Pattern of Antitumor Activity of a Novel Camptothecin, ST1481, in a Large Panel of Human Tumor Xenografts¹

Camptothecins are among the most promising antitumor drugs. Their interest is related to the wide spectrum of antitumor activity and to the fact that topoisomerase I is a useful target for the design of effective antitumor agents. The clinically available camptothecins such as irinotecan and topotecan are active in a number of solid tumors, including colorectal carcinoma and lung cancer (1). Despite their specificity for the unique cellular target, some general limitations of camptothecins include the reversibility of drug-target interaction (i.e., instability of drug-induced enzyme-DNA cleavage complexes); the opening of the lactone ring, resulting in an inactive carboxylate form; and preferential binding of the open form to human serum albumin (2). Chemical modifications of camptothecins aimed at favoring rapid drug uptake and intracellular accumulation and stabilizing the labile drug-target interaction are promising strategies to improve the pharmacological profile of these agents. Recently, a novel series of camptothecins substituted at position 7 with lipophilic chains was developed (3, 4). Among the analogues of this series, the compound ST1481 was selected for further preclinical development, based on the cytotoxic potency and topoisomerase-inhibitory activity (4). Several advantages of oral ST1481 versus oral topotecan, in terms of antitumor efficacy, potency, and improvement of the therapeutic index, were documented in preclinical studies (5). The novel camptothecin ST1481 is not recognized by the recently described breast cancer resistance-associated protein, which is implicated in cellular resistance to topotecan, SN-38, and mitoxantrone (6). Based on the favorable therapeutic features of ST1481, this study was undertaken to investigate its preclinical profile of antitumor efficacy in a large panel of human tumor xenografts, with particular attention to tumor types known to be resistant to chemotherapy.

The results of cellular pharmacology and in vivo studies support the therapeutic potential of the novel 7-substituted camptothecin in the treatment of specific tumor types and its overall interest for clinical development.

ST1481 (7-t-butoxyiminomethylcamptothecin, gimatecan; Sigma-Tau, Pomezia, Rome) was dissolved in DMSO and stored at −20°C until use. The synthesis of the molecule has been reported elsewhere (4). On the days of treatment, the drug was thawed and suspended in sterile distilled water (10% DMSO for in vivo studies). Topotecan, kindly supplied by Smith-Kline Beecham Pharmaceuticals (King of Prussia, PA), was dissolved in sterile distilled water. CPT-11 (Campto; Aventis) was ready to use.

Human tumor cell lines were used in the study. The IGROV-1 ovarian carcinoma line, the U2OS osteosarcoma line, three glioblastoma lines (GBM, SW1783, and U-87-MG), and the LP melanoma line, as well as two sublines selected for resistance to cisplatin (IGROV-1/Pt1 and U2OS/Pt), were all maintained as monolayers in RPMI 1640 supplemented with 10% FCS. The p53 gene status of the cell lines has been described elsewhere (7, 8, 9, 10). For convenience, the information is presented in Table 2.

Cellular sensitivity to drugs was evaluated by growth inhibition assay after a 1-h drug exposure. Cells in the logarithmic growth phase were seeded in duplicates into 6-well plates. Twenty-four h after seeding, the drug was added to the medium. Cells were harvested 72 h after drug exposure and counted with a cell counter. IC₅₀ is defined as the drug concentration causing a 50% reduction of cell number compared with that of untreated control.

Cells were seeded in 75-cm² flasks, and 24 h later, they were exposed to ST1481 for 24 h. Cells were then harvested, fixed, and stained with a propidium iodide-containing solution (30 μg/ml propidium iodide and 66 units/ml RNase in PBS) for 30 min. Cell cycle perturbations were analyzed by a FACScan flow cytometer equipped with an argon laser (Becton Dickinson, Mountain View, CA). Five thousands cells were analyzed for DNA content. Cell cycle distribution was calculated using the Lysis II software (Becton Dickinson).

