Trial Watch

Abstract

Toll-like receptors (TLRs) have long been known for their ability to initiate innate immune responses upon exposure to conserved microbial components such as lipopolysaccharide (LPS) and double-stranded RNA. More recently, this family of pattern recognition receptors has been attributed a critical role in the elicitation of anticancer immune responses, raising interest in the development of immunochemotherapeutic regimens based on natural or synthetic TLR agonists. In spite of such an intense wave of preclinical and clinical investigation, only three TLR agonists are currently licensed by FDA for use in cancer patients: bacillus Calmette–Guérin (BCG), an attenuated strain of Mycobacterium bovis that operates as a mixed TLR2/TLR4 agonist; monophosphoryl lipid A (MPL), a derivative of Salmonella minnesota that functions as a potent agonist of TLR4; and imiquimod, a synthetic imidazoquinoline that activates TLR7. One year ago, in the August and September issues of OncoImmunology, we described the main biological features of TLRs and discussed the progress of clinical studies evaluating the safety and therapeutic potential of TLR agonists in cancer patients. Here, we summarize the latest developments in this exciting area of research, focusing on preclinical studies that have been published during the last 13 mo and clinical trials launched in the same period to investigate the antineoplastic activity of TLR agonists.

Keywords: :

• CpG oligodeoxynucleotides
• Hiltonol™
• damage-associated molecular patterns
• lipopolysaccharide
• picibanil
• resiquimod

Introduction

Although the Toll gene was originally identified as a controller of the dorsal-ventral embryonic polarity of Drosophila melanogaster as early as in 1985,^1,2 its critical role in the response of fruit flies to fungal infections became clear only 10 y later.³ Approximately in the same period, human orthologs of Toll begun to be characterized^4,5 and implicated in innate immune responses to bacterial lipopolysaccharide (LPS).^6-8 Since then, the murine genome has been shown to encode 13 distinct Toll-like receptors (TLRs), 10 of which are also coded by the human genome, and members of the TLR family have been discovered in evolutionarily distant organisms such as plants and fish.^9-11

TLRs are enzymatically-inactive single membrane-spanning proteins best known for their ability to detect so-called “microbe-associated molecular patterns” (MAMPs), conserved microbial products including (but not limited to) bacterial LPS and derivatives thereof (which generally operate as mixed TLR2/TLR4 agonists),^12-14 components of the bacterial cell wall, such as lipoteichoic acid (a specific activator of TLR2),¹⁵ bacterial flagellin (a pure TLR5 agonist),^16-19 microbial DNA (mostly functioning as a TLR9 agonist),²⁰ microbial single-stranded RNA (ssRNA, which can be detected by both TLR7 and TLR8)^21-23 and viral double-stranded RNA (dsRNA, which specifically activates TLR3).^24-26 Of note, TLRs that detect nucleic acids (i.e., TLR3, TLR7, TLR8 and TLR9) are localized to the endosomal compartment, while TLRs that mainly detect proteo-lipidic structures (i.e., TLR1, TLR2, TLR4, TLR5, TLR6 and TLR10) are exposed on the cell surface.^27,28 As an exception to this general pattern, TLR2 and TLR10 (the sole orphan TLR in humans) have been shown to co-localize at phagosomes, perhaps indicating that TLR10 shares some binding specificity with TLR2. Compelling evidence in support of this hypothesis, however, is missing. Along similar lines, the actual role of murine Tlr11, Tlr12 and Tlr13 has just begun to emerge (see below).

Several TLRs have recently been shown to sense not only exogenous MAMPs but also endogenous “damage-associated molecular patterns” (DAMPs), i.e., molecules released or exposed by stressed, dying or dead cells to convey a danger signal.^29-32 These DAMPs include, but presumably are not limited to: several heat-shock proteins (e.g., HSP60, HSP70),^33,34 uric acid³⁵ and surfactant protein A,³⁶ all of which function as mixed TLR2/TLR4 agonists; the non-histone chromatin-binding protein high mobility group box 1 (HMGB1) and the Ca²⁺- and Zn²⁺-binding protein S100A9, both operating as TLR4 agonists;^37-41 multiple components and breakdown products of the extracellular matrix, which mainly activate TLR4;⁴² and mitochondrial DNA (mtDNA), a pure TLR9 agonist.^43,44 Interestingly nuclear DNA from eukaryotic cells can also be recognized by TLR9 if the latter is ectopically expressed at the plasma membrane (rather than in endosomes).⁴⁵ This suggests that TLR9 might specifically respond to exogenous (as opposed to self) DNA because of its own subcellular localization rather than due to the methylation state and frequency of CpG islands on its ligand (as originally thought).^20,46 A detailed description of the signaling cascades triggered by TLRs in response to MAMPs or DAMPs exceeds the scope of the present Trial Watch and can be found in refs. 27 and 47–50.

