Establishment of an in vitro 3D model for neuroblastoma enables preclinical investigation of combined tumor-stroma drug targeting.

The majority of anti-cancer therapies target the proliferating tumor cells, while the tumor stroma, principally unaffected, survives, and provide a niche for surviving tumor cells. Combining tumor cell and stroma-targeting therapies thus have a potential to improve patient outcome. The neuroblastoma stroma contains cancer-associated fibroblasts expressing microsomal prostaglandin E synthase-1 (mPGES-1). mPGES-1-derived prostaglandin E-2(PGE(2)) is known to promote tumor growth through increased proliferation and survival of tumor cells, immune suppression, angiogenesis, and therapy resistance, and we, therefore, hypothesize that mPGES-1 constitutes an interesting stromal target. Here, we aimed to develop a relevant in vitro model to study combination therapies. Co-culturing of neuroblastoma and fibroblast cells in 3D tumor spheroids mimic neuroblastoma tumors with regard to the cyclooxygenase/mPGES-1/PGE(2)pathway. Using the spheroid model, we show that the inhibition of fibroblast-derived mPGES-1 enhanced the cytotoxic effect of doxorubicin and vincristine and significantly reduced tumor cell viability and spheroid growth. Cyclic treatment with vincristine in combination with an mPGES-1 inhibitor abrogated cell repopulation. Moreover, inhibition of mPGES-1 potentiated the cytotoxic effect of vincristine on established neuroblastoma allografts in mice. In conclusion, we established a 3D neuroblastoma model, highlighting the potential of combining stromal targeting of mPGES-1 with tumor cell targeting drugs like vincristine.