F14512, a polyamine-vectorized anti-cancer drug, currently in clinical trials exhibits a marked preclinical anti-leukemic activity.

Chemotherapy remains mainly used for the treatment of acute myeloid leukemia (AML). However, in the past 3 decades limited progress has been achieved in improving the long-term disease-free survival. Therefore the development of more effective drugs for AML represents a high level of priority. F14512 combines an epipodophyllotoxin core targeting topoisomerase II with a spermine moiety introduced as a cell delivery vector. The polyamine moiety facilitates F14512 selective uptake by tumour cells via the polyamine transport system, a machinery overactivated in cancer cells. F14512 has been characterized as a potent drug candidate and is currently in Phase I clinical trials. Here, we demonstrated marked survival benefit and therapeutic efficacy of F14512 treatments in a series of human AML models, established either from AML cell lines or from patient AML samples. Furthermore, we reported in vitro synergistic anti-leukemic effects of F14512 in combination with cytosine arabinoside (Ara-C), doxorubicin, gemcitabine, bortezomib or SAHA. In vivo combination of suboptimal doses of F14512 with Ara-C also resulted in enhanced anti-leukemic activity. We further showed that F14512 triggered both senescence and apoptosis in vivo in primary AML models, but not autophagy. Overall, these results support the clinical development in onco-hematology of this novel promising drug candidate.

A fluorescent biomarker of the polyamine transport system to select patients with AML for F14512 treatment.

The polyamine transport system (PTS), hyperactive in cancer cells, can constitute a gate to deliver F14512, a novel spermine epipodophyllotoxin conjugate recently selected for clinical development in AML phase I. We investigated in vitro the high antiproliferative effect of F14512 against 13 leukemia cell lines, and demonstrated a statistically significant correlation with the level of PTS activity, using a novel fluorescent marker F96982. This labelling protocol was then adapted for clinical applications for blood, bone marrow and AML samples with CD45 gating. Within the patient samples, the PTS activity varied significantly in AML cells, as compared to normal lymphocytes. In conclusion, the identification of PTS-positive AML with F98982 probe offers new perspectives to select patients prone to respond to F14512.

Description of two new human ovarian carcinoma models in nude rats suitable for laparoscopic experimentation.

BACKGROUND: Experimental laparoscopic trials require relevant models of ovarian carcinomatosis. METHODS: Female nude rats were inoculated intraperitoneally either with the IGR-OV1 or the NIH:OVCAR-3 human adenocarcinoma cell lines. Serial clinical checks and sacrifices were used to evaluate the rates of tumor take, survival, and patterns of tumor spread. Finally, laparoscopies with various pneumoperitoneum pressures were performed to verify the "surgical" relevancy of out models. The learning curve was measured. RESULTS: The best results were obtained when twenty-seven 106 IGR-OV1 cells and thirty-six 106 NIH:OVCAR-3 cells were injected in 28-day-old rats. The IGR-OV1 model provided a mean survival of 17.8 days (range, 13-22 days), with a high take rate (94%). The NIH:OVCAR-3 model resulted in a longer mean survival (59 days; range, 49-77) and also a high take rate (83%). The two models differed in their patterns of tumor spread: solid bulky omental metastasis having a diffuse microscopic peritoneal carcinomatosis with the IGR-OV1 line (the weight of the omental cake correlated significantly with the stage of development) and diffuse macroscopic peritoneal carcinomatosis having no large solid tumor, but visceral and paraaortic metastases, with the NIH:OVCAR-3 line. In both models, CA125 was high. Anesthesia could be performed and repeated in healthy and tumor-bearing rats. Laparoscopy was feasible, with pneumoperitoneum pressures as high as 8 mmHg lasting 1 h. Laparoscopy provided a reliable evaluation of the tumor spread into the peritoneal cavity. The plateau of the learning curve was soon obtained for take rate and survival after laparoscopy. CONCLUSION: We report two new human ovarian carcinoma xenografts in nude rats suitable for laparoscopy. The IGR-OV1 model mimics an advanced stage of the disease, and the NIH:OVCAR-3 model presents an earlier stage. These two models appear useful for experiments involving laparoscopy.

