidermal growth factor receptor 2 (HER2) status for a specimen from a labeled immunohistochemistry (IHC) image corresponding to a sample of the subject;

obtaining, via the electronic network using the pre-processing application layer, HER2 status for a specimen from a labeled fluorescence in-situ hybridization (FISH) image corresponding to the sample;

identifying, via one or more processors using a pipeline application layer of the pathway analysis computing system, discordant HER2 status result between the HER2 status labeled in the IHC image and the HER2 status labeled in the FISH image; and

in response to the identification of the discordant HER2 status:

I) obtaining, via the electronic network using the pipeline application layer, gene expression data corresponding to the sample;

II) processing, via the one or more processors using the pipeline application layer, the gene expression data using a molecular subtype model to generate a HER2 status for the specimen,

wherein the molecular subtype model is accessible by the pipeline application layer and trained using bulk gene expression data from multiple patient specimens each associated with gene expression data and a positive or negative HER2 status; and

III) generating, via the one or more processors using the pipeline application layer, a comprehensive HER2 status by classifying the processed gene expression data using the molecular subtype model;

receiving, via a graphical user interface of a digital display using a molecular subtype processing and report generation application layer of the pathway analysis computing system, one or more user-settable preferences configuring a report template; and

storing, via the one or more processors using the molecular subtype processing and report generation application layer, the user-settable preferences.

2 . The computer-implemented pathway analysis method of claim 1 , further comprising:

generating, via the one or more processors using the molecular subtype processing and report generation application layer, a HER2 discordance status report including an indication of the molecular subtype model used to identify the comprehensive HER2 status,

wherein generating the HER2 discordance status report including the indication of the molecular subtype model used to identify the comprehensive HER2 status is based on the report template configured by the one or more user-settable preferences.

3 . The computer-implemented pathway analysis method of claim 2 ,

wherein the molecular subtype model comprises a linear gene expression model or a pathway gene expression model.

4 . The computer-implemented pathway analysis method of claim 3 ,

wherein the molecular subtype model comprises the pathway gene expression model, and

wherein generating, via the one or more processors using the molecular subtype processing and report generation application layer, the HER2 discordance status report including the indication of the molecular subtype model used to identify the comprehensive HER2 status includes generating a listing of pathways identified by the pathway gene expression model.

5 . The computer-implemented pathway analysis method of claim 2 ,

wherein the molecular subtype model includes a multiple gene linear regression gene expression model, and

wherein generating, via the one or more processors using the molecular subtype processing and report generation application layer, the HER2 discordance status report including the indication of the molecular subtype model used to identify the comprehensive Her2 status includes generating a listing of genes identified by the multiple gene linear regression gene expression model.

6 . The computer-implemented pathway analysis method of claim 1 , wherein generating the HER2 discordance status report includes a molecular subtype determined from the gene expression data and/or biologic pathways determined from the gene expression data.

7 . The computer-implemented pathway analysis method of claim 1 , wherein the molecular subtype is a HR+ subtype, a HR+/HER2+ subtype, a HR−/HER2+ subtype, a HER2− subtype, ER+, ER−, PR+, PR−, or a triple negative subtype.

8 . The computer-implemented pathway analysis method of claim 1 , further comprising:

adjusting a therapeutic treatment protocol based on patterns in gene expression data.

9 . The computer-implemented pathway analysis method of claim 1 , wherein processing, via the one or more processors using the pipeline application layer, the gene expression data using the molecular subtype model to generate the HER2 status for the specimen comprises:

receiving for each of a plurality of biological pathways in the gene expression data, a respective pathway score;

preparing a summary score for the plurality of biological pathways, based upon the respective pathway score for each biological pathway;

comparing the summary score to one or more enrichment scores each associated with a pre-determined molecular subtype; and

determining a molecular subtype of the gene expression data as corresponding to the HER2 status, based on the comparison of the summary score and the one or more enrichment scores.

10 . The computer-implemented pathway analysis method of claim 1 , wherein the molecular subtype model trained on the bulk gene expression data is a machine learning model.

11 . The computer-implemented pathway analysis method of claim 1 , wherein the sample comprises a first slice portion and a second slice portion each taken from a biopsy block, wherein the labeled IHC image is of the first slice portion and wherein the labeled FISH image is from the second slice portion.

