Bone invading NSCLC cells produce IL-7: mice model and human histologic data

We used NOD/SCID mice implanted with human bone. After bone engraftment, two groups of mice were injected subcutaneously with A549, a human NSCLC cell line, either close or at the contralateral flank to the human bone implant, while a third control group did not receive cancer cells. Tumor and bone vitality and IL-7 expression were assessed in implanted bone, affected or not by A549. Serum IL-7 levels were evaluated by ELISA. IL-7 immunohistochemistry was performed on 10 human bone NSCLC metastasis biopsies for comparison.At 12 weeks after bone implant, we observed osteogenic activity and neovascularization, confirming bone vitality. Tumor aggressive cells implanted close to human bone invaded the bone tissue. The bone-aggressive cancer cells were positive for IL-7 staining both in the mice model and in human biopsies. Higher IL-7 serum levels were found in mice injected with A549 cells close to the bone implant compared to mice injected with A549 cells in the flank opposite to the bone implant.We demonstrated that bone-invading cells express and produce IL-7, which is known to promote osteoclast activation and osteolytic lesions. Tumor-bone interaction increases IL-7 production, with an increase in IL-7 serum levels. The presented mice model of bone invasion by contiguous tumor is suitable to study bone-tumor cell interaction. IL-7 plays a role in the first steps of metastatic process.Lung cancer is a major cause of cancer-related deaths [1]. More than 90% of deaths due to lung cancer can be attributed to metastases [2], which are frequently observed at the time of diagnosis. After liver and brain, bone represents the most common target organ [3]. Osteolytic bone metastases worsen prognosis, causing morbidity [4] and also having substantial financial implications for the health-care providers [5]. The low success rate in treating lung cancer depends on a still incomplete understanding of the metastatic process and on the lack of sensitive markers to predict and mon