Consolidation cetuximab after concurrent triplet radiochemotherapy+cetuximab in patients with advanced head and neck cancer: A randomized phase II study.

• Consolidation cetuximab after concurrent triplet radiochemotherapy+cetuximab did not improve the 2-year locoregional control (LRC) rate. • Toxicity was similar in both arms except a tendency for increased late G3/4 mucositis in the arm with consolidation cetuximab. • High serum levels of IL-8, Osteopontin and FasL2 were negatively associated with the primary endpoint 2-year LRC. • The first radiomics model prognostic not only for local control but for combined local and nodal (locoregional) control could be built. • No correlation was found between serum and radiomics markers. Preclinical data suggest that cetuximab should be continued after end of concurrent radiotherapy+cetuximab due to its efficacy against residual tumor cells in the irradiated tumor bed. Based on this concept the phase II add-on cetuximab (AOC) study was designed. Altogether 63 patients with advanced head and neck cancer were treated with radiochemotherapy (70 Gy, cisplatin 40 mg/m2 weekly) in combination with concurrent cetuximab (loading dose 400 mg/m2, then 250 mg/m2 weekly). Thereafter patients were randomized to cetuximab consolidation (500 mg/m2 biweekly × 6) or no further treatment. The primary endpoint was the 2-year locoregional control (LRC) rate. As translational research endpoints serum markers were analyzed before and during treatment and CT-based quantitative image analysis (radiomics) was performed. Median follow-up was 24 months. The 2-year LRC rates were 67.9% and 67.7% in the treatment arms with and without consolidation cetuximab, respectively. Higher than median levels of three serum markers were negatively associated with the 2-year LRC rate in the overall patient cohort: Osteopontin, IL8 and FasL2 (p ≤ 0.05). A radiomics model consisting of two radiomics features could be built showing that higher entropy and higher complexity of tumor Hounsfield unit distribution indicates worse LRC (concordance index 0.66). No correlation was found between biological and imaging markers. There was no evidence that consolidation cetuximab would improve the 2-year LRC rate. Prognostic biological and imaging markers could be identified for the overall patient cohort. Studies with larger patient numbers are needed to correlate biological and imaging markers. [ABSTRACT FROM AUTHOR]