The spatiotemporally defined emission of specific DAMPs by dying cells has been proposed to constitute the essence of immunogenic cell death (ICD), a peculiar type of apoptosis that activates adaptive immune responses.^39,51,52 So far, only a few bona fide inducers of ICD have been identified: specific chemotherapeutic agents such as mitoxantrone, doxorubicin and oxaliplatin, ionizing irradiation ad some types of photodynamic therapy. DAMPs that play a prominent role in ICD include (but presumably are not limited to) the endoplasmic reticulum (ER) chaperone calreticulin (CRT), ATP, HSP70 and HMGB1.^39,52,53 Importantly, both HSP70 and HMGB1 appear to exert immunostimulatory functions by activating TLR4 on the surface of antigen-presenting cells, hence promoting the cross-priming of antigen-specific T lymphocytes.^54-56 Thus, TLRs appear to play a prominent role not only in the orchestration of innate immune responses against infectious pathogens, but also in anticancer immunity, be it spontaneous or elicited by (chemo)therapeutic interventions.^23,27,57,58 In accord with this notion, functionally relevant polymorphisms in the genes encoding several TLRs (i.e., TLR1, TLR2, TLR3, TLR4, TLR6, TLR9 and TLR10) have been shown to influence the natural development of a wide array of neoplasms, including tumors that are not associated with a microbial etiology,^59-77 as well as to affect the response of cancer patients to chemotherapy and immunotherapy, at least in some settings.^54,78-80 Moreover, the expression of several TLRs including TLR2, TLR4, TLR7 and TLR9 by malignant cells appear to evolve not only along with oncogenesis and tumor progression, but also in response to microenvironmental cues,^81-90 suggesting that, at least in some types of cancer, TLRs may influence disease progression in a direct fashion rather than as a consequence of immunological effects.

Irrespective of the great preclinical and interest orbiting around TLRs since the late 1990s, however, only three TLR agonists are nowadays approved by FDA for use in cancer patients: bacillus Calmette–Guérin (BCG, an attenuated strain of Mycobacterium bovis initially developed as a vaccine against tuberculosis), which is currently approved for the immunotherapy of in situ bladder carcinoma;^57,91 monophosphoryl lipid A (MPL), a derivative of the LPS of Salmonella minnesota, which is currently licensed as part of Cervarix^®, a vaccine against human papillomavirus 16 and 18 (the etiological determinants of >70% cases of cervical carcinoma);^92,93 and imiquimod (a small imidazoquinoline derivative originally developed as a topic antiviral agent), which is currently used (as a 5% cream) for the treatment of actinic keratosis, superficial basal cell carcinoma and external genital or perianal warts (condylomata acuminata).⁵⁷ Of note, while both BCG and MPL function as mixed TLR2/TLR4 agonists,^14,94,95 imiquimod mainly exerts immunostimulatory effects in a TLR7-dependent manner.^96,97 Interestingly, also the Coley toxin, a mixture of killed Streptococcus pyogenes and Serratia marcescens,⁹⁸ is thought to mediate therapeutic effects by activating TLR2 and/or TLR4.^99-101 Nonetheless, the use of the Coley toxin as an anticancer medication has been discontinued in the 1960s, mostly due to concerns raised by the thalidomide case.¹⁰²

One year ago, in the August and September issues of OncoImmunology, we presented the main functions of human TLRs in innate and cognate immunity and discussed the progress of recent clinical studies evaluating the safety and immunostimulatory activity of TLR agonists in cancer patients.^50,103 Here, along the lines of our monthly Trial Watch series,^50,103-117 we review the latest developments in this area of research, focusing on preclinical studies that have been published during the last 13 mo and clinical trials initiated in the same period to assess the antineoplastic potential of hitherto experimental TLR activators as well as of FDA-approved TLR agonists employed as “off-label” medications against cancer.

Literature Update

Clinical studies

During the last 13 mo, the results of no less than 28 clinical trials investigating the immunostimulatory potential of TLR agonists (all confounded) in cancer patients have been published. Seventeen of these studies involved FDA-approved agents (i.e., BCG, MPL and imiquimod), most often employed as “on-label” medications, while the other 11 trials assessed the safety and therapeutic profile of hitherto experimental TLR agonists (source www.clinicaltrials.gov). Interestingly, none of the clinical studies published since May 1, 2012, have investigated the immunostimulatory potential of BCG in off-label oncological settings. Rather, the intravesical instillation BCG was chosen as a reference approach for the treatment of non muscle-invasive bladder carcinoma (NMIBC), and hence given to the control arm of the cohort,^118,119 or alternative dosing schedules were investigated.^120,121 Thus, the perioperative instillation of mitomycin C (a DNA crosslinker) or gemcitabine (a nucleoside analog) appears not to improve the therapeutic profile of intravesical BCG given as a standalone intervention.^118,119 Moreover, it has been shown that (1) a 2-week intravesical BCG maintenance regimen is virtually as efficient as a standard 3-week course;¹²⁰ (2) BCG administered at 1/3 of the standard dose causes the same side effects than the full-dose regimen among intermediate- and high-risk NMIBC patients;¹²¹ and (3) individuals affected by high-risk (but not intermediate-risk) NMIBC benefit from a 3-y full-dose BCG maintenance therapy only in terms of tumor recurrence, but not of disease progression and overall survival.¹²¹ Finally, the presence of cytogenetically abnormal cells as detected by fluorescence in situ hybridization has been proposed as a means of predicting the response of NMIBC patients to standard-of-care BCG immunotherapy.¹²²

