Patients with advanced head and neck squamous cell carcinoma (HNSCC) may benefit from organ-preservation treatment based on combination of chemotherapy and radiotherapy without compromising disease-free and overall survival. In patients with initially advanced regional disease, there is controversy about the place of routine planned lymph node neck dissection after chemoradiotherapy, especially in responding patients without clinically invaded residual lymph nodes. There is uncertainty about the lymph nodes status after chemoradiation because the structural imaging modalities (CT, MRI) lack sensitivity and specificity : small positive lymph nodes are not detected, and residual large lymph nodes can be sterilized ( " ghosts nodes " with no sign of viable tumor cells at histopathology). Despite the absence of evidence based on prospective study, in numerous institutions currently, head and neck surgeons are quite reluctant to operate on for neck dissection patients with a complete clinical and radiological response following chemoradiation. Metabolic imaging of tumors using PET and the glucose analog FDG has proven effective in head and neck SCC, especially after treatment to differentiate disease progression from radiation-induced inflammation.1 Several studies have shown that the metabolic response could predict the presence or absence of residual tumor cells in the primary tumor as well as the probability of relapse .2-4 Conflicting results have been reported on the use of PET to predict the pathological nodal status after chemoradiation, with negative predictive values ranging from 14 % to 100 %.5,6 Discrepancies observed might be due to the fact that PET was performed at variable time points after the end of radiotherapy. Ideally, PET should be performed as late as possible so that tumor regrowth can begin and become detectable, increasing the sensitivity of the procedure.