Prodrug-based intracellular delivery of anticancer agents

Advanced Drug Delivery Reviews 63 (2011) 3–23

Contents lists available at ScienceDirect

Advanced Drug Delivery Reviews j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / a d d r

Prodrug-based intracellular delivery of anticancer agents L. Bildstein, C. Dubernet, P. Couvreur ⁎ University Paris-Sud, UMR CNRS 8612, IFR 141-ITFM, Faculté de Pharmacie, Châtenay-Malabry, 92296, France CNRS, Châtenay-Malabry, 92296, France

a r t i c l e

i n f o

Article history: Received 22 October 2010 Accepted 21 December 2010 Available online 13 January 2011 Keywords: Prodrug Anticancer Bioconjugate Immunoconjugate Immunoliposome Endocytosis Passive diffusion Spacer Stimuli-responsive Self-immolative

s u m m a r y There are numerous anticancer agents based on a prodrug approach. However, no attempt has been made to review the ample available literature with a specific focus on the altered cell uptake pathways enabled by the conjugation and on the intracellular drug-release mechanisms. This article focuses on the cellular interactions of a broad selection of parenterally administered anticancer prodrugs based on synthetic polymers, proteins or lipids. The report also aims to highlight the prodrug design issues, which are key points to obtain an efficient intracellular drug delivery. The chemical basis of these molecular concepts is put into perspective with the uptake and intracellular activation mechanisms, the in vitro and in vivo proofs of concepts and the clinical results. Several active targeting strategies and stimuli-responsive architectures are discussed throughout the article. © 2011 Elsevier B.V. All rights reserved.

Contents 1.

2.

Prodrug endocytosis . . . . . . . . . . . . . . . . . . . . . . . . 1.1. Fluid-phase endocytosis . . . . . . . . . . . . . . . . . . . 1.1.1. Passive targeting: the EPR effect . . . . . . . . . . . 1.1.2. Drug–polymer conjugates under clinical trials . . . . 1.1.3. Dendrimers as promising alternative architectures . . 1.2. Receptor-mediated endocytosis . . . . . . . . . . . . . . . 1.2.1. Small molecular ligands . . . . . . . . . . . . . . . 1.2.2. Peptides . . . . . . . . . . . . . . . . . . . . . . 1.2.3. Antibodies . . . . . . . . . . . . . . . . . . . . . 1.2.4. Hyaluronic acid . . . . . . . . . . . . . . . . . . . 1.2.5. Synthetic lipoproteins and low-density lipoproteins. . Passive diffusion through the plasma membrane . . . . . . . . . . . 2.1. Amphiphilic prodrugs administered alone. . . . . . . . . . . 2.2. Amphiphilic prodrugs encapsulated within targeted liposomes .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

. . . . . . . . . . . . . .

4 4 4 6 8 8 8 10 10 11 13 13 13 14

Abbreviations: 5-FdU, 5-fluoro-2′-deoxyuridine; Ag, Antigen; ApoE, Apolipolipoprotein E; AraC, 1-β-D-Arabinofuranosylcytosine (cytarabine); BE, Benzyl elimination; CALI, NAcgamma calicheamycin; CPP, Cell-penetrating peptide; CPT, Camptothecin; DAUN, Daunorubicin; dFdC, 2′,2′-difluorodeoxycytidine (gemcitabine); DOX, Doxorubicin; DMPC, Dimyristoylphosphatidylcholine; DSC, Differential scanning calorimetry; EPR, Enhanced permeability and retention; FA, Folic acid; FA-R, Folic acid receptor; FITC, Fluorescein isothiocyanate; HA, Hyaluronic acid; HPMA, N-(2-hydroxyhropyl)methacrylamide; LDL, Low-density lipoprotein; LDL-R, LDL Receptor; LHRH, Luteinizing hormone-releasing hormone; MAb, Monoclonal antibody; MDR, Multidrug resistance; MMAE, Monomethyl auristatine E; MMC, Mitomycin C; MTD, Maximum tolerated dose; MTX, Methotrexate; Mw, Molecular weight; NOAC, N4-octadecyl-AraC; NSCLC, Non-small-cell lung cancer; PAA, poly(aspartic acid); PABC, Para-aminobenzyl carbamate; PAMAM, Poly(amidoamine); PCT, Paclitaxel; PEG, Poly(ethylene glycol); P-gp, P-glycoprotein; PHEG, Poly-[N5-(2-hydroxyethyl)-L-glutamine]; PLA2, Phospholipase A2; RME, Receptor-mediated endocytosis; SMVT, Sodium-dependent multi-vitamin transporter; SQ, Squalene; SST, Somatostatin; TML, Trimethyl lock lactonisation. ⁎ Corresponding author at: University Paris-Sud, UMR CNRS 8612, IFR 141-ITFM, Faculté de Pharmacie, Châtenay-Malabry, 92296, France. E-mail address: [email protected] (P. Couvreur). 0169-409X/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.addr.2010.12.005

4

L. Bildstein et al. / Advanced Drug Delivery Reviews 63 (2011) 3–23

3.

Controlled intracellular release of the promoiety 3.1. Stimuli-responsive groups . . . . . . . 3.1.1. pH-labile hydrazone linkers . . 3.1.2. Hindered disulphide bonds . . 3.1.3. Enzyme-sensitive spacers . . . 3.2. Double prodrug approach . . . . . . . 3.2.1. Benzyl elimination. . . . . . . 3.2.2. Trimethyl lock lactonisation . . 4. Summary and outlook . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

Cancer is a multifaceted disease that represents one of the leading causes of mortality in developed countries. Due to the societal and economical implications of this pathology, tremendous efforts have been made over the past decades to improve the available therapeutic options. Although a large number of potent chemotherapeutic anticancer agents have been identified and successfully used in clinical practice, considerable research activity is devoted to discover more potent treatments, while minimising their toxic side effects. Indeed, most anticancer agents display a narrow therapeutic window due to their lack of selectivity against cancer cells. Besides, the ability of the anticancer compounds to actually reach their target is often impaired by a number of physiological barriers (i.e., tumour interstitial pressure, diffusion through the tumour endothelium and/ or extracellular matrix and so on) as well as by metabolisation/ degradation phenomena such as conversion into inactive metabolites. Due to the constant progress accomplished in the fields of chemistry, soft-matter science and nanotechnology and in the understanding of the biological mechanisms of cancer diseases, several drug delivery approaches have been developed to enhance the efficacy of existing anticancer agents. One of them, the ‘pro-drug’ strategy, was devised 50 years ago to help drugs to cross physiological barriers. The concept consists in grafting a molecule (termed ‘promoiety’) onto an active drug molecule that will help it in reaching the pharmacological target, while ensuring that the promoiety can afterwards be removed to regenerate the biologically active compound [1]. A well-thought-out prodrug strategy may overcome various obstacles such as poor drug solubility, the systemic conversion into inactive metabolites, a lack of site specificity or an inefficient cell uptake. This last point is especially critical in the case of anticancer therapy, as an alteration of drug transport across the cell membrane is a common mechanism of resistance to chemotherapy. Many prodrugs described in the literature have a different cell uptake pathway than their parent drug and may ensure an efficient intracellular drug release. However, no attempt has been made to review this ample literature with a specific focus on these altered cell uptake pathways and on the intracellular drug-release mechanisms. Thus, the aim of this report is to discuss the relevant cellular mechanisms and to identify the critical design features that lead to an efficient intracellular delivery of the active molecule when administered as a prodrug. We aimed to cover the broadest possible selection of prodrugs based on synthetic polymers, lipids or proteins, and coupled to small-molecular-weight cytotoxic anticancer agents. The scope has been restricted to parenterally administered systems because prodrugs designed for non-parenteral routes usually aim to address the absorption of the compound rather than its cellular uptake in the target tissue. We have also excluded from this report several clinically important anticancer prodrugs (e.g., cyclophosphamide) [2] and two-step prodrug strategies such as enzyme- and andibody-directed enzyme prodrug therapy, which have been reviewed elsewhere [3,4] and did not fit our focus on conjugated small-molecular-weight anticancer agents. Given the magnitude of the available literature, only reports of drug-delivery strategies that provided mechanistic insight or illustrated the critical design issues were selected, at the expense of exhaustiveness.

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

15 15 15 16 16 17 17 17 19 20 20

This article is organised as follows. Each of the first two sections will describe one of the available uptake pathways for anticancer prodrugs: (1) the endocytosis of macromolecular and/or targeted constructs and (2) the passive diffusion through the plasma membrane enabled by a hydrophobic promoiety. Successful technical options to trigger the intracellular release of the active drug will be discussed in a third section. The desirable design features identified throughout the report will be summarised and put into perspective in the last section. 1. Prodrug endocytosis Endocytosis refers to the deformation or invagination of a cell's plasma membrane, which results in the internalisation of solutes or material bound to the cell membrane or present in its vicinity. This generic concept encompasses several distinct mechanisms, such as phagocytosis, clathrin-mediated endocytosis, caveolar endocytosis or macropinocytosis [5]. The complex molecular basis and physiological relevance of the respective pathways have been reviewed in depth by several authors [6–8], and will not be detailed here. Endocytotic pathways have been used for over 30 years to deliver various payloads inside cells by means of nanometre-scale carriers [5]. In this first section of the article, we will report prodrug architectures that result in an uptake through an endocytotic pathway. To facilitate the presentation of the prodrug architectures reported throughout the article, the chemical formulae of all anticancer agents will be replaced by their cartoon representation, to better focus on the chemical moieties relevant to the conjugation processes. The complete formulae of the drugs and their graphical placeholders are reported in Fig. 1. 1.1. Fluid-phase endocytosis Over the past decades, different types of hydrophilic polymers have been covalently linked to various anticancer agents to improve their solubility or to alter their transport properties [9]. However, the important size and hydrophilicity of those conjugates prohibit their spontaneous diffusion through cellular membranes. Nonetheless, the fluid-phase endocytosis of the drug–polymer conjugates present in the vicinity of the cells through any of the endocytotic mechanisms mentioned above constitutes a valid cell uptake mechanism. ‘Fluidphase endocytosis’ is a generic term that designates the internalisation of materials present near the cell surface (or adsorbed onto it) over the course of physiological or baseline endocytotic processes (Scheme 1). Noteworthy, to maximise the impact of this aspecific endocytosis pathway, it is essential to maximise the prodrug concentration within the tumour tissue, in the vicinity of the target cells. 1.1.1. Passive targeting: the EPR effect Fortunately, colloidal systems such as polymer–drug conjugates display the interesting property of allowing the passive targeting of solid tumours. Indeed, the architecture of the abundant neovasculature required for tumour growth is known to be incomplete. This results in a superior permeability with respect to healthy vessels,

L. Bildstein et al. / Advanced Drug Delivery Reviews 63 (2011) 3–23

Fig. 1. Names, abbreviations, chemical formulae, and graphical placeholders of the anti-cancer agents described throughout the report. The circled parts of the molecules correspond to the coupling sites with the promoieties. 5

6

L. Bildstein et al. / Advanced Drug Delivery Reviews 63 (2011) 3–23

Scheme 1. Endocytosis of untargeted and targeted anticancer prodrugs, and intracellular stimuli leading to the drug release by stimuli-responsive spacers.

which enhances the leakage of macromolecules into the interstitial fluid of tumours [10]. Due to their chaotic growth, tumours also lack a mature lymphatic system that would help in removing the macromolecules from the interstitium. The combination of these physiological characteristics of solid tumours was termed the enhanced permeability and retention (EPR) effect [11]. It constitutes one of the cornerstones of carrier-based anticancer drug delivery strategies. One of the first crucial steps towards the design of efficient polymer–drug conjugates was to determine the molecular requirements of the EPR effect. The first investigations of the relationship between the molecular weights (Mws) of polymers and their biodistribution were undertaken in the absence of conjugated drugs. A preferential accumulation of hydrophilic polymers in solid tumours was evidenced for Mws as small as 10 kDa [12], although the approximately 45 kDa glomerular filtration cut-off strongly opposed the use of Mws smaller than 20 kDa due to their rapid urinary excretion [13]. The EPR effect was reported to reach a maximum in the 100–200 kDa Mw range for hydrophilic polymers [14]. These findings were confirmed in the case of polmer–drug conjugates. Noteworthy, crosslinked N-(2-hydroxypropyl)methacrylamide (HPMA)–doxorubicin (DOX) conjugates of Mw up to approximately 1000 kDa displayed a considerable tumour accumulation and anticancer efficacy [15]. Although increasing the size of the colloidal carrier favours tumour targeting via the EPR effect, it also has adverse effects such as impairing the diffusion of the conjugate in the extracellular matrix or decreasing the efficacy of the endocytosis [16]. The need to balance these different effects to maximise the efficacy of the conjugate explains why the most successful polymer prodrugs of anticancer agents currently under clinical investigation display Mws between 25 and 50 kDa [17]. 1.1.2. Drug–polymer conjugates under clinical trials Numerous anticancer drug–polymer conjugates have been developed over the past 20 years, several of which are currently under clinical investigation [17]. The common denominator of all their architectures was an uptake by fluid-phase endocytosis. Table 1

summarises the chemical structure of the clinically investigated prodrugs described throughout this article, along with their most advanced clinical status to date. The HPMA–DOX conjugate named PK1 (Fig. 2(A)) [18] was the first to enter clinical trials in 1994, as it initially displayed an extremely promising preclinical activity while decreasing the DOXrelated toxicity [19]. As indicated in Fig. 2(A), each macromolecule bears several DOX molecules, leading to an optimised drug loading of 8% (w/w). Noteworthy, this prodrug also dramatically improved the efficacy of DOX against resistant cancer cell lines [18], thanks to its fluid-phase endocytotic uptake followed by lysosomotropic metabolisation that minimised the impact of the multidrug resistance (MDR) membrane efflux pumps, as reported for other polymer–drug conjugates [20,21]. Phase I clinical trials indicated that the maximum tolerated dose (MTD) of the conjugate was 5 times higher (i.e., less toxic) than that of free DOX, with a considerable reduction of the cardiotoxicity of the anthracycline [22]. Ongoing phase II studies revealed a partial response in patients bearing breast cancer or nonsmall-cell lung cancer (NSCLC), which seemed to confirm the activity of the conjugate [23]. HPMA was also coupled to other anticancer

Table 1 Chemical structure, name and clinical status of the most advanced prodrugs described throughout the review. Prodrug structure (and name)

Status

Ref.

