In pancreatic tumors of mice, radiation causes infiltration by macrophages that acquire an immune-suppressive phenotype that disables T-cell–mediated anti-tumor responses, researchers report in the June issue of Gastroenterology. Blocking colony stimulating factor (CSF1 or MCSF) prevents this process, allowing radiation to slow tumor growth, they show.
Most patients with pancreatic ductal adenocarcinoma (PDA) are treated with chemotherapy, sometimes in combination with radiation therapy, which induces cancer cell death. However, there is controversy over the survival benefit that radiation provides to patients with locally advanced PDA. Although efficacy has been reported for some doses and combinations, benefits are modest and some regimens can produce worse outcomes.
In patients and mouse models of PDA, foci of invasive tumors are almost always surrounded by dysplastic pancreatic intraepithelial neoplasia (PanIN) lesions. Unlike most adenocarcinomas, which are mostly made up of transformed epithelial cells, PDA comprises fibro-inflammatory elements interspersed with islands of neoplastic epithelium.
Although radiation can kill malignant epithelial cells, its efficacy is believed to be limited because it suppresses the anti-tumor immune response.
Lena Seifert et al investigated whether radiation causes inflammatory cells to acquire an immune-suppressive phenotype in mouse models of PDA.
They investigated the effects of radiation in p48Cre;LSL-KrasG12D (KC) mice, which develop PanIN lesions, and p48Cre;LSLKrasG12D;LSL-Trp53R172H (KPC) mice, which develop PDA. They also studied C57BL/6 mice with orthotopic tumors grown from KPC mice. Some mice were given neutralizing antibodies against MCSF—a cytokine that controls the production, differentiation, and function of macrophages.
Seifert et al found that KC mice exposed to radiation developed more advanced PanIN lesions than mice not exposed to radiation, as well as numerous foci of invasive cancer. Pancreata from mice exposed to radiation had dense fibrous and inflamed desmoplasia, and contained macrophages of an M2 phenotype.
M2 macrophages produce immune-suppressive cytokines and induce differentiation of T-helper (Th) 2 and regulatory T (Treg) CD4+ cells. The authors found pancreata from mice exposed to radiation to have fewer CD8+ T cells than controls, and greater numbers of Th2 and Treg cells.
In the KC mice, radiation reduced survival time by more than 6 months. However, radiation did not affect a model of chronic pancreatitis or alter the pancreatic architecture of C57BL/6 (non-KC) mice.
Radiation slowed growth of KPC-derived orthotopic tumors and induced tumor necrosis. However, similar to the KC mice, tumors were infiltrated by M2-type macrophages after radiation.
MCSF expands macrophage populations and promotes differentiation of M2 macrophage. Siefert et al found an approximately 5-fold increase in MCSF expression in KPC orthotopic tumors of mice exposed to radiation, compared with unexposed mice. Immunohistochemical analysis showed increased expression of MCSF by transformed pancreatic ductal cells on day 3 after radiation in pancreata of KC mice.
Adoptive transfer of T cells from irradiated mice with PDAs to tumors of non-irradiated mice with PDAs accelerated tumor growth (see figure). A neutralizing antibody against MCSF prevented radiation from altering the phenotype of macrophages in tumors, increased the anti-tumor T-cell response, and slowed tumor growth. However, despite reducing macrophage infiltration, the antibody against MCSF did not slow tumor growth in the absence of radiation exposure. This could be because in the absence of radiation, PDAs do not contain large numbers of M2 macrophage.
Siefert et al conclude that radiation of PDAs leads to immune suppression within the tumor microenvironment. This occurs via tumor cell production of MCSF, which promotes expansion of immune-suppressive tumor-associated macrophage (M2), resulting in T-cell anergy.
They say their findings are similar to those from mouse models of breast cancer, in which macrophage depletion significantly delayed tumor regrowth after radiotherapy.
The authors noticed that the effects of radiation on recruitment of macrophage and their M2 polarization were only observed at early time points after radiation, and not at 8 weeks, so the stimulus for macrophage recruitment and programming seems to disappear by later time points. Tumor expression of MCSF was reduced to baseline levels by 8 weeks after radiation.
The authors also found radiation to induce up-regulation of PD-L1 on PDA-infiltrating macrophage—another potential mechanism of immune suppression by irradiated PDAs.
Siefert et al propose that strategies to block MCSF might be developed as an adjuvant to increase the efficacy of radiation therapy for patients with locally advanced unresectable PDAs.