Reply

Abstract

The results of our study suggest that inactivation of the TGF-β signaling pathway is a common and important event in colon cancer pathogenesis and that mutation of TGF-βRII is the responsible mechanism in one third of all colon cancer cases. The mechanism causing TGF-β resistance in the remaining cases is still unknown and is an important area for active investigation. Mutations in the Smad family members, hSmad2, 3, and 4, may explain a portion of the additional TGF-β–resistant tumors; however, the low incidence thus far detected of hSmad 2, 3, and 4 mutations in human colon cancer implies hSmad mutations alone will not account for all the remaining cases. The results of our study raise the question of what pathophysiological advantage accrues to tumors from inactivating the TGF-β signaling pathway. Work from our group has shown RII mutations develop late in the adenoma to carcinoma sequence concurrent with the acquisition of malignant behavior of the tumor (Cancer Res 1998;58:3101-3104). This observation suggests TGF-βRII mutations promote cancer formation through mechanisms beyond release from TGF-β–induced growth inhibition, perhaps by promoting neoangiogenesis and/or tumor invasiveness. Strikingly, colon cancers often produce significant amounts of TGF-β. In an informative breast cancer model, Yin et al. recently reported that the cancer, although resistant to TGF-β growth inhibition, preserved autocrine TGF-β–mediated induction of PTHrp expression that favored the formation of bony metastatic disease (J Clin Invest 1999;103:197-205). In contrast, we find that colon cancers that are resistant to TGF-β–mediated growth inhibition have also lost TGF-β–induced transcriptional responses. Nonetheless, secreted TGF-β could contribute to the pathogenesis of TGF-β–resistant colon cancer via paracrine effects on neighboring stromal cells that could alter extracellular matrix, induce the growth of new blood vessels, and suppress the local activity of immune cells in the tumor environment. If such a model ultimately is verified, it would paradoxically imply a potential role for TGF-β antagonists as therapeutic agents in the treatment of TGF-β–resistant colon cancer. More directly, the discovery of RII mutations in a substantial portion of MSS as well as MSI colon cancers suggests the opportunity for novel therapeutic interventions such as gene reconstitution with TGF-βRII or the development of pharmaceutical agents that could reactivate postreceptor TGF-β signaling elements in these cancers. Unpublished observations in our laboratory suggest that gene reconstitution strategies may prove most successful in MSS colon cancer, because the MSI tumors rapidly mutate the reconstituted wild-type RII tumor-suppressor gene. Currently, we have developed a mutation-resistant version of wild-type RII and are evaluating its effectiveness as a tumor-suppressor of MSI colon cancer. The design of pharmaceutical agents that could reactivate postreceptor TGF-β signaling will become a potential reality as progress continues to be made in elucidating the gene targets that mediate TGF-β–induced growth inhibition and the mechanisms causing TGF-β resistance in colon cancers in which receptors are still wild-type. In addition to potential new treatment approaches, diagnostic assays that exploit the presence of mutations on RII could be developed. The mutant RII found in MSI colon cancer results from a frameshift mutation in the RII gene that prematurely truncates the protein, resulting in loss of the intracytoplasmic and transmembrane component of the receptor. Consequently, the mutant protein is predicted to be secreted by the tumor cells and should be detectable in the blood. Thus, the identification of RII mutations in colon cancer has not only provided insight into the mechanisms that contribute to the pathogenesis of colon cancer but also offers the promise of new therapeutic and diagnostic strategies to help treat this common malignancy.