Our objective was to evaluate in vitro and in vivo the effect of the combination of trabectedin (Yondelis, ET-743) and irinotecan (CPT-11) or its major metabolite SN-38 in a human rhabdomyosarcoma cell line. The schedule trabectedin (1 h) followed by irinotecan or SN-38 (24 h) and the opposite sequence (irinotecan or SN-38 24 h followed by trabectedin 1 h) were analyzed in a rhabdomyosarcoma cell line. In vivo studies were conducted with trabectedin and irinotecan at the doses of 0.2 and 20 mg/kg, respectively, simultaneously administered with a q4d×3 schedule. In vitro studies indicated an overall additive effect [combination index (CI) relatively close to 1.0], with the former schedule slightly superior to the latter (at the IC50 effect levels: CI=0.89 versus 1.07). Neither transcription nor expression of DNA topoisomerase I was affected by trabectedin treatment. In vivo the therapeutic results of the combination were certainly more impressive: trabectedin and irinotecan combination caused a strong and long-lasting effect on tumor growth (tumor volume inhibition=89%, log10 cell kill=1.6), whereas each drug given as a single agent was only marginally active. The discrepancy between the in vitro and in vivo results suggests possible mechanisms involving host cells, other than tumor cells. The striking effects of the combination observed in vivo could be related to a combination of a direct cytotoxic and an anti-inflammatory indirect effect. The very marked and long-lasting effect of the trabectedin and irinotecan combination in vivo suggests a basis for a clinical evaluation in pediatric patients with rhabdomyosarcoma.

ET-743 (YondelisTM, Trabectedin) isolated from the tunicate Ecteinascidia turbinata, is being tested in phase II clinical trials in Europe and the United States of America (USA). Studies with different solid tumours have shown antitumour activity in advanced, pre-treated sarcomas as well as in drug-resistant breast and ovarian cancer. The primary mechanism of action for ET-743 has not been fully elucidated and different models have been suggested to explain its molecular mechanism of action. ET-743 binds tightly to the minor groove of DNA and previous data have suggested that ET-743 acts by interfering with RNA transcription. To further investigate the mechanism of in vitro drug resistance, we evaluated the gene expression profile in ovarian and chondrosarcoma cell lines selected for resistance to ET-743. We found 70 genes whose expression was modulated in both drug-resistant cell lines when compared with their respective parental drug-sensitive cell lines. This pattern of gene expression seems to be selective for ET-743-resistant cells, since ovarian cancer cells resistant to paclitaxel did not share the same gene expression changes. Data presented in this study reveal different molecular pathways that could be involved in the cellular mechanism of ET-743 resistance.

It was previously suggested that the peculiar mechanism of action of the novel anticancer drug Yondelis (ET-743, trabectedin) was due to part of the molecule, units A and B, binding to DNA in the minor groove, causing an alkylation at the N2 of guanine, while unit C protrudes out of DNA, possibly interacting with transcription factors or other DNA binding proteins. To test this hypothesis, we have compared the biological activity and the mode of action of Yondelis with its analogue ET-637, which has the same chemical structure except for the lack of the C ring. Yondelis and ET-637 showed similar cytotoxic potency and cell cycle perturbations. As already reported for Yondelis, the UV-96 cell line, deficient in ERCC-1, was less sensitive to ET-637 than the parental cell line. The binding of Yondelis or ET-637 to DNA-oligonucleotides was demonstrated by gel shift assay and SDS did not reverse the binding. Both compounds blocked the temperature-induced activation of the HSP40 promoter in the range of 1-10 nM. This study indicates that ET-637 acts similarly to Yondelis and demonstrates that the C ring of Yondelis may not be required for its biological activity.

The antineoplastic compound aplidine, a new marine-derived depsipeptide, has shown preclinical activity in vitro on haematological and solid tumour cell lines. It is currently in early phase clinical trials. The exact mechanism of action of this anticancer agent still needs to be clarified. We have previously reported that aplidine blocks the secretion of the angiogenic factor vascular endothelial growth factor (VEGF) by the human leukaemia cells MOLT-4, suggesting a possible effect on tumour angiogenesis. This study was designed to investigate the antiangiogenic effect of aplidine. In vivo, in the chick embryo allantoic membrane (CAM) assay, aplidine inhibited spontaneous angiogenesis, angiogenesis elicited by exogenous VEGF and FGF-2, and induced by VEGF overexpressing 1A9 ovarian carcinoma cells. In vitro, at concentrations achievable in the plasma of patients, aplidine inhibited endothelial cell functions related to angiogenesis. It affected VEGF- and FGF-2-induced endothelial cell proliferation, inhibited cell migration and invasiveness assessed in the Boyden chamber and blocked the production of matrix metalloproteinases (MMP-2 and MMP-9) by endothelial cells. Finally, aplidine prevented the formation of capillary-like structures by endothelial cells on Matrigel. These findings indicate that aplidine has antiangiogenic activity in vivo and inhibits endothelial cell functional responses to angiogenic stimuli in vitro. This effect might contribute to the antineoplastic activity of aplidine.