Marked antitumor activity of a new potent acronycine derivative in orthotopic models of human solid tumors.

S 23906-1 is a novel acronycine derivative selected on the basis of its potency in vitro. We investigated the antitumor activity of S 23906-1 against several murine transplantable tumors (C38 colon carcinoma, P388 leukemia, B16 melanoma, and Lewis lung carcinoma) and in orthotopic models of human lung (NCI-H460 and A549), ovarian (IGROV1 and NIH:OVCAR-3), and colorectal cancers (HCT116 and HT-29). Against established C38 colon carcinoma, S 23906-1 administered twice i.v. from 1.56-6.25 mg/kg markedly inhibited tumor growth. Treatment at the optimal dose (6.25 mg/kg) induced tumor regression in all of the mice. Acronycine was 16-fold less potent and only moderately active at the maximum tolerated dose, 100 mg/kg. Against other murine tumors of the former National Cancer Institute panel, S 23906-1 was either only moderately active or totally inactive. When evaluated in human orthotopic models, S 23906-1 given p.o. or i.v. demonstrated a marked antitumor activity against human carcinomas. In the two human lung cancer models, S 23906-1 increased the survival of the animals in a dose-dependent manner and induced treated versus control values of 162% (NCI-H460) and 193% (A549). Vinorelbine was less active, with treated versus control values of 119% and 174%, respectively. A significant survival benefit was also observed against the two i.p. ovarian tumors in which S 23906-1 was as active as paclitaxel, inducing 80% long-term survivors in the NIH:OVCAR-3 model. Lastly, S 23906-1 inhibited the growth of primary HT-29 and HCT116 colon tumors grafted onto the cecum as efficiently as irinotecan and eradicated the formation of lymph node, hepatic, and pulmonary metastases in the aggressive HCT116 model. The novel spectrum of activity of S 23906-1 compared with existing anticancer agents warrants further preclinical investigation.

Effect of pneumoperitoneum on circulating tumor DNA.

BACKGROUND: Release and circulation of tumor DNA could be favored by surgery. No data is available for the effect of laparoscopy on this phenomenon. MATERIAL AND METHODS: The aim of this study was to assess the impact of CO2 laparoscopy on circulating tumor DNA. Two xenografts of ovarian cancer were obtained by intraperitoneal inoculation (IP) of IGR-OV1 or NIH:OVCAR-3 cells in nude rats. CO2 laparoscopy (L), gasless laparoscopy (GL), midline laparotomy (ML) or general anesthesia as a control (C) were randomly carried out when the tumor graft was present in the peritoneal cavity. A sterile blood sample was taken in each case as soon as the experiment was completed. DNA was subsequently extracted and amplified (PCR, primers HLA GH 26 and HLA GH 27 specific for human DNA). In each model, we compared the influence of each surgical approach on circulating tumor DNA. Statistics were performed with the Wilcoxon test and Fisher exact test. 1: RESULTS: Eighteen rats were included in each group. Our protocol could detect an amount of tumor DNA equivalent to 10 cells/ml of blood. This technique was specific. Circulating tumor DNA was frequently observed in the IGR-OV1 model (45 to 50%), without significant difference between groups (p=0.6). In the NIH: OVCAR-3 model, the detection rate ranged from 22% (control group) to 64% (gasless group); but the overall comparison between the four groups was not significant (p = 0.2). CONCLUSION: In this experimental trial, CO2 laparoscopy had no deleterious effect on circulating tumor DNA. Biologic characteristics of tumors could also play a role.

Synthesis and cytotoxic and antitumor activity of benzo[b]pyrano[3, 2-h]acridin-7-one analogues of acronycine.

