Effects of entrapment in liposomes on the distribution, degradation and effectiveness of methotrexate in vivo

Novel Delivery Systems for Interferons

Interferons, IFNs, are among the most widely studied and clinically used biopharmaceuticals. Despite their invaluable therapeutic roles, the widespread use of IFNs suffers from some inherent limitations, mainly their relatively short circulation lifespan and their unwanted effects on some non-target tissues. Therefore, both these constraints have become the central focus points for the research efforts on the development of a variety of novel delivery systems for these therapeutic agents with the ultimate goal of improving their therapeutic end-points. Generally, the delivery systems currently under investigation for IFNs can be classified as particulate delivery systems, including micro- and nano-particles, liposomes, minipellets, cellular carriers, and non-particulate delivery systems, including PEGylated IFNs, other chemically conjugated IFNs, immunoconjugated IFNs, and genetically conjugated IFNs. All these strategies and techniques have their own possibilities and limitations, which should be taken into account when considering their clinical application. In this article, currently studied delivery systems/techniques for IFN delivery have been reviewed extensively, with the main focus on the pharmacokinetic consequences of each procedure.

Lymphatic delivery and pharmacokinetics of methotrexate after intramuscular injection of differently charged liposome-entrapped methotrexate to rats

The lymph node targeting ability and pharmacokinetics of methotrexate (MTX) after intramuscular (i.m.) injection of differently charged liposome-entrapped [3H]MTX to rats were evaluated using [3H]MTX as a tracer. Neutral liposomes were prepared with a mixture of phosphatidylcholine, cholesterol and alpha-tocopherol (8:4:0.1, molar ratio). Positively and negatively charged liposomes were also prepared by incorporation of stearylamine (8:4:0.1, molar ratio) and dicetylphosphate (8:4:0.1:1, molar ratio) into neutral liposomes respectively. The encapsulation efficiency (as expressed in terms of radioactivity) in liposomes was increased as alpha-tocopherol was incorporated into the lipid bilayer. The disappearance of [3H]MTX from the i.m. injection site was rapid and essentially complete after 30 min. On the other hand, the disappearance of radioactivity of liposome-entrapped [3H]MTX was much slower when compared to free drug. The area under the drug concentration-time curve (AUC) of liposome-entrapped [3H]MTX in lymph nodes was significantly increased when compared to free [3H]MTX. It suggested that liposomes injected by the i.m. route entered into the lymphatics and only drug released from liposomes diffused directly into the systematic circulation. The liposome-entrapped [3H]MTX by i.m. route was markedly localized in the lymph nodes. The concentration of MTX-equivalents in regional lymph node after i.m. injection of liposome-entrapped [3H]MTX was > 100-350 fold higher when compared to the plasma concentration. These values are more than 10-20 fold higher compared to the i.m. injection of free [3H]MTX. The positively charged liposomes were more localized in lymph nodes compared to neutral and negatively charged ones. While liposomes injected by i.v. route were localized in liver, spleen and lung compared to free [3H]MTX, it was evident that i.m. administration of liposomes resulted in enhanced localization of MTX in the lymphatic system but decreased deposition in kidney, liver and other non-targeting tissues compared to free [3H]MTX. The targeting ability and carrier properties of liposome-entrapped anticancer drugs with varying surface charge, lipid compositions and route of administration are of significant importance to alter biodistribution in chemotherapy.

Liposomes as carriers of cancer chemotherapy. Current status and future prospects

Chemotherapy is a modality of cancer therapy that needs much improvement. Development of a new chemical entity is very costly and time consuming, but improvements in delivery of existing agents may yield more rapid clinical results. Liposomes and other lipid-based drug delivery systems have the advantage, in this context, of utilising no new chemical entities. In terms of mechanism of action, tumour targeting has been the focus of much work in liposomal drug delivery. The recent development of liposomes with longer circulation times has led to improved tumour targeting in animal studies. Other mechanisms of action, such as release from drug depot formulations, heat-triggered local drug release, and transfection of genetic materials, may prove to be useful in humans. Liposomal formulations of more than a dozen antineoplastic agents have shown promise in vitro and in animal models. Somewhat mundane, but nevertheless crucial, issues of medical rationale and formulation engineering, and commercial considerations, have slowed testing in patients with cancer. However, 3 antineoplastic agents, doxorubicin, daunorubicin and cytarabine, are in advanced stages of clinical testing in humans. One or more of these should prove to be a medically useful and commercially viable product within the next few years.

