Enhancement of in Vivo Antitumor Activity of Classical Anticancer Agents by Combination with the New, Glutathione-Interacting DNA Minor Groove-Binder, Brostallicin

Brostallicin ({N-[5-({[5-({[2-(aminoiminomethyl)amino]-ethyl}amino)carbonyl]-1-methyl-1H-pyrrol-3-yl}amino) carbonyl]-1-methyl-1H-pyrrol-3-yl}-4-[({4-[(2-bromo-1-oxo-2-propenyl)amino]-1methyl-1H-pyrrol-2-yl}carbonyl)amino]-1-methyl-1H-pyrrole-2-carboxamide, PNU-166196) is a new derivative of a distamycin A-like frame (1, 2, 3, 4), which has shown very promising activity in experimental tumor models (5, 6) and is currently under clinical investigation (7, 8).

In addition, brostallicin has shown a particularly favorable therapeutic index in preclinical studies, with myelotoxicity as dose-limiting toxicity (4, 6); its cytotoxic activity is not affected by mechanisms of resistance associated to mismatch repair deficiency or to treatment with alkylating agents and topoisomerase I inhibitors (4, 6, 9).

As for its mechanism of action, brostallicin is a DNA minor groove-binding agent of which the in vitro and in vivo activity is increased in tumor cells with higher GSH² and/or GST levels (4, 10, 11). The α-bromoacrylic moiety of brostallicin was found to react with GSH, in a reaction catalyzed by GST, with the possible formation of a highly reactive GSH-complex able to bind covalently to DNA. Among the different GST isoenzymes tested, the GST-π isoform was found to be the strongest activator of brostallicin-GSH reaction. Furthermore, transfection into tumor cells of the human cDNA encoding for GST-π resulted in an increased activity of the compound both in in vitro and in vivo experimental tumor models (10). This finding is particularly interesting, considering that GSTs, especially GST-π isoform, are overexpressed with high frequency in a wide variety of tumors, whereas GST levels in normal or surrounding tissues are low (12, 13, 14, 15, 16), thus, potentially representing a therapeutic advantage for brostallicin, rather than a mechanism of resistance as it has been observed for established antitumor agents such as platinum derivatives, alkylating agents, anthracyclines, and to a certain extent for topoisomerase I inhibitors (17, 18, 19, 20). Moreover it has been shown that anticancer drugs such as cDDP, DX, or CPT-11 are able to promote the expression or to increase the nuclear level of GST-π in tumor cells (21, 22) 

The peculiar mechanism of chemical interaction of brostallicin with GSH and particularly its enhanced activity in cells with high GST/GSH content, prompted us to investigate its efficacy in combination with drugs of which the mechanism of resistance is because of high GST/GSH levels. Theoretically these combinations should increase the number of killed cancer cells.

Because the combined use of anticancer agents has generally been the most effective approach for treating metastatic cancer than the use of single agents, most current chemotherapy regiments are based on combinations.

In the current article, we report data obtained testing the antitumor efficacy of brostallicin in combination with cDDP, CPT-11, DX, and Taxotere. We show that indeed brostallicin is suitable for combination treatments with other anticancer agents and shows synergistic activity in different tumor models.

Brostallicin, DX, and CPT-11 were provided by Pharmacia (Milan, Italy). cDDP was purchased from Sigma Chemical Co. (St. Louis, MO), and Taxotere was obtained from Aventis Pharmaceuticals. Drugs were dissolved and diluted just before use.

Four to 6-week-old female Swiss Ncr nu/nu mice (Charles River, Calco, Lecco, Italy), weighing 20–25 g, were used in experiments involving human tumors. Two to 3-month-old female DBA2 and CDF1 mice (Charles River), weighing 18–20 g, were used for L1210 cells. Mice were maintained under specific pathogen-free conditions, and provided food and water ad libitum.

Procedures involving animals and their care are conducted in conformity with the institutional guidelines that are in compliance with national (D.L. n.116, G.U., suppl.40, 18 febbraio 1992, Circolare No.8, G.U., 14 luglio 1994) and international laws and policies (EEC Council Directive 86/609, OJ L 358,1, Dec 12, 1987; Guide for the use of Laboratory Animals, United States National Research Council, 1996).

