Targeting Death-Inducing Receptors in Cancer Therapy

Home , CD137, CD27, Death receptor 4, Fas receptor, Osteoprotegerin, RANK

K Takeda1,2, J Stagg2, H Yagita1, K Okumura1 and MJ Smyth2

1Department of Immunology, Juntendo University School of Medicine, Tokyo, Japan and 2Cancer Immunology Program, Peter MacCallum Cancer Centre, East Melbourne, Victoria, Australia

Deregulated cell death pathways may lead to the stimuli, including growth factor withdrawal, UV light, development of cancer, and induction of tumor cell irradiation or chemicals, and the latter two have been apoptosis is the basis of many cancer therapies. Knowl- significantly utilized in cancer therapy (Nicholson, 2000; edge accumulated concerning the molecular mechanisms Johnstone et al., 2002). There are two major signalling of apoptotic cell death has aided the development of new pathways leading programmed cell death in mammalian therapeutic strategies to treat cancer. Signals through cells: the intrinsic pathway and the extrinsic pathway. death receptors of the tumor necrosis factor (TNF) The intrinsic pathway is initiated at the mitochondrial superfamily have been well elucidated, and death receptors level and plays a substantial role in chemotherapy- or are now one of the most attractive therapeutic targets irradiation-induced cell death. By contrast, the extrinsic in cancer. In particular, DR5 and DR4, death receptors death pathway is initiated through death receptor- of TNF-related apoptosis-inducing ligand (TRAIL mediated signals on the cell surface (Ashkenazi, 2002). or Apo2L), are interesting targets of antibody-based Cell death induced through the extrinsic pathway has therapy, since TRAIL may also bind decoy receptors that been studied extensively following signals mediated by may prevent TRAIL-mediated apoptosis, whereas TRAIL various members of tumor necrosis factor (TNF) ligand itself selectively induces apoptosis in cancer cells. receptor superfamily. The TNF superfamily is charac- Here, we review the potential therapeutic utility of terized by a sequence of two to five cysteine-rich agonistic antibodies against DR5 and DR4 and discuss extracellular repeats (Figure 1) (Ashkenazi and Dixit, the possible extension of this single-antibody-based 1998; Ashkenazi, 2002). The death receptors belonging strategy when combined with additional modalities that to the TNF superfamily share a homologous, intracel- either synergizes to cause enhanced apoptosis or further lular death domain of about 80 amino acids, which is engage the cellular immune response. Rational design essential for the transduction of apoptotic signals of antibody-based therapies combining the induction of (Itoh and Nagata, 1993; Tartaglia et al., 1993). tumor cell apoptosis and activation of tumor-specific Targeting death receptors has been thought a useful adaptive immunity enables promotion of distinct steps of approach, since extrinsic pathway alternatively or the antitumor immune response, thereby enhancing tumor- complementally induces apoptosis in cancer cells as well specific lymphocytes that can eradicate TRAIL/DR5- as the intrinsic pathway-mediated apoptosis in chemo- resistant mutating, large established and heterogeneous therapeutics and irradiation. CD95, so-called Fas or tumors in a manner that does not require the definition of Apo1, is the best-known death receptor belonging to the individual tumor-specific antigens. TNF receptor superfamily (Schmitz et al., 2000; Timmer Oncogene (2007) 26, 3745–3757. doi:10.1038/sj.onc.1210374 et al., 2002); however, before long it was appreciated that systemic treatment with Fas-targeting therapy Keywords: TNF superfamily; TRAIL; DR5; DR4; causes severe liver damage (Ogasawara et al., 1993; apoptosis; anti-tumor immune response Galle et al., 1995). The therapeutic potential of TNFa was also tempered by the demonstration that the systemic administration of recombinant TNFa resulted in a significant systemic toxicity associated with its strong pro-inflammatory activity, including fever, lung Targeting the TNF receptor superfamily in cancer or liver failure, increased blood clotting and hypoten- sion (Creaven et al., 1987; Creagen et al., 1988; Hersh Programmed cell death (apoptosis) is a controlled et al., 1991; Vassalli, 1992). Thus, the use of Fas or cellular mechanism, where the organism maintains TNFa receptors as cancer targets appeared fruitless, and cellular homeostasis in normal tissue compartments more recently efforts have been made to focus on and eliminates disordered cells (Steller, 1995; Thompson, minimizing unfavorable side effects by using local 1995; Hanahan and Weinberg, 2000; Danial and administration, inducible FasL- or TNF-expressing Korsmeyer, 2004). Apoptosis can be induced by various vectors, and tumor targeting gene delivery methods (Kircheis et al., 2002). Alternatively, the TNF-related Correspondence: Dr K Takeda, Department of Immunology, Juntendo apoptosis-inducing ligand (TRAIL)/TRAIL receptor University School of Medicine, Tokyo, Japan. pathway is regarded as most promising since apoptosis E-mail: [email protected] or [email protected] of tumor cells may be possible without life-threatening Anti-TRAIL-R mAb in cancer therapy K Takeda et al 3746 toxicity (Ashkenazi, 2002; Smyth et al., 2003; Yagita (Figure 1) (Ashkenazi, 2002). To date, four human et al., 2004). receptors specific for TRAIL have been identified: the death receptors, DR4 (DR4/TRAIL receptor 1/TRAIL- R1) and DR5 (TRAIL receptor 2/TRAIL-R2/KILL- ER), and the decoy receptors, DcR1/ TRAIL-R3/TRID TRAIL and its receptors and DcR2/TRAIL-R4. In mice, there is only one death- inducing receptor, a homolog of human DR5 (mDR5/ TRAIL was originally identified and cloned based on its mTRAIL-R2), and two decoy receptors (mDcTRAIL- sequence homology to the extracellular domain of CD95 R1/mDcR1 and mDcTRAIL-R2/mDcR2) have been ligand (CD95L, FasL) and TNF (Wiley et al., 1995; Pitti identified to bind mouse TRAIL (Schneider et al., 2003). et al., 1996). Like other TNF superfamily members, Ligation of TRAIL homotrimer to death receptors, TRAIL forms homotrimers, which cross link three DR4 or DR5, induces receptor trimerization leading to receptor molecules on the surface of target cells. the formation of a death-inducing signaling complex. Although it was reported that TRAIL binds to Trimerization of the death domain of the death receptor osteoprotegerin (OPG) (soluble inhibitor of receptor molecules results in recruitment of the adaptor molecule activator of NF-kB (RANK) ligand) at low affinity Fas-associated death domain (FADD), which in turn (Truneh et al., 2000), TRAIL predominantly interacts recruits and activates caspase-8 to execute apoptosis via with two types of receptors: death receptors triggering- the extrinsic pathway. In some cell types (type I), and decoy receptors inhibiting-TRAIL-induced apoptosis activation of caspase 8 is sufficient for subsequent

Figure 1 Members of the TNF and TNF receptor superfamily that induce apoptosis. Ligands are shown in their schematic transmembrane form. Arrows indicate receptor interactions with solid lines for strong-afﬁnity binding and dashed lines for low-afﬁnity binding. Dark gray boxes represent receptor cysteine-rich domains and black boxes are cytoplasmic receptor death domains. In humans, TRAIL induces apoptosis via DR5 or DR4. DcR1 with a GPI anchor and DcR2 with a truncated death domain do not trigger an apoptosis signal. TRAF binding motifs (dotted box) in the cytoplasmic region of human DR5 and DcR2 may be responsible for NF-kB activation by these receptors. In mice, TRAIL induces apoptosis via mDR5. mDcR1 with a GPI anchor and two alternative splicing forms of mDcR2 may act decoys.

