Preclinical Antitumor Activity of a New Vinca Alkaloid Derivative, S … · [CANCER RESEARCH 51....

[CANCER RESEARCH 51. 2312-2318. May 1, 1991]

Preclinical Antitumor Activity of a New Vinca Alkaloid Derivative, S 12363Alain Pierre, Laurence Kraus-Berthier, Ghanem Atassi, Suzy Gros, Marie-France Poupon, Gilbert Lavielle,Maryse Berlion, and Jean-Pierre Bizzari1

InstituÃde Recherches Servier, Division de CancÃ©rologieExpÃ©rimentale,II rue des Moulineaux, Suresnes, France [A. P., L. K-B., G. L.J; UniversitÃ©Libre de Bruxelles,Laboratoire de Pharmacologie, Boulevard du Triomphe, 1050 Bruxelles, Belgique Â¡G.A.]; CNRS, Laboratoire de Pharmacologie et de Toxicologie Fondamentales, 205roule de Narbonne, 31078 Toulouse, France [S. CJ; CNRS, Institut de Recherches sur le Cancer, 7 rue Guy Moquet, 94802 Villejuif, France [M-F. P.]; Institut deRecherches Internationales Servier, 6 place des PlÃ©iades,92415 Courbevoie Cedex, France Â¡M.B., J-P. B.]

ABSTRACT

S 12363 is a new l ineu alkaloid derivative obtained by appending anoptically active a-aminophosphonate at the ( '... position of O*-deacetyl

vinblastine. S 12363 was evaluated for cytotoxic and antitumor activityagainst a spectrum of murine and human tumors. This compound was,respectively, on average, 72- and 36-fold more cytotoxic than wereMin-rÃ¬simi-and vinblastine, when tested on a panel of 2 murine and 37

human tumor cell lines using the microculture tetrazolium assay. S 12363exhibited significant antitumor activity against murine transplantabletumors (i.p. and s.c. P388 leukemia, i.p. I 121(1leukemia, i.p. and i.v.B16 melanoma, i.p. M5076 sarcoma, and s.c. colon adenocarcinoma 38),while no activity was observed on s.c. Lewis lung carcinoma. S 12363,when administered i.p., showed moderate activity on human NCI-H460lung and PANC-1 pancreas tumor xenografts in nude mice. However,when it was administered i.v., it exerted a significant activity againsthuman MT-29 colon, NCI-H460 lung, NC1-H125 lung, PANC-1 pancreas, and A-431 vulvar tumor xenografts. S 12363 was also active invivo against a P388 leukemia subline resistant to vincristine. On the invivo panel of tumors used in this study, S 12363 was at least as active asreference compounds, while its optimal dosage was 10- to 40-fold lowerthan that of vinblastine, depending on the models studied. The effects ofschedule and route of administration on the antitumor activity of S 12363were studied in both i.p. inoculated P388 leukemia and B16 melanoma,in which the activity was improved by single and intermittent treatment(Days 1, 8, and 15) and i.p. route. S 12362, which differs only by theconfiguration of the asymmetric carbon atom of the side chain, was 300-fold less cytotoxic and 1000-fold less potent in vivo than was S 12363.These results suggest that S 12363 could present a therapeutic advantageover its congeners and deserves further pharmacological evaluations.

INTRODUCTION

Vinca alkaloids are an important class of antitumor agents,widely used in combination chemotherapy regimens for thetreatment of leukemia and solid tumors. There has been muchresearch on this class of agents, which led initially to VBL2 and

following minor modifications to this molecule to VCR andVDS which have very different clinical behavior (1, 2). Recently,chemists have grafted amino acid derivatives onto vinblastine,so as to facilitate transport of these large molecules. This workled to the synthesis of molecules with interesting pharmacological properties, namely vinglycinate (3) and vintryptol (4), thatare structural analogues of vindesine. Taking into considerationthe possibility of drastically modulating the toxicity and anti-tumor activity of these compounds by introducing relativelyslight structural modifications, we synthesized a series of a-aminophosphonate derivatives of vinblastine (5). The most

Received 11/9/90: accepted 2/19/91.The costs of publication of this article were defrayed in part by the payment

of page charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

' To whom requests for reprints should be addressed.2The abbreviations used are: VBL. vinblastine sulfate; VCR. vincristine sulfate;

VDS. vindesine sulfate; ADM, doxorubicin; MTA, microculture tetrazoliumassay; MST. median survival time; MTW, median tumor weight; SGD. specificgrowth delay; NCI, National Cancer Institute; DCT, Division of Cancer Treatment; IC50.50% inhibitory concentration; T/C. treated/control; LD,0, 10% lethaldose; TI, therapeutic index.

active compound of this series, called S 12363, consists of anaminophosphonate, a bioisoster of the amino acid valine,grafted onto the vindoline moiety.

In this paper, we present the in vitro cytotoxic activity andin vivo antitumor effect of this new Vinca alkaloid derivative,S 12363, which has been tested both according to the usual andthe most recent strategy (called disease-oriented strategy)adopted by the NCI (6-8). This work points to the high potencyand stereospecificity of this compound compared with its epi-mer S 12362, which differs only by the absolute configurationof the asymmetric carbon atom of the side chain.

