Stealth Liposomes

(CANCER RESEARCH 52. 2431-2439. May 1. I992|

Stealth Liposomes: An Improved Sustained Release System for 1-/?-D-ArabinofuranosyIcytosine ' Theresa M. Allen,2 Tarun Mehra, Christian Hansen, and Yeen Cheel Chin Department of Pharmacology, University of Alberta, Edmonton, Alberta, Canada, T6G 2H7

ABSTRACT Newly developed liposomes with prolonged circulation half-lives and dose-independent pharmacokinetics (Stealth" liposomes) have been tested for their efficacy as a slow release system for the rapidly degraded, schedule-dependent, antineoplastic drug 1-0-D-arabinofuranosylcytosine (ara-C) in the treatment of murine L1210/C2 leukemia. Mice were given injections of either 10* cells or 10' cells by either the i.v. or the i.p. routes. Leukemia-bearing mice were treated with either i.v. or i.p. injec tions of free drug, i.v. or i.p. injections of liposome-entrapped drug, or 24-h i.v. infusions of free drug. Long-circulating liposomes contained, as the stealth component, either monosialoganglioside or polyethylene glycol-distearoylphosphatidylethanolamine. Liposomes lacking the stealth components (non-stealth liposomes) were also injected for comparison. At lower dose ranges, stealth liposomes were superior to non-stealth liposomes in prolonging mean survival times of the mice, and all liposome preparations were superior to injections of the free drug. Drug entrapped in stealth liposomes, when administered at or near the maximum tolerated dose of 100 mg/kg ara-C were considerably superior to 24-h free drug infusions given at the same total drug dose. Therapeutic effect was related to the half-life of leakage of ara-C from the liposome formulations, as well as to circulation half-life, with maximum therapeutic effect achieved with long circulation half-lives and more rapid leakage rates. The thera peutic efficacy of non-stealth liposomes increased with increasing lipo some (and drug) dose as a result of saturation of liposome uptake by the mononuclear phagocyte system, which resulted in longer circulation halflives for these liposomes at higher doses (Michaelis-Menten pharmaco kinetics). Liposome entrapment can protect rapidly degraded drugs from breakdown in n'TO,with release of the drugs in a therapeutically active form over periods of up to several days. The dose-independent pharma cokinetics and reduced mononuclear phagocyte system uptake of stealth liposomes gives them distinct advantages over non-stealth liposomes.

INTRODUCTION There have been at least 2 major drawbacks to the use of liposomes as sustained release systems for drugs in vivo, (a) The high affinity of conventional liposome formulations (termed here "non-stealth" liposomes) for the MPS' (reticuloendothelial system) leads to their rapid removal from circula tion resulting in adverse effects on this important host defense system (reviewed in Refs. 1 and 2) and reduced availability of liposome-entrapped drug to other tissues, (b) Avid uptake into Received 7/31/91; accepted 2/19/92. 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 in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1This work was supported by a grant from the Medical Research Council of Canada (MA-9127) and by Liposome Technology. Inc., Menlo Park, CA. who provided the PEG-DSPE and HSPC for the experiments. T. M. was supported by a graduate studentship, and Y. C. C. by a summer studentship, from the Alberta Heritage Foundation for Medical Research. 2To whom requests for reprints should be addressed. 3The abbreviations used are: MPS, mononuclear phagocyte system (reticuloendothelial system); ara-C, 1-ff-D-arabinofuranosylcj tosine; PC. egg phosphatidylcholine; HSPC, hydrogenated soy phosphatidylcholine; SM, bovine brain sphingomyelin; CH, cholesterol; PC, egg phosphatidylglycerol; GMi. monosialo ganglioside; PEG, polyethylene glycol; PEG-DSPE. polyethylene glycol (M, 2, average) derivative of dislcaroylphosphatidylethanolamine; REV, large unilamel lar liposomes prepared by the reverse-phase evaporation technique; USP. United Stales Pharmacopoeia; IIS, percent increase in the mean survival times of test mice as compared to control mice; q3h. every 3 h; LDÂ«.SO^c lethal dose; T,/2, half-life.