Benzo¿bacronycine (6-methoxy-3,3,14-trimethyl-3, 14-dihydro-7H-benzo¿bpyrano¿3,2-hacridin-7-one, 4), an acronycine analogue with an additional aromatic ring linearly fused on the natural alkaloid basic skeleton, was synthesized in three steps, starting from 3-amino-2-naphthalenecarboxylic acid (5). Eight 1, 2-dihydroxy-1,2-dihydrobenzo¿bacronycine esters and diesters (17-24) were obtained by catalytic osmic oxidation, followed by acylation. All these compounds were significantly more cytotoxic than acronycine, when tested against L1210 leukemia cells in vitro. The potency of the cyclic carbonate 24 was in the range of the most active drugs currently used in cancer chemotherapy. Two selected diesters (17 and 24) were evaluated in vivo against P388 leukemia and colon 38 adenocarcinoma implanted in mice. Both compounds were markedly active at doses 16-fold lower than the dose of acronycine itself. Against colon 38 adenocarcinoma, compounds 17 and 24 were highly efficient, inhibiting tumor growth by more than 80%. Diacetate 17 was the most active, inhibiting tumor growth by 96% at 6.25 mg/kg, with two of seven mice being tumor-free on day 43.

Histology and sensitivity to anticancer drugs of two human non-small cell lung carcinomas implanted in the pleural cavity of nude mice.

We have established two metastatic models of human non-small cell lung carcinoma (NSCLC)-the NCI-H460 large-cell carcinoma and the A549 adenocarcinoma-by inoculating tumor cells into the pleural space of nude mice. The objectives of this work were as follows: (a) to study the histological characteristics and growth and dissemination patterns of these tumors in nude mice; (b) to assess their sensitivity to drugs that have demonstrated significant clinical therapeutic effect in the treatment of NSCLC; and (c) to investigate the antitumor activity of S 16020-2, a new olivacine derivative, currently in Phase II clinical evaluation. In each of the two models, all animals developed lung tumors, resulting in 100% mortality. Histopathological study showed that these two tumors spread locally to contiguous structures, including the mediastinal pleura and diaphragm, with histological characteristics consistent with the human pathology. Anticancer drugs used for the treatment of NSCLC, such as cisplatin, doxorubicin, vinblastine, and etoposide, enhanced the life span of treated mice in the two models and were more active in the NCI-H460 than in the A549 model. The increases of survival time as compared to control groups were from 60 (P < or = 0.05) to 83% (P < or = 0.01) and from 21 to 40% for NCI-H460 and A549, respectively. Vinorelbine, paclitaxel, and irinotecan showed similar activities in the two models and increased the survival of treated mice by between 38 and 79% (P < or = 0.001) and between 58 (P < or = 0.01) and 78% in the NCI-H460 and A549 models, respectively. However, none of these drugs was curative, reflecting the resistance of this disease to chemotherapy. S 16020-2 exhibited a remarkable antitumor activity, increasing the survival by 82% (P < or = 0.01) for NCI-H460 and by 126% (P < or = 0.001) for A549. This drug was among the most active compounds in these models, thereby indicating its potential for the chemotherapy of this disease.

Biological and pharmacological characterisation of three models of human ovarian carcinoma established in nude mice: use of the CA125 tumour marker to predict antitumour activity.

In an attempt to improve the relevance of human tumour xenografts to the clinical situation, we have established 3 models of human ovarian carcinoma (IGROV1, A2780 and NIH:OVCAR-3) in nude mice in which progressive peritoneal carcinomatosis resulted in the death of tumour-bearing animals (median survival times: 32, 40 and 64 days, respectively). Histological analyses revealed both common and different characteristics in growth patterns and dissemination profiles. In each case, three stages of the disease were defined (early, intermediate and late). The antitumour activities of adriamycin, cisplatin, cyclophosphamide and paclitaxel were then compared when administered at the early stage where small multifocal tumour nodules were detectable in the peritoneal cavity of the animals. Significant antitumour activities of cisplatin and particularly paclitaxel were noted in terms of increase in survival time of the treated mice (T/C values for IGROV1, A2780 and NIH:OVCAR-3 respectively: 152%, 167%, and 187% for cisplatin and 211%, 179% and >283% for paclitaxel), paclitaxel being curative against the NIH:OVCAR-3 xenograft. These results reflect the high efficacy of these two drugs in the clinic in the treatment of ovarian carcinoma. The clinically used CA125 tumour marker, not detectable in healthy mice, was measured in the serum of mice bearing IGROV1 and NIH:OVCAR-3 tumours. CA125 serum levels increased as a function of time and were well correlated to disease progression. Moreover, treatment with cisplatin and paclitaxel led to significant decreases in these levels of between 58% and 100%. This human serum marker could be used to predict early on the efficacy of chemotherapy in these two models. In conclusion, the three experimental ovarian carcinomas possess several important characteristics of the human disease and may thus be used as a screen to select new antitumour drugs potentially active in this pathology.

