Abstract
BACKGROUND: Approximately 5% of melanoma pts develop LMDz. There are essentially no models of LMDz available for therapeutic development. Here we report, the in-vitro & in-vivo culturing of CSF-CTCs. METHODS: CSF-CTCs were detected by the Veridex CellSearch® System. Cell-free DNA and cell-associated DNA were extracted, sequenced and profiled. Expanded ex-vivo CSF-CTCs were grown in-vitro and tested for drug sensitivity. CSF-CTCs were grown successfully in-vivo from 1 pt; labeled human Braf V600E WM164 cells were injected IT in as a control. RESULTS: CSF-CTCs: 12 LMDz pts and 8 melanoma pts without LMDz were studied. All but 1 LMDz pts (92%) had CSF-CTCs (avg: 2148.60; range 23 - 3055 CTCs/ml). In contrast, 3/8 (37%) melanoma Brain Mets pts without LMDz had CSF-CTCs but fewer of them (avg: 0.31; range 0.13 - 0.6 CTCs/ml CSF). CSF-CTCs Profile: These had BrafV600E (83%), and GNAQ Q209P & NRAS Q61R in 1 pt each. Ex-vivo culture of CSF-CTCs and PDX model: After lengthy optimization of conditions we successfully expanded CSF-CTCs in-vitro (~25% of pts), and in-vivo in immunodeficient mice from 1 pt (~10% of samples). Ceritinib, used as a FAK inhibitor, with MEKi was effective in-vitro (p=3.17e-6) and prolonged survival in-vivo in LMDz (median survival: >32 days vs control: 18 days; p=7.81e-5). CONCLUSIONS: Though the sample size is small, this is the first report of the successful in-vitro & in-vivo culture of CSF-CTCs from pts with LMDz. Single cell analysis to determine how representative these models are and further in-vivo testing are in progress.