The mechanism by which aplidine, a marine natural product in early clinical development as an anticancer agent, induces cell growth inhibition and apoptosis has been investigated in the human leukemia cell line MOLT-4. This cell line is characterized not only by the ability to secrete VEGF, but also for the presence on its surface of the VEGF receptor-1 (VEGFR-1). Previous studies from our laboratory concerned with evaluating early changes in gene expression induced by aplidine in MOLT-4 cells have shown that the drug decreases the expression of VEGFR-1 (Marchini et al . Proc Am Assoc Cancer Res 2000; 41: 833). Here, we report the ability of aplidine to block the VEGF/VEGFR-1 loop. We found that aplidine blocked VEGF secretion that was temporally followed by a decrease in both VEGF and VEGFR-1 production. Aplidine did not directly affect either VEGF transcription or stabilization of its mRNA. Transfection of MOLT-4 cells with an antisense VEGF cDNA construct, resulted in inhibition of colony formations. One clone, transfected with sense VEGF cDNA, secreting 8–10 times more VEGF than parental cells, was less sensitive to aplidine-induced cytotoxicity and apoptosis than control cells. Moreover, addition of VEGF in the medium decreased the activity of aplidine in MOLT-4 cells. These data demonstrate that aplidine inhibits the growth and induces apoptosis in MOLT-4 cells through the inhibition of VEGF secretion which blocks the VEGF/VEGFR-1 autocrine loop necessary for the growth of these cells.

The cytotoxic effect of Aplidin was investigated on fresh leukaemia cells derived from children with B-cell-precursor (BCP) acute lymphoblastic leukaemia ( ALL) by using stromal-layer culture system and on four cell lines, ALL-PO, Reh, ALL/MIK and TOM-1, derived from patients with ALL with different molecular genetic abnormalities. In ALL cell lines Aplidin was cytotoxic at nanomolar concentrations. In the ALL cell lines the drug-induced cell death was clearly related to the induction of apoptosis and appeared to be p53-independent. Only in ALL-PO 20 nM Aplidin treatment caused a block of vascular endothelial growth factor (VEGF) secretion and downregulation of VEGF-mRNA, but Aplidin cytotoxicity does not seem to be related to VEGF inhibition since the sensitivity of ALL-PO cells to Aplidin is comparable to that observed for the other cells used. Aplidin induced a G(1) and a G(2)M block in ALL cell lines. In patient-derived leukaemia cells, Aplidin induced a strong cytotoxicity evidenced in a stroma-supported immunocytometric assay. Cells from children with genetic abnormalities such as t(9;22) and t(4;11) translocations, associated with an inferior treatment outcome, were sensitive to Aplidin to the same extent as that observed in other BCP-ALL cases. Aplidin exerted a strong cell killing effect (>88%) against primary culture cells from five relapsed ALL cases, at concentrations much lower than those reported to be achieved in plasma of patients receiving Aplidin at recommended doses. Taken together these data suggest that Aplidin could be a new anticancer drug to be investigated in ALL patients resistant to available therapy.

Yondelis™ (trabectidin, ET-743) is a marine natural product that has shown activity both in preclinical systems and in human malignancies such as soft tissue sarcoma and ovarian cancers that are resistant to previous chemotherapies. Molecular pharmacological studies indicated that Yondelis interacts with DNA and DNA repair systems in a way that is different from Cisplatin (DDP). The current study was designed to investigate the effects of the combination of Yondelis and DDP in human cancer cell lines and in xenografts derived from different tumours. The in vitro studies performed in human TE-671 rhabdomyosarcoma, Igrov-1 and 1A9 human ovarian carcinoma cell lines showed additive effects or slight synergism. Several human tumour xenografts, such as TE-671 rhabdomyosarcoma, SK-N-DX neuroblastoma, FADU head and neck, LX-1 non-small cell lung cancer (NSCLC), H-187 melanoma and SKOV HOC 8 ovarian carcinoma, showed an antitumour effect for the combination that was greater than that of each drug when given as a single agent. No consistent changes in the activity were observed if Yondelis and DDP were given 1 h apart in sequence or simultaneously. An orthotopically transplanted human ovarian cancer HOC 8 growing in the peritoneal cavity of nude mice was used that is insensitive to Yondelis alone and only moderately sensitive to DDP alone. The combination of the two drugs produced a dramatic increase of survival lasting several months. In conclusion, the combination of Yondelis and DDP is synergistic in vivo (i.e. the antitumour effect is greater than that of each drug used as a single agent at the maximum tolerated dose (MTD)) in different human tumour xenografts. The two drugs can be combined at the MTD of each drug, thus indicating there are no overlapping toxicities. These results provide a rationale for testing the combination of Yondelis and DDP in the clinic.