Oncogene Anti-TRAIL-R mAb in cancer therapy K Takeda et al 3747 activation of the effector caspase 3 to execute apoptosis (mDcR2L). It has recently been reported that the (extrinsic pathway). However, in another cell type (type inhibitory mechanism by which DcR2 prevents DR5- II), amplification through the intrinsic pathway (the mediated apoptosis is not generally due to competitive mitochondrial pathway) is required (Ashkenazi, 2002; binding to TRAIL, but rather by DcR2 association with Kelley and Ashkenazi, 2004). The relative contribution DR5, via preligand assembly domain interactions, to of the extrinsic and intrinsic pathways to TRAIL- form a death–inhibitory complex (Clancy et al., 2005). induced cell death is variable, depending on the cellular Thus, several mechanisms of inhibition may allow decoy context (Figure 2). In contrast to DR4 and DR5, decoy receptors to prevent TRAIL-mediated apoptosis. receptors do not signal apoptosis. DcR1/TRAIL-R is a glycosyl phosphatidylinositol (GPI)-anchored membrane protein and DcR2/TRAIL-R4 contains a truncated, nonfunctional death domain. In mice, mDcR1 is Targeting DR4/DR5 in cancer therapies a GPI-anchored membrane protein, and mDcR2 can be expressed as two alternatively spliced variants, a TRAIL has been reported to preferentially induce secreted form (mDcR2S) and a transmembrane form apoptotic cell death in a wide variety of tumor cells and transformed cells, but not in most normal cells. Thus, recombinant TRAIL (rTRAIL) is an attractive anti cancer therapeutic. Preclinical studies in mice and non-human primates have shown that the administration of soluble forms of rTRAIL suppressed the growth of TRAIL-sensitive human tumor xenografts, with no apparent systemic toxicity (Ashkenazi et al., 1999; Walczak et al., 1999), supporting the potential utility of rTRAIL in vivo. Clearly, according to their decoy receptor expression, some tumor cells expressing DR5 or DR4 may still be protected from rTRAIL-induced apoptosis. Furthermore, monoclonal antibodies (mAbs) generally have longer half-life in vivo than recombinant proteins. Thus, specific targeting of death-inducing TRAIL receptors (DR4 and/or DR5) by agonistic mAbs may theoretically be a more effective approach than using the recombinant TRAIL ligand. Alterna- tively, although cancer cells express DR5 and DR4 more preferentially than normal cells (Daniels et al., 2005), anti-DR4 or anti-DR5 mAbs may be more toxic than rTRAIL, since the decoy receptors might protect DR4- and/or DR5-expressing normal cells from TRAIL- inducing killing (Sheridan et al., 1997; Degli-Esposti, 1999; Kim et al., 2000). To date, it has been reported that agonistic anti-human DR4 or DR5 mAbs exhibited potent tumoricidal activities against human tumor xenografts in nude or SCID mice without apparent toxicity (Griffith et al., 1999; Chuntharapai et al., 2001; Ichikawa et al., 2001; Motoki et al., 2005; Pukac et al., 2005). Either mAb possessing intrinsic agonistic activity or protoagonistic types of mAb requiring crosslinking of their Fc domains for agonistic activity show apoptosis- Figure 2 TRAIL-induced apoptosis signaling pathway and mediated tumoricidal activity in vivo. Interestingly, the resistant mechanisms. Trimerization of DR5 or DR4 by a TRAIL tumoricidal activity of anti-human DR4 mAb in mice trimer leads to recruitment of the adaptor FADD, which in turn recruits and activate caspase 8. In certain cells (type I cells), bearing human tumor xenografts was reported to be activation of caspase 8 is sufficient for activation of caspase 3, markedly influenced by their isotype, suggesting a which executes apoptosis (extrinsic pathway). In other cells (type II critical contribution of Fc receptors on some effector cells), amplification through the mitochondrial pathway is initiated functions in vivo (Chuntharapai et al., 2001). In contrast by cleavage of Bid by caspase 8 and translocation of the truncated to these observations, some human mAbs against Bid to mitochondria. This leads to Bax/Bak-mediated release of cytochrome-c (cyt-c), and subsequent caspase 9 activation by human DR4 and DR5 induced apoptosis in cultured Apaf-1, is then required for caspase 3-mediated apoptosis (intrinsic normal human hepatocytes in vitro (Mori et al., 2004), pathway). DcR1 and DcR2 can act as decoy receptors competing reminiscent of the different cytotoxic activity of with DR4 and DR5 for TRAIL. FLIP can prevent the recruitment recombinant TRAIL against normal cells due to altered of caspase 8. Bcl-2 and Bcl-xL can suppress the Bax/Bak-mediated release of cyt-c and Smac/DIABLO from mitochondria. Inhibitor forms (Jo et al., 2000; Lawrence et al., 2001; Qin et al., of apoptosis proteins (IAPs) can attenuate the activation of caspase 2001). Thus, the nature of mAbs (epitope, affinity, 9 and caspase 3, although Smac/DIABLO can counteract IAPs. isotype, etc.) and the status of normal cells may be

Oncogene Anti-TRAIL-R mAb in cancer therapy K Takeda et al 3748 critical as to whether apoptosis is inducible in normal tumor growth (Pollard, 2004); however, they may be cells (hepatocytes, endothelial cells, dendritic cells substantial effector cells that induce apoptosis following (DCs), etc.), and thus these parameters should be anti-DR5 mAb treatment. Thus, not only Fab-mediated further analysed to safely treat in a clinical setting. tumor cell death but also Fc-mediated activation of innate immune cells infiltrating the tumor site may be key initial steps in the antitumor activity of anti-DR5 mAb. This activating Fc-receptor-mediated immune Preclinical assessment of anti-DR5-mediated therapy activation has been also revealed an advantage of antibody-based targeting therapy in other tumor-target- We have generated an agonistic hamster anti-mouse ing mAb therapies (Clynes et al., 1998, 2000; Zhang DR5-specific mAb (MD5-1) and examined the anti- et al., 2004). tumor effects against syngeneic tumors in mice (Takeda Surprisingly, anti-DR5 mAb treatment was shown to et al., 2004). This mAb induced apoptosis in TRAIL- eradicate small tumor burdens in mice and indeed sensitive tumor cells when crosslinked in vitro. All concurrently generate tumor-specific cytolytic T lym- examined DR5-expressing TRAIL-resistant cell lines phocytes (CTLs). Subsequent tumor challenge revealed were resistant to anti-DR5 mAb-induced apoptosis. We that TRAIL-resistant tumor variants could be elimi- also observed, using a large variety of tumor target cells, nated in a perforin-dependent manner. Induction of that the sensitivity to anti-DR5 mAb-induced apoptosis tumor-specific CTL required primary DR5-mediated closely correlated with TRAIL-induced apoptosis. Thus, tumor cell death, as CTL never developed when FLIP- it would appear that mouse TRAIL decoy receptors overexpressing tumors were treated primarily. Hence, only make a minor contribution to protecting tumor cell following anti-DR5 mAb treatment, the recruitment of from TRAIL-mediated killing, a finding consistent with Fc receptor-expressing professional antigen-presenting the observation in human tumor cells (Griffith et al., cells (APCs) into the tumor site resulted in the efficient 1999; Zhang et al., 1999; LeBlanc and Ashkenazi, 2003). induction of tumor-specific T cells (Figure 3). Thus, Anti-mouse DR5 (MD5-1) mAb exerted potent anti-DR5 mAb may be more beneficial than rTRAIL antitumor effects against TRAIL-sensitive tumor cells for cancer therapy, as anti-DR5 mAb not only primarily in vivo. Consistent with the requirement for crosslinking eliminates most of the TRAIL-sensitive tumor cells at to induce apoptosis in vitro, Fc-receptor expressing the time of administration but also secondarily induces innate immune cells, including macrophages and NK tumor-specific effector and memory T cells that can cells, were required for the antitumor effect of this mAb. eradicate TRAIL-resistant variants and provide a long- The MD5-1-induced antitumor effect was demonstrated term protection from tumor recurrence. Undoubtedly, to depend mainly on direct apoptosis induction, rather the emergence of TRAIL-resistant variants is a critical than antibody-dependent cellular cytotoxicity (ADCC) problem for mTRAIL-based therapy (LeBlanc et al., or complement-dependent cytotoxicity (CDC), as DR5- 2002). Further experiments with other anti-DR5 mAb expressing and Fas-associated death domain (FADD)- that do not require crosslinking to induce apoptosis, like IL-1b-converting enzyme (FLICE) inhibitory other isotypes of anti-DR5 mAb or mouse rTRAIL are protein (FLIP)-transfected apoptosis-resistant tumor essential to definitely prove that engagement of innate cell variants were completely resistant to MD5-1 in vitro immunity is a critical feature of effective mAb-based and in vivo. Interestingly, apoptosis was similarly therapy in this context. induced by this anti-DR5 mAb after crosslinking with either negative or positive signaling Fc receptors in vitro; however, the antitumor effect of this mAb was significantly reduced in mice deficient for the activating Apoptosis in other mAb-based therapies that inhibit Fc-receptor (FcR common g-deficient mice lacking Fc g tumor growth RI and Fc g RIII) compared with activity in wild-type or inhibitory Fc-receptor-deficient mice (FcRIIB-deficient In the past decade, a number of tumor-targeting mice) in vivo. We observed recruitment of FcR-expres- antibodies have been approved for cancer treatment. sing innate immune cells, macrophages and DCs into Multiple mechanisms, including ADCC, CDC, opsoni- the early tumor site in wild-type mice, but not in sation and phagocytosis, have been reported to gen- mice deficient for the activating Fc-receptors. Thus, we erally contribute to the antitumor effects of various assume that crosslinking of anti-DR5 mAb by ubiqui- tumor-targeting antibodies (Gelderman et al., 2004). tously expressed inhibitory Fc receptors primarily Apoptosis can be induced by growth factor withdrawal, triggers apoptosis of cancer cells conjugated with Fab and thus antibodies to growth signal-mediating recep- portion of anti-DR5 mAb. At the same time, the tors on tumor cells might in part act by inducing binding of Fc portion of the mAb to an activating apoptosis in tumor cells. Epidermal growth factor Fc receptor concurrently facilitates the recruitment receptors (EGFRs) are integral components of principal additional Fc receptor-expressing cells to tumor site. signaling cascades involved in regulating solid tumor cell This amplifies the number of effector cells capable of proliferation and angiogenesis (de Bono and Rowinsky, crosslinking anti-DR5 mAb in tumor site and augment- 2002). Therefore, EGFR blockade is a rational thera- ing tumor cell apoptosis. Macrophages are thought to peutic approach to treat some malignancies (Salomon normally suppress antitumor immunity and facilitate et al., 1995). Anti-EGFR mAbs (Cetuximab, etc.) block

