CLINICAL INVESTIGATION Head and Neck Cancer IMAGING TUMOR PERFUSION AND OXIDATIVE METABOLISM IN PATIENTS WITH HEAD-AND-NECK CANCER USING 1- [ 11 C]-ACETATE PET DURING RADIOTHERAPY: PRELIMINARY RESULTS AIJUN SUN, M.D.,* y SILVIA JOHANSSON, M.D., PH.D.,* z INGELA TURESSON, M.D., PH.D., z ALEXANDRU DAS ¸U,PH.D., x AND JENS S€ ORENSEN, M.D., PH.D.* *Section of Nuclear Medicine, Department of Medical Sciences, Uppsala University Hospital, Uppsala, Sweden; y Section of Oncology, Department of Radiation Sciences, Ume  a University Hospital, Ume  a, Sweden; z Section of Oncology, Department of Medical Sciences, Uppsala University Hospital, Uppsala, Sweden; x Department of Radiation Physics, Norrlands University Hospital, Ume  a, Sweden Background: A growing body of in vitro evidence links alterations of the intermediary metabolism in cancer to treatment outcome. This study aimed to characterize tumor oxidative metabolism and perfusion in vivo using dynamic positron emission tomography (PET) with 1- [ 11 C]-acetate (ACE) during radiotherapy. Methods and Materials: Nine patients with head-and-neck cancer were studied. Oxidative metabolic rate (k mono ) and perfusion (rF) of the primary tumors were assessed by dynamic ACE-PET at baseline and after 15, 30, and 55 Gy was delivered. Tumor glucose uptake (Tglu) was evaluated with [ 18 F]-fluorodeoxyglucose PET at baseline. Patients were grouped into complete (CR, n = 6) and partial responders (PR, n = 3) to radiotherapy. Results: The 3 PR patients died within a median follow-up period of 33 months. Baseline k mono was almost twice as high in CR as in PR (p = 0.02) and Tglu was lower in CR than in PR (p = 0.04). k mono increased during radiotherapy in PR (p = 0.004) but remained unchanged in CR. There were no differences in rF between CR and PR at any dosage. k mono and rF were coupled in CR (p = 0.001), but not in PR. Conclusions: This study shows that radiosensitive tumors might rely predominantly on oxidative metabolism for their bioenergetic needs. The impairment of oxidative metabolism in radioresistant tumors is potentially reversible, suggesting that therapies targeting the intermediary metabolism might improve treatment out- come. Ó 2012 Elsevier Inc. 1- [ 11 C]-acetate PET, Perfusion, Oxidative metabolism, Head-and-neck cancer, Radiotherapy. INTRODUCTION In locally advanced head-and-neck cancer, the 5-year progression-free survival is about 50% with combined radio- therapy, chemotherapy, and surgery (1). Further improvement of current outcome rates could be achieved by carefully study- ing and targeting the factors that are related to treatment failure. Among these, tumor hypoxia is an important factor de- termining treatment response (2), because poor tumor oxygen- ation correlates to radioresistance and local failure (3). Intracellular oxygen is involved in increasing the DNA dam- age induced by radiation (4), but is also necessary for the tumor to maintain oxidative phosphorylation (5). Lack of ox- ygen resulting from insufficient perfusion would force the tumor cells to switch from cellular respiration toward anaero- bic glycolysis for survival. However, it has been well known since the days of Warburg that many cancers share a common glycolytic phenotype, even in the presence of oxygen (6). War- burg attributed this phenomenon to a deranged mitochondrial function, causing impaired oxidative phosphorylation and disease progression. In vitro studies along this line seem to confirm Warburg’s observation. Tumor cells with deficient oxidative metabolic capacity represent a more malignant phenotype (7), and oxidative metabolism may be a key factor in controlling cancer growth (8). Increased tumor glycolysis is detectable in vivo by [ 18 F]-fluorodeoxyglucose (FDG)-posi- tron emission tomography (PET) (9) and quantification of tumor FDG uptake using PET appears to carry prognostic in- formation (10). However, glucose uptake appears unrelated to the distribution of hypoxia (11). These findings imply that imaging tumor oxidative metabolism and perfusion in vivo might provide insights into the bioenergetic mechanisms and ultimately predict tumor response. Reprint requests to: Jens S€ orensen, M.D., Ph.D., Department of Nuclear Medicine, Uppsala University Hospital, SE-75185 Up- psala, Sweden. Tel: +4618666809; Fax: +4618666819; E-mail: jens.sorensen@medsci.uu.se Conflicts of interest: none Acknowledgments—The authors wish to thank Dr Lena Cederblad and the staff at Uppsala Imanet PET Center for expert assistance in performing the studies. This work was supported by grants from Uppsala University Amersham PET Research Fund, Uppsala University Hospital and Laryngfonden Sweden. This article was presented in part at the annual meeting of Society of Nuclear Medicine, June 2009. Received May 27, 2008, and in revised form Aug 27, 2010. Accepted for publication Nov 17, 2010. 554 Int. J. Radiation Oncology Biol. Phys., Vol. 82, No. 2, pp. 554–560, 2012 Copyright Ó 2012 Elsevier Inc. Printed in the USA. All rights reserved 0360-3016/$ - see front matter doi:10.1016/j.ijrobp.2010.11.007