Influence of stromal-epithelial interactions on breast cancer in vitro and in vivo.

Stromal cell-secreted chemokines including CCL2 have been implicated in the primary tumor microenvironment, as mediators of tumor cell migration, proliferation, and angiogenesis. Expression of CCL2 and its principal receptor CCR2 was analyzed by RQ-PCR in primary tumor cells and breast cancer cell lines. Breast cancer cell lines (MDA-MB-231, T47D) were co-cultured directly on a monolayer of primary breast tumor and normal stromal cells, retrieved using EpCAM+ magnetic beads, and changes in expression of CCL2, CCR2, MMP11, ELK1, VIL2, and Ki67 detected by RQ-PCR. Epithelial cell migration and proliferation in response to stromal cell-secreted factors was also analyzed. In vivo, tumor xenografts were formed by co-injecting T47D cells with primary tumor stromal cells. Following establishment, tumors were harvested and digested, epithelial cells retrieved and analyzed by RQ-PCR. Whole tumor tissue was also analyzed by immunohistochemistry for CD31 and the VIL2 encoded protein Ezrin. Tumor stromal cells expressed significantly higher levels of CCL2 than normal cells, with no CCR2 expression detected. Primary epithelial cells and breast cancer cell lines expressed elevated CCL2, with relative expression of CCR2 found to be higher than the ligand. Interaction of breast cancer epithelial cells with primary tumor, but not normal stromal cells, stimulated increased expression of CCL2 (8-fold), ELK1 (6-fold), VIL2 (6-fold), and MMP11 (17-fold). Factors secreted by stromal cells, including CCL2, stimulated a significant increase in epithelial cell migration, with no effect on cell proliferation in vitro observed. In vivo, the presence of stromal cells resulted in tumors of increased volume, mediated at least in part through neoangiogenesis demonstrated by immunohistochemistry (CD31). Admixed tumor xenografts exhibited increased expression of Ki67, MMP11, VIL2, and ELK1. Elevated Ezrin protein was also detected, with increased cytoplasmic localization. The results presented highlight mechanisms through which breast cancer epithelial cells can harness stromal cell biology to support tumor progression.

C-reactive protein as a predictor of low trough infliximab concentrations in patients who lose response to infliximab.

OBJECTIVE: Low serum infliximab concentrations are associated with an increased risk of loss of response in inflammatory bowel disease (IBD). The objective of this study was to evaluate the test characteristics of C-reactive protein (CRP) in identifying low serum infliximab concentrations in patients with IBD. METHODS: We measured serum infliximab concentrations and CRP levels in patients who experienced deteriorating symptoms while on infliximab (the reactive cohort). Receiver operating characteristic (ROC) curves were used to determine the CRP concentration threshold that identified an infliximab concentration <3 µg/mL at the time of loss of response. These CRP thresholds for infliximab concentration <3 µg/mL were then tested in a separate validation cohort. RESULTS: The reactive cohort contained 111 patients and the validation cohort contained 139 patients. In 41% of participants, serum infliximab concentration was <3 µg/mL. In the reactive cohort, the area under the ROC curve for CRP to identify an infliximab concentration <3 µg/mL was 0.70 (95% confidence interval [CI] 0.50-0.80, P = 0.02). A CRP level above 12 mg/L in the preceding 90 days provided a 90% specificity for the later detection of infliximab concentration <3 µg/mL. These test characteristics were similar in the validation cohort. CONCLUSION: CRP levels over 12 mg/L exhibit a high specificity for identifying patients with an infliximab concentration <3 µg/mL. CRP may be cost-effective for identifying patients with low concentrations of infliximab at the time of, or at risk of, loss of response.