Because overexpression of biomarkers in the later phasis of tumor progression.

Biomarkers have higher role in the survival phase also in begining phasis of the tumor.

I would suggest a few points.

1. Not all genes that are robustly overexpressed in tumors lead to poor prognosis. This is because of multiple known and unknown mechanisms working behind. A simple theory can be these normal properties/ mechanisms of these cells are still working i.e regulated cell growth and even the tumor suppressor may get overexpressed to counterbalance the "bad/ oncogenic forces. Seems good?

2. Practically, when we are analyzing bulk tumor RNA expression data (having a mix of both tumor and some normal tissues), even the stress response genes which are overexpressed in surrounding normal cells will show up in your differential analysis.

Therefore if you still want to go further, look for cell-specific expression/ laser capture microdissection data.

Nevertheless, the biomarker feature is still there, but if you want to publish it in a combination with other data, it will be a little difficult to tackle reviewer's questions

Dear Mohit,

Thank you for your answer. I checked the link you provided.

Thanks,

Amal.