Treatment plans are mainly tailored by the involved lymph node site and its association with probable primaries (Table 11.1), stage of neck (Table 11.2), surgical margin and extranodal extension (ECE) status, and presence/absence of residual disease. Neck dissection (ND) is the preferentially recommended treatment option for N1 patients with SCC histology, RT being an alternative Category 2B. RT is also indicated in medically unfit patients or in technically unresectable N1 disease. In surgically treated patients, adjuvant treatment indications depend on the status of gross residue, surgical margins, ECE, and pathological nodal involvement. For N1 and ECE (−) patients, both observation and RT are options. For N2-3 and ECE (−) cases, RT or CRT (Category 2B) may be alternatives. If ECE (+), CRT is the treatment of choice (Category 1), but RT may be an option for cases unsuitable for chemotherapy. If gross residue left or surgical margins are positive, re-resection and/or CRT should be performed. In any patient planned to receive RT or CRT at upfront/adjuvant settings, it is mandatory to consider the tumor size, involved nodal station(s), and HPV and EBV status for the design of RT field. Salvage ND is indicated in clinical or metabolic evidence of residual disease following RT or CRT.

CRT is the preferred upfront treatment option for all medically fit patients with N2–3 disease (Category 2B). Induction chemotherapy (Category 3) followed by RT or CRT is alternative recommendation for such patients. Salvage ND should be considered in patients with incomplete response.

In N1 and early N2A disease, either ND or RT alone provides excellent regional control rates. For ECE (−) N1-2a disease, neck recurrence rates following either modality ranges from 0 to 15 % [1, 2]. In adequately dissected ECE (−) cases, postoperative RT is not necessary as does not increase local control rates or survival [1, 3]. In ECE (+) patients, postoperative concurrent CRT is mandatory to increase local control rates [1].

In patients with advanced neck disease (N2b-N3), combined modality treatment with postoperative CRT or definitive CRT followed by planned ND for residual, unresponsive, or progressive tumors is the standard treatment approach with the best neck control and survival outcomes. In a series of 224 patients treated with RT alone, Grau et al. reported 5-year neck control and overall survival rates of 50 and 37 %, respectively [4]. In two recent smaller series with 40 and 60 patients, Aslani et al. [5] and Lu et al. [6] reported encouraging 5-year local control (76.3 and 65.3 %) and overall survival (77.8 and 68.5 %) rates, which may be associated with significant improvements in diagnostics and RT techniques. As a reflection of site specific data from advanced HNC which suggest significantly better locoregional control and survival rates with CRT than RT alone, it is rational to anticipate similar outcomes for MCCUP patients [7–9]. In series of definitive concurrent CRT or postoperative CRT, the 2- to 5-year rates of 89–100 % neck control and 74–92 % overall survival are excellent [10].

A summary of results of RT and CRT series of MCCUP are summarized in Table 11.3.

Although extensive-field RT including the bilateral neck and potentially involved mucosal sites is the widely practiced RT option, the question whether the irradiation be uni- or bilateral neck and whether potential mucosal sites be irradiated has not answered yet. Data from two literature reviews favor bilateral neck irradiation and inclusion of mucosal sites over unilateral neck irradiation alone [10, 11], but care must be given as different studies may have suffer from biases of inclusion of patients with poor overall health status or heavy/inoperable necks in unilateral neck irradiation group.

RT technique of choice is another question. However, a recent study by Ligey et al. demonstrated that use of 3D-RT or IMRT were associated with significantly superior regional control (p = 0.026) and survival outcomes (p = 0.029) compared to 2D-RT [12]. Based on the available data on specific HNC sites, IMRT has the similar potential to offer lesser acute and late toxicity rates for MCCUP patients, especially for cases in whom bilateral neck and/or mucosal sites are planned to be irradiated [13, 14]. Therefore, we recommend the use of 3D-RT as minimum standard, preferably IMRT where available.

Primary tumor emergence at mucosal sites is generally in the range of 0–12 % [15], increasing by time and decreasing with involvement of mucosal sites in the RT portal (Table 11.3). In series of Wallace et al., mucosal control rates for neck only and selective mucosal irradiation were 92 and 100 %, respectively [16]. Because majority of unknown primaries emerge at tonsils and base of tongue, the authors included nasopharynx and oropharynx in their selective radiation portal and included larynx and hypopharynx only in cases with level 3 nodal involvements. On the other hand, as their incidence rates are similar, it is rather difficult to discriminate true mucosal emergence of MCCUP from second primaries of head and neck area [17].