Oncogene Anti-TRAIL-R mAb in cancer therapy K Takeda et al 3749

Figure 4 Antitumor effect of MD5-1 versus anti-Her2 mAb. Tumor cells (5 Â 105 cells) derived from a spontaneously arising tumor in rat NeuT transgeneic BALB/c mouse were injected subcutaneously to 4-week-old tumor-free NeuT transgeneic BABL/ c mice. Groups of four mice were treated with three intraperitoneal injections of 100 mg of anti-DR5 mAb (clone MD5-1) (anti-DR5; closed diamond) at days 0, 4 and 8, or continuously with 100 mgof anti-rat Her2 mAb (clone 7.16.4) (anti-rHer2; closed triangle) or control Ig (clone UC8-1B9) (Ctr Ig; open square) every 4 days from day 0. Hybridoma cells producing 7.16.4 mAb were a generous gift from Dr Mark I Greene, University of Pennsylvania School of Medicine. Standard errors are shown. Figure 3 Mechanisms of antitumor effect induced by anti-DR5 mAb (MD5-1). Administration of anti-DR5 mAb (MD5-1) primarily induced apoptosis selectively in TRAIL-sensitive tumor cells by cross-linking on inhibitory Fc-g receptor (FcgR) and/or inhibition of cell growth (Hinoda et al., 2004). Thus, activating FcgR on recruited activating FcgR-expressing innate immune cells (macrophages, NK cells and DCs) (Pulendran et al., apoptosis of cancer cells is implicated as a common 2001) (a). The apoptotic tumor cells coated with anti-DR5 mAb are mechanism employed by a number of clinically useful incorporated efficiently by FcgR-mediated endocytosis into profes- antitumor antibodies. sional APCs as suggested in the effector mechanisms of other One of the most obvious and important questions is tumor-specific mAb (Selenko et al., 2001; zum Buschenfelde et al., 2002) (b). These APCs cross-present the incorporated tumor whether mAb reactive with cell death-inducing recep- antigens to T cells on MHC class I and class II, leading to the tors, which directly induce apoptosis, can prevent tumor development of tumor specific CTL (Albert et al., 1998; Dhodapkar growth more effectively than mAb targeting other types et al., 2002; Rafiq et al., 2002; Akiyama et al., 2003), which can lyse of molecules that both arrest cell proliferation and TRAIL-resistant variants in perforin-mediated manner in anti- subsequently cause apoptosis in tumor cells. Our DR5 mAb-treated tumor-bearing host (c). preliminary data suggest an improved antitumor efficacy of anti-DR5 mAb compared with an anti-rat HER2 ligand binding to the EGFR or dimerize and internalize mAb against rat HER2-expressing tumors derived from EGFR, which results in inhibition of EGF-stimulated rat HER2-transgenic mice (Figure 4). Although further EGFR tyrosine kinase activity. Apoptosis is reported to extensive comparative studies are required and should contribute to the antitumor effect of anti-EGFR mAbs, be validated in clinical settings, our observation suggests although the major mechanism is thought to be cell cycle that anti-DR5 mAb therapy may be comparable to arrest in the G1 phase (Bianco et al., 2005). Anti-CD20 anti-Her2 mAb therapy against TRAIL-sensitive Her2- mAb (Rituximab) has been clinically used to treat B-cell expressing tumor cells. chronic lymphocytic leukemia and non-Hodgkin’s lymphoma; although this mAb requires Fc receptor crosslinking and exhibits its antitumor effects through multiple mechanisms, it has also been reported to induce Combination therapies apoptosis (Cragg et al., 1999; Reed et al., 2002; Smith, 2003). Anti-HER2 mAb specific for the cellular proto- Enhancing apoptosis oncogene p185HER2/neu (Carter et al., 1992; Pegram Most of the DNA-damaging chemotherapeutic drugs et al., 1998) was the first mAb approved for use in solid and radiation induce tumor cell apoptosis by triggering tumors and it has already developed clear therapeutic p53-mediated activation of the intrinsic pathway. As benefit in the clinical treatment of breast cancer. TRAIL-induced apoptosis is mediated through both the Some anti-HER2 mAb can induce apoptosis as well as extrinsic and intrinsic pathways in a p53-independent

Oncogene Anti-TRAIL-R mAb in cancer therapy K Takeda et al 3750 manner (Johnstone et al., 2002), it is possible that 2004). Agonistic anti-DR5 mAb triggered tumor cell combination therapy of TRAIL with chemotherapeutic death directly via caspase activation, but also seconda- drugs or radiation may potentially deliver greater rily evoked tumor-specific CTL that could also eliminate apoptosis signals in cancer and results in excellent anti-DR5-resistant variants (Takeda et al., 2004). antitumor efficacy. In concert, many previous studies Despite a variety of pathways of cell death induction have shown that combinations of rTRAIL, anti-DR5 by mAbs, both primary tumor cell death and secondary mAb or anti-DR4 mAb with chemotherapeutic agents induction of tumor-specific T-cell immunity have been or irradiation additively or synergistically induced implicated in the successful antitumor effects of such tumor cell apoptosis in vitro and in vivo (Sheikh et al., mAbs. Thus, combining primary tumor cell death with 1998; Ashkenazi et al., 1999; Gliniak and Le, 1999; T-cell activation may be a very attractive approach to Chinnaiyan et al., 2000; Takimoto and El-Deiry, 2000; augment the therapeutic effect of the anti-DR5 mAb Chuntharapai et al., 2001; LeBlanc et al., 2002; Wajant and other mAb-based tumor targeting (Pardoll and et al., 2002; Buchsbaum et al., 2003; Ohtsuka et al., Allison, 2004). 2003; Voelkel-Johnson, 2003; Shankar and Srivastava, Based on this hypothesis, we have combined anti- 2004). Combinations are particularly attractive when DR5 mAb treatment with agonistic anti-CD40 mAb various cancers are naturally resistant to TRAIL, and agonistic anti-CD137 (4-1BB) mAb treatments, and chemotherapy or radiation. However, the treatments coined the combination ‘trimAb’. Both CD40 and for sensitizing tumor cells to TRAIL may also sensitize CD137 are members of the TNF receptor superfamily normal cells to TRAIL, underlining the need for that do not have a death domain, but alternatively they particular caution with combination therapies. contain homologous sequence motifs that mediate Proteasome inhibitors are suggested to affect TRAIL- interaction with specific adaptors from the TNFR- induced apoptosis at many different levels including associated factor (TRAF) family, thereby causing NF-kB-dependent and independent pathways (Sayers NF-kB and mitogen-activated protein kinase activation and Murphy, 2006). Indeed, inhibition of the protea- (Wallach et al., 1999; Croft, 2003a). CD40-stimulated some by bortezomib in combination with rTRAIL DC produce IL-12 that enhances NK cell and NKT cell enhanced apoptosis of a variety of human tumor cell activation, T helper type I responses and CTL induc- lines, as well as primary tumor cells, but not normal tion. The administration of agonistic anti-CD40 mAb cells, in vitro (Mitsiades et al., 2001; An et al., 2003; induces potent antitumor effects in vivo (Diehl et al., Ganten et al., 2005). Further experiments will hopefully 1999; French et al., 1999; Sotomayor et al., 1999). provide evidence that the combination of a proteasome Triggering CD137 signaling is also reported to elicit inhibitor and TRAIL or anti-DR5/DR4 mAb can robust antitumor immune responses in vivo (Melero suppress tumors in vivo without any unfavorable side et al., 1997; Watts, 2005). This effect is largely attributed effects. to CD137 signaling of tumor-specific T cells, enhancing Co-treatment of breast cancer cells with both inter- their proliferation and cytotoxic activity, and preventing feron-g and retinoic acid resulted in synergistic levels of their activation-induced cell death and immune tole- TRAIL expression that induced the death of hetero- rance (Pollok et al., 1993; Shuford et al., 1997; Takahashi geneous cancer cells in a paracrine fashion (Clarke et al., et al., 1999; Wilcox et al., 2002). Additionally, it was 2004). IFNs augment TRAIL sensitivity by modulating reported that CD137 signaling induces the activation of the death-inducing signal pathway (Varela et al., 2001; NK cells and DC (Melero et al., 1998; Futagawa et al., Fulda and Debatin, 2002; Chawla-Sarkar et al., 2003), 2002). With trimAb therapy, we proposed that a and not surprisingly the antitumor effect of rTRAIL, sequence of events may occur as follows: (1) anti-DR5 anti-DR5 mAb or anti-DR4 mAb was augmented by mAb induces apoptosis in tumor cells, recruits innate IFN (Merchant et al., 2004; Papageorgiou et al., 2004). immune cells into the tumor site and possibly augments Cytokines or bacterial components that stimulate IFN antigen-presenting function via activating Fc-receptors; production might also be favorable candidates in (2) anti-CD40 mAb augments APC function of antigen- combination with TRAIL-receptor targeting therapy. loaded DC and enables IL-12 production; (3) anti- Increased knowledge of the molecular components of CD137 mAb facilitates the induction, activation and the apoptotic pathway has encouraged the development survival of tumor-antigen specific CD8 þ T cells, and of more specific agents that can be combined with death therefore tumor-specific CTL would be effectively receptor targeting therapy (Fulda et al., 2002; Li et al., induced. Indeed, trimAb promptly induced tumor- 2004). specific T cells in the draining lymph node and resulted in the complete rejection of established TRAIL-sensitive tumors (Uno et al., 2006). Moreover, trimAb therapy Combination with other immunomodulators could even induce complete regression of tumor masses It has been reported that tumor-specific T-cell responses containing 90% apoptosis-resistant variants. Efforts to critically contributed to successful antibody-based tumor- replace anti-DR5 with TRAIL ligands in combination targeting therapy that causes cell death through with anti-CD40 and anti-CD137 mAbs were compar- inhibition of growth signals and Fc receptor-mediated ably unsuccessful (data not shown). Thus, due to the ADCC and/or CDC (Cragg et al., 1999; Clynes et al., ability of mAb to engage innate immunity and down- 2000; Selenko et al., 2001; Trcka et al., 2002; zum stream immune responses, mAb-based death receptor- Buschenfelde et al., 2002; Smith, 2003; Gelderman et al., targeting may have advantages in combinations with

