Anti-Breast Cancer Potential of SS5020, a Novel Benzopyran Antiestrogen

Abstract

Treatment with tamoxifen (TAM) increases the risk of developing endometrial cancer in women. The carcinogenic effect is thought to involve initiation and/or promotion resulting from DNA damage induced by TAM as well as its estrogenic action. To minimize this serious side-effect while increasing the anti-breast cancer potential, a new benzopyran antiestrogen, 2E-3-{4-[(7-hydroxy-2-oxo-3-phenyl-2H-chromen-4-yl)-methyl]-phenyl}-acrylic acid (SS5020), was synthesized. Unlike TAM, SS5020 exhibits no genotoxic activity to damage DNA. Furthermore, SS5020 does not present significant uterotrophic potential in rats; in contrast, the structurally related compounds, TAM, toremifene, raloxifene (RAL) and SP500263 all have uterotrophic activity. At the human equivalent molar dose of TAM (0.33 or 1.0 mg/kg), SS5020 had much stronger antitumor potential than those same antiestrogens against 7,12-dimethylbenz(a)anthracene-induced mammary carcinoma in rats. The growth of human MCF-7 breast cancer xenograft implanted into athymic nude mice was also effectively suppressed by SS5020. SS5020, lacking genotoxic and estrogenic actions, could be a safer and stronger antiestrogen alternative to TAM and RAL for breast cancer therapy and prevention.

Introduction

Tamoxifen (TAM; the structure in Fig. 1a) has been widely used since 1973 as an adjuvant therapy for early-stage breast cancer with positive estrogen receptors (ER) (1) and since 1998 as a prophylactic agent for women at high risk of developing this disease (2). However, long-term administration of TAM has been associated with several adverse effects, including endometrial cancer in women (2-4). The carcinogenic effect may be caused through initiation and/or promotion due to DNA damage induced by TAM as well as the drug’s estrogenic action (reviewed by 5 and 6). Several groups including ours established that α-hydroxylation of TAM and its subsequent O-sulfonation are essential for DNA-adduct formation (5, 6). TAM-induced DNA adducts were detected in rodent liver (7, 8) and in the endometrium of women treated with TAM (9, 10). Because TAM-DNA adducts are highly mutagenic (11) and not rapidly repaired (12), the DNA adducts likely contribute to the initiation of endometrial cancer; in fact, K-ras mutations were detected frequently in the endometria of women treated with TAM (13). TAM is a partial ER agonist in uterine tissue (14); such an estrogenic effect may also contribute to promoting endometrial cancer (15). This drug has been listed as a human carcinogen by the International Agency for Research on Cancer (16).

Fig. 1.

Synthesis of SS5020. (a) Structures of SS5020 and its related antiestrogens. (b) Synthetic method for SS5020.

Some other antiestrogens are fully or partially used for early-stage breast cancer therapy. Toremifene (TOR; Fig. 1a), a chlorinated TAM derivative, was approved in 1987 by the Food and Drug Administration (FDA) for breast cancer therapy. Although the metabolic fate of TOR is similar to that of TAM, TOR does not promote DNA adducts (8, 17, 18) or hepatocarcinoma in rats (8, 17). In fact, no K-ras mutations were observed in the endometria of patients receiving TOR (13). On the contrary, a review paper (19) in a personal communication stated that a clinical trial comparing TAM and TOR, after a mean follow-up time of > 12 years, did not show significant difference in the incidence of endometrial cancers in women taking these drugs. The estrogenic activity of TOR is similar to that of TAM and the clinical efficacy of TOR for breast cancer patients is also similar (20); therefore, this drug is not frequently used in the United States. Raloxifene (RAL; Fig. 1a) has been used since 1998 for treating osteoporosis, and it was approved in 2007 by the FDA as a chemopreventive agent for postmenopausal women at high risk for invasive breast cancer (21). With this drug, an increased incidence of endometrial cancer was not observed in women (22). However, RAL retains a weak proliferative effect on the uterus in postmenopausal women (23). Moreover, venous thrombo-embolic events, liver dysfunction, hot flashes, leg cramps and peripheral edema were still observed as adverse events in patients treated with RAL (24, 25). Therefore, a new safer alternative is required to diminish the side-effects and to increase clinical efficacy.