A single study of those mentioned above dealt with Cervarix^®, which contains the MPL-based adjuvant AS04 (MPL + aluminum salts). This work actually demonstrated the immunogenicity and safety of Cervarix^® co-administered with Twinrix^®, an FDA-approved vaccine consisting in inactivated hepatitis A and B viral particles, in girls aged 9–15 y.¹²³ In addition, the hitherto experimental formulation AS15, that is, MPL + QS21 (a water soluble saponin extracted from the South American tree Quillaja saponaria Molina)¹²⁴ + CpG oligodeoxynucleotides (ODNs), has been employed to boost the immunogenicity of a v-erb-b2 erythroblastic leukemia viral oncogene homolog 2, neuro/glioblastoma derived oncogene homolog (avian) (ERBB2)-targeting vaccine in patients with trastuzumab-resistant ERBB2-overexpressing metastatic breast carcinoma.¹²⁵ In this Phase I clinical trial, the co-administration of ERBB2-derived peptides with lapatinib, a relatively unspecific inhibitor of ERBB2 tyrosine kinase activity,¹²⁶ was well tolerated, triggered detectable ERBB2-specific immune responses in a majority of patients, and was associated with promising clinical benefits, warranting the initiation of Phase II/III studies.¹²⁵

Imiquimod (also known as R-837) still stands out as the TLR agonist that generates the broadest clinical interest of all, presumably owing not only to its approval status, but also to the increasing amount of data from clinical studies and pharmacosurveillance demonstrating that—at least in its 5% cream formulation—imiquimod has an exceptional safety profile.^97,103 In line with this notion, no less than 12 studies investigating the immunostimulatory potential of imiquimod in cancer patients have been published during the last 13 mo (8 of which employing imiquimod as an off-label medication). Conversely, no reports on the immunostimulatory activity of resiquimod (R-848), another synthetic imidazoquinoline that operates as a TLR7 agonist, in cancer patients have been made in the same period, though the clinical interest in this compounds remains high (see below).

The combination of imiquimod 5% cream (commercialized by 3M Pharmaceuticals under the label of Aldara^®) with aminolevulinate-based photodynamic therapy has been shown to provide therapeutic benefits to patients affected by actinic keratosis and basal cell carcinoma over the use of either intervention alone.^127,128 In addition, imiquimod 5% cream has been evaluated as a means to reduce the size of nodular basal cell carcinomas of the face, and hence limit the aesthetic impact of subsequent Mohs micrographic surgery, with encouraging results.¹²⁹ Besides being tested in such on-label indications, imiquimod has been investigated for its ability to exert immunostimulatory effects as an off-label medication, for instance (1) in melanoma patients receiving imiquimod as a standalone intervention or as an adjuvant to a nanoparticle-based vaccine;^130,131 (2) in subjects with cutaneous metastases of breast carcinoma or melanoma, receiving imiquimod alone or in combination with 5-fluorouracil (a nucleoside analog), respectively;^132,133 (3) in prostate and renal cancer patients, receiving imiquimod as an adjuvant to a dual peptide-based vaccine;¹³⁴ and (4) in women with high-grade cervical intraepithelial neoplasms and in HIV-1-infected men bearing anal intraepithelial tumors, two settings in which imiquimod was employed as a standalone therapeutic intervention.^135,136 Moreover, the safety and immunostimulatory potential of the intravesical instillation of a liquid formulation of imiquimod (TMX-101) have been assessed in a cohort of NMIBC patients.¹³⁷ Although imiquimod appeared to be inferior to other therapeutic modalities (notably electrocautery) in the treatment of HIV1-associated anal intraepithelial neoplasms,¹³⁶ these studies confirmed the safety of imiquimod and its ability to stimulate (natural or vaccine-induced) immune responses that—at least in a fraction of patients—engender clinical benefits.