To determine the potential of aplidin as a cytotoxic agent in pediatric leukemia, we tested bone marrow (BM) and peripheral blood (PB) samples ( n =72) of children with different types of leukemia and healthy children in the methyl-thiazol-tetrazolium assay. Also, we compared these results with other cytotoxic drugs. Aplidin was cytotoxic in vitro at nanomolar concentrations, in a dose-dependent fashion. L-carnitine, that is applied in clinical studies to prevent myotoxicity caused by aplidin, had no effect on aplidin cytotoxicity in vitro . Aplidin cytotoxicity in vitro was not different when initial and relapsed acute lymphoblastic leukemia (ALL) or initial ALL and initial acute myeloid leukemia were compared. However, normal BM ( n =19) and PB ( n =13) cells were more resistant to aplidin than leukemic cells (median two- to seven-fold, P =0.001 and median four- to 11-fold, P<0.0001, respectively). In leukemia samples, no significant crossresistance between aplidin and other cytotoxic drugs was found, except for a trend for correlation with 2′,2′-difluorodeoxycytidine ( ρ =0.71, P =0.02). In normal BM samples, significant crossresistance with the epipodophyllotoxins was found, which is not readily explained by the currently known mechanisms of action of aplidin. In conclusion, we show that aplidin has selective cytotoxicity in vitro towards childhood leukemia cells and generally lacks crossresistance with other known cytotoxic drugs, which warrants clinical studies.

A combination therapy protocol using a marine chemotherapeutic and an antiangiogenic molecule was tested in a mouse tumor xenograft model for the ability to curtail the growth of a human chondrosarcoma (CHSA). Ecteinascidin-743 (ET-743), a marine-derived chemotherapeutic, was effective at slowing the growth of a primary CHSA. Plasminogen-related protein B, which antagonizes various endothelial cell activities, also elicited a significant inhibition of neoplastic growth, albeit with reduced effectiveness. The combination of the two agents resulted in only a modest further repression of tumor growth over that associated with ET-743 treatment alone, as measured by tumor volume (82% versus 76% inhibition, respectively). However, analysis of the extent of tumor necrosis and vascularization of the tumor revealed that the coadministration of the two compounds was clearly more effective, eliciting a 2.5-fold increase in tumor necrosis relative to single-agent treatment. The combination therapy also was most effective at antagonizing tumor-associated microvessel formation, as assessed by CD31 immunostaining, suggesting that combination therapy may hold promise for treating CHSA. Tumor necrosis produced by combination therapy of ET-743 and recombinant plasminogen-related protein B was also significantly greater than that produced by conventional doxorubicin treatment, further corroborating the efficacy of combination therapy.

Objective. In addition to neutropenias and/or thrombocytopenias as a short-term effect, antineoplastics also can produce long-term effects as a consequence of damage to the hematopoietic stem cells. The aim of the present study was to evaluate the toxicity of three marine-derived antineoplastics on murine hematopoietic stem cells. These antitumoral compounds currently are being evaluated in patients in phase 11 (aplidin and kahalalide F) and phase II/III (trabectedin) clinical trials.Materials and Methods. Long-term competitive repopulating assays were performed in mice to analyze toxic effects on the hematopoietic stem cells responsible for the multipotential long-term repopulation of hematopoiesis. Furthermore, granulocytic and T- and B-lymphoid lineages were studied, as well as myeloid (CFU-GM) and megakaryocytic (CFU-Meg) progenitors.Results. When cells were treated in vitro for 24 hours with CFU-GM IC50 dose of trabectedin (9.59 +/- 4.96 nM), no significant effects were observed in the stem cells. The dose of trabectedin that produced 90% of inhibition in CFU-GM (IC90: 23.71 +/- 1.27 nM) only inhibited 45% survival of stem cells. Doses of aplidin that produced reductions of 50% (56.9 +/- 13.32 nM) or 90% (195.88 +/- 21.39 nM) in myeloid progenitors did not show any effect on hematopoietic stem cells. Kahalalide F did not show any toxic effect in either short-term or long-term repopulating cells up to 10 muM.Conclusions. Our data show that the hematopoietic stem cells effects of antitumoral drugs can be properly characterized by the murine competitive repopulating assays. Our results suggest that long-term myelosuppression as a consequence of trabectedin, aplidin, or kahalalide F treatment would not be expected. (C) 2003 International Society for Experimental Hematology. Published by Elsevier Inc.