MATERIALS AND METHODS

Drugs. S 12363 is (15)-l-[3-(O"-deacetyl-3-demethoxycarbonyl vin-caleukoblastinyl)carbonylamino]-2-methyl-propylphosphonic acid die-thylester sulfate, and S 12362 is the (Itf)epimer. S 12363 and S 12362were synthesized as described (5). These 2 compounds were separatedby high-performance liquid chromatography and were approximately98 to 99% pure. VCR sulfate was purchased from Roger BelIon(France), and VBL sulfate and VDS were from Eli Lilly (France). Thestructural formula of S 12363 is presented in Fig. 1.

Mice and Tumors. Inbred and hybrid mice, weighing 20 to 23 g, wereobtained from Charles River, France. They were provided with foodand water ad libitum. Athymic nude mice (5 to 6 wk old, Swissbackground) were obtained from IFFA CREDO, France, and weremaintained in macroisolation (ISO CONCEPT, France). Cages, water,and food were sterilized by paracetic acid.

The murine tumors, Â¡.p.P388 leukemia and LI210 leukemia, wereprovided by Mario Negri Institute (Milan, Italy). B16 melanoma wasobtained from the DCT Tumor Repository (NCI, Frederick, MD).Colon adenocarcinoma 38, Lewis lung carcinoma, M5076 sarcoma,and s.c. P388 leukemia were supplied by the National Cancer Institute(Bethesda, MD).

The human tumor cell lines HT-29, PANC-1, and A-431 wereobtained from the American Tissue Culture Collection (Bethesda, MD);the cell lines NCI-H460 and NCI-HI25 were provided by the DCTTumor Repository. These human cell lines were grafted into nude mice(IO7 cells grafted s.c.) and used for chemosensitivity assays after 2 to 3

passages.In Vitro Studies. All cell lines were from the American Tissue Culture

Collection except NCI-H460, NCI-H125, and B16 which were providedby the DCT Tumor Repository. Tumor cells were cultivated in RPMI1640 medium supplemented with 10% fetal calf serum (Gibco), 2 mMi-glutamine, 100 units/ml of penicillin, 100 Mg/ml of streptomycin, and10 mM 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid buffer, pH7.4. The MTA was used as previously described (6, 9) with minormodifications. In brief, adherent tumor cells were plated into 96-wellmicrotiter plates and allowed to adhere for 48 h before drug addition,while nonadherent cells were distributed just before the experiment.

The cell densities were selected in order to maintain the cells in anexponential phase of growth and to obtain a linear relation betweenabsorbance and cell number. Cells were then incubated for l h in thepresence of graded concentrations of drugs (9 concentrations in triplicate). The cells were washed and then incubated in drug-free mediumfor about 4 doubling times limited to 7 days for slowly growing cells.The cell densities used are listed in Table 1. At the end of the incubationtime, 3-(4,5-dimethylthiazol-2y)-2,5-diphenyl-2//-tetrazolium bromide

2312

on March 14, 2021. © 1991 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

ANTITUMOR ACTIVITY OF S 12363

**-./

CH30

xCHv. H

CH3 CH3

Fig. 1. Structural formula of S 12363.

(0.5 mg/ml; Sigma) was added, and cells were incubated for 4 h at37Â°C.The plates were centrifuged (1600 x g) for 5 min, and culture

medium supernatant was removed from wells by slow aspiration andreplaced with 100 M'of dimethyl sulfoxide. The plates were shaken andread in a MCC Multiskan spectrophotometer (Flow Laboratories) at540 nm. Results are expressed as IC50 (mean Â±SEM), the drugconcentration reducing by 50% the absorbance in treated cells, withrespect to untreated cells.

In Vivo Studies. VCR, VDS, and VBL were solubilized and dilutedin saline solution, and S 12363 was solubilized in absolute ethanol andthen diluted in saline solution. The drugs were administered in thisvehicle by the i.p., i.V.,or p.o. (by gavage) routes.

To evaluate the sensitivity of transplantable murine tumors toS 12363, C57BL/6 x DBA/2F, (hereafter called B6D2F,) mice wereinoculated with IO6P388 leukemia cells i.p., IO5L1210 leukemia cellsi.p., 0.5 ml of 1:10 B16 melanoma brei i.p., or 5 x IO4B16 melanomacells i.V., or the mice were grafted s.c. with 70-mg fragments of colon38 carcinoma. C57BL/6 mice were implanted s.c. with 1-^1 fragmentsof Lewis lung carcinoma. BALB/c x DBA/2F, (hereafter called CD2F,)mice were implanted s.c. with IO6 P388 leukemia cells. C57BL/6 xC3HF] (hereafter called B6C3F.) mice were inoculated i.p. with IO6

M5076 sarcoma cells. Experiments and antitumor testing were conducted according to published NCI protocols (10, 11). For Â¡.p.and s.c.P388, L1210, i.p. B16, and M5076 tumors, antitumor activity wasdetermined by comparing the MST of the treated group with that ofthe control group, and results are expressed as T/C where

T/C = MST of treated gro"P x ,00MST of control group

For Lewis lung and C38 tumors, the MTW of the treated group wascompared with the MTW of the control group at the end of theexperiment (Day 20).

To evaluate the effect of S 12363 on the occurrence of artificialmÃ©tastasesin the i.v. B16 melanoma model, mice were sacrificed onDay 20 after tumor cell inoculation, and lungs were excised andexamined using an image analysis system (VIDS IV; SystÃ¨mesAnalytiques, France). The mÃ©tastaseswere counted on the pulmonary surface,and the result was expressed as the percentage of diminution in thenumber of lung mÃ©tastasesin treated animals with respect to controls.