HPMA-DOX (PK1) HPMA-CPT HPMA-PCT HPMA-platinum derivatives poly(L-glutamic acid)-PCT (CT-2103) PEG-CPT Anti-CD33 MAb-Calicheamycin (Mylotarg) Anti-LewisY MAb-DOX Fatty acid-AraC (CP-4055) Fatty acid-dFdC (CP-4126) MAb-maytansine derivative

Phase II Phase I Phase I Phase I / II Phase III Phase II Market Phase I Phase I Phase I Phase I

[23] [24] [21] [25] [17] [23,29] [59] [67] [104] [105] [132,135]

L. Bildstein et al. / Advanced Drug Delivery Reviews 63 (2011) 3–23

7

Fig. 2. Drug-polymer conjugates. Gray boxes represent aminoacids. The bracketed numbers recall the corresponding literature references. A: Structure of PK1, an HPMA-DOX conjugate encompassing a tetrapeptide spacer. B: CT-2103: pol(L-glutamic acid) conjugated through ester bonds to either of the two secondary hydroxyl moieties of paclitaxel. C: Structure of a hyaluronic acid-paclitaxel conjugate, additionally labeled with FITC through the same linker as paclitaxel (§ 1.2.4). D: General structure of a PHLG-spacer-MMC conjugate encompassing the GFAL spacer (§ 3.1.3).

agents such as camptothecin (CPT) [24] and paclitaxel (PCT) [21] but the conjugation failed to improve the toxicological profile of the drugs, which led to severe dose-limiting toxicities and to a poor antitumour activity. On the contrary, two platinum derivatives conjugated to HPMA, currently under phase I/II clinical evaluation, reduced the platinum-related toxicity and displayed promising anticancer efficacy [25]. Another interesting example is that of the PCT derivative CT-2103 (Fig. 2(B)), which is undergoing a phase III evaluation in combination

with standard chemotherapy against NSCLC and breast cancer [17]. Its polymeric backbone is constituted of the fully biodegradable poly(Lglutamic acid), which resulted in a mainly intracellular metabolisation after fluid-phase endocytosis, although a moderate extracellular release of PCT was reported [26]. One of the main advantages of this architecture is its extremely high PCT drug loading (37% w/w). Despite the excellent biocompatibility of poly(ethylene glycol) (PEG) and the numerous applications of this polymer in the field of nanomedicine, only one example of a PEG-anticancer agent prodrug

8

L. Bildstein et al. / Advanced Drug Delivery Reviews 63 (2011) 3–23

has reached clinical trials. This may be explained by the fact that each linear PEG chain can bear at most one active molecule at each of its extremities, which results in significantly lower drug loadings than with the multifunctional polymers reported above [23]. For instance, Prothecan, a PEG prodrug of the DNA damaging agent CPT, was conceived by attaching CPT molecules on both extremities of a glycine-bifunctionnalised 40-kDa PEG, yielding a drug loading of only approximately1.7% (w/w) [27]. However, the conjugation to PEG considerably enhanced CPT solubility and bioavailability at the tumour site, thanks to the EPR effect [28]. Phase I clinical studies underlined partial response in some cases and indicated that the conjugation to PEG notably improved the pharmacokinetics of the compound [29], and this conjugate is currently ongoing phase II studies [23].

tolerated by the animals than free DOX, and was considerably more efficient at its MTD against DOX-resistant colon tumour xenografts than free DOX or Doxil®, its commercial liposomal formulation. Aside from DOX, other anticancer agents have also been conjugated to dendrimers. For example, CPT was grafted onto the aspartic acid-functionalised, PEG-chain-bearing periphery of a symmetrical poly(lysine) dendrimer [37]. The approximately 30-h plasmatic half-life of the conjugate was considerably larger than that of free CPT, which almost disappeared from the blood in only 30 min. Besides, biodistribution studies revealed that the tumour accumulation of the prodrug was 14.5 times higher than that of free CPT. This resulted in a considerable gain of anticancer activity against murine and human colon carcinoma xenografts, in vivo. 1.2. Receptor-mediated endocytosis

1.1.3. Dendrimers as promising alternative architectures One of the most significant design issues of the above polymer– drug conjugates is the control of their molecular weight and polydispersity [30]. Dendritic polymers offer a valuable alternative to random-coil linear polymers, as the unimolecular, hyperbranched structures are generated by a step-by-step process, which ideally allows an extremely well-controlled morphology. Besides, dendrimers can be further functionalised and chemically modified to generate core-shell structures [17]. One of the most prevalent drawbacks of cationic dendrimers, such as the widely used poly (amidoamine) (PAMAM) polymer, is that their interaction with cellular membranes can result in cell lysis [31]. However, this can often be offset by masking the dendrimer end groups with anionic or neutral moieties. In this perspective, the functionalisation of dendrimers with hydrophilic and biocompatible polymers such as PEG often serves the double purpose of increasing the circulation time of the construct by delaying its hepatic capture and of lowering its toxicity [31,32]. All in all, dendrimer-based carriers are envisioned as promising drug-delivery systems and are currently receiving considerable attention [33]. Although several teams focus on dendrimerbased systems that possess a hydrophobic core in which active drugs can be easily encapsulated [31], covalent systems falling within the scope of the present article have been reported. These systems present the advantage of allowing a better control of the intracellular drug release than carriers relying solely on hydrophobic encapsulation [34]. Kono et al. developed a dendrimer–DOX conjugate from symmetric PAMAM dendrimers bearing glutamic acid residues on their chain ends [35]. PEG chains were grafted onto the main amino acid chain, and DOX was covalently linked to the side chains of glutamic acid residues via amide or hydrazone bounds (Fig. 3(A)), resulting in a dendritic carrier in which the active drug was shielded by biocompatible PEG chains. The in vitro evaluation of these dendritic conjugates indicated that their activity was similar on sensitive and DOX-resistant cell lines, which suggested that the prodrug circumvented the action of the MDR-associated efflux transporters, as reported above in the case of linear polymer–drug conjugates. Besides, the lysosomal pH-labile hydrazone bond led to a considerably higher cytotoxicity than the amide one. These observations were interpreted as an uptake mechanism involving the endocytosis of the whole prodrug, followed by a progressive lysosomal DOX release. Other dendrimer-based PEGylated prodrugs have displayed a promising in vivo efficacy. A biodegradable polyester bow-tie-shaped dendrimer was grafted on one end with eight 5-kDa PEG chains and, on the other end, with DOX via hydrazone bonds (Fig. 3(B)) [36]. Despite the asymmetric nature of the globular core, the PEG chains were presumed to wrap around the whole structure and shield the drug-bearing part of the dendrimer, efficiently impairing interconjugate aggregation while reducing the hepatic capture of the prodrug. The approximately 45 kDa Mw enabled an excellent EPRmediated tumour targeting. This DOX-based dendrimer was better

As described above, the use of high-Mw polymers to provide passive tumour accumulation and the PEGylation of prodrugs to delay their hepatic capture represent useful tools to improve the tissue targeting of prodrugs. Active targeting moieties can also be incorporated in the prodrug architectures to specifically enhance their internalisation by the target cells, while minimising the action on healthy tissue. Most of the ligands used to achieve this aim do so by binding onto specific structures overexpressed by neoplasic cells. Subsequent receptor-mediated endocytosis (RME) then leads to the uptake of the targeting moiety, along with its covalently attached drug payload. Noteworthy, this process is considerably faster and more efficient than fluid-phase endocytosis [38]. In the following subsections, we will illustrate the various types of targeting moieties that have been used to promote the RME of prodrugs and, thereby, increase their anticancer activity. A summary of these architectures is reported in Scheme 1. 1.2.1. Small molecular ligands Folic acid (FA) is a small-Mw compound that binds with a high affinity with membrane receptors that were proven to be overexpressed in a variety of tumours [39]. As a result, FA has been extensively used as a targeting moiety in drug-delivery applications. In this subsection, we will report several FA-containing prodrugs, some of them similar to the architectures reported in the previous section, but additionally encompassing FA as a targeting moiety. Self-assembled polymeric micelles were prepared from diblock PEG–poly(aspartic acid) (PAA) copolymers bearing FA on the extremity of the hydrophilic PEG chain, and multiple DOX molecules conjugated to the hydrophobic PAA part via pH-sensitive hydrazone linkers [40]. Surface-plasmon resonance measurements clearly proved that FA-bearing micelles could interact with the FA receptor (FA-R). Flow cytometry and in vitro cytotoxicity experiments on the FA-R-expressing KB cancer cell line confirmed that FA considerably enhanced the cell uptake and cytotoxicity of the micelles, respectively, over untargeted ones containing an identical DOX payload, presumably due to an enhanced endocytosis of the construct. A system based on a similar block copolymer but additionally encompassing a commercial, biodegradable, globular, aliphatic, polyester dendritic core (Bolorn H40) was recently reported (Fig. 3 (C) [41],). Such an ~150-kDa architecture would be extremely stable in vivo due to its entirely covalent structure. Although some secondary inter-micelles’ aggregation was observed, the cellular uptake of the ~25-nm monomolecular objects was greater than that of FA-free constructs. The in vitro cytotoxicity of the constructs was negatively influenced by the addition of free FA in the cell supernatant, presumably due to a competition for the binding to the membrane FA-R, which strongly suggested that the uptake pathway was RME, as expected. Smaller (26 kDa) PAMAM dendrimers were coupled to FA, and radiolabelled or conjugated either to a fluorophore or to methotrexate

L. Bildstein et al. / Advanced Drug Delivery Reviews 63 (2011) 3–23 Fig. 3. Dendrimer-based prodrugs. The bracketed numbers recall the corresponding literature references. A: DOX-containing PAMAM-PEG dendrimers, doxorubicin was grafted on Glu residues at the periphery of the PAMAM core, either through (a) an amide bond or through (b) a pH-labile hydrazone bond. B: [G-3]-(PEG5k)8-[G-4]-OH16 bow-tie dendrimer conjugated to doxorubicin through hydrazone bonds. C: H40-p(LA-DOX)-b-PEG-OH/FA conjugate (§ 1.2.1). 9

10

L. Bildstein et al. / Advanced Drug Delivery Reviews 63 (2011) 3–23

(MTX), respectively, to evaluate the carrier biodistribution, to explicit the tumour uptake and subcellular localisation or to assess the in vivo anticancer efficacy of the construct against FA-R-overexpressing tumours [42]. As expected, an increased tumour accumulation of the FA-conjugated carrier was observed, and was found to be negatively influenced by the administration of free FA to the animals. Confocal microscopy indicated a cytoplasmic accumulation of the carrier, consistent with RME. The dendrimers induced no acute or chronic toxicity in mice, and the dendrimer-conjugated form of MTX was considerably less toxic than free MTX. Finally, intravenously administered MTX–PAMAM–FA led to an excellent antitumour efficacy in mice bearing human cervix carcinoma xenografts overexpressing the FA-R. Another FA-based prodrug consisted of DOX–PEG–FA conjugates nanoprecipitated in the presence of free DOX [43]. The resulting nanoaggregates comprised a molecular DOX core that also contained the DOX-bearing extremity of the polymer, stabilised and solubilised by a FA-bearing PEG corona. Although these objects were readily internalised through FA-dependent RME, their good in vitro cytotoxicity and in vivo efficacy on FA-R-overexpressing cell lines probably rather stemmed from the encapsulated DOX than from an intracellular release of the DOX covalently linked to the polymer, as the amount of encapsulated DOX was much larger than the amount of conjugated one. Biotin is another B-group vitamin, which has numerous applications in drug delivery and biotechnology due to its extreme affinity for avidin proteins, thanks to which near-covalent interactions can be obtained under mild conditions with a plethora of compounds. The sodium-dependent multivitamin transporter (SMVT), which mediates the cell uptake of biotin, was found overexpressed in a number of aggressive cancer cell lines. Its overexpression was found to be superior to that of the FA-R in several lung, renal, colon and breast cancer cell lines [44]. Besides, investigations in the field of oral delivery have clearly proven that the presence of biotin on various constructs could promote their uptake by RME. These considerations made the biotinylation of anticancer prodrugs a worthwhile strategy to enhance tumour targeting, while providing a means to target some SMVT-overexpressing but non-FA-R-overexpressing tumours. For example, the functionalisation of a labelled HPMA polymer with biotin improved its tumour accumulation in several SMVToverexpressing tumours in mice, after intravenous injection, in the absence of conjugated drug [45]. As a consequence, the anticancer efficacy of the corresponding HPMA–DOX conjugates against colon carcinoma xenografts was greatly enhanced by biotin functionalisation, in vivo. This targeting strategy was also applied to PEG prodrugs of CPT [46]. In this case, biotin targeting improved the cytotoxicity and apoptosis induction in SMVT-expressing sensitive and MDR ovarian cell lines, in vitro, presumably thanks to an enhanced endocytotic uptake. An increased cellular uptake by the HeLa human cervix carcinoma cell line was also recently reported with biotinylated PAMAM dendrimers, in comparison with the corresponding nontargeted prodrugs [47].

penetration, and they can tolerate harsher chemical processes than more fragile targeting moieties such as monoclonal antibodies (MAbs) [53]. As the profile of expressed receptor subtypes considerably varies from one tumour to another, a critical issue in the design of a successful peptide-based prodrug is to ensure that the targeting moiety has a high affinity towards the broadest possible array of receptor subtypes [54]. As in all targeted strategies, the optimisation of peptide–drug conjugates revolves around the preservation of the binding affinity, as it is a crucial requirement to the efficient internalisation of the prodrug, which, in turn, is essential for the downstream drug activity. Most authors adopted a trial-and-error empirical approach by combining different peptide analogues, linker chemistries and anticancer drugs. For example, in the case of LHRH, a slightly modified version of the peptide ([D-Lys6]LHRH) was generated to allow the conjugation with anticancer agents. The peptide analogue retained the same affinity as LHRH for its receptor. The grafting of CPT, MTX or DOX (Fig. 4(A)) on this modified peptide did not decrease the binding of the construct to the LHRH receptor, comparatively to the original peptide [55]. In nude mice, the DOX-conjugated LHRH analogue was considerably more active than free DOX against xenografted LHRHreceptor-expressing cell lines originating from breast, ovarian and prostate human cancers [49]. The preservation of the binding affinity after peptide conjugation reported above for LHRH analogues is, however, not a general rule. Indeed, two octapeptide analogues of SST, initially developed for their intrinsic effect on native SST targets, were coupled via a glutarate spacer to DOX or to its more potent derivative 2-pyrrolino-DOX [56]. In this case, binding studies demonstrated that both drugs diminished peptide affinity in a similar fashion, and the various binding affinities of the four conjugates were directly correlated to their cytotoxicities on several cancer cell lines, in vitro. The optimised 2-pyrrolino–DOX– SST analogue conjugate was proven active against a variety of cancer cell lines xenografted into nude mice, including tumours of breast, ovarian, renal, prostate, lung, brain, pancreatic, colorectal and gastric origins [49]. In another example, three functionalisable peptides were derived from a synthetic BN analogue known to bind to the three main subtypes of the BN receptor family, and subsequently conjugated to CPT by means of two different self-eliminating linkers [54]. The best peptide-linker couple (Fig. 4(B)) was identified based on its binding affinity with the three main BN receptor subtypes. Surprisingly, the affinity of this construct for the three subtypes of the BN receptor was even superior to that of the unconjugated peptide analogue, although the authors failed to explain this phenomenon. More in-depth investigations confirmed that the prodrug was internalised by endocytosis, which was more intense than that of the unconjugated peptide analogue as a result of the superior binding affinity of the conjugate. The anticancer efficacy of the prodrug was confirmed on a panel of human cancer cell lines, in vitro, and on human NSCLC and pancreatic xenografts, in vivo [48].

1.2.2. Peptides Peptide ligands with a high affinity towards receptors naturally present at the surface of cells can also be used as targeting moieties. Noteworthy, the receptors of several hypothalamic peptides such as somatostatin (SST), luteinising hormone-releasing hormone (LHRH) and members of the bombesin (BN) family were found to be overexpressed in a variety of neoplasic tissue [48,49]. It was proven that most of the subtypes of SST and BN receptors could internalise by endocytosis, various cargoes covalently coupled to their specific peptide [50,51]. As a result, many prodrugs of anticancer agents coupled to these peptides or to their analogues have been designed [52]. Although peptides suffer from a rapid clearance due to their plasmatic proteolysis, their small size facilitates a good tumour

1.2.3. Antibodies MAbs are large glycoproteins able to recognise a variety of targets and bind them with a high affinity. Many MAbs have been investigated as potential cancer treatments, as standalone drugs. As most antibodies lack a direct cytotoxic effect, MAb–anticancer drug conjugates were among the first targeted-drug-delivery approaches to be investigated [57]. After binding to their cancer-cell-specific membrane antigen (Ag), most antibodies are internalised by endocytosis [58], which makes antibody-targeted prodrugs a valid strategy to improve the selectivity and cellular uptake of anticancer agents. Indeed, numerous MAb-conjugated cytotoxic agents have been reported, the most notorious being the clinically approved Mylotarg® (Fig. 5(A), Table 1) [59].

