Method of predicting response to immunotherapy

A method of predicting a likelihood that a cancer patient will respond positively to immunotherapy, the method comprising:

• (A) scoring a sample comprising tumor tissue taken from the cancer patient comprising:
• using the sample comprising tumor tissue taken from the cancer patient, determining an interaction score representative of a spatial proximity between at least one pair of cells, a first member of the at least one pair of cells expressing PD-1 and a second member of the at least one pair of cells expressing PD-L1;
• and recording the interaction score;
• (B) deriving a value for % biomarker positivity (PBP) for all HLA-DR positive cells in a field of view of the sample expressing HLA-DR+IDO-1+, and recording the value for PBP;
• and
• (C) comparing the interaction score to a first threshold value and comparing the value for PBP to a second threshold value;
• wherein if (1) either the interaction score is greater than or equal to the first threshold value or the value for PBP is greater than or equal to the second threshold value, or (2) the interaction score is greater than or equal to the first threshold value and the value for PBP is greater than or equal to the second threshold value, then the cancer patient is likely to respond positively to immunotherapy.

A method of predicting a likelihood that a cancer patient will respond positively to immunotherapy, the method comprising:

• (A) determining a score representative of a spatial proximity between at least one pair of cells selected from among a plurality of cells present in a predetermined number of fields of view available from a sample comprising tumor tissue, which sample is taken from the cancer patient, the method comprising:
• selecting a predetermined number of fields of view available from the sample comprising tumor tissue taken from the cancer patient, which is stained with a plurality of fluorescence tags, which selection is biased toward selecting fields of view that contain a greater number of cells that express PD-L1 relative to other fields of view;
• for each of the selected fields of view, dilating fluorescence signals attributable to PD-L1 by a margin sufficient to encompass proximally located cells expressing PD-1;
• dividing a first total area for all cells from each of the selected fields of view, which express PD-1 and are encompassed within the dilated fluorescence signals attributable to the cells expressing PD-L1, with a normalization factor, and multiplying the resulting quotient by a predetermined factor to arrive at a spatial proximity score;
• and
• recording the spatial proximity score;
• (B) deriving a value for % biomarker positivity (PBP) for all HLA-DR+ cells present in a field of view expressing HLA-DR+IDO-1+, comprising:
• generating an image of first fluorescence signals representative of nuclei of all cells present in a field of view, and dilating the first fluorescence signals to a diameter of that of an entire cell to construct a first mask of all cells present in the field of view;
• constructing a second mask of second fluorescence signals representative of all areas present in the field of view, which express HLA-DR+;
• constructing a third mask of third fluorescence signals representative of all areas present in the field of view, which express IDO-1+;
• combining said first and second masks in a manner that provides a fourth mask comprising fluorescence signals representative of all cells in the field of view, which also express HLA-DR+;
• combining said first and third masks in a manner that provides a fifth mask comprising fluorescence signals representative of all cells in the field of view, which also express IDO-1+;
• combining said fourth and fifth masks in a manner that provides a sixth mask comprising fluorescence signals representative of all cells in the field of view, which express HLA-DR+ and IDO-1+;
• deriving the value for PBP for all cells present in the field of view expressing HLA-DR+IDO-1+ by dividing the total area of the sixth mask by the total area of the fourth mask;
• and
• recording the value for PBP;
• and
• (C) comparing the spatial proximity score to a first threshold value and comparing the value for PBP to a second threshold value;
• wherein if (1) either the spatial proximity score is greater than or equal to the first threshold value or the value for PBP is greater than or equal to the second threshold value, or (2) the spatial proximity score is greater than or equal to the first threshold value and the value for PBP is greater than or equal to the second threshold value, then the cancer patient is likely to respond positively to immunotherapy.

A method of predicting a likelihood that a cancer patient will respond positively to immunotherapy, the method comprising:

• (A) determining a score representative of a spatial proximity between at least one pair of cells selected from among a plurality of cells present in a predetermined number of fields of view available from a sample comprising tumor tissue, which sample is taken from the cancer patient, the method comprising:
• selecting a predetermined number of fields of view available from the sample comprising tumor tissue taken from the cancer patient, which is stained with a plurality of fluorescence tags, which selection is biased toward selecting fields of view that contain a greater number of cells that express PD-1 relative to other fields of view;
• for each of the selected fields of view, dilating fluorescence signals attributable to PD-1 by a margin sufficient to encompass proximally located cells expressing PD-L1;
• dividing a first total area for all cells from each of the selected fields of view, which express PD-L1 and are encompassed within the dilated fluorescence signals attributable to the cells expressing PD-1, with a normalization factor, and multiplying the resulting quotient by a predetermined factor to arrive at a spatial proximity score;
• and
• recording the spatial proximity score;
• (B) deriving a value for % biomarker positivity (PBP) for all HLA-DR+ cells present in a field of view expressing HLA-DR+IDO-1+, comprising:
• generating an image of first fluorescence signals representative of nuclei of all cells present in a field of view, and dilating the first fluorescence signals to a diameter of that of an entire cell to construct a first mask of all cells present in the field of view;
• constructing a second mask of second fluorescence signals representative of all areas present in the field of view, which express HLA-DR+;
• constructing a third mask of third fluorescence signals representative of all areas present in the field of view, which express IDO-1+;
• combining said first and second masks in a manner that provides a fourth mask comprising fluorescence signals representative of all cells in the field of view, which also express HLA-DR+;
• combining said first and third masks in a manner that provides a fifth mask comprising fluorescence signals representative of all cells in the field of view, which also express IDO-1+;
• combining said fourth and fifth masks in a manner that provides a sixth mask comprising fluorescence signals representative of all cells in the field of view, which express HLA-DR+ and IDO-1+;
• deriving the value for PBP for all cells present in the field of view expressing HLA-DR+IDO-1+ by dividing the total area of the sixth mask by the total area of the fourth mask;
• and
• recording the value for PBP;
• and
• (C) comparing the spatial proximity score to a first threshold value and comparing the value for PBP to a second threshold value;
• wherein if (1) either the spatial proximity score is greater than or equal to the first threshold value or the value for PBP is greater than or equal to the second threshold value, or (2) the spatial proximity score is greater than or equal to the first threshold value and the value for PBP is greater than or equal to the second threshold value, then the cancer patient is likely to respond positively to immunotherapy.