Oncogene Anti-TRAIL-R mAb in cancer therapy K Takeda et al 3751 immunomodulators when compared with recombinant MCA-induced oncogenic transformation. Tumor hete- protein ligands. rogeneity that develops during carcinogenesis and Importantly, trimAb therapy also induced complete following natural immune selection (immunoediting) tumor rejection in a large proportion of mice bearing may cause many immunotherapeutics to fail (Khong established MCA-induced sarcomas, despite the known and Restifo, 2002; Dunn et al., 2004). Therefore, a heterogeneity of these tumors and their capacity to great advantage of the trimAb therapy is that tumor evade natural immunity (Basombrio, 1970; Khong and antigens do not have to be defined for therapeutic Restifo, 2002; Dunn et al., 2004). This trimAb-induced application. Moreover, the trimAb therapy approach rejection of MCA-initiated sarcomas was mediated by may concurrently induce a number of different CTL CD8 þ T cells, although it might be expected to be specificities against various tumor antigens expressed on primarily triggered by apoptosis of TRAIL-sensitive developing heterogeneous tumors, and thus may be variants that appear in the MCA-induced tumor mass suitable for coping with the genetic instability of tumors (Cretney et al., 2002; Takeda et al., 2002). The (Figure 5). cytotoxicity mediated by CTL from individual mice We have also recently found that trimAb combined that rejected tumors after therapy was variable and this with CD4 þ T-cell activation using an agonistic anti- might reflect the spectrum of tumor antigens expressed OX40 mAb may further augment the rejection by such heterogeneous tumor cells developing, following frequency and allow rejection of larger tumor burdens

Figure 5 Effector mechanisms of trimAb, a strategy combining apoptosis induction with immune activation. Anti-DR5 mAb induces apoptosis directly in tumor cells via crosslinking by FcgRs on innate immune cells including APCs. However, anti-DR5 mAb alone is not sufficient to induce strong tumor-specific immune responses capable of completely rejecting established tumors. Anti-CD40 mAb and anti-CD137 mAb, antibodies to co-stimulating TNF receptor superfamily members, cooperate to induce tumor-specific immune responses. These mAbs are less effective without a source of tumor antigen. In trimAb, both strategies complement one another. Anti- DR5 mAb induced apoptosis results in a rapid tumor antigen supply to FcgR- and CD40-activated APCs. These activated APCs promptly induce tumor-specific CTLs, and anti-CD137 mAb activates CTL responses and maintains CTL survival for long periods. The cooperation of these three mAbs results in rejection of a variety of established tumors by CTL utilizing perforin and IFN-g effector mechanisms. The predominant presentation of tumor antigens in the local draining lymph node may limit any potential autoimmunity and toxicity of this approach.

Oncogene Anti-TRAIL-R mAb in cancer therapy K Takeda et al 3752 (data not shown). OX40 is another member of TNF trials, and it was reported that several patients with receptor superfamily inducing CD4 þ T-cell activation refractory or heavily pretreated disease have experienced (Croft, 2003b), and has been reported to facilitate robust stable disease during anti-DR5 mAb (HGS-ETR2) CD8 þ and CD4 þ T-cell responses against tumors (Lee treatment (presentations at the AACR-NCI-EORTIC et al., 2004). Blockade of CTLA-4 mediated inhibitory International Conference. Molecular Targets and Can- signals to tumor-specific CD8 þ T cells and immune cer Therapeutics, 2005). Although these early trials are suppression by regulatory T cells is another interesting promising, it is important to recognize that the utility of approach to further augment mAb-based antitumor recombinant TRAIL and agonistic anti-DR5/DR4 Ab effects (Egen et al., 2002). We have also reported that therapies is limited to patients with TRAIL-sensitive IL-21 enhanced the induction of tumor-specific CTL by tumors. It is also obvious that TRAIL sensitivity will anti-DR5 mAb in mice (Smyth et al., 2006). Thus, be various among individual cancer patients, even if cytokines enhancing CTL expansion, effector function preferential TRAIL-sensitivity and DR5/DR4 expres- and/or survival are possibly favorable reagents to sion in certain cancer cell types were reported. The combine with antideath receptor therapy. efficacy of DR5/DR4-targeting therapies will be improved dramatically, when diagnostic methods deter- mining TRAIL sensitivity of clinically detectable human cancers are developed. Clinical application

