S622
Clinical – Head & neck
ESTRO 2026
was 38.5 months. All data were collected using the DKTK-RadPlanBio platform [2]. For analysis of the trial results, all primary tumours and lymph node volumes were segmented centrally, including the first site of locoregional relapse to assure that the failure occurred within the radiation dose volume specified by the protocol and does not represent out-of-field- failure. The baseline signature derived from our previous exploratory retrospective study [1] served as hypothesis for this confirmatory validation trial. It included the tumour volume (lnGTV), p16 status, CD44 expression, the 15-gene hypoxia-associated classifier [3] and the interaction of the hypoxia classifier with the tumour volume. Results: All parameters of the previously published baseline model were significantly associated with LRC in univariable analysis: Increasing tumour volume was associated with lower LRC (p=0.020). Patients with p16-positive HNSCC and low CD44 expression showed higher LRC (p=0.010; p=0.034). Tumours with high hypoxia-associated gene expression showed low LRC (p=0.029); the interaction between hypoxia classification and tumour volume was significantly associated with LRC (p=0.005). Additional relevant parameters included T- and N-stage, smoking, alcohol abuse and tumour localisation. The hypothesis of the prospective HNprädBio biomarker trial, that is, the multivariable baseline model derived from the previous retrospective exploratory study was successfully validated. The C-index for predicting LRC was 0.64 (95% confidence interval 0.55-0.71), which was similar as published for the retrospective cohort (C-index=0.65). Patient stratification into three risk groups, as defined in the hypothesis-generating study [1], was confirmed leading to significant differences between low- and high-risk groups (Figure 1). Similar performance was observed for OS.
Conclusion: The prospective and confirmatory DKTK HNprädBio clinical biomarker trial, successfully validated the baseline signatures for LRC and OS hypothesized from the previous exploratory DKTK study. The study, which is one of the first prospective biomarker trials in radiation oncology, provides high-level and robust evidence for stratifying patients in future interventional trials towards personalized prescriptions in radiation oncology. References: 1. Löck S, Linge A, Lohaus F, Ebert N, Gudziol V, Nowak A, et al. Biomarker signatures for primary radiochemotherapy of locally advanced HNSCC - Hypothesis generation on a multicentre cohort of the DKTK-ROG. Radiother Oncol. 2022;169:8–14.2. Skripcak T, Just U, Simon M, Büttner D, Lühr A, Baumann M. Toward Distributed Conduction of Large- Scale Studies in Radiation Therapy and Oncology: Open-Source System Integration Approach. EEE Journal of Biomedical and Health Informatics. 2016;20(5):1397–403.3. Toustrup K, Sorensen BS, Nordsmark M, Busk M, Wiuf C, Alsner J, et al. Development of a hypoxia gene expression classifier with predictive impact for hypoxic modification of radiotherapy in head and neck cancer. Cancer Res. 2011;71(17):5923–31. Keywords: HNSCC, individualized radiotherapy, trial
Made with FlippingBook - Share PDF online