Hello,

I have a question regarding the method to choose for stratifying the patients as high and low expressing groups. As far as I read, it seems that there are two main methods for bifurcating patients as either low or high expressing for a particular gene of interest.

1. Method 1 - Median value method. Get the required data with time, event and expression in a single table. Select a median value of the modified counts. Counts below median value are classified as low and those above as high expressing. Plot KM curve.

2. Method 2 - Percentile method. Get the required data with time, event and expression in a single table. Segregate data as high High, medium and low expressing groups based on percentiles. Plot KM curve.

Can anyone explain which method to use and why? Thanks a lot for your time.

Best regards, Manav Gandhi