Several antiestrogen compounds have been applied in clinical studies for treatment of breast cancer (reviewed by 26-28). However, idoxifene, levormeloxifene (29), and arzoxifene (30) were dropped from clinical testing because of their undesirable effects on the uterus and/or no significant clinical benefit beyond that of TAM. Information regarding any clinical evaluations currently being undertaken for breast cancer therapy with other antiestrogens was not found. Development of new, safer antiestrogen alternatives having significant clinical benefits superior to TAM and RAL is urgently needed for breast cancer therapy and prevention.

To avoid the genotoxic action induced by TAM, the compound’s ethyl moiety has to be modified to prevent its α-hydroxylation and subsequent O-sulfonation that could produce an intermediate capable of reacting with DNA. Moreover, a novel carboxylic side chain present in GW5638 (Fig. 1a) has shown to have powerful antiestrogenic activity, expecting to minimize undesirable effects on the uterus (31). Applying the ideas described above, we have designed and synthesized a new triphenylethylene antiestrogen SS1020 reported recently (32) and a new benzopyran antiestrogen, 2E-3-{4-[(7-hydroxy-2-oxo-3-phenyl-2H-chromen-4-yl)-methyl]-phenyl}-acrylic acid (SS5020). In the present studies, the genotoxic, estrogenic, and anti-breast cancer potentials of SS5020 were examined and compared with those of antiestrogens (TAM, RAL, and SP500263) now being used in the clinic or undergoing clinical trials. Our results indicate that SS5020 lacks estrogenic and genotoxic actions and has higher antitumor activity than TAM, RAL and SP500263, making SS5020 a safer alternative for breast cancer therapy and prevention.

Materials and Methods

Chemicals

TAM, 17β-estradiol (E₂), and 7,12-dimethylbenz(a)anthracene (DMBA) were purchased from Sigma (St. Louis, MO). E₂ (0.72 mg) pellet was obtained from Innovative Research of America (Sarasota, FL). TOR and RAL hydrochloride were obtained from LKT Laboratories, Inc. (St. Paul, MN). SP500263 was prepared following the established procedure (33).

Synthesis of SS5020

SS5020 was prepared as described in Fig. 1b. The chromen-2-one intermediate was synthesized by using a protocol similar to that for preparing SP500263 (33). Briefly, commercially available 3-methoxy phenol 1 was acylated under Fries reaction conditions using ZnCl₂ and POCl₃ to produce the desired benzophenone 3. Formation of the triflate 4, followed by the Heck reaction (34) and deprotection gave SS5020 7 in good overall yield (~75%). NMR did not detect any significant impurity in SS5020 within the detection limit (<3% impurity). By HPLC/UV analysis with XTerra MS C18 column chromatography, the purity of SS5020 was determined to be >98%.

2-(4-Hydroxybenzylacetone)-5-methoxyphenol (3): POCl₃ (50 ml, 0.8 mol) was added to a mixture of 3-methoxy phenol (1) (25 g, 0.20 mol), 4-hydroxyphenylacetic acid (2) (35.5 g, 0.23 mol) and ZnCl₂ (87 g, 0.64 mol), and the mixture was stirred at 65 °C for 2 hrs, poured into ice water (1 L) and stirred until the ice melted. The clear supernatant was decanted and the residue was rinsed with water (500 ml) and partitioned between EtOAc and water. The organic layer was washed with brine, dried over sodium sulfate, filtered and concentrated. The crude was purified by chromatography (silica gel, 20 % EtOAc /hexane), followed by crystallization to give 3 as a white solid (17 g, 33 %). ¹H-NMR (CD₃OD): 3.77 (s, 3H); 4.08 (s, 2H); 6.37 (d, 1H, J=2.8Hz); 6.43 (dd, 1H, J= 2.6Hz & 9Hz); 6.71-6.73 (m, 2H); 7.07 (d, 2H, J=8.8Hz); 7.83 (d, 1H, J=9.2Hz). ¹³C-NMR (CD₃OD): 204.77, 167.83, 166.88, 157.53, 134.10, 131.56, 127.13, 116.57, 114.39, 108.53, 102.06, 56.25 & 44.99. Mass: m/z 259 (M⁺+1); Melting point: 137-138°C; R_f: 0.157 (hex: EtOAc: 8:2).

