Establishment of a Metastasizing Human Cell Line Based Multiple Myeloma Model in Immunodeficient Mice.

Multiple myeloma (MM) accounts for 10% of all hematological malignancies, leads to devastating bone destruction, bone pain, renal failure, and hematopoietic insufficiency. In order to allow a better understanding of MM, the establishment of functional and reproducible in vivo models is pursued worldwide. Of available models, xenograft models in immunodeficient mice (IDM) reproduce the clinical situation best. Using the different MM cell lines (MMCLs) L363 and RPMI8226, we tested their ability to engraft in IDM under different conditions. As the supportive effect of bone marrow (BM) stroma has been suggested to be vital, we injected L363 or RPMI8226 into a freshly prepared mouse tibia which was implanted subcutaneously (sc) into IDM. In a second approach, we injected MMCLs directly into the tibia of the recipient mouse. Furthermore, the influence of IL-6, matrigel, pre-treatment with an anti-mouse CD122-antibody (Ab) and/or the use of NOD/SCID-IL2-receptor-gamma-chain−/− (IL2−/−) mice, instead of conventional NOD/SCID mice, was evaluated. Tumor growth was monitored by a) multiparameter flow-cytometry (FACS; detection of human HLA-DR, HLA-A,B,C, CD45 and CD138), b) immunohistological detection of human CD138+ cells in tumor implants, BM and spleen and c) a fluorescence-based in vivo imaging system. L363 and RPMI8226 engrafted reliably at the injection site and in distant organs, independent of the experimental conditions (using IL-6, matrigel and/or pre-treatment with anti-CD122-Ab). L363 cells showed higher take- and metastases-rates compared to RPMI8226. The knock-down of NK cell activity, either by Ab-treatment or by genetic engineering, enhanced the tumor take rate with both MMCLs in 8/8 mice. With RPMI8226, the BM infiltration rate was 1.5–1.8% in all examined murine models. In contrast, BM-infiltration rates of L363 cells negatively correlated with the NK cell activity of the host: L363 cells metastasized particularly well to the BM when injected into anti-CD122-pretreated mice or intratibialy into IL-2−/− mice. Engraftment of circulating cells into the peripheral blood and spleen was found with both MMCLs irrespective of the mouse strain, pretreatment or implantation site. Matrigel or IL-6 showed no relevant engraftment effect, neither sc, nor with the intratibial approach. We conclude that the establishment of a metastasizing cell line-based human MM in vivo model was successfully pursued. We observed that the knock-down of NK cell activity was essential for both MMCLs, whereas the influence of matrigel or IL-6 treatment could be neglected. Furthermore, the intratibial approach optimized especially the L363 model in terms of infiltration rate to clinically relevant sites. With this optimized transplant approach, we aim to determine whether this allows MM engraftment using primary patient specimens, which should also enable us to test novel anti-myeloma agents.