Picibanil (OK-432) is a lyophilized preparation of Streptococcus pyogenes (operating as a mixed TLR2/TLR4 agonist) that has been approved for use in cancer patients by the Japanese Ministry of Health and Welfare as early as in 1975.^42,138 In line with this notion, the immunostimulatory activity of picibanil is being intensively investigated in Japan, but very much less so in the US and Europe. During the last 13 mo, picibanil has been shown to be well tolerated and effective as an adjuvant (1) to immature DCs administered intratumorally to resectable pancreatic cancer patients;¹³⁹ (2) to cisplatin (a DNA-damaging agent frequently associated with chemoresistance)¹⁴⁰ and hyperthermotherapy, in a cohort of individuals with malignant pleural effusions;¹⁴¹ and (3) to a NY-ESO-1-based vaccine, in a single lung cancer patient selected out of a previously completed clinical trial¹⁴² for the study of multiple immunological parameters.¹⁴³ In the same period, additional TLR2/TLR4 agonists have been shown to be safe and to exert therapeutically relevant immunostimulatory effects in cancer patients, including OM-174 (also known as CRX-527), a water soluble, diphosphorylated and triacetylated form of lipid A from Escherichia coli,¹⁴⁴ which has been tested in patients with refractory solid tumors;¹⁴⁵ and IMM-101, a preparation of heat-killed Mycobacterium obuense¹⁴⁶ that has been investigated as a standalone therapeutic intervention against melanoma.¹⁴⁷ Moreover, recently completed trials have evaluated the safety and immunostimulatory profile of (1) the TLR3 agonist Hiltonol, a particular formulation of polyriboinosinic polyribocytidylic acid (polyI:C) that includes carboxymethylcellulose and poly-L-lysine as stabilizing agents,¹⁴⁸ employed as adjuvant of a peptide-derived vaccine against ovarian cancer;¹⁴⁹ (2) the imidazoquinoline 852A, a TLR7 agonist, administered subcutaneously for a prolonged period to patients with advanced hematologic malignancies;¹⁵⁰ (3) the TLR9 agonist IMO-2055 (an immunomodulatory oligonucleotide also known as EMD1201081),^151,152 given in combination with 5-fluorouracil, cisplatin, and cetuximab (an FDA-approved monoclonal antibody specific for the epidermal growth factor receptor, EGFR)¹⁵³ as a first-line palliative treatment to patients with recurrent/metastatic head and neck squamous cell carcinoma (HNSCC);¹⁵⁴ and (4) Agatolimod (also known as CpG-7909, PF-3512676 and Promune^®), an unmethylated CpG ODN that also activates TLR9,¹⁵⁵ given in combination with tremelimumab (an experimental monoclonal antibody targeting the immune checkpoint regulator cytotoxic T-lymphocyte-associated protein 4, CTLA4)^156,157 to patients affected by advanced solid tumors (including melanoma),¹⁵⁸ or combined with local irradiation in subjects bearing mycosis fungoides.¹⁵⁹

Preclinical studies

During the last 13 mo, TLR agonists have been the subject of an intense wave of preclinical investigation, resulting in more than 200 hundreds scientific publications (source www.ncbi.nlm.nih.gov/pubmed/). A large fraction of these studies has confirmed the ability of multiple TLR agonists to mediate—alone or combined with radio-, chemo- or immunotherapy—prophylactic or therapeutic effects against a variety of tumors, including (but not limited to) lymphoma,^160-162 mastocytoma,¹⁶² glioma,¹⁶³ breast carcinoma,^164,165 thymoma,¹⁶⁶ fibrosarcoma,¹⁶⁷ head and neck cancer,^168,169 melanoma,^170-174 lung cancer,^174-176 colorectal carcinoma,^174,177-179 renal cancer^177,180 and ovarian carcinoma.^181,182 For the most part, such robust antineoplastic effects have been ascribed to the ability of TLR agonists to induce de novo or boost pre-existing (natural or therapy-elicited) immune responses. This said, accumulating preclinical evidence suggests that TLR-targeting agents can also influence tumor progression in a direct manner as they interact with TLRs expressed on the surface of malignant cells. Thus, activators of TLR3 (e.g., dsRNA, polyI:C),^183-186 TLR4 (e.g., picibanil, LPS)^187,188 and TLR7 (e.g., imiquimod)¹⁸⁹ have all been shown to arrest the proliferation or induce the death of various cancer cell lines in vitro, in the absence of immune effectors. However, the direct effects of TLR agonists on malignant cells exhibit a consistent degree of context-dependency. In line with this notion, TLR-conveyed signals have also been shown to confer malignant cells with a proliferative advantage,^190-201 with an increased invasive/metastatic potential,^{193,194,196-200,202-208} with a profound resistance to environmental and chemotherapeutic cues,^{192,195,209-212} or with the ability to secrete immunosuppressive cytokines.^201,211,213 Of note, endothelial cells have been shown to respond to the release of peroxiredoxin 1 (a TLR4-interacting DAMP) from dying prostate cancer cells by secreting vascular endothelial growth factor, thus stimulating angiogenesis.²¹⁴ Thus, TLR agonists may promote tumor progression not only as they directly alter the behavior of malignant cells, but also as they engage signaling circuitries that involve the tumor stroma. Taken together, these observations suggest that the therapeutic potential of a given TLR agonist in a given clinical setting should be carefully evaluated in view of its (more or less pronounced) propensity to mediate direct or indirect pro-tumor effects.