the MPS of non-stealth liposomes leads to saturable (nonlinear) pharmacokinetics for the carrier, which complicates calcula tions of the amount of liposome-entrapped drug necessary to achieve a given therapeutic drug dose (3-6). In addition, many conventional liposome formulations are susceptible to the ac tion of plasma proteins and other biological fluids in vivo, leading to rapid loss of their drug contents (7-9). New liposomal formulations have now been developed that have considerably reduced MPS uptake, remain in circulation for extended periods of time (9-17) with dose-independent pharmacokinetics (6), and have reduced susceptibility to pro tein-induced leakage (13, 18). These improved liposome for mulations, termed Stealth4 liposomes for their ability to avoid detection by the MPS (12), would be expected to be of utility as drug-sustained release systems for drugs that normally ex perience rapid degradation in vivo. The model system chosen to test this proposed therapeutic application was one in which the drug ara-C, free or liposomeentrapped, was tested for its ability to prolong survival times of mice bearing L1210/C2 leukemia. ara-C was chosen because it is a schedule-dependent antineoplastic drug, used clinically in the treatment of acute leukemia, that is rapidly inactivated in vivo by cytidine deaminases with an initial half-life of 16 to 20 min in mice, close to the value found in humans (19, 20). It is an inexpensive model for other drugs, e.g., the immune modu lators, which are candidates for delivery by drug-sustained release systems. In addition, several investigators have previ ously tested non-stealth liposomal preparations of ara-C and have found good therapeutic efficacy for their formulations (21-28). These prior experiments provide data against which the efficacy of stealth liposomes can be compared, although the comparison will be complicated by the nonlinear pharmacoki netics of the previous preparations. Measurement of mean survival times of tumor-bearing mice in response to various treatment regimens is a simple means of comparing the thera peutic efficacy of the different therapeutic groups. The objective of our initial experiments was to compare the therapeutic efficacy of the various treatments, not to produce long-term survivors, although this is achievable at high doses of ara-C at low tumor burdens. The presence of long-term survivors prevents statistical comparisons. Therefore, initially, low drug doses of ara-C were chosen, so that single injections of free drug had marginal effects on increasing mean survival times of the mice. The same total dose given by infusion prolonged the mean survival times, but did not result in cures. In the final experiments, the objective was to evaluate the therapeutic efficacy, and the maximum tolerated dose, of a newly developed, inexpensive stealth liposome formulation con sisting of HSPC, CH, and PEG-DSPE. Single or multiple injections of liposome-entrapped drug were given in order to determine the conditions that resulted in long term survivors. In addition, drug leakage, pharmacokinetic, and tissue distri bution studies were carried out in order to provide data that 4 Stealth is a registered trademark of Liposome Technology Inc.. Menlo Park. CA.

2431

Downloaded from cancerres.aacrjournals.org on January 26, 2016. © 1992 American Association for Cancer Research.