Toxicity and antitumor activity of liposome-entrapped retinoid Ro13-7410

We compared the toxicity of free and liposome-entrapped retinoid, Ro13-7410 in C2DF1 mice. Mice received 7 daily i.p. injections of liposome-entrapped Ro13-7410 at doses of 5, 10, 100, 500, or 1000 micrograms/kg bw/day. For comparison, two groups of mice were used as controls, one group received Ro13-7410 in corn oil and a second group received liposome-entrapped Ro13-7410 that had been solubilized with detergent. The liposomes were then tested for chemotherapeutic activity against human myelocytic leukemia (HL-60/MRI) implanted in athymic NCr-nu mice. The doses used in the chemotherapy experiment (20, 50, and 100 micrograms/kg bw/day) were selected based on the results of the toxicity experiment in CD2F1 mice. CD2F1 mice were marginally protected from toxicity after receiving retinoid in liposomes relative to controls. There were 2/5 survivors in the 1000 micrograms/kg bw/day Ro13-7410-liposome group after 7 daily i.p. doses compared to 0/5 for both the corn oil and solubilized liposome groups, and 4/5 survivors in the 500 micrograms/kg bw/day Ro13-7410-liposome group after 7 daily i.p. doses compared to 2/5 for both the corn oil and solubilized liposome groups. We observed no dramatic differences in toxicity among the treatment groups over the range of doses administered. There were 2/6 long-term tumor-free survivors in athymic mice receiving liposome-entrapped retinoid, at 50 micrograms/kg bw/day for 7 days, compared with 0/6 and 0/9 survivors in groups receiving empty liposomes or no treatment.(ABSTRACT TRUNCATED AT 250 WORDS)

Liposomes as drug carrier system for cis-diamminedichloroplatinum (II). II. Antitumor activity in vivo, induction of drug resistance, nephrotoxicity and Pt distribution

In this study we have investigated the use of liposomes as a drug carrier system for cis-diamminedichloroplatinum(II) (cis-DDP) in order to reduce the nephrotoxicity with preservation of antitumor activity. Liposomes (PC/PS/Chol 10:1:4) were prepared using hydration media containing no or a relatively low concentration of NaCl. It was found that cis-DDP containing liposomes (lip cis-DDP) injected i.v. to IgM immunocytoma-bearing LOU/M rats at a dose of 1 mg cis-DDP/kg (cumulative dose 7 mg cis-DDP/kg) showed less antitumor activity than the free drug. The optimal cumulative dose of free cis-DDP for induction of antitumor activity in this tumor system is 7 mg/kg (7 X 1 mg/kg). At a dose of 2 mg lip cis-DDP/kg (cumulative dose 14 mg cis-DDP/kg) the antitumor activity could not be increased by choosing another phospholipid composition of the liposomes [DPPC/DPPG/Chol (10:1:10)]. cis-DDP incorporated in DPPC/DPPG/Chol liposomes showed a similar antitumor activity to cis-DDP incorporated in PC/PS/Chol liposomes. After an i.v. dose of 2 mg lip cis-DDP/kg (PC/PS/Chol) kidney damage was less compared to the treatment with free cis-DDP (1 mg/kg). However, after a single dose of 2 mg cis-DDP/kg or a cumulative dose of 8 or 16 mg cis-DDP/kg, kidneys of rats treated with lip cis-DDP contained twice as much Pt as after treatment with free cis-DDP. Moreover, after treatment with lip cis-DDP, a twofold increase of the amount of Pt in tumor tissue was measured. In vitro studies with Pt recovered from spleens obtained from rats treated with lip cis-DDP i.v. showed that based on the equal amounts of Pt recovered the antitumor activity of the recovered Pt was reduced, indicating inactivation of cis-DDP in vivo. As during treatment with free cis-DDP, recurrence of the tumor was observed during the continued treatment with lip cis-DDP. It was found that these recurrent tumors were resistant to further therapy with cis-DDP. In conclusion, cis-DDP encapsulation into liposomes decreased the nephrotoxicity. The antitumor activity of cis-DDP is preserved by liposome encapsulation when it was used at a dose of 2 mg/kg, but it was reduced in terms of earlier onset of regrowth.

