A novel biomarker combination and its association with resistance to chemotherapy combinations with bevacizumab: First results of the PERMAD trial

Abstract

Background: Antiangiogenic agents are frequently used in the systemic treatment of metastatic colorectal cancer (mCRC). Since these strategies target not the tumor but the tumor associated endothelium establishing positive or negative predictive biomarkers is challenging. Recent research indicated a potentially predictive value of cytokines and angiogenic factors (CAF) for early detection of progression during treatment with chemotherapy and bevacizumab. The PERMAD trial is a two-phase trial with the primary objective to evaluate of the impact of a personalized, marker-driven treatment approach with early detection of progression and modification of treatment. The aim of the first phase is to establish a combination of CAFs that allows early detection/prediction of disease progression (PD) under a treatment with FOLFOX plus bevacizumab (Bev). The second phase is a randomized part with marker-driven switch of anti-angiogenic agent and maintenance of the chemotherapy backbone until definite radiological PD compared to a conventional treatment approach of changing chemotherapy and antiangiogenic agent at time of radiologic PD. Here we report the results from the first phase examining samples of 50 patients. Method(s): During the run-in phase 102 CAFs were established in blood samples taken prior to treatment and q2w thereafter until PD. Using machine learning we aimed at establishing a combination of 5 out of the 102 CAFs that fulfilled these criteria: cytokines should be affected by PD, but not by treatment itself and should indicate PD at least 2 months prior to the time of PD as determined by MD-CT which was performed every 2 months. Result(s): Using our machine learning approach we could establish a combination of markers from 30 patients that is associated with > 80% accuracy with PD 2 months prior to radiological PD under a treatment with FOLFOX plus Bev. This combination will be corroborated in another 20 patients (data presented at the meeting) and used for the second, randomized part of the trial. Conclusion(s): Using advanced bioinformatics we have identified a biomarker combination that is associated with subsequent PD with a high accuracy under a treatment with FOLFOX plus Bev.