During the time while these first observations were being made, the group of J.P. Allison described the antitumor effects of systemic administration of monoclonal antibodies that interfered with the function of CTLA-4 (CD152) (Leach et al., 1996). A side by side comparison with anti-CD137 mAb was made that indicated that the therapeutic activity against tumor mouse models was comparable, with certain superiority for the anti-CD137 in some tumors (Melero et al., 1997a). The obvious attempts at finding additive or synergistic effects have failed both at inducing stronger immune responses and at achieving more effective therapeutic activity. In other words, equal or even less tumor rejections were observed in the combined treatment group when compared to each antibody given separately (A. Arina et al. unpublished observations). The reasons underlying the lack of effectiveness of the combination are still unknown but might be related to the differential in vivo effects reported for anti-CD137 mAb on CD4 and CD8 T cells. With regard to potential toxicity, none of these antibodies appeared to cause any serious adverse effects in mice. However, the striking autoimmune phenotype of CTLA-4-/- mice (Tivol et al., 1995; Waterhouse et al., 1995) warned somehow about the serious autoimmnunity that was eventually reported in early clinical trials with anti-CTLA-4 mAb (Hodi et al., 2003; Phan et al., 2003; Sanderson et al., 2005). The soothing early safety observations with anti-CD137 mAb have received a further boost when it was repeatedly documented that anti-CD137 mAb not only did not induce autoimmunity, but on the contrary, prevented or successfully treated experimental autoimmune conditions in mice (Mittler, 2004; Seo et al., 2004; Sun et al., 2002a, 2002b). This became a puzzling paradox in the field of CD137 research because the observations of immune tumor rejections were difficult to reconcile with the prevention of autoimmune conditions by the very same therapeutic agents. Early observations had suggested that, although anti-CD137 mAb costimulated CD4+ T lymphocytes in culture, in vivo treatment with the antibodies resulted in a decrease of CD4 T cell helper functions, at least for antibody production against certain antigens (Shuford et al., 1997). Several hypotheses to explain this paradox have been raised including the ability of anti-CD137 agonistic antibodies to enhance the proliferation of regulatory T cells (Zheng et al., 2004), as well as the ability of anti-CD137 mAb to induce the immunosuppressive enzyme indoleamine 2,3-dioxygenase (IDO) expression (Seo et al., 2004), through a mechanism dependent on IFN-y overproduction (Seo et al., 2004). Although the existence of these conflicting mechanisms reassures safety in clinical translational research, it provides a rationale to temporarily tamper with these anti-CD137-dependent immunosuppresive mechanisms in order to further enhance antitumor immunity, even at the risk of inducing autoimmunity.
Other specificities of mAb that enhanced antitumor responses were eventually described. Agonistic anti-CD40 mAbs are another set of promising agents of this kind (French et al., 1999). Side by side comparisons with anti-CD137 were difficult to perform and there were large differences depending on the tumor model, although in general anti-CD137 mAbs seemed to be more potent in the tested instances (A. Arina et al. unpublished observations). Unfortunately again, the antitumor effects of anti-CD137 and anti-CD40 mAbs were not mutually enhanced, as it had been anticipated based on the mechanism of action of anti-CD40 mAb that are postulated to act by licensing and activating dendritic cells for full CTL priming (French et al., 1999; Lanzavecchia, 1998).
In contrast to our unpublished observations at the time of going to press two important studies have shown that anti-CD137 mAbs can synergize with anti-CTLA-4 mAb (Kocak et al., 2006) and anti-CD40 (Uno et al., 2006). In the case of anti-CTLA-4 the synergy was observed against the MC38 colon carcinoma even with a human anti-CD40 mAb given to huCD40 knock-in mice. In the case of agonistic anti-CD40 mAbs the synergistic activity with anti-CD137 mAbs has been described against the 4T1 breast carcinoma. The therapeutic effect of the anti-CD137 plus anti-CD40 combination is further enhanced by an anti-DR5 mAb that causes tumor cell apoptosis and presumably increase cross-presentation of tumor antigens.
Was this article helpful?