Consistent with the selective sensitivity of cancer cells to TRAIL-mediated apoptosis, augmented expression of Possible side effects of TRAIL and anti-DR4/DR5 mAb DR5 and DR4 was reported in cancer cells, particularly in melanoma, lung cancer and gastrointestinal cancers It is clear that the toxicity of therapy versus the efﬁcacy (Daniels et al., 2005). Sensitivity to rTRAIL-induced of tumor treatment will be limiting in patients. As apoptosis was reported in cell lines derived from human mentioned above, apoptosis induction in normal human colon, lung, breast, kidney, brain and skin cancer cells (hepatocytes, keratinocytes, etc.) by some recombi- (Ashkenazi et al., 1999). Till date, effective apoptosis nant TRAIL, anti-DR5 mAb or anti-DR4 mAb was induction by rTRAIL has been reported in human reported in vitro (Jo et al., 2000; Lawrence et al., 2001; melanoma, leukemia, multiple myeloma, breast, blad- Qin et al., 2001; Mori et al., 2004). Thus, administration der, prostate, renal or colon cancer in single treatments of recombinant TRAIL or anti-DR5/DR4 mAb might and/or combination therapy with chemotherapeutic induce autoimmune type disorders due to direct agents (Chawla-Sarkar et al., 2002; An et al., 2003; cytotoxic activities toward normal cells. To date, no Buchsbaum et al., 2003; van Geelen et al., 2003; Voelkel- critical toxicities have been reported in any experimental Johnson, 2003; Merchant et al., 2004; Georgakis et al., setting, including hepatotoxicity (Hao et al., 2004). As 2005; Kaufmann and Steensma, 2005; Kurbanov an exception, hepatotoxicity with elevated serum aspa- et al., 2005; Pukac et al., 2005; Bucur et al., 2006; Marini tate aminotransferase (AST), alanine aminotransferase et al., 2006; O’Kane et al., 2006; Zeng et al., 2006). Thus, (ALT) and bilirubin was reported in a few subjects when these cancers are the most promising indications for treated with higher doses (20 mg/kg) of anti-DR5 mAb clinical trials of TRAIL and anti-DR5/DR4 mAb. (HGS-ETR2). Even if this unfavorable effect was Genentech Incorporated (South San Francisco, CA, observed only in a dose-limiting manner and using one USA) and Amgen (Thousand Oaks, CA, USA) are now anti-human DR5 mAb, the mechanisms underling undertaking phase I clinical trials with soluble recombi- hepatotoxicity should be revealed, since this will provide nant human TRAIL in cancer patients and the results novel information for safer treatment with anti-DR5 are greatly anticipated. Human Genome Sciences (HGS) mAb in humans in the future. in association with Cambridge Antibody Technologies Alternatively, immature human and mouse DCs are are entering phase II trials with anti-DR4 mAb (HGS- sensitive to TRAIL-mediated apoptosis in vitro (Leverkus ETR1) and further phase I trials with an anti-DR5 mAb et al., 2000; Hayakawa et al., 2004), and eliminated by (HGS-ETR2) in cancer patients are planned. In addi- TRAIL expressed on NK cells in vivo (Hayakawa et al., tion, a second anti-DR5 mAb (HGS-TR2J) developed 2004). Moreover, negative regulatory functions of DR5 by HGS in collaboration with the pharmaceutical on innate immune responses have been shown using division of the Kirin Brewery Company is in a phase I TRAIL-receptor deﬁcient mice (Diehl et al., 2004). clinical trial (Rowinsky, 2005). These reports suggested the possible immune suppres- Phase 1 trials have involved patients with advanced sive effect of rTRAIL or anti-DR4/DR5 mAb treat- solid malignancies (and/or non-Hodgkins lymphomas in ments, which might result in an increased frequency of the case of the anti-DR4 mAb (HGS-ETR1)), antibody infectious disease during therapy. To date, there is no administration has been intravenous and dosing report showing severe immune suppression by TRAIL repeated every 14–28 days. A phase 1 study of anti- or anti-DR5/DR4 mAb treatments in any experimental DR4 mAb (HGS-ETR1) in combination with gemcita- settings and clinical trials. However, it may be even bine and cisplatin has also been undertaken. To date, no more important to monitor immune status during major toxicities have been observed at least in response TRAIL or anti-DR5/DR4 mAb therapy, when com- to the doses administered during these initial clinical bined with chemotherapies and/or radiotherapy.

Oncogene Anti-TRAIL-R mAb in cancer therapy K Takeda et al 3753 Combination of antibody therapy in targeting dif- pathway would not be an effective target in cancer ferent steps that collectively induce antitumor immunity therapy. However, quite the contrary, preclinical data is one of most rational strategies to eradicate established suggest the TRAIL-R might be a good target on tumor tumor masses. However, we clearly caution the use of cells and many clinical trials using TRAIL receptor- such approaches given the immense power of immune speciﬁc mAbs are now in progress. Even if monotherapy system and attention must direct at assessing unfavor- with these antibodies is largely ineffective, it may still be able side effects, including autoimmune reactions. possible to obtain signiﬁcant therapeutic effects when Moreover, it has been noted that too much activation combining these TRAIL-R reactive mAbs with other of immune responses occasionally results in various therapies (radiotherapy, chemotherapy and/or other unfavorable side effects including hepatotoxicity (e.g. mAb-based tumor targeting therapy). In particular, concanavalin A-induced hepatitis). Moreover, it has combination with immune activation therapy seems been reported that hepatocytes had augmented TRAIL both logical and promising, given that antibodies sensitivity during viral infections, alcohol intake and naturally activate immune responses via Fc receptors. cholestasis (Higuchi et al., 2002; Janssen et al., 2003; However, careful analysis of unfavorable side effects due Mundt et al., 2003; Mundt et al., 2005). Thus, disorders to apoptosis induction in normal cells and over of hepatic function are the most expected side effects activation of immune responses will also be required in during DR5/DR4-targeting mono or combination parallel to make sure these approaches are safe and therapy. Regardless, targeting TRAIL receptors will effective. be relatively safer and more useful compared with cancer therapies that target other members of the death- Abbreviations inducing TNF receptor superfamily. ADCC, antibody-dependent cellular cytotoxicity; APCs, antigen-presenting cells; CDC, complement-dependent cytotoxicity; DCs, dendritic cells; EGFRs, epidermal growth factor Conclusion and perspective receptors; mAbs, monoclonal antibodies; rTRAIL, recombinant TRAIL; TRAIL, TNF-related apoptosis-inducing ligand. Evidence obtained in experimental mouse models clearly suggests that TRAIL is one of important immune Acknowledgements effector molecules in the surveillance and elimination of developing tumors (Cretney et al., 2002; Takeda et al., This work is supported by the Ministry of Education, Science and Culture, Japan. JS is supported by a fellowship from the 2002). Besides, preferential resistance of tumor cells Canadian Institutes of Health Research (CIHR). MJS is escaping immunity, termed ‘immunoediting’, might supported by a National Health and Medical Research explain the failure of many immunotherapeutic Council of Australia Program Grant and Research Fellowship. approaches to cancer treatment (Dunn et al., 2004). We thank the Susan G Komen Breast Cancer Foundation for One might therefore expect that the TRAIL/TRAIL-R their funding support.

References

Akiyama K, Ebihara S, Yada A, Matsumura K, Aiba S, Bianco R, Daniele G, Ciardiello F, Tortora G. (2005). Nukiwa T et al. (2003). Dendritic cells acquire antigen Monoclonal antibodies targeting the epidermal growth from apoptotic cells and induce class I-restricted CTLs. factor receptor. Curr Drug Targets 6: 275–287. J Immunol 170: 1641–1648. Buchsbaum DJ, Zhou T, Grizzle WE, Oliver PG, Hammond CJ, Albert ML, Sauter B, Bhardwaj N. (1998). Dendritic cells Zhang S et al. (2003). Antitumor efﬁcacy of TRA-8 anti- acquire antigen from apoptotic cells and induce class DR5 monoclonal antibody alone or in combination with I-restricted CTLs. Nature 392: 86–89. chemotherapy and/or radiation therapy in a human breast An J, Sun YP, Adams J, Fisher M, Belldegrun A, Rettig MB. cancer model. Clin Cancer Res 9: 3731–3741. (2003). Drug interactions between the proteasome inhibitor Bucur O, Ray S, Bucur MC, Almasan A. (2006). APO2 bortezomib and cytotoxic chemotherapy, tumor necrosis ligand/tumor necrosis factor-related apoptosis-inducing factor (TNF) a, and TNF-related apoptosis-inducing ligand ligand in prostate cancer therapy. Front Biosci 11: in prostate cancer. Clin Cancer Res 9: 4537–4545. 1549–1568. Ashkenazi A. (2002). Targeting death and decoy receptors of Carter P, Presta L, Gorman CM, Ridgway JB, Henner D, the tumour-necrosis factor superfamily. Nat Rev Cancer 2: Wong WL et al. (1992). Humanization of an anti- 420–430. p185HER2 antibody for human cancer therapy. Proc Natl Ashkenazi A, Dixit VM. (1998). Death receptors: signaling Acad Sci USA 89: 4285–4289. and modulation. Science 281: 1305–1308. Chawla-Sarkar M, Leaman DW, Jacobs BS, Borden EC. Ashkenazi A, Pai RC, Fong S, Leung S, Lawrence DA, (2002). Apoptosis and interferons: role of interferon- Marsters SA et al. (1999). Safety and antitumor activity stimulated genes as mediators of apoptosis. J Immunol of recombinant soluble Apo2 ligand. J Clin Invest 104: 169: 847–855. 155–162. Chawla-Sarkar M, Lindner DJ, Liu YF, Williams BR, Basombrio MA. (1970). Search for common antigenicities Sen GC, Silverman RH et al. (2003). IFN-b pretreatment among twenty-ﬁve sarcomas induced by methylcholan- sensitizes human melanoma cells to TRAIL/Apo2 ligand- threne. Cancer Res 30: 2458–2462. induced apoptosis. Apoptosis 8: 237–249.