3-Phenyl-4(4-hydroxybenzyl)-7-methoxycoumarin (4): Carbonyl diimidazole (CDI) (6.6 g, 41 mmol) was added in several portions over 5 min to a solution of phenyl acetic acid (5.05 g, 37.3 mmol) in 60 ml of dimethylformamide (DMF) at room temperature. The reaction mixture was stirred at 40 °C for 10 min and cooled to room temperature. The ketone 3 (4.91 g, 19 mmol), potassium carbonate (7.85 g, 57 mmol) and N,N-dimethyl aminopyridine (DMAP) (0.47 g, 3.8 mmol) were added, and the reaction mixture was heated to 80 °C for 2 hrs. The reaction mixture was cooled and poured into water (100 ml) and extracted with methylene chloride. The organic layer was dried and concentrated under vacuum. Crystallization with methanol gave the product 4 as colorless crystals (3.86 g, 57 %). ¹H-NMR (acetone-d₆): 3.89 (s, 3H); 4.01 (s, 2H); 6.73 (d, 2H, J=8.8Hz); 6.80 (dd, 1H, J=2.4 & 8.8 Hz); 7.01 (d, 2H, J=8.4Hz); 7.32-7.42 (m, 5H); 7.55 (d, 1H, J=8.8Hz). ¹³C-NMR (acetone-d₆): 163.30, 161.38, 156.94, 156.01, 150.16, 136.17, 130.98, 129.97, 129.85, 129.06, 128.84, 128.77, 126.43, 116.44, 116.36, 113.92, 112.81, 101.41, 56.32 & 35.27. Mass: m/z 357 (M⁺−1); Melting point: 236-238 °C; R_f: 0.07 (hex: EtOAc: 8:2).

4-[(7-Methoxy-2-oxo-3-phenyl-2H-chromen-4-yl)-methyl]-phenyl-trifluoromethane sulfonate (5): To a solution of the phenol 4 (3.86 g, 10.8 mmol) in methylene chloride (97 ml) was added pyridine (9.43 g, 119 mmol) and the solution was cooled to 0 °C. Triflic anhydride (3.87 g, 13.7 mmol) was added drop wise and stirred at room temperature for 2 hrs. The reaction mixture was diluted with methylene chloride (100 ml) poured into cold water, and separated. The organic phase was washed with water, dried and evaporated to give the desired product 5 as a white solid (5.24 g, 98 %). ¹H-NMR (CDCl₃): 3.87 (s, 3H); 4.08 (s, 2H); 6.77 (dd, 1H, J=2.4Hz & 9Hz); 6.89 (d, 1H, J=2.4Hz); 7.10-7.17 (m, 4H); 7.21-7.24 (m, 2H); 7.33 (d, 1H, J=8.8Hz); 7.34-7.42 (m, 3H). ¹³C-NMR (CDCl₃): 162.34, 161.14, 154.85, 148.00, 147.90, 138.54, 133.95, 129.57, 129.54, 128.48, 128.27, 126.87, 125.99, 121.53, 112.57, 112.43, 100.85, 55.63 & 34.67. ¹⁹F-NMR (CDCl₃): −73.29. Mass: m/z 491 (M⁺+1); Melting point: 148-149 °C; R_f: 0.3 (dichloromethane; DCM).