Synthesis of 9-O-substituted derivatives of 9-hydroxy-5, 6-dimethyl-6H-pyrido[4,3-b]carbazole-1-carboxylic acid (2-(dimethylamino)ethyl)amide and their 10- and 11-methyl analogues with improved antitumor activity.

Analogues of the antitumor drug S 16020-2 modified at the 9, 10, or 11 position were synthesized and evaluated in vitro and in vivo on the P388 leukemia and B16 melanoma models. Starting from 9-methoxy-5, 11-dimethyl-6H-pyrido[4,3-b]carbazole-1-carboxylic acid ethyl ester, the 11-CH3 analogue of 9-hydroxy-5,6-dimethyl-6H-pyrido[4, 3-b]carbazole-1-carboxylic (2-(dimethylamino)ethyl)amide (1), compound 4, was synthesized using a four-step sequence, whereas its 10-CH3 analogue 5 was prepared using a two-step pathway, starting from compound 1. Finally starting from the 9-OH compounds 1, 4, and 5, a series of variously 9-O-substituted derivatives were synthesized. In these series, the most active compounds resulted from esterification of the 9-OH group with various aliphatic diacids, which led to 9-O-CO-( )-COOH derivatives of 1, 4, and 5. For these compounds, the number of long-term surviving mice obtained at the optimal dose were 60-100% in the ip/iv P388 leukemia and 10-35% in the ip/ip B16 melanoma, corresponding to an improved therapeutic index with respect to 1 and 4. This high antitumor activity, with curative examples in both models, was not due to a higher cytotoxicity since these compounds were equally or slightly less potent in vitro than 1 and 4. The most active compounds were thus selected for further in vivo evaluation.

Synthesis, properties and biological evaluation of substituted furo[3,2-e] and pyrano[3,2-e]pyrido[4,3-b]indoles.

Furo[3,2-e]- and pyrano[3,2-e]pyrido[4,3-b] indoles were synthesized from 1,4,5-trisubstituted 8-hydroxy-5H-pyrido[4,3-b]indoles. The intermediates, 10-chloro-6H-furo[3,2-e]pyrido[4,3-b]indole (11), 10-chloro-2,6-dihydro-1H-furo[3,2-e]pyrido-[4,3-b]indole (10) and 11-chloro-2,3-dihydro-3H,7H-pyrano[3,2-e]pyrido[4,3-b]indole (15), were substituted by diamines under thermal conditions (180 degrees C). In contrast, 11-chloro-3H,7H-pyrano[3,2-e]pyrido[4,3-b]indole (14), 9-allyl-1-chloro-4,5-dimethyl-5H-pyrido[4,3-b]indole (9a) and 8-propargyloxy-4,5-dimethyl-5H-pyrido[4,3-b]indole (8) led mainly to 1-aminosubstituted 8-hydroxy-5H-pyrido[4,3-b]indole derivatives resulting from an unexpected C3 unit elimination. When examined in three tumour cell lines (L1210 leukaemia, the B16 melanoma and the MCF7 breast adenocarcinoma) the new amino substituted furo[3,2-e]-, dihydrofuro[3,2-e]- and dihydropyrano[3,2-e]-pyrido[4,3-b]indole derivatives revealed cytotoxic properties, especially important for the 2,6-dihydro-1H-furo[3,2-e]pyrido[4,3-b]indole series. The most active compound (12b) significantly inhibits both DNA topoisomerases I and II, and is as potent as Adriamycin at inhibiting cell proliferation and inducing a massive accumulation of L1210 cells in the G2 + M phase of the cell cycle. However, 12b was less active than Adriamycin when tested in vivo against P388 leukaemia or the B16 melanoma tumour models.

Eriochrome Black T, structurally related to suramin, inhibits angiogenesis and tumor growth in vivo.

