Diversity Supplement to Expansion of Tumoroid Models for Precise Treatment of the Rectal Cancer Patient

点击翻译按钮获取中文摘要

We propose to develop rectal cancer organoids (tumoroids) as individualized models and to build a large rectal cancer tumoroid repository. Research on rectal cancers is hampered by the paucity of models. Of the few existing in vivo models of rectal cancer, none place the tumors in the rectal lumen, so the models fail to mimic the correct anatomic environment and local invasion. The existing models also have not been observed to metastasize. Another problem is that we lack accurate means to predict whether individual rectal cancer patients will respond to chemotherapy or radiation, both of which are part of the current standard of care. We believe that both the paucity of models and the lack of predictive tools can be addressed by patient-derived tumoroids. Tumoroids can be grown in 3-dimensional culture ex vivo or implanted into mice, so they offer a flexible research platform. In preliminary results, we have derived tumoroid lines from multiple patients' rectal cancers and found them to resemble the corresponding patient tumors. The tumoroids, when implanted in mice endoluminally (i.e. in the rectum), formed locally invasive tumors capable of metastasis. Moreover, we found tumoroids to have clinically relevant responses to chemotherapy and radiation. Thus, drawing from these preliminary data, we hypothesize that rectal cancer tumoroids mirror the traits of their original tumors, can be used to predict patients' response to therapy, and, when implanted endoluminally into mice, can serve as an optimal model of rectal cancer. We plan to develop 100 new tumoroids, which we expect to encompass much of the diversity of human rectal adenocarcinoma. The tumoroids will be analyzed in ex vivo culture and in two mouse models: the endoluminal implantation model and a conventional flank injection model. In these settings, we will test whether the tumoroid accurately reflects its tumor of origin in terms of mutations, histology, and gene expression. We will determine whether response of the tumoroids to patient-specific chemotherapy and radiation can predict the corresponding patient's response. Of particular interest is whether individual human rectal cancers are more accurately modeled by endoluminal implantation than by flank injection. Finally, to integrate our findings into a comprehensive platform for broad use, we will develop a rectal cancer tumoroid biorepository seamlessly integrated with online pathologic, genomic, and model-specific information. The online platform will be built within our institution's cancer genomics portal, then integrated into the NCIP Hub. We have assembled a collaborative team with expertise in colorectal surgical oncology, radiation oncology, and pathology; organoids; mouse models; biostatistics; and bioinformatics. We anticipate that the proposed research will credential tumoroids as accurate models for rectal cancer research and for predicting patient responses to therapy. The large tumoroid biorepository is likely to stimulate research on new treatments for rectal cancer. The ultimate result will be new treatment options and better treatment selection for patients affected by this deadly disease.