The discovery/development of novel TLR agonists is also a very active area of research and during the past 13 mo several new compounds and strategies to activate specific TLRs have been reported. Great interest has gathered around the use of cationic preparations, notably liposomes, as a means to deliver immunotherapeutic agents (e.g., TLR agonists, vaccines) to neoplastic lesions.^215-217 Indeed, cationic agents may per se exert immunostimulatory effects by binding to, and hence activating TLR4.²¹⁶ In addition, several new TLR agonists have been identified, including endogenous DAMPs as well as exogenous chemicals. Thus, heparan sulfate, a component of the extracellular matrix, has turned out to bind TLR4 on the surface of DCs, hence promoting their maturation (in vitro), and to be involved in the etiology of the graft-vs.-host disease.²¹⁸ Along similar lines, multiple mature microRNAs have been demonstrated to activate natural killer (NK) cells, in vitro and in vivo, in a TLR1-dependent manner, hence protecting mice from a challenge with A20 lymphoma cells,²¹⁹ while two specific microRNAs secreted by tumor cells, namely, miR-21 and miR-29a, have been reported to mediate pro-metastatic effects by stimulating a TLR7-dependent inflammatory response.²²⁰ A novel, chemically-defined LPS derivative has been reported to reinstate the immunogenicity of HGMB1-deficient cancer cells, thus compensating for cell-intrinsic alterations that may compromise the therapeutic efficacy of ICD inducers.²²¹ Finally, extracts of Larix kaempferi (a popular Japanese larch) have been shown to exert promising immunostimulatory effects in a murine model of thymoma as they activate TLR2 and TLR4,²²² while distinct isothiocyanates have been reported to differentially modulate (either promote or inhibit) TLR3-dependent signal transduction cascades.²²³

Several papers published since May 1, 2012, have investigated the fundamental mechanisms whereby TLRs exert a prominent influence not only on immune effector cells but also on malignant and stromal cells. Among dozens of top quality reports, we have found the following works of particular interest. Shi et al. have identified a role for Tlr11 in the control of Salmonella spp infection at the level of Peyer patches,²²⁴ while Oldenburg and colleagues have discovered (one of) the natural ligand(s) of Tlr13, namely, bacterial 23S rRNA.²²⁵ Several groups have demonstrated that specific microRNAs play a critical function in the signaling cascades elicited by various TLRs, including TLR4 and TLR9, in both immune and malignant cells.^198,226-228 Dibra et al. have demonstrated that TLR9 signaling in macrophages and CD3 signaling in T cells underpin a cellular crosstalk that result in the secretion of high levels of interleukin (IL)-30.²²⁹ Balamurugan and colleagues have unveiled a molecular cascade linking the oncosuppressive activity of FBXW7 to the transcriptional repression of TLR4,²³⁰ whereas Aksoy et al. have reported that the subcellular compartmentalization of TLR4 is under the control of phosphoinositide-3-kinase (PI3K).²³¹ As PI3K is frequently hyperactivated in malignant cells,²³² these findings lend further support to the notion that TLR4 may be implicated in tumor-cell intrinsic oncogenic signaling cascades.^{79,188,211,233} By generating induced pluripotent stem cells from the dermal fibroblasts of TLR3-deficient patients and healthy individuals, Lafaille and coworkers have implicated TLR3 in the cell-intrinsic immunity of neurons and oligodendrocytes to herpes simplex virus 1.²³⁴ The research group lead by Richard Gallo has demonstrated that UV rays induce alterations in the double-stranded domains of some non-coding RNAs, endowing them with the ability to operate as DAMPs and activate TLR3.²³⁵ Luger and colleagues have shown that TLR4-activated DCs switch from an initial pro-inflammatory mode, characterized by the release of T_H1 cytokines including IL-12 and interferon (IFN)γ, to an anti-inflammatory one, featuring the autocrine secretion of IL-10.²³⁶ During this latter functional phase, DCs reportedly become able to inhibit the proliferation of T cells and to convert them into IL-10, FOXP3-expressing regulatory T cells (Tregs).^236-239 Karbach and coworkers have demonstrated that the use of Agatolimod is associated with the formation of neutralizing antibodies in significant fraction of patients, de facto abolishing its long-term therapeutic potential.²⁴⁰ According to Lin et al., TLR2 is required for the removal of senescent hepatocytes by immune effector cells, hence restraining the development of hepatocellular carcinoma.^241,242 Ochi et al. have implicated TLR4 and TLR7 in pancreatic carcinogenesis, in both mice and humans, presumably linked to their robust pro-inflammatory functions.^243-245 Hodven and collaborators have reported that picibanil not only activates TLR2 and TLR4, but also exert TLR3-dependent immunostimulatory effects.²⁴⁶ Finally, Walter et al. have discovered that Aldara^® stimulates inflammatory responses also in an imiquimod- and TLR7-independent manner.²⁴⁷ Indeed, an abundant component of Aldara^®, isostearic acid, appears to trigger the activation of the inflammasome, hence stimulating the release of IL-1β and IL-18,²⁴⁸ even in the absence of imiquimod. Altogether, these findings suggest (1) that the biological functions of TLRs are complex and exhibit a significant degree of context-dependency, and (2) that the signal transduction cascades originating from TLRs are intimately intertwined with several other cell-intrinsic and cell-extrinsic signaling pathways. Much work is still needed to characterize this functional crosstalk, which has profound implications for cancer (immuno)therapy.