ARA-C IN STEALTH LIPOSOMES

would aid in our understanding of the mechanism by which liposome-entrapped drug was having its therapeutic effect. MATERIALS AND METHODS Materials. PC, SM, and PC were purchased from Avanti Polar Lipids (Birmingham. AL). CH and ara-C were purchased from Sigma Chem ical Co. (St. Louis, MO) and [5-'H]ara-C (0.55-1.1 TBq/mmol) was purchased from Amersham (Oakville, Ontario, Canada). HSPC and PEG-DSPE were generous gifts of Liposome Technology, Inc. (Menlo Park, CA). The PEG used in these experiments was an average of A/r 2. The synthesis of PEG-DSPE has recently been described (17). GMr was purchased from Makor Chemicals (Jerusalem, Israel). Pyrogenfree saline (0.9% USP) was obtained from Travenol Canada, Inc. (Mississauga, Ontario, Canada). Liposome Preparation. Liposomes were prepared by the REV method (29) with an aqueous solution of 247 Â¿IM ara-C (60 mg/ml) in distilled water (290 mOsm) containing [5-'H]ara-C such that the specific activity of the final ara-C solution was 37 kBq/ml. For some experiments involving the effect of osmolarity on leakage of ara-C from liposomes, the concentration of ara-C was 150 mg/ml (850 mOsm). The liposomes were then extruded (Extruder; Lipex Biomembranes, Vancouver, Brit ish Columbia, Canada) 10 times through 2 stacked 0.4-^m Nuclepore polycarbonate filters (Nuclepore Corp., Pleasanton, CA) either at room temperature, or at 5 to 10Â°Cabove the phase transition temperature in the case of high phase transition phospholipids (30, 31). Concentrations of lipids are expressed as molar ratios of phospholipids. Liposome-entrapped ara-C was separated from free ara-C in one of 3 ways depending on the desired concentration of liposomal ara-C. For experiments involving low doses of liposomal ara-C, liposomes were chromatographed over Sephadex G-50 in sterile, pyrogen-free NaCI (0.9%, USP) containing 10 mM /V-tris(hydroxymethyl)methyl-2-aminoethanesulfonic acid, pH 7.4 (buffer). For higher doses of liposomal ara-C, where dilution of the liposomes was undesirable, liposomes were dialyzed through Spectrapore dialysis tubing (cutoff, M, 12) against several changes of buffer at 4Â°C.For the highest doses of liposomal ara-C, liposomes were concentrated by centrifugation at 100,000 x g for 1 h, and free ara-C was removed by washing the liposomes 3 times in buffer. Following the final washing step, the liposome pellet was made up to the desired concentration by the addition of buffer. In some liposome preparations, liposomes were pelleted by centrifugation, the supernatant containing free ara-C was removed, but the pellet was not washed. The pellet was resuspended in buffer to the desired drug concentration. In those preparations, the total amount of free ara-C was equivalent to the total amount of liposomal ara-C, as determined by chromatography of the samples over Sephadex G-50. Samples of the liposome preparations were taken before and after separation of the free drug, and the concentration of ara-C in the preparations was calculated from the specific activity of [5-'H]ara-C. Extruded vesicles were sized by quasielastic light scattering using a BI-90 particle sizer (Brookhaven Instruments, Holtsville, NJ). Phospholipid concentration was determined using the method of Bartlett (32), and trapped volumes of the liposomes were determined from the specific activity of [3H]ara-C. The average mean diameter of the lipo somes was approximately 190 to 220 nm and the trapped volume was in the range of 4 to 6 liters/mol phospholipid. These results could be reliably reproduced from sample to sample. Leakage of ara-C from Liposomes. ara-C and [5-3H]ara-C were en

