Fig. 4

MRI and BLI images of one representative animal for each group. a Animals from the ganciclovir (GCV)-treated group displayed significantly smaller tumors at the end of GCV treatment compared to phosphate-buffered saline (PBS)-treated and sham-operated (SHAM) animals (*p < 0.05). Some substantial variability was noticed, however, between individual animals so subgroup analysis was performed (green bar: mOct4- BM-MAPC steretactical injection; red bar: treatment phase). b Representation of subgroup tumor development over time for sham-operated, PBS- and GCV-treated animals. No statistically significant differences could be found on day 14 and 16 between the different groups (green bar: mOct4- BM-MAPC steretactical injection; red bar: treatment phase). c Statistical analysis of the tumor volumes at the end of treatment (day 30) showed a statistically significant difference in tumor size between GCV responding (n = 11) animals and sham-operated (n = 5)/PBS-treated animals (n = 8). Furthermore, some GCV-treated animals (n = 7) did not respond to therapy whereas some PBS-treated (n = 4) animals also showed a reduced tumor size at the end of treatment. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. d mOct4⁻ BM-MAPC viability measurements showed a reduced cell viability for the GCV but not the PBS treated group. *p < 0.05, ****p < 0.0001. e MR images of animals of different groups show a comparable tumor growth prior to mOct4⁻ BM-MAPC injection on T2-weighted coronal MR images (upper row for each group) whereas there is little hypointense contrast visible on three-dimensional T2* MR images. In sham-operated animals this mild hypointense contrast was maintained as tumors grew larger although some increase in the hypointense voxel volume, due to the development of necrosis and bleedings, was observed. mOct4⁻ BM-MAPC-injected animals (PBS and GCV) could be detected by three-dimensional T2* MRI (lower row for each group) on day 1 after injection. For animals which developed tumors, the hypointense voxels got more dispersed over time as tumors grew whereas mice responding to GCV treatment had little tumors where labeled mOct4⁻ BM-MAPCs could still be detected at the end of GCV treatment. f BLI measurements showed a persistence of the BLI signal in PBS-treated animals indicating survival of the mOct4⁻ BM-MAPCs whereas GCV-treated animals showed a decreased viability after GCV treatment. Sham operated animals were used as negative controls. g Histological overview images of brain sections from the respective animals of each treatment group stained with trichrome staining. d Day