All experiments were carried out using 8–10-week-old female athymic Swiss nude mice (Charles River, Calco, Italy). The mice were kept in laminar flow rooms at constant temperature and humidity. They had free access to food and water. Experimental protocols were approved by the Ethics Committee for Animal Experimentation of the Istituto Nazionale Tumori of Milan, according to the United Kingdom Coordinating Committee on Cancer Research Guidelines (11).

The characteristics of the human tumor lines used in the study are reported in Table 1. They are representative of different tumor types known to be highly resistant to chemotherapy. Tumor lines were maintained by s.c. passages of tumor fragments (about 2 × 2 × 2 mm) in healthy mice according to previously reported procedures (12).

For chemotherapy experiments, tumor fragments were used for all tumor lines, except in the case of the GEO colon tumor, for which cells (7 × 10⁶) from in vitro cultures were injected s.c. into mice. Each experimental group included four/six mice bearing bilateral tumors. Tumors were implanted on day 0, and tumor growth was followed by biweekly measurements of tumor diameters with a Vernier caliper. TV³ was calculated according to the following formula: TV (mm³) = d² × D/2, where d and D are the shortest diameter and the longest diameter, respectively. In all but one experiment, drug treatment started when tumors were just measurable (mean TV <100 mm³). A comparison of ST1481 and CPT-11 efficacy against the NCI-H460 model was also performed in animals bearing very large tumors (∼550 mm³). All drugs were delivered by gavage in a volume of 10 ml/kg, except for CPT-11 (350 mg/kg; 17.5 ml/kg). Drugs were administered every fourth day for four times (q4d×4) at their MTD (roughly corresponding to LD₁₀), i.e., 15 mg/kg for topotecan and 2 or 3 mg/kg for ST1481 (5). Two doses of CPT-11 were also investigated. In all experiments, control mice were treated with the drug solvent.

Drug efficacy was assessed as described below.

(a) TVI in drug-treated versus control mice was expressed as follows: TVI (%) = 100 − (mean TV treated/mean TV control × 100). TVI was evaluated 5–15 days after the last treatments according to the doubling time of the control tumors. A TVI value higher than 80% is indicative of active compounds.

(b) LCK was calculated using the following formula: LCK = (T − C)/3.32 × DT where T and C are the mean time (in days) required for treated (T) and control (C) tumors, respectively, to reach a determined volume, and DT is the doubling time of control tumors. A LCK value greater than 1 is indicative of active compounds.

(c) CR was defined as disappearance of the tumor lasting at least 10 days after the end of treatments. Tumors that had not regrown by the end of the experiment were considered “cured.”

For statistical comparison, TVs of ST1481-treated versus topotecan-treated mice were compared by Student’s t test on the day TVI was assessed. The Fischer test was used to compare the CR rates. When not specifically indicated, Ps between the two groups were >0.05 (i.e., not significant).

Toxic effects of drug treatment were assessed as described below.

(a) BWL was calculated as follows: BWL (%) = 100 − (mean body weight day x/mean body weight day 1 × 100), where day 1 is the first day of treatment, and day x is any day thereafter. The highest (maximum) BWL is reported in the tables. Mice were weighed twice a week throughout the period of experimentation.

(b) Lethal toxicity was defined as any death in treated groups occurring before any control death. Mice were inspected daily for mortality.

The antiproliferative activities of ST1481 and topotecan were compared in a panel of human tumor cell lines including cells selected for resistance to cisplatin (Table 2). In all tested cell systems, the novel analogue ST1481 showed marked cytotoxic potency. Whereas among the examined cell lines a considerable difference in cell sensitivity to topotecan was found (IC₅₀ range, 80–980 ng/ml), the same tumor cell lines, with the exception of U-87-MG glioblastoma cells, showed comparable sensitivity to ST1481 with IC₅₀ in the range between 8 and 30 ng/ml. The most sensitive cell line was U2OS osteosarcoma. Among the sublines selected for resistance to cisplatin, only U2OS/Pt cells exhibited partial cross-resistance to both agents.