Pharmacokinetics and chemotherapeutic efficacy of adriamycin encapsulated in immunoliposomes against avian myeloblastosis virus infection

Immunoliposomes were prepared using rabbit anti-AMV gp80 IgG for the targeted chemotherapy of avian myeloblastosis virus infection. Adriamycin was encapsulated into immunoliposomes and used for in vivo studies. Comparative pharmacokinetics of free drug, drug encapsulated in free liposomes and of drug encapsulated in immunoliposomes in the virus-infected cells revealed that (i) the drug encapsulated in liposomes was cleared from the plasma slowly, and (ii) the drug encapsulated in immunoliposomes accumulated in the target tissue, the bone marrow, 5- and 8.5-fold more than the drug encapsulated in free liposomes and free drug, respectively. The drug encapsulated in immunoliposomes inactivated the virus and exhibited more chemotherapeutic efficacy as compared to controls when injected up to 24 h post-infection. However, when injected 48 h post-infection the drug encapsulated in immunoliposomes did not offer any protection against the virus infection. There is no detectable antibody response against immunoliposomes in the infected animals.

Porous biodegradable microspheres for controlled drug delivery. I. Assessment of processing conditions and solvent removal techniques

Microspheres containing methylene blue and prednisolone acetate were prepared by one of three methods: freeze-drying, evaporation, and solvent-extraction-precipitation. An extremely porous structure was obtained by the freeze-dry and solvent-extraction-precipitation procedures. The specific surface area of 6.33-microns particles was 20.6 m2/g, or 35 times that of a particle devoid of pores, and the void space was 59-61%. The sphericity, size, and yields of the microspheres were influenced by the preparation procedure, surfactant type and concentration, temperature of the continuous phase, polymer concentration in the dispersed phase, and ratio of marker to polymer. The most suitable processing conditions were a polymer concentration of 5-10%, a marker loading of 10%, 0.1% sorbitan sesquioleate as the surfactant, and temperature adjustment of the continuous phase from 15 to 50 degrees C following the addition of the dispersed phase. Complete release of the highly water soluble methylene blue occurred within 72 hr, while the less soluble prednisolone acetate released much more slowly, i.e., 90% after 7 days. The microspheres remained relatively intact during the in vitro release of methylene blue, confirming that the incorporated agent was confined to the walls of the porous network. Collapse of the polymer structure was evident after 7 days. The release therefore was believed to be governed principally by the solubility of the drug and the porosity of the matrix.

The effect of niosomes and polysorbate 80 on the metabolism and excretion of methotrexate in the mouse

The effect of non-ionic surfactant vesicle (niosome) encapsulation on the metabolism and urinary and faecal excretion of methotrexate (MTX) in mice has been studied following oral and intravenous administration, and compared with the effects of co-administration of free drug and polysorbate 80, which does not form vesicles. Niosome entrapment reduces the excretion of MTX into urine and bile whereas polysorbate 80 increases its excretion. Monitoring of the levels of MTX and its 7-hydroxy metabolite indicates that entrapped MTX is protected from rapid metabolism in vivo, particularly in niosomes but to a small degree in the micellar systems formed by polysorbate.

The effect of non-ionic surfactant vesicle (niosome) entrapment on the absorption and distribution of methotrexate in mice

Non-ionic surfactant vesicles (niosomes) prepared from a non-ionic surfactant, cholesterol and dicetyl phosphate and containing methotrexate (MTX) have been administered to mice. Given intravenously the niosomes prolong the levels of MTX in the blood, large amounts of the drug being taken up by the liver. There was also an increased uptake of MTX into the brain, perhaps due to an effect of the niosome components on the permeability of the blood brain barrier. Absorption of the drug from the gastrointestinal tract following oral ingestion, appeared to be increased at some doses; most of the entrapped MTX was taken up by the liver, but uptake of MTX into the brain was also increased. The metabolic profile of the drug is altered by the niosomes which appear to prevent the rapid formation of 7-hydroxy methotrexate.