age in the presence of plasma, all therapeutic experiments and all subsequent leakage experiments using other liposome compositions were done with isoosmolar ara-C. Pharmacokinetic Experiments. The time course for the elimination of liposomes from the circulation of C57BL/6J x DBA/2J (B6D2F,) mice was determined by either i.p. or i.v. injection of liposomes con taining entrapped ['"Ijtyraminylinulin (33), a marker of intact lipo somes (10, 13, 33). Free label was separated from entrapped label, prior to injection, by gel filtration over Ultragel AcA-34 columns (IBF Biotechnics, Columbia, VtD). Mice were sacrificed at various time points postinjection, and blood, liver, spleen, and remaining carcass were sampled. These tissues were corrected for blood volume using correction factors determined previously (12). Bone marrow levels (cpm/mg) were determined as described previously (13). Results are expressed as percent of in vivo cpm per organ, which gives total organ uptake, corrected for leakage of drug from the liposomes (6). Because total bone marrow weight could not be determined, percent of in vivo cpm could not provided for bone marrow. Results are also expressed as cpm/mg tissue, normalized to IO6 injected cpm, which is propor tional to the concentration of liposomes in each tissue. Similar experi ments were performed in mice that had received IO6 L1210/C2 leuke mia cells 24 h previous in order to determine whether the presence of the tumor cells affected the pharmacokinetics of the liposomes. Survival Experiments. L1210/C2 leukemia cells (34) were passaged in vivo by weekly i.p. transplants in either male or female B6D2F, hybrid mice. After 25 weeks of in vivo passage, the passage line was restarted from frozen stock to preclude the possibility of genetic drift. Groups of 5-10 mice of either sex (2-4 months, 18-30 g) were given injections by either the i.p. or the i.v. (tail vein) route with either 10s or 10" L1210/C2 cells. Twenty-four h after implantation of cells, treatment began. Treatment consisted of either single or multiple injections of free drug or liposome-entrapped drug by either the i.v. or the i.p. route (Table 1). Control mice received injections of sterile saline (0.9%, USP). Other groups of control mice received empty (i.e., con taining no drug) liposomes. Some groups of mice received free ara-C over 24 h by tail-vein infusion according to described procedures (35, 36). Infused drug was administered in a total volume of 3 ml at a flow rate of 0.0021 ml/min using a Harvard Infusion Pump (model 940). Survival times were noted for all mice. Survival times of mice in the various treatment groups were compared by analysis of variance. Some times the results were expressed as %ILS.

RESULTS The rates of leakage from liposomes of ara-C at isoosmolar and hyperosmolar concentrations were determined for lipo somes composed of PC:CH 2:1, or SM:PC:CH:GM, 1:1:1:0.14. This was to determine whether the presence of hyperosmolar concentrations of ara-C, which was desirable in order to reduce the total amount of lipid administered to the mice at the higher ara-C doses, would affect the rate of ara-C leakage from the liposomes in the presence of buffer and plasma. In buffer, at 37Â°C,the rates of leakage of ara-C from liposomes of both compositions were low and independent of the concentration of the entrapped ara-C, with terminal half-lives averaging 442 h for POCH liposomes and 475 h for SM:PC:CH:GMi lipo somes (data not shown). In the presence of 25% plasma in

trapped in liposomes, as above, at concentrations of 60 or 150 mg/ml, corresponding to osmolarities of 290 and 850 mOsm (Westcor Vapour Pressure Osmometer, model 5500; Westcor, Logan, UT). The effect of osmolarity on leakage was determined for liposomes composed of SM:PC:CH:GMi, 1:1:1:0.14 or PC:CH, 2:1 (molar ratios of phospho lipid). The rate of leakage was determined at 37Â°Cin either buffer or 25% pooled human plasma in buffer, by measuring the free and en trapped ara-C peaks following chromatography of samples over Seph adex G-50 columns. Half-times for leakage were calculated using Graphpad software (ISI, Philadelphia. PA). Because liposomes contain ing hyperosmolar concentrations of ara-C had increased contents leak

Table 1 Experimental protocols for treatment ofLI2IO/C2 mice with ara-C Route of injection of cells (10'or 10'cells/mouse)

leukemia-bearing

Route of drug treatment ( 10-100 mg/kg ara-C)

