JOURNAL OF
`PHARMACEUTICAL
`SCIENCES
`
`A publication of the American Pharmaceutical Association
`
`November 1990
`Volume 79, Number 11
`
`REVIEW ARTICLE
`
`Drug Targeting in Cancer Chemotherapy: A Clinical Perspective
`
`PRAMOD K. GUPTA*
`Accepted for
`Received A ril 28, 1989, from the Co//ege of Pharmacy, Universi of Kentucky, Rose Street, Lexington, KY 4053M082.
`‘Present address: Drug Delivery gystems Research, Abbott Laboratories, Department 99P, AP-4,
`publication Larch 27, 1990.
`Abbott Park, IL 60064.
`
`Abstract 0 Drug targeting is a phenomenon which maneuvers the
`distribution of drug in the body in such a manner that the major fraction
`of the drug interacts exclusively with the target tissue at a cellular or
`subcellular level. Numerous strategies have been developed to accom-
`plish this goal; some of them have been tried clinically for improving
`cancer chemotherapy. This review updates the current status of research
`in the area of targeted drug delivery, with particular emphasis on its
`application in the clinical management of carcinomas.
`
`Currently, the development of techniques which could
`selectively deliver drug molecules to diseased vasculature,
`without concurrent increase in its levels in the healthy tissues
`of the body, is one of the most enthusiastically and exhaus-
`tively pursued areas of research in experimental pharmacol-
`ogy and therapeutics. This interest is particularly focused
`towards optimizing the delivery of chemotherapeutic agents,
`which possess little intrinsic ability to discriminate cancer
`cells from the healthy cells.1 As a result of this nonselectivity,
`the iv administration of chemotherapeutic agents almost
`invariably causes dose-dependent systemic toxicities, often
`warranting discontinuation of the treatment, and thus fails to
`allow successful eradication of cancer cells.24
`For successful cancer treatment, “total-cell kill” is imper-
`ative.5 This means total excision of tumor for surgical cure,
`and complete destruction of all cancer cells by chemotherapy
`and/or radiotherapy. Failure to accomplish this goal has been
`shown to permit development of minimal residual neoplastic
`disease.6 To date, oncologists face the challenge of developing
`strategies and treatment schedules which may provide an-
`swers to these problems. In practice, the treatment of cancer
`cells has been complicated by several factors [e.g., existence
`of subpopulations of neoplastic cells within a tumor which
`considerably differ in their morphology, immunogenicity, rate of
`growth, capacity to metastasize, and response towards chemo-
`therapeutic agent(s)l.5 In addition, the total blood flow, the rate
`of perfusion, and vessel permeability may vary within different
`regions of the same tumor tissue.’-15 These physiological bar-
`riers severely limit the possibility of predicting an effective
`dosage regimen for the treatment of most tumors.
`
`Following repeated disappointments with iv administra-
`tion of cytotoxic agent(s), both in terms of acute immediate
`toxicities and delayed dose-related complications, several
`workers evaluated the therapeutic benefits of different routes
`of cytotoxic drug administration. Initially, it was proposed
`that the intra-arterial (ia) administration of cytotoxic agents
`may improve cancer therapy because it delivers a major
`fraction of the dose directly to the target cells.1618 However,
`information on the pharmacokinetics of ia drug delivery
`eventually suggested that this route of drug administration is
`beneficial only in situations when cLBIQ > 1, where CLB is
`the total body clearance of drug and Q is the arterial blood
`flow rate.19.20 According to this relationship, the higher this
`quotient, the greater is the therapeutic advantage. However,
`these predictions often do not consider the consequences of
`drug loss due to the existence of vascular shunts and leaky
`capillaries in the tumor tissue. This phenomenon, called drug
`streaming, has often been suggested to offset the advantages
`anticipated as a result of the ia administration of drug.21
`The clinical implications of intrathecal drug administra-
`tion for the chemotherapy of brain metastases, and ip drug
`administration for treating the carcinomas of the peritoneum,
`have also been evaluated.19 It has been suggested that these
`routes of regional delivery are useful only for drugs which
`exhibit much higher CLB relative to their permeability in
`target tissue (K), because their therapeutic advantage ap-
`proximates (CLBIfi) + 1, where A is the surface area
`constant.