L. Bildstein et al. / Advanced Drug Delivery Reviews 63 (2011) 3–23

11

Fig. 4. Peptide-based prodrugs. The bracketed numbers recall the corresponding literature references. A: Example of a LHRH analog-DOX architecture. B: Structure of an optimized bombesin analog- camptothecin conjugate.

This conjugate of the potent anticancer agent NAc-gamma calicheamycin perfectly illustrates the common strategy that consists in taking advantage of the specificity provided by MAb targeting to use highly potent anticancer agents, too toxic to be administered in free form, rather than classic chemotherapeutic drugs (such as DOX, MTX or PCT) [60]. This approach results from the fact that cancerassociated surface epitopes are present in a relatively limited number (typically 105 per cell) [61]. This must be compensated either by a large number of active drug molecules per antibody, which is technically difficult to achieve, or by the use of highly potent drugs such as calicheamyicin, geldanamycin, auristatin and their derivatives [58]. On the other hand, the choice of the cellular target is governed by several factors. The overexpression of the MAb's epitope on certain types of neoplasic cells with respect to healthy tissue is an obvious requirement to ensure the selectivity of the prodrug. The structure, nature and physiological properties of potential epitopes also have a profound impact on their usefulness in a drug-delivery perspective [62]. The chosen membrane target must be rapidly internalised and recycled to allow for a sufficient uptake of anticancer agent to induce cytotoxicity [58]. These physiological considerations may actually affect the activity of MAb–drug conjugates considerably more than the expression level of the membrane target itself [63]. The uptake efficiency is especially critical in the case of solid malignancies, as large proteins such as MAbs diffuse slowly in the tumour interstitium. Therapeutically relevant epitopes that fulfil the above-mentioned requirements and have therefore been used in prodrug strategies include CD33 (acute myeloid leukemia) [64], CD30 (Hodgkin lymphoma), [63] CD22 (B cell lymphoma), [65] CD47 (ovarian carcinoma) [66] and LewisY antigen (tumours of epithelial origin) [67]. As immunogenicity concerns represent one of the main limitations of MAb-based prodrugs, humanised antibodies are preferentially used in the context of MAb–drug conjugates, due to their lower immunogenicity [57]. The affinity of the antibodies must also be finely balanced to ensure sufficient binding levels to their epitope, while minimising the ‘binding site barrier’ effect in the case of solid tumour-targeting MAbs [68]. This theory states that MAbs of excessive affinity bind massively to the first cellular targets encoun-

tered in the tumours and sterically hinder the diffusion of more immunoconjugates, thereby lowering tumour access and impairing the overall anticancer activity [69]. The overall properties of the prodrug are considerably affected by the nature of the active drug and by drug loading [70]. Noteworthy, the preservation of the binding affinity must be ensured after grafting the active anticancer drug onto the MAb. An interesting example to illustrate this design issue was reported in the case of an immunoconjugate of DOX and the anti-LewisY MAb, named BR96. It was proven that the reduction of interchain or even intrachain disulphide bonds of the MAb, and their substitution with up to eight DOX molecules per MAb (Fig. 5(B)) did not alter its binding efficacy [71]. In a subsequent report, the authors further increased the drug loading to 12–15 DOX molecules per MAb by the use of lysosomally cleavable branched peptide linkers grafted on the primary amine moiety of the DOX molecule, again without the loss of binding affinity [72]. At such drug loadings, however, the conjugation of short PEG chains through a hydrazone bond on the carbonyl residue of DOX (Fig. 5(C)) was required to avoid hydrophobicinteraction-driven prodrug dimerisation. Polymers such as dextran or HPMA, bearing multiple drug molecules, were also conjugated to MAbs in an attempt to further increase the drug loading of MAb-based conjugates [58]. In these systems, a key issue was to avoid the masking of the antibody's binding site by the drug-bearing polymer, which could be achieved by carefully selecting the polymer's grafting locus. For example, grafting HPMA– drug conjugates on an oxidised carbohydrate residue of the MAb's hinge region (far away from the MAb's binding site) or on the free sulphhydryl group of the fragment antigen binding (Fab) better preserved the MAb's binding affinity, leading also to more reproducible architectures than grafting the polymer on a random lysine residue of the MAb [73]. 1.2.4. Hyaluronic acid Hyaluronic acid (HA) is a biocompatible natural polysaccharide that is recognised by a specific membrane receptor, CD44 [74]. Interestingly, it was shown that the expression of CD44 in various neoplasic cell lines was correlated with their invasive properties and disseminative potential, due to the role of the CD44 in metastatic processes [75]. Consequently, the design of HA-based conjugates simultaneously allows the passive targeting of solid tumours via the

12

L. Bildstein et al. / Advanced Drug Delivery Reviews 63 (2011) 3–23

Fig. 5. Immunoconjugates. The bracketed numbers recall the corresponding literature references.A: Chemical structure of Mylotarg®, a clinically approved immunoconjugate of the potent anticancer agent NAc-gamma calicheamycin, conjugated to the anti-CD33 MAb P67.6 through a spacer containing a hydrazone function and a hindered disulfide bond. B: Linker chemistry of a BR96 MAb-DOX immunoconjugate. C: Chemical structure of DOX attached to PEG by a pH-labile bond, and conjugated to the BR96 MAb via a lysosomallycleavable branched peptide bond. D: Alternative spacer architecture for NAc-gamma calicheamicin immunoconjugates: amide bond with random lysine residues (§ 3.2.2).

EPR effect and the active targeting of CD44-bearing cancer cells without requiring additional targeting ligands. HA-based prodrug strategies can also alter the cell uptake mechanism of anticancer agents, as the binding of HA to the CD44 receptor results in the RME of the polymer, along with its drug payload [76]. As a result, HA–drug conjugates have been designed by several research groups. HA conjugated to the DNA crosslinker mitomycin C (MMC) was found considerably more active than free MMC in CD44expressing lung carcinoma, in vivo, due to an improved biodistribution and a more efficient endocytotic cell uptake, as proven by in vitro experiments on fluorescein isothiocyanate (FITC)-labelled HA [77]. Sodium butyrate, a fatty acid known to inhibit cancer cell proliferation [78], was conjugated to HA in an attempt to improve the poor pharmacokinetics of this anticancer agent. The conjugate exhibited a

better in vitro anti-proliferative effect on a CD44(+) breast cancer cell line than free sodium butyrate [79]. Interestingly, the authors proved that an excessive substitution of HA residues with their active drug reduced the cell uptake and subsequent cytotoxicity of their conjugate, due to the steric hindrance of the HA–CD44 interaction that is essential to the RME. HA–PCT conjugates were also developed. The binding and internalisation of FITC-labelled HA–PCT conjugates (Fig. 2(C)) were shown to be mediated by the CD44, as evidenced by the decrease in uptake and in vitro cytotoxicity in the presence of free HA or of an anti-CD44 MAb [80]. The CD44-mediated uptake and the potential of this drug delivery strategy were confirmed on other CD44 (+) cell lines, in vivo [81]. Alternative linker chemistries of HA–PCT prodrugs were reported, which also led to promising in vivo results [82,83].

L. Bildstein et al. / Advanced Drug Delivery Reviews 63 (2011) 3–23

1.2.5. Synthetic lipoproteins and low-density lipoproteins The various classes of lipoproteins (LPs) play a central role in the physiological transport of lipids. The low-density lipoprotein (LDL) receptor (LDL-R) is overexpressed in several cancer cell lines, as a consequence of the rapid growth of neoplasic cells and their corresponding needs for structural lipids [84]. The LDL-R recognises the Apolipoprotein B100 present at the surface of the LDL, as well as Apolipoprotein E (ApoE) [85,86]. This receptor can mediate the internalisation of apolipoproteins through RME [87]. Taken together, these considerations indicate that the LDL uptake pathway can be exploited to promote the endocytosis of anticancer prodrugs. In this subsection, we will describe two strategies that were followed to promote a LDL-R-dependent uptake pathway. A first approach consists in conceiving Apolipoprotein-functionalised carriers encapsulating hydrophobic prodrugs of an anticancer agent. A strongly hydrophobic prodrug of daunorubicin (DAUN) was prepared by covalent coupling of this anticancer compound to a cholesteryl oleate moiety via a peptide linker. Then, the prodrug was encapsulated in liposomes bearing recombinant ApoE on their surface [88]. In vitro binding and competition studies on B16 melanoma cells confirmed that the liposome was taken up by endocytosis, along with its prodrug payload. Biodistribution studies indicated that DAUN could be addressed to tumour cells, thanks to this strategy, although the drug accumulation was still proportionally more important in the LDL-R-expressing organs such as the liver, adrenals and spleen, than in the solid tumour [89]. As hydrophobic drugs can partition into LPs, which greatly influence their pharmacological and pharmacokinetic properties [90], the second approach consists in the administration of an amphiphilic prodrug, encapsulated in a way that maximises its partition into LPs in the bloodstream. In this strategy (i.e., contrarily to the former approach), it is essential that the prodrug only exhibits a moderate affinity for the liposomal carrier, to facilitate its transfer to LPs. An example of this indirect targeting was reported by the group of Schwendener, who designed a potent amphiphilic analogue of 1-β-Darabinofuranosyl-cytosine (AraC), N4-octadecyl-AraC (NOAC) [91]. The authors reported the conversion of this conjugate into AraC in vivo, effectively making NOAC an AraC prodrug [92]. When NOACloaded liposomes were incubated in full blood ex vivo, a considerable proportion of the NOAC was transferred to LPs, especially LDLs [93]. They later reported that prodrug-loaded LDLs retained their affinity for the LDL-R, and that NOAC was much more cytotoxic when encapsulated in LPs than in liposomes, as a consequence of its LDL-Rmediated uptake. As the NOAC transfer to LPs would be expected to occur in vivo following the administration of NOAC-loaded liposomes, this prodrug was considered especially suitable for the treatment of LDL-R-overexpressing malignancies. Noteworthy, blocking the LDL–RME did not completely suppress the uptake of NOAC, indicating that the passive diffusion of NOAC from LPs to cellular membranes constituted a secondary cell uptake mechanism [94]. Indeed, the passive diffusion of similar amphiphilic prodrugs through the plasma membrane can be expected in a number of cases. In this perspective, the next section of this article focusses on prodrug-based strategies aiming to enhance the passive diffusion of the anticancer agents through the plasma membrane of cancer cells. 2. Passive diffusion through the plasma membrane Enhancing the passive diffusion through the cell membrane of anticancer agents is particularly challenging in the case of small hydrophilic molecules such as nucleoside analogues that are otherwise unable to cross the plasma membrane without the help of specialised membrane transporters, whose inactivation/down-regulation can be a source of resistance [95]. This could be achieved in most cases by coupling various natural lipids such as fatty acids, phospholipids or terpenes with hydrophilic anticancer agents. This

13

procedure yielded (predominantly hydrophobic) amphiphilic conjugates able, in general, to improve the intracellular delivery of the attached drug. However, the gain in terms of uptake rate had also to be put into perspective with the lower aqueous solubility of these conjugates, which may negatively influence their bioavailability. A summary of the uptake pathways reviewed in this section is reported in Scheme 2. 2.1. Amphiphilic prodrugs administered alone In this subsection, we review several examples of amphiphilic prodrugs of very hydrophilic and poorly diffusing anticancer compounds. In a first example, several nucleoside analogues, once coupled to phospholipids, generated prodrugs that self-assembled into micellar structures upon sonication [96]. Noteworthy, the antimetabolite AraC was conjugated to a thioether–lipid moiety (Fig. 6(A)) to generate thermodynamically stable micellar disks of 10–40 nm in diameter [97]. The in vivo pharmacological evaluation of this compound on several leukaemia and solid tumour models in vivo indicated a better anticancer efficacy than AraC, which was attributed to the enhanced retention of the conjugate in cancer cells due to its diffusive cell uptake [98]. In another example, 5-fluoro-2′-deoxyuridine (5FdU) was coupled to lipophilic 5FdU derivatives by a phosphonate bond or to phospholipids to generate five different amphiphilic hetero-dinucleoside phosphates or liponucleotides [99]. The in vitro activity of these prodrugs was investigated on the 60 human cancer cell lines of the US National Cancer Institute. The architecture of the prodrugs had a considerable impact on their cytotoxicity. A promising compound more active than 5FdU on several cancer cell lines was even identified (5-fluoro-2'-deoxyurididylyl-(3'→5')-5-fluoro-N4-octadecyl-2'deoxycytidine; Fig. 6(B)). The passive diffusion of the compounds through cellular membranes was assumed as structurally similar, less hydrophobic heterodinucleoside prodrugs of antiviral nucleoside analogues had previously been proven to enter cells by a transporter-independent passive diffusion [100]. An example of standalone amphiphilic prodrug that actually reached clinical trials stemmed from 11 different C16–C20 fatty acids (containing up to three unsaturations), esterified onto the 5′-OH moiety of AraC to generate a prodrug library [101]. These compounds were evaluated on solid tumour and leukaemia cell lines, and on their resistant variants resistant to AraC or its functional and structural analogue. A clear structure–activity relationship was found in sensitive cell lines, the prodrug activity being favoured by short and unsaturated chains, due to differences in intracellular hydrolysis rates. Thanks to their diffusive uptake that bypasses membrane transporters, these amphiphilic prodrugs considerably lowered the resistance factor of the AraC- or gemcitabine-resistant cell lines (2′,2′-difluorodeoxycytidine, dFdC). The C18 mono-unsaturated (C18:1) AraC prodrug (CP-4055) identified over the course of this study was investigated in vivo, and displayed a good anticancer efficacy [102]. Additional studies conducted on this compound and a similar C18:1 prodrug of dFdC (CP-4126) clearly confirmed the activity of both prodrugs on a transporter-deficient leukaemia cell line, in vitro [103]. This suggested a potential application in the field of resistant cancers caused by a deficit in membrane transport activity. Both compounds are currently under phase I clinical trials (Table 1), and preliminary results indicated a good patient tolerance and safety profile [104,105]. Noteworthy, a stabilisation of the disease was observed in a few patients bearing ovarian, NSCLC or malignant melanoma for CP-4055. Similarly, CP-4126 stopped the disease progression in a few cases of advanced pancreas, colon and ovarian cancer. Squalene (SQ) is a natural acyclic molecule that has the unique property of being transformed into the cyclic derivative lanosterol (a precursor of cholesterol) by spontaneously passing through a highly coiled, compact conformation, without the aid of either coenzymes or

14

L. Bildstein et al. / Advanced Drug Delivery Reviews 63 (2011) 3–23

Scheme 2. Passive diffusion of anticancer prodrugs, and their intracellular metabolization scheme.