Oncogene Anti-TRAIL-R mAb in cancer therapy K Takeda et al 3754 Chinnaiyan AM, Prasad U, Shankar S, Hamstra DA, T-lymphocyte tolerance and augments anti-tumor vaccine Shanaiah M, Chenevert TL et al. (2000). Combined effect efficacy. Nat Med 5: 774–779. of tumor necrosis factor-related apoptosis-inducing ligand Dunn GP, Old LJ, Schreiber RD. (2004). The immunobiology and ionizing radiation in breast cancer therapy. Proc Natl of cancer immunosurveillance and immunoediting. Immu- Acad Sci USA 97: 1754–1759. nity 21: 137–148. Chuntharapai A, Dodge K, Grimmer K, Schroeder K, Egen JG, Kuhns MS, Allison JP. (2002). CTLA-4: new Marsters SA, Koeppen H et al. (2001). Isotype-dependent insights into its biological function and use in tumor inhibition of tumor growth in vivo by monoclonal antibodies immunotherapy. Nat Immunol 3: 611–618. to death receptor 4. J Immunol 166: 4891–4898. French RR, Chan HT, Tutt AL, Glennie MJ. (1999). CD40 Clancy L, Mruk K, Archer K, Woelfel M, Mongkolsapaya J, antibody evokes a cytotoxic T-cell response that eradicates Screaton G et al. (2005). Preligand assembly domain- lymphoma and bypasses T-cell help. Nat Med 5: 548–553. mediated ligand-independent association between TRAIL Fulda S, Debatin KM. (2002). IFNg sensitizes for apoptosis by receptor 4 (TR4) and TR2 regulates TRAIL-induced upregulating caspase-8 expression through the stat1 path- apoptosis. Proc Natl Acad Sci USA 102: 18099–18104. way. Oncogene 21: 2295–2308. Clarke N, Jimenez-Lara AM, Voltz E, Gronemeyer H. (2004). Fulda S, Wick W, Weller M, Debatin KM. (2002). Smac Tumor suppressor IRF-1 mediated retinoid and interferon agonists sensitize for Apo2L/TRAIL- or anticancer drug- anticancer signaling to death ligand TRAIL. EMBO J 23: induced apoptosis and induce regression of malignant 3051–3060. glioma in vivo. Nat Med 8: 808–815. Clynes R, Takechi Y, Moroi Y, Houghton A, Ravetch JV. Futagawa T, Akiba H, Kodama T, Takeda K, Hosoda Y, (1998). Fc receptors are required in passive and Yagita H et al. (2002). Expression and function of 4-1BB active immunity to melanoma. Proc Natl Acad Sci USA and 4-1BB ligand on murine dendritic cells. Int Immunol 14: 95: 652–656. 275–286. Clynes RA, Towers TL, Presta LG, Ravetch JV. (2000). Galle PR, Hofmann WJ, Walczak H, Schaller H, Otto G, Inhibitory Fc receptors modulate in vivo cytoxicity against Stremmel W et al. (1995). Involvement of the CD95 (APO-1/ tumor targets. Nat Med 6: 443–446. Fas) receptor and ligand in liver damage. J Exp Med 182: Cragg MS, French RR, Glennie MJ. (1999). Signaling 1223–1230. antibodies in cancer therapy. Curr Opin Immunol 11: Ganten TM, Koschny R, Haas TL, Sykora J, Li-Weber M, 541–547. Herzer K et al. (2005). Proteasome inhibition sensitizes Creagan ET, Kovach JS, Moertel CG, Frytak S, Kvols LK. hepatocellular carcinoma cells, but not human hepatocytes, (1988). A phase I clinical trial of recombinant human tumor to TRAIL. Hepatology 42: 588–597. necrosis factor. Cancer 62: 2467–2471. Gelderman KA, Tomlinson S, Ross GD, Gorter A. (2004). Creaven PJ, Plager JE, Dupere S, Huben RP, Takita H, Complement function in mAb-mediated cancer immu- Mittelman A et al. (1987). Phase I clinical trial of recom- notherapy. Trends Immunol 25: 158–164. binant human tumor necrosis factor. Cancer Chemother Georgakis GV, Li Y, Humphreys R, Andreeff M, O’Brien S, Pharmacol 20: 137–144. Younes M et al. (2005). Activity of selective fully human Cretney E, Takeda K, Yagita H, Glaccum M, Peschon JJ, agonistic antibodies to the TRAIL death receptors Smyth MJ. (2002). Increased susceptibility to tumor TRAIL-R1 and TRAIL-R2 in primary and cultured initiation and metastasis in TNF-related apoptosis-inducing lymphoma cells: induction of apoptosis and enhancement ligand-deficient mice. J Immunol 168: 1356–1361. of doxorubicin- and bortezomib-induced cell death. Br J Croft M. (2003a). Co-stimulatory members of the TNFR Haematol 130: 501–510. family: keys to effective T-cell immunity? Nat Rev Immunol Gliniak B, Le T. (1999). Tumor necrosis factor-related 3: 609–620. apoptosis-inducing ligand’s antitumor activity in vivo is Croft M. (2003b). Costimulation of T cells by OX40, 4-1BB, enhanced by the chemotherapeutic agent CPT-11. Cancer and CD27. Cytokine Growth Factor Rev 14: 265–273. Res 59: 6153–6158. Danial NN, Korsmeyer SJ. (2004). Cell death: critical control Griffith TS, Rauch CT, Smolak PJ, Waugh JY, Boiani N, points. Cell 116: 205–219. Lynch DH et al. (1999). Functional analysis of TRAIL Daniels RA, Turley H, Kimberley FC, Liu XS, Mongkolsapaya J, receptors using monoclonal antibodies. J Immunol 162: Ch’En P et al. (2005). Expression of TRAIL and TRAIL 2597–2605. receptors in normal and malignant tissues. Cell Res 15: Hanahan D, Weinberg RA. (2000). The hallmarks of cancer. 430–438. Cell 100: 57–70. de Bono JS, Rowinsky EK. (2002). Therapeutics targeting Hao C, Song JH, Hsi B, Lewis J, Song DK, Petruk KC et al. signal transduction for patients with colorectal carcinoma. (2004). TRAIL inhibits tumor growth but is nontoxic Br Med Bull 64: 227–254. to human hepatocytes in chimeric mice. Cancer Res 64: Degli-Esposti M. (1999). To die or not to die –the quest of the 8502–8506. TRAIL receptors. J Leukoc Biol 65: 535–542. Hayakawa Y, Screpanti V, Yagita H, Grandien A, Ljunggren Dhodapkar KM, Krasovsky J, Williamson B, Dhodapkar MV. HG, Smyth MJ et al. (2004). NK cell TRAIL eliminates (2002). Antitumor monoclonal antibodies enhance cross- immature dendritic cells in vivo and limits dendritic cell presentation of cellular antigens and the generation of vaccination efficacy. J Immunol 172: 123–129. myeloma-specific killer T cells by dendritic cells. J Exp Med Hersh EM, Metch BS, Muggia FM, Brown TD, Whitehead RP, 195: 125–133. Budd GT et al. (1991). Phase II studies of recombinant Diehl GE, Yue HH, Hsieh K, Kuang AA, Ho M, Morici LA human tumor necrosis factor a in patients with malignant et al. (2004). TRAIL-R as a negative regulator of innate disease: a summary of the southwest oncology group immune cell responses. Immunity 21: 877–889. experience. J Immunother 10: 426–431. Diehl L, den Boer AT, Schoenberger SP, van der Voort EI, Higuchi H, Bronk SF, Taniai M, Canbay A, Gores GJ. (2002). Schumacher TN, Melief CJ et al. (1999). CD40 activation Cholestasis increases tumor necrosis factor-related apoptosis- in vivo overcomes peptide-induced peripheral cytotoxic inducing ligand (TRAIL)-R2/DR5 expression and sensitizes