The relatively low sensitivity of U-87-MG glioma cells to ST1481 compared with slowly proliferating GBM cells is rather unexpected on the basis of the proliferation rate. The expression of transport systems is not expected to influence the cytotoxicity of ST1481 (5, 6). This led us to examine drug-induced perturbation of the cell cycle in view of the fact that the efficiency of the S-phase DNA damage checkpoint could critically influence the cellular outcome. In fact, whereas U-87-MG cells were blocked in S phase, GBM cells were arrested in G₂ (Fig. 1). The profiles of cell cycle perturbation induced by topotecan were similar to those found after exposure to ST1481 (data not shown). However, the lower sensitivity of GBM cells to topotecan as compared with U-87-MG cells was likely related to expression of P-glycoprotein, because, in contrast to ST1481, topotecan is a substrate for this transport system (5).

The pattern of antitumor activity of ST1481 was investigated in a large panel of s.c. growing human tumor xenografts representative of various tumor types, with particular reference to intrinsically resistant tumors. Under optimal treatment conditions (e.g., when protracted daily schedules are used), camptothecins may be curative against several human tumor xenografts growing in nude mice (13). Use of a less effective, intermittent treatment schedule may be expected to single out the most effective agents. In the present study, for the purpose of comparison, the drugs were delivered p.o. according to a q4d×4 schedule. Table 3 summarizes the antitumor effects of ST1481 and topotecan (used as a reference drug) at their MTDs, i.e., 2–3 mg/kg and 15 mg/kg, respectively. Of the three lung carcinoma models, NCI-H460 and LX-1 tumors were highly responsive to the tested camptothecins, which produced almost complete inhibition of tumor growth (TVI ≥ 98%). The slowly growing tumor A549 was markedly less responsive to both agents. Although the two drugs produced comparable effects in terms of TVI, in the two responsive lung tumors ST1481 achieved a significantly higher CR rate and higher LCK values than topotecan. When used against LX-1 lung carcinoma, ST1481 (3 mg/kg) achieved CR in all tumors (100%; P < 0.05 versus topotecan-treated tumors). Moreover, in 50% of treated tumors, no tumor regrowth was observed by the end of the experiment (day 90).

An unexpected finding of the present study was the outstanding efficacy of camptothecins against human melanoma models. LP and LM melanomas originating from primary and metastatic lesions of the same patient were much more responsive to ST1481 than to topotecan when the TVI and LCK values were considered. The LM tumor from the metastatic lesion was substantially less responsive than the tumor from the primary lesion, despite the increased proliferation rate of the metastatic cell line (Table 1). In particular, ST1481 induced a CR of LP and BLM melanomas in all treated animals, and in >50% of the tumors there was no evidence of regrowth at the end of the experiment. In both tumor lines, under the same conditions, topotecan produced only a 50% CR rate, and all tumors regrew during the period of experimentation.

An interesting observation in this study was the marked efficacy of both camptothecins in the treatment of s.c. growing glioma models. The most responsive tumor was SW1783, in which ST1481 induced a high CR rate. The high responsiveness of the glioblastoma GBM was unexpected, considering the slow growth of the tumor (Table 1). A correlation was found for in vivo responsiveness and cell sensitivity of glioma models to ST1481 because the less responsive U-87-MG model also exhibited a reduced cell sensitivity to this agent. Such a correlation was not evident for topotecan.

In the two investigated ovarian carcinoma models, ST1481 was more effective than topotecan in inhibiting tumor growth, but the difference between drug effects was significant only in the IGROV-1 tumor (P < 0.05).

Both camptothecins administered at their MTD showed relevant and comparable efficacy against tumors of gastrointestinal tract, i.e., GEO, HT29, and CoBA colon carcinomas and MKN 28 gastric carcinoma. ST1481 was more effective than topotecan against the HT29 colon tumor. In addition, ST1481 was also more effective in the treatment of the bladder carcinoma HT1376.