.V..V..V..p.â€¢p..p.i.v.Lp.Infusioni.p.I.V.Infusion

2432



buffer, the rates of leakage of hyperosmolar ara-C were rapid compared to isoosmolar ara-C (Fig. 1), and leakage of ara-C from PC:CH liposomes was more rapid than from SM:PC:CH:GMi liposomes under both conditions of osmolarity. The leakage of hyperosmotic ara-C from SM:PC:CH:GMi liposomes (Fig. 1, solid squares) was characterized by an initial rapid loss of contents, followed by a slow leakage, with a halflife similar to that seen with isoosmotic conditions. All thera peutic experiments were performed under isoosmotic condi tions because of the susceptibility of liposomes containing hyperosmotic ara-C to plasma-induced loss of contents. We used large liposomes, formed by the REV technique, in these experiments in order to increase the trapped volume and decrease the administered lipid dose, particularly at the higher doses of ara-C. The in vivo behavior of the larger stealth liposomes used in these experiments was characterized in order to aid our understanding of how liposome-entrapped drug was exerting its therapeutic effect in vivo. Blood, liver, spleen, and remaining carcass levels, as a function of time postinjection, were determined for PC:CH, SM:PC:CH:GM,, and HSPC:CH:PEG-DSPE liposomes (REV, 0.4 Mmextruded) fol lowing either i.v. or i.p. injection. The results for blood, follow ing i.v. injection of a phospholipid dose of 0.5 /umol/mouse, are shown in Fig. 2. Liposomes containing PEG-DSPE or G\n were removed from circulation with similar pharmacokinetics at a rate that was significantly slower than that seen for lipo somes lacking stealth components (Fig. 2). PC:CH liposomes reached undetectable levels in blood by 48 h postinjection. Following i.p. injection, intact liposomes of all compositions began to move from the peritoneal cavity into blood approxi mately 2 h postinjection. Following their appearance in blood, liposomes were removed with a time course very similar to that seen following i.v. injection (data not shown), as has been previously seen for smaller liposomes (17). For liposomes composed of HSPC:CH:PEG-DSPE, blood and liver levels were not significantly different for mice receiv ing either 0.5 or 4 Â¿imolphospholipid/mouse, the latter dose being the average phospholipid dose injected into mice receiving 50 mg/kg ara-C for liposomes of this composition (Table 2). For non-stealth liposomes composed of HSPC:CH:PG, at the higher dose, liver levels were significantly lower (P > 0.001) and the blood levels were significantly higher (P> 0.001) than at the lower dose, due to saturation of MPS uptake of the

Fig. I. Leakage of [3H)ara-C from liposomes, as a function of time, at 37Â°Cin 25% human plasma. Liposomes were composed of SM:PC:CH:GMi 1:1:1:0.14 (D. â€¢¿) or PC:CH 2:1 (O. â€¢¿) and were REV-extruded through 0.4-^m Nuclepore filters (0.5 Â¿imolphospholipid/ml 25'V plasma). ara-C was entrapped in liposomes al isoosmolar (â€¢,G) or hyperosmolar (O. â€¢¿) conditions.

Time post-injection (hours)

Fig. 2. Blood levels in B6D2F, mice as a function time postinjection for liposomes (0.4 firn extruded REV) of different compositions. Mice (3 per group) were given 0.5 ^mol/mouse injections of liposomes composed of PC:CH 2:1 (O), SM:PC:CH:GMi 1:1:1:0.14 (d). or HSPC:CH:PEG DSPE 2:1:0.1 (A). Results are given as percent of in rivo cpm of liposome-entrapped [';'l|lyraminylinulin. Data are mean Â±SD.