Preparation of 2E-tert-butyl 3-{4-[(7-methoxy-2-oxo-3-phenyl-2H-chromen-4-yl) methyl]-phenyl}-acrylate (6): To a solution of the triflate 5 (0.47 g, 0.96 mmol), in anhydrous DMF (5.54 ml) in a pressure tube was added t-butyl acrylate (0.71 ml, 4.8 mmol, 5eq), triethyl amine (TEA) (0.24 ml, 1.75 mmol, 2eq) and catalyst Pd(PPh₃)₂Cl₂ (84.2 mg, 0.12 mmol, 12.5 mol%), and the reaction mixture was degassed with nitrogen for 1 min. The pressure tube was sealed and heated at 100 °C for 24 hrs. The tube was cooled to room temperature and poured into ice water and extracted with ethyl acetate. The organic layer was washed with water and dried and evaporated. The crude was triturated with methanol and the solid was filtered and washed with methanol to give 6, a pale yellow solid (380 mg, 85%). ¹H-NMR (CDCl₃): 1.51 (s, 9H), 3.83 (s, 3H); 4.06 (s, 2H); 6.30 (d, 1H, J=16Hz); 6.78 (dd, 1H, J=2.4Hz & 8.8Hz); 6.88 (d, 1H, J=2.4Hz); 7.06 (d, 2H, J=8.4 Hz); 7.26-7.40 (m, 7H); 7.51 (d, 1H, J=16Hz). ¹³C-NMR (CDCl₃): 28.15, 35.48, 55.70, 80.48, 100.82, 112.40, 112.92, 120.08, 125.90, 127.19, 128.27, 128.35, 128.47, 129.69, 133.10, 134.16, 140.12, 142.79, 148.40, 154.86, 161.34, 162.27, 166.18. Mass: m/z 469 (M⁺+1); Melting point: 185-186 °C; R_f: 0.11 (DCM).

2E-3-{4-[(7-hydroxy-2-oxo-3-phenyl-2H-chromen-4-yl)-methyl]-phenyl}-acrylic acid (7; SS5020): To a solution of the ester-ether (0.38 g, 0.81 mmol) in DCM (31 ml) at 0 °C was added 1 M boron tribromide in DCM (11 ml). The mixture was stirred at room temperature for 3 hrs. The reaction mixture was quenched with water and ice and stirred for 30 min and then filtered and dried to give 7, a white solid (290 mg, 90%). ¹H-NMR (CD₃OD): 4.09 (s, 2H); 6.41(d, 1H, J=16Hz); 6.69 (dd, 1H, J=2.4 & 8.8Hz); 6.77 (d, 1H, J=2.4Hz); 7.13 (d, 2H, J=8.4Hz); 7.25-7.27 (m, 2H); 7.29-7.41 (m, 3H); 7.44 (d, 1H, J=8.8Hz); 7.47 (d, 2H, J=8.4Hz); 7.59 (d, 1H, J=16Hz). ¹³C-NMR (CD₃OD): 36.25, 103.57, 113.41, 114.52, 119.48, 126.12, 129.15, 129.34, 129.58, 129.67, 129.81, 131.16, 134.37, 136.07, 142.26, 145.75, 151.43, 156.39, 162.74, 163.78 & 170.61. Mass: m/z 399 (M⁺+1); Melting point: 212-215°C; R_f: 0.5 (DCM: MeOH: 9:1).

Determination of DNA adducts in rats treated with antiestrogens

The level of DNA adduct induced by antiestrogens was determined, according to the same protocol established previously (32, 35). Briefly, SS5020 (3 rats/dose) was administered orally to rats (Sprague-Dawley, 8-week-old females) for 7 days. Because large amounts of TAM-DNA adducts were detected in hepatic DNA of rats treated with TAM [20 mg (54 μmol)/kg/day] (35), the same molar dose was used for SS5020. Control animals received vehicle only. The animals were euthanized by CO₂ asphyxiation. The hepatic DNA extracted using a Qiagen kit was subjected to ³²P-postlabeling/gel electrophoresis (³²P-postlabeling/PAGE) analysis for determination of DNA adducts (32, 35). The detection limit for 5 μg DNA sample was ~7 adducts/10⁹ nucleotides. Animals were used in compliance with guidelines established by the NIH Office of Laboratory Animal Welfare.

Determination of uterotrophic potential

The uterotrophic potential of SS5020 and its related compounds were determined, applying the same protocol reported recently (32). Briefly, ovariectomized (OVX) rats (Sprague-Dawley, 6-week-old females, Taconic; 5 rats/dose) were treated orally for 3 days with an antiestrogen and the uterine weight was measured 1 day after the final treatment. A dose (0.27, 2.7, or 27 μmol/kg/day) molar equivalent to TAM (0.1, 1.0, or 10 mg/kg/day) was used for SS5020 and its related antiestrogens (TAM, TOR, RAL, and SP500263). The control rats received vehicle only. Uterine wet-weight/body-weight ratios were compared with that obtained for the OVX-rats treated subcutaneously with E₂ (0.3 μg/100 μl corn oil/rat/day) as a positive control.