`Despite extensive experimental investigations and numer-
`ous theoretical considerations,~~~20~~~-~~
`the overall success in
`routine cancer therapy has been nominal. Often aggressive
`dosing regimens involving multiple cytotoxic agents, with or
`without radiotherapy, have been tried with the hope of
`eradicating the complete tumor population.30-36 However, in
`most cases the combinations of chemotherapeutic agents,
`their doses and dosing interval, and the dose and duration of
`radiotherapy, have been empirical.37 This makes the evalu-
`ation and comparison of two different treatment schedules,
`directed towards the same cancer cells, practically difficult.
`Chemotherapy, particularly in the case of cervical, pancre-
`atic, renal, and small cell lung cancer, has been disappoint-
`
`0022-3549/90/1100-0949$0 1 . OO/O
`0 1990, American Pharmaceutical Association
`
`Journal of Pharmaceutical Sciences I 949
`Vol. 79, No. 1 7 , November 1990
`
`IMMUNOGEN 2183, pg. 1
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`ing.a.39 Acute side-effects and complications as a result of
`radiotherapy have also been reported.40 These difficulties
`have urged the need for developing “specialized systems”
`which may allow selective delivery of one or more chemo-
`therapeutic agents to cancer cells, without affecting the
`physiology of the normal cells. To meet these requirements,
`numerous targeted drug delivery systems have been proposed
`over the last two decades. Detailed discussion on each of these
`systems is beyond the scope of this review; however, specific
`information on their development, evaluation, and current
`status in therapeutics can be found in the literature.4142
`Selective or Targeted Drug Delivery
`Theoretically, selective or targeted drug delivery systems
`can improve the outcome of chemotherapy due to the follow-
`ing processes: (1) by allowing a maximum fraction of the
`delivered drug molecules to react exclusively with the cancer
`cells, without having any hannful effect on the normal cells;
`and (2) by allowing preferential distribution of drug to the
`cancer cell. The first process can be classified as absolute drug
`targeting. In the second process, complete eradication of
`cancer cells is not possible without some degree of destruction
`to the normal cells; this process therefore falls in the category
`of partial drug targeting. This classification is simpler and
`more general, particularly in terms of clinical response,
`compared with previously defined active versus passive or
`vascular versus extravascular drug targeting processes7-6-
`because the ultimate aim of targeted drug delivery in che-
`motherapy is the eradication of cancer cells. Hence, a delivery
`system may allow extravascular and possibly intracellular
`transport of drug, and the process may therefore be considered
`as the highest degree of selective drug delivery theoretically
`possible; but, unless it eradicates the total population of
`target cells, without hampering the viability of normal cells,
`it cannot be considered as absolute drug targeting from the
`standpoint of therapeutic response.
`Depending on the site of delivery of drug in the target
`tissue, (i.e., intracellular or extracellular), Heath et al.69 have
`defined targeting as being a carrier-dependent or carrier-
`independent phenomenon. In the former case, after its local-
`ization in the target tissue, the drug camer is taken up by the
`target cell(s) and the drug is released intracellularly in a
`controlled manner; however, in the latter case, the drug is
`released from the carrier extracellularly and hence the drug
`action inside the target cells is not influenced by the ability
`(or inability) of carrier to be taken up by these cells. In lieu
`of this discussion, it is clear that carrier-dependent targeted
`delivery may allow utilization of drugs which are active
`intracellularly, but are normally discarded due to their poor
`intracellular influx. In such situations, appropriate selection
`of drug carrier may allow greater influx of drug to the
`intracellular components and hence increase the overall
`efficacy of drug delivery.
