br Results Patient disease and treatment characteristics of
Results Patient, disease, and treatment characteristics of the study cohort are described in Table 1. All patients were stage IIIA with pathologically documented N2 disease per AJCC 7th edition. EP4 nuclear staining 0–1 versus 2+ was significantly associated with overall survival (OS) (44.3 vs. 18 mo; HR=0.41, p=0.024) and numerically superior progression free survival (PFS) (16.4 vs. 10.2 mo, p=0.16), Fig. 1. EP4 cytoplasmic staining did not correlate with OS (0–1 vs. 2+, 23.8 vs. 28.8 mo; HR=1.2, p=0.81). We explored whether there was an association of EP4 nuclear staining with histology. Only three of ten squamous cell specimens stained 2+ while 14 of 31 non-squamous specimens had 2+ staining, (p=0.48, Fisher’s exact test). There was an association of histology with overall survival, favoring non-squamous squamous (33.7 vs. 13.5 months, p=0.064 log rank, p=0.035 Wilcoxon). The differences in outcome by histology and the relatively small number of squamous cell patients, led us to perform an analysis restricted to the nonsquamous subset. EP4 nuclear staining 0–1 versus 2+ was again significantly associated with OS, 61.7 vs. 19 months, HR=0.2 (p=.0006) as well as numerically superior PFS (16.4 months vs. 10.2 months, HR=0.61, p=0.17). There was also a numerical, though not a statistically significant advantage for the patients who underwent surgery for those who did not in terms of OS (44 vs. 13.8 months, HR=0.5, p=0.07). There were no differences in the distribution of patients with tumors with EP4 nuclear staining who underwent surgery vs. those who were treated non-operatively. Of the 21 patients who underwent surgery, 12 had tumors with 0–1 nuclear EP4 expression and 9 had tumors with 2–4 expression. Of the 20 patients who were treated non-operatively, 12 had tumors with 0–1 expression and 8 had tumors with 2–4 expression. Given the potential role of EP4 in promoting metastasis, we evaluated the patterns of relapse. This was scored as no relapse or locoregional (at the site of the primary, ipsilateral lung or mediastinal lymph nodes) vs. systemic (contralateral lung, extrathoracic disease) relapse. There was no correlation of relapse pattern with EP4 nuclear staining (p=1.0, X2).
Discussion PGE2 induces migration of tumor pkc inhibitor in vitro and endogenous EP4 signaling promotes metastasis in vivo. EP4 is highly expressed on human and mouse breast cancer cell lines. Studies have shown that metastasis is inhibited in both syngeneic and xenograft breast cancer models by selective EP4 antagonists (AH23848, ONO-AE3-208, RQ15986, Frondoside) , , , , , , . Additionally, EP4 gene silencing by EP4 shRNA can inhibit lung colonizing ability . In addition to preclinical models of breast cancer, investigators have evaluated animal models of lung and colon cancer. Yang et al. found that EP4 antagonism significantly reduced metastatic disease in both malignancies . Mice were injected with i.v. Lewis lung carcinoma (3LL) and then treated with ONO-AE3-208, a specific EP4 antagonist or when EP4 expression was reduced with RNA interference, resulted in a significant reduction in tumor growth and dissemination. Conversely, treatment with an EP4-specific agonist (AE1-734) caused an increase in these same tumor cell traits. Given the recent clinical validation of immune checkpoint inhibition in NSCLC, it is important to note that EP4 receptors on both tumor and immune cells may result in a tumor enhancing immunologic environment. Natural killer (NK) cells play a major role in cytokine release, migration, and cytolytic activity. Studies have shown that these functions can be inhibited by PGE2 since NK cells have EP4 receptors , , . Furthermore, it has been demonstrated that NSCLC patient have high levels of myeloid derived suppressor cells (MDSCs). MDSCs suppress the immune system through multiple mechanisms including the stimulation of T-regulatory cells, drive to type-2 tumor-promoting phenotype, and inhibition of CD4+ and CD8+ T cells and NK cytotoxicity ,  PGE2 interacts with receptors including EP4 on MDSC precursors which stimulates the differentiation and growth of MDSCs ,  Zhang et al. showed that PGE2 also plays a role in Fas signaling which recruits MDSCs, thereby promoting tumor growth in lung cancer . Depletion of MDSCs has been demonstrated in animal models to enhance the efficacy of anti-tumor therapy, including vaccines and chemotherapy . These cells have been shown to inhibit T-cell proliferation and cytotoxic lymphocytes in an MHC- and antigen-independent manner . Myeloid cells expressing similar markers in humans have been found to be increased fivefold in patients with head and neck squamous cell cancer, renal cell cancer, breast cancer, and NSCLC . Depletion of MDSCs has been shown to reduce tumor progression and to improve immune-based cancer therapies , .