Background: CALGB 80405 was a randomized phase III trial comparing FOLFOX or FOLFIRI + Bev, Cetux or both. Results showed no significant differences between Bev and Cetux in progression-free survival (PFS) or overall survival (OS). The primary objective of our study was to validate 7 biomarkers previously identified to be predictive for outcome for either Bev or Cetux. Methods: Baselineplasma samples were analyzed via multiplex ELISA technology for Ang-2, sCD73, sHER3, HGF, IL-6, SDF-1, and VEGF-D. The markers were correlated with PFS and OS using univariate Cox proportional hazards models stratified by chemotherapy (chemo); predictive markers were tested by adding a treatment by analyte interaction term to the model. Exploratory analyses included a chemo by biologic by marker term. Results: Baseline samples were available from 715 KRAS wild type pts treated with Bev or Cetux. No prespecified biomarkers were predictive for OS or PFS after multiple testing correction. However, lower levels of VEGF-D were associated with greater benefit from Bev for PFS (uncorrected interaction p = 0.049; HR 0.69 [95% CI 0.48-1.00]). Based on exploratory analyses, a potential 3-way interaction between chemo, biologic, and VEGF-D levels with respect to PFS was observed (p = 0.028). In subset analyses, VEGF-D was found to predict for benefit from Bev in pts treated with FOLFOX (p = 0.0028; HR 0.50 [95% CI 0.31-0.79]), but not in pts treated with FOLFIRI (p = 0.53; HR 1.27 [95% CI 0.59-2.73]). IL-6 was strongly prognostic for PFS and OS (p < 0.0001). Conclusions: While the pre-specified criteria for validation of these markers was not met, these data suggest a role for VEGF-D in mediating resistance to Bev, particularly for pts receiving FOLFOX. These findings are consistent with prior observations and highlight the potential importance of alternate VEGF ligands in response to anti-VEGF therapeutics. In addition, IL-6 was found to be strongly prognostic, highlighting the importance of inflammation in mCRC. Additional analyses are ongoing. Support: U10CA180821, U10CA180882. ClinicalTrials.gov Id: NCT00265850