`The literature on drug targeting indicates that in most
`cases the efforts involving the design and development of
`targeted drug delivery systems have been aimed at accom-
`plishing absolute drug targeting. Limited success with some
`of the initially designed delivery systems-5
`eventually led
`to the development of more complex devices, often involving
`multiple processes both during their formulation and thera-
`peutic action.-B A classical example of such a complexity is
`the development of heat-sensitive immunoliposomes.~7 The
`formulation of this delivery system involves synthesis of
`drug-loaded liposomes using heat-sensitive lipoidal compo-
`nents, followed by their coating or conjugation with target-
`cell specific antibodies. After the removal of unconjugated
`antibodies by gel chromatography, the formulation is injected
`in such a manner that it reaches predominantly to the target
`
`950 I h m a l of Pharmaceutical Sciences
`Vd. 79, No. 11, November 1990
`
`area and the interaction between target cell antigens and
`liposomal antibodies may be conferred. This process is com-
`plemented by external heating of the target tissue so that
`near the lipid transition temperature the drug is released
`from the liposomal bilayers.67 Efforts have also been made to
`develop pH-sensitive immunoliposomes which destabilize
`and fuse following endocytosis by the target cells due to the
`acidic nature of the microenvironment.69.70 Theoretically,
`such systems, and several others involving similar multiple
`processes, are appealing because they address some of the
`physiological difficulties and offer the possibility of selective
`delivery of chemotherapeutic agents. However, the overall
`clinical success of such complex systems may diminish in
`practice due to the high degree of perfection and specificity
`and the reproducibility of these processes, necessary a t each
`stage, starting from their formulation to the release of drug
`and specific drug-cell interaction desired in vivo. Moreover, it
`is important to consider that if such a system did prove to be
`successful in a particular experimental situation, the chances
`of it being commercially available and the probability of it
`being applicable to more frequently encountered cancers are
`scant. Considering these facts, it is not surprising why so few
`targeted drug delivery systems have reached the stage of
`clinical investigations.
`Clinical Evaluation of Targeted Drug Delivery
`Systems
`Water-in-oil emulsions (wlo), liposomes, starch micro-
`spheres, ethylcellulose microcapsules, albumin microspheres,
`and polymethacrylic nanoparticles are notable examples of
`the systems which have been clinically tested for their
`feasibility in the targeted delivery of chemotherapeutic
`agents (see Table I). The current status of each of these
`systems is described below. Other experimental approaches
`which have shown promise in optimizing the delivery of
`chemotherapeutic agents are briefly discussed towards the
`latter part of this review.
`local administration water-in-oil
`Emulsions-Following
`(wlo) or fat emulsions generally localize at the site of injection.
`Hence, these formulations are expected to minimize rapid
`distribution of incorporated drug to the systemic tissues.71.72
`With this rationale, Takahashi et al.71 investigated the
`usefulness of a wlo emulsion (composition: 65% sesame oil,
`30% aqueous drug solution, and 5% Span 80) and a multiple
`water-oil-water emulsion (wlolw; composition: 65% water,
`19.5% sesame oil, 9% aqueous drug solution, 5% Pluronic F68,
`and 1.5% Span 80) for the delivery of bleomycin and mito-
`mycin C in patients suffering from squamous cell carcinoma
`of skin or recurred adenocarcinoma of breast. The doses of