Liposomal Methotrexate in the Treatment of Murine L1210 Leukemia

To assess the therapeutic effectiveness of methotrexate (MTX) when administered in a liposome carrier, mice bearing intracranial L1210 leukemia were tested with liposomal MTX, free MTX, or saline. Single i.p. injections of liposomal MTX at doses of 5 mg/kg and 2.5 mg/kg prolonged survival of mice bearing intracranial L1210 leukemia. The same doses of the free drug did not prolong survival of the tumor-bearing mice. This system may have clinical application not only for MTX, but also other polar anticancer agents in the treatment for central nervous system malignancy.

In vitro uptake and therapeutic application of liposome-encapsulated methotrexate in mouse hepatoma 129

The biological activities of liposome-encapsulated and non-encapsulated methotrexate (MTX) were compared in vitro and in vivo using mouse Hepatoma 129 ascites tumor cells. Under in vitro conditions, cells accumulated up to 29% of [3H]-MTX when the drug was incorporated in the lipid bilayers of positively charged, unilamellar liposomes. There was no significant uptake of non-encapsulated MTX under the same conditions. A single intraperitoneal injection of liposome-encapsulated MTX (3 mg MTX/kg) increased the mean survival time of tumor-bearing mice to 42.5 +/- 11.2 days, compared to 23.5 +/- 2.2 days for untreated controls. Non-encapsulated MTX had no significant effect on survival time. Thus the in vivo treatment studies appear to agree with the in vitro uptake measurements. Addition of galactolipids to the lipid bilayers of liposomes did not increase in vitro uptake of encapsulated MTX and gave no additional improvement in therapeutic effectiveness. Encapsulation of MTX in liposomes might thus be used to increase uptake of the drug in cells which may be deficient in MTX transport.

Interactions of liposomes with the reticuloendothelial system. II: Nonspecific and receptor-mediated uptake of liposomes by mouse peritoneal macrophages

Liposomes are taken up as intact vesicles by mouse peritoneal macrophages in a process which is temperature sensitive and is affected by inhibitors of glycolytic metabolism and of microfilament activity. Macrophages take up negatively charge vesicles more readily than positively charged vesicles (2-fold) or neutral vesicles (4-fold). Macrophages take up similar amounts of multilamellar liposomes, reversed phase liposomes and small unilamellar liposomes in terms of lipids, however this corresponds to vastly different numbers of particles and amounts of trapped volume. Coating the liposomes with macromolecular ligands capable of interacting with macrophage surface receptors can markedly promote liposome uptake. Thus, formation of an IgG-antigen complex on the liposome surface results in a 10(2)-fold enhancement of liposome uptake, while coating the vesicles with fibronectin results in a 10-fold augmentation of uptake. Uptake via IgG-mediated and fibronectin-mediated processes seem to be independent since excess unlabelled, IgG-coated liposomes will inhibit the uptake of radioactively-labelled IgG-coated liposomes much more effectively than the uptake of radioactively-labelled fibronectin-coated liposomes. Cell-bound liposomes can readily be visualized on and inside of the macrophages using fluorescence microscopy techniques.

Increased therapeutic efficiency of a lipid-soluble alkylating agent incorporated in liposomes

A highly hydrophobic alkylating agent, 1-N,N-bis(beta-bromoethyl) amino-3-methylnaphthalene, given as the free drug in oil, cured a substantial proportion of mice bearing the PC6 myeloma in the dose range 2-7 mg/kg. However, these doses were toxic, and the LD50 was 6-7 mg/kg. When incorporated in liposomes, similar curative effects were obtained at doses of 10-41 mg/kg without material toxicity, even at the highest dose. Liposome entrapment therefore greatly increases the therapeutic efficiency of this agent.

Encapsulation of immunoadjuvant [14C]peptidoglycan monomer into liposomes. Effect on metabolism and immune response in mice

14C-labeled peptidoglycan monomer was encapsulated into negatively charged, multilamellar liposomes composed of egg phosphatidylcholine, cholesterol and dicetylphosphate. Excretion and tissue distribution of the label in mice were studied after intravenous injections. Encapsulation of peptidoglycan monomer into liposomes as compared to free peptidoglycan monomer, resulted in increased retention of the label, particularly in the liver and to a lesser extent in spleen. The excretion was drastically reduced and delayed even after 4 days when cholesterol-rich (phosphatidylcholine/cholesterol, 7:5 molar ratio) liposomes were used for encapsulation of peptidoglycan monomer. Peptidoglycan monomer and liposomes, when tested separately, stimulate the immune response to sheep erythrocytes in mice. However, there was no significant additive or synergistic effect when peptidoglycan monomer was encapsulated into liposomes.