biochemical energy (adenosine triphosphate, ATP). Due to the nontoxic and biocompatible characteristics of SQ, Couvreur et al [103]. have taken advantage of this unique molecular conformation to use this natural lipid as a carrier for the delivery of nucleoside analogues, including the anticancer compound gemcitabine (Fig. 6 (C)) [106]. In aqueous media, this bioconjugate self-assembled into stable nanoparticles, which displayed a well-defined inner supramolecular organisation on the nanoscale level [107]. Mechanistic studies revealed that SQ-dFdC nanoassemblies did not enter cells by endocytosis, but that these dynamic structures exchanged SQ-dFdC molecules with plasmatic proteins, which resulted in a diffusive uptake of the prodrug modulated by the concentration and nature of extracellular proteins [108]. X-ray diffraction studies showed that SQdFdC strongly interacted with phospholipids’ bilayers [109]. This supported the fact that the prodrug was embedded within cellular membranes, which constituted an intracellular reservoir of the prodrug. Cellular metabolisation experiments confirmed that the cleavage occurred intracellularly, and in vitro cytotoxicity assays underlined the potential of the membrane-transporter-independent uptake pathway of SQ-dFdC against cancer cell lines resistant to dFdC, due to a lack of membrane transporters [110]. The authors also demonstrated that dFdC released in the cell cytoplasm could exit the cancer cells through concentrative membrane transporters, which opened up the possibility of a ‘bystander’ effect on neighbouring cells, thereby potentiating the anticancer effects of the nanoassemblies (Scheme 2). Two design features of squalenoyl-gemcitabine were especially interesting from an in vivo perspective. First, due to the protection of the sensitive amine function of dFdC by the SQ moiety,

SQ-dFdC exhibited a more prolonged plasmatic half-life than its parent drug [111]. Second, the self-assembly properties of SQ-dFdC and of other nucleoside analogues bioconjugated to SQ addressed the issue of the prodrug's limited aqueous solubility without the need for additional surface-active agents; the nanoparticulate state of the bioconjugate allowed a passive targeting of solid tumours, thanks to the EPR effect. Indeed, biodistribution studies revealed that SQ-dFdC accumulation was sixfold higher than that of dFdC inside the spleen, which is one of the main metastasis sites in the case of haematological malignancies. Taken together, these two features resulted in a much higher anticancer efficacy of SQ-dFdC with respect to dFdC in vivo, in mice either inoculated with an aggressive leukaemia model or bearing a subcutaneously grafted solid tumour model [112,113]. Noteworthy, a very recent study has also shown that the concept of ‘squalenoylation’ may be applied to other non-nucleoside anticancer molecules, such as PCT, whose conjugation with SQ also led to the spontaneous formation of nanoparticles in water [114]. 2.2. Amphiphilic prodrugs encapsulated within targeted liposomes An important shortcoming of many amphiphilic prodrugs presented above is that they provide no active targeting towards neoplasic cells. Several groups have therefore encapsulated amphiphilic prodrugs of anticancer agents in targeted or environmentsensitive liposomes to address this issue. Contrary to the strategy described earlier for NOAC, an amphiphilic prodrug of AraC, whose partition into LPs was desired, it is here essential that the promoiety ensures the stable encapsulation of the prodrug until reaching the

L. Bildstein et al. / Advanced Drug Delivery Reviews 63 (2011) 3–23

Fig. 6. Amphiphilic prodrugs of hydrophilic drugs. The bracketed numbers recall the corresponding literature references. A: Structure of a thioether phospholipid prodrug of AraC that self-assembles into micellar disks upon sonication. B: Structure of a potent heterodimeric amphiphilic prodrug of 5FdU. C: Structure of squalenoyl gemcitabine, a prodrug that self-organizes into nanoparticles. D: Structure of the amphiphilic methotrexate prodrug designed for encapsulation within phospholipase-sensitive liposomes. E: Dipalmitoyl derivative of 5-FdU encapsulated within immunoliposomes.

pharmacological target where subsequent passive diffusion through the plasma membrane should occur. Below are discussed examples of drug-delivery strategies that illustrate how these delicate specifications may, or may not, be fulfilled. 5-FdU was conjugated to fatty acids of different chain lengths via an ester bond on the 5′-OH moiety of the sugar. Liposomes containing the resulting prodrugs were prepared, and functionalised with an MAb directed against CAR-3, an epitope present on colon cancer cells [115]. 5′-Palmitoyl-5-FdU led to the best encapsulation efficiency. The resulting construct was found to be more cytotoxic in vitro than free 5-FdU. In vivo assays confirmed the anticancer efficacy of the liposomal prodrug as well as the improvement in tumour control brought by the active targeting approach. The RME of the immunoliposomes was considered unlikely as they measured approximately 250 nm in diameter, which is considerably larger than the cellular structures that mediate the endocytotic uptake [5]. The passive diffusion of the prodrug through the cell membrane after leakage from the liposomal bilayers was suggested to be the most likely cell uptake pathway. An amphiphilic prodrug of the antimetabolite MTX (Fig. 6(D)) was also encapsulated into PEGylated dipalmitoylphosphatidylcholine/ dipalmitoylphosphatidylglycerol/dipalmitoylphosphatidylethanolamine-PEG2k liposomes sensitive to phospholipase A2 (PLA2) degradation [116]. Noteworthy, this secretory enzyme is overexpressed in the interstitium of solid tumours, allowing a specific degradation of

15

the carrier with subsequent release of the prodrug in close vicinity of the cancer cells. Additional cytotoxicity investigations have validated that the uptake of the non-encapsulated prodrug could occur, through a mechanism that the authors assumed to be passive diffusion. The in vitro evaluation of the whole construct on a PLA2-secreting cell line confirmed such an anticancer activity. Although the addition of free PLA2 increased the cytotoxicity of the construct, thereby proving the whole targeting concept, the inhibition of PLA2 activity failed to suppress the cytotoxicity. This observation indicated that the degradation of the carrier was not required for the uptake of the prodrug, which suggested that the conjugate could also diffuse spontaneously from the liposomes to the cellular membranes. Differential scanning calorimetry (DSC) measurements confirmed this observation, as a diffusive transfer of the prodrug was observed upon incubation of prodrug-loaded liposomes with pure dimyristoylphosphatidylcholine (DMPC) and initially drug-free liposomal probes, as evidenced by a perturbation of the main phase transition of DMPC. These elements called for a redesign of the prodrug to provide an increased anchoring stability, to avoid a premature release with possible adverse effects on healthy tissue. PEGylated immunoliposomes directed against a metastatic colon cancer epitope were loaded with a dipalmitoyl derivative of 5-FdU in an attempt to target metastases (Fig. 6(E)) [117]. In vitro studies indicated that the immunoliposomes actively targeted colon cancer cell lines, but that, due to their excessive size, they bound to the plasma membrane without being internalised and/or metabolised within lysosomes [118]. More interestingly, however, endocytosis inhibitors considerably reduced the intracellular metabolisation of the prodrug. Taking this observation into account, the proposed uptake mechanism was that dipalmitoyl-5FdU: (1) diffused from the immunoliposomes into the plasma membrane, (2) was internalised over the course of fluid-phase endocytotic processes and (3) was then activated within the lysosomes (Scheme 2). As mentioned above for the SQ-dFdC prodrug, the authors reported that the intracellularly released 5-FdU could exit the cells, and be taken up by neighbouring cells to induce a ‘bystander effect’ [118]. This could be a key feature in the case of cells devoid of the MAb's epitope, which would otherwise not have been affected by the immunoliposomes. Indeed, different phenotypes may be encountered between cancer cells within the same tumour, which can be a significant drawback in the case of targeted therapies that do not encompass a ‘bystander effect’ [119]. An additional advantage of this feature is that it may damage the tumour neovasculature, which is essential to the growth of the tumour while not expressing cancerspecific surface epitopes [120]. 3. Controlled intracellular release of the promoiety After an uptake of the prodrug by endocytosis or by passive diffusion, the active drug must, in most cases, be cleaved from the promoiety to exert its anticancer activity. The design of linkers able to ensure simultaneously the extracellular stability of the prodrug and a rapid intracellular cleavage is, therefore, a central issue in prodrugbased drug delivery strategies. In this section, chemical moieties responsive to intracellular stimuli will be reviewed, followed by the description of two self-immolative linkers that further improve the versatility of stimuli-responsive systems. The intracellular stimuli are recalled in Scheme 1. 3.1. Stimuli-responsive groups 3.1.1. pH-labile hydrazone linkers Anticancer conjugates internalised by endocytosis will end up inside endosomes that are progressively converted to lysosomes, whose pH ranges between 4.5 and 5 [121]. Many researchers have taken advantage of this pH decrease to generate linkers that undergo a

16

L. Bildstein et al. / Advanced Drug Delivery Reviews 63 (2011) 3–23

spontaneous hydrolysis in an acidic environment, while remaining stable at pH N 7 in the general circulation. Most of these approaches encompassed the pH-sensitive hydrazone function. This moiety has been successfully employed in a variety of architectures, including some of those described in the first part of the present article. Noteworthy, hydrazone-containing prodrugs have been designed using linear polymers [122], MAbs [71] or polymeric micelles [40]. Measuring the release kinetics of the anticancer agent at various pH was a key methodology in all these studies. In general, b5% drug release was observed at pH ≥ 7, and at least 30–50% release occurred at pH ~ 5, over 48–72 h. Noteworthy, it was observed that hydrazone-based prodrugs induced a progressive intracellular drug release over several hours or days rather than a burst release effect [123]. An interesting investigation on the relationship between linker chemistry and prodrug intracellular activation was reported during the development of the marketed anticancer agent Mylotarg, the antibody–calicheamicin conjugate already mentioned previously (Fig. 5(A)). Calicheamicin was conjugated to various hydrazidebearing spacers through a disulphide bond, and the resulting constructs were coupled to aldehyde-bearing oxidised carbohydrates of the MAb's hinge region, through hydrazone bonds [59]. The nature of the groups in alpha of the hydrazone function had a considerable impact on the pH sensitivity of the linkage. No direct correlation between the in vitro cytotoxicity and the hydrolysis rate at pH 4.5 was observed. However, the most active prodrugs in vivo were stable at pH 7.4 and rapidly cleaved at pH 4.5, as expected. An alternative linker encompassing a non-pH-labile amide bond, coupled to lysine residues of the MAb, was also reported [124]. In this system, the release of calicheamicin was supposed to rely solely on the intracellular oxidation of a hindered disulphide bond (see next subsection). Although this conjugate had the same affinity for CD33 as Mylotarg, it was hundreds of times less cytotoxic in vitro and considerably less active in vivo and ex vivo. This clearly indicated that the incorporation of an optimised pH-labile hydrazone bond was compulsory for the effective intracellular activation of the calicheamicin–MAb conjugate.

3.1.2. Hindered disulphide bonds Thiol concentrations are lower in serum than inside the cells, due to the important intracellular concentrations of natural antioxidants such as glutathione [125]. Consequently, disulphide linkers are cleaved considerably faster in the reductive intracellular environment than in the general circulation, thereby providing a triggered intracellular release opportunity for prodrugs encompassing a disulphide bond. Early studies conducted in the 1980s indicated that sterically hindered disulphide linkers were stable enough in the bloodstream to ensure that most of the cleavage occurred intracellularly [126,127]. This strategy has been notably investigated in the field of MAb–drug conjugates [57]. From a design perspective, an obvious requirement of MAb–drug conjugates encompassing a disulphide linker is the presence of a thiol moiety on the active drug. A few anticancer agents naturally possess a thiol ‘handle’ (e.g., calicheamicin). In the case of ‘thiol-less’ cytotoxic molecules, a chemical modification of the molecule was therefore required, while ensuring that this alteration did not hamper the anticancer activity of the drug. An example of this strategy was reported for taxoid derivatives [61]. The authors generated several PCT analogues containing the methyldisulphanyl group required for the disulphide linkage to a MAb. The choice of the coupling site on the PCT molecule was guided by previous structure–activity relationship studies [128]. Powerful derivatives with an IC50 below 1 nmol l–1 were identified and conjugated to the functionalised lysine residues of MAbs directed against a tumour-specific antigen (epidermal growth factor). The optimised bioconjugate displayed a satisfying binding affinity and cytotoxicity in vitro together with an impressive

inhibition of tumour growth in vivo, which resulted from an efficient intracellular release of the PCT derivative. Two different prodrugs of NAc-gamma calicheamicin were also investigated that only differed in their linker chemistry. In both cases, NAc-gamma calicheamicin was attached to the linker through a sterically hindered disulphide bond. One prodrug was linked to oxidised carbohydrate residues of an MAb (specific of the MUC1 antigen) through a hydrazone bond (linker chemistry identical to that of Mylotarg®, Fig. 5(A)), while the other was attached through a nonpH-labile amide linkage on lysine residues (Fig. 5(D) [129]). While both prodrugs displayed a similar activity in vitro on non-resistant cell lines, the amide–disulphide conjugate was significantly more cytotoxic than the hydrazone–disulphide prodrug, on a P-gp-overexpressing MDR cell line. The authors unfortunately failed to provide a satisfactory explanation for this observation. Besides, both immunoconjugates displayed a significant antitumour efficacy in vivo on a resistant cell line known to possess high intracellular glutathione levels, which was consistent with the proposed reductive intracellular release of the active drug. Noteworthy, these results strongly contrasted with those reported in the previous subsection for the anti-CD33 conjugate Mylotarg: although lysine–amide–disulphide prodrugs of both antibodies were prepared similarly, the activity of the anti-CD33 amide conjugate was weaker than the hydrazonecontaining prodrug, whereas the anti-MUC1 amide conjugate was as active as its carbohydrate–hydrazone analogue, or even superior against resistant cell lines [124]. This example illustrates that the nature of the MAb can considerably influence the overall properties of the resulting immunoconjugate, and underlines the importance of optimising the chemistry of the linker respectively to the entire prodrug architecture. In a last example reported in the early 1990s, the already extremely potent microtubule-targeting agent maytansine (Fig. 1) was derivatised into analogues that encompassed a thiol ‘handle’ to generate disulphide-based conjugates, which was at the time the most reliable option for specific intracellular release [130]. The derivative was conjugated to an MAb specific to a mucin-type glycoprotein expressed to various extents in different human cancer cells. An in vitro competition-binding assay revealed that the specific affinity of the MAb was not altered by its conjugation to the maytansine derivative, and the conjugate was extremely active in vitro. The substitution of the disulphide bond by a non-cleavable thioether decreased the in vitro activity by a factor of 200, thereby proving that the disulphide prodrug was, indeed, cleaved intracellularly. After impressive complete remissions of tumour xenograft models in vivo [131], the compound entered clinical trials and the first phase I studies showed a good toxicology profile and an encouraging preliminary activity (Table 1) [132]. However, pharmacokinetic studies revealed that the clearance of maytansine was more rapid than that of the MAb, which was symptomatic of a premature prodrug cleavage that considerably reduced the therapeutic window [133]. These negative results suggested that a redesign of the conjugate should be undertaken, with a linker more stable in the bloodstream. 3.1.3. Enzyme-sensitive spacers Lysosomes contain dozens of different enzymes that are able to efficiently metabolise proteins, fats and carbohydrates [134]. Among them, cathepsins B and D are respectively cysteyl- and aspartylproteases that cleave a variety of peptides, which can easily be incorporated as spacers in prodrugs’ constructs. When secreted extracellularly, cathepsins contribute to the degradation of the extracellular matrix, which enhances the invasive capability of tumours [135]. This is why the cellular expression of cathepsins is often correlated to the invasivity and metastatic potential of tumour cells. Consequently, these proteases can contribute to improve the specificity of anticancer prodrugs for their target cells, as cathepsincleavable prodrugs should undergo a superior metabolisation inside

L. Bildstein et al. / Advanced Drug Delivery Reviews 63 (2011) 3–23

cathepsin-overexpressing neoplasic cells than inside healthy cells. Noteworthy, this strategy is only valid in the case of anticancer agents that are able to tolerate the harsh chemical environment inside lysosomes. Although the cathepsin-mediated cleavage of amphiphilic prodrugs able to passively diffuse through cellular membranes has been reported [136], this intracellular release mechanism is especially suitable for prodrugs internalised by endocytosis. Indeed, a considerable number of reports have described the design of macromolecule-drug conjugates by means of peptide spacers in the early 1980s [137,138]. The group of Kopecek has put in a great deal of effort into optimising the peptide linkers of their HPMA prodrugs. The lysosomal release rate of a model drug coupled to HPMA via different peptide spacers led to the identification of the Gly–Phe–Leu–Gly (GFLG) sequence (Fig. 2(A)) [139,140]. It was shown that a tetrapeptide was the optimal spacer length, and that GFLG was, indeed, cathepsin B cleavable. Other studies indicated that the nature of the active drug coupled to the GFLG spacer could considerably influence the cleavage rate [141]. In this perspective, a model was required to prove the cleavability of the GFLG spacer, when conjugated to various polymer– drug combinations. The most usual strategy was to compare the cleavability of a GFLG-based prodrug with that of a Gly–Gly (GG)based prodrug, as the GG dipeptide is not cleaved by lysosomal proteases [142]. Since then, these two spacers have been used as a joint model to confirm that the activity of various drug–polymer conjugates actually resulted from the lysosomal cleavage of the promoiety [143,144]. Noteworthy, this lysosomal metabolisation of HPMA–GFLG–DOX conjugates was shown to moderate the impact of MDR phenotypes, thus overcoming resistance [18]. A similar approach was reported for poly-[N5-(2-hydroxyethyl)-Lglutamine] (PHEG) prodrugs of MMC, where several oligopeptide spacers were investigated (Fig. 2(D)) [139]. In this case, a Gly–Phe– Ala–Leu (GFAL) spacer turned out to be the best compromise, as it was rapidly cleaved inside lysosomes, and exhibited a better stability than GFLG in the extracellular medium. In vitro studies pointed out a clear correlation between the hydrolysis rate of the oligopeptide and the cytotoxicity of the conjugate [145]. The PHEG–GFAL–MMC conjugate exhibited a better in vivo efficacy against leukaemia and colorectal cancer models than free MMC did. The comparison of this study with the case of HPMA–DOX conjugates, where GFLG was the optimised spacer sequence, once more underlines the importance of tailoring the chemistry of the linker to each promoiety–drug couple.

