Umibe5690,
From what I have learned about the classifications they are all subject to a continuum. Just like a patient is considered M+ or M- but there is a continuum of the gradient so a patient may be slightly M+ or slightly M-, each tumor is partly this classification and partly that classification. I am not sure how far along they are regarding how to use these classifications in total, which is why I think there is varying reports of which classification has what tendency. I know there are some proven ones like M+ is more prone to respond to TMZ for known cellular reasoning and also to IO. But, there still appears to be a lot of knowledge to be gained before complete generalizations can be of much value. I also think sampling is of limited value unless the test results show heavily tilted results or unless there is truly strongly representative sampling of each tumor. I am familiar with sampling techniques and the statistics which are employed to help make reasonable deductions. I am not sure how or if such techniques are or can be employed to strengthen the classification deductions for phenotype classifications. Lastly I am pretty sure there are biomarkers which are not yet exploited which will someday assist more toward the knowledge of how to determine which type of treatments will tend to be better for which combinations. There is still a lot to develop here, and, the surface is being scratched is what I am seeing, which makes it difficult to make any sweeping conclusions, just really gross tendencies.
News 
Market Data 
Discover 