liposomes (6). Bone marrow uptake, expressed as cpm/mg bone marrow, for both compositions at both doses was not signifi cantly different. Although these larger stealth liposomes did not have halflives in circulation as long as those seen for smaller stealth liposomes of the same composition (13, 17), their half-life in circulation of approximately 12 h was significantly longer than the 15-min initial half-life seen for PC:CH liposomes (Fig. 2). Liposomes containing GM1 (data not shown) or PEG-DSPE (Table 2) had decreased uptake into liver and spleen compared to PC:CH (data not shown) or HSPC:CH:PG liposomes (Table 2) as has been reported previously (10-13), although liver uptake, and particularly spleen uptake, were higher for the larger liposomes than had previously been seen for smaller liposomes, accounting for the decrease in circulation half-lives. Leukemic mice, given injections of liposomes 24 h after receiv ing 10" LI 210 cells, did not show any significant differences in liposome pharmacokinetics as compared to nonleukemic mice (data not shown). Fig. 3 shows the survival data for mice given either i.v. or i.p. injections of IO5 L1210/C2 leukemia cells and treated either i.v. or i.p. with 10 mg/kg ara-C as single injections of free drug, as 24-h infusions of free drug, or as single injections of each of 2 liposome compositions (PC:CH or SM:PC:CH:GMi). It can be readily seen that the stealth liposomes are superior to the non-stealth liposomes, which are in turn superior to single injections of free drug, at the same drug dose, in prolonging the survival time of the mice. All of these treatment groups are significantly different from each other (P > 0.001). The optimum dosage schedules for free ara-C are reported to be either by infusion or by frequent injection5 (37). We have compared our results to either 24-h infusions or to literature values for frequent injections. Free drug given by 24-h infusion (10 mg/kg total dose) was significantly better than 10 mg/kg ara-C entrapped in stealth liposomes (P > 0.001) against i.v. tumor (Fig. 3O), but not significantly better than stealth lipo somes (P < 0.05) against i.p. tumor (Fig. 3fi). The injection of 'The single i.v. dose LD,â€žfor ara-C is 2500 mg/kg and (his provides a cytotoxic body fluid level of ara-C for approximately 6 h. Optimum dosage schedules for injection have been determined to be every 3 h. The LD,0 for a q3h (x8) schedule was 244 mg/kg. The best dosing regimen for free ara-C was q.lh (X8); q4d (X4) (15 mg/kg/dose) which resulted in 5'i cures when the cell burden was i.v. 10*cells (37).

2433



Table 2 Tissue distribution in B6D2FÂ¡mice of'^l-laoeled

liposomes of 2 compositions (0.4 Â¡imextruded REV) as a function of liposome dose, 24 h postinjection via the tait vein

Results are expressed either as percent of m vivo cpm (mean Â±SD, n = 6) for liver, spleen, and blood, or as cpm/mg tissue (mean Â±SD, n = 6, except n = 3 for bone marrow), normalized to 10* cpm of injected counts. % of m vivo cpm remaining in body 24 h postinjection

(/Â¿mol)0.5

ratio)HSPC:CH:PG(2:1:0.1) Liposome composition (M

4 HSPC:CH:PEG-DSPE

(cpm/mg)

(2:1:0.1)Dose

0.54Liver67.3

marrowND(107.1 (370.1 55.9 (247.7 40.0 (220.3 43.2 (205.2

Â±4.6 Â±3.1 Â±41.2) (1068.5 Â±247.1) Â±4.0 22.4 Â±3.9 Â±37.5) (1101. 1 Â±198.7) Â±3.0 29.3 Â±3.2 Â±45.8) (2121.8 Â±509.2) Â±1.9 18.1 Â±2.3 Â±68.0)Spleen15.9(1219.1 Â±275.7)Blood0.5

(1.7 2.2 (7.1 12.4 (48.1 17.3 (78.0

Â±0.3 Â±1.3) Â±0.9 Â±2.4) Â±3.6 Â±17.3) Â±6.4 Â±42.5)Bone

Â±16.2) ND (88.5 Â±11.4) ND (7 1.4 Â±6.4) ND (115.5Â± 20.7)

IDO

75

50

25

B 10

15

20

25

20

25

Days survived

100 r

100 r

25

10

15

10

20

15

Days survived

Days survived

Fig. 3. Percentage of surviving B6D2F, mice (5 to 10 per group), as a function of days postinoculation of tumor, for mice receiving either i.p. (A and B) or i.v. (C and D) injections of 10* L12IO/C2 leukemia cells. Twenty-four h later the mice were treated either i.p. (A and C) or i.v. (B and D) with single injections of sterile isotonic saline (A). 10 mg/kg free ara-C (O), or 10 mg/kg of ara-C entrapped in liposomes composed of SM:PC:CH:GM, 1:1:1:0.14 (D) or PC:CH 2:1 (V). Mice received approximately 1 Â¿imolof lipid. Some mice received 24-h tail-vein infusions of a total of 10 mg/kg ara-C (O).

