Reducing liver metastases of colon cancer in the context of extensive and minor surgeries through β-adrenoceptors blockade and COX2 inhibition

Abstract

Liver metastases are a major cause of colorectal cancer death, and the perioperative period is believed to critically affect the metastatic process. Here we tested whether blocking excess release of catecholamines and prostaglandins during surgical procedures of different extent can reduce experimental liver metastasis of the syngeneic CT26 colon cancer in female and male BALB/c mice. Animals were either treated with the beta-blocker, propranolol, the COX-2 inhibitor, etodolac, both drugs, or vehicle. The role of NK cells in controlling CT26 hepatic metastasis and in mediating the effect of the drugs was assessed by in vivo depletion or stimulation of NK cells, using anti-asialo GM1 or CpG-C, respectively. Surgical extent was manipulated by adding laparotomy to small incision, extending surgical duration, and enabling hypothermia. The results indicated that combined administration of propranolol and etodolac, but neither drug alone, significantly improved host resistance to metastasis. These beneficial effects occurred in both minor and extensive surgeries, in both sexes, and in two tumor inoculation approaches. NK cell-mediated anti-CT26 activity is involved in mediating the beneficial effects of the drugs. Specifically, CpG-C treatment, known to profoundly activate mice marginating-hepatic NK cytotoxicity, reduced CT26 hepatic metastases; and NK-depletion increased metastases and prevented the beneficial effects of the drugs. Overall, given prevalent perioperative psychological and physiological stress responses in patients, and ample prostaglandin release by colorectal tumors and injured tissue, propranolol and etodolac could be tested clinically in laparoscopic and open colorectal surgeries, attempting to reduce patients' metastatic disease.

Keywords: COX-2 inhibition; CT26 colon-cancer; Colorectal cancer; Etodolac; Liver metastasis; Peri-operative period; Propranolol; β-Adrenergic blocker.

Fig. 1

(A&B) The beneficial effects of propranolol, etodolac, and their combined administration on the number of surface hepatic metastases and liver weight: A total of 57 mice (32 males and 25 females) were injected with propranolol (n = 13), etodolac (n = 16), both drugs (n = 14), or vehicles (n = 14), and 30 min later underwent one of two surgical procedures (small incision with or without laparotomy, approximately half from each drug treatment) for the injection of CT26 tumor cells into the spleen. Twenty-one days later, surface hepatic metastases were counted and livers were weighted. (A) Only the combined drug treatment, but not each drug alone, significantly reduced the number of surface-hepatic metastases compared to no drug treatment (vehicle), as indicated by an *, in both surgical procedures (PLSD p < 0.05 in all comparisons). (B) Liver weight showed the exact same pattern of results, and livers without tumors weigh ~1 g. (C) Comparison of corticosterone levels between the two surgical procedures: 14 females and 10 males were subjected to either a small incision or to laparotomy and small incision, and 6 h later blood corticosterone levels were assessed. Significantly lower levels of corticosterone were evident in mice undergoing the minor surgical procedure compared to mice undergoing the more extensive surgical procedure. Data are presented as mean ± SEM.

Fig. 2

Females present markedly fewer metastases than males, and drug administration is effective in both sexes: A total of 47 males and 41 females were injected with propranolol and etodolac (n = 42), or with vehicles (n = 46), and 30 min later underwent the minor incision surgical procedure for the injection of CT26 tumor cells into the spleen. Twenty-one days later, mice were sacrificed with an overdose of isoflurane, livers were removed and weighed, and surface hepatic metastases were counted. Drug treatment significantly reduced the number of surface-hepatic metastases compared to control (vehicle) levels in both sex categories (p < 0.05 for both), and females presented markedly fewer metastases than males, as indicated by * (p < 0.05). Data are presented as mean ± SEM.

Fig. 3

Using the portal vein inoculation approach – Drugs treatment significantly reduced the number of surface-hepatic metastases, and females presented fewer metastases than males: CT26 tumor cells were injected through the hepatic portal vein to male (n = 34) or female (n = 28) mice without removing the spleen. 30 min before tumor injection, mice were injected with propranolol and etodolac, or with vehicles. Twenty-one days later, mice were sacrificed, livers were removed and surface hepatic metastases were enumerated. Drug treatment significantly reduced the number of surface-hepatic metastases compared to control (vehicle) levels, and females presented significantly fewer metastases than males (ANOVA main effects). * indicate a significant effect for drug separately in each group. Data are presented as mean ± SEM.

Fig. 4

NK cells are significant in controlling CT26 hepatic metastasis: To verify the capacity of anti-asialo GM1 to deplete NK cells, a separate study compared the number of circulating and splenic NK cells in mice treated with anti-asialo GM1 or vehicle control (rabbit serum or PBS). (A) Anti-asialo GM1 significantly eliminated NK cells. (B) A total of 27 female mice were injected with either anti-asialo GM1 (n = 10), CpG-C (n = 7) (known to activate NK cells), or vehicle (n = 10). Twenty-four hours later, all mice were inoculated through the spleen with CT26 tumor cells. Twenty-one days later, all mice were sacrificed, their livers were removed and weighed, and surface-hepatic metastases were counted. CpG-C reduced the number of metastases from an average of 10 in the vehicle treated group to 0.1 (0 in all mice except for one animal that exhibit one metastasis) (p = 0.044). Anti-asialo GM1 caused an 8-fold significant increase in the number of metastases. Similar effects were evident in liver weight (not shown). Data are presented as mean ± SEM.

Fig. 5

NK cells are significantly involved in the beneficial effects of the combined drug treatment: (A) A total of 44 male mice were injected with either anti-asialo GM1 (n = 20), or vehicle (n = 24). Twenty-four hours later, each group was further sub-divided to receive either propranolol and etodolac, or vehicles. Thirty minutes later, all mice were inoculated through the spleen with CT26 tumor cells. Twenty-one days later, all mice were sacrificed, their livers were removed and weighed, and surface-hepatic metastases were counted. (B) The same design and procedure were used (a total of 33 male mice), with the exception that during tumor injection procedure, axillary blood was withdrawn through a heparinized capillary (10–20 ml) for assessment of circulating NK cell number. (A&B) A significant main effect for depletion was evident in both studies, which indicates the effect of NK-depletion across drug treatment. Drug treatment significantly reduced the number of CT26 metastases in NK-intact animals (t-test), as indicated by *, but no effects for drug treatment were evident in NK-depleted animals. Data are presented as mean ± SEM.