Ecteinascidine-743 (ET-743) and aplidine are two marine-derived antineoplastics currently in phase II development. With the aim of evaluating whether in vitro haematopoietic studies can predict the toxicity of these two drugs in patients, human bone marrow (BM) samples were incubated with these drugs under conditions which mimicked the administration exposures used in the clinics. As it was observed in different cancer cell lines, ET-743 was more toxic on an equimolar basis in human hematopoietic progenitors (inhibitory concentration reducing the viability to 50% after 24 h exposures; IC50(24h): 10-50 nM) compared with doxorubicin (IC50(24h) values: 280-460 nM), used as a control anticancer drug. In contrast to the high haematotoxic effects observed for ET-743, similar IC values were obtained for aplidine (IC50(24h): 150-530 nM) compared with doxorubicin. For both ET-743 and aplidine, the megakaryocytic progenitor was the most sensitive, compared with the other haematopoietic progenitors (IC50 values were 3- to 5-fold lower in the CFU-Megs compared with the CFU-GMs). The observation that the Cmax observed in patients treated with the aplidine maximum tolerated dose (MTD) (7.1 nM) was 21-75 fold lower than the IC50(24h) value observed for the different haematopoietic progenitors is highly consistent with the lack of haematotoxicity observed in patients treated with this drug. In the case of ET-743, differences between the Cmax value corresponding to the MTD (2.6 nM) and the in vitro IC50 values corresponding to the different progenitors were much lower (4-19-fold), also consistent with the haematotoxicity that was observed in patients treated at recommended doses (RDs) and MTDs. Although CFU-Megs were more sensitive than CFU-GM progenitors to ET-743 in vitro, clinical data showed that neutropenic events were more frequent than thrombocytopenic episodes. Aiming to further improve the predictive value of in vitro IC values corresponding to the different haematopoietic progenitors, additional refinement parameters derived from pharmacokinetic and animal studies are proposed.

This paper describes the current knowledge of the primary mode of action of a natural product, ecteinascidin 743 (ET-743), derived from the marine tunicate Ecteinascidia turbinata. ET-743 was initially selected for preclinical development because of its potent antitumor activity observed against several human solid tumor types. In vitro, the drug is cytotoxic in the nanomolar range, and in the case of some very sensitive cell lines, in the picomolar range. The large potency differences observed among several solid tumor types indicate that this compound possesses some tumor selectivity, but the molecular basis of these differential effects remains to be elucidated. The present studies were undertaken to evaluate the mechanism of action of ET-743 in this context. The available information on ET-743 binding to DNA and its effects on transcriptional regulation point to a unique behavior of this drug, as it independently affects specific gene transcription in a promoter-dependent way. In addition, ET-743 shows a peculiar pattern of selectivity in cells with different defects in their DNA-repair pathways. These results highlight a unique property of ET-743, possibly explaining why it possesses antitumor activity against tumors that are refractory to standard anticancer drugs, all of which certainly act by mechanisms that are different from that of ET-743.

Microarray technique was employed to study differences in gene expression profile induced by Aplidine treatment in the Molt-4 human leukemic T cell line. Aplidine is a novel marine compound purified from caribbean tunicate (sea squirt) Aplidium Albicans. Despite promising anti-tumor activity, few data are available on its mechanism of action. Exponentially growing cells were treated with Aplidine concentrations close to its 5IC50 for 1 hour and RNA samples collected after 0.5, 1, 6 and 24 hours of recovery in drug free medium. 32P labelled cDNAs were hybridized against Atlas Human Cancer arrays onto which 588 cDNAs were spotted. Genes involved in different cellular pathways, (such as growth factors, signal transduction or transcription factors) were found modulated by the drug. Even if the data obtained in the present study cannot be conclusive, several hypothesis on Aplidine's mechanism of action are indicated that will be the subject of future studies.