Both agents showed striking efficacy against U2OS osteosarcoma: all tumors regressed completely after drug treatment, and most mice were tumor free at the end of the experiment (day 120). The responsiveness of U2OS/Pt osteosarcoma was somewhat reduced compared with that of the parental tumor line. Nevertheless, ST1481 still retained a relevant efficacy against the cisplatin-resistant tumor subline.

A comparison of ST1481 and CPT-11 efficacy was also performed against the NCI-H460 tumor (Table 4). Both drugs were very active in inhibiting tumor growth when the treatment was started in the presence of small tumors. The observed difference in CR rate (3 of 10 versus 5 of 8, for CPT-11 and ST1481, respectively) was not significant. A comparable efficacy between the two drugs was shown even when bulky tumors were treated (TV ≅ 550 mm³), but under these conditions, a lower cumulative dose was tolerated.

To better exploit the therapeutic potential of camptothecins in the treatment of slowly growing tumors, a comparative study was performed, using a daily protracted treatment schedule, in A549 non-small cell lung carcinoma and HT29 colon carcinoma (Table 5). When used against the A549 tumor, the same total dose of topotecan, i.e., 60 mg/kg, delivered according to an intermittent or daily treatment schedule produced a marginal antitumor effect. However, a 2-fold higher dose (4 versus 2 mg/kg/injection) given according to the daily schedule was well tolerated and resulted in a superior effect in terms of TVI or LCK values. As for ST1481, a daily dose of 0.5 mg/kg could be safely administered for up to 6 weeks. Compared with the best results obtained with the intermittent administration schedule (q4d×4), ST1481 delivered daily resulted in better antitumor activity in terms of tumor growth inhibition and persistence of the effect (increased LCK value). The dose used in the protracted treatment should be considered approximately the maximum tolerated dose because a higher dose level, i.e., 0.8 mg/kg, proved to be toxic (2 of 5 deaths due to toxicity) after only 10 treatments. The effects of topotecan against the HT29 tumor were clearly improved with daily as compared with intermittent administration. Daily treatment with ST1481 at a low dose (0.25 mg/kg) produced a very strong and long-lasting effect that persisted even after the end of treatment.

The 7-substituted camptothecin analogue ST1481 was selected for preclinical development on the basis of its potent inhibition of topoisomerase I (4) and its cytotoxic and antitumor potency (4, 5). The lipophilic nature of the compound could allow rapid and efficient drug uptake by tumor cells. Therefore, the antiproliferative effects were determined after a short exposure (1 h) to the drug. The novel analogue exhibited an at least 10-fold increased cytotoxic potency with respect to topotecan in most examined cell lines. The pattern of cell sensitivity to ST1481 was markedly different from that of topotecan. In fact, whereas there was a marked variability in the sensitivity to topotecan (10-fold range) among the cells examined, the variability in the sensitivity for ST1481 was low (5-fold range). This behavior is rather unexpected, considering the variable proliferation rate and p53 mutational status of the cell lines used. We have previously shown that ST1481 is not a substrate for transport systems involved in the multidrug resistance phenotype (5, 6). The transport systems may have a variable influence on the efficacy of topotecan (14), and it is conceivable that multiple factors are implicated as determinants of chemosensitivity to topotecan. Among the factors recognized to play a role in the sensitivity to camptothecins, the DNA damage checkpoint causing S-phase arrest is likely implicated in modulating the sensitivity to ST1481, as suggested by the behavior of glioma cells in response to drug treatment (Fig. 1). This interpretation is supported by recent evidence that loss of this checkpoint function sensitizes cells to the cytotoxic effects of camptothecins (15). Indeed, a role for efficiency of DNA damage checkpoint is expected in determining cell sensitivity, although the relative contribution may be dependent on the biological context. The influence of the S-phase checkpoint on the sensitivity to topotecan was not evident in response to topotecan; in fact, the expression of MDR in GBM cells (16) might mask the contribution of the S-phase checkpoint in determining a differential sensitivity of GBM and U-87-MG cells to topotecan.

