(A) Flow chart for producing conditioned medium from various adipocytes. (B) The viability of adipocyte-conditioned medium–treated GBC cells in response to gemcitabine (GEM). SNU-308 and RCB-1130 cells were treated with serum-free medium, adipocyte CM, and GEM for 24 hours in each sample. Cell viability was assessed by the MTT assay (n = 3; ***P < 0.001, 1-way ANOVA followed by Tukey’s multiple comparisons). (C) Inhibition of GEM-induced apoptosis by adipocyte CM treatment. SNU-308 and RCB-1130 cells were plated in 100 nM GEM or 100 nM GEM in combination with adipocyte CM and treated for 24 hours in microtiter plates. Cell apoptosis was analyzed by the TUNEL assay. Blue spots represent cell nuclei by DAPI staining, and green spots represent apoptotic bodies by TUNEL staining (n = 3; **P < 0.01, ***P < 0.001; 1-way ANOVA followed by Tukey’s multiple comparisons). Scale bar: 100 μm. (D) Four-week-old male NOD/SCID mice were fed a normal diet for 12 weeks. Then, mice were coinjected with gallbladder cancer RCB-1130 cells (4 × 10⁶/mice) and human primary adipocytes (1 × 10⁶/mice). Once the tumors attained a size of approximately 100 mm³, animals were randomized to receive gemcitabine treatment or were left untreated (control). Mice were treated twice per week by intraperitoneal injection of gemcitabine (400 mg/kg) diluted in 150 μL PBS, and tumor volume was measured for 4 weeks using a caliper. Tumor volume was calculated according to the formula (length × width²)/2 (n = 4; ***P < 0.001, 1-way ANOVA followed by Tukey’s multiple comparisons).