Use Of Flim Histology-Based Her2-Her3 Heterodimer Quantification And A Bayesian Latent Class Proportional Hazards Model To Predict Cetuximab Response In The Coin And New Epoc Trials.

e14535 Background: Up-regulation of the EGFR-HER3 dimer may limit the efficacy of anti-EGFR treatment in breast cancerand may constitute a pathway of resistance to Cetuximab in colorectal cancer. The question of whether the HER receptor dimers (in different configurations), may account for the poor response to Cetuximab is insufficiently explored in the clinical context. Methods: We developed a new fluorescence lifetime imaging (FLIM)-based histology assay for accurately quantifying the HER2-3 heterodimer in FFPE patient tissues (with FLIM being the gold standard for measuring FRET within the u003c 10nm protein proximity range). A Bayesian latent class proportional hazards model was used to carry out the multivariate analysis on the prediction of survival in two published patients cohorts: [i] advanced/metastatic colorectal cancer patients in the COIN trial (N = 144) and [ii] patients with KRAS wild-type resectable colorectal liver metastases in a neo-adjuvant chemotherapy study (new EPOC) (N = 126). Results:...