The human tumors (HCT-116 colon carcinoma and A549 NSCLC) were obtained from American Type Culture Collection (Rockville, MD) and were maintained in vitro as continuous cultures. HCT-116 and A549 cells were grown in Iscove’s medium or Ham’s F-12, respectively, at 37°C, 5% CO₂ in medium supplemented with 10% FCS.

For drug testing, tumor cells were implanted s.c. into the left flank of recipient athymic mice (5 × 10⁵ cells/mouse). When the tumor was palpable (∼0.2 g), animals were divided randomly into test groups consisting of at least 8 mice each (day 0). L1210 murine leukemia (National Cancer Institute, Frederick, MD) was maintained by weekly i.p. transplants (10⁵ cells/mouse) in DBA/2 mice, according to Geran et al. (23). For combination experiments, CD2F1 mice were injected i.v. with 10⁶ cells/mouse. Treatments started the day after tumor inoculum.

All of the drugs were administered i.v. in a volume of 10 ml/kg of body weight, according to the indicated schedules. For solid tumors, the length (L) and width (W) of the solid tumor mass were measured by caliper twice weekly, and the tumor volume (TV) was calculated as: TV = (L× W²)/2.

The tumor volume at day n was expressed as RTV according to the following formula RTV = TVn/TV₀, where TVn is the tumor volume at day n and TV₀ is the tumor volume at day 0.

The percentage of T/C% was determined by calculating RTV as: T/C% = 100×(mean RTV of treated group)/(mean RTV of control group). According to the National Cancer Institute standards, a T/C <42% is the minimum level for activity. The LCK was calculated from the following formula: LCK = T-C/3,32×Td; T-C, the tumor growth delay, was calculated as the median time (in days) required for the treated-group tumors less the median time required for the control-group tumors to reach a predetermined size (usually 1g), whereas Td is the tumor volume doubling time, in days, measured from a best-fit straight line of the control group in exponential growth. The compound was considered active when the LCK value was >0.7.

Drug activity for leukemia models was calculated as ILS% of treated animals compared with the control group.

Toxicity was evaluated on the basis of weight loss. Mice body weight was recorded the same day of tumor measurement and was taken as a measure of toxicity, together with gross autopsy findings, mainly in terms of reduction of spleen and liver size. Antitumor activity was evaluated at the highest nontoxic dose, which is the highest dose that can be administered without causing death or undue toxicity. A dose producing a weight loss nadir of ≥20% was considered as excessively toxic.

Tumor volumes of all of the groups were compared using Fisher’s test.

Analyses were computed using the StatView statistical package by SAS Institute Inc. (Third edition).

The antitumor activity of the combination therapy of brostallicin plus cDDP, CPT-11, and Taxotere in solid tumor-bearing mice was analyzed by the fractional product method as described (24, 25).

FTV relative to controls is calculated as the ratio between the mean tumor volume of the experimental groups and the mean tumor volume of the control group. The calculation was performed during the time after the last treatment. The expected FTV of the combination is the product of the mean FTV of the two drugs given as single drugs. The ratio was obtained by dividing the expected FTV by the observed FTV of the combination. A ratio of >1 indicates a more than additive effect, and a ratio of <1 indicates a less than additive effect (24, 25). The antitumor activity of the combination therapy in L1210-bearing mice of brostallicin plus DX was evaluated as reported by Izumi et al. (26).

Exponentially growing HCT-116 cells were seeded in T-25 flasks 48 h before treatment. Cells were exposed to cDDP (12, 25, and 50 μm) for 1 h and then incubated in drug-free medium for 24 and 48 h. Treated and untreated cells were harvested. Total GST activity was determined using 1-chloro-2–4-dinitrobenzene as a substrate (27). Reaction was performed using cytosolic extracts and measuring the conversion of 1-chloro-2–4-dinitrobenzene by GST using a spectrophotometer. The data were normalized for the amount of proteins present in each sample and expressed as nmol of dinitrophenylglutathione formed/min/mg protein at 37°C, using the extinction coefficient of 9.6 nm⁻¹cm⁻¹.

The antitumor activity of brostallicin in combination with classical anticancer agents acting with different mechanisms has been evaluated in vivo in different experimental systems.

