Phase I study of OM-174, a lipid A analogue, with assessment of immunological response, in patients with refractory solid tumors

Despite a downward trend of mortality rates [1‐5], cancer remains the second cause of death, after cardiovascular diseases, in industrialized countries accounting for 570,000 deaths in the USA in 2010 [6, 7].

Whereas the primary tumor can, in most cases, be efficiently treated by a combined therapeutic approach, preventing the metastatic spread of the disease is often not effective. The eradication of disseminated tumor cells present in the blood circulation and micro-metastases in distant organs therefore represents another promising approach of cancer immunotherapy. The immune system could recognize tumor-specific antigens, and eliminate cancer cells. Furthermore it has been highlighted that some cancers, such as colorectal cancer, cause direct inhibition of the host's immune response with a detrimental effect upon prognosis suggesting that stimulation of the immune system might offer a therapeutic strategy to counteract these effects [8]. This has been observed with Toll-like receptor 2 (TLR2), TLR4 and TLR9 agonists, such as BCG and paclitaxel, which have demonstrated significant benefit for the treatment of bladder and breast cancers [9‐11].

Toll-like receptors (TLRs) are important sensors of foreign microbial components as well as products of damaged or inflamed self-tissues. TLRs are a family of type I transmembrane proteins that are the major pattern recognition receptors, and using TLR proteins, host can recognize the conserved molecular structures found in pathogens called pathogen-associated molecular patterns (PAMPs). In addition to their recognition of PAMPs, several TLRs have also been shown to recognize endogenous ligands associated with inflammation, which have been termed danger-associated molecular patterns (DAMPs). Then, TLRs have an important role in maintaining tissue homeostasis by regulating the inflammatory and tissue repair in responses to injury. There are 13 TLRs described in mammals with a broad expression profile but TLR11, 12 and 13 are not expressed in the human genome. TLR1, TLR2, TLR4, TLR5 and TLR6 reside in the plasma membrane and recognize extracellular ligands [12]. It is now well recognized that TLR4 is signalling receptor for lipopolysaccharides (LPS) and more specifically lipid A [13, 14].

LPS are major cell wall components of Gram-negative bacteria and were found to be able to induce regression of several tumor types in animal models [15]. This antitumor activity is related to lipids A, a component of LPS with oligosaccharide core and polysaccharide chain [16‐19]. Lipids A do not induce direct cytotoxicity on tumor cells but trigger the immune system [16, 19] stimulating expression and secretion of several cytokines [20, 21], and activating the inducible nitric oxide synthase (NOSII) [22]. They also induce activation of various immune cells which infiltrate the tumors [15, 17, 23, 24].

OM-174 is an analogue of lipid A that exerts, in experimental models, anti-tumoral effect against different tumor types including colon and breast cancers and melanoma [17, 25]. In a model of peritoneal carcinomatosis induced in BDIX rats by intraperitoneal injection of syngeneic PROb colon cancer cells, intravenous (IV) administration of OM-174 (1 mg/kg, 15 times every third day), induced complete regression of tumors and hemorrhagic ascitis in 95% of cases [17]. This anti tumoral activity was associated with cytokine secretion and NOSII activation [22].

Based on these experimental data showing that several administrations are needed to achieve antitumor effect, as well as on the overall favourable safety profile of lipid A analogues observed in phase I studies [26‐28] including two with single IV-administration of OM-174, one conducted in healthy volunteers and the other in patients with solid cancer, we designed a phase I study with repeated dose of OM-174. The aim of this study was to determine tolerability, toxicities, the maximum tolerated dose (MTD) based upon predetermined dose limiting toxicities (DLT) and the recommended phase II dose of OM-174 administered as a 15 minutes intravenous infusion, repeated twice weekly for 5, 10 or 15 administrations.

This new phase one was deemed necessary as animal data suggested that repeated administration of OM-174 were needed to observe anti-tumor effect, with a maximal effect observed after 15 injections. This study also aimed to assess the pharmacokinetic profile and the biological response induced by OM-174 in the multidose administration schedule.

This was a single center phase I study conducted in the clinical research unit (Centre Georges-François Leclerc - INSERM CIC-P 803) in Dijon France. The study protocol and its amendments were approved by Ethics Committee and Health Authorities (NCT01800812 (ClinicalTrials.gov Identifier)). The trial was designed according to the current Declaration of Helsinki [29] and conducted in accordance with Good Clinical Practice Guidelines. Written informed consent was obtained from each participating patient before study entry. The study was supported by a grant from the French Ministry of health (Projet Hospitalier de Recherche Clinique, PHRC 2002). OM-174 was a generous gift from OM-Pharma (Geneva, Switzerland).

Our study strengthens data on the overall very favourable safety profile of OM-174, as administered for up to 15 injections as a biweekly infusion and up to 1000 μg/m² which is the recommended dose. Chills (70%) and fever (41%) were the most common adverse advents, with no grade 3 or 4, explaining that the MTD could not be determined, and were reversible with acetaminophen, NSAID or steroid treatments. Furthermore no increase in severity occurred as injections were repeated. It might be argued that giving acetaminophen or NSAIDs to patients with grade 1 or 2 fever might impair antitumoral effect as it has been suggested that response to IL-2 plus interferon-alpha might be altered by acetaminophen in patients with metastatic melanoma [34]. Nevertheless our patients were included prior to this report that has never been published as a full paper and only a trend was suggested. The number of patients with disease stabilization observed in our study (3) is too small to allow for an appropriate assessment of the influence of acetaminophen or NSAIDs use in tumour response, but it has to be noticed that these 3 patients all received acetaminophen.