3.2. Double prodrug approach Because, as discussed before, the linkers have to combine an adequate stability in the extracellular medium with a rapid intracellular cleavage, the cleavage site must coincide with the chemical bond between the linker and the active drug. This considerably reduces the available chemical options, and may be difficult to achieve in the case of cumbersome drugs with reduced accessibility to the cleavage site. The use of a ‘double’ prodrug approach, that is, making a prodrug of a prodrug, provides a mean to overcome this problem [146]. As schematised in Fig. 7, ‘double’ prodrugs are tripartite entities. The key feature of this strategy is the introduction of an additional linker between the stimuli-sensitive moiety (specifier) and the active drug. This linker must exhibit the two following properties: (1) be attached to the drug in a stable fashion as long as the specifier-linker bond is intact and (2) rapidly and spontaneously release the active drug, once the specifier–linker bond is hydrolysed (e.g., following exposure to one of the abovementioned stimuli) [147]. The main advantages of this strategy are that the presence of the linker reduces the steric influence of the drug on the cleavage site and provides more options in terms of specifier chemistries than in the case of ‘simple’ prodrugs.

17

3.2.1. Benzyl elimination Benzyl elimination (BE)-based linkers self-eliminate because of a cascade shift in electron pairs following the cleavage of the specifier moiety (Fig. 7). The most widely used BE linker is para-aminobenzyl carbamate (PABC), which can be easily conjugated to a variety of amine-bearing drugs through a carbamate bond [148]. PCT, MMC and DOX were conjugated to a cathepsin B-cleavable Phe–Lys dipeptide through a PABC spacer [149]. These conjugates were extremely stable in human plasma, and their half-lives in the presence of cathepsin B or rat liver lysosomes ranged between 8 and 66 min, which was considered acceptable. The importance of the PABC spacer was clearly evidenced, as a Phe–Lys–DOX conjugate lacking PABC was not cleaved by cathepsin B, due to the steric influence of DOX on the cleavage site. In subsequent studies, the authors prepared DOX immunoconjugates by grafting branched spacers on reduced disulphide bonds borne by the cancer-targeting MAb BR96 (Fig. 5(C)). The Phe–Lys–PABC–DOX sequence optimised in the previous study constituted the extremity of the branched spacers [150]. As expected, the compound was stable in plasma and rapidly released DOX intracellularly, leading to a good in vitro cytotoxicity. Unfortunately, no investigation concerning the in vivo efficacy of this complex immunoconjugate was reported. Another PABC-containing immunoconjugate directed against the CD30 antigen, a marker of Hodgkin disease, was reported. A derivative of the extremely potent and cytotoxic auristatin E was conjugated through a cathepsin-sensitive Val-Citrullin-PABC spacer (Fig. 7, example A) [63]. This architecture retained a good antigen specificity, displayed an excellent stability in plasma over 10 days and exhibited a considerable in vivo efficacy and low toxicity, which clearly proved the tumour cell-specific intracellular release of the anticancer agent. The group of Greenwald designed several prodrugs of DAUN where a BE spacer was linked to a 40-kDa PEG by ester, carbonate, carbamate or amide bonds [151]. Ester and carbonate prodrugs were rapidly cleaved in plasma in less than 1 h. The introduction of steric hindrances in the vicinity of the specifier–BE linker junction, however, increased the plasmatic half-life of the conjugates to several hours, which was sufficient to induce a moderate anticancer efficacy in vivo. Amide bonds were extremely stable in plasma, but were inactive in vivo. However, the addition of a peptide spacer within the specifier of amide conjugates increased the lysosomal cleavage of these compounds, which resulted in some anticancer activity in vivo. Noteworthy, the prodrug that led to the best in vivo anticancer efficacy, even better than that of free DAUN, displayed a plasmatic half-life of 4 h. This intermediate stability was a compromise between circulating half-life and the ability to actually release DAUN intracellularly. This study illustrated the versatility of the BE platform and its ability to yield vast compound libraries for the screening and optimisation of prodrugs. A similar study was reported with PEG prodrugs of DOX, which were tested in vivo against a xenograft murine model very responsive to DOX [148]. Surprisingly, the linker structure that led to the most active PEG–DOX conjugate was a PEG–Gly–amide–PABC linker, which had actually displayed little activity in the case of DAUN prodrugs. This underlines once more the critical importance of ligand chemistry optimisation, even in the case of closely related drugs such as DOX and DAUN.

3.2.2. Trimethyl lock lactonisation Several linkers rely on an intramolecular cyclisation reaction to rapidly release active drugs following the cleavage of their specifier [152]. Trimethyl lock lactonisation (TML), for example, is a system that leads to the rapid release of amine or hydroxyl drugs in response to a variety of stimuli [153]. As shown in Fig. 7, the cleavage of the specifier leads to the intramolecular lactonisation of the compound, which results in the drug release. Due to the conformational influence

18 L. Bildstein et al. / Advanced Drug Delivery Reviews 63 (2011) 3–23 Fig. 7. Double prodrug approach: scheme of principle, mechanisms, and examples. The bracketed numbers recall the corresponding literature references. A: Structure of a BE-based auristatin derivative immunoconjugate. B: Chemical structure of a successful PEG- DAUN conjugate.

L. Bildstein et al. / Advanced Drug Delivery Reviews 63 (2011) 3–23

of the three neighbouring methyl groups (hence the name ‘trimethyl lock’), the cyclisation occurs in a matter of minutes only [154]. This linker was used to generate PEG prodrugs of DAUN [155]. These studies indicated that the cyclisation by-products were nontoxic, and that the introduction of substituents in ortho- and meta- of the phenol function considerably altered the half-life of the conjugate, due to steric effects. Like in the case of BE linkers, a successful conjugate (Fig. 7, example B) more active than free DAUN was identified during in vivo experiments [156]. The authors reported that although linking the drug to the spacer by means of an amide bond was easier than through the carbamate used in BE linkers, the adjustment of the drug-release rates was more complex in the case of the TML platform [148]. 4. Summary and outlook In the broad field of cancer treatment, the ‘grail’ of drug delivery strategies is to address the active molecule in an exclusive and specific manner to the tumour cell/tissue. In this article, we have shown that numerous prodrug-based strategies could contribute to this aim, while enhancing the uptake of anticancer agents and ensuring the intracellular release of the active compounds with an improved anticancer efficacy in vivo. While few anticancer prodrugs have actually reached the market, numerous prodrugs of cytotoxic anticancer agents are currently under clinical trials. This suggests that several other anticancer conjugates could be approved in the next few years. Indeed, the need for efficient delivery strategies such as prodrugs will keep growing in the future because new anticancer drugs become increasingly complex and difficult to deliver [1]. Noteworthy, due to the constant progress in the field of polymer chemistry, the versatile polymer–drug conjugates are viewed as one of the most promising prodrug families [23]. In a nutshell, four critical design issues have been discussed over the course of this report: (1) the targeting ability of the promoiety, (2) the stability of the construct until it reaches its target, (3) the efficiency of the cell uptake and (4) the intracellular release of the promoiety. The main conclusions of this article relatively to each of these four issues are reported in the following summary, along with technical perspectives to further improve the design of future anticancer prodrugs. First, the tissue and cell targeting of anticancer agents towards tumours is a highly desirable feature to maximise the activity of the medicines, while minimising the adverse effects on healthy tissues. As such, most of the prodrugs reported throughout the article had a builtin targeting ability. Macromolecular and self-assembled prodrugs allowed the passive targeting of solid tumours via the EPR effect. The active targeting was achieved, thanks to various moieties that displayed a high affinity for tumour-specific features. Even untargeted prodrugs designed to enter cells by passive diffusion could be advantageously encapsulated within carriers designed to deliver them in the vicinity of their pharmacological targets. An inherent advantage of the so-called active targeted strategies is that they allow the use of extremely potent anticancer agents that could not otherwise be administered, due to their important toxicity. The hydrodynamic size and sterical influence of the targeting moiety represented critical design issues. While antibodies or other macromolecules did provide efficient targeting, their access to the inside of solid tumours may be limited by their large size and poor diffusion in the tumour interstitium. Maintaining the affinity of the active targeting moieties after their chemical modification with a linker is another central concern, as such chemical modification may reduce target recognition. We also reported that the specificity of the targeting moieties had to be balanced to avoid the ‘binding site barrier’ phenomenon in the case of solid tumours. Noteworthy, although antibodies have been largely used for targeting purposes, their immunogenicity and relative fragility constitute an incentive to

19

use other types of active targeting moieties. The most thrilling perspectives in this area include new generations of peptides recognised by a variety of targets often overexpressed by cancer cells such as integrins or receptors of growth factors and hormones [157]. The alternative strategy represented by aptamers also represents a very promising targeting approach [158,159]. Regarding the stability of prodrugs in the bloodstream, we showed that most, if not all, of the research teams addressing the subject devoted a considerable energy to the evaluation of spacer libraries to avoid a premature cleavage of the promoiety in the extracellular medium. A trial-and-error approach was used mainly because the stability of the prodrug was influenced by the nature of both the promoiety and the active drug. The most widespread answer to a premature drug release consisted in increasing the steric hindrances in the vicinity of the cleavage site. Two major internalisation routes have been described for anticancer conjugates: the endocytotic uptake of the prodrug, either by fluid-phase endocytosis or RME, and the passive diffusion of amphiphilic prodrugs through the cellular membrane. As mentioned above, passive or active targeting is a central requirement for an altered cell uptake pathway of the drug. The combination of the targeting with efficient endocytotic or diffusive processes may considerably increase the drug concentration within the target tumour cells. Of course, this greatly contributes to improve the in vitro/in vivo anticancer activity, in comparison with the parent free drug. Another advantage of altering the uptake mechanism of anticancer agents with prodrugs is the possibility to overcome the resistance to the currently available treatments. The endocytotic entry followed by a lysosomal metabolisation or the passive diffusion of hydrophobic conjugates through the cellular membrane may, indeed, circumvent common resistance mechanisms such as altered membrane transport or increased efflux activity through MDR efflux transporters. Noteworthy, the development of cell-penetrating peptides (CPPs) has recently opened considerable prospects in further tailoring the cellular uptake of anticancer agents. Although the precise mechanism of action of these peptides is still subject to debate [157], they already are used to enhance the cytoplasmic delivery of proteins and other sensitive molecules [160]. However, due to the lack of inherent specificity of CPPs for neoplasic cells, the design of advanced nano-sized constructs encompassing both cell-targeting and cellpenetrating moieties will probably be required before CPPs can be applied to the field of anticancer treatment by cytotoxic drugs, with a broader perspective than the proofs of concept reported recently [161]. Finally, the triggered intracellular release of the anticancer agent, indispensable to the efficacy of most prodrug strategies, has received a great deal of attention. We have reported that several types of spacers could be cleaved in response to intracellular stimuli such as the low lysosomal pH or enzymatic activity and the reductive environment of the cytoplasm. An empirical modification of the linkers was generally used to adjust the intracellular cleavage rate of the promoieties, based on structure–activity investigations whenever possible. In particular, appropriate spacers were shown to facilitate the action of enzymes on cleavable bonds, and the lability of self-hydrolysing functions such as the hydrazone bond was adjusted by the addition of various chemical moieties in their vicinity. Self-eliminating spacers provided additional leverage in the design of efficient and versatile linkers, as they remove the requirement for a labile function to reside directly under the steric influence of the active drug. This concept was recently taken one step further with the design of ‘self-immolating dendrimers’ that, through a chain reaction, allow the release of multiple drug molecules in response to the cleavage of a single stimuli-responsive moiety [162]. Although the optimisation of these polymers is still ongoing, due to various development issues [163], proofs of concept of the triggered release of anticancer drugs from self-immolative dendrimers have been reported in biological media [164]. The massive responsiveness

20

L. Bildstein et al. / Advanced Drug Delivery Reviews 63 (2011) 3–23

of these structures, allied with the extreme architectural control potentially provided by dendrimers, may lead to considerable future advances in the field of the prodrug-based drug delivery of anticancer agents.

[23]

Acknowledgements

[25]

The financial support from the European Research Council under the European Community's Seventh Framework Programme (FP7/ 2007-2013 Grant Agreement no. 249835) and of the French ‘Agence Nationale de la Recherche’ (ANR, grant SYLIANU) is acknowledged.