For evaluation of the sensitivity of human tumor xenografts to S12363 and reference compounds, 2- to 3-mm1 fragments were im

planted s.c. into the right flank of nude mice (tutmi), and in situmeasurements of the length and width were done twice a week followingNCI protocols (12). Treatment started when the tumor reached 50mm3, and tumors volumes were estimated by the formula length (mm)x width (mm)2/2. The relative tumor volume was expressed as the VJ

Vuindex, where K, is the tumor volume on a given day of measurement,and V0 is the initial volume of the same tumor at the start of thetreatment. For calculations of the T/C percentage, the following formula was applied at each day of tumor measurements.

T/C (%) = mean VJV0 treatedmean VJV0 control

x 100 (13)

The lowest (i.e., the optimal) T/C within 5 wk after the last injectionwas selected. Specific growth delay was calculated using the formula

SCO =treated - TÂ¿control

TÂ¿control

where TÂ¿is tumor doubling time. The number of control and treatedmice varied, depending on the tumor model and the size of the experiment (5 to 10 for the treated group and 10 to 30 for control groups).In each experiment, S 12363 was evaluated at several dosage levels andwas administered according to the defined schedule (Table 2). Tumor-bearing control mice were given injections of the vehicle only (0.9%NaCl solution). Activity of S 12363 was compared with that of referencecompound VBL, VCR, or VDS.

The effects of schedule and route of administration on the antitumoractivity of S 12363 were studied in mice bearing P388 leukemia andB16 melanoma, both inoculated i.p. S 12363 was administered according to the schedules and routes listed in Table 3.

Cross-resistance. Cross-resistance to S 12363 was evaluated in vitroand in vivo using P388 sublines resistant to VCR (P388/VCR) andADM (P388/ADM). These sublines were obtained from the JulesBordet Institute (Bruxelles, Belgium).

In this experiment, cytotoxic activity was evaluated by determinationof cell number with an electronic particle counter (Model ZBI; CoulterCoultronics) after 48 h of exposure to graded concentrations of drugs.Results were expressed as IC50(mean Â±SEM).

Evaluation of cross-resistance in vivowas conducted in CD2Fi mice,weighing 18 to 22 g, implanted i.p. with IO6leukemia cells. Mice were

observed for life span; MST and T/C were evaluated.

RESULTS

Cytotoxic Activity for Tumor Cells. Table 1 shows the sensitivity of the 2 murine and the 37 human tumor cell lines tested,i.e., 9 leukemia, 8 lung cancer, 5 colon cancer, 2 kidney cancer,2 ovarian cancer, 5 breast cancer, 2 melanoma, and 4 miscellaneous cancer cell lines. The mean (Â±SEM) of the IC50s foreach tissue was calculated, showing a better activity of S 12363against leukemia, lung, breast, ovarian, and melanoma tumorsthan against colon or kidneys tumors. S 12363 is a highlypotent cytotoxic drug. The mean of all the IC50 values (humancells) is 35 nivi, a very low value if we consider that the cellswere exposed to the drugs for only 1 h. On average, S 12363 is72- and 36-fold more cytotoxic than are VCR and VBL, respectively. Results presented in Table 1 show that no directrelation exists between the rate of proliferation (i.e., the doubling time) and the sensitivity to S 12363. Nevertheless, S12363 is cytotoxic for some cell lines, having a slow rate ofproliferation (e.g., CaLu-3, Caov-3, all the breast cell lines).

We calculated, for a given compound and for each cell line,the difference between the mean of the logarithm of all the IC50and the log of the IC50. These values indicating the relativesensitivity of a cell line to the drug, with respect to the meansensitivity, were displayed according to the method of Paull etal. (14). The comparison of the patterns obtained with the threedrugs showed that the pattern of differential growth inhibitionof S 12363 resembled that of VBL and was very different fromthat of VCR (data not shown).

Antitumor Activity in the Murine Tumor System. The toxicityof the compounds was first evaluated in healthy mice, after one

2313




Table I Cylotoxic activity of S 12363 and reference compounds on murine and human tumor cells (MTA, l-h contact)

CelllinesMurine

L1210B16HumanLeukemia

HL-60K562RPMI

8226CESSCCRF-CEMCCRF-SBMOLT-4U-937MC

116Lung

CaLu-1CaLu-3CaLu-6A549A-427NCI-H125NCI-H460SK-MES-1Colon

HT-29WiDrCOL0320

DMSW480SW620Kidney

A-498A-704Ovary

Caov-3SK-OV3Breast

Hs 578TZR-75-1ZR-75-30BT-20MCF-7Melanoma

SK-MEL-1SK-MEL-2Miscellaneous

A-431PANC-1T24U-20SDoublingtime

(In92044234742322731194427803520442721442530263025391407844351038112070387236222327No.ofcells/well1,500750,500,500,500,500.500.500.500,500,5007503,0003003007501.5004503003006009006006001503,0007503001.5001,5003.0001.5001.5003,0001.5007501,500150750VCR121.3