`drug varied between 2 and 15 mg per injection and up to 13
`injections were tried. After treatment, three patients had
`complete remission of tumor, and others demonstrated mod-
`erate to partial regression of the tumor lesions. No side-effects
`(e.g., leukopenia or anorexia) were observed in any patient;
`however, the injections were ~ a i n f u l . ~ l
`Hashida, Sezaki, and co-workers have experimentally com-
`pared the efficacy of oil alone, olw emulsion, wlo emulsion, and
`gelatin-based wlo emulsion for improving the transfer of
`incorporated radiomarker to lymph nodes.TS75 The rate of
`transfer of radiomarker increased in the following rank order:
`oil > olw emulsion > w/o emulsion > gelatin-based wlo
`emulsion.73 They also found that compared with the aqueous
`solution of marker, the injection of wlo and gelatin-based wlo
`emulsion into stomach wall increased the area under the
`concentration-time curve (AUC) for the regional lymph nodes
`by 1.7 and 5.5 times, respectively.74 However, the inability of
`these systems to reproducibly control the in vivo distribution
`of drug and their inadequate pharmaceutical stability have
`
`IMMUNOGEN 2183, pg. 2
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`Table I-Summary of Cllnlcally Investigated Mlcrocarrler Systems for Cancer Chemotherapy'
`Microcarrier System
`Used for Drug
`Targeting
`Liposomes
`
`Agent
`
`Chemical Structure
`
`Adriamycin
`
`P
`
`I
`OH
`
`
`
`FOCH,OtI
`
`in Water
`
`Solubleb
`
`Site or Type of Treatment
`
`Reference(s)
`
`Lung and pancreatic
`adenocarcinoma; breast,
`colon or multiple
`myeloma
`
`99-104,106
`
`Starch microspheres
`
`Polymethacrylic
`nanoparticles
`Liposomes
`
`Insoluble
`
`Colon and hepatic
`carcinoma
`Hepatoma
`
`Non-small cell lung cancer;
`head and neck cancer;
`breast cancer;
`mesothelioma
`
`136, 137
`
`161
`
`120
`
`Slightd
`
`Starch microspheres
`
`Metastatic liver cancer
`
`133
`
`Solubleb
`
`w/o Emulsion
`
`0
`
`Liposomes
`
`Breast carcinoma; neck
`cystic hygroma
`Cerebral glioma
`
`71, 72
`
`97, 105
`
`BCNU"
`
`Bleomycin
`
`tN+~~2
`
`Solubleb
`
`Albumin microspheres Hepatic sarcoma
`
`153, 154
`
`5-Flurouracil
`
`I'
`
`Ethylcellulose
`microcapsules
`Starch microspheres
`
`Maxillary sinus carcinoma;
`squamous carcinoma
`Hepatic carcinoma
`
`145
`131
`
`Slightd
`
`Solubleb
`
`Starch microspheres
`
`Colon and hepatic
`carcinoma
`
`136. 137
`
`Mitomycin C
`
`OH H
`
`Hepatic and colon
`carcinoma
`Breast, cervical, gastric,
`hepatic, prostatic, renal,
`urinary bladder
`carcinomas
`152
`Hepatic carcinoma
`Albumin microspheres
`a An investigational cancer chemotherapeutic agent (NSC-251635) has been clinically tested for targeted drug delivery via liposomes (see refs 79
`and 80 for details). Solubility 25% (w/v) in water. Refers to 1,3-bis(2-chloroethyl)-l-nitrosourea or carmustine. Solubility 5 1 % (w/v) in water.
`
`Solubleb
`
`Starch microspheres
`
`Ethylcellulose
`microcapsules
`
`1 35-1 37
`139-144
`
`Journal of Pharmaceutical Sciences / 951
`Vol. 79, No. 7 7 , November 1990
`
`IMMUNOGEN 2183, pg. 3
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`hampered their wide clinical investigation.
`Liposomes-Liposomes are unilamellar or multilamellar
`systems consisting of alternative lipid and aqueous layers;
`hence, they allow the incorporation of lipophilic as well as
`hydrophilic drugs within their matrix.76 Current liposome
`formulation technology permits manufacture in the size
`range such that their in vivo deposition may be controlled in
`favor of liver, spleen, or lung. The liposomal formulations can
`be sterilized by membrane filtration (0.20 or 0.45 pm)77,78 or
`yradiation,79@J The detrimental effects of the radiation pro-
`cess on the stability of liposomes are not fully understood.