[26]

[24]

[27]

References [28] [1] V.J. Stella, A case for prodrugs, in: V.J. Stella, R.T. Borchardt, M.J. Hageman, R. Oliyai, H. Maag, J. Tilley (Eds.), Prodrugs: Challenges and Rewards Part 1, Springer-AAPS Press, New York, 2007. [2] W.A. Denny, Prodrug strategies in cancer therapy, European Journal of Medicinal Chemistry 36 (2001) 577–595. [3] K.D. Bagshawe, Antibody-directed enzyme prodrug therapy, in: V.J. Stella, R.T. Borchardt, M.J. Hageman, R. Oliyai, H. Maag, J. Tilley (Eds.), Prodrugs: Challenges and Rewards Part 1, Springer-AAPS Press, New York, 2007, pp. 525–540. [4] C. Altaner, Prodrug cancer gene therapy, Cancer letters 270 (2008) 191–201. [5] H. Hillaireau, P. Couvreur, Nanocarriers’ entry into the cell: relevance to drug delivery, Cellular and Molecular Life Sciences 66 (2009) 2873–2896. [6] G.J. Doherty, H.T. McMahon, Mechanisms of endocytosis, Annual Review of Biochemistry 78 (2009) 857–902. [7] S.A. Mousavi, L. Malerød, T. Berg, R. Kjeken, Clathrin-dependent endocytosis, Biochemical Journal 377 (2004) 1–16. [8] L. Pelkmans, Secrets of caveolae-and lipid raft-mediated endocytosis revealed by mammalian viruses, Biochimica et Biophysica Acta-Molecular Cell Research 1746 (2005) 295–304. [9] C. Larsen, J. Østergaard, S.W. Larsen, Controlled Release-Macromolecular Prodrugs, in: V.J. Stella, R.T. Borchardt, M.J. Hageman, R. Oliyai, H. Maag, J. Tilley (Eds.), Prodrugs: Challenges and Rewards Part 1, Springer-AAPS Press, New York, 2007. [10] W.A. Marinaro, V.J. Stella, Macromolecular Prodrugs of Small Molecules, in: V.J. Stella, R.T. Borchardt, M.J. Hageman, R. Oliyai, H. Maag, J. Tilley (Eds.), Prodrugs: Challenges and Rewards Part 1, Springer-AAPS Press, New York, 2007, pp. 289–321. [11] H. Maeda, Y. Matsumura, Tumoritropic and lymphotropic principles of macromolecular drugs, Critical Reviews in Therapeutic Drug Carrier Systems 6 (1989) 193–210. [12] Y. Murakami, Y. Tabata, Y. Ikada, Tumor accumulation of poly (ethylene glycol) with different molecular weights after intravenous injection, Drug Delivery 4 (1997) 23–31. [13] L.W. Seymour, R. Duncan, J. Strohalm, J. Kopecek, Effect of molecular weight (Mw) of N-(2-hydroxypropyl) methacrylamide copolymers on body distribution and rate of excretion after subcutaneous, intraperitoneal, and intravenous administration to rats, Journal of Biomedical Materials Research 21 (1987) 1341–1358. [14] Y. Noguchi, J. Wu, R. Duncan, J. Strohalm, K. Ulbrich, T. Akaike, H. Maeda, Early phase tumor accumulation of macromolecules: a great difference in clearance rate between tumor and normal tissues, Cancer Science 89 (1998) 307–314. [15] J.G. Shiah, M. Dvorak, P. Kopeckova, Y. Sun, C.M. Peterson, J. Kopecek, Biodistribution and antitumour efficacy of long-circulating N-(2-hydroxypropyl)methacrylamide copolymer-doxorubicin conjugates in nude mice, European Journal of Cancer 37 (2001) 131–139. [16] R. Savic, A. Eisenberg, D. Maysinger, Block copolymer micelles as delivery vehicles of hydrophobic drugs: Micelle-cell interactions, Journal of Drug Targeting 14 (2006) 343–355. [17] R. Haag, F. Kratz, Polymer therapeutics: Concepts and applications, Angewandte Chemie-International Edition 45 (2006) 1198–1215. [18] T. Minko, P. Kopecková, J. Kopecek, Comparison of the anticancer effect of free and HPMA copolymer-bound adriamycin in human ovarian carcinoma cells, Pharmaceutical Research 16 (1999) 986–996. [19] T. Minko, P. Kopeckova, J. Kopecek, Efficacy of the chemotherapeutic action of HPMA copolymer-bound doxorubicin in a solid tumor model of ovarian carcinoma, International Journal of Cancer 86 (2000) 108–117. [20] Y. Miyamoto, T. Oda, H. Maeda, Comparison of the cytotoxic effects of the highand low-molecular-weight anticancer agents on multidrug-resistant Chinese hamster ovary cells in vitro, Cancer Research 50 (1990) 1571–1575. [21] M. Terwogt, M. Jetske, W.W. ten Bokkel Huinink, J.H.M. Schellens, M. Schot, I.A.M. Mandjes, M.G. Zurlo, M. Rocchetti, H. Rosing, F.J. Koopman, J.H. Beijnen, Phase I clinical and pharmacokinetic study of PNU166945, a novel water-soluble polymer-conjugated prodrug of paclitaxel, Anti-Cancer Drugs 12 (2001) 315–323. [22] P.A. Vasey, S.B. Kaye, R. Morrison, C. Twelves, P. Wilson, R. Duncan, A.H. Thomson, L.S. Murray, T.E. Hilditch, T. Murray, S. Burtles, D. Fraier, E. Frigerio, J. Cassidy, Phase I clinical and pharmacokinetic study of PK1 [N-(2-hydroxypropyl) methacrylamide copolymer doxorubicin]: first member of a new class of

[29]

[30]

[31] [32]

[33] [34]

[35]

[36]

[37]

[38] [39]

[40]

[41]

[42]

[43]

[44]

[45]

[46]

[47]

[48]

chemotherapeutic agents-drug-polymer conjugates. Cancer Research Campaign Phase I/II Committee, Clinical Cancer Research 5 (1999) 83–94. R. Duncan, Polymer conjugates as anticancer nanomedicines, Nature Reviews Cancer 6 (2006) 688–701. D. Bissett, J. Cassidy, J.S. De Bono, F. Muirhead, M. Main, L. Robson, D. Fraier, M.L. Magne, C. Pellizzoni, M.G. Porro, R. Spinelli, W. Speed, C. Twelves, Phase I and pharmacokinetic (PK) study of MAG-CPT (PNU 166148): a polymeric derivative of camptothecin (CPT), British Journal of Cancer 91 (2004) 50–55. J.M. Rademaker-Lakhai, C. Terret, S.B. Howell, C.M. Baud, R.F. de Boer, D. Pluim, J.H. Beijnen, J.H.M. Schellens, J.P. Droz, A Phase I and pharmacological study of the platinum polymer AP5280 given as an intravenous infusion once every 3 weeks in patients with solid tumors, Clinical Cancer Research 10 (2004) 3386–3395. S.A. Shaffer, C.B. Lee, A. Kumar, J. Singer, Proteolysis of Xyotax by Lysosomal Cathepsin B; Metabolic Profiling in Tumor Cells Using LC-MS, European Journal of Cancer 38 (2002) 428. C.D. Conover, R.B. Greenwald, A. Pendri, C.W. Gilbert, K.L. Shum, Camptothecin delivery systems: enhanced efficacy and tumor accumulation of camptothecin following its conjugation to polyethylene glycol via a glycine linker, Cancer Chemotherapy and Pharmacology 42 (1998) 407–414. D. Yu, P. Peng, S.S. Dharap, Y. Wang, M. Mehlig, P. Chandna, H. Zhao, D. Filpula, K. Yang, V. Borowski, Antitumor activity of poly (ethylene glycol)–camptothecin conjugate: The inhibition of tumor growth in vivo, Journal of Controlled Release 110 (2005) 90–102. E.K. Rowinsky, J. Rizzo, L. Ochoa, C.H. Takimoto, B. Forouzesh, G. Schwartz, L.A. Hammond, A. Patnaik, J. Kwiatek, A. Goetz, L. Denis, J. McGuire, A.W. Tolcher, A phase I and pharmacokinetic study of pegylated camptothecin as a 1-hour infusion every 3 weeks in patients with advanced solid malignancies, Journal of Clinical Oncology 21 (2003) 148–157. O.L. Padilla De Jesus, H.R. Ihre, L. Gagne, J.M.J. Frechet, F.C. Szoka, Polyester dendritic systems for drug delivery applications: In vitro and in vivo evaluation, Bioconjugate Chemistry 13 (2002) 453–461. S. Svenson, Dendrimers as versatile platform in drug delivery applications, European Journal of Pharmaceutics and Biopharmaceutics 71 (2009) 445–462. M. El-Sayed, M. Ginski, C. Rhodes, H. Ghandehari, Transepithelial transport of poly (amidoamine) dendrimers across Caco-2 cell monolayers, Journal of Controlled Release 81 (2002) 355–365. M. Liu, J.M.J. Fréchet, Designing dendrimers for drug delivery, Pharmaceutical Science & Technology Today 2 (1999) 393–401. J.B. Wolinsky, M.W. Grinstaff, Therapeutic and diagnostic applications of dendrimers for cancer treatment, Advanced Drug Delivery Reviews 60 (2008) 1037–1055. K. Kono, C. Kojima, N. Hayashi, E. Nishisaka, K. Kiura, S. Watarai, A. Harada, Preparation and cytotoxic activity of poly (ethylene glycol)-modified poly (amidoamine) dendrimers bearing adriamycin, Biomaterials 29 (2008) 1664–1675. C.C. Lee, E.R. Gillies, M.E. Fox, S.J. Guillaudeu, J.M. Frechet, E.E. Dy, F.C. Szoka, A single dose of doxorubicin-functionalized bow-tie dendrimer cures mice bearing C-26 colon carcinomas, Proceedings of the National Academy of Sciences of the United States of America 103 (2006) 16649–16654. M.E. Fox, S. Guillaudeu, J.M.J. Frechet, K. Jerger, N. Macaraeg, F.C. Szoka, Synthesis and In Vivo Antitumor Efficacy of PEGylated Poly(L-lysine) DendrimerCamptothecin Conjugates, Molecular Pharmaceutics 6 (2009) 1562–1572. L.M. Bareford, P.W. Swaan, Endocytic mechanisms for targeted drug delivery, Advanced Drug Delivery Reviews 59 (2007) 748–758. S.D. Weitman, R.H. Lark, L.R. Coney, D.W. Fort, V. Frasca, V.R. Zurawski Jr., B.A. Kamen, Distribution of the folate receptor GP38 in normal and malignant cell lines and tissues, Cancer Research 52 (1992) 3396–3401. Y. Bae, W.D. Jang, N. Nishiyama, S. Fukushima, K. Kataoka, Multifunctional polymeric micelles with folate-mediated cancer cell targeting and pH-triggered drug releasing properties for active intracellular drug delivery, Molecular Biosystems 1 (2005) 242–250. M. Prabaharan, J.J. Grailer, S. Pilla, D.A. Steeber, S. Gong, Amphiphilic multi-armblock copolymer conjugated with doxorubicin via pH-sensitive hydrazone bond for tumor-targeted drug delivery, Biomaterials 30 (2009) 5757–5766. J.F. Kukowska-Latallo, K.A. Candido, Z. Cao, S.S. Nigavekar, I.J. Majoros, T.P. Thomas, L.P. Balogh, M.K. Khan, J.R. Baker Jr., Nanoparticle targeting of anticancer drug improves therapeutic response in animal model of human epithelial cancer, Cancer Research 65 (2005) 5317. H.S. Yoo, T.G. Park, Folate-receptor-targeted delivery of doxorubicin nanoaggregates stabilized by doxorubicin-PEG-folate conjugate, Journal of Controlled Release 100 (2004) 247–256. G. Russell-Jones, J. Meewan, Amplification of Biotin-Mediated targeting, US patent 2006/0127310 A1, application no. 10/535, 269, in, Access Pharmaceuticals Australia PTY LTD., USA, 2003. G. Russell-Jones, K. McTavish, J. McEwan, J. Rice, D. Nowotnik, Vitamin-mediated targeting as a potential mechanism to increase drug uptake by tumours, Journal of Inorganic Biochemistry 98 (2004) 1625–1633. T. Minko, P.V. Paranjpe, B. Qiu, A. Lalloo, R. Won, S. Stein, P.J. Sinko, Enhancing the anticancer efficacy of camptothecin using biotinylated poly(ethyleneglycol) conjugates in sensitive and multidrug-resistant human ovarian carcinoma cells, Cancer Chemotherapy and Pharmacology 50 (2002) 143–150. W. Yang, Y. Cheng, T. Xu, X. Wang, L. Wen, Targeting cancer cells with biotin– dendrimer conjugates, European Journal of Medicinal Chemistry 44 (2009) 862–868. T.W. Moody, L.C. Sun, S.A. Mantey, T. Pradhan, L.V. Mackey, N. Gonzales, J.A. Fuselier, D.H. Coy, R.T. Jensen, In vitro and in vivo antitumor effects of cytotoxic

L. Bildstein et al. / Advanced Drug Delivery Reviews 63 (2011) 3–23

[49] [50]

[51] [52]

[53] [54]

[55]

[56]

[57]

[58] [59]

[60] [61]

[62]

[63]

[64]

[65]

[66]

[67]

[68]

[69]

[70]

[71]

[72]

