Create the indicator variables, the raw data does not include these.
You should calculate a difference in the gscores between the consecutive rows; this is how we know if the Gleason score has increased since the last biopsy, which is one of the indicators for disease progression.
In addition, we need an indicator variable for pcore (number of positive cores) being greater than three, and an indicator variable for the ratio of pcore and tcore (total number of cores) being greater than 0.5.
Based on the biopsy file, can you create three indicator variables for the three criteria for disease progression used in the paper? These were (1) increased gleason score compared to previous biopsy (in either of gscore1 or gscore2), (2) more than three positive cores (pcore), or (3) more than 50% of positive cores (the ratio of pcore to tcore). If any of these indicators receive value one, this means disease progression.
After this, to create the time-to-event outcome for the Kaplan-Meier analysis, for each patient, we should identify (1) the time of the first (baseline) biopsy, (2) time of the first positive biopsy (if it exists), and (3) for those patients without positive biopsy, time of the last biopsy.