`These developments in liposome technology have prompted
`researchers to collect information regarding their fate in
`humans as a function of the route of administration.81-88
`Several studies have clinically used radiolabeled liposomes
`for diagnostic purposes (e.g., to compare the phagocytic
`activity of tumor versus normal cells).89-94 Using 99m-Tc-
`labeled multilamellar vesicles (MLV), it has been shown that
`the patients with Hodgkin's disease accumulate higher levels
`of radioactivity in bone marrow and lungs than those suf-
`fering from other malignancies. This differential distribution
`pattern of liposomes has been suggested to be due to altered
`activity of the reticuloendothelial system ( R E 3 in patients
`with Hodgkin's disease.83 Present et a1.95 have shown that
`In111-labeled small unilamellar vesicles (SUV) generally
`accumulate more readily in tumors than in normal tissues.
`Their study demonstrated a tumor:blood and tumor: fat dis-
`tribution ratio in the range of 10 to 14. Similar results have
`also been demonstrated by Gregoriadis and co-workers,76
`using human serum albumin entrapped and 131I-labeled
`liposomes. Large doses (up to 300 mg) of labeled liposomes
`have been tried to improve tumor localization following
`saturation of the RES,B' and the blood profiles of patients have
`generally demonstrated biexponential decline in the levels of
`radioactivity with the initial and the terminal half-life (t1,2)
`approximating 3 and 10 h, respectively.81
`More recently, attempts have been made towards the
`clinical investigation of liposomes for the selective delivery of
`chemotherapeutic agents (e.g., adriamycin, bleomycin, NSC-
`251635),79,80.96-106 antifungal agents,l07-112 biological re-
`sponse modifiers,113-115 enzymes,ll6 and other pharmaceuti-
`cal agents.84.117 In one study, the intracerebral injection of
`liposomal bleomycin in a patient with cerebral glioma con-
`siderably reduced the blood and urinary levels of the drug as
`compared with the levels observed following the administra-
`tion of free drug.105 When 0.1 mg of drug was given intratu-
`morally as a free solution, both blood levels as well as urinary
`excretion of drug could be monitored. However, when the
`same dose of drug was given via liposomes, neither blood nor
`urinary levels could be detected. In general, compared with
`the free drug, liposomal drug reduced blood levels and urinary
`excretion by 3 to 10 times (see Figure 11.105 In addition, no
`adverse effects were recorded with liposomal drug delivery
`over the period of 4 weeks either in terms of focal neurological
`deficits or changes in intracranial pressure.105
`The patients suffering from hematological malignancies
`are known to develop fungal infections involving the RES.
`Amphotericin B is the drug of choice for the treatment of this
`infection; however, its conventional use is associated with
`acute as well as chronic toxicities. Hence, the feasibility of
`liposomal delivery of this compound has been investigated.
`Using a population of 12 patients with fungal infection
`resistant to the free drug, Lopez-Berestein and co-workers107
`intravenously administered 0.8 to 1.0 mglkg of drug en-
`trapped in liposomes every 1 to 3 days. The liposomes were
`composed of 7:3 dimyristoyl phosphatidylcholine (DMPC) and
`dimyristoyl phosphatidylglycerol (DMPG) and were 0.5 to 6.0
`pm in size. The number of injections per patient varied
`between 4 and 43, and the concentration of drug in each
`
`952 I Journal of Pharmaceutical Sciences
`Vol. 79, No. 11, November 1990
`
`0.1
`
`0.45
`
`2
`
`@IB
`r
`n n
`
`r
`
`60
`
`O J
`
`(***I
`2
`0.45
`0.1
`Dose of Bleomycin (mg)
`Flgure 1-Peak concentration (A) and urinary excretion (B) of bleomycin
`in 24 h following the intracerebral injection of increasing doses of free
`drug and drug entrapped in negatively charged liposomes. All data
`represent values obtained from one patient with cerebral glioma. Key:
`(113) free drug; (w) liposomal drug; (') value <1 ng/mL; (") these values
`are mean of data obtained from two patients; (,**) drug levels could not
`be detected. (Adapted in part from ref 105, with permission from the
`authors.)