The polyanionic species suramin is a potential anti-cancer agent of narrow therapeutic index. Among other pharmacological characteristics, suramin is an inhibitor of angiogenesis. We have targeted its angiostatic properties as part of a program to discover less toxic analogs. From screening a series of commercially available compounds, structurally related to suramin and containing a sulfonic acid substituted naphthylamine moiety, we discovered a new lead, Eriochrome Black T (EBT). EBT is a novel inhibitor of angiogenesis, more potent and less toxic than suramin in the chick chorioallantoic membrane assay. EBT was more active than suramin in inhibiting endothelial cell proliferation in primary culture and in inhibiting proliferation of three tumor cell lines, A431, L1210 and M5076 (IC50 10-100 microM). Cell cycle studies on the A431 line showed that both EBT and suramin caused an accumulation of cells in the S phase, EBT being 10-fold more potent. We suggest that this cell cycle perturbation is linked to inhibition of topoisomerase II catalytic activity. EBT was found to be a moderate but significant inhibitor of matrix metalloproteinases (10 microM range), more efficient than suramin. In a s.c. M5076 sarcoma model in mice, EBT had similar efficacy to suramin both by the i.p. or s.c. route and was moreover better tolerated. Combined pharmacological results show that EBT compared favorably with suramin in all assays, and that in ovo and in vivo, EBT is an analog of suramin with diminished toxicity.

Synthesis and antitumor activity of the metformin platinum (IV) complex. Crystal structure of the tetrachloro(metformin)platinum (IV) dimethylsulfoxide solvate.

The synthesis of (metformin) tetrachloroplatinum (IV) was investigated (metformin is N,N-dimethylbiguanide). It crystallizes with one dimethylsulfoxide molecule as solvate in the monoclinic system, space group P2(1)/n (No. 14) with Z = 4. The cell dimensions are: a = 13.136(7), b = 9.424(2), c = 14.009(8) A, beta = 111.96(4) degrees, V = 1608.4(2) A3. Of the 4269 independent nonzero reflections collected, 1979 with I > 3 sigma (I) were considered and used in the calculations. The structure was refined to R = 0.043 and wR = 0.045. The platinum coordination is octahedral, built up from four chloride anions and one bidentate chelating ligand via the two imine nitrogen atoms cis position. The distances and angles are typical of six-membered rings that have similar donor atoms. The complex was evaluated in vitro and in vivo on murine P388 leukemia. It was found to be as potent as cis-dichlorodiammine platinum (II), CDDP, in inhibiting the proliferation of the sensitive P388 cells. However the resistant P388/CDDP cells were threefold more sensitive to the compound than to CDDP. The two compounds induced a similar perturbation in the G2+M phases of the cell-cycle. The complex was less active than CDDP in vivo on P388 leukemia when administered i.p. (intra peritoneal) on day 1.

Antitumor activity of S 16020-2 in two orthotopic models of lung cancer.

S 16020-2, a new olivacine derivative selected on the basis of its cytotoxicity in vitro and antitumor activity in vivo, was evaluated against the human A549 and the murine Lewis lung tumor models implanted s.c. and i.v. Against Lewis lung carcinoma implanted s.c., S 16020-2 was found to be curative, with an activity and therapeutic index (Ti = 4) similar to that of cyclophosphamide. S 16020-2 administered weekly demonstrated a high therapeutic efficacy against A549 non-small cell lung carcinoma implanted s.c. in nude mice and induced tumor regression at 80 mg/kg. When A549 tumor cells were injected i.v. in SCID mice, experimental metastases rapidly developed and the progressive invasion of the lung tissue by tumor preceded the death of animals. In this model, S 16020-2 administered at 40 mg/kg i.v. following an early (days 8, 18 and 28) or delayed (days 20, 30 and 40) treatment schedule prolonged the survival of tumor-bearing mice with T/C values of 150 and 145%, respectively. Against the i.v. Lewis lung carcinoma, S 16020-2 was also highly active since when administered at 60 mg/kg on days 5, 9 and 13 it totally inhibited tumor growth and cured up to 89% of mice. When administered on days 11, 15 and 19 to animals with established tumors, S 16020-2 was still active but not curative. In the presented studies, S 16020-2 antitumor activity was superior to that of adriamycin and comparable or superior to cyclophosphamide (used as reference compounds). Our results demonstrate the efficacy of S 16020-2 against these highly aggressive and chemoresistant tumor models.

