Enhancement of Tumor Response to -Radiation by an Inhibitor of Cyclooxygenase-2 Enzyme

Abstract

Prostaglandins are arachidonate metabolites produced in virtually all mammalian tissues and possess diverse biologic capabilities, including vasoconstriction, vasodilatation, stimulation or inhibition of platelet aggregation, and immunomodulation, primarily immunosupression (1–4). They are implicated in the promotion of development and growth of malignant tumors (4–7). They are also involved in the response of tumor and normal tissues to cytotoxic agents such as ionizing radiation (8). Prostaglandin production is mediated by two cyclooxygenase enzymes: cyclooxygenase-1 and cyclooxygenase-2. Cyclooxygenase-1 is constitutively expressed and is ubiqui tous , and cyclooxygenase-2 is induced by diverse inflammatory stimuli (7,9). Nonsteroidal anti-inflammatory drugs (NSAIDs) or agents inhibit cyclooxygenase enzymes and consequently can prevent, inhibit, or abolish the effects of prostaglandins. Increasing evidence shows that NSAIDs can inhibit the development of cancer in both experimental animals and in humans (7), can reduce the size of established tumors (6–8), and can increase the efficacy of cytotoxic anticancer agents (8). Our own investigations have demonstrated that the NSAID indomethacin prolongs tumor growth delay and increases the tumor cure rate in mice after radiotherapy (8,10,11). Commonly used NSAIDs, including indomethacin, inhibit both cyclooxygenase-1 and cyclooxygenase-2. However, treatment with these agents may be limited by toxicity to normal tissue, particularly by ulcerations and bleeding in the gastrointestinal tract ascribed to the inhibition of cyclooxygenase-1. Recently developed selective cyclooxygenase-2 inhibitors exert potent anti-inflammatory activity but cause fewer unwanted side effects (7,9,12,13). These compounds may thus be safer than those NSAIDs that are in common use. A recent report (7) shows that cyclooxygenase-2-specific inhibitors can prevent carcinogenesis in experimental animals, but their efficacy in enhancing in vivo tumor response to radiation has not been established. By use of the mouse sarcoma NFSA, we investigated the potential of 4-[5-(4chlorophenyl)-3-(trifluoromethyl)-1Hpyrazol-l-yl]benzenesulfonamide (SC8236), a selective cyclooxygenase-2 inhibitor (14,15) (supplied by Searle, G. D. & Co., Skokie, IL), to enhance response of tumor to local g-irradiation. All studies reported had institutional approval and all guidelines for appropriate animal treatment were followed. We have reported earlier (6) that the NFSA sarcoma is a nonimmunogenic and prostaglandin-producing tumor that spontaneously developed in C3Hf/Kam mice. This tumor exhibits an increased radioresponse if indomethacin is given prior to tumor irradiation (10,11). In experiments described in this communication, solitary tumors were generated in the right hind legs of mice by the injection of 3 × 10 viable NFSA tumor cells. When tumors were 8 mm in diameter, they were locally irradiated with 25–80 Gy single-dose g-radiation. Treatment with SC-8236 (6 mg/kg body weight, given in the drinking water) was started when tumors were approximately 6 mm in diameter, and the treatment was continued for 10 consecutive days. In some experiments, tumor irradiation was performed 3–8 days after initiation of the treatment with SC-8236. The end points of the treatment were tumor growth delay (days) and TCD50 (tumor control dose 50, defined as the radiation dose yielding local tumor cure in 50% of irradiated mice 120 days after irradiation). Treatment of mice with SC-8236 alone significantly inhibited tumor growth (inset in Fig. 1, A). Tumor diameter doubling time, based on tumor growth from 6 to 12 mm in diameter, was increased from 7.3 days (95% confidence interval [CI] 4 6.4–8.1 days) to 14.8 days (95% CI 4 11.5–18.1 days) (P<.0001). The effect of SC-8236 was evident already within 1 day from the start of the treatment. SC-8236 treatment dramatically increased the effect of tumor irradiation, as shown by both tumor growth delay (Fig. 1, A and B) and tumor cure rate (Fig. 1, C). The growth delay after the combined treatment was more than the sum of growth delays caused by either irradiation alone or SC-8236 alone (Fig. 1, A). Tumors in control mice required 4.6 days (95% CI 4 3.9–5.4 days) to grow from 8 to 12 mm in diameter. Mice treated with SC-8236 required 7.1 days (95% CI 4 5.0–9.2 days) (P 4 .003), mice treated with 30 Gy required 13.6 days (95% CI 4 10.5–16.7 days), and mice treated with both agents required 43.5 days (95% CI 4 30.8–56.2 days) (P 4 .001 compared with radiation-only group). The efficacy of irradiation was enhanced by a factor of 3.64 (95% CI 4 3.42–3.86), determined from the curves in Fig. 1, B, which plot the magnitude of tumor growth delay as a function of radiation dose with or without treatment with SC-8236 (see legend to Fig. 1, A and B). This compound also greatly enhanced the tumor cure rate after irradiation (Fig. 1, C): The TCD50 value was reduced from 69.2 Gy (95% CI 4 65.7– 72.7 Gy) in the radiation-only group to 39.2 Gy (95% CI 4 31.1– 44.6 Gy) in the combination-treatment group. The enhancement factor was 1.77 (95% CI 4 1.51–1.98), obtained by dividing the TCD50 value of the radiation-alone group by the combination-treatment group. The 95% CIs were obtained by use of Fieller’s theorem (16). Because prostaglandins are known to stimulate angiogenesis (17), the possibility that SC-8236 inhibited tumor angiogenesis was investigated. In an intradermal assay for angiogenesis developed in our laboratory (18), mice received intradermally injections of 10