Figure 2: Host-derived FGF2 is upregulated in the intrathoracic Y-MESO-14 tumours resistant to bevacizumab treatment.

(a) A comparison of mouse mRNA expression of pro-angiogenic factors in the intrathoracic Y-MESO-14 tumours treated with or without bevacizumab (bev). The fold-changes in mRNA expression of the bevacizumab-treated tumours compared with control tumours are shown. *P<0.05 by the Mann–Whitney-U test. The changes in the amounts of mouse (b) FGF2 (n=8 per group) and (c) VEGF (n=6 per group) in the pleural effusion of tumour (Y-MESO-14)-bearing mice treated with or without bevacizumab (bev). *P<0.05 by the Mann–Whitney-U-test. (d) Representative images of FGF2-positive cells (green) in the sections from Y-MESO-14 intrathoracic tumours treated with or without bevacizumab (bev). Scale bar, 100 μm. (e) The number of FGF2-positive cells in the Y-MESO-14 tumours treated with bevacizumab (bev) compared with the control group (n=9 per group). *P<0.01 by the Mann–Whitney-U-test. (f) The survival curves of the tumour (Y-MESO-14)-bearing mice treated with bevacizumab (bev), BGJ-398 (an FGFR inhibitor), or both. Bevacizumab was continuously given from day 7 after tumour cell injection. BGJ-398 was given from days 14–28. (g) Representative images of the CD31-positive vessels in the tumours harvested from each group. Scale bar, 200 μm. (h) Evaluation of microvessel density (MVD, quantified as the total number of CD31⁺ structures/HPF) in the tumours from each group (n=12 per group). *P<0.01 by a one-way ANOVA. (i) Representative images of the CD31-positive vessels in the tumours harvested from mice treated with bevacizumab (bev) in combination with FGF2 Ab or control IgG. Scale bar, 200 μm. (j) Evaluation of MVD in each group (n=12 per group). *P<0.01 by the Mann–Whitney-U-test. All data are shown as the means±s.e.m.