Last month's publication of a review on the role of the immune system in tumor development has revealed a persistent divide between some groups of the cancer research community. The long-standing disagreement centers on the ability of the immune system to intercept tumors of non-viral origin.

Model of elimination phase of cancer immunoediting process Credit: Printed with permission from Nature Immunology 3, 991–998 (2002) Dunn et al doi:10.1038/ni1102-991

Several animal studies in the 1970s failed to show that immunosurveillance protects against the effects of non-viral carcinogens, and the immune system lost much of its status as a protector against tumor formation. But Robert Schreiber, whose review in Nature Immunology (3, 991; 2002) combines data from his laboratory with other evidence from around the world, says that the mice in those studies were not truly immunodeficient. “Experiments were done with mice that had spontaneous mutations affecting the immune system, but which did not delete it,” says Schreiber. “That could not have been known back then.”

Using knockout mice, Schreiber, professor of pathology at Washington University School of Medicine, and his colleagues have now shown that cancers occur more frequently in immunodeficient animals than in those with a largely intact immune system. Moreover, it seems that the immune system actually plays a role in shaping the type of tumor formed, because tumorigenic cells undergo natural selection to develop proteins that evade immunosurveillance, a process that Schreiber calls 'immuno-editing'.

Not everyone is convinced. Although cancer expert Robert Weinberg of the Whitehead Institute agrees that immunosurveillance protects against cancers with a viral etiology—such as non-Hodgkin lymphoma, Kaposi sarcoma and cancers of the genitourinary system—he doubts that other tumors are recognized as foreign by the immune system. “Virtually all the proteins made by cancer cells are normal proteins,” he says. “Whereas viruses are invaders.”

Schreiber hopes his findings will stimulate new research into the phenomenon. “We have to realize that when we try to attack tumors, we should understand that they have already undergone shaping by the immune system,” he says. “We need to boost our immunity very strongly in order to react against tumors with reduced immunogenicity.”