Lipid A ONO

As is implied by its name, the first TNF-a-dependent mechanism described was the induction of tumor necrosis in vivo through its role in tumor vasculature. However the mechanisms of the in vitro toxicity of TNF-a to tumor cells imply apoptosis rather than necrosis [97], Tumor necrosis in SCID (severe combined immuno-deficiency) mice treated with LPS does not lead to the rejection of tumors [98], Furthermore, necrosis and tumor regression must be dissociated since anti-IFN-y antibodies inhibit LPS-induced regression of Meth A sarcoma in mice, but not the necrotic hemorrhage attributed to TNF-a. It is now accepted that the antitumoral effect of TNF-a is indirect and dependent on acquired immune response. Matsumoto et al. [99] reported that, while TNF-a itself has no effect on hepatoma KDH-8 tumor cells in vitro, the antitumoral effect of the lipid A ONO-4007 against KDH-8 tumors in vivo is inhibited by anti-TNF-a antibodies in WKAH rat, showing an indirect effect of TNF-a. 

A well-known effect of LPS in mice is mitogenicity to B lymphocytes. Lipid A from P. gingivalis is shown to be mitogenic in vitro even to splenic B cells from LPS-unresponsive C3H/HeJ mice [31]. The lipid A ONO-4007 is mitogenic to spleen cells from BALB/c mice [135], The role of antibodies in tumor regression has solely been documented in the... 

Other cytokines are involved in the effect of lipid A on tumor growth. OM-174 decreases TGF-P mRNA and protein in rat PROb tumors [107]. ONO-4007 increases TNF-a, IL-ip, IL-12, and 11-6 while decreasing TGF-P levels [108]. The secretion of IL-12 by dendritic cells is also induced by microbial products like LPS [109], These cytokines have further effects on other cell types, which could be considered as indirect effects of LPS. In this respect, 11-6 and IL-12 increase LAK induction [110]. 

Here, we will emphasize on treatments with lipid A, considering only curative treatments beginning after tumor cell injection. After a review of the literature, we will detail the effects and mechanisms of DT-5461, ONO-4007 and OM-174, the three lipids A which have been mostly documented. 

In conclusion, various lipid A have been tested as treatments for tumorbearing animals, using different routes (intratumoral, intraperitoneal, intravenous, intradermic). While the intradermic route permits the use of greater doses without toxicity [76,77], most of the studies were performed using several i.v. injections of lipid A. Optimum doses range from 1 to 5 mg/kg for the 3 lipids A DT 5461, ONO 4007 and OM-174 in rats and mice. In our model of colon carcinoma in rats, we showed that the i.v. treatment is more efficient than an intraperitoneal one [81]. In general, most studies showed that the treatments increase survival, or slow the growth of established tumors in mice... 

Lipids A are generally considered to act through TNF-a secretion [112], For instance ONO-4007 was shown to be efficient only on TNF-a-sensitive tumors [171,172], therefore, only well vascularized tumors can be affected. However, in our model, we showed that TNF-a is probably not involved since this cytokine peakes after the first two injections and then returns to basal levels before tumor regression. The efficacy of OM-174 relies on an indirect induction of an autotoxic production of NO by the tumor cells [104], Perhaps a more important aspect is the immunogenicity of tumor cells. After apoptosis or necrosis of tumor... 

In a recent phase I trial the synthetic lipid A analog ONO-4007 was given by i.v. injections to patients with cancer unresponsive to the standard therapy. The limited systemic toxicity disappeared within 24 hours. The MTD was defined as 125 mg/patient [e.g. (125 65=2 mg/kg]. The lipid A increased serum concentrations of TNF-a and IL-6, without affecting the concentrations of GM-CSF, IFN-y and neopterin. There was a significant drop in lymphocyte counts after injections, but no effect on clotting parameters [190]. 

Three lipids A have been more intensively studied in animal models, all of them having indirect effects, mediated in vivo by the immune system. For two of them, DT-5461 and ONO-4007, TNF-a is an important mediator acting at the vascular level that provokes tumor necrosis. For the third one, OM-174, the treatment induces the accumulation of IFN-y and EL-1 P in tumors, which activate NOS II transcription in tumor cells that produce autotoxic NO, which then provokes the apoptosis of tumor cells. At the same time this treatment inhibits the production of TGF-pi by tumor cells which reduces the TGF-pi induced immunosuppression and enhances NO production. Acquired immune response, probably completes the tumor regression started by the apoptosis process and, most probably induces specific memory. 

Several lipids A have been tested in cancer patients MPLA, SDZ MRL 953, and ONO-4007 were injected i.v. in phase I trials. The maximal tolerated dose found is lower than or close to the optimal dose defined in animals. Humans are more sensitive to lipid A than rodents so it is possible that similarly to the toxic dose, the effective dose is lower in humans than in animals. 

Yoshida, K., Shimizugawa, T., Ono, M., and Fumkawa, H. 2002. Angiopoietin-like protein 4 is a potent hyperlipidemia-inducing factor in mice and inhibitor of lipoprotein lipase. J. Lipid Res. 43, 1770-1772. 

---