Dynamical Models of Cetuximab Resistance in HNSCC Based on Serial Genomics Data

描述（由申请人提供）：头颈部鳞状细胞癌（HNSCC）是全球第六大常见癌症，尽管采用了联合治疗方式，但治愈率仅为50%。表皮生长因子受体（EGFR）的治疗靶向提高了一个子集的患者的生存，虽然敏感性的分子预测目前还很难。此外，有反应的患者往往会产生耐药性，最终死于疾病。区分在细胞信号传导过程和随机进化压力中复杂的串扰中驱动这种治疗抗性的特定分子过程需要从串行数据建立动态模型。因此，在本申请中，我们开发了新的计算算法来从西妥昔单抗耐药的HNSCC的体外和体内模型中推断西妥昔单抗耐药的分子机制。具体而言，我们将研究以下假设：（1）短期时间过程数据提高了硅中建模技术推断西妥昔单抗的靶向和脱靶信号应答的能力;（2）HNSCC细胞在长期暴露于西妥昔单抗后的表观遗传、转录后和基因组变化导致获得性耐药;和（3）对个体间和个体内异质性进行建模将辨别特定的细胞信号传导过程，所述细胞信号传导过程被激活以在HNSCC的细胞系异种移植模型中驱动体内获得的西妥昔单抗抗性。该项目的结果将最终有助于选择西妥昔单抗治疗的患者和克服获得性西妥昔单抗耐药的替代分子靶点。开发的算法也将直接适用于推断其他癌症中治疗抗性的分子驱动因素。

DESCRIPTION (provided by applicant): Head and neck squamous cell carcinoma (HNSCC) is the sixth most frequent cancer worldwide, with only a 50% cure rate in spite of combined treatment modalities. Therapeutic targeting of the epidermal growth factor receptor (EGFR) improves the survival in a subset of patients, although molecular predictors of sensitivity are currently elusive. Moreover, responsive patients often acquire resistance and ultimately succumb to their disease. Distinguishing the specific molecular processes that drive such therapeutic resistance amid complex cross-talk in cell signaling processes and stochastic evolutionary pressures requires dynamical models built from serial data. Therefore, in this application, we develop novel computational algorithms to infer the molecular mechanisms underlying cetuximab resistance from in vitro and in vivo model of cetuximab resistant HNSCC. Specifically, we will investigate the hypotheses that: (1) short-term time course data improve the ability of in silicon modeling techniques to infer both on- and off-target signaling responses to cetuximab; (2) combined epigenetic, post-transcriptional, and genomic changes in HNSCC cells upon chronic exposure to cetuximab result in acquired resistance; and (3) modeling inter and intra-individual heterogeneity will discern the specific cellular signaling processes that are activated to drive in vivo acquired cetuximab resistance in cell- line xenograft models of HNSCC. The results from this project will ultimately contribute to the selection of patients for cetuximab treatment and alternative molecular targets to overcome acquired cetuximab resistance. The algorithms developed will also be directly applicable to inference of molecular drivers of therapeutic resistance in additional cancers.