Oncogene Anti-TRAIL-R mAb in cancer therapy K Takeda et al 3755 the liver to TRAIL-mediated cytotoxicity. J Pharmacol Exp Marini P, Denzinger S, Schiller D, Kauder S, Welz S, Ther 303: 461–467. Humphreys R et al. (2006). Combined treatment of colorec- Hinoda Y, Sasaki S, Ishida T, Imai K. (2004). Monoclonal tal tumours with agonistic TRAIL receptor antibodies antibodies as effective therapeutic agents for solid tumors. HGS-ETR1 and HGS-ETR2 and radiotherapy: enhanced Cancer Sci 95: 621–625. effects in vitro and dose-dependent growth delay in vivo. Ichikawa K, Liu W, Zhao L, Wang Z, Liu D, Ohtsuka T et al. Oncogene 25: 5145–5154. (2001). Tumoricidal activity of a novel anti-human DR5 Melero I, Johnston JV, Shufford WW, Mittler RS, Chen L. monoclonal antibody without hepatocyte cytotoxicity. Nat (1998). NK1.1 cells express 4-1BB (CDw137) costimulatory Med 7: 954–960. molecule and are required for tumor immunity elicited Itoh N, Nagata S. (1993). A novel protein domain required for by anti-4-1BB monoclonal antibodies. Cell Immunol 190: apoptosis. Mutational analysis of human Fas antigen. J Biol 167–172. Chem 268: 10932–10937. Melero I, Shuford WW, Newby SA, Aruffo A, Ledbetter JA, Janssen HL, Higuchi H, Abdulkarim A, Gores GJ. (2003). Hellstrom KE et al. (1997). Monoclonal antibodies against Hepatitis B virus enhances tumor necrosis factor-related the 4-1BB T-cell activation molecule eradicate established apoptosis-inducing ligand (TRAIL) cytotoxicity by increas- tumors. Nat Med 3: 682–685. ing TRAIL-R1/death receptor 4 expression. J Hepatol 39: Merchant MS, Yang X, Melchionda F, Romero M, Klein R, 414–420. Thiele CJ et al. (2004). Interferon g enhances the effective- Jo M, Kim TH, Seol DW, Esplen JE, Dorko K, Billiar TR ness of tumor necrosis factor-related apoptosis-inducing et al. (2000). Apoptosis induced in normal human hepato- ligand receptor agonists in a xenograft model of Ewing’s cytes by tumor necrosis factor-related apoptosis-inducing sarcoma. Cancer Res 64: 8349–8356. ligand. Nat Med 6: 564–567. Mitsiades CS, Treon SP, Mitsiades N, Shima Y, Richardson P, Johnstone RW, Ruefli AA, Lowe SW. (2002). Apoptosis: a Schlossman R et al. (2001). TRAIL/Apo2L ligand link between cancer genetics and chemotherapy. Cell 108: selectively induces apoptosis and overcomes drug resistance 153–164. in multiple myeloma: therapeutic applications. Blood 98: Kaufmann SH, Steensma DP. (2005). On the TRAIL of a new 795–804. therapy for leukemia. Leukemia 19: 2195–2202. Mori E, Thomas M, Motoki K, Nakazawa K, Tahara T, Kelley SK, Ashkenazi A. (2004). Targeting death receptors Tomizuka K et al. (2004). Human normal hepatocytes are in cancer with Apo2L/TRAIL. Curr Opin Pharmacol 4: susceptible to apoptosis signal mediated by both TRAIL-R1 333–339. and TRAIL-R2. Cell Death Differ 11: 203–207. Khong HT, Restifo NP. (2002). Natural selection of tumor Motoki K, Mori E, Matsumoto A, Thomas M, Tomura T, variants in the generation of ‘tumor escape’ phenotypes. Nat Humphreys R et al. (2005). Human normal hepatocytes are Immunol 3: 999–1005. susceptible to apoptosis signal mediated by both TRAIL-R1 Kim K, Fisher MJ, Xu SQ, el-Deiry WS. (2000). Molecular and TRAIL-R2. Clin Cancer Res 11: 3126–3135. determinants of response to TRAIL in killing of normal and Mundt B, Kuhnel F, Zender L, Paul Y, Tillmann H, cancer cells. Clin Cancer Res 6: 335–346. Trautwein C et al. (2003). Involvement of TRAIL and its Kircheis R, Wightman L, Kursa M, Ostermann E, Wagner E. receptors in viral hepatitis. FASEB J 17: 94–96. (2002). Tumor-targeted gene delivery: an attractive strategy Mundt B, Wirth T, Zender L, Waltemathe M, Trautwein C, to use highly active effector molecules in cancer treatment. Manns MP et al. (2005). Tumour necrosis factor related Gene Therapy 9: 731–735. apoptosis inducing ligand (TRAIL) induces hepatic Kurbanov BM, Geilen CC, Fecker LF, Orfanos CE, Eberle J. steatosis in viral hepatitis and after alcohol intake. Gut 54: (2005). Efficient TRAIL-R1/DR4-mediated apoptosis in 1590–1596. melanoma cells by tumor necrosis factor-related apoptosis- Nicholson DW. (2000). From bench to clinic with apoptosis- inducing ligand (TRAIL). J Invest Dermatol 125: 1010–1019. based therapeutic agents. Nature 407: 810–816. Lawrence D, Shahrokh Z, Marsters S, Achilles K, Shih D, O’Kane HF, Watson CJ, Johnston SR, Petak I, Watson RW, Mounho B et al. (2001). Differential hepatocyte toxicity of Williamson KE. (2006). Targeting death receptors in recombinant Apo2L/TRAIL versions. Nat Med 7: 383–385. bladder, prostate and renal cancer. J Urol 175: 432–438. LeBlanc H, Lawrence D, Varfolomeev E, Totpal K, Morlan J, Ogasawara J, Watanabe-Fukunaga R, Adachi M, Matsuzawa A, Schow P et al. (2002). Tumor-cell resistance to death Kasugai T, Kitamura Y et al. (1993). Lethal effect of the receptor-induced apoptosis through mutational inactivation anti-Fas antibody in mice. Nature 364: 806–809. of the proapoptotic Bcl-2 homolog bax. Nat Med 8: Ohtsuka T, Buchsbaum D, Oliver P, Makhija S, Kimberly R, 274–281. Zhou T. (2003). Synergistic induction of tumor cell LeBlanc HN, Ashkenazi A. (2003). Apo2L/TRAIL and its apoptosis by death receptor antibody and chemotherapy death and decoy receptors. Cell Death Differ 10: 66–75. agent through JNK/p38 and mitochondrial death pathway. Lee SJ, Myers L, Muralimohan G, Dai J, Qiao Y, Li Z et al. Oncogene 22: 2034–2044. (2004). 4-1BB and OX40 dual costimulation synergistically Papageorgiou A, Lashinger L, Millikan R, Grossman HB, stimulate primary specific CD8 T cells for robust effector Benedict W, Dinney CP et al. (2004). Role of tumor necrosis function. J Immunol 173: 3002–3012. factor-related apoptosis-inducing ligand in interferon- Leverkus M, Walczak H, McLellan A, Fries HW, Terbeck G, induced apoptosis in human bladder cancer cells. Cancer Brocker EB et al. (2000). Maturation of dendritic cells leads Res 64: 8973–8979. to up-regulation of cellular FLICE-inhibitory protein and Pardoll D, Allison J. (2004). Cancer immunotherapy: breaking concomitant down-regulation of death ligand-mediated the barriers to harvest the crop. Nat Med 10: 887–892. apoptosis. Blood 96: 2628–2631. Pegram MD, Lipton A, Hayes DF, Weber BL, Baselga JM, Li L, Thomas RM, Suzuki H, De Brabander JK, Wang X, Tripathy D et al. (1998). Phase II study of receptor- Harran PG. (2004). A small molecule Smac mimic enhanced chemosensitivity using recombinant humanized potentiates TRAIL- and TNFa-mediated cell death. Science anti-p185HER2/neu monoclonal antibody plus cisplatin in 305: 1471–1474. patients with HER2/neu-overexpressing metastatic breast