Experimental antitumour activity of S 16020-2 in a panel of human tumours.

The antitumour activity of S 16020-2, a new topoisomerase II inhibitor, was evaluated in comparison with doxorubicin against 13 human tumours, including colon (HT-29, Colo320DM), breast (MCF7, MDAMB-231), ovary (SK-OV-3, A2780, NIH:OVCAR-3), non-small cell lung (NCI-H460, A549, Calu-6, NCI-H125) and small-cell lung (NCI-H69, SCLC6) cancers. S 16020-2 was administered weekly intravenous within a dose range of 20-90 mg/kg for 3 weeks. Antitumour responses were obtained in all the tumour types tested except in the two colon cancers. S 16020-2 produced significant growth delays in nine tumour models and induced regressions of all A549 lung tumours. The antitumour activity of S 16020-2 was superior to that of doxorubicin against the NCI-H460, A549, NCI-H69, SCLC6 and NIH:OVCAR-3 xenografts. These results demonstrate the broad spectrum of antitumour activity of S 16020-2 in a large panel of in vivo experimental models and confirm its interest as a potential agent in the treatment of malignant disease.

Synthesis and cytotoxic and antitumor activity of esters in the 1,2-dihydroxy-1,2-dihydroacronycine series.

Seven 1,2-dihydroxy-1,2-dihydroacronycine and 1,2-dihydroxy-1,2-dihydro-6-demethoxyacronycine esters and diesters were synthesized via osmic oxidation of acronycine or 6-demethoxyacronycine followed by acylation. The 6-demethoxyacronycine derivatives were found to be inactive, whereas in contrast, all of the acronycine derivatives were more potent than acronycine itself when tested against L1210 cells in vitro. Four selected acronycine derivatives (17,19, 21, and 22) were evaluated in vivo against murine P388 leukemia and colon 38 adenocarcinoma implanted in mice. All compounds were markedly active against P388 at doses 4-16-fold lower than acronycine itself. Against the colon 38 adenocarcinoma, the three compounds 17, 21, and 22 were highly efficient. 1,2-Diacetoxy-1,2-dihydroacronycine (17) was the most active, all the treated mice being tumor-free on day 23.

Synthesis and biological evaluation of 6-(9-hydroxy-5-methyl (and 5,6-dimethyl)-6H-pyrido[4,3-b]carbazol-1-yl)picolinic amides as new olivacine derivatives.

Starting from 2-(6-methoxy-1-methylcarbazol-2-yl)ethylamine and diethyl-2,6-pyridine dicarboxylate, the title compounds were obtained through five or six steps. The new compounds retained significant cytotoxicity towards various tumor cell lines, but in vivo studies on murine P388 leukemia, B16 melanoma and Lewis lung carcinoma showed a lowered antitumor activity with respect to that of the related olivacine lead compound 1.

New purines and purine analogs as modulators of multidrug resistance.

A series of 36 purine and purine analog derivatives have been synthesized and tested for their ability to modulate multidrug resistance in vitro (P388/VCR-20 and KB-A1 cells) and in vivo (P388/VCR leukemia). Compounds were compared to S9788, a triazine derivative which has already shown some activity during phase 1 clinical trials and also a limiting cardiovascular side effect possibly linked to its calcium channel affinity. The fact that active compounds increase adriamycin accumulation in the resistant KB-A1 cells, and not in the sensitive KB-3-1 cells, suggests they act predominantly by inhibiting the P-glycoprotein-catalyzed efflux of cytotoxic agents. No direct relation was found between the affinity for the phenylalkylamine binding site of the calcium channel and in vitro sensitization of resistant cells. In vivo, when administered po in association with vincristine (0.25 mg/kg), five compounds (3, 4, 9, 25, and 26), of very differing calcium channel affinities (Ki from 5 to 560 nM), fully restored (T/V > or = 1.4) the sensitivity of P388/VCR leukemia to vincristine.