Effect of liposomally trapped antitumour drugs on a drug-resistant mouse lymphoma in vivo

A TLX-5 mouse lymphoma which was resistant to 1-β-D-arabinofuranosyl cytosine (AraC) was used in vivo to study the possibility of using liposomes as drug-delivery vehicles in order to overcome drug resistance.The effects of free drugs (AraC, AraCTP and methotrexate) and the liposomally associated drugs on the survival time of tumour-bearing mice were determined.As a more sensitive measure of cell survival, (125)IUdR was incorporated into the DNA of the ascites TLX-5 cells before i.p. injection. Cell survival and the cytotoxic effects of the drugs on the tumour cells were determined by using a double-headed gamma counter to measure the retention of the (125)I label.Both AraC and AraCTP, either as the free drugs or liposomally associated, had no effects on the tumour. Due to the lack of response of tumour cells to these drugs, further studies were initiated with free and liposomally associated methotrexate (MTX), a drug to which the cells were known to be sensitive. It was found that the liposomally associated MTX, at a 5-10-fold lower dose than the free drug, was (a) more effective in prolonging the survival of tumour-bearing mice and (b) as effective as the free drug in killing tumour cells (as measured by the (125)I retention).In vivo MTX was more effective in the liposomally associated form, whereas liposomally entrapped AraC and AraCTP were ineffective. It is proposed that in vivo liposomally associated drugs may be acting not by actively localizing in the tumour cells, but by the liposomes providing a slow-release drug depot, improving the pharmacokinetic properties of MTX.

Altered pharmacological properties of liposome-associated human interferon-alpha

Human interferon-alpha was associated in different ways with positively (stearylamine) and negatively (phosphatidylserine) charged phosphatidylcholine multilamellar vesicles, depending on the presence or absence of a cholesterol component. Inclusion of cholesterol resulted in interferon that was significantly (P = 0.0001) more deeply internalized within the liposomes, such that detergent disruption was necessary before most of the interferon activity was expressed. Interferon was stably associated with stearylamine-containing liposomes, both with and without a cholesterol component. However, inclusion of cholesterol in the phosphatidylserine-containing liposomes was necessary for stable association of the interferon for more than 2 days at 4 degrees C or for more than 24 h at 37 degrees C. After intramuscular injection into mice, liposome-associated interferon in reverse-phase evaporation vesicles was retained at the local site of injection significantly longer than free interferon. Even 3 days after intramuscular injection, stearylamine-containing liposomes with or without cholesterol resulted in local interferon levels that were comparable to the peak levels obtained 2 to 4 h after free interferon was injected. In contrast, free interferon was not detectable in the local muscles 24 h after injection of 10(4.6) U. Liposomes containing phosphatidylserine and cholesterol resulted in intermediate levels of local interferon retention; without a cholesterol component, phosphatidylserine-containing liposomes resulted in no increased local interferon retention compared with the results when free interferon was injected.

Interactions of liposomes with the reticuloendothelial system. Effects of reticuloendothelial blockade on the clearance of large unilamellar vesicles

Large unilamellar liposomes (also called reversed phase vesicles or REVs) composed of DPPC and cholesterol are cleared from the circulation of the rat by a process which closely resembles the clearance of colloidal particles by the reticuloendothelial system. Thus, increasing the total amount of REVs administered by giving a loading dose of unlabelled REVs slows the clearance of a test dose of radioactively labelled REVs. This resembles the reticuloendothelial 'blockade' induced by large doses of colloids. Administration of other types of particles known to induce reticuloendothelial blockade, such as latex beads and xenogeneic red cells, also slows the clearance of radioactively labelled REVs. Administration of small unilamellar liposomes (SUVs) can also cause blockade of REV clearance, but the onset of blockade is delayed until a substantial fraction of the SUVs have been removed from the circulation. Blockade caused by the administration of large doses of REVs seems to result from a direct action on reticuloendothelial cells rather than from depletion of opsonic factors in the blood. Partial blockade of REV clearance produces a modest alteration of the tissue distribution of REVs, with enhanced uptake in the lungs. These results suggest that both REVs and SUVs are taken up by the reticuloendothelial system via a process which closely resembles the clearance of other types of colloids.

