Dynamic contrast enhanced magnetic resonance imaging for hypoxia mapping and potential for brachytherapy targeting

Background and purpose: Dynamic contrast enhanced magnetic resonance imaging (DCE-MRI) may be used to visualize tumor hypoxia, and was in this work explored in treatment planning of hypoxia-guided brachytherapy of patients with locally advanced cervical cancer (LACC). Materials and methods: Pharmacokinetic ABrix maps were derived from DCE-MR images taken prior to chemoradiotherapy of 78 patients with LACC. A logistic regression procedure was used to segment the tumor volume fraction from the ABrix maps that showed the strongest association with patient survival, denoted biological target volume (BTV) fraction. A hypoxia gene score was calculated from a biopsy-based gene signature and correlated against the BTV fraction. Brachytherapy planning based on the ABrix maps was performed, for 23 patients. A general planning aim was a minimum D90 dose of 7.5 Gy to the tumor per brachytherapy fraction. Two planning approaches were explored: (1) a conventional uniform and (2) a non-uniform approach targeting the BTV to the highest dose possible. Results: The segmented BTV fraction was significantly associated local and locoregional control (P = 0.025) and the hypoxia gene score (P = 0.002). Comparing brachytherapy approaches 1 and 2, it was possible to dose escalate the BTV with 0.4 Gy per fraction in median (D90; cohort range [0, 3.8]). Some tumors could not be dose escalated without violating the dose constraints to the organs at risk. Conclusions: Tumor regions associated with hypoxia may be targeted with brachytherapy. The presented methodology may become useful in future strategies to improve cure probability of resistant tumors. Keywords: Hypoxia, Uterine cervical cancer, Computer-assisted image analysis, Magnetic resonance imaging, Brachytherap