In vivo antitumor activity of S 16020-2, a new olivacine derivative.

The antitumor activity of S 16020-2, a new olivacine derivative, was investigated in vivo and compared with that of Adriamycin and elliptinium acetate in a panel of murine (P388 leukemia, M5076 sarcoma, Lewis lung carcinoma, and B16 melanoma) and human (NCI-H460 non-small-cell lung and MCF7 breast carcinomas) tumor models. S 16020-2 given i.v. was active against P388 leukemia implanted i.p., s.c., or intracerebrally. The therapeutic effect of an intermittent schedule (administration on days 1, 5, 9) was superior to that of single-dose treatment, allowing the i.v. administration of high total doses of S 16020-2 and resulting in the cure of 60% of mice in the i.p. P388 model. In this model, S 16020-2 was more active than elliptinium acetate and showed a better therapeutic index than Adriamycin: > or = 8 versus 2. A good therapeutic effect of S 16020-2 was also observed in three P388 leukemia sublines displaying the classic multidrug-resistance phenotype, namely, P388/VCR, P388/VCR-20, and P388/MDRC.04, the latter being totally insensitive to vincristine and Adriamycin. However, S 16020-2 was not active against the P388/ADR leukemia, a model highly resistant to adriamycin in vivo. S 16020-2 was both more active than Adriamycin and curative in the M5076 sarcoma and Lewis lung carcinoma implanted s.c. In the B16 melanoma implanted i.p. or s.c., S 16020-2 was less active than Adriamycin. Against the NCI-H460 human tumor xenograft, S 16020-2 demonstrated activity superior to that of Adriamycin (T/C = 20% versus 43% on day 21). Against the MCF7 breast cancer xenograft, S 16020-2 was active, but less so than Adriamycin (T/C = 23% versus 9% on day 21), whereas elliptinium acetate was marginally active (T/C = 49% on day 24). The hematological toxicity of S 16020-2 given to B6D2F1 mice at pharmacological dose appeared to be less severe than that of Adriamycin, particularly in bone-marrow stem cells. These results demonstrate that S 16020-2 is a highly active antitumor drug in various experimental tumor models and is markedly more efficient than elliptinium acetate. Because of its pharmacological profile, which is globally different from that of Adriamycin, S 16020-2 is considered an interesting candidate for clinical trials.

Synthesis and evaluation of 9-hydroxy-5-methyl-(and 5,6-dimethyl)-6H-pyrido[4,3-b]carbazole-1-N- [(dialkylamino)alkyl]carboxamides, a new promising series of antitumor olivacine derivatives.

Starting from 2-(2-aminoethyl)-6-methoxy-1-methylcarbazole, ethyl 9-methoxy-5-methyl-6H-pyrido[4,3-b]carbazole-1-carboxylate was obtained through a three-step sequence. This compound and its 6-methyl derivative react with (dialkylamino)alkylamines to provide various 9-methoxy-5-methyl-6H-pyrido[4,3-b]carbazole-1-(N-substituted carboxamides) whose boron tribromide demethylation afforded corresponding 9-hydroxy-1-(N-substituted carbamoyl)-olivacines. The same pathway but starting from 2-(2-aminoethyl)-6-methoxy-1,4-dimethylcarbazole led to ethyl 9-methoxy-5,11-dimethyl-6H-pyrido[4,3-b]carbazole-1-carboxylate which did not normally react with amines. It provided either the recovered starting material at 120 degrees C or 9-methoxyellipticine resulting from an unexpected decarboethylation in a steel vessel at 180 degrees C. Biological testing of the newly obtained 1-carbamoylolivacine derivatives showed that 9-hydroxylated compounds displayed high cytotoxicity for cultured L1210 and colon 38 cells (IC50 range 5-10 nM) and good antitumor activity in vivo in the P388 leukemia and colon 38 models when administered by the iv route. The most active compound in these series is 9-hydroxy-5,6-dimethyl-1-[N-[2-(dimethylamino)ethyl]carbamoyl]-6H- pyrido[4,3-b]carbazole which was selected for further evaluation on murine solid tumors and for toxicological studies.