The MLH1-deficient human colon carcinoma cell line HCT-116, transplanted s.c. in nude mice, was used to determine the effect of brostallicin given alone or in combination with the platinum derivative cDDP (Table 1; Fig. 1). Mice were treated when tumors reached ∼200 mg. Brostallicin and cDDP were given i.v. three times every 7 days (q7dx3). Two treatment schedules were tested: cDDP followed by brostallicin 2 days after (schedule A) or brostallicin followed by cDDP 2 days after (schedule B).

The rationale in selecting these two administration schedules was based on the observation that tumor cells treated with cDDP showed an increased activity of GST.

In fact, when GST activity was analyzed in HCT-116 cells, 24 and 48 h after a 1 h exposure to cDDP (Fig. 2), a statistical significant increase of GST activity was observed after 1-h treatment with cDDP (25–50 μm) and 48-h recovery in drug-free medium.

In all of the experiments performed (Table 1), drugs alone produced a dose-dependent inhibition of tumor growth. The highest nontoxic doses were 2 and 0.4 mg/kg/day for cDDP and brostallicin, respectively. When these two doses were combined, no addition of toxicity was found for both schedules.

As far as the antitumor activity is concerned, the effect obtained with the schedule A was significantly different from the extent of growth inhibition observed after the inverted administration schedule. The analysis of the results obtained with these two combination schedules are reported in Table 1. When cDDP administration was followed by brostallicin 48 h after, the ratio between the expected FTV and the observed FTV of the combination ranged from 1.0 to 2.6, which is clearly indicative of an increased activity of the combination over single drugs. Conversely, with the opposite schedule, the ratio ranged from 0.7 to 1, suggesting at best an additive effect of the combination.

Furthermore, schedule A was associated with a significant LCK increase (see Table 1) compared with both cDDP and brostallicin given alone.

On the contrary, LCK did not change significantly in schedule B in comparison to either drug alone.

The same tumor model was used to evaluate the efficacy of brostallicin in combination with the topoisomerase I inhibitor CPT-11. Drugs were given on the same day (q7dx3), with brostallicin administered 1 h after the i.v. injection of CPT-11. The results obtained are shown in Table 2. The highest dosages evaluated, 0.4 and 60 mg/kg/injection for brostallicin and CPT-11, respectively, were well tolerated with weight loss <5% either when the drugs were given alone or in combination. The combination therapy exhibited more potent antitumor activity than single-drug treatment. The LCK values for CPT-11 40 mg/kg, brostallicin 0.2 mg/kg, and the combination of the two were in fact 0.4, 0.1, and 1.1, respectively. Higher doses, still not toxic, such as CPT-11 60 mg/kg, brostallicin 0.4 mg/kg, and the combination of these two, gave a LCK of 0.8, 0.4, and 1.7, respectively. The combined use of brostallicin and CPT-11 appeared to be superior to single drugs against this tumor model, being the ratio between expected and observed FTV >1 for all of the tested doses (Table 2).

The antileukemic effect of brostallicin in combination with the anthracycline DX alone was studied in the i.v. injected murine L1210 leukemia. Three administration schedules were tested alone and in combination: simultaneous administration at day 1 after tumor inoculum (schedule 1) or at 24 h interval, with DX given at day 1 and brostallicin at day 2 (schedule 2) or vice versa (schedule 3).

Table 3 reports the data obtained testing brostallicin, DX, and the combination of the two drugs. At all of the tested doses the CI values were determined (Table 3). The antileukemic activity of brostallicin and DX administered alone was dose dependent. The highest nontoxic doses were 0.8 and 15 mg/kg/injection for brostallicin and DX, respectively. The combination therapy of the two drugs showed a higher ILS% compared with either drug alone, especially in the DX or brostallicin-preceding treatments (schedules 2 and 3). The CI values for the three schedules ranges between 0.67 and 1, 1 and 1.52, and 1.1 and 1.65 for schedules 1, 2, and 3, respectively. These results suggest that the sequential administration of brostallicin and DX gave an antitumor effect higher than that obtained with simultaneous therapy. In all of the treatment schedules both drugs were well tolerated.