empty liposomes had no therapeutic effect (data not shown). Liposomes administered by one route of injection were effective against tumors given by the other route route of injection, i.e., i.v. liposomal ara-C was effective against i.p. tumor and vice versa (Fig. 3, B and C). A single i.v. injection of ara-C at a dose of 1000 mg/kg against IO5 LI210 tumor cells is reported to result in a %ILS of 30% (37). This was superior to 10 mg/kg ara-C administered in PC:CH liposomes, which resulted in a %ILS of 11%, but inferior to SM:PC:CH:GM, liposomes (%ILS = 52%) (Fig 3fi). Very similar relationships to those seen in Fig. 3 were ob

served between the various therapeutic groups at higher tumor burdens (IO6 cells/mouse) and at higher drug doses (data not shown). For subsequent experiments, even though experiments were done for all 4 injection groups (Table 1), the results for i.v. tumor and i.v. treatment will primarily be presented since these data best mimic the clinical situation. As the drug dose increased, the liposomes became superior in their therapeutic effect to the free drug given by 24-h infusion. These results are shown in Fig. 4 for i.v. treatment against i.v. leukemia. For mice receiving IO6 L1210/C2 cells and stealth liposomes, both by the i.v route, treatment with 10 mg/kg ara-C was signifi-

2434



of 80 mg/kg ara-C. Here, stealth and non-stealth liposomes were equivalent (data not shown). At i.v. doses of 80 mg/kg liposome-entrapped ara-C, PC:CH liposomes were therapeutically superior (%ILS = 141%) to 24-h free drug infusions (P > 0.001 ) against i.v. tumor, but still significantly inferior to stealth liposomes (P> 0.001). The maximum tolerated dose of liposomal ara-C was found to be on the order of 100 to 120 mg/kg ara-C, and did not appear to vary in any significant way between stealth and nonstealth liposomes (data not shown). This value can be compared to a 10% lethal dose of 95 mg/kg for an optimized i.v. injection schedule of q3h (xl6) (37) or a LD 0.001), resulting in a %ILS of 192%, compared to 122% for infusion at the same total dose (Fig. 4, A versus B). The results at 80 mg/kg for stealth liposomes can be compared against those for mice, bearing a tumor burden of IO6i.v. leukemia cells, injected with free ara-C at a multiple injection schedule of i.v. drug every 3 h for 3 days (24 injections), at a slightly higher total ara-C dose of 105 mg/kg, resulting in a %ILS of 142% (37). In other words, at the highest drug dose used in this series of experiments, the liposomal preparations were superior to an optimized multiple dose schedule of free ara-C administered over 3 days, and to 24-h infusions of free drug, suggesting that the liposomes were able to maintain therapeutic drug levels for periods longer than 24 h. As the dose of liposome-entrapped drug increased, the effi cacy of non-stealth liposomes, for all 4 injection groups, was significantly inferior to stealth liposomes (range = P > 0.05 to 0.001) with the exception of i.p. drug versus i.p. tumor at adose

