Hybrid tracer-based synchronization of quantification of cellular receptor-expression and in vivo receptor-targeted imaging

Objectives Current efforts on clinical implementation of personalized receptor-targeted imaging and therapeutic approaches are hampered by the unavailability of a clear strategy on which of the vast amount of receptor targeted biomarker-specific agents should be used during treatment. Synchronization between target quantification and receptor targeted imaging could provide a means to predict the feasibility of in vivo tumor visualization.

Methods For proof of principle the chemokine receptor 4 (CXCR4) was used as a model-receptor. A hybrid imaging label consisting out of a DTPA chelate and a Cy5 fluorescent dye was coupled to a CXCR4 targeting peptide (Ac-TZ14011-DTPA-Cy5, KD = 228.3 ± 20.2 nM). This tracer was used in both in vitro assessment of receptor expression (fluorescence label) in MDAMB231 and MDAMB231 X4 cells and in vivo SPECT (radiolabel) of tumor bearing Balb/c nude mice (n=4/group). Membranous binding was confirmed using a Leica SP8 fluorescence confocal microscope at 1hr post addition of hybrid tracer to the medium of the cells (incubation at 40C); CY5: λex= 633nm/ λem= 650-700nm), intrinsic GFP-tagged CXCR4 in MDAMB231 X4: λex= 488nm/ λem=500-550nm. Membranous receptor expression levels were quantified using a calibration line based on pre-calibrated Cy5-fluorescent beads. In vivo tracer distribution and tumor visibility (tumor-to-muscle (T/M)-ratio) was determined at 24hr post intravenous injection using SPECT imaging of tumor bearing Balb/c nude mice (n=3/cell line; 50ug 111In labelled hybrid CXCR4 peptide). Quantitative assessment of tracer distribution was determined via gamma counting and determination of the percentage of the injected dose per gram (%ID/g) of tissue. Antibody-based immunohistochemical staining was used as a control.

Results The amount of receptors present on the membrane of basal CXCR4 expressing MDAMB231 cells was calculated at 5.7[asterisk]103 ± 82. While 1.5[asterisk]106 ± 2.0[asterisk]105 receptors were shown to be present on the high CXCR4 expressing MDAMB231 X4 cell line. Fluorescence confocal microscopy confirmed the membranous localization of CXCR4 and targeting of hybrid CXCR4 targeting tracer, staining intensity increased with the number of receptors present. In MDAMB231X4 cells discrimination could be made between the stained membranous component of CXCR4 on the membrane and the unstained component of CXCR4 within the cytoplasm, based on co-staining between the Cy5-based tracer staining and the intrinsic GFP signal. In vivo SPECT imaging using an 111In labelled version of the hybrid tracer showed a comparable biodistribution between animals, with the distinction that only the high CXCR4 expressing cells could be visualized; In MDAMB231 X4 tumor bearing mice T/M of 5.95 ± 0.39 was achieved, whereas the T/M ratio in MDAMB231 tumor bearing mice was only 1.02 ± 0.2.

Conclusions Hybrid tracer-based quantification of CXCR4 expression levels prior to SPECT imaging can be used to predict the feasibility of successful in vivo tumor visualization.