12 . The computer-implemented pathway analysis method of claim 1 , wherein obtaining the gene expression data corresponding to the sample comprises:

obtaining the gene expression data from a second sample obtained from the subject.

13 . A pathway analysis computing system, comprising:

a digital display;

one or more processors;

an electronic network;

a pre-processing application layer including computer-executable instructions configured to be executed by the one or more processors;

a pipeline application layer including computer-executable instructions configured to be executed by the one or more processors;

a molecular subtype processing and report generation application layer including computer-executable instructions configured to be executed by the one or more processors; and

a molecular subtype model,

accessible by the pipeline application layer and trained using bulk gene expression data from multiple patient specimens each associated with gene expression data and a positive or negative HER2 status,

wherein the molecular subtype model is configured to analyze gene expression data of a specimen and generate a corresponding HER2 status;

wherein the computer-executable instructions of the pre-processing application layer are configured to, when executed by the one or more processors:

obtain, via the electronic network, human epidermal growth factor receptor 2 (HER2) status for a specimen from a labeled immunohistochemistry (IHC) image corresponding to a sample of a subject; and

obtain, via the electronic network, HER2 status for a specimen from a labeled fluorescence in-situ hybridization (FISH) image corresponding to the sample of the subject;

wherein the computer-executable instructions of the pipeline application layer are configured to, when executed by the one or more processors:

identify, via the one or more processors, discordant HER2 status result between the HER2 status labeled in the IHC image and the HER2 status labeled in the FISH image; and

in response to the identification of discordant HER2 status,

 obtain, via the electronic network, gene expression data corresponding to the sample;

 process, via the one or more processors, the gene expression data using the molecular subtype model to generate a HER2 status for the specimen; and

 generate, via the one or more processors, a comprehensive HER2 status by classifying the processed gene expression data using the molecular subtype model; and

wherein the molecular subtype processing and report generation application layer is configured to:

receive, via a graphical user interface of the digital display, one or more user-settable preferences configuring a report template; and

store, via the one or more processors, the user-settable preferences.

14 . The pathway analysis computing system of claim 13 ,

wherein the molecular subtype processing and report generation application layer is further configured to:

generate, via the one or more processors, a HER2 discordance status report indicating a molecular subtype determined from the gene expression data and/or indicating biologic pathways in gene expression data,

wherein the HER2 discordance status report is based on the report template configured by the one or more user-settable preferences.

15 . The pathway analysis computing system of claim 13 ,

wherein the molecular subtype processing and report generation application layer is further configured to:

generate, via the one or more processors, a HER2 discordance status report including an indication of the molecular subtype model used to identify the discordant HER2,

wherein the HER2 discordance status report is based on the report template configured by the one or more user-settable preferences.

16 . The pathway analysis computing system of claim 13 ,

wherein the molecular subtype model comprises a linear gene expression model or a pathway gene expression model.

17 . The pathway analysis computing system of claim 16 , wherein the molecular subtype model comprises the pathway gene expression model,

wherein the molecular subtype processing and report generation application layer is further configured to:

generate a HER2 discordance status report including a listing of pathways identified by the pathway gene expression model,

wherein the HER2 discordance status report is based on the report template configured by the one or more user-settable preferences.

18 . The pathway analysis computing system of claim 13 , wherein the molecular subtype model comprises a multiple gene linear regression gene expression model,

wherein the molecular subtype processing and report generation application layer is further configured to:

generate a HER2 discordance status report including a listing of genes identified by the multiple gene linear regression gene expression model,

wherein the HER2 discordance status report is based on the report template configured by the one or more user-settable preferences.

19 . The pathway analysis computing system of claim 13 ,

wherein the molecular subtype processing and report generation application layer is further configured to:

receive, for each of a plurality of biological pathways in the gene expression data, a respective pathway score;

prepare a summary score for the plurality of biological pathways, based upon the respective pathway score for each biological pathway;

compare the summary score to one or more enrichment scores each associated with a pre-determined molecular subtype; and

determine a molecular subtype of the gene expression data as corresponding to the HER2 status, based on the comparison of the summary score and the one or more enrichment scores.

20 . The pathway analysis computing system of claim 13 , wherein the molecular subtype model trained on the bulk gene expression data is a machine learning model.