A comparison between the antitumor activity of brostallicin and Taxotere alone and in combination against A549 NSCLC transplanted in nude mice is reported in Table 4. Drugs were given simultaneously (q7dx3), with brostallicin administered 1 h after the i.v. injection of Taxotere. The highest dosages evaluated, 0.4 and 20 mg/kg/injection for brostallicin and Taxotere, respectively, were well tolerated when drugs were administered alone. Conversely, when drugs were given in combination at these dosages, some initial toxicity could be observed. Against this tumor model, Taxotere was very active (LCK of 8.3 at 20 mg/kg/injection), and brostallicin did not show any significant antitumor activity. When combination therapy was performed with the highest tolerated doses of both drugs, the antitumor effect was additive.

Brostallicin is a novel DNA minor groove binder anticancer agent in clinical development. The combined use of anticancer agents is a common strategy in clinical chemotherapy and, thus, the possibility to identify optimal combinations of the antitumor agents should be examined. The results of our study suggest that brostallicin is not only an active compound in single-agent therapy but is also a promising agent in combination with other anticancer agents.

Preclinical findings showing that brostallicin has increased activity in tumors with higher GSH/GST levels/expression (4, 10, 11), overcomes resistance to alkylating agents and topoisomerase I inhibitors, and is fully effective against mismatch repair-deficient tumor cells (23), stimulated our interest in studying the combined activity of brostallicin with drugs of which the antitumor activity is reduced by these mechanisms of resistance (13, 14, 15, 16, 17, 18, 19, 20). The involvement of GSTs in tumor resistance to a wide range of chemotherapeutic agents has been described extensively (13, 14, 15, 16, 17, 18, 19, 20). In addition, it has been shown that GST-π expression or amount into the nucleus increases in tumor cells in response to cDDP, DX, and CPT-11 treatment (21, 22).

The interaction and the biochemical mechanism of synergy between cDDP and brostallicin have been explored on a mismatch repair-deficient model (HCT-116; Ref. 28). The kinetic of GST activity in HCT-116 cells exposed to cDDP was evaluated and results confirmed the previously reported ability of cDDP in increasing the expression of GST-π. Because results showed a higher level of GST activity after 48-h recovery in drug-free medium, two combination protocols were investigated giving brostallicin 48 h before or after cDDP treatment. The sequence of drug administration in the combination influenced the response. cDDP/brostallicin interaction leads to a more than additive antitumor effect without additional toxicity, only giving cDDP before the brostallicin administration. These findings support our hypothesis that cDDP could enhance brostallicin activity through GST-π-stimulated overexpression. We found that the antitumor activity of CPT-11 is enhanced by brostallicin cotreatment. Although a clear understanding of the mechanism by which the antitumor activity of these two drugs benefits of their combination is not clearly understood, the combined use of brostallicin and CPT-11 is an interesting approach for colon cancer treatment considering that brostallicin overcomes camptothecin resistance and, therefore, can kill cells that can survive to CPT-11 treatment.

Furthermore, brostallicin preserves its activity in mismatch repair-deficient tumors, a condition found in human colon cancer with relatively high frequency (29, 30, 31).

In in vivo systemic L1210 murine leukemia model a more than additive antitumor effect was observed when DX/brostallicin were sequentially administered. The reasons for these results still remain to be determined, and other mechanisms, different from those involving the GSH/GST system, could be operative and should be investigated for designing more effective treatments regimens.

The last study reported in this article regards the combination of brostallicin and Taxotere. The choice of this agent was guided by the brostallicin-clinical plan in which NSCLC has been selected. Among the effective drugs in clinical use against NSCLC, Taxotere as single agent or in combination therapy provides a clinical benefit over or comparable with other agents (32, 33). Preclinical data showed that brostallicin/Taxotere simultaneous combination is feasible without any antagonistic effect.

Combination chemotherapy using conventional cytotoxic agents accomplishes several important objectives not possible with single-agent treatment. Here we report four promising brostallicin combination treatments. Brostallicin plus cDDP or CPT-11 or DX showed hints of therapeutic synergy, and brostallicin plus Taxotere activity was certainly additive. Altogether these results, together with the peculiar mechanism of action of the drug, indicate brostallicin as a new promising candidate for clinical use in combination therapy.