Prospective validation of Ang-2 and Tie-2 plasma levels as predictors of benefit from regorafenib in metastatic colorectal cancer patients: REGOLAND study

Background Regorafenib is a treatment option for refractory metastatic colorectal cancer (mCRC) patients. The identification of markers to predict or monitor the efficacy of regorafenib represents a key issue in this palliative setting. In a previous exploratory study, the early increase of Ang-2 plasma levels during treatment with regorafenib seemed to predict benefit from this agent and low baseline Ang-2 and Tie-2 plasma levels were associated with good prognosis. REGOLAND is a translational study, which aims to prospectively validate these retrospective findings. Methods Ang-2 and Tie-2 were assessed by ELISA on plasma samples collected at baseline (d1) and after 15 days (d15) of treatment in a cohort of mCRC patients receiving regorafenib, as per indication. To detect a HR for PFS of 0.50 in favour of the early increase (Δd15-d1) of Ang-2 levels, setting two-sided α=0.05 and β=0.10, 87 events were required according to Schoenfeld design. Comparisons among concentrations of each marker at d1 and d15 were performed by Wilcoxon test. d1 median cut-off values of Ang-2 and Tie-2 were adopted to discriminate patients with low versus high plasma levels and were analysed for their correlation with outcome. Results One hundred patients were included. Median PFS and OS were 2.5 and 6.7 months, respectively. As compared to d1, Tie-2 levels decreased at d15 (P=0.007), while no significant early modulation was reported for Ang-2. Patients with Ang-2 early increased levels (n=42) had a trend for a better outcome (HR for PFS: 0.72 [95%CI:0.48-1.08], P=0.095; HR for OS: 0.77 [95%CI:0.51-1.16], P=0.204) than those with Ang-2 early decreased levels (n=58). No difference was detected for d1 Tie-2 levels in terms of outcome, while d1 low levels of Ang-2 were associated with longer PFS (HR: l0.59 [95%CI:0.39-0.89], P=0.005) and OS (HR:0.62 [95%CI:0.41-0.94], P=0.017). In the multivariate model, the association of d1 Ang-2 levels with PFS was confirmed (HR:0.48 [95%CI:0.31-0.76], P=0.001), but not in OS (HR: 0.80 [95%CI:0.49-1.28], P=0.351). Conclusions This experience met to prospectively validate Ang-2 baseline levels as prognostic marker and its early modulation as a predictor of benefit from regorafenib in mCRC patients. Whether the early modulation of Ang-2 levels suggests that the successful Tie-2 inhibition by regorafenib leads to a compensatory increase in Ang-2 and correlates with anti-tumour activity deserves further investigation. Legal entity responsible for the study The author. Funding ARCO Foundation. Disclosure E. Campi: Research grant / Funding (institution): Istituto Nazionale dei Tumori, Istituto di Pisa. A. Falcone: Honoraria (self): Amgen, Lilly, Merck, Roche, Servier; Advisory / Consultancy: Amgen, Bayer, Bristol-Myers Squibb, Lilly, Merck, Roche, Servier; Research grant / Funding (institution): Amgen, Bayer, Merck, MSD, Roche, Sanofi, Servier; Travel / Accommodation / Expenses: Amgen, Bayer, Merck, Roche, Servier. All other authors have declared no conflicts of interest.