The T staging of primary HNSCC, according to the American Joint Committee on Cancer (AJCC) tumor, node, metastasis (TNM) staging manual, version 9, is determined through a combination of clinical examination, CT, and/or MR imaging. MR imaging is preferred if there is suspicion for skull base/intracranial/intraorbital invasion, perineural spread, or bone marrow involvement; to differentiate tumor from paranasal inflammatory disease; and if CT images are degraded by dental hardware artifact. CT is useful for assessing bone erosion/destruction. Although PET is highly accurate in the detection of primary tumors, its poor spatial resolution limits its clear delineation of tumor margins. False negative results from small superficial tumors and those located in/obscured by areas of high physiologic activity such as nasopharyngeal, palatine, and lingual lymphoid tissues. These areas may also give rise to false positive results. These errors may be ameliorated largely by obtaining contrast-enhanced CT (CECT) or MR imaging for attenuation correction. PET/CT may be useful in oral cavity SCC where dental artifact may limit precise T staging. SCC from an unknown primary tumor, presenting as cervical lymphadenopathy on clinical or imaging evaluation ( Fig. 2 A–D ) may account for 2.0% to 5.0% of all head and neck cancers. Unknown primary may be a transient definition with small primary tumors being often identified as HPV-positive oropharyngeal SCC. Its identification enables targeted treatment thus reducing treatment-related morbidity. Although PET/CT plays a key role in this regard, its accuracy is limited by the small size of these tumors and their presence in areas of intrinsically high metabolic activity such as the base of the tongue and palatine tonsils. Various studies have evaluated the efficacy of PET/CT in the detection of primary tumors with sensitivity ranging from 79.2% to 91.5% and specificity ranging from 70.4% to 87%. ^,,

A 54-year-old man with right neck mass. Fine needle aspiration (FNA) showing keratinizing SCC. ( A , B ) Axial postcontrast CT shows no evident mass in the tonsillar fossae ( circles ) with a partially necrotic level II lymphadenopathy ( arrow ). Initial oral examination was negative suggesting occult primary. ( C , D ) Follow-up axial fused ¹⁸ F-FDG PET/CT show a focal asymmetrical uptake in right tonsillar fossa ( arrow in C ) suspicious for the primary site with metabolic activity in the metastatic right level II lymphadenopathy ( arrow in D ). Re-evaluation with direct laryngoscopy under anesthesia revealed a small ulcer in the right tonsillar fossa and biopsy was positive for SCC.

Staging of cervical nodal metastases in HNSCC incorporates nodal size, level, laterality, and extracapsular extension (ENE) of disease. This impacts treatment recommendations, with the goal of reducing overtreatment for HPV-positive patients with limited nodal involvement. A meta-analysis by Sun and colleagues revealed that pooled per-patient, per-neck-side, and per-neck-level sensitivities/specificities for ¹⁸ F-FDG-PET/CT were 0.91/0.87, 0.84/0.83, and 0.80/0.96, respectively. They concluded that ¹⁸ F-FDG-PET/CT has good diagnostic performance for the detection of regional nodal metastasis. PET is especially useful in the detection of nodal metastatic disease in the clinically/radiologically node negative (cN0) neck. A recent meta-analysis by Guedj and colleagues showed that ¹⁸ F-FDG-PET/CT has a high specificity and negative predictive value for ruling out nodal involvement in cN0 necks, but a limited sensitivity. It also revealed that ¹⁸ F-FDG-PET/CT missed nearly a third of cN0 patients who were found to be pN + upon pathologic analysis, suggesting that ¹⁸ F-FDG-PET/CT alone may be insufficient as a decision-making tool for elective neck dissection. The presence ENE is a poor prognostic indicator often necessitating chemoradiation and is more accurately detected by using CECT than PET/CT. It has been suggested that using a SUV _max cutoff of 3.0 may provide appropriate diagnostic value in identifying ENE. ^,

Distant metastatic disease from HNSCC occurs in less than 5% of cases at initial presentation, usually to the lungs, liver, bones, and rarely to the brain. There are several factors that correlate with increased risk for distant metastasis at the time of initial staging and are summarized in Box 1 . Despite the presence of distant metastases portending poor outcome, the emergence of new chemotherapeutic and immunotherapeutic agents and local ablative techniques offers hope. FDG-PET is highly accurate in the detection of such spread, especially osseous metastatic lesions that may not be apparent on conventional imaging. Limitations include small metastatic lung nodules, which may require CT, and intracranial lesions that may be obscured by the intense physiologic uptake of brain parenchyma. PET/CT is also highly accurate in ruling out the presence of a second primary tumor, encountered in 5% to 10% of patients with HNSCC (especially in HPV-negative tumor and in smokers) with other upper aerodigestive tract and lungs being the most common sites of origin.