Cancer risk stratification based on histopathological tissue slide analysis

Abstract

The subject disclosure presents systems and computer-implemented methods for providing reliable risk stratification for early-stage cancer patients by predicting a recurrence risk of the patient and to categorize the patient into a high or low risk group. A series of slides depicting serial sections of cancerous tissue are automatically analyzed by a digital pathology system, a score for the sections is calculated, and a Cox proportional hazards regression model is used to stratify the patient into a low or high risk group. The Cox proportional hazards regression model may be used to determine a whole-slide scoring algorithm based on training data comprising survival data for a plurality of patients and their respective tissue sections. The coefficients may differ based on different types of image analysis operations applied to either whole-tumor regions or specified regions within a slide.

1 .- 8 . (canceled) 
     
     
         9 . A computer-implemented method for early-stage cancer prognosis, the method comprising:
 mapping one or more tumor regions annotated on an image of a first tissue slide to each of a plurality of other images of histopathological tissue slides, wherein the plurality of other images of tissue slides correspond to serial sections from a tissue block;   scoring each of the plurality of other images of histopathological tissue slides based on an expression of one or more tumor markers or immune or stromal cells on each tissue slide; and   computing a risk stratification score to combine the slide scores from an included subset of individual marker tissue slides using a set of coefficients;   wherein a patient may be stratified into a low or high recurrence risk group based on a cut-off point of the risk stratification score.   
     
     
         10 . The method of  claim 9 , further comprising selecting a field of view on the image of the first tissue slide to compute slide scores. 
     
     
         11 . The method of  claim 10 , wherein the field of view comprises a whole tumor. 
     
     
         12 . The method of  claim 10 , wherein the field of view comprises a hot spot selection. 
     
     
         13 . The method of  claim 10 , wherein slide scoring further comprises analyzing the image and scoring each slide based on the field of view. 
     
     
         14 . The method of  claim 10 , wherein the set of coefficient values is selected based on the field of view. 
     
     
         15 . The method of  claim 14 , wherein the set of coefficient values are derived from fitting a Cox proportional hazard model to the slide scores computed for a considered set of tissue slides to the actual survival data for a plurality of patients included in a training cohort. 
     
     
         16 . The method of  claim 15 , wherein the training cohort comprises the actual survival data and comprises tissue marker slides having specific staining and scanning protocols, annotation generation, image and data analysis workflows corresponding to a plurality of patients. 
     
     
         17 . The method of  claim 15 , wherein the cut-off point is determined based on a statistical fit. 
     
     
         18 .- 28 . (canceled) 
     
     
         29 . A system for early-stage cancer prognosis, the system comprising:
 a processor;   and a memory coupled to the processor, the memory to store computer-readable instructions that, when executed by the processor, cause the processor to perform operations comprising:
 mapping one or more tumor regions annotated on an image of a first tissue slide to each of a plurality of other images of histopathological tissue slides, wherein the plurality of other images of tissue slides correspond to serial sections from a tissue block; 
 scoring each of the plurality of other images of histopathological tissue slides based on an expression of one or more tumor markers or immune or stromal cells on each tissue slide; and 
 computing a risk stratification score to combine the slide scores from an included subset of individual marker tissue slides using a set of coefficients; 
   wherein a patient may be stratified into a low or high recurrence risk group based on a cut-off point of the risk stratification score.   
     
     
         30 . The system of  claim 29 , wherein the operations further comprise selecting a field of view on the image of the first tissue slide to compute slide scores. 
     
     
         31 . The system of  claim 30 , wherein the field of view comprises a whole tumor. 
     
     
         32 . The system of  claim 30 , wherein the field of view comprises a hot spot selection. 
     
     
         33 . The system of  claim 30 , wherein slide scoring further comprises analyzing the image and scoring each slide based on the field of view. 
     
     
         34 . The system of  claim 30 , wherein the set, of coefficient, values is selected based on the field of view. 
     
     
         35 . The system of  claim 34 , wherein the set of coefficient values are derived from fitting a Cox proportional hazard model to the slide scores computed for a considered set of tissue slides to the actual survival data for a plurality of patients included in a training cohort. 
     
     
         36 . The system of  claim 35 , wherein the training cohort comprises the actual survival data and comprises tissue marker slides having specific staining and scanning protocols, annotation generation, image and data analysis workflows corresponding to a plurality of patients. 
     
     
         37 . The system of  claim 35 , wherein the cut-off point is determined based on a statistical fit. 
     
     
         38 . A tangible non-transitory computer-readable medium configured to store computer-readable code that when executed by a processor causes the processor to perform operations comprising:
 mapping one or more tumor regions annotated on an image of a first tissue slide to each of a plurality of other images of histopathological tissue slides, wherein the plurality of other images of tissue slides correspond to serial sections from a tissue block;   scoring each of the plurality of other images of histopathological tissue slides based on an expression of one or more tumor markers or immune or stromal cells on each tissue slide; and   computing a risk stratification score to combine the slide scores from an included subset of individual marker tissue slides using a set of coefficients;   wherein a patient may be stratified into a low or high recurrence risk group based on a cut-off point of the risk stratification score.   
     
     
         39 . The tangible non-transitory computer-readable medium of  claim 38 , wherein the operations further comprise selecting a field of view on the image of the first tissue slide to compute slide scores. 
     
     
         40 . The tangible non-transitory computer-readable medium of  claim 39 , wherein the field of view comprises a whole tumor. 
     
     
         41 . The tangible non-transitory computer-readable medium of  claim 39 , wherein the field of view comprises a hot spot selection. 
     
     
         42 . The tangible non-transitory computer-readable medium of  claim 39 , wherein slide scoring further comprises analyzing the image and scoring each slide based on the field of view. 
     
     
         43 . The tangible non-transitory computer-readable medium of  claim 39 , wherein the set of coefficient values is selected, based on the field of view. 
     
     
         44 . The tangible non-transitory computer-readable medium of  claim 43 , wherein the set of coefficient values are derived from fitting a Cox proportional hazard model to the slide scores computed for a considered set of tissue slides to the actual survival data for a plurality of patients included in a training cohort. 
     
     
         45 . The tangible non-transitory computer-readable medium of  claim 44 , wherein the training cohort comprises the actual survival data and comprises tissue marker slides having specific staining and scanning protocols, annotation generation, image and data analysis workflows corresponding to a plurality of patients. 
     
     
         46 . The tangible non-transitory computer-readable medium of  claim 44 , wherein the cut-off point is determined based on a statistical fit.