A novel PET scanner with semiconductor detectors may improve diagnostic accuracy in the metastatic survey of head and neck cancer patients
- 333 Downloads
Our research group developed new PET scanner with semiconductor detectors for high spatial resolution with low scatter noise. On head and neck cancer (HNC) surgery, FDG-PET may often provide false-positive findings in cervical node involvements. Accordingly, we assessed diagnostic accuracy using this new scanner in the HNC patients as compared with the conventional lutetium oxyorthosilicate (LSO) PET.
We prospectively studied FDG imaging in 35 HNC patients by both semiconductor PET and LSO-PET. At 60 min after 18F-FDG injection, two PET scans were obtained using both scanners consecutively and in random order. Two nuclear medicine specialists scored FDG abnormalities using 5 point scale system for receiver operating characteristic (ROC) curve analysis.
63 suspected of metastatic or recurrent lesions were evaluated and correlated by the final confirmation by pathological findings or clinical courses (malignant 26/benign 37). Semiconductor PET showed sensitivity of 92.3 % (24/26), specificity of 51.4 % (19/37), and accuracy of 68.2 % (43/63), while LSO-PET showed sensitivity of 84.6 % (22/26), specificity of 16.2 %(6/37), and accuracy of 44.4 % (28/63), respectively. Especially, semiconductor PET accurately diagnosed as true negative in the 13 of 14 lesions only detected by LSO-PET. ROC analyses revealed the diagnostic superiority of semiconductor PET from location of- and area under curve particularly in the study of small (≤10 mm) lesions.
A new novel semiconductor PET scanner can increase diagnostic accuracy with reduction in false positive findings in the HNC patients mainly due to higher spatial resolution and lower noise than the LSO-PET. This new technology can lead to more accurate diagnosis and the more optimal therapeutic tactics in head and neck surgery.
KeywordsSemiconductor scanner PET Head and neck cancer
We thank nuclear medicine technologists, Hidehiko Omote and Hiroshi Arai. We are grateful to Dr Kenji Hirata and Naoya Hattori for clinical study.
Conflict of interest
- 5.Darambara DG, Todd-Pokropek A. Solid state detectors in nuclear medicine. J Nucl Med. 2002;46(1):3–7.Google Scholar
- 27.Casey M. Point spread function reconstruction in PET. Oxford: Siemens Molecular Imaging; 2007.Google Scholar
- 33.de Jong HWAM, Knoess C, Lammertsma AA, et al. Performance characteristics of the high resolution research tomograph comparison of three prototypes. Paper presented at Nuclear Science Symposium Conference Record, 2004 IEEE, pp 16–22; 2004.Google Scholar
- 36.Antoch G, Saoudi N, Kuehl H, et al. Accuracy of whole-body dual-modality fluorine-18-2-fluoro-2-deoxy-d-glucose positron emission tomography and computed tomography (FDG-PET/CT) for tumor staging in solid tumors: comparison with CT and PET. J Clin Oncol. 2004;22(21):4357–68.PubMedCrossRefGoogle Scholar
- 39.Knausl B, Hirtl A, Dorbrozemsky G, et al. PET based volume segmentation with emphasis on the iterative treu X algorithm. Z Med Phys (in press).Google Scholar
- 41.Hass HG, Schmidt A, Nehls O, Kaiser S. DNA ploidy, proliferative capacity and intratumoral heterogeneity in primary and recurrent head and neck squamous cell carcinomas (HNSCC)—potential implications for clinical management and treatment decisions. Oral Oncol. 2008;44(1):78–85.PubMedCrossRefGoogle Scholar