DMBA-induced rat mammary carcinoma model

Mammary carcinoma (~8 mm diameter tumor) was induced by treating rats (Sprague-Dawley, 8-week-old females) with a single oral dose (50 mg/kg) of DMBA, following the established protocol (32, 36, 37). The rats were then treated orally for 4 weeks with antiestrogens, including SS5020 (5 rats/dose). Because 20 or 40 mg TAM is generally used as a daily dose for breast cancer patients [20 or 40 mg/60 kg B.W. = 0.33 mg (0.9 μmol)/kg or 0.66 mg (1.8 μmol)/kg], a similar molar dose equivalent to TAM was used for the antiestrogens examined. Controls received vehicle only. The size of the tumors was recorded once a week, using the two perpendicular dimensions. The volume of the tumors (TV) was estimated from equation 1: TV (mm³) = (length) × (width)²/2, calculated by measuring the width and length of the tumor. The relative tumor volume (RTV) was calculated as the ratio of the TV on day n to that on day 0, according to equation 2: RTV = (TV on day n)/(TV on day 0) × 100.

Athymic nude mice implanted with MCF-7 human mammary tumor xenograft

OVX-nude mice (nu/nu-BALB/c, 8-week-old females, Taconic) supplemented with an E₂ (0.72 mg) pellet were injected with MCF-7 cells (4 × 10⁷ cells) s.c. into the shoulder region, following an established protocol (32, 38); the tumors were allowed to grow for 6 weeks (up to ~6 mm in diameter). The nude mice were then treated orally for 4 weeks with SS5020 or TAM at a dose molar equivalent to TAM [3.0 (8.1 μmol) or 10 mg (27 μmol)/kg/day, p.o.; 4 mice/dose]. The control mice received vehicle only. The size of the tumors was measured once a week, as described for the studies of DMBA-induced mammary tumors.

Statistical Analysis

Statistical analysis was performed to evaluate the significance of the differences in treatment effects at a given time point using the F-test. The Tukey procedure was used to adjust for multiple pairwise comparisons between treatments using least-square means. All analyses were performed using the PROC MIXED procedure of SAS software (Cary, NC). A two-tailed p < 0.05 is regarded as statistically significant.

Results

Determination of DNA adducts induced by SS5020 in rats

Synthetic procedures of SS5020 (Fig. 1b) and its purity are described in the Materials and Methods. SS5020 was designed not to produce DNA damage. The capability of forming DNA adducts was used as an indicator to evaluate the genotoxic potential of antiestrogens. Sensitive ³²P-postlabeling/PAGE analysis with the detection limit of 7 adducts/10⁹ nucleotides (35, 39) was used to determine the level of DNA adducts generated by SS5020, and compared to TAM as a positive control. Rats were treated orally for 7 days with SS5020 or TAM at a dose molar equivalent to TAM [20 mg (54 μmol)/kg/day], as studied previously with TAM (40). A high level (~1 adduct/10⁵ nucleotides) of hepatic DNA adducts was observed in TAM-treated rats whereas no DNA adducts were detected in rats treated with SS5020 or rats receiving the vehicle only (Fig. 2). Using ³²P-postlabeling/PAGE analysis, any bulky DNA adducts including adducts derived from benzo[a]pyrene diol epoxide or 2-acetylaminofluorene can be detected even though their migration on the gel is varied (35, 39). Any DNA adducts derived from SS5020 were not detected on the gel at the regions different from that of TAM-DNA adducts. As expected, SS5020 may be a compound free of genotoxic effects.

Fig. 2.

Formation of hepatic DNA adducts in rats treated with antiestrogen. Rats were treated orally for 7 days with TAM or SS5020 at a dose molar equivalent to TAM (20 mg/kg/day). Control animals received vehicle only. The hepatic DNA samples were extracted using a Qiagen kit and analyzed by the ³²P-postlabeling/PAGE method for determination of the level of DNA adducts, as described in the Materials and Methods.