+40.4354.8Â±76.527.1

Â±3.745.5Â±12.228.5Â±7.239.2Â±18.4153.3Â±41.212.9

Â±2.723.0Â±4.318.9

Â±0.333.0Â±8.142.4

+14.2*28.5'2.166.4

Â±848.8342.6Â±34.7412.4Â±

104.05,653.7Â±500.871.0

Â±9.589.7Â±23.11.057.1

Â±378.5799.7Â±208.21,324.1

Â±664.0*606.0'214.5

Â±20.5231.5+26.756,552.0Â±15.841.5239.8Â±48.2288.9

Â±41.111,505.4+11,261.7*239.8'10,420.8

+3,860.650.000.0+0.030,210.4'90.1

+24.32,226.4Â±626.11,058.3'237.8

Â±36.4173.0Â±19.0225.5

+44.0336.1Â±58.8140.1

Â±21.2222.5Â±33.5*225.5'122.2

+24.052.4Â±14.687.3'68.4

Â±10.15.939.7Â±853.91.327.5+317.72,4

54.6Â±910.02447.6Â±1,262.0*1,891.0'1CÂ«,

(nM)Â°VBL180.5

Â±31.8193.3+52.450.7

Â±12.2272.2Â±94.9320.7+

121.9161.1Â±22.4912.0

+261.140.5Â±13.7185.9

Â±48.283.4Â±2.8342.7Â±122.3263.2Â±89.3189.9661.

8 Â±278.4149.0+25.9179.0Â±

56.0476.2Â±21.071.5

Â±21.8112.7Â±37.6340.4

Â±107.5235.0Â±60.9278.2Â±71.8207.0215.3

Â±58.8228.7+54.33,577.3Â±673.3622.7Â±238.7585.4+

156.31,045.9+638.6585.41,556.3

+539.49,593.0Â±211.05,574.7103.5

Â±39.8308.3Â±86.1205.9153.1

Â±38.3136.0Â±21.2300.6

+39.4211.5Â±62.799.7

Â±25.1180.2Â±35.1153.145.2

Â±15.168.9+22.457.141.5

Â±13.4532.0Â±111.4834.6+108.31,043.5Â±580.5612.9

Â±217.5683.3S

1236314.4

Â±1.327.4Â±2.351.9

Â±0.76.1Â±1.85.3Â±0.95.4Â±1.216.3Â±5.23.6Â±0.82.4Â±0.41.8

Â±0.512.8+1.56.2+

1.75.322.4

+4.28.1Â±1.26.7Â±1.723.0Â±3.26.0Â±1.84.7+0.84.6Â±1.011.3

Â±4.610.9Â±2.77.46.5

Â±1.56.5Â±1.3361.7+79.119.9+3.717.4

Â±4.282.4+69.917.411

3.7Â±37.6492.9+49.6303.37.1

Â±2.322.3+4.614.73.6

Â±0.78.8Â±0.63.5Â±0.515.7Â±1.37.7Â±1.57.9Â±2.27.73.00

Â±0.31.95Â±0.42.52.4

Â±0.310.7Â±1.321.7

Â±5.830.2+8.416.2+6.116.2

" Mean Â±SEM of 2 to 7 replicate assays.* Mean Â±SEM for each histological type of tumor.' Median.

i.p. administration. The LD,o values were 4.42 mg/kg for VBL,4.37 mg/kg for VCR, and 0.20 mg/kg for S 12363.

Primary screening was assessed on P388 leukemia and thenpursued on a panel of murine tumors. Results are listed inTable 2. The antitumor activity of S 12363 was compared withthat of VBL or VCR. Good activity was observed on i.p. P388leukemia, i.p. LI210 leukemia, i.p. B16 melanoma, and s.c.colon adenocarcinoma 38 as well as on disseminated modelss.c. P388 leukemia, i.v. B16 melanoma, and i.p. M5076 sarcoma. In all experiments, the optimal dosage was 10 to 40times lower than that of VBL, depending on the models studied.S 12363 had no effect on growth of s.c. Lewis lung carcinoma.The TI, defined as the ratio of LD10 to the lowest dose givingT/C of 135% (i.p. P388, one administration), was higher for S12363 (TI = 18) than VBL (TI = 11) or VCR (TI = 8.7). Theinfluence of the mode of administration on the activity of S

12363 was studied on P388 leukemia and B16 melanomainoculated i.p. Antitumor activity of S 12363 was more clearlymarked when the compound was administered by the i.p. routeagainst i.p.-implanted tumors (Table 3). Nevertheless, the activity remained significant by the i.v. and p.o. routes.

The influence of schedule of treatment on the therapeuticeffect was investigated in two independent experiments (Table3). In Experiment 1, single administration was compared with3 administrations 7 days apart starting from Day 1. The T/Cwas slightly, but not significantly, improved with its intermittent treatment, while the number of long-term survivors wasslightly higher following a single administration. Similar resultswere obtained with i.p. B16 melanoma. In Experiment 2, singleadministration was compared with 3 administrations 4 daysapart starting from Day 1. This dosing schedule (Days 1, 5, and9) did not improve antitumor activity.

2314




Table 2 Activity of S 12363 and reference compounds on a panel of murine tumors recommended by the NCI

TreatmentTumor

DrugSchedule(days)P388

leukemiai.p.S12363VCRVBLS

12363VCRVBLS.C.

S12363VCRVBLL1210

leukemia Â¡.p. S12363VBLB16

melanomaÂ¡.p.S12363VBLi.v.