camptothecin-bombesin conjugates are mediated by specific interaction with cellular bombesin receptors, Journal of Pharmacology and Experimental Therapeutics 318 (2006) 1265–1272. A.V. Schally, A. Nagy, Chemotherapy targeted to cancers through tumoral hormone receptors, Trends in Endocrinology and Metabolism 15 (2004) 300–310. L.H. Wang, S.A. Mantey, J.T. Lin, H. Frucht, R.T. Jensen, Ligand binding, internalization, degradation and regulation by guanine nucleotides of bombesin receptor subtypes: a comparative study, Biochimica et Biophysica Acta 1175 (1993) 232–242. Y.C. Patel, Somatostatin and its receptor family, Frontiers in Neuroendocrinology 20 (1999) 157–198. A.V. Schally, A. Nagy, New approaches to treatment of various cancers based on cytotoxic analogs of LHRH, somatostatin and bombesin, Life Sciences 72 (2003) 2305–2320. S.M. Okarvi, Peptide-based radiopharmaceuticals and cytotoxic conjugates: Potential tools against cancer, Cancer Treatment Reviews 34 (2008) 13–26. T.W. Moody, S.A. Mantey, T.K. Pradhan, M. Schumann, T. Nakagawa, A. Martinez, J. Fuselier, D.H. Coy, R.T. Jensen, Development of high affinity camptothecin-bombesin conjugates that have targeted cytotoxicity for bombesin receptor-containing tumor cells, Journal of Biological Chemistry 279 (2004) 23580–23589. A.V. Schally, A. Nagy, Cancer chemotherapy based on targeting of cytotoxic peptide conjugates to their receptors on tumors, European Journal of Endocrinology 141 (1999) 1–14. A. Nagy, A.V. Schally, G. Halmos, P. Armatis, R.Z. Cai, V. Csernus, M. Kovács, M. Koppán, K. Szepesházi, Z. Kahán, Synthesis and biological evaluation of cytotoxic analogs of somatostatin containing doxorubicin or its intensely potent derivative, 2-pyrrolinodoxorubicin, Proceedings of the National Academy of Sciences of the United States of America 95 (1998) 1794–1799. P.D. Senter, D.L. Meyer, Monoclonal Antibody Drug Conjugates for Cancer Therapy, in: V.J. Stella, R.T. Borchardt, M.J. Hageman, R. Oliyai, H. Maag, J. Tilley (Eds.), Prodrugs: Challenges and Rewards Part 2, Springer-AAPS Press, New York, 2007, pp. 507–524. M.C. Garnett, Targeted drug conjugates: principles and progress, Advanced Drug Delivery Reviews 53 (2001) 171–216. P.R. Hamann, L.M. Hinman, I. Hollander, C.F. Beyer, D. Lindh, R. Holcomb, W. Hallett, H.R. Tsou, J. Upeslacis, D. Shochat, A. Mountain, D.A. Flowers, I. Bernstein, Gemtuzumab ozogamicin, a potent and selective anti-CD33 antibody-calicheamicin conjugate for treatment of acute myeloid leukemia, Bioconjugate Chemistry 13 (2002) 47–58. P.D. Senter, Potent antibody drug conjugates for cancer therapy, Current Opinion in Chemical Biology 13 (2009) 235–244. I. Ojima, X. Geng, X. Wu, C. Qu, C.P. Borella, H. Xie, S.D. Wilhelm, B.A. Leece, L.M. Bartle, V.S. Goldmacher, R.V.J. Chari, Tumor-Specific Novel Taxoid- Monoclonal Antibody Conjugates, Journal of Medicinal Chemistry 45 (2002) 5620–5623. E.R. Boghaert, L. Sridharan, D.C. Armellino, K.M. Khandke, J.F. DiJoseph, A. Kunz, M.M. Dougher, F. Jiang, L.B. Kalyandrug, P.R. Hamann, P. Frost, N.K. Damle, Antibody-targeted chemotherapy with the calicheamicin conjugate hu3S193-Nacetyl calicheamicin dimethyl hydrazide targets Lewisy and eliminates Lewisypositive human carcinoma cells and xenografts, Clinical Cancer Research 10 (2004) 4538–4549. J.A. Francisco, C.G. Cerveny, D.L. Meyer, B.J. Mixan, K. Klussman, D.F. Chace, S.X. Rejniak, K.A. Gordon, R. DeBlanc, B.E. Toki, C.L. Law, S.O. Doronina, C.B. Siegall, P.D. Senter, A.F. Wahl, cAC10-vcMMAE, an anti-CD30-monomethyl auristatin E conjugate with potent and selective antitumor activity, Blood 102 (2003) 1458–1465. R.H.C. van der Jagt, C.C. Badger, F.R. Appelbaum, O.W. Press, D.C. Matthews, J.F. Eary, K.A. Krohn, I.D. Bernstein, Localization of radiolabeled antimyeloid antibodies in a human acute leukemia xenograft tumor model, Cancer Research 52 (1992) 89–94. J.F. DiJoseph, D.C. Armellino, E.R. Boghaert, K. Khandke, M.M. Dougher, L. Sridharan, A. Kunz, P.R. Hamann, B. Gorovits, C. Udata, J.K. Moran, A.G. Popplewell, S. Stephens, P. Frost, N.K. Damle, Antibody-targeted chemotherapy with CMC-544: a CD22-targeted immunoconjugate of calicheamicin for the treatment of B-lymphoid malignancies, Blood 103 (2004) 1807–1814. F.P. Lindberg, D.M. Lublin, M.J. Telen, R.A. Veile, Y.E. Miller, H. Donis-Keller, E.J. Brown, Rh-related antigen CD47 is the signal-transducer integrin-associated protein, Journal of Biological Chemistry 269 (1994) 1567–1570. M.N. Saleh, S. Sugarman, J. Murray, J.B. Ostroff, D. Healey, D. Jones, C.R. Daniel, D. LeBherz, H. Brewer, N. Onetto, A.F. LoBuglio, Phase I trial of the anti-Lewis Y drug immunoconjugate BR96-doxorubicin in patients with lewis Y-expressing epithelial tumors, Journal of Clinical Oncology 18 (2000) 2282–2292. K. Fujimori, D.G. Covell, J.E. Fletcher, J.N. Weinstein, Modeling analysis of the global and microscopic distribution of immunoglobulin G, F (ab') 2, and Fab in tumors, Cancer Research 49 (1989) 5656–5663. G.P. Adams, R. Schier, A.M. McCall, H.H. Simmons, E.M. Horak, R.K. Alpaugh, J.D. Marks, L.M. Weiner, High affinity restricts the localization and tumor penetration of single-chain fv antibody molecules, Cancer research 61 (2001) 4750–4755. K.J. Hamblett, P.D. Senter, D.F. Chace, M. Sun, J. Lenox, C.G. Cerveny, K.M. Kissler, S.X. Bernhardt, A.K. Kopcha, R.F. Zabinski, D.L. Meyer, J.A. Francisco, Effects of drug loading on the antitumor activity of a monoclonal antibody drug conjugate, Clinical Cancer Research 10 (2004) 7063–7070. P.A. Trail, D. Willner, S.J. Lasch, A.J. Henderson, S. Hofstead, A.M. Casazza, R.A. Firestone, I. Hellstrom, K.E. Hellstrom, Cure of xenografted human carcinomas by BR96-doxorubicin immunoconjugates, Science 261 (1993) 212–215. H.D. King, G.M. Dubowchik, H. Mastalerz, D. Willner, S.J. Hofstead, R.A. Firestone, S.J. Lasch, P.A. Trail, Monoclonal antibody conjugates of doxorubicin prepared

[73]

[74] [75] [76]

[77]

[78]

[79]

[80] [81]

[82]

[83]

[84] [85]

[86]

[87] [88]

[89]

[90]

[91]

[92]

[93]

[94]

[95]

[96]

[97]

21

with branched peptide linkers: inhibition of aggregation by methoxytriethyleneglycol chains, Journal of Medicinal Chemistry 45 (2002) 4336–4343. V. Omelyanenko, P. Kopeckova, C. Gentry, J.G. Shiah, J. Kopecek, HPMA copolymer-anticancer drug-OV-TL16 antibody conjugates. 1. Influence of the method of synthesis on the binding affinity to OVCAR-3 ovarian carcinoma cells in vitro, Journal of Drug Targeting 3 (1996) 357–373. R. Stern, Hyaluronan catabolism: a new metabolic pathway, European Journal of Cell Biology 83 (2004) 317–325. Z. Rudzki, S. Jothy, CD44 and the adhesion of neoplastic cells, Molecular Pathology 50 (1997) 57–71. Q. Hua, C.B. Knudson, W. Knudson, Internalization of hyaluronan by chondrocytes occurs via receptor-mediated endocytosis, Journal of Cell Science 106 (1993) 365–375. K. Akima, H. Ito, Y. Iwata, K. Matsuo, N. Watari, M. Yanagi, H. Hagi, K. Oshima, A. Yagita, Y. Atomi, I. Tatekawa, Evaluation of antitumor activities of hyaluronate binding antitumor drugs: Synthesis, characterization and antitumor activity, Journal of Drug Targeting 4 (1996) 1–8. N.F. Guilbaud, N. Gas, M.A. Dupont, A. Valette, Effects of differentiation-inducing agents on maturation of human MCF-7 breast cancer cells, Journal of Cellular Physiology 145 (1990) 162–172. D. Coradini, C. Pellizzaro, G. Miglierini, M.G. Daidone, A. Perbellini, Hyaluronic acid as drug delivery for sodium butyrate: improvement of the anti-proliferative activity on a breast-cancer cell line, International Journal of Cancer 81 (1999) 411–416. Y. Luo, M.R. Ziebell, G.D. Prestwich, A Hyaluronic Acid-Taxol Antitumor Bioconjugate Targeted to Cancer Cells, Biomacromolecules 1 (2000) 208–218. E. Auzenne, S.C. Ghosh, M. Khodadadian, B. Rivera, D. Farquhar, R.E. Price, M. Ravoori, V. Kundra, R.S. Freedman, J. Klostergaard, Hyaluronic acid-paclitaxel: antitumor efficacy against CD44 (+) human ovarian carcinoma xenografts, Neoplasia (New York, NY) 9 (2007) 479–486. F. Leonelli, A. La Bella, A. Francescangeli, R. Joudioux, A.L. Capodilupo, M. Quagliariello, L.M. Migneco, R.M. Bettolo, V. Crescenzi, G. De Luca, D. Renier, A new and simply available class of hydrosoluble bioconjugates by coupling paclitaxel to hyaluronic acid through a 4-hydroxybutanoic acid derived linker, Helvetica Chimica Acta 88 (2005) 154–159. A. Rosato, A. Banzato, G. De Luca, D. Renier, F. Bettella, C. Pagano, G. Esposito, P. Zanovello, P.F. Bassi, HYTAD1-p20: a new paclitaxel-hyaluronic acid hydrosoluble bioconjugate for treatment of superficial bladder cancer, Annual Meeting of the Society of Urologic Oncology, Elsevier, 2005, pp. 207–215. R.A. Firestone, Low-density lipoprotein as a vehicle for targeting antitumor compounds to cancer cells, Bioconjugate Chemistry 5 (1994) 105–113. T.L. Innerarity, R.E. Pitas, R.W. Mahley, Binding of arginine-rich (E) apoprotein after recombination with phospholipid vesicles to the low density lipoprotein receptors of fibroblasts, Journal of Biological Chemistry 254 (1979) 4186–4190. R.C. Kowal, J. Herz, K.H. Weisgraber, R.W. Mahley, M.S. Brown, J.L. Goldstein, Opposing effects of apolipoproteins E and C on lipoprotein binding to low density lipoprotein receptor-related protein, Journal of Biological Chemistry 265 (1990) 10771–10779. A.M. Salter, D.N. Brindley, The biochemistry of lipoproteins, Journal of Inherited Metabolic Disease 11 (1988) 4–17. A.J. Versluis, E.T. Rump, P.C. Rensen, T.J. Van Berkel, M.K. Bijsterbosch, Synthesis of a lipophilic daunorubicin derivative and its incorporation into lipidic carriers developed for LDL receptor-mediated tumor therapy, Pharmaceutical Research 15 (1998) 531–537. A.J. Versluis, P.C. Rensen, E.T. Rump, T.J. Van Berkel, M.K. Bijsterbosch, Lowdensity lipoprotein receptor-mediated delivery of a lipophilic daunorubicin derivative to B16 tumours in mice using apolipoprotein E-enriched liposomes, British Journal of Cancer 78 (1998) 1607–1614. K.M. Wasan, S.M. Cassidy, Role of plasma lipoproteins in modifying the biological activity of hydrophobic drugs, Journal of Pharmaceutical Sciences 87 (1998) 411–424. R.A. Schwendener, H. Schott, Treatment of L1210 murine leukemia with liposome-incorporated N 4-hexadecyl-1-b-D-arabinofuranosyl cytosine, International Journal of Cancer 51 (1992) 466–469. R.A. Schwendener, K. Friedl, H. Depenbrock, H. Schott, A.R. Hanauske, In vitro activity of liposomal N4octadecyl-1- -D-arabinofurano-sylcytosine (NOAC), a new lipophilic derivative of 1- -D-arabino-furanocylcytosine on biopsized clonogenic human tumor cells and hematopoietic precursor cells, Investigational new drugs 19 (2001) 203–210. S.K.M. Koller-Lucae, H. Schott, R.A. Schwendener, Interactions with human blood in vitro and pharmacokinetic properties in mice of liposomal N4-octadecyl-1beta-D-arabinofuranosylcytosine, a new anticancer drug, Journal of Pharmacology and Experimental Therapeutics 282 (1997) 1572. S.K.M. Koller-Lucae, H. Schott, R.A. Schwendener, Low density lipoprotein and liposome mediated uptake and cytotoxic effect of N4-octadecyl-1- -D-arabinofuranosylcytosine in Daudi lymphoma cells, British Journal of Cancer 80 (1999) 1542. L.P. Jordheim, C. Dumontet, Review of recent studies on resistance to cytotoxic deoxynucleoside analogues, Biochimica et Biophysica Acta-Reviews on Cancer 1776 (2007) 138–159. M. MacCoss, J.J. Edwards, P. Lagocki, Y.E. Rahman, Phospholipid-nucleoside conjugates. 5. The interaction of selected 1-beta-D-arabinofuranosylcytosine-5'diphosphate-L-1, 2-diacylglycerols with serum lipoproteins, Biochemical and Biophysical Research Communications 116 (1983) 368–374. C.I. Hong, R.J. Bernacki, S.W. Hui, Y. Rustum, C.R. West, Formulation, Stability, and Antitumor-Activity of 1-Beta-D-Arabinofuranosylcytosine Conjugate of Thioether Phospholipid, Cancer Research 50 (1990) 4401–4406.

22

L. Bildstein et al. / Advanced Drug Delivery Reviews 63 (2011) 3–23

[98] R. Herrmann, W.E. Berdel, Therapeutic Activity of a Thioether-Lipid Conjugate of 1-Beta-D-Arabinofuranosylcytosine in Human Colorectal-Cancer Xenografts, Cancer Research 52 (1992) 1865–1867. [99] P.S. Ludwig, R.A. Schwendener, H. Schott, Synthesis and anticancer activities of amphiphilic 5-fluoro-2 '-deoxyuridylic acid prodrugs, European Journal of Medicinal Chemistry 40 (2005) 494–504. [100] P.A. Peghini, R. Zahner, H. Kuster, H. Schott, R.A. Schwendener, In vitro antihuman immunodeficiency virus and anti-hepatitis B virus activities and pharmacokinetic properties of heterodinucleoside phosphates containing AZT or ddC, Antiviral Chemistry & Chemotherapy 9 (1998) 117–126. [101] A.M. Bergman, C.M. Kuiper, D.A. Voorn, E.M. Comijn, F. Myhren, M.L. Sandvold, H.R. Hendriks, G.J. Peters, Antiproliferative activity and mechanism of action of fatty acid derivatives of arabinofuranosylcytosine in leukemia and solid tumor cell lines, Biochemical Pharmacology 67 (2004) 503–511. [102] K. Breistol, J. Balzarini, M.L. Sandvold, F. Myhren, M. Martinsen, E.D. Clercq, O. Fodstad, Antitumor activity of P-4055 (elaidic acid-cytarabine) compared to cytarabine in metastatic and sc human tumor xenograft models, Cancer Research 59 (1999) 2944–2949. [103] C.M. Galmarini, F. Myhren, M.L. Sandvold, CP-4055 and CP-4126 are active in ara-C and gemcitabine-resistant lymphoma cell lines, British Journal of Haematology 144 (2009) 273–275. [104] S. Aamdal, S. Dueland, M.J. Lind, H. Thomas, C.R. Franks, M.L. Sandvold, W. Rasch, Phase I trial of a nucleoside analog CP-4055 given daily for 5 days every 3 weeks in patients (pts) with advanced solid tumors-preliminary results, American Society of Clinical Oncology, 2004, p. 2049. [105] B. Nilsson, A. Hendlisz, M. Castella, S. Aamdal, S. Dueland, M. Nyakas, J. Evans, B. Venugopal, W. Rasch, A. Awada, First-in-human study of a novel nucleoside analogue, CP-4126, in patients with advanced solid tumors, American Society of Clinical Oncology, 2009, p. 2577. [106] P. Couvreur, B. Stella, L.H. Reddy, H. Hillaireau, C. Dubernet, D. Desmaele, S. Lepetre-Mouelhi, F. Rocco, N. Dereuddre-Bosquet, P. Clayette, V. Rosilio, V. Marsaud, J.M. Renoir, L. Cattel, Squalenoyl nanomedicines as potential therapeutics, Nano Letters 6 (2006) 2544–2548. [107] P. Couvreur, L.H. Reddy, S. Mangenot, J.H. Poupaert, D. Desmaele, S. LepetreMouelhi, B. Pili, C. Bourgaux, H. Amenitsch, M. Ollivon, Discovery of new hexagonal supramolecular nanostructures formed by squalenoylation of an anticancer nucleoside analogue, Small 4 (2008) 247–253. [108] L. Bildstein, V. Marsaud, H. Chacun, S. Lepêtre-Mouelhi, D. Desmaële, P. Couvreur, C. Dubernet, Extracellular-protein-enhanced cellular uptake of squalenoyl gemcitabine from nanoassemblies, Soft Matter 6 (2010) 5570–5580. [109] B. Pili, C. Bourgaux, H. Amenitsch, G. Keller, S. Lep tre-Mouelhi, D. Desmaële, P. Couvreur, M. Ollivon, Interaction of a new anticancer prodrug, gemcitabinesqualene, with a model membrane: Coupled DSC and XRD study, Biochimica et Biophysica Acta (BBA)-Biomembranes 1798 (2010) 1522–1532. [110] L. Bildstein, C. Dubernet, V. Marsaud, H. Chacun, V. Nicolas, C. Gueutin, A. Sarasin, H. Bénech, S. Lepêtre-Mouelhi, D. Desmaële, P. Couvreur, Transmembrane diffusion of gemcitabine by a nanoparticulate squalenoyl prodrug: An original drug delivery pathway, Journal of Controlled Release 147 (2010) 163–170. [111] L.H. Reddy, H. Khoury, A. Paci, A. Deroussent, H. Ferreira, C. Dubernet, X. Decleves, M. Besnard, H. Chacun, S. Lepetre-Mouelhi, D. Desmaele, B. Rousseau, C. Laugier, J.C. Cintrat, G. Vassal, P. Couvreur, Squalenoylation favorably modifies the in vivo pharmacokinetics and biodistribution of gemcitabine in mice, Drug Metabolism and Disposition 36 (2008) 1570–1577. [112] L.H. Reddy, C. Dubernet, S.L. Mouelhi, P.E. Marque, D. Desmaele, P. Couvreur, A new nanomedicine of gemcitabine displays enhanced anticancer activity in sensitive and resistant leukemia types, Journal of Controlled Release 124 (2007) 20–27. [113] L.H. Reddy, J.M. Renoir, V. Marsaud, S. Lepetre-Mouelhi, D. Desmae Test, article sample title placed here, Molecular Pharmaceutics 6 (2009) 1526–1535. [114] F. Dosio, L.H. Reddy, A. Ferrero, B. Stella, L. Cattel, P. Couvreur, Novel Nanoassemblies Composed of Squalenoyl-Paclitaxel Derivatives: Synthesis, Characterization, and Biological Evaluation, Bioconjugate Chemistry 21 (2010) 1349–1361. [115] P. Crosasso, P. Brusa, F. Dosio, S. Arpicco, D. Pacchioni, F. Schuber, L. Cattel, Antitumoral activity of liposomes and immunoliposomes containing 5-fluorouridine prodrugs, Journal of Pharmaceutical Sciences 86 (1997) 832–839. [116] T. Kaasgaard, T.L. Andresen, S.S. Jensen, R.O. Holte, L.T. Jensen, K. Jørgensen, Liposomes containing alkylated methotrexate analogues for phospholipase A2 mediated tumor targeted drug delivery, Chemistry and Physics of Lipids 157 (2009) 94–103. [117] G.A. Koning, A. Gorter, G.L. Scherphof, J. Kamps, Antiproliferative effect of immunoliposomes containing 5-fluorodeoxyuridine-dipalmitate on colon cancer cells, British Journal of Cancer 80 (1999) 1718. [118] G.A. Koning, J. Kamps, G.L. Scherphof, Efficient intracellular delivery of 5fluorodeoxyuridine into colon cancer cells by targeted immunoliposomes, Cancer Detection and Prevention 26 (2002) 299–307. [119] T.M. Allen, Ligand-targeted therapeutics in anticancer therapy, Nature Reviews Cancer 2 (2002) 750–763. [120] Y.V. Kovtun, C.A. Audette, Y. Ye, H. Xie, M.F. Ruberti, S.J. Phinney, B.A. Leece, T. Chittenden, W.A. Blattler, V.S. Goldmacher, Antibody-drug conjugates designed to eradicate tumors with homogeneous and heterogeneous expression of the target antigen, Cancer Research 66 (2006) 3214–3221. [121] S. Ohkuma, B. Poole, Fluorescence probe measurement of the intralysosomal pH in living cells and the perturbation of pH by various agents, Proceedings of the National Academy of Sciences of the United States of America 75 (1978) 3327–3331. [122] W.M. Choi, P. Kopeckova, T. Minko, J. Kopecek, Synthesis of HPMA copolymer containing adriamycin bound via an acid-labile spacer and its activity toward