Determination of the uterotrophic potential of SS5020

The uterotrophic activity was used as an indicator to evaluate the estrogenic potential of antiestrogens. The uterotrophic activity of SS5020 was determined in OVX-rats and the results compared with those for TAM, TOR, RAL or SP500263, compounds all being used clinically or being considered for clinical trials. OVX-rats were treated orally for 3 days with each compound with a dose molar equivalent of TAM [10 mg (27 μmol)/kg/day] and the uterine wet weight [mg/g body weight (B.W.)] was measured one day after the final treatment. The dose given to rats was approximately 30 times higher than the human equivalent dose of TAM (20 mg/day). The uterine weight of the untreated OVX-rats was 0.255 mg/gB.W. and the uterine weight of OVX-rats treated subcutaneously with E₂ (0.3 μg/day for 3 days) was 1.074 mg/gB.W (Fig. 3). TAM had high uterotrophic activity, showing 54% of that observed for E₂-treated OVX-rats; even at doses of 0.1 and 1.0 mg/kg TAM, the uterotrophic activities were still 33% and 34%, respectively. The uterotrophic potential of TOR (46%) was not significantly different from that of TAM. RAL showed partial uterotrophic activity (14%) similar to that observed with SP500263 (15%), indicating that these antiestrogens have weak estrogenic activity. In contrast, SS5020 did not have significant uterotrophic activity at doses of 1.0 mg/kg (data not shown) and 10 mg/kg (Fig. 3), indicating that the newly synthesized compound has little or no estrogenic action.

Fig. 3.

Uterotrophic potential of antiestrogens on OVX-rats. OVX-rats (5 rats/group) were treated orally for 3 days with each compound (TAM, TOR, RAL, SP500263 or SS5020) at a dose molar equivalent to TAM (10 mg/kg/day). The control rats received vehicle only. At one day after the final treatment, uterine wet-weight/body-weight ratios were measured and compared with that obtained for OVX-rats treated subcutaneously with E₂ (0.3 μg/rat/day) as a positive control, as described in the Materials and Methods. Statistical analysis (t-test) was performed for multiple comparisons to evaluate the difference; p < 0.05 (SS5020 vs RAL or SP500263), p < 0.00001 (SS5020 vs E2, TAM or TOR).

Antitumor effect of SS5020 against DMBA-induced mammary tumor

Rats bearing mammary tumors were treated orally for 4 weeks with SS5020, TAM, RAL or SP500263 at the equivalent molar dose of TAM [1.0 mg (2.7 μmol)/kg BW/day], which is slightly higher than a human equivalent dose of TAM [0.33 mg (0.9 μmol) or 0.66 mg (1.8 μmol)/kg/day]. The tumors in the control rats grew rapidly, becoming approximately 11 times larger after 4 weeks (Fig. 4A). With SS5020, the tumor volume did not increase; rather, it decreased to 70% of the initial volume (***, p <0.001 and ****, p < 0.0001 versus the control), as indicated by statistical analysis. In contrast, using the same molar dose for TAM or RAL, the tumor growth was not statistically inhibited, compared with the control (Fig. 4A). Statistical significance was observed for SS5020 versus TAM or RAL at day 21 and 28 after treatment; for example, at day 28, p < 0.0001 for SS5020 vs. TAM or RAL. SP500263 suppressed the tumor growth, but the tumor inhibitory effect was significantly lower than that observed with SS5020. With a lower dose (0.9 μmol/kg/day) of SS5020, the tumor growth was suppressed significantly, compared with TAM and the control (Fig. 4B). Thus, SS5020 showed strong antitumor potential superior to TAM, RAL or SP500263 against DMBA-induced mammary tumors.

Fig. 4.