Perturbation of lipid membranes by organic pollutants

The ability of a range of organic pollutants--hexachlorobenzene, mirex(1,1a,2,2,3,3a,4,5,5,5a,5b,6-dodecachlorooctahydro-1,3,4-metheno-1H cyclobuta(cd) pentalene), 1,3,5-trichlorobenzene, 2,4,6-trichlorophenol, p-nitrophenol, p-chlorophenol, DDT, and pentachlorophenol--to perturb liposomes of dipalmitoyl phosphatidylcholine (DPPC) has been measured by differential scanning calorimetry. The degree of perturbation was measured by the increase in breadth of the main DPPC phase transition in both heating and cooling scans. DDT and the phenol derivatives were effective perturbers of phospholipid, broadening the transition by as much as 12-fold. Hexachlorobenzene and mirex did not perturb at all when mixed with DPPC at concentrations as high as 20 mol%, although 1,3,5-trichlorobenzene caused slight broadening of the main transition at this concentration. Perturbation is facilitated by the presence of a hydroxyl group on the benzene ring and hindered by increasing degrees of chloride substitution. An apparent correlation exists between the extent of phospholipid perturbation measured by differential scanning calorimetry and LD50 values for these compounds taken from the literature. This suggests the possibility of formulating an "index of perturbation" which could be used to screen certain classes of organic compounds for potential biological toxicity on a routine basis.

Binding and capture of human interferon-alpha by reverse evaporation vesicles, multilamellar vesicles, and small unilamellar vesicles

Liposomes of various physical and chemical compositions were prepared and their ability to externally bind and internally capture human interferon alpha (HuIFN-alpha) was determined. HuIFN-alpha was bound by preformed liposomes composed of either dipalmitoyl phosphatidyl choline alone or dipalmitoyl phosphatidic acid (calcium salt), and cholesterol, with or without a phosphatidyl choline component. HuIFN-alpha could also be internally captured within liposomes both of compositions that bound or did not bind interferon. The interferon could be associated with liposomes in at least three manners: bound to the outside surface of the liposome, associated partially within the liposomal membranes, or completely internalized either within the aqueous compartments of the liposome or completely buried within the liposomal bilayer membrane. HuIFN-alpha was stably-associated with reverse-evaporation vesicles, multilamellar vesicles, and small unilamellar vesicles for 30 days at 4 degrees C. Incubation of the multilamellar vesicles at 37 degrees C caused an initial decrease in the amount of externally bound interferon, and incubation with mouse serum caused a further dissociation of the externally-bound interferon. Depending on liposome composition, incubation at 37 degrees C either had little effect on stability of internalized interferon or caused leakage of internalized interferon.

pH-sensitive liposomes: possible clinical implications

When pH-sensitive molecules are incorporated into liposomes, drugs can be specifically released from these vesicles by a change of pH in the ambient serum. Liposomes containing the pH-sensitive lipid palmitoyl homocysteine (PHC) were constructed so that the greatest pH differential (6.0 to 7.4) of drug release was obtained near physiological temperature. Such liposomes could be useful clinically if they enable drugs to be targeted to areas of the body in which pH is less than physiological, such as primary tumors and metastases or sites of inflammation and infection.

Erythrocytes and lymphocytes as drug carrier systems: techniques for entrapment of drugs in living cells

Mouse thymocytes and erythrocytes are loaded electrically with drugs in isotonic solution. The loaded cells are used for targeting the drugs to specific sites in the organism in order to achieve a controlled drug release in time and space. The field technique used for the loading of the cells is based on the dielectric breakdown of the cell membrane which is observed when cell suspensions are subjected to external field pulses of 2-20 kV/cm for short time intervals (ns to microseconds). When an apparent membrane potential of about 1 V is reached in response to the external field, the membrane breaks down reversibly. The breakdown of the membrane is associated with a remarkable and reversible permeability increase of the cell membrane. The increase in permeability depends on the strength and the duration of the field pulse.

Liposomes and local hyperthermia: selective delivery of methotrexate to heated tumors

Liposomes with phase transitions a few degrees above physiological temperature delivered more than four times as much methotrexate to murine tumors heated to 42 degrees C as to unheated control tumors. Most of the accumulated drug appeared to be intracellular and bound to dihydrofolate reductase, the enzyme blocked by methotrexate in its role as an antineoplastic agent.