Figure 1.
Methodological approach of this study.
The two step strategy of analysis applied in this study.
Table 1.
Percentages of loss, disomy, gain by type of tumor, probe analyzed and test applied.
Figure 2.
Statistical analysis by a Poisson model.
Estimated count of signals observed in each probe (with 95% confidence interval) obtained in each type of tumor, accounting for clustering. The reported p-values refer to the comparison between FFPE and FIN methods. Panel A = LGNI (9 pts); panel B = HGNI (3 pts); panel C = HGIN (9 pts).
Figure 3.
CNA collection evidenced by array-CGH.
CNAs of 10 TCCs samples: 6 infiltrating tumors (IN-TCCs: 70CR09, 81CR09, 04CR10, 09CR10, 10CR10, 26CR10) on the left, and 4 non- infiltrating tumors (NI-TCCs: 28CR09, 75CR09, 80CR09, 82CR09) on the right. Each dot/bar corresponds to one sample. Losses are evidenced in green while gains in red.
Table 2.
The most frequent CNAs evidenced by array-CGH in this study and comparison with data from literature.
Figure 4.
Examples of concordant and non-concordant results of Urovysion test.
Comparison between results on two selected tumoral areas of the same section of FFPE: (A): the two most concordant tumors (070CR09 and 081CR09); (B): the two most discordant tumors (009CR10 and 026CR10).
Figure 5.
Urovysion test applied to: (A): FIN sample 032CR07 (HG NI); (B): FFPE sample 080CR09 (LG NI). (C) FISH with PPARγ probe on 028CR09 (HGNI). Urovysion versus array-CGH data: example of concordant data (D), (sample 080CR09); and non-concordant data (E), (sample 004CR10).
Table 3.
Comparison between results of Urovysion FISH (in two selected tumoral areas of the same FFPE section) and array-CGH.