Accelerated metastasis after short-term treatment with a potent inhibitor of tumor angiogenesis

Abstract

Herein we report that the VEGFR/PDGFR kinase inhibitor sunitinib/SU11248 can accelerate metastatic tumor growth and decrease overall survival in mice receiving short-term therapy in various metastasis assays, including after intravenous injection of tumor cells or after removal of primary orthotopically grown tumors. Acceleration of metastasis was also observed in mice receiving sunitinib prior to intravenous implantation of tumor cells, suggesting possible "metastatic conditioning" in multiple organs. Similar findings with additional VEGF receptor tyrosine kinase inhibitors implicate a class-specific effect for such agents. Importantly, these observations of metastatic acceleration were in contrast to the demonstrable antitumor benefits obtained when the same human breast cancer cells, as well as mouse or human melanoma cells, were grown orthotopically as primary tumors and subjected to identical sunitinib treatments.

Figure 1. Accelerated Experimental Metastasis and Decreased Survival after Short-Term Sunitinib Treatment before and after Intravenous Tumor Inoculation

(A) 231/LM2-4^LUC+ cells were injected into the tail vein of SCID mice that received vehicle (group A) or short-term sunitinib treatment daily for 7 days either before (group B) or after tumor inoculation (group C). Quantification of bioluminescence showed accelerated tumor growth in groups B and C compared with controls. A representative experiment is shown. Group A, n = 10; group B, n = 5; group C, n = 10. Data are presented as mean ± SD. (B) Kaplan-Meier survival curve shows significantly decreased median survival of mice in group B (log-rank test, p = 0.0055) and group C (log-rank test, p < 0.0001) compared with group A. Data represent a summary of multiple experiments. Group A, n = 31; group B, n = 11; group C, n = 19. 0.001 < **p < 0.01; ***p < 0.001. (C) Representative images for each group taken 1, 7, and 27 days following tumor implantation, with increased metastasis visible in sunitinib-treated mice. Sunitinib dose and treatment schedule were performed as illustrated in (A).

Figure 2. Short-Term Sunitinib Treatment Increases Spontaneous Metastasis and Decreases Survival after Removal of Primary Human Xenograft Tumors

(A) Orthotopically grown 231/LM2-4^LUC+ tumors were surgically removed, and SCID mice were treated daily with vehicle (group A) or short-term sunitinib therapy (group B). Biweekly quantification of bioluminescence showed accelerated tumor growth and increased spontaneous metastasis in group B compared with group A. Data are presented as mean ± SD. (B) Representative bioluminescence images visualizing tumor cells before and after primary tumor resection (days 5 and 30 after resection). (C) Kaplan-Meier survival curves of the corresponding mice show significantly decreased median survival in group B (log-rank test, p = 0.0024) compared with group A. 0.001 < **p < 0.01. (D) Resected tumors were weighed prior to sorting into groups A and B to ensure equal tumor burden between groups. Sunitinib dose and treatment schedule were performed as illustrated in (A).

Figure 3. Increased Multiorgan Metastasis in Mice after Short-Term Sunitinib Treatment

(A) Following the same experimental design as described in Figure 1A, SCID mice were sacrificed at day 27 to compare increases in overall tumor burden after short-term sunitinib treatment (inset) and corresponding increased bioluminescence in multiple organs (groups A, B, and C). Data are presented as mean ± SEM. Instances of highly divergent bioluminescence values did not permit statistical significance to be reached in all groups. 0.01 < *p < 0.05 by one-way ANOVA. (B) Micrometastases were confirmed by immunostaining for human vimentin in organs of the 231/LM2-4^LUC+ tumor model in (A) (upper panel) or in nu/nu mice receiving vehicle (group A) or short-term sunitinib therapy (group B) prior to intravenous (i.v.) inoculation with human MeWo melanoma cells (lower panel). Tissue sections were scored as positive or negative based on the presence or absence of detectable micrometastases. NT = not tested. (C) Representative examples of micrometastases in spleen, liver, kidney, and brain shown using human-specific vimentin antibodies. (D) Excised lungs from 231/LM2-4^LUC+ and MeWo tumor models were scored visually for surface tumor nodules, with confirmation of macrometastasis by hematoxylin and eosin (H&E) and anti-vimentin immunostaining (representative images shown). For groups A, B, and C in the 231/LM2-4^LUC+ tumor model, n = 10 per group. For groups A and B in the MeWo tumor model, n = 5 per group.

Figure 4. Differentiating Opposing Efficacies of Short-Term and Sustained Sunitinib Treatment in Primary and Metastatic Disease

(A) nu/nu mice bearing orthotopically grown 231/LM2-4^LUC+ tumors received either vehicle (group A) or sustained sunitinib therapy (group B) when tumors reached an average volume of 200 mm³. A third group received short-term sunitinib therapy (group C) starting the day after tumor implantation. Group A reached tumor volume endpoint (1500 mm³) at 27 days, with comparative tumor volume significantly reduced in group B (p = 0.0002 by Student’s t test) and group C (p = 0.0027 by Student’s t test) at the same time point. Mice receiving sustained sunitinib therapy showed greater tumor growth inhibition compared to mice receiving short-term therapy (41 to 62 days to endpoint, respectively). n = 5 for all groups. (B) nu/nu mice injected i.v. with231/LM2-4^LUC+cells were treated daily with vehicle (group A) or received short-term sunitinib therapy for 7 days either before (group B) or after tumor inoculation (group C). Quantification of bioluminescence showed accelerated metastatic tumor growth in groups B and C after short-term sunitinib therapy. Mice in groups D and E received sustained sunitinib therapy starting on day 8, with group D also receiving short-term sunitinib therapy similar to group C. (C) Corresponding Kaplan-Meier survival curves show that median survival was significantly decreased in groups B, C, and D (p=0.0011, p=0.0022, and p=0.0365 by log-rank test, respectively) and was not significantly different in group E (p=0.4485 by log-rank test) compared to control mice in group A. For (B) and (C): group A, n = 7; group B, n = 7; group C, n = 7; group D, n = 5; group E, n = 5. (D) In a syngeneic tumor model, C57BL/6 mice bearing mouse B16 melanoma tumors grown orthotopically showed delayed primary tumor growth after sustained sunitinib therapy compared with control mice (groups A and B, with time to endpoint 17 and 28 days, respectively). Group A, n = 4; group B, n = 4. (E) C57BL/6 mice receiving short-term sunitinib therapy prior to i.v. inoculation with the same mouse B16 melanoma tumor cells showed accelerated experimental metastasis and decreased survival compared to controls (group B; p = 0.0014 by log-rank test). Delayed metastasis and increased survival were observed in mice receiving short-term sunitinib treatment followed by sustained sunitinib treatment (group D; p = 0.0047 by log-rank test). Mice receiving sustained sunitinib therapy 7 days after tumor implantation showed no difference in survival compared to vehicle-treated control mice (group E; p=0.6368 by log-ranktest). Mice receiving short-term sunitinib therapy following tumor implantation exhibited a bimodal response that included either accelerated metastasis and reduced survival or extended survival (group C; p= 0.3391 by log-rank test). Group A, n = 10; group B, n=10; group C, n=9; group D, n=5; group E, n= 5. Sunitinib dose and treatment schedule were performed as illustrated. Data are presented as mean ± SD. 0.01 < *p < 0.05; 0.001 < **p < 0.01; ***p < 0.001.