Others adverse events included nauseas, vomiting, fatigue, headache and diarrhoea and were generally of mild to moderate intensity. The more severe adverse event was an episode of bronchospasm which occurred for a patient with a progressive hepatocarcinoma after the third infusion of 800 μg/m².

The safety profile of OM-174 we describe here is similar to that observed in others studies with other lipids A [26‐28], whether the lipid A was of synthetic origin as SDZ MRL 953 [26] and ONO-4007 [27], or prepared from S typhimurium and minnnesota as in the phase I trial performed by Vosika et al. Similarly to our study, Vosika et al. described bronchospasm in one patient, considered as a severe adverse event. In this study another patient experienced, at the highest dose level (500 μg/m²) a severe hypotension which was considered by the authors as a DLT. It has to be noted that we assessed twice higher dose and number of injections (15 vs. 8) of lipid A than in the study by Vosika et al [28].

OM-174 PK parameters appeared to be independent of dose within the dose range of 600 – 1000 μg/m². A trend toward higher OM-174 concentrations was observed when administered doses increased, but the difference was not significant. OM-174 is a drug with a low systemic clearance (0.11 L/h) and a small distribution volume (4.6 L). Our results are in agreement with those of de Bono et al. [27] who described that the pharmacokinetic of ONO-4007, another synthetic lipid A analogue, appeared to be independent of dose and showed linearity with respect to time. ONO-4007 was described to have a low systemic clearance (approximately 1.3 ml/min) and a small volume of distribution (5-8 liters) with a long t1/2 of 74-95 h. These data validated the twice weekly administration protocol. It has previously been reported that shorter intervals between repeated injection of LPS lead to the development of LPS tolerance [35] and similar tolerance might have occurred with Lipid A analogues even if it has never been studied.

Pharmacodynamic studies showed OM-174 IV-infusion induced a significant increase in IL-8, IL-10, IL-6 and TNF-α concentrations. IL-8 and IL-10 concentrations increased after each OM-174 injection whereas TNF-α and IL-6 concentrations were highest after the first OM-174 injection and declined thereafter as OM-174 injections were repeated. This response was not correlated to the dose of OM-174 and is known as tolerance. The occurrence of tolerance or host desensitization in animals and humans administered multiple doses of microbial products has long been recognized [36]. It was reported with OM-174 in preclinical studies. In the study by De Bono et al. [27], an increase in IL-1α, IL-6, and TNF-α concentration was observed whereas it was not the case with SDZ MRL 953 in the study by Kiani et al. [26]. Our study is in partial agreement with both studies as we observed an increase in TNF-α, IL-6, IL-8 and IL-10, but not in IL-1β or IL-12. This increase of secretion of TNF-α is in the same magnitude as that observed with ONO-4007 (1970 versus 1200 pg/ml), suggesting a similar biological activity and confirming that OM-174 activates the TLR4 receptor as has been demonstrated in preclinical models. The variations of cytokines secretion could be due to the dose (from 2 to 125 mg/patient), or administration schedules (every day for SDZ MRL 953, every week for ONO-4007) differences, between the three clinical assays. However, the best clinical tolerance of OM-174 makes it a drug of choice for clinical development.

Interestingly, our study also suggests that a sufficient number of OM-174 injections has to be administered to reach a biological activity since a trend toward an increase in NK cells number and activity was observed only in patients receiving 15 infusions. As only three patients received 15 infusions these findings deserve further studies to be confirmed.

As OM-174 has been shown, in mice, to induce TNF-α production via TLR4 receptors, a family of receptors which recognized pathogen-associated microbial structure and are involved in innate immune cell activation and antitumor immune response inducing tumor cell death [37], we search for possible mutations that may explain inter patient variability in terms of biological and clinical activities. Sixteen out of seventeen patients in our study carried the wild type suggesting that TLR4 responsiveness was unaltered in our population. Since no functional analysis was performed, we cannot formally rule-out a TLR4-mediated variability in clinical response, but in the absence of genomic variability this is unlikely to be the case. However, the finding of TNF secretion after administration of OM-174 in our study as has been observed in mice suggested that OM-174 activates TLR4 well.

We did not observed objective tumour responses, as it was the case in the Vosika [28] or de Bono [27] studies, but disease stabilisation occurred in 3 patients including 2 patients with colo-rectal carcinoma and one patient with non-small cell lung carcinoma. As disease stabilisation was observed at 3 different dose levels including three different doses and number of infusion, and as no patients were included in some dose levels, no trend can be suggested toward the most interesting dose level. Nevertheless, even if these stabilizers have no significant value, they deserve our interest on the mechanisms involved, including the induction of cytokine.

It has to be agreed that lipid A analogues, such as OM-174, do not appear very promising when used alone in cancer patients. Nevertheless our team has been working extensively in animal model and our knowledge of underlying mechanism or tumor growth inhibition led us to assess the possible synergy of OM-174 with different drugs. Like this, we have shown (unpublished data) that whereas treatment by either oxaliplatin or OM-174 alone doesn’t lead to regression of voluminous peritoneal tumors, induced in BDIX rats by intraperitoneal injection of syngeneic PROb colon cancer cells. Their combination induces the disappearance of carcinomatosis.

We have shown that the lipid A analog, OM-174, administered twice weekly, for up to 15 injections and 1000 μg/m² offers a predictable and overall favourable safety profile and good tolerability at biologically active concentrations. Clinical and biological activities deserve further experiments, likely in conjunction with chemotherapy, as preclinical experiments conducted in our unit suggest that this combination might be synergistic. To confirm this hypothesis, we have designed a phase Ib study in association with a derivate of platine that will soon be enrolling patients.