Antitumor Activity in a Human Multiple Myeloma (MM) Model Using Sorafenib, Bortezomib, and Dexamethasone Treatment.

For a better understanding of myeloma disease and biology, the establishment of reproducible in vivo models is pursued worldwide. We have established a cell linebased, disseminated myeloma model in NOD/SCID-IL2-receptor-gamma-chain^−/− (IL2^−/−) mice. In the current study, this model was validated in various treatment groups, using 1. bortezomib (0.7mg/kg/day (d); d0, d4, d11), 2. sorafenib (200mg/kg/d; d0–11), 3. dexamethasone (3mg/kg/d, d0–4 + 7–11), in comparison with 4. a control group. L363 cells were injected intratibialy (it) into IL^2−/− mice and respective therapies were started 7 days after L363-it-injection (d0). Tumor growth was monitored with

• daily monitoring of MM symptoms,

• fluorescence-based in vivo imaging (FI) performed every 2. week and

• flow-cytometry (FACS; detection of human HLA−A, B, C + CD138) performed once weekly by sacrificing 2 mice per group and analyzing bone marrow (BM), spleen, peripheral blood (PB) and liver.

Based on the FACS data, tumor inhibition was calculated as the median percentage of MM cells at respective compartments of the test- vs. control-group multiplied by 100 (optimal test/control (T/C) in %). Furthermore, hollow bones of the injected mice were retrieved when mice were sacrificed for FACS analyses, cells flushed out and MM cells purified by MACS microbeads. Total RNA was isolated from these cells and gene expression profiles will be analyzed using the HG-U133 Plus 2.0 array (Affymetrix) and the Expressionist software (Genedata AG, Basel). L363 engrafted reliably (take rate=100%) at the injection site and in distant organs, such as BM (100%), spleen (38%) and rarely liver (8%). Control mice developed MM symptoms, such as hind limb pareses, weight loss and osteolyses. L363 cells were detected by FACS and FI, not only at injection sites, but also in the BM, hollow bones and spleen. Primary tumor development was markedly reduced by sorafenib (optimal T/C of 23% on d14), as well as with dexamethasone and bortezomib, albeit to a much lesser extend (optimal T/C: 81% + 62% on d14, respectively). BM metastases were also significantly reduced by sorafenib with an optimal T/C value of 67% on d28. Dexamethasone and bortezomib, the latter possibly due to subclinical doses (determined after titration and toxicity experiments), had no relevant influence on BM metastases suppression (97% + 100% optimal T/C on d28, respectively). Thus, L363 engraftment into IL^2−/− is a valuable in vivo model for MM which exhibits high reproducibility, take- and metastases-rates and closely mimics the clinical situation. Collection of whole-body FI data proved to be a time- and animal-saving analysis that allows to closely monitor MM growth. Further investigations will validate the very promising antitumor activity of sorafenib and evaluate the potentially synergistic effect of bortezomib and sorafenib. Amongst others, a detailed characterization of the antitumor activity of both compounds will be provided by the gene expression profile of L363 cells isolated from untreated as well as treated mice. The evaluation of new therapeutic approaches in comparison to standard agents was thus successfully conducted, suggesting that our model serves as a valuable tool in the development of new anticancer strategies.