PET-based quantification of statistical properties of hypoxic tumor subvolumes in head and neck cancer

Purpose: PET hypoxia imaging of head and neck cancer (HNC) has the potential to stratify the response to radiochemotherapy. The aim of this work was to quantify the statistical properties of hypoxic tumor subvolumes measured by PET, and their impact on failure rate of standard chemoradiation. Methods: A PubMed search was undertaken to identify relevant publications between 2001 and July 2015, containing original data regarding the properties of HNC hypoxic subvolumes and their evolution during therapy, measured using specific PET tracers. Results: The number and intensity of hypoxic voxels is significantly reduced during therapy. Patients are 4.2 times more likely to have negative outcome if baseline hypoxic. The change of the hypoxic volume during therapy, as well as the statistical distribution of the hypoxic fraction (HF), is quantified and analyzed. HF strongly correlates with the median T/M (tumor-to-muscle) SUV (standard uptake value) ratio, but not with gross tumor volume. Hypoxia is 2.2 times more frequently manifest in T3 + T4 than in T1 + T2 primary tumors. Calculated quantities are presented for primary and nodal tumors separately, where available. Conclusions: Although hypoxia diminishes during chemoradiation, it is a major predictor of outcome. HF, if large enough, can be predicted from the median T/M SUV ratio. CT-delineated gross tumor size does not influence the percentage of hypoxic voxels. Primary tumors are less likely hypoxic at an earlier stage. (C) 2015 Associazione Italiana di Fisica Medica. Published by Elsevier Ltd. All rights reserved.