DIfference in survival times present in TCGA and cBIoPortal

Question

[sandhya.p]

Hi All,

I am working on correlating clinical information with mutation data. I have downloaded clinical data related to Prostate Adenocarcinoma (PRAD) from TCGA firehose. But, survival times seem to be different from that of cBioPortal. For example, for TCGA-G9-6336, days_to_last_followup in TCGA is 1669, which comes to 54.87 months. But in cBioPortal, Overall Survival Time is given as 67.9 months. Any ideas which one is correct?

Thanks

[Devon Ryan]

I'm not sure those numbers disagree. The overall survival time should refer to either the survival time since admittance to the study or since first diagnosis. The "day_to_last_followup" should just be the time since the last follow-up appointment with the study. I would expect that there will be cases where the two values are equal, but there's no reason to expect them to be.

That's my quick reading of things at least.

Answer 1

[sandhya.p]

I think, now I understand the scenario. I was taking days_to_last_followup from "PRAD.merged_only_clinical_clin_format.txt" file located in "gdac.broadinstitute.org_PRAD.Merge_Clinical.Level_1.2014101700.0.0.tar.gz". But, when I looked into "PRAD.clin.merged.picked.txt" file present in "gdac.broadinstitute.org_PRAD.Clinical_Pick_Tier1.Level_4.2014101700.0.0.tar.gz", it is same as that of given in cBioPortal. I do not know why the information given in these two files are different.

Answer 2

[cying]

May be the survival time given by the cBioPortal is the nearest follow-up. Here(https://tcga-data.nci.nih.gov/tcgafiles/ftp_auth/distro_ftpusers/anonymous/tumor/) is the link of updated clinical information for all types of tumors. The follow up data is updated with the time.Anyway, you can count on the latest follow-up.