Update on Clinical Trials

When this Trial Watch was being redacted (May 2013), official sources listed only 32 clinical trials launched after May 1, 2012, to investigate the safety and therapeutic potential of experimental and FDA-approved TLR agonists (source www.clinicaltrials.gov). Of these, 7 involved BCG (1 study), AS04 (3 studies) or imiquimod (3 studies) as fully “on-label” medications, and therefore will not be discussed further here. Of the remaining 25 clinical trials, 10 aimed at assessing the immunostimulatory potential of BCG or imiquimod in “off-label” oncological settings, and 15 were launched to test the safety and antineoplastic activity of hitherto investigational TLR agonists (Table 1).

Table 1. Recent clinical trials evaluating Toll-like receptor agonists in cancer patients*

Agent	Target	Cancer type	Status	Phase	Route	Notes	Ref.
AS15	TLR2 TLR4 TLR9	NSCLC	Not yet recruiting	II	i.m.	As single agent	NCT01853878
BCG	TLR2 TLR4	Melanoma	Recruiting	I	i.t.	Combined with ipilimumab	NCT01838200
				II/III	n.a.	Combined with irradiated melanoma cells and rhGM-CSF	NCT01729663
CBLB502	TLR5	HNSCC	Not yet recruiting	I	s.c.	Combined with cisplatin and IMRT	NCT01728480
GNKG168	TLR9	ALL AML	Active, not recruiting	I	i.v.	As single agent	NCT01743807
Hiltonol^TM	TLR3	AML MDS	Recruiting	I	s.c.	Combined with recombinant vaccine and decitabine	NCT01834248
		Melanoma	Recruiting	I	s.c.	Combined with a NY-ESO-1- based vaccine and ipilimumab	NCT01810016
		MM	Active, not recruiting	I	i.m.	Combined with a multi- peptide vaccine (PVX-410)	NCT01718899
		NSCLC	Recruiting	I/II	s.c.	Combined with MUC1- targeting vaccination	NCT01720836
		Pancreatic cancer	Recruiting	0	i.t.	Combined with immature DCs	NCT01677962
		Solid tumors	Active, not recruiting	II	n.a.	Combined with autologous DCs	NCT01734564
Imiquimod	TLR7	Anal dysplasia	Not yet recruiting	IV	Local	As single agent	NCT01663558
		Astrocytoma GBM Glioma	Not yet recruiting	I	Topical	Combined with DC- based vaccination	NCT01808820
		Ependymoma	Recruiting	n.a.	Topical	Combined with HLA-A2- restricted TAA-derived peptides	NCT01795313
		Glioma	Recruiting	0	Topical	Combined with a tumor cell lysate-based vaccine	NCT01678352
		Lentigo maligna melanoma	Recruiting	III	Topical	As a standalone neoadjuvant intervention	NCT01720407
		Sarcoma	Recruiting	I	Topical	Combined with DC-based vaccination ± gemcitabine	NCT01803152
		VIE	Not yet recruiting	III	Topical	As an alternative to primary surgery	NCT01861535
Resiquimod	TLR7 TLR8	CTCL	Recruiting	I/II	Topical	As single agent	NCT01676831
		Melanoma	Recruiting	n.a.	Topical	Combined with MLANA- targeting vaccination	NCT01748747
		nBCC	Recruiting	I/II	Topical	As single agent	NCT01808950
SD-101	TLR9	Lymphoma	Recruiting	I	i.t.	Combined with radiotherapy	NCT01745354
TMX-101	TLR7	NMIBC	Recruiting	II	Intravesical	As single agent	NCT01731652
VTX-2337	TLR8	HNSCC	Not yet recruiting	II	n.a.	Combined with conventional chemotherapy plus cetuximab	NCT01836029
		Reproductive tract cancer	Recruiting	II	n.a.	Combined with PLD	NCT01666444

ALL, acute lymphoblastic leukemia; AML, acute myeloid leukemia; BCG, bacillus Calmette-Guérin; CTCL, cutaneous T-cell lymphoma; DC, dendritic cell; GBM, glioblastoma; HNSCC, head and neck squamous cell carcinoma; i.m., intra musculum; IMRT, intensity-modulated radiation therapy; i.t., intra tumorem; i.v., intra venam; MLANA, melan-A; MDS, myelodysplastic syndrome; MM, multiple myeloma; MUC1, mucin 1; n.a., not available; nBCC, nodular basal cell carcinoma; NMIBC, non-muscle-invasive bladder carcinoma; NSCLC, non-small cell lung carcinoma; PLD, pegylated liposomal doxorubicin; rhGM-CSF, recombinant human granulocyte-macrophage colony-stimulating factor; s.c., sub cutem; TAA, tumor-associated antigen; TLR, Toll-like receptor; VIE, vulvar intraepithelial neoplasia. *Started after May 1, 2012.