`
`injection varied between 0.4 and 1.5 mglkg. The cumulative
`dose of liposomal amphotericin B per patient varied between
`0.145 and 3.1 g, and that of lipid varied between 1.47 and 66.5
`g. The liposomal drug delivery was associated with mild fever
`and chills in two patients which were considerably less than
`that routinely observed with the equitoxic doses of free
`drug.107 Of the total patients treated, 25% (3/12) demon-
`strated complete remission and 42% (5/12) demonstrated
`partial remission of the infection. The remaining patients
`(33% or 4/12) did not respond to this therapy. This study did
`not reveal any indication of either short- or long-term toxicity
`during the treatment schedule.lO7 Another study by this
`group repeated the same liposomal drug dosing strategy in
`eight patients with fungal infections and demonstrated 100%
`regression of lesions in all the patients.112 These studies
`provided strong evidence to support the application of the
`liposomal drug delivery approach for the treatment of man-
`ifestations involving the RES.
`In a separate study, 15 patients received 0.2 to 1 g of
`amphotericin B via liposomes.118 The daily volume of lipo-
`somes infused varied between 0.09 and 0.26 L, and the
`duration of treatment ranged between 1 and 20 days.
`Whereas 1.2 mglkg of free drug resulted in peak serum levels
`of 2.5 pg/mL, the liposomal drug delivery resulted in peak
`levels z 5 pg/mL which increased to 10 pg1mL after multiple
`dosing.118 In fact, a single dose of 0.5 to 1.5 mgkg of
`amphotericin B via liposomes maintained sustained drug
`levels of 2 pg/mL in serum for 3 to 4 days.118 The liposomal
`therapy was generally well tolerated with no cardiac, pulmo-
`
`IMMUNOGEN 2183, pg. 4
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`nary, or CNS toxicity during or post treatment. Three pa-
`tients from this study could not be evaluated for overall
`response due to early deaths which were related to noninfec-
`tious reasons. Five patients failed to respond to this therapy.
`Of the remaining seven patients, three demonstrated definite
`improvement and others demonstrated partial response
`towards the therapy.118 A recent study from the same group,
`which involved repeated doses of amphotericin B (2 to 4
`mglkglday) via liposomes to eight patients for 4 to 16 days, has
`demonstrated a positive response in at least three subjects.119
`One patient failed to respond and in other cases the outcome
`could not be evaluated. The cumulative dose of drug per
`patient in this study varied between 0.96 and 2.5 g and the
`volume of liposomes varied between 1.9 and 8 L.119
`Coune, Sculier, and ~o-workers79~80 have investigated the
`clinical utility of liposomes in the delivery of a water-
`insoluble chemotherapeutic agent, NSC-251635. In one study
`which involved five patients with adenocarcinoma and eryth-
`roleukemia, iv infusion of 139 to 300 mg of drug via liposomes
`did not demonstrate any therapeutic advantage.79 In another
`study involving 14 patients, the increase in drug dose to up to
`450 mg/m2 body surface area, which was administered as
`infusion via liposomes (100 to 500 mL/h), did not improve the
`antitumor activity of this delivery system.80 However, these
`studies did reveal that liposomes can be safely administered
`to patients in large volumes.79.80
`Adriamycin is a potent anticancer agent with activity
`against a variety of carcinomas. However, the successful use
`of this compound is severely limited by its toxic side-effects,
`particularly dose-related cardiomyopathy. Hence, efforts
`have been made to use liposomes for alleviating the car-
`diotoxic effects of this drug without affecting its potency as an
`anticancer agent. In one study, 5 to 24 mg of adriamycin was
`given via SUV or MLV as a hepatic arterial infusion to 10
`patients with hepatic metastatic tumors.106 Only 20% (2110)
`of the patients demonstrated a positive response towards this
`drug delivery and the overall antitumor activity was limited,
`possibly due to small amounts of drug administered per dose
`(only one dose was given per patient).106 Another Phase I
`studyloo involving 19 patients with primary and metastatic
`liver cancer resistant to free adriamycin investigated 20- to
`70-mg/m2 doses of this drug administered as an iv infusion via
`liposomes. Only 10% (2/19) of the patients responded to this
`therapy after two courses of treatment.100 In a Phase I study
`by Sells et al.,e9 three patients were treated with 15 to 20
`mg/mz of liposomal adriamycin every week for up to 4 weeks.