S12363VBLLewis

lung carcinoma s.c. S12363VBLonlyonlyonlyonly-onlyonly,5,9,5,9,5,9onlyonlyonlyonlyonlyonly-5-5Colon

adenoearcinoma 38 s.c. S 123632,9VBL2.9S

123632,9VBL2,9M5076

sarcoma i.p. S 12363 1, 5, 9,13VCR1,5,9,13Routei.p.i.p.i.p.i.v.i.v.i.v.i.v.i.v.i.v.i.p.i.p.i.p.i.p.i.p.i.p.i.p.i.p.i.V.i.v.i.p.i.p.Â¡.p.i.p.Optimal

dosage"

(mg/kg/day)0.150330.2240.2001.0003.0000.3003.0000.1503.0000.2003.0000.075'1.0000.2003.0000.1003.0000.1001.000Mediansurvival time(%ofT/C)244*171167186127138173.0f172.0141.0164.0r136.0173.0'145.0(88.0)''(65.6)165f141Median

60-day tumor wtsurvivors (% ofT/C)13/450/173/651/251/153/25100.8760.0(/027.0'13.0

* Optimal dosage is defined as the dosage giving the best T/C.4 T/C a 175% considered significant activity.c T/C a 150% considered significant activity.Â¿Numbers in parentheses, percentage of inhibition of the number of mÃ©tastases.' Maximally tolerated dose for nonresponsive tumors.

< 42% considered moderate activity and T/C Â£10% considered significant activity.

Table 3 Effects of schedule and route of administration on the activity ofS 12363 against i.p. P388 leukemia and i.p. BI6 melanoma

Tumor"P388

leukemiai.p.Experiment1Experiment

2B16

melanoma i.p.Schedule(days)1

only1only1,8,151only1only1only1,5,91

only1,8,151

onlyRoutei.p.i.v.i.p.i.p.i.v.P.O.i.p.i.p.i.p.i.V.Optimaldosage*(mg/kg/day)0.1500.200O.I

000.1250.1250.5000.0800.1500.1500.150Mediansurvivaltime(%ofT/C)f244183276>28018517726817320015360-daysurvivors2/51/51/56/100/100/102/101/60/50/5

Â°B6D2F, mice were inoculated i.p. on Day 0 with either IO6 P388 leukemia

cells or 0.5 ml of 1:10 B16 melanoma brei.* Optimal dosage is defined as the dosage giving the best T/C.CT/C 2 175% for P388 leukemia and T/C > 150% for B16 melanoma are

considered significant activity.

Cross-resistance. The cytotoxic activity of S 12363 was evaluated in vitro on P388 leukemia cell lines resistant to VCR(P388/VCR) and resistant to ADM (P388/ADM) by comparison with the sensitive P388 cell line (Table 4). Murine resistanttumor cells P388/VCR and P388/ADM were 4- and 9-foldmore resistant to S 12363 than were the parental P388 cells,whereas P388/VCR and P388/ADM cells were 4- and 20-foldmore resistant to VBL and 6- and 68-fold more resistant toVCR than were the parental cells. Multidrug-resistant P388/ADM cells were relatively more sensitive to S 12363 than toVBL or VCR. The same experiment was conducted in vivo onP388 cell lines, both sensitive and resistant to VCR (Table 5).

Table 4 In vitro cross-resistance pattern ofP388/ VCR and P388/ADM cell linesto S 12363, VBL, and VCR

IC,o (niu) Resistance factor0

Cell line S 12363 VBL VCR S 12363 VBL VCRP388 5.9 Â±0.6* 11.9 Â±1.1 9.9 Â±1.1P388/VCR 23.0 Â±4.9 45.3 + 3.6 61.6 Â±4.7 4 46P388/ADM 53.2 Â±6.7 239e 678e 9 20 68

Â°Determined by dividing the IC50 for the resistant line by the 1CÂ»

sensitive line.* Mean Â±SEM of 3 to 10 experiments.c Mean of 2 experiments.

for the

S 12363 showed less cross-resistance than the reference compounds VBL and VCR with a significant activity on the resistant line.

Stereospecificity. The epimer of S 12363, S 12362, wasinvestigated. Cytotoxic activity was compared in murine tumorcell line LI 210. The IC50of S 12362 (l h of contact) was 4342nM, against 14.4 HM of S 12363. Its epimer, S 12362, is thus300-fold less cytotoxic. The tests performed in vivo on P388leukemia led to similar conclusions (Table 6); S 12362 becamesignificantly active on P388 leukemia at a dose 1000 timeshigher than the minimum active dose of S 12363.

Antitumor Activity in Human Tumor Xenograft Models. Toevaluate the antitumor activity of S 12363 against human tumorxenografts, 5 human tumor cell lines, among the most sensitivein vitro, were transplanted into nude mice. The influence of thedosage and schedule of administration by the i.p. route wasfirst studied (Table 7).

Three schemes of administration were investigated: Days 0,4, and 8; Days 0, 4, 8, and 12; and Days 0, 3, 6, and 9. To avoidtoxic effects of S 12363, doses were adapted to the schedule

2315




Table 5 Antitumor activity of S 12363, VBL, and VCR on P388/VCR leukemiaand on sensitive P388 leukemia

Tumor"P388

leukemiaP388/VCR

leukemiaOptimal

dosage*

Drug(mg/kg/day)S

12363VCRVBLS

I2363VCRVBL0.08

1.001.000.08

1.001.00Median

survival time(%ofT/C)268

237236190135

16630-day

survivors2/10

0/100/100/10

0/100/10

Â°P388/VCR and P388 cells (lO'/mouse) were implanted i.p. on Day 0, and

drugs were administered i.p. on Days 1, 5, and 9.* Optimal dosage is defined as the dosage giving the best T/C.