Oncogene Anti-TRAIL-R mAb in cancer therapy K Takeda et al 3756 cancer refractory to chemotherapy treatment. J Clin Oncol apoptosis by a family of signaling and decoy receptors. 16: 2659–2671. Science 277: 818–821. Pitti RM, Marsters SA, Ruppert S, Donahue CJ, Moore A, Shuford WW, Klussman K, Tritchler DD, Loo DT, Chalupny J, Ashkenazi A. (1996). Induction of apoptosis by Apo-2 Siadak AW et al. (1997). 4-1BB costimulatory signals ligand, a new member of the tumor necrosis factor cytokine preferentially induce CD8+ T cell proliferation and lead to family. J Biol Chem 271: 12687–12690. the amplification in vivo of cytotoxic T cell responses. J Exp Pollard JW. (2004). Tumour-educated macrophages promote Med 186: 47–55. tumour progression and metastasis. Nat Rev Cancer 4: Smith MR. (2003). Rituximab (monoclonal anti-CD20 anti- 71–78. body): mechanisms of action and resistance. Oncogene 22: Pollok KE, Kim YJ, Zhou Z, Hurtado J, Kim KK, Pickard RT 7359–7368. et al. (1993). Inducible T cell antigen 4-1B. Analysis of Smyth MJ, Hayakawa Y, Cretney E, Zerafa N, Sivakumar P, expression and function. J Immunol 150: 771–781. Yagita H et al. (2006). IL-21 enhances tumor-specific CTL Pukac L, Kanakaraj P, Humphreys R, Alderson R, Bloom M, induction by anti-DR5 antibody therapy. J Immunol 176: Sung C et al. (2005). HGS-ETR1, a fully human TRAIL- 6347–6355. receptor 1 monoclonal antibody, induces cell death in Smyth MJ, Takeda K, Hayakawa Y, Peschon JJ, van den multiple tumour types in vitro and in vivo. Br J Cancer 92: Brink MR, Yagita H. (2003). Nature’s TRAIL –on a path 1430–1441. to cancer immunotherapy. Immunity 18: 1–6. Pulendran B, Banchereau J, Maraskovsky E, Maliszewski C. Sotomayor EM, Borrello I, Tubb E, Rattis FM, Bien H, Lu Z (2001). Modulating the immune response with dendritic cells et al. (1999). Conversion of tumor-specific CD4+ T-cell and their growth factors. Trends Immunol 22: 41–47. tolerance to T-cell priming through in vivo ligation of CD40. Qin J, Chaturvedi V, Bonish B, Nickoloff BJ. (2001). Avoiding Nat Med 5: 780–787. premature apoptosis of normal epidermal cells. Nat Med 7: Steller H. (1995). Mechanisms and genes of cellular suicide. 385–386. Science 267: 1445–1449. Rafiq K, Bergtold A, Clynes R. (2002). Immune complex- Takahashi C, Mittler RS, Vella AT. (1999). 4-1BB is a bona mediated antigen presentation induces tumor immunity. fide CD8 T cell survival signal. J Immunol 162: 5037–5040. J Clin Invest 110: 71–79. Takeda K, Smyth MJ, Cretney E, Hayakawa Y, Kayagaki N, Reed JC, Kitada S, Kim Y, Byrd J. (2002). Modulating Yagita H et al. (2002). Critical role for tumor necrosis apoptosis pathways in low-grade B-cell malignancies using factor-related apoptosis-inducing ligand in immune surveil- biological response modifiers. Semin Oncol 29: 10–24. lance against tumor development. J Exp Med 195: 161–169. Rowinsky EK. (2005). Targeted induction of apoptosis in Takeda K, Yamaguchi N, Akiba H, Kojima Y, Hayakawa Y, cancer management: the emerging role of tumor necrosis Tanner JE et al. (2004). Induction of tumor-specific T cell factor-related apoptosis-inducing ligand receptor activating immunity by anti-DR5 antibody therapy. J Exp Med 199: agents. J Clin Oncol 23: 9394–9407. 437–448. Salomon DS, Brandt R, Ciardiello F, Normanno N. (1995). Takimoto R, El-Deiry WS. (2000). Wild-type p53 transacti- Epidermal growth factor-related peptides and their recep- vates the KILLER/DR5 gene through an intronic sequence- tors in human malignancies. Crit Rev Oncol Hematol 19: specific DNA-binding site. Oncogene 19: 1735–1743. 183–232. Tartaglia LA, Ayres TM, Wong GH, Goeddel DV. (1993). A Sayers TJ, Murphy WJ. (2006). Combining proteasome novel domain within the 55 kd TNF receptor signals cell inhibition with TNF-related apoptosis-inducing ligand death. Cell 74: 845–853. (Apo2L/TRAIL) for cancer therapy. Cancer Immunol Thompson CB. (1995). Apoptosis in the pathogenesis and Immunother 55: 76–84. treatment of disease. Science 267: 1456–1462. Schmitz I, Kirchhoff S, Krammer PH. (2000). Regulation of Timmer T, de Vries EG, de Jong S. (2002). Fas receptor- death receptor-mediated apoptosis pathways. Int J Biochem mediated apoptosis: a clinical application? J Pathol 196: Cell Biol 32: 1123–1136. 125–134. Schneider P, Olson D, Tardivel A, Browning B, Lugovskoy A, Trcka J, Moroi Y, Clynes RA, Goldberg SM, Bergtold A, Gong D et al. (2003). Identification of a new murine Perales MA et al. (2002). Redundant and alternative roles tumor necrosis factor receptor locus that contains two for activating fc receptors and complement in an antibody- novel murine receptors for tumor necrosis factor- dependent model of autoimmune vitiligo. Immunity 16: related apoptosis-inducing ligand (TRAIL). J Biol Chem 861–868. 278: 5444–5454. Truneh A, Sharma S, Silverman C, Khandekar S, Reddy MP, Selenko N, Maidic O, Draxier S, Berer A, Jager U, Knapp W Deen KC et al. (2000). Temperature-sensitive differential et al. (2001). CD20 antibody (C2B8)-induced apoptosis of affinity of TRAIL for its receptors. DR5 is the highest lymphoma cells promotes phagocytosis by dendritic cells affinity receptor. J Biol Chem 275: 23319–23325. and cross-priming of CD8+ cytotoxic T cells. Leukemia 15: Uno T, Takeda K, Kojima Y, Yoshizawa H, Akiba H, 1619–1626. Mittler RS et al. (2006). Eradication of established tumors in Shankar S, Srivastava RK. (2004). Enhancement of therapeu- mice by a combination antibody-based therapy. Nat Med tic potential of TRAIL by cancer chemotherapy and 12: 693–698. irradiation: mechanisms and clinical implications. Drug van Geelen CM, de Vries EG, Le TK, van Weeghel RP, Resist Updat 7: 139–156. de Jong S. (2003). Differential modulation of the TRAIL Sheikh MS, Burns TF, Huang Y, Wu GS, Amundson S, receptors and the CD95 receptor in colon carcinoma cell Brooks KS et al. (1998). p53-dependent and -independent lines. Br J Cancer 89: 363–373. regulation of the death receptor KILLER/DR5 gene Varela N, Munoz-Pinedo C, Ruiz-Ruiz C, Robledo G, expression in response to genotoxic stress and tumor Pedroso M, Lopez-Rivas A. (2001). Interferon-g sensitizes necrosis factor a. Cancer Res 58: 1593–1598. human myeloid leukemia cells to death receptor-mediated Sheridan JP, Marsters SA, Pitti RM, Gurney A, Skubatch M, apoptosis by a pleiotropic mechanism. J Biol Chem 276: Baldwin D et al. (1997). Control of TRAIL-induced 17779–17787.

Oncogene Anti-TRAIL-R mAb in cancer therapy K Takeda et al 3757 Vassalli P. (1992). The pathophysiology of tumor necrosis of a new member of the TNF family that induces apoptosis. factors. Annu Rev Immunol 10: 411–452. Immunity 3: 673–682. Voelkel-Johnson C. (2003). An antibody against DR4 Yagita H, Takeda K, Hayakawa Y, Smyth MJ, Okumura K. (TRAIL-R1) in combination with doxorubicin selectively (2004). TRAIL and its receptors as targets for cancer kills malignant but not normal prostate cells. Cancer Biol therapy. Cancer Sci 95: 777–783. Ther 2: 283–290. Zeng Y, Wu XX, Fiscella M, Shimada O, Humphreys R, Wajant H, Pfizenmaier K, Scheurich P. (2002). TNF-related Albert V et al. (2006). Monoclonal antibody to tumor apoptosis inducing ligand (TRAIL) and its receptors necrosis factor-related apoptosis-inducing ligand receptor 2 in tumor surveillance and cancer therapy. Apoptosis 7: (TRAIL-R2) induces apoptosis in primary renal cell 449–459. carcinoma cells in vitro and inhibits tumor growth in vivo. Walczak H, Miller RE, Ariail K, Gliniak B, Griffith TS, Int J Oncol 28: 421–430. Kubin M et al. (1999). Tumoricidal activity of tumor Zhang M, Zhang Z, Garmestani K, Goldman CK, Ravetch JV, necrosis factor-related apoptosis-inducing ligand in vivo. Nat Brechbiel MW et al. (2004). Activating Fc receptors are Med 5: 157–163. required for antitumor efficacy of the antibodies directed Wallach D, Varfolomeev EE, Malinin NL, Goltsev YV, toward CD25 in a murine model of adult T-cell leukemia. Kovalenko AV, Boldin MP. (1999). Tumor necrosis factor Cancer Res 64: 5825–5829. receptor and fas signaling mechanisms. Annu Rev Immunol Zhang XD, Franco A, Myers K, Gray C, Nguyen T, Hersey P. 17: 331–367. (1999). Relation of TNF-related apoptosis-inducing ligand Watts TH. (2005). TNF/TNFR family members in costimula- (TRAIL) receptor and FLICE-inhibitory protein expression tion of T cell responses. Annu Rev Immunol 23: 23–68. to TRAIL-induced apoptosis of melanoma. Cancer Res 59: Wilcox RA, Flies DB, Zhu G, Johnson AJ, Tamada K, 2747–2753. Chapoval AI et al. (2002). Provision of antigen and CD137 zum Buschenfelde CM, Hermann C, Schmidt B, Peschel C, signaling breaks immunological ignorance, promoting Bernhard H. (2002). Antihuman epidermal growth factor regression of poorly immunogenic tumors. J Clin Invest receptor 2 (HER2) monoclonal antibody trastuzumab 109: 651–659. enhances cytolytic activity of class I-restricted HER2- Wiley SR, Schooley K, Smolak PJ, Din WS, Huang CP, specific T lymphocytes against HER2-overexpressing tumor Nicholl JK et al. (1995). Identification and characterization cells. Cancer Res 62: 2244–2247.

Oncogene