Recently initiated clinical trials are testing BCG, either in combination with the FDA-approved immunostimulatory monoclonal antibody ipilimumab^78,249-251 (NCT01838200) or as an adjuvant to irradiated allogeneic melanoma cells plus recombinant granulocyte macrophage colony-stimulating factor (GM-CSF, Molgramostin^®) (NCT01729663), only in advanced melanoma patients. Ad odds with such an apparent decrease in the clinical interest generated by BCG, the possibility to use of imiquimod as an immunostimulatory agent in patients affected by a variety of tumors remains under intensive investigation. In particular, during the last 13 mo clinical trials have been initiated that aim at assessing the safety and therapeutic efficacy of imiquimod (1) employed as a standalone intervention to minimize the risk of recurrence as well as the esthetic impact of surgery in patients bearing lentigo maligna melanomas on the head and face (NCT01720407); (2) given in support of DCs loaded with tumor cell lysates to pediatric and adult sarcoma patients who have optionally been pre-conditioned with gemcitabine (to inhibit myeloid-derived suppressor cells)^252-255 (NCT01803152); (3) used as a topic support to a DC-based vaccine in adult and pediatric patients affected by anaplastic astrocytoma, glioma or glioblastoma multiforme (NCT01808820), or to a tumor cell lysate-based vaccine in individuals with high-grade/recurrent glioma²⁵⁶ (NCT01678352); (4) given as an adjuvant to HLA-A2-restricted tumor-associated antigen (TAA)-derived peptides in children bearing recurrent ependymomas (NCT01795313); (5) used as a standalone intervention in the form of anal suppositories (as an alternative to ablative therapy) to men who had sex with HIV-1-infected men (NCT01663558); or (6) employed as a primary measure in alternative to surgery in women affected by vulvar intraepithelial neoplasia (NCT01861535). Moreover, following the encouraging results of a recent Phase I clinical trial,¹³⁷ a Phase II study has been initiated to test the therapeutic activity of a liquid imiquimod preparation (TMX-101) administered intravesically to NMIBC patients (NCT01731652).

Also the general interest in experimental TLR agonists appears to decrease, with the notable exception of Hiltonol™, a particular formulation of polyriboinosinic polyribocytidylic acid (polyI:C, Ampligen™, Rintatolimod) that includes carboxymethylcellulose and poly-l-lysine as stabilizing agents.¹⁴⁸ Thus, the safety and immunostimulatory potential of Hiltonol™, which operates as a TLR3 agonist, are being investigated in (1) acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS) patients receiving a recombinant vaccine consisting of full-length NY-ESO-1 fused to an anti-LY75 monoclonal antibody in combination with 5-aza-2'-deoxycytidine (an FDA-approved analog of cytidine also known as decitabine)²⁵⁷ (NCT01834248); (2) multiple myeloma (MM) patients treated with a multi-peptide vaccine (PVX-410) (NCT01718899); (3) melanoma patients administered with a NY-ESO-1 based vaccine plus ipilimumab (NCT01810016); (4) unresectable pancreatic carcinoma patients receiving immature DCs²⁵⁸ (NCT01677962); (5) individuals administered with a mucin 1 (MUC1)-derived peptide vaccine for the treatment of non-small cell lung carcinoma (NSCLC) (NCT01720836); and (6) subjects bearing not better specified neoplasms and allocated to receive autologous DCs (NCT01734564).

The TLR5 agonist CBLB502 (a derivative of Salmonella flagellin also known as Entolimod) is being tested in Stage III–IV HNSCC patients receiving cisplatin and radiation therapy (NCT01728480), reflecting the robust immunostimulatory and radioprotective activities that have been ascribed to this compound in preclinical settings.^259-262 The safety and immunostimulatory profile of the TLR7 agonist resiquimod are being assessed (1) in cohorts of cutaneous T-cell lymphoma (CTCL) and nodular basal cell carcinoma (nBCC) patients, two settings in which resiquimod is administered topically as a standalone therapeutic intervention (NCT01676831; NCT01808950); as well as (2) in patients with recurrent or advanced melanoma, receiving resiquimod or an HIV-1-derived peptide as alternative adjuvants to a melan-A (MLANA)-targeting vaccine (NCT01748747). VTX-2337, a small molecule that specifically binds and activates TLR8,¹⁶⁸ is under investigation for its ability to improve the therapeutic response of HNSCC patients to conventional chemotherapy plus cetuximab (NCT01836029) and that of women bearing reproductive tract neoplasms to pegylated liposomal doxorubicin (NCT01666444). Finally, a few clinical trials have recently been initiated to evaluate the immunostimulatory potential of TLR9 agonists in cancer patients. In particular, (1) AS15 (which actually operates as a mixed TLR4/TLR9 agonist, as it contains both MPL and Agatolimod) is being tested for its ability to boost immune responses elicited by full-length recombinant preferentially expressed antigen in melanoma (PRAME) in NSCLC patients upon tumor resection (NCT01853878); (2) the safety and therapeutic profile of GNKG168 (a CpG ODN),^263,264 given intravenously as a standalone agent, are being assessed in a cohort of relapsed acute lymphoblastic leukemia (ALL) or AML patients (NCT01743807); and (3) the intratumoral administration of SD-101 (a phosphorothiolate CpG ODN)²⁶³ is being evaluated as a means to exacerbate the antineoplastic effects of local irradiation in subjects bearing Hodgkin and non-Hodgkin lymphoma (NCT01745354).