Table 6 Antitumor activity of S 12362 and S 12363 on i.p. P3S8 leukemia

MedianDosage survival time

Drug" (mg/kg)(days)Control

0S

123630.031250.062500.125000.25000S

12362 0.31250.62501.25002.50005.000010.000010.721.322.3>30411.011.412.112.813.414.5%

ofT/C*100199208>280Toxic10310611311912513530-daysurvivors0/270/101/107/101/100/100/100/100/100/100/9Body

wtchange'

(g)+

1.2-0.1-0.2-2.6-4.3+0.8+0.1+0.1+0.4+0.2+0.2

" Drugs were administered i.p. on Day 1.*T/C, MST of treated groups/MST of control group x 100.' Difference in body weight (g) between Days 5 and 1.

used: 0.1 mg/kg/day (Days 0, 4, and 8 and Days 0, 4, 8, and12); 0.075 mg/kg/day in Days 0, 3, 6, and 9. For comparativereasons, two reference drugs were included, VBL and/or YDS.Following i.p. administrations, S 12363 was moderately, butsignificantly, active on NCI-H460 (lung carcinoma) and PANC-I (pancreatic carcinoma) and inactive on NCI-HI25 (lungcarcinoma), HT-29 (colon adenoacarcinoma), and A-431 (vul-var carcinoma).

When administered by the i.v. route, S 12363 showed a lowertoxicity, reflected by a lower body weight loss, fewer hepaticdamages (data not shown), and a better tolerance to drugtreatment. Different dosages and schedules of administrationwere studied using NCI-H460 xenograft: 0.15 mg/kg of S12363 given intermittently every fourth day for 4 treatments,0.3 mg/kg administered as a single dose, and 0.25 mg/kgadministered every 7 days for 3 treatments (Table 7). Theevolution of the T/C values as a function of time was represented in Fig. 2. The best scheme of administration was Days0, 7, and 14. In these conditions, the best T/C value obtainedwas 10% and remained under 50% at the end of the experiment(Day 34). This schedule allowed us to increase the dosage of S12363 without increasing its toxicity.

We have thus chosen this scheme (Days 0, 7, and 14) for theother 4 xenografts. In all 5 xenograft models, S 12363 showeda significant antitumor activity with optimal T/C values under25% (Table 7).

Tumor growths were also significantly delayed following S12363 administration as shown by the SGD values (Table 7).In three tumor cell lines, we observed a complete disappearanceof the tumor only in mice treated with S 12363. This disappearance started from Days 15 to 23 and was irreversible untilDay 77 (day of sacrifice). This occurred in 2 of 10 mice forPANC-1 and A-431 tumors and in 3 of 10 mice for the NCI-

HI 25 tumor. The reference compound VBL was less active and18-fold less potent than was S 12363 in these 5 xenografts.

All the preceding results were summarized in Table 8, whichshows the comparison of the antitumor activity of S 12363 andreference compounds.

DISCUSSION

The biindole alkaloids of Catharanthus roseus (Vinca rosea)result from the linkage of 2 alkaloids: a derivative of catharan-thine and vindoline. From a chemical point of view (15), transformations carried out on the catharanthine moiety have usuallyled to compounds devoid of antitumor activity apart fromdehydration derivatives of the tertiary alcohol on the C4' atom

(vinorelbine) (16). In contrast, modification of the vindolinemoiety led to the formation of more active compounds whichare more easily synthesized, using vinblastine as starting material (17). The region modified in these latter compounds islocated around the indolic nitrogen atom. The chemical modifications in this part of the molecule can have at least twoimportant consequences: the interaction of the functional groupwith biological molecules and the modification of the physico-chemical properties (polarity and partition coefficient). Theimportance of optimal values of polarity for the diffusion andthe passage of these compounds through biological membraneswas previously discussed (15).

The appending of an a-aminophosphonic acid, bioisoster ofa natural amino acid, gives to S 12363 an optimal polarity, thisacid being used in the form of diester. In addition, we hypothesized two favorable properties for this molecule: (a) a highuptake by the cells, due to the vectorization conferred by theamino acid; and (b) an improved retention by the cells, theefflux of the compound being hindered by a metabolic transformation of the neutral phosphonic ester to charged acid, whichwould therefore be incapable of recrossing the cell membrane.

For the evaluation of the antitumor activity of S 12363, weused first a classical approach based upon the common murinetransplantable tumors and then the more recent approach developed by the NCI, called "Disease Oriented Strategy." This

latter strategy consists of an in vitro prescreening performed onhuman tumor cell lines followed by an in vivo evaluation on thesame lines xenografted into nude mice.

In conventional models, S 12363 was found to be at least asactive as the reference compounds, but at doses 10 to 40 timeslower. This drug has been shown to have significant activity byi.v. and p.o. routes as compared with the usual i.p. route. Theantitumor activity was slightly improved by the intermittentschedule.

The fact that S 12363 was more active than were referencecompounds on tumor cells resistant to VCR or ADM, if confirmed by other studies, may lead to major interest in clinicalsituations. S 12363 was found to be very potent in the in vitropanel system as shown by comparison with its activity withreference compounds. On average, S 12363 is 72- and 36-foldmore cytotoxic than are VCR and VBL, respectively. It isdifficult, considering the limited number of cell lines per tissue,to draw conclusions on the specificity of S 12363 for a givenhistolÃ³gica! type. However, results presented in Table 1 showthat a better activity was obtained in leukemia, lung, and breastthan in colon or kidney cancer cell lines. From the panel of ourin vitro human cell lines, we selected 5 lines for the in vivoevaluation of S 12363, based on (a) their adequate growth innude mice and (b) their in vitro sensitivity to S 12363 (the IC50s

2316




Table 7 Activity of S 12363 and reference compounds on human tumor xenografts

Â¡.p.administrationTumor

celllineNCI-H

460NCI-H

125HT-29PANC

1A-431Histological

typeLung

carcinomaLung

carcinomaColon

adenocarcinomaPancreas

carcinomaVulvar

carcinomaDrugSI

2363

VBLSI

2363VBLVDSS

12363VBLVDSS

I2363VBLS

12363VBLVDSSchedule

(days)0,4.8.

