MENU

TGF beta drives immune evasion in genetically reconstituted colon cancer metastasis

Tauriello, Daniele V. F.; Palomo-Ponce, Sergio; Stork, Diana; Berenguer-Llergo, Antonio; Badia-Ramentol, Jordi; Iglesias, MAR; Sevillano, Marta; Ibiza, Sales; Canellas, Adria; Hernando Momblona, Xavier; Byrom, Daniel; Matarin, Joan A.; Calon, Alexandre; Ri

NATURE
2018
VL / 554 - BP / 538 - EP / +
abstract
Most patients with colorectal cancer die as a result of the disease spreading to other organs. However, no prevalent mutations have been associated with metastatic colorectal cancers(1,2). Instead, particular features of the tumour microenvironment, such as lack of T-cell infiltration(3), low type 1 T-helper cell (T(H)1) activity and reduced immune cytotoxicity(2) or increased TGF beta levels(4) predict adverse outcomes in patients with colorectal cancer. Here we analyse the interplay between genetic alterations and the tumour microenvironment by crossing mice bearing conditional alleles of four main colorectal cancer mutations in intestinal stem cells. Quadruple-mutant mice developed metastatic intestinal tumours that display key hallmarks of human microsatellite-stable colorectal cancers, including low mutational burden(5), T-cell exclusion(3) and TGF beta-activated stroma(4,6,7). Inhibition of the PD-1-PD-L1 immune checkpoint provoked a limited response in this model system. By contrast, inhibition of TGF beta unleashed a potent and enduring cytotoxic T-cell response against tumour cells that prevented metastasis. In mice with progressive liver metastatic disease, blockade of TGF beta signalling rendered tumours susceptible to anti-PD-1-PD-L1 therapy. Our data show that increased TGF beta in the tumour microenvironment represents a primary mechanism of immune evasion that promotes T-cell exclusion and blocks acquisition of the THI-effector phenotype. Immunotherapies directed against TGF beta signalling may therefore have broad applications in treating patients with advanced colorectal cancer.

AccesS level

Green accepted

MENTIONS DATA