The comparable sensitivity of cells of various tumor types to ST1481 could be indicative of the ability of the drug to overcome multiple resistance mechanisms that could play a role in determining the variable sensitivity to topotecan. This interpretation is supported by antitumor efficacy studies indicating a relevant efficacy (TVI > 80%) in most tumors of our panel. Only two tumors (A549 lung carcinoma and MKN 28 gastric carcinoma) showed reduced responsiveness, probably reflecting a relatively low proliferation rate (doubling time = 6.1 and 7.6 days, respectively).

The panel of human tumor xenografts investigated in our study included tumor types (glioma, colon carcinoma, and ovarian carcinoma) in which high preclinical activity had already been reported for other camptothecins (17, 18, 19, 20), as well as other tumor types. The most interesting observation of the preclinical evaluation of ST1481 was the striking antitumor effect against human lung carcinoma and human melanoma xenografts. The superior efficacy of ST1481 over topotecan was clearly documented by the CR rate and the presence of “cured” tumors at the end of the observation time. Its outstanding efficacy (100% CR) in two of three of the tested melanoma models was surprising because this tumor type is known to be intrinsically resistant to conventional agents, and the tumor lines investigated were only marginally responsive to cisplatin or doxorubicin (TVI < 50%) in similar experimental conditions. Among the melanoma models, LP and LM tumors, derived from a primary tumor and a metastatic lesion of the same patient, presented a strikingly different responsiveness to camptothecins, suggesting the development of resistance during progression. In any event, clear superiority of ST1481 over topotecan was observed.

This preclinical study was performed using topotecan as a reference drug because, as for ST1481, it is not a prodrug. However, the comparison of ST1481 and CPT-11 performed in the NCI-H460 lung carcinoma clearly indicated an activity of ST1481 at least comparable to that of CPT-11. Potential advantages of ST1481 could be: (a) high potency after oral administration, likely related to lack of recognition by breast cancer resistance-associated protein (6) and good intestinal absorption; (b) good tolerability, likely related to lack of intestinal toxicity; and (c) favorable pharmacokinetic profile and tissue distribution, including ability to cross the blood-brain barrier (5).

Both ST1481 and topotecan were very effective in the treatment of U2OS osteosarcoma. Complete tumor response was achieved in all treated animals, with 70% “cured” tumors. The tumor variant selected for resistance to cisplatin was still responsive to camptothecin treatment, but the degree and the persistence of tumor growth inhibition were considerably less pronounced. The reduction of efficacy against U2OS/Pt was more marked for topotecan. Relevant to this point are the following observations: (a) U2OS cells were the most sensitive cells among the cell systems growing in vitro (Table 2); and (b) partial cross-resistance to both camptothecins was also evident at the cellular level. Thus, the in vivo responsiveness of osteosarcoma models closely reflected the relative cellular sensitivity.

In the tumor types known to be responsive to camptothecins in preclinical systems (ovarian and colorectal carcinoma), the responsiveness to ST1481 was lower than expected, which might suggest that the analogue has a spectrum of efficacy somewhat different from that of other camptothecins. In the aforementioned tumors, with the exception of HT29 colon carcinoma, the efficacy of ST1481 and topotecan was similar. The efficacy of ST1481 against glioma models might have clinical implications because the novel lipophilic analogue is known to be able to cross the blood-brain barrier (5).

In conclusion, the results of our study, obtained with an unfavorable administration schedule for conventional camptothecins, documented a unique profile of antitumor activity of the novel camptothecin ST1481. The use of a large panel of human tumor xenografts evidenced an exquisite efficacy against lung carcinoma and melanoma models. A major challenge in the development of a novel camptothecin is to optimize its efficacy. For example, a protracted daily schedule could be used to exploit the therapeutic potential in the treatment of slowly growing tumors such as the A549 lung carcinoma and the HT29 colon carcinoma. On the basis of the present and previous preclinical studies, ST1481 represents a promising candidate for clinical development. Phase I studies are ongoing.

We thank Enrica Favini for technical assistance and Laura Zanesi for editorial assistance.