12

0.4.8,120,4,8,

120,4,8, 120,4,8,120,4,8

0,4,80.

3. 6. 90, 3, 6,90,4,8,

120,4,8, 120.4,8, 12Dosage

(mg/kg/day)0.100

1.0000.100

1.5001.5000.100

1.5000.075

1.0000.100

1.5001.500Optimal

T/C%'45

8361

4550107

69494766

8664SGD*-0.2

(-)

-0.4(-)+0.5

(-)

O(-)+0.1

(-)+2.3I++)+2.4

(++)+

1.0 (+)+0.8 (-)+0.7 (-)Schedule

(days)0,4.8,

1200,7, 140,7,140.7.

140, 7.140,

7, 140, 7.140,7,

140, 7,140,7,

140, 7, 14i.v.

administrationDosage

(mg/kg/day)0.150

0.3000.2504.5000.250

4.5000.250

4.5000.250

4.5000.250

4.500OptimalT/C'

';;.16

17106122423

2932312

17SGD+2.9

(++)+2.7(++)+0.4

(-)+7.9

(+++)+5.1

(+++)+4.1

(+++)+2.5

(++)+2.5 (++)

" T/C values were determined after the

significant activity.* Ta treated - 7"dcontrol/7",, control (Td.

end of treatment. T/C = (mean VJV, treated/mean y,/ya control) x 100. T/C < 50%, moderate activity; T/C < 25%,

tumor doubling time). SDG Â£1 expressed as -; >1 as +; >2 as ++; and >4 as +++.

10 20 30 40

DAYSFig. 2. Effects of schedule of administration of S 12363 in the NCI-H460

xenograft line. Nude mice were grafted s.c. with tumor fragments (3-mm size).Treatment started when the tumor reached 50 mm3 (Day 0), and tumor volumeswere measured twice a week. S 12363 was administered by the i.v. route at Days0. 4. 8, and 12 (0.15 mg/kg/day) (A): Days 0, 7. and 14 (0.25 mg/kg/day) (â€¢);and Day 0 (0.3 mg/kg) (x). T/C, on a given day, was expressed as the meanrelative tumor volume of the treated group divided by that of the control groupx 100.

Table 8 Comparison of the antitumor activity of S I2363 and referencecompounds

Activity rating"

Tumor model S 12363 VBL

Murine tumorsi.p. P388 leukemiai.p. 11-10 leukemiai.p. B16 melanomas.c. Lewis lung carcinomas.c. Colon 38s.c. P388 leukemiai.v. B16 melanomai.p. M5076 sarcoma

Human tumor xenograftsi.v. treatment

NCI-H 460 lungNCI-H 125 lungHT-29 colonPANC-1 pancreasA-431 vulvar

i.p. treatmentNCI-H 460 lungNCI-H 125 lungHT-29 colonPANC-1 pancreasA-431 vulvar

â€¢H-

" -, no activity; +, moderate activity (>lower limit of significance); ++,

significant activity (>upper limit of significance); +++, high activity.* Reference compound is VCR.' Reference compound is VDS.

obtained after l h of exposure being in the range of 2.4 to 10.7nM).

When administered in nude mice by the i.v. route followingthe intermittent schedule (Days 0, 7, and 14), S 12363 reducedby more than 75% the tumor volume, with some mice beingtumor free at least 77 days later in 3 xenografts (PANC-1, A-431, and NCI-H 125). These complete regressions of the tumorswere not observed in mice treated with VBL. S 12363 is thusmore active than VBL and about 20-fold more potent. Theseresults, obtained on only 5 xenografts, might be extended toother xenografts, since 23 of the 37 cell lines used in this panelwere sensitive to S 12363 with an IC50below 10.7 nM. Interestingly, and in opposition to the previous results obtained withi.p.-implanted tumors, S 12363 is much more active when given

by the i.v. than the i.p. route.

Considering the pharmacological properties of S 12363, threecharacteristics are to be noted/ (a) S 12363 stands out from thereference compounds by its optimum active doses that were 10to 40 times lower, (b) The stereoselectivity is particularly impressive since S 12362, which differs only by the absoluteconfiguration of the asymmetric carbon atom of the side chain,is 300-fold less cytotoxic and 1000 times less potent in vivo, (c)The fact that S 12363 is cytotoxic, after l h of contact, forsome cell lines having a slow rate of proliferation may conferto this drug an advantage for the treatment of solid tumors.Such a possibility is supported by the excellent activity observedin the 5 xenografts tested.

The precise mechanism of action of S 12363 is not known.One can postulate either a better interaction with the tubulindimer (for example, a higher affinity to the VBL site or the

2317




formation of a more stable complex) or a better cellular accumulation and retention by tumor cells. This latter possibility isparticularly attractive and may explain the relative activity ofthe drug against resistant P388 leukemia and the superiority ofboth single-dose and intermittent schedule with long intervalsover an intermittent schedule with short intervals.

ACKNOWLEDGMENTS

The authors thank M. Anstett, V. Perez, M. H. Rouillon, M. Visalli,V. Boutin, M. J. Calatayud. and T. Dunn for technical assistance.

