Establishment and characterization of residual breast cancer patient-derived xenografts resistant to neo-adjuvant therapy.

Background: In HER2 positive and triple-negative breast cancer subgroups, residual disease after neoadjuvant therapy is associated with higher risk of metastatic recurrence compared to patients achieving a pathological complete response. Residual tumor analysis after neoadjuvant treatment is a major and under-explored field to identify resistance mechanisms. To develop patient-derived xenografts (PDX) of residual breast cancer we started a program of residual tumor engraftment in nude mice, following the same procedures previously published for PDX of human breast cancer (Marangoni et al, 2007 and Reyal et al, 2012). Methods: 26 residual breast tumors and 2 residual metastatic axillary lymph nodes were engrafted in swiss nude mice immediately after surgery. Expression of Ki67, HER2, PTEN, P-AKT, P-S6, MET, RET and KIT were analyzed in xenografts by immunohistochemistry, western blot and RT-PCR analyses. Brain, lungs, liver and bones of xenografts were systematically formalin-fixed to search for human metastasis. The in vivo drug response of established xenografts was determined for the following treatments: adryamicin+cyclophosphamide (AC), docetaxel, capecitabine, cisplatin, irinotecan, everolimus, trastuzumab and lapatinib (for the HER2+ PDX). PDX tumors were additionally mechanically dissociated to establish cell lines. Results: Seven PDX were established (tumor take of 25%), 5 triple-negative and 2 HER2+. Six out of seven PDX were metastatic in the lungs. Two xenografts were established from lymph node metastasis. The in vivo drug responses were concordant with the response to neo-adjuvant treatments in patients. Histological analyses showed that xenografts’ tumors recapitulated the patients’ tumor morphology. Residual tumor xenografts expressed high level of Ki67 protein and tumor latency during the first tumor passages was found to be shorter when compared to tumor latency of non pre-treated breast cancers. In 5/5 triple-negative breast cancer PDX the PTEN protein was lost and the PI3 kinase pathway activated. The mTOR inhibitor Everolimus was tested in 2 triple-negative PDX: one was resistant and one was responding, with a tumor growth inhibition of 80%. Triple-negative PDX show expression of “druggable” tyrosin kinase receptors (MET, RET, KIT) providing relevant models to test new target therapies in these models. One cell line was established from a highly metastatic triple-negative breast cancer xenograft. When re-injected into mice, the cell line was tumorigenic, however the tumor architecture was changed and the xenograft was not metastatic. Conclusions: we have established a panel of metastatic PDX models of breast cancer resistant to neo-adjuvant therapies. These models provide a valuable preclinical tool to investigate mechanisms of resistance to neo-adjuvant treatments and for the preclinical testing of new targeted agents. Citation Information: Mol Cancer Ther 2013;12(11 Suppl):A9. Citation Format: Elisabetta Marangoni, Dalila Labiod, Franck Assayag, Rania El Botty, Rana Hatem, Sophie Richon, Sophie Chateau-Joubert, Marine Carlus, Helene Bonsang-Kitzis, Alice Pinheiro, Cecile Laurent, Ivan Bieche, Fabien Reyal. Establishment and characterization of residual breast cancer patient-derived xenografts resistant to neo-adjuvant therapy. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2013 Oct 19-23; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2013;12(11 Suppl):Abstract nr A9.