100

80

40

20

B 0

10

20

30

40

50

60

70

80

Days survived

Fig. 5. Percentage of surviving B6D2F, mice (5 per group), as a function of days postinoculation of tumor and ara-C dose, for mice receiving 10* L1210/C2 leukemia cells by the i.v. route or the i.p. route and treated 24 h later with single (A) or multiple (B) injections of ara-C entrapped in liposomes composed of HSPC:CH:PEG-DSPE, 2:1:0.1 (0.4 ^m extruded REV). A. mice received i.v. tumor and single i.v. injections of sterile saline (A), 25 mg/kg (O), 50 mg/kg (+), or 100 mg/kg (D) liposome-entrapped ara-C. Some mice received 50 mg/kg (O) or 100 mg/kg (V) 24-h infusions of free ara-C. Phospholipid doses averaged approximately 2. 4. and 10 jjmol/mouse. B. mice received tumor by cither the i.v. (O, D) or the i.p. (O, V) route and 3 injections of 50 mg/kg liposome-entrapped ara-C (-arabinofuranosyladenine cytotoxicity to mouse leukaemia L1210 in vitro by 2'-deoxycoformycin. Cancer Res., 36: 1486-1491, 1976. 35. Paul, M. A., and Dave, C. A simple method for long-term drug infusion in mice: evaluation of guanazole as a model (38488). Proc. Soc. Exp. Biol. Med., 148: 118-122, 1975. 36. Danhauser. L. L., and Rustum, Y. M. A method for continuous drug infusion in unrestrained rats: its application in evaluating the toxicity of 5-fluorouracil/thymidine combination. J. Lab. Clin. Med.. 93: 1047-1053, 1979. 37. Skipper, H. E. Ara-C and Cyclophosphamide: A Closer Look at the Influence of Dose Intensity and Treatment Duration on Host Toxicity and Therapeutic Response (Experimental Data). Booklet 13, pp. 1-133. Birmingham, AL: Southern Research Institute, 1986. 38. Menten, J.. Van der Schueren, E., and Ang, K. K. The toxicity of cytosine arabinoside in mice treated with continuous infusion or push-injections. European Organization for Research on Treatment of Cancer Symposium of Continuous Infusion Chemotherapy, Brussels, Abstract 7. 1985. In: Cancergram, Antitumor and Antiviral Agents, Series CB20, No. 85/11, p. 16. Philadelphia, PA: Cancer Information Dissemination and Analysis Center. 1985. 39. Juliano. R. L., and Stamp. D. The effect of particle size and charge on the clearance rates of liposomes and liposome encapsulated drugs. Biochem. Pharmacol., 27:21-27, 1978. 40. Skipper, H. E. Analyses and Interpretations of the Influence of Dose and Schedule Variables with an S-Phase Specific Drug (Ara-C) on (1) Toxicity, and (2) the Degree and Duration of Response and Cure Rate of Animals Bearing Known Burdens of LI210 Leukemia Cells. Booklet 4, pp. 1-57. Birmingham, AL: Southern Research Institute, 1988. 41. Mayhew, E., Rustum, Y. M.. and Szoka, F. Therapeutic efficacy of cytosine arabinoside trapped in liposomes. In: G. Gregoriadis. J. Senior, and A. Trouet (eds.). Targeting of Drugs. NATO Advanced Study Institutes Series, pp. 249-260. New York: Plenum Publishing Corp., 1982. 42. Gabizon. A., Shiota. R.. and Papahadjopoulos. D. Pharmacokinetics and tissue distribution of doxorubicin encapsulated in stable liposomes with long circulation times. J. Nati. Cancer Inst., 81: 1484-1488. 1989. 43. Kreis, W., Chaudhri. F.. Chan. K.. Allen, S., Budman, D. R., Schulman, P., Weiselberg, L., Freeman, J., Deere, M., and Vinciguerra, V. Pharmacoki netics of low-dose 1-ff-D-arabinofuranosylcytosine given by continuous intra venous infusion over twenty-one days. Cancer Res., 45: 6498-6501. 1985.

2439


Stealth Liposomes: An Improved Sustained Release System for 1- β-d-Arabinofuranosylcytosine Theresa M. Allen, Tarun Mehra, Christian Hansen, et al. Cancer Res 1992;52:2431-2439.

Updated version

E-mail alerts Reprints and Subscriptions Permissions

Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/52/9/2431

Sign up to receive free email-alerts related to this article or journal. To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at [email protected]. To request permission to re-use all or part of this article, contact the AACR Publications Department at [email protected].