REFERENCES

1. Johnson, I. S.. Armstrong, J. Ci., Gorman. M.. and Burnett, J. P. The Vincaalkaloids: a new class of oncolytic agents. Cancer Res.. 23:1.190-1426, 1963.

2. Barnett, C. J., Cullinan. G. J., Gerzon. K.. Hoying. R. C., Jones, W. E.,Newlon, W. M., Poore, G. A., Robison. R. I., Sweeney. M. J., Todd, G. C.,Dyke. R. \\.. and Nelson, R. L. Structure activity relationships of dimeric( aiharanthu1, alkaloids. 1. Dcacetyl-vinblastinc amide (vindesine) sulfate. J.Med. Chem.. 21: 88-96. 1978.

3. Armstrong. J. Ci.. Dyke. R. \\.. Fouls. P. J.. Hawthorne. J. J., Jansen, C. J.,and Peabody, A. M. Initial clinical experience with vinglycinate sulfate, amolecular modification of vinblastine. Cancer Res.. 27: 221-227, 1967.

4. Bhushana Rao. K. S. P.. Collard. M. P.. and Trouet. A. Vinca-2Ã¬oyl aminoacid derivatives: as new anticancer agents (review). Anticancer Res.. 5: 379-386. 1985.5. Lavielle, G.. Hautefaye. P.. Schaeffer, C'.. Boutin, J. A., Cudennec, C. A.,

and Pierre. A. New i>-aminophosphonic acid derivatives of vinblastine: chemistry and antitumor activity. J. Med. Chem.. in press, 1991.

6. Alley, M. C.. Scudiero. D. A., Monks, A., Hursey, M. L., Czerwinski. M. J.,Fine, D. L.. Abbott, B. J.. Mayo. J. G., Shoemaker. R. H.. and Boyd. M. R.Feasibility of drug screening with panels of human tumor cell lines using

microculture tetrazolium assay. Cancer Res., 48: 589-601, 1988.7. Shoemaker, R. H., McLemore, T. L., Abbott, B. J., Fine, D. L., Gorelik, E.,

Mayo, J. G., Fodstad. O.. and Boyd. M. R. Human tumour xenograft modelsfor use with an in n'rro-based. disease-oriented antitumour drug screening

program. In: B. Winograft, M. J. Peckham, and H. M. Pinedo (eds.). HumanTumor Xenografts in Anticancer Drug Development. Berlin: Springer-Verlag, 1988.

8. Instruction 2711: Summary of the Usual Characteristics of Selected MurineModel Used under the Auspices of the NCI. Bethesda. MD: Division ofCancer Treatment, 1985.

9. Carmichael, J., Degraff, W. G., Gazdar. A. F., Minna, J. D., and Mitchell,J. B. Evaluation of a tetrazolium-based semiautomated colorimetrie assay:assessment of chemosensibility testing. Cancer Res., 47: 936-942, 1987.

10. Driscoll, J. S. The preclinical new drug research program of the NationalCancer Institute. Cancer Treat. Rep., 68: 63-75, 1984.

11. Geran, R. I., Greenberg, N. H., MacDonald, M. M., Schumacher, A. M.,and Abbott, B. J. Protocols for screening chemical agents and naturalproducts against animal tumors and other biological systems. Ed. 3. CancerChemother. Rep. Part III. 3: 1-100, 1972.

12. Developmental Therapeutics Program, Division of Cancer Treatment, National Cancer Institute. In rivo cancer model 1976-1982. In: NIH Publication84 2635. Washington, DC: Government Printing Office, 1984.

13. Boven, E., Winograd, B., Fodstad, O., Lobbezoo, M. W.. and Pinedo, H. M.Preclinical Phase 11studies in human tumor lines: a European multicancerstudy. Eur. J. Cancer Clin. Oncol., 24: 567-573, 1988.

14. Paull, K. D., Shoemaker, R. H., Hodes. L., Monks, A., Scudiero, D. A.,Rubinstein, L., Plowman. J., and Boyd, M. R. Display and analysis ofpatterns of differential activity of drugs against human tumor cell lines:development of mean graph and compare algorithm. J. Nail. Cancer lust..81: 1088-1092, 1989.

15. Gerzon, K. Anticancer agents based on natural products models. In: Medicinal Chemistry, Vol 16, Chap. 8. New York: Academic Press, 1980.

16. Potier. P. Synthesis of the antitumor dimeric indole alkaloids from Cathar-anthus species (vinblastine group). J. Nati. Proc., 43: 72-86, 1980.

17. Bhushana Rao, K. S. P., Collard, M. P., De Jonghe, J. P., Atassi, G.,Hannart, J. A., and Trouet, A. Vinblaslin-23-oyl amino acid derivatives:chemistry, physicochemical data, toxicity, and antitumor activities againstP388 and L1210 leukemias. J. Med. Chem., 28: 1079-1088, 1985.

2318



1991;51:2312-2318. Cancer Res Alain Pierré, Laurence Kraus-Berthier, Ghanem Atassi, et al. Derivative, S 12363

AlkaloidVincaPreclinical Antitumor Activity of a New

Updated version

http://cancerres.aacrjournals.org/content/51/9/2312

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/51/9/2312To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



Preclinical Antitumor Activity of a New Vinca Alkaloid Derivative, S … · [CANCER RESEARCH 51....

Documents

Transcript of Preclinical Antitumor Activity of a New Vinca Alkaloid Derivative, S … · [CANCER RESEARCH 51....