[123]

[124]

[125] [126]

[127]

[128]

[129]

[130]

[131]

[132]

[133]

[134] [135]

[136]

[137]

[138]

[139]

[140]

[141] [142]

[143]

[144]

[145]

human ovarian carcinoma cells, Journal of Bioactive and Compatible Polymers 14 (1999) 447–456. Y. Bae, S. Fukushima, A. Harada, K. Kataoka, Design of environment-sensitive supramolecular assemblies for intracellular drug delivery: polymeric micelles that are responsive to intracellular pH change, Angewandte Chemie International Edition 42 (2003) 4640–4643. P.R. Hamann, L.M. Hinman, C.F. Beyer, D. Lindh, J. Upeslacis, D.A. Flowers, I. Bernstein, An Anti-CD33 Antibody- Calicheamicin Conjugate for Treatment of Acute Myeloid Leukemia. Choice of Linker, Bioconjugate Chem 13 (2002) 40–46. B. Ketterer, B. Coles, D.J. Meyer, The role of glutathione in detoxication, Environmental Health Perspectives 49 (1983) 59–69. P.E. Thorpe, P.M. Wallace, P.P. Knowles, M.G. Relf, A.N.F. Brown, G.J. Watson, R.E. Knyba, E.J. Wawrzynczak, D.C. Blakey, New coupling agents for the synthesis of immunotoxins containing a hindered disulfide bond with improved stability in vivo, Cancer Research 47 (1987) 5924–5931. N.R. Worrell, A.J. Cumber, G.D. Parnell, A. Mirza, J.A. Forrester, W.C. Ross, Effect of linkage variation on pharmacokinetics of ricin A chain-antibody conjugates in normal rats, Anti-Cancer Drug Design 1 (1986) 179–188. I. Ojima, J.C. Slater, E. Michaud, S.D. Kuduk, P.Y. Bounaud, P. Vrignaud, M.C. Bissery, J.M. Veith, P. Pera, R.J. Bernacki, Syntheses and Structure-Activity Relationships of the Second-Generation Antitumor Taxoids: Exceptional Activity against Drug-Resistant Cancer Cells, Journal of Medicinal Chemistry 39 (1996) 3889–3896. P.R. Hamann, L.M. Hinman, C.F. Beyer, L.M. Greenberger, C. Lin, D. Lindh, A.T. Menendez, R. Wallace, F.E. Durr, J. Upeslacis, An anti-MUC1 antibodycalicheamicin conjugate for treatment of solid tumors. Choice of linker and overcoming drug resistance, Bioconjugate Chemistry 16 (2005) 346–353. R.V.J. Chari, B.A. Martell, J.L. Gross, S.B. Cook, S.A. Shah, W.A. Blattler, S.J. McKenzie, V.S. Goldmacher, Immunoconjugates containing novel maytansinoids: promising anticancer drugs, Cancer Research 52 (1992) 127–131. C. Liu, B.M. Tadayoni, L.A. Bourret, K.M. Mattocks, S.M. Derr, W.C. Widdison, N.L. Kedersha, P.D. Ariniello, V.S. Goldmacher, J.M. Lambert, W.A. Blättler, R.V.J. Chari, Eradication of large colon tumor xenografts by targeted delivery of maytansinoids, Proceedings of the National Academy of Sciences of the United States of America 93 (1996) 8618–8623. A.W. Tolcher, L. Ochoa, L.A. Hammond, A. Patnaik, T. Edwards, C. Takimoto, L. Smith, J. de Bono, G. Schwartz, T. Mays, Z.L. Jonak, R. Johnson, M. DeWitte, H. Martino, C. Audette, K. Maes, R.V.J. Chari, J.M. Lambert, E.K. Rowinski, Cantuzumab mertansine, a maytansinoid immunoconjugate directed to the CanAg antigen: a phase I, pharmacokinetic, and biologic correlative study, Journal of Clinical Oncology 21 (2003) 211–222. J. Rodon, M. Garrison, L.A. Hammond, J. de Bono, L. Smith, L. Forero, D. Hao, C. Takimoto, J.M. Lambert, L. Pandite, M. Howard, H. Xie, A.W. Tolcher, Cantuzumab mertansine in a three-times a week schedule: a phase I and pharmacokinetic study, Cancer Chemotherapy and Pharmacology 62 (2008) 911–919. A.J. Barrett, M.F. Heath, Lysosomal enzymes, Lysosomes: a laboratory handbook, American Elsevier Pub. Co, New York, 1972, pp. 46–135. E. Liaudet-Coopman, M. Beaujouin, D. Derocq, M. Garcia, M. Glondu-Lassis, V. Laurent-Matha, C. Prébois, H. Rochefort, F. Vignon, Cathepsin D: newly discovered functions of a long-standing aspartic protease in cancer and apoptosis, Cancer Letters 237 (2006) 167–179. M.L. Immordino, P. Brusa, F. Rocco, S. Arpicco, M. Ceruti, L. Cattel, Preparation, characterization, cytotoxicity and pharmacokinetics of liposomes containing lipophilic gemcitabine prodrugs, Journal of Controlled Release 100 (2004) 331–346. A. Trouet, M. Masquelier, R. Baurain, D. Deprez-De Campeneere, A covalent linkage between daunorubicin and proteins that is stable in serum and reversible by lysosomal hydrolases, as required for a lysosomotropic drugcarrier conjugate: in vitro and in vivo studies, Proceedings of the National Academy of Sciences of the United States of America 79 (1982) 626–629. A. Trouet, P. Pirson, R. Steiger, M. Masquelier, R. Baurain, J. Gillet, Development of new derivatives of primaquine by association with lysosomotropic carriers, Bulletin of the World Health Organization 59 (1981) 449–458. H. Soyez, E. Schacht, S. Vanderkerken, The crucial role of spacer groups in macromolecular prodrug design, Advanced Drug Delivery Reviews 21 (1996) 81–106. P. Rejmanová, J. Kopecek, J. Pohl, M. Baudys, V. Kostka, Polymers containing enzymatically degradable bonds. 8. Degradation of oligopeptide sequences in N-(2-hydroxypropyl) methacrylamide copolymers by bovine spleen cathepsin B, Die Makromolekulare Chemie 184 (1983) 209–220. J. Kopecek, Controlled biodegradability of polymers–a key to drug delivery systems, Biomaterials 5 (1984) 19–24. V. Subr, J. Strohalm, K. Ulbrich, R. Duncan, I.C. Hume, Polymers containing enzymatically degradable bonds. XII: Effect of spacer structure on the rate of daunomycin and adriamycin from poly[N-(2-hydroxypropyl)- methacrylamide] copolymer drug carriers in vitro and antitumour activity measured in vivo, Journal of controlled release 18 (1992) 123–132. V. Omelyanenko, P. Kopeckova, C. Gentry, J. Kopecek, Targetable HPMA copolymer-adriamycin conjugates. Recognition, internalization, and subcellular fate, Journal of Controlled Release 53 (1998) 25–37. F. Searle, S. Gac-Breton, R. Keane, S. Dimitrijevic, S. Brocchini, E.A. Sausville, R. Duncan, N-(2-Hydroxypropyl) methacrylamide Copolymer- 6-(3-Aminopropyl)ellipticine Conjugates. Synthesis, in Vitro, and Preliminary in Vivo Evaluation, Bioconjugate Chemistry 12 (2001) 711–718. L.W. Seymour, H. Soyez, A. De Marre, M.A. Shoaibi, E.H. Schacht, Polymeric prodrugs of mitomycin C designed for tumour tropism and sustained activation, Anti-cancer drug design 11 (1996) 351–365.

L. Bildstein et al. / Advanced Drug Delivery Reviews 63 (2011) 3–23 [146] H. Bundgaard, The double prodrug concept and its applications, Advanced Drug Delivery Reviews 3 (1989) 39–65. [147] P.L. Carl, P.K. Chakravarty, J.A. Katzenellenbogen, A novel connector linkage applicable in prodrug design, Journal of Medicinal Chemistry 24 (1981) 479–480. [148] R.B. Greenwald, H. Zhao, Poly (ethylene gylcol) prodrugs: altered pharmacokinetics and pharmacodynamics, in: V.J. Stella, R.T. Borchardt, M.J. Hageman, R. Oliyai, H. Maag, J. Tilley (Eds.), Prodrugs: Challenges and Rewards Part 2, Springer-AAPS Press, New York, 2007. [149] G.M. Dubowchik, K. Mosure, J.O. Knipe, R.A. Firestone, Cathepsin B-sensitive dipeptide prodrugs. 2. Models of anticancer drugs paclitaxel (Ttaxol (R)), mitomycin C and doxorubicin, Bioorganic & Medicinal Chemistry Letters 8 (1998) 3347–3352. [150] G.M. Dubowchik, S. Radia, H. Mastalerz, M.A. Walker, R.A. Firestone, H. Dalton King, S.J. Hofstead, D. Willner, S.J. Lasch, P.A. Trail, Doxorubicin immunoconjugates containing bivalent, lysosomally-cleavable dipeptide linkages, Bioorganic & Medicinal Chemistry Letters 12 (2002) 1529–1532. [151] R.B. Greenwald, A. Pendri, C.D. Conover, H. Zhao, Y.H. Choe, A. Martinez, K. Shum, S. Guan, Drug delivery systems employing 1, 4-or 1, 6-elimination: poly (ethylene glycol) prodrugs of amine-containing compounds, Journal of Medicinal Chemistry 42 (1999) 3657–3667. [152] V. de Groot, Targeting: Cancer—Small molecules, in: V.J. Stella, R.T. Borchardt, M.J. Hageman, R. Oliyai, H. Maag, J. Tilley (Eds.), Prodrugs: Challenges and rewards Part 2, Springer-AAPS Press, New York, 2007. [153] D. Shan, M.G. Nicolaou, R.T. Borchardt, B. Wang, Prodrug strategies based on intramolecular cyclization reactions, Journal of Pharmaceutical Sciences 86 (1997) 765–767. [154] B. Wang, S. Gangwar, G.M. Pauletti, T.J. Siahaan, R.T. Borchardt, Synthesis of a novel esterase-sensitive cyclic prodrug system for peptides that utilizes a “trimethyl

[155] [156]

[157]

[158] [159]

[160] [161]

[162] [163]

[164]

23

lock”-facilitated lactonization reaction, Journal of Organic Chemistry 62 (1997) 1363–1367. R.B. Greenwald, PEG drugs: an overview, Journal of Controlled Release 74 (2001) 159–171. R.B. Greenwald, Y.H. Choe, C.D. Conover, K. Shum, D. Wu, M. Royzen, Drug delivery systems based on trimethyl lock lactonization: poly (ethylene glycol) prodrugs of amino-containing compounds, Journal of Medicinal Chemistry 43 (2000) 475–487. E. Vivès, J. Schmidt, A. Pèlegrin, Cell Penetrating and Cell Targeting Peptides in Drug Delivery, Biochimica et Biophysica Acta - Reviews on Cancer 1786 (2008) 126–138. G. Kaur, I. Roy, Therapeutic applications of aptamers, Expert Opinion on Investigational Drugs 17 (2008) 43–60. E. Levy-Nissenbaum, A.F. Radovic-Moreno, A.Z. Wang, R. Langer, O.C. Farokhzad, Nanotechnology and aptamers: applications in drug delivery, Trends in Biotechnology 26 (2008) 442–449. G.P.H. Dietz, M. Bohr, Delivery of bioactive molecules into the cell: the Trojan horse approach, Molecular and Cellular Neuroscience 27 (2004) 85–131. A. Nori, K.D. Jensen, M. Tijerina, P. Kopeckova, J. Kopecek, Tat-conjugated synthetic macromolecules facilitate cytoplasmic drug delivery to human ovarian carcinoma cells, Bioconjugate Chemistry 14 (2003) 44–50. E.W. Meijer, M.H.P. Van Genderen, Dendrimers set to self-destruct, Nature 426 (2003) 128–129. M. Avital-Shmilovici, D. Shabat, Enzymatic activation of hydrophobic selfimmolative dendrimers: The effect of reporters with ionizable functional groups, Bioorganic & Medicinal Chemistry Letters 19 (2009) 3959–3962. D. Shabat, Self-immolative dendrimers as novel drug delivery platforms, Journal of Polymer Science Part A-Polymer Chemistry 44 (2006) 1569–1578.