Antitumor potential of SS5020 against DMBA-induced mammary carcinoma. Rats (5 rats/dose) bearing DMBA-induced mammary carcinomas were treated orally for 4 weeks (A) with TAM, RAL, SP500263, or SS5020 at a molar equivalent dose of TAM [1.0 mg (2.7 μmol)/kg/day] and (B) with TAM or SS5020 at a molar equivalent dose of TAM [0.33 mg (0.9 μmol)/kg/day]. Controls received vehicle only. The size of the tumors (TV) was recorded once a week, using the two perpendicular dimensions, as described in the Materials and Methods. The RTV (%) was calculated as the ratio of the TV on day n to that on day 1. Based on the F-test, the relative tumor volumes were compared at each time point. Pairwise comparison results (p-values) were obtained based on Tukey’s adjustment and using least-square means. (A); *, p < 0.05, ***, p < 0.001 and ****, p < 0.0001 vs control. a) p < 0.05 vs SP500263, p < 0.0001 vs TAM or RAL; b) p < 0.0001 vs TAM, RAL or SP500263; c) p < 0.05 vs RAL, p < 0.01 vs TAM; d) p < 0.001 vs TAM, p < 0.0001 vs RAL. (B); ****, p < 0.0001 vs control. e) p < 0.05 vs TAM.

Antitumor potential of SS5020 against human MCF-7 breast cancer xenograft

SS5020 was administered orally for 4 weeks to athymic nude mice bearing an ER-positive MCF-7 human breast cancer xenograft (Fig. 5). The tumor volume of the control mice increased by approximately 7.1 times in 4 weeks. Compared with the control, TAM [10 mg (27 μmol)/kg/day] and SP500263 at the equivalent molar dose of TAM did not significantly inhibit the tumor growth. In contrast, with SS5020, the tumor growth was suppressed significantly during the period of treatment compared with other antiestrogens (p < 0.05 for SS5020 vs. TAM, SP500263 or the control). Even at a low dose equivalent to TAM [3 mg (8.1 μmol)/kg], the tumor growth was inhibited similarly. Evidently, SS5020 has higher antitumor potential than TAM and SP500263 against human breast cancer xenograft.

Fig. 5.

Antitumor potential of SS5020 against human MCF-7 breast cancer xenograft. OVX-nude mice bearing MCF-7 xenograft (4 mice/dose) were treated orally for 4 weeks with SS5020, TAM or SP500263 at a dose molar equivalent to TAM [3.0 mg (8.1 μmol) or 10 mg (27 μmol)/kg/day]. The control received vehicle only. Based on the F-test, the relative tumor volumes were compared at each time point. Pairwise comparison results (p-values) were obtained based on Tukey’s adjustment and using least-square means. *, p < 0.05 vs control. a) p < 0.05 vs TAM or SP500263.

Discussion

Undesirable effects on the uterus were major reasons to halt clinical studies of antiestrogens. To minimize such adverse effects, we gave the highest priority in our study to finding a safer compound lacking genotoxic and estrogenic potential in animals. To avoid genotoxic action observed with TAM, SS5020 was designed as a new benzopyran antiestrogen that does not include an ethyl moiety that is subject to α-hydroxylation and subsequent O-sulfonation, leading to cellular DNA damage. With our sensitive ³²P-postlabeling/PAGE analysis, SS5020, unlike TAM, did not produce DNA-adducts in the liver of rats given a dose (20 mg/kg/day) 60 times higher than the human equivalent dose of TAM. In addition, to increase the antiestrogenic potential, an acrylate moiety (−CH=CHCOOH) present in GW5638 was incorporated into SS5020. SS5020 did not present significant estrogenic activity in comparison to the untreated control. In contrast, SP500263, a benzopyran antiestrogen, still retained weak uterotrophic potential similar to that of RAL, as reported previously by another research laboratory (41) (Fig. 3). Although long-termed treatment of SS5020 to animals is required to determine the ability of inducing gene mutations and chromosomal aberrations, SS5020, lacking both DNA adduct formation and estrogenic action, could be a safer candidate for clinical trials on breast cancer therapy. The estrogenic properties of TAM and RAL in tissues other than the uterus provided significant potential for the prevention of osteoporosis as well as cardiovascular diseases (42, 43). The pharmacological properties of SS5020 on bone density and/or the cardiovascular system should also be determined to explore the additional benefits of this compound.

