In vivo E-cadherin expression assessed with radiolabeled E-cadherin antibodies: a potential target to image epithelial-to-mesenchymal transition in vivo.

Objectives The epithelial cell-to-cell adhesion molecule cadherin 1 (CDH1), also known as epithelial cadherin (E-cadherin), is a transmembrane glycoprotein that is expressed in tight junctions between normal epithelial cells. Many ductal breast cancers anarchically overexpress E-cadherin, which might expose this protein as an imaging target. The loss of E-cadherin expression is linked to mesenchymal transition and propensity to metastasize, yet E-cadherin can be re-expressed in metastases. Normal expression in normal tissues might create high background activity, which could preclude the use of this protein as an imaging target. The goal of this study was to evaluate the feasibility of imaging E-cadherin expression in tumors in vivo, using antibodies that cross-react to human and mouse E-cadherin.

Methods Western blot and flow cytometry analyses were performed to assess E-cadherin expression in breast cancer cells with different anti E-cadherin antibodies (anti E-cad) to identify suitable antibody candidates. Female mice were inoculated with E-cadherin expressing cell lines following the implantation of an estrogen pellet. Anti-E-cad (Cell Signalling 24E10 or ebioscience DECMA-1; 50ug/mouse; reacting to both mouse and human E-cadherin) were conjugated with p-SCN-Bn-DTPA and radiolabeled with ¹¹¹InCl₃, purified with a 50 kDa filter and verified via size exclusion HPLC. Micro SPECT/CT imaging was performed at 1, 3 and 5 days post injection of the ¹¹¹In-DTPA-anti-E-cad followed by bio-distribution at day 3 and 5.

Results E-cadherin expression was confirmed with western blot and flow cytometry in Zr-75, MCF-7, T47D and MDA-MB-468 breast cancer cell lines with 3 anti-E-cad. ¹¹¹In- DTPA-anti-E-cad radiolabeling was confirmed with SEC HPLC analysis. SPECT/CT Images showed uptake at 24 h in the respective tumours; although tumour uptake decreased over the course of the study. High tumour to background contrast was maintained. For the 24E10 antibody day 5 bio-distribution data of tumour uptake was 8.87 ± 4.30, 11.35 ± 2.19 and 5.09 ± 1.17 %ID/g for Zr-75, MCF-7 and T47-D tumours, respectively. Tumour-to-muscle ratios increased to 11.99 ± 4.66, 25.47 ± 5.60 and 11.43 ± 2.76 for the respective tumours from day 3 to 5. The DECMA-1 antibody did not fare as well with day 5 tumour uptake of 1.64 ± 0.48, 3.13 ± 0.40 and 2.12 ± 0.43 %ID/g for Zr-75, MCF-7 and T47-D tumours, respectively. Although tumour uptake was not as impressive, tumour-to-muscle ratio was maintained from day 3 to day 5 with ratios of 10.05 ± 1.31, 6.74 ± 0.59 and 3.52 ± 0.87 for the respective tumours at day 5.

Conclusions In vivo E-cadherin expression was assessed using two ¹¹¹In radiolabeled anti-E-cadherin antibodies in three different E-cadherin expressing breast cancer xenografts. High tumour to background ratio suggests E-cadherin imaging is possible in vivo, and further studies are underway to assess the specificity of antibody uptake.