Of note, during the last 13 mo, 11 clinical trials investigating the immunostimulatory activity of TLR agonists in cancer patients that had been registered at www.clinicaltrials.gov after January 1, 2008, are now listed as terminated (2 studies), suspended (1 study) or completed (9 studies). In particular, NCT01013623 (testing BCG as a standalone agent in melanoma patients) and NCT01396018 (assessing the activity of VTX-2337 in combination with radiotherapy in B-cell lymphoma patients) have been terminated owing to a lack of accrual and to an excessively slow rate of recruitment, respectively, while NCT01400672 (investigating the immunostimulatory potential of a tumor cell lysate-based vaccine adjuvanted with imiquimod in individuals bearing glioma or glioblastoma) has been suspended for safety concerns. The results of only 1 (NCT00707174) of the 9 studies that are nowadays listed as completed (NCT00006352; NCT00079300; NCT00581425; NCT00707174; NCT00773097; NCT00785122; NCT00952692; NCT01161888; NCT01219348) have been published, indicating that the administration of topical retinoid may reduce the risk of local recurrence among lentigo maligna melanoma patients treated with imiquimod 5% cream.²⁶⁵

Concluding Remarks

There is an abundant literature demonstrating that TLR agonists can exert potent immunostimulatory effects in vivo, hence triggering de novo or boosting pre-existing (natural or therapy-elicited) anticancer immune responses.^{23,27,47,52,58,266} Nonetheless, only a few TLR agonists are nowadays licensed by international regulatory agencies for use in cancer patients, i.e., BCG, MPL, imiquimod (all of which are approved by the US FDA) and picibanil (which is approved by the Japanese Ministry of Health and Welfare). In addition, the number or clinical trials that are initiated to test the safety and therapeutic profile of TLR agonists in oncological indications is steadily decreasing. Thus, the trend that we have delineated one year ago, in the August and September issues of OncoImmunology,^50,103 appears to be confirmed. Such a decline has surely been influenced by limited availability of clinical grade TLR agonists, prompting academic researchers to focus on surrogate compounds.^267,268 Moreover, most (if not all) TLR agonists activate a complex set of signal transduction cascades that involve not only immune effectors but also malignant cells and components of the tumor stroma. As the biological outcomes of such a cell-intrinsic and cell-extrinsic signaling network exhibit an elevated degree of context-dependency, it is possible—yet remains to be formally demonstrated—that TLR agonists induce therapeutic responses only in specific patient subsets. Thus, the future of these immunostimulatory agents might depend not only on the precise elucidation of the signaling pathways that they activate at the cell-intrinsic and cell-extrinsic level, but also on the identification of biological markers that predict the propensity of individual cancer patients to obtain a clinical benefit from TLR agonists.

Abbreviations:
ALL	=	acute lymphoblastic leukemia
AML	=	acute myeloid leukemia, BCG, bacillus Calmette-Guérin
CRT	=	calreticulin
CTCL	=	cutaneous T-cell lymphoma
CTLA4	=	cytotoxic T-lymphocyte-associated protein 4
DAMP	=	damage-associated molecular pattern
DC	=	dendritic cell
dsRNA	=	double-stranded RNA
EGFR	=	epidermal growth factor receptor
ER	=	endoplasmic reticulum
ERBB2	=	v-erb-b2 erythroblastic leukemia viral oncogene homolog 2, neuro/glioblastoma derived oncogene homolog (avian)
GM-CSF	=	granulocyte macrophage colony-stimulating factor
HMGB1	=	high mobility group box 1
HNSCC	=	head and neck squamous cell carcinoma
HSP	=	heat-shock protein
ICD	=	immunogenic cell death
IFN	=	interferon
IL	=	interleukin
LPS	=	lipopolysaccharide
MAMP	=	microbe-associated molecular pattern
MDS	=	myelodysplastic syndrome
MLANA	=	melan-A
MPL	=	monophosphoryl lipid A
mtDNA	=	mitochondrial DNA
MUC1	=	mucin 1
nBCC	=	nodular basal cell carcinoma
NK	=	natural killer
NMIBC	=	non muscle-invasive bladder carcinoma
NSCLC	=	non-small cell lung carcinoma
ODN	=	oligodeoxynucleotide
PI3K	=	phosphoinositide-3-kinase
polyI:C	=	polyriboinosinic polyribocytidylic acid
PRAME	=	preferentially expressed antigen in melanoma
ssRNA	=	single-stranded RNA
TAA	=	tumor-associated antigen
TLR	=	Toll-like receptor
Treg	=	regulatory T cell

Acknowledgments

Authors are supported by the European Commission (ArtForce); Agence National de la Recherche (ANR); Ligue Nationale contre le Cancer; Fondation pour la Recherche Médicale (FRM); Institut National du Cancer (INCa); Association pour la Recherche sur le Cancer (ARC), LabEx Immuno-Oncologie; Fondation de France; Fondation Bettencourt-Schueller; AXA Chair for Longevity Research; Cancéropôle Ile-de-France, Paris Alliance of Cancer Research Institutes (PACRI) and Cancer Research for Personalized Medicine (CARPEM).

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.