Table 3.
Overview of univariate and multivariate results of intergroup comparisons
UVA P |
MVA P |
Supplemental material | |
---|---|---|---|
GVL vs noGVL | Table 5A | ||
Figure 2B CD8+ 1/D | .0051 | .0309 | Table 5B |
Figure 3B Δ no. of clones top 25% | .0172 | .0326 | Table 5C |
Figure 3B Δ no. of clones top 50% | .0200 | .0380 | Table 5D |
Figure 3B Δ no. of clones top 75% | .0179 | .0480 | Table 5E |
Figure 3B Δ no. of clones whole repertoire | .0287 | .5600 | Table 5F |
Figure 4B Δ CD8+ TRB repertoire occupation by top 20 clones | .0051 | .0069 | Table 5G |
Figure 4B Δ CD8+ TRB repertoire occupation by top 21-50 clones | .1700 | .6273 | Table 5H |
Figure 4B Δ CD8+ TRB repertoire occupation by top 51-N clones | .0248 | .0186 | Table 5I |
Figure 5B Δ CD8+ TRB repertoire occupation by top 20 overlapping clones | .0442 | .0306 | Table 5J |
Figure 5B Δ CD8+ TRB repertoire occupation by top 21-N overlapping clones | .8400 | .6098 | Table 5K |
Figure 5B Δ CD8+ TRB repertoire occupation by all overlapping clones | .0018 | .0121 | Table 5L |
Figure 5F Δ overlapping DLI CD8+ clones | .0128 | .0024 | Table 5M |
Figure 5G Δ new DLI CD8+ clones | .7780 | .0415 | Table 5N |
No relapse vs relapse | Table 6A-B | ||
Figure 6B CD8+ 1/D | .0142 | .0066 | Table 6C |
Figure 6D Δ no. of clones top 25% | .0107 | .0034 | Table 6D |
Figure 6D Δ no. of clones top 50% | .0242 | .0133 | Table 6E |
Figure 6D Δ no. of clones top 75% | .4363 | .1815 | Table 6F |
Figure 6D Δ no. of clones whole repertoire | .8633 | .4480 | Table 6G |
Figure 6F Δ CD8+ TRB repertoire occupation by top 20 clones | .0244 | .0073 | Table 6H |
Figure 6F Δ CD8+ TRB repertoire occupation by top 21-50 clones | 1.0000 | .4050 | Table 6I |
Figure 6F Δ CD8+ TRB repertoire occupation by top 51-N clones | .2581 | .1220 | Table 6J |
Figure 6H Δ CD8+ TRB repertoire occupation by top 20 overlapping clones | .0503 | .0308 | Table 6K |
Figure 6H Δ CD8+ TRB repertoire occupation by top 21-N overlapping clones | .5613 | .7643 | Table 6L |
Figure 6H Δ CD8+ TRB repertoire occupation by all overlapping clones | .0770 | .0660 | Table 6M |
CD8+ expansion vs no expansion | Table 7A-B, D | ||
Figure 7A CIR | .0040 | .0080 | Table 7C |
Figure 7B Relapse-free survival | .0150 | .0210 | Table 7E |
Given are the UVA and MVA results for the intergroup comparisons of Figures 2-7. UVA for Figures 2-6 were done with Mann-Whitney test (two-tailed, exact). UVA for Figure 7A was done with Gray’s test and for Figure 7B with univariate Cox regression analysis. MVA for Figures 2-6 employed analysis of variance testing, either with all variables (categorized to 2 levels, when applicable) for the GVL vs noGVL comparison or using only variables significantly correlating with dependent variables at P < .25 for the no relapse vs relapse comparison. MVA of Figure 7 used multivariate Cox regression models. Variables found to be significant with P < .25 in respective UVA were used in MVA. Backward elimination procedure was used to reduce the number of variables until remaining variables were fitting a model at P < .05. Details of MVA are shown in the supplemental material.