Treatments for rheumatoid arthritis have come a long way over the past decade - they are "less toxic, better tolerated and more effective" than older drugs, which essentially were carefully dosed poisons, Bassil Dahiyat told BioWorld Today. "But they do come at a cost - they suppress the immune system more indiscriminately than you'd like."

Fighting autoimmune disease such as rheumatoid arthritis without maiming the immune system's regular function is, unsurprisingly, a tricky task. But in a paper in the Aug. 1, 2007 issue of the Journal of Immunology, scientists at Xencor, where Dahiyat is president and CEO, and of the Centre National de la Recherche Scientifique in Orleans, France, reported success in selectively ameliorating the symptoms of arthritis in mice without rendering them more susceptible to infection.

Tumor necrosis factor, or TNF, is the target of several newer rheumatoid arthritis drugs, including the blockbusters drugs Enbrel (etanercept, Amgen Inc.), Remicade (infliximab, Centocor Inc.) and Humira (adalimumab, Abbott).

TNF comes in two forms: soluble TNF travels the bloodstream to induce inflammatory responses, while transmembrane TNF remains attached to the cell that produced it, acting more locally to coordinate innate immune system functions. Current anti-TNF therapies on the market target both forms.

In the studies published in the Journal of Immunology, the researchers used a so-called dominant negative form of TNF to selectively inhibit the soluble form of TNF while leaving the transmembrane version unaffected. TNF consists of three subparts, and those parts can separate and reassemble in different configurations, like dancers in a group dance. But when dominant negative TNF joins, the dance is soon over. The molecule is mutated in such a way that TNF containing a DN-TNF subpart cannot bind to the TNF receptor.

Using both mouse and human cell cultures, the scientists found that DN-TNFs selectively swapped subunits with soluble TNF, and subsequently blocked the activation of enzymes and cytokine release that occurs when soluble TNF binds to its receptors. DN-TNF also reduced inflammation in two separate mouse models of arthritis.

In contrast, transmembrane TNF remained unaffected by the DN-TNF, and consequently, so did innate immunity: Mice treated with DN-TNF and then exposed to Listeria monocytogenes bacteria were no more likely to succumb to infection than controls.

The data presented in the Journal of Immunology suggested, as the authors wrote, that "maintenance of transmembrane TNF activity may improve the therapeutic index of future anti-inflammatory agents."

Dahiyat noted that on the basic science side, the data also might explain a rare but puzzling side effect of TNF-alpha blockers: "They can cause other autoimmune diseases, such as lupus and demyelinating diseases, which are not exactly effects one would expect from an immunosuppressant." But transmembrane TNF, he said, might have a more generalized regulatory role rather than a simple activating one.

Why transmembrane TNF is impervious to the mutated DN-TNF is not clear, but Dahiyat said that the anchoring to the membrane may confer stability on the membrane-bound version so that it is less likely to dissociate and mix subunits with the DN-TNF.

Dahiyat said that Xencor is "on the cusp" of starting clinical trials with DN-TNF. The company has an open IND, and is "seeking partnerships to enable clinical development."

Initial trials will be in the area of rheumatoid arthritis, which he termed "pretty clearly the best understood of the autoimmune diseases."

Dahiyat said he believes that DN-TNF ultimately could find roles in the treatment of other autoimmune diseases, besides rheumatoid arthritis, and even other indications beyond autoimmune diseases.

He named oncology and graft-versus-host disease as two areas that could benefit from specific targeting of soluble TNF.

"Because it is selective, DN-TNF potentially opens up new indications altogether," Dahiyat added.