`Therapeutically, this treatment schedule compared well with
`60- to 75-mg/m2 doses of free adriamycin given every 3 to 4
`weeks; however, the liposomal adriamycin was associated
`with fewer side-effects. The maximum tolerable dose of this
`delivery system is not known to date.99
`Sculier et a1.120 have also conducted a Phase I study with
`liposomally delivered 6-aminochrysene. In this study involv-
`ing 13 patients, the drug dosage was escalated from 30 to 200
`mg/m2 in steps of 10 mg/m2. The number of infusions per
`patient varied between 1 and 10, with 2 weeks between two
`infusions. A maximum of 12 g of lipids were infused per
`patient. All doses of liposomal drug were well tolerated,
`without any limiting toxicity. However, only one patient, who
`had lung adenocarcinoma with brain metastasis, demon-
`strated an objective response.120 Ten patients had progression
`of cancer, although they generally survived >6 months. One
`patient could not be evaluated as he died 3 weeks after
`treatment and another patient had no change in her mesothe-
`lioma.120
`So far, z 200 patients have received different formulations of
`liposomes, either for the treatment of life-threatening diseases
`(e.g., cancer) or to improve tumor imaging.= Compared with the
`iv administration of drug as a solution, the administration of
`
`equitoxic dose of drug via liposomes has generally allowed
`higher and sustained levels to be maintained in the 1iver.w
`Systemic drug toxicities, particularly those related to cardiac
`and renal tissues, are therefore reduced. Hence, patients have
`been shown to tolerate much higher cumulative doses of lipo-
`some-delivered chemotherapeutic agents in Phase I and II
`studies than that usually observed with free drugs.103 Lipid
`doses up to 1 g per injection have shown little toxic effects.=
`Although the toxicity of liposomal drug to the RES has been an
`a priori concern, so far, none of the studies has shown any
`damage to this tissue. The major side-effects of liposome-
`delivered drug include fever (50-70%), rigors (50-60%), and
`nausea and vomiting (3050%).1@JJ13J14 At times, acute lumbar
`pain, bronchospasm, respiratory distress,79.m and leucocytope-
`nia” have also been observed.
`As a result of frequent positive therapeutic response with
`liposomal delivery systems, either in terms of sustained drug
`levels and/or reduced systemic toxicities, several pharmaceu-
`tical companies worldwide are currently involved in bringing
`these innovative products to the market. In the US., at least
`five companies are in the process of Phase I, 11, or 111 clinical
`trials with liposomal systems for the delivery of chemother-
`apeutic agents, antifungal agents useful in malignancies, or
`radioactive compounds for tumor imaging.121 Nonetheless, it
`is important to realize that several chemotherapeutic agents
`are intrinsically hepatotoxic,122-124 and frequently patients
`with hepatic metastases exhibit altered reticuloendothelial
`phagocytic activity.125-130 These observations therefore raise
`the possibility that the administration of drug via liposomes
`may in fact increase the delivery of drug towards the popu-
`lation of hepatic cells performing normal metabolic activities,
`or that the “target” hepatic cells may fail to capture the
`“effective” fraction of liposomal drug. A more frequently
`encountered limitation of liposomal delivery systems is their
`relatively low stability, both in vitro and in vivo.SO-62,66 The
`attempts to formulate fresh batches of drug-loaded liposomes
`to meet the needs of Phase I clinical trials have not been very
`promising.80 Until these aspects of liposomal drug delivery
`system are addressed, its routine application may be difficult.
