Figure 2

Kaplan-Meier analysis of pre-treatment BM VDJ% and EFS. The VDJ%, dichotomized by the median, shows a significant correlation with EFS (A) - those who fall below the median (light line) have longer EFSs than those who fall above the median (dark line). Statistical determination of the optimal cut point for dichotomizing into two groups gave a value nearly identical to the median (18.6%). There were 39 relapses out of 69 patient samples below the median and 43 of 70 above the median. When patient samples were stratified to exclude those receiving an ASCT (the group of non-transplant patients), the significance remains (B). When the remaining patients who did receive an ASCT were analyzed, the significant correlation was lost (C). D represents the Kaplan-Meier analyses for three significantly different EFS categories: lightest line=all ASCT treated patients, medium grey line= non-transplant patients below the VDJ% median, and the dark line= non-transplant patients above the VDJ% median, with respectively 20 relapses out of 50 samples, 30 out of 45 and 32 out of 44. E summarizes the logrank p-value, hazard ratio and median survivals for the above plots.