The antitumor potential of SS5020 was determined by using rats bearing DMBA-induced mammary carcinoma. At the same molar dose equivalent to TAM [1.0 mg (2.7 μmol)/kg B.W./day], TAM and RAL did not show any significant antitumor potential, compared with the control (Fig. 4). Supporting the results, effective regressions against DMBA-induced mammary tumor were observed at minimum with 2 mg (5.4 μmol)/kg oral treatment of TAM (44). Oral treatment of RAL, named originally as keoxifene (LY156758), also showed inhibitory effect against DMBA-induced mammary tumor only at a high dose [20 mg (39.2 μmol) or 40 mg (78.4 μmol)/kg/day] (45). Therefore, TAM or RAL may not inhibit significantly the tumor growth at the lower doses (0.9 or 2.7 μmol/kg) we examined. In sharp contrast, SS5020 completely suppressed the tumor growth during the entire period of treatment, indicating that the antitumor potency of SS5020 was superior to that observed with TAM or RAL. The antitumor activity of SP500263 was also higher than that of TAM and RAL, but significantly lower than that of SS5020. Replacement of a piperidinyl alkyl moiety [−OCH₂CH₂N(CH₂)₅] in SP500263 with an acrylate moiety (−CH=CHCOOH) apparently enhances the antitumor activity. With the human-equivalent molar dose of TAM [0.33 mg (0.9 μmol)/kg/day], SS5020 effectively suppressed tumor growth. In addition, SS5020 showed a strong inhibitory effect against human MCF-7 breast cancer xenograft in nude mice, compared with TAM and SP500263 (Fig. 5). Based on the results of the animal experiments, antiestrogen representing lower uterotrophic activity tends to provide higher antitumor activity against ER-positive mammary tumors. These results indicate that SS5020 is a superior alternative to TAM, RAL or SP500263 for breast cancer therapy.

Using a similar experimental system with nude mice, intraperitoneal administration of SP500263 (3 or 30 mg/kg) reduced the growth of MCF-7 tumor by approximately 65% and 85%, respectively (46). However, oral administration of SP500263 (10 mg/kg/day) in our studies did not significantly suppress the tumor growth (Fig. 5), suggesting that the oral bioavailability of SP500263 in mice may be poor. The antitumor potential of SS5020 against the growth of DMBA-induced mammary tumor and MCF-7 human breast cancer xenograft was much higher than that observed with SP500263. The non-estrogenic potential observed with SS5020, by replacing a piperidinyl alkyl moiety [−OCH₂CH₂N(CH₂)₄] in SP500263 by an acrylate moiety (−CH=CHCOOH), may explain the enhanced antitumor activity. The molecular mechanisms including interactions of SS5020 with receptors binding sites should be determined to explore such beneficial effects.

In this study, SS5020, a novel benzopyran antiestrogen, did not have any detectable DNA adduct formation and estrogenic effects on rats, as observed recently with a new triphenylethylene antiestrogen SS1020 (32). Like SS1020, SS5020 showed much higher antitumor potency than that of TAM, RAL and SP500263 against DMBA-induced mammary tumor in rats and human MCF-7 breast cancer xenograft implanted in athymic nude mice. Although the antitumor potential of SS5020 against MCF-7 breast cancer xenograft was slightly weaker than that observed with SS1020 (32), SS5020 could also be used as a safer benzopyran for breast cancer therapy and prevention. Although significant side-effects have so far not been observed in animals treated with SS5020 or SS1020, the results from further animal studies to evaluate the long-term toxicity/safety of SS5020 and SS1020 may give us the priority of being considered their pre-clinical and clinical studies.

Abbreviations

SS5020

2E-3-{4-[(7-hydroxy-2-oxo-3-phenyl-2H-chromen-4-yl)-methyl]-phenyl}-acrylic acid

SS1020

E-3-{4-[(E)-4-chloro-1-(4-hydroxyphenyl)-2-phenylbut-1-enyl]-phenyl} acrylic acid

TAM

tamoxifen

TOR

toremifene

RAL

raloxifene

E₂

1 7β-estradiol

DMBA

7,12-dimethylbenz(a)anthracene

D C M

dichloromethane

DMF

dimethylformamide

ER

estrogen receptor

OVX

ovariectomized

TV

tumor volume

RTV

relative tumor volume

bw

body weight