KM metastatic vs primary query

[corrie painter]

Hello,

I am just starting to look at expression data using the KM feature. However, before I look at my genes of interest, I wanted to get a sense for how the system worked, so I generated a KM using metastatic and primary as the left most clinical feature. The KM that comes out of that shows patients with metastatic disease having better outcomes. Can anyone help me make sense of this or teach me how to set up my query so that I get the correct KM output?

Thank you,

Corrie Painter

[Jing Zhu]

Hi Corrie,

You probably looking at TCGA melanoma, since it has a large number of mets.  There are a lot of patients in the mets group are censored (still alive) and there are only 6 events. In this case, I think you need to download the data, import into R to get a p value, the difference maybe be significant. 

However, we observed the survival difference, in some (not all) TCGA cancer cohorts, between primary tumor vs solid normals. We don't really know why that's the case that the patient contributed solid normals has worse survival than patient contributed tumors. 

Jing

[corrie painter]

Hi Jing,

Thank you for the reply, I'll try your idea.

Corrie