`Starch Microspheres-Transcatheter vessel occlusion is a
`popular technique for the control of hemorrhage, tumor
`palliation, and redistribution of blood supply. This technique
`has lately been adapted for improving regional drug delivery
`and, due to their bioacceptability and biodegradability, starch
`microspheres have been increasingly used for this pro-
`cess.131-137 It has been shown that following the ia infusion of
`180 x lo6 starch microspheres (40 pm in diameter) over a
`period of 3 min to four patients with liver metastasis under-
`going hepatic arterial chemotherapy, the microspheres pro-
`voked transient diversion of hepatic arterial blood flow from
`higher to lower flow areas.132
`Conceptually, for improved regional drug delivery, the
`starch microspheres (15 to 80 pm in diameter) are infused into
`the artery(s) which supply the target-tissue. The micro-
`spheres block the blood supply temporarily because, depend-
`ing on the degree of their crosslinking, they degrade within 10
`to 30 min. Hence, the ia administration of a drug as a solution,
`following the establishment of embolism, provides a unique
`opportunity for controlling its distribution to the target area.
`It shares the benefits of arterial occlusion and ia chemother-
`apy. Apart from the direct effect of ischemia on neoplasm,
`arterial occlusion probably also incmases tissue permeability
`due to anoxia.138 Hence, the cancer cells are exposed to the
`delivered cytotoxic agent in highconcentrations, for a prolonged
`period of time, and with minimal “spill-over” to the systemic
`circulation. This technique has been successfully tried in nu-
`merous patients with hepatic metastases for the delivery of
`1,3-bis(2-chloroethyl)-l-nitrosourea (BCNU),133 5-fluoroura-
`cil,131 mitomycin C,135-137 adriamycin, and floxuridine.136J~7
`
`Journal of Pharmaceutical Sciences I 953
`Vol. 79, No. 11, November 7990
`
`IMMUNOGEN 2183, pg. 5
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`Aronsen et al.131 investigated starch microspheres (size: 92
`pm; half-life: 1 h) for improving the delivery of 5-fluorouracil
`in 12 patients with secondary hepatic neoplasms. In this
`study, 10 to 30 mL of starch microsphere suspension (1.5 g in
`100 mL of saline) was infused three times per day for 2 weeks.
`This was followed by the infusion of 5-fluorouracil solution (10
`mg/mL) through the same catheter. One patient demon-
`strated complete eradication of lesions and six patients had
`moderate regression of their lesions. Other patients failed to
`respond to this therapy.131 In another study,133 the infusion of
`10 mL of microsphere suspension (40 pn; 9 x lo6 micro-
`spheres/mL) into the hepatic artery of five patients tran-
`siently reduced the hepatic blood flow by almost 80 to 100%
`for almost 30 min. The infusion of 50 mg/m2 of BCNU over a
`1-min period along with the microspheres reduced the peak
`drug concentration and the systemic exposure of drug by 30
`to 90% (see Table IIl.133 Similar results were observed in 10
`patients following the hepatic arterial administration of 10
`10 5 starch microspheres.136 The systemic drug exposure was
`m /m2 of mitomycin C over a 1-min period along with 90 x
`generally 10 to 70% lower when it was infused in the presence
`of microspheres. Although the drug (or drug plus micro-
`spheres) was infused only over a period of 1 min, the systemic
`drug exposure increased with time in a linear fashion (see
`Figure 2).135 This is particularly interesting for treatments
`involving microspheres since it suggests that the microsphere
`distribution in the microcirculation of target tissue remained
`largely unaltered at least up to 60 min postinfusion. This
`study also revealed that at a constant drug dose, the reduction
`in the microsphere dose from 90 x los to 36 x lo6 resulted in
`comparably less reduction in systemic drug exposure, and
`therefore suggested the existence of dose-dependent shunting
`of microspheres.136
`Another study by F'feifle, Howell, and co-workers136 has
`found that following one to four courses of ia administration
`of 210 mg/m2 of starch microspheres (40 ? 5 pm) mixed with
`500 mg/mz of fl