Initial Experience with Oropharynx-targeted Radiation Therapy for Metastatic Squamous Cell Carcinoma of Unknown Primary of the Head and Neck

Discussion

With conflicting data and recommendations surrounding the extent of radiation volume necessary for the treatment of patients with MUP, examination of therapeutic options that balance the efficacy of treatment with potential toxicities is critical. Traditionally, MUP has been managed by comprehensive RT fields that covered bilateral neck nodes and mucosal sites of putative origin (4). In recent years, modified mucosal volume coverage RT that only covers the nasopharynx and oropharynx has been applied based on the premise that primaries of the hypopharynx and larynx are rare (9). In a retrospective study, Barker et al. reviewed treatment results from 17 patients who underwent larynx-sparing RT with and without neck dissection (10). Analysis from this study revealed the 5-year cause-specific rate was 88% and OS rate was 82%, where no patient developed SCC in the mucosal site or distant metastasis after RT. Good LRC was also reported, where only one patient (6%) experienced persistent nodal disease and another patient (6%) had recurrent nodal disease one year after completing RT. The authors concluded that LRC and survival of patients undergoing larynx-sparing RT was comparable to that of comprehensive RT, with the benefit of reduced irradiation and toxicities of the spared mucosal sites.

Combined modality radiation therapy has been shown to yield outcomes superior to RT-alone in head and neck cancer (11-14). While these studies specifically excluded patients with MUP, the positive outcome of these studies led to the wide use of CCRT for the treatment of MUP, which turned out to have an excellent locoregional control and survival (15, 16). Recently, Chen et al. examined whether the addition of concurrent chemotherapy to RT influences outcome and found that CCRT is associated with significant toxicity without a clear advantage for OS, LRC and PFS in the treatment of MUP (17). Since radiation fields for MUP have classically covered all potential mucosal disease sites, limiting the toxicity imposed by RT becomes even more crucial in the setting of added toxicity associated with CCRT.

Frank et al. reported the outcome of 52 patients with MUP treated with IMRT (100%) with or without concurrent chemotherapy and targeted RT (18). Only when the disease location or histological findings strongly suggested a primary nasopharyngeal tumor origin, the larynx and hypopharynx were excluded from RT coverage in 17 patients (33%). Grade 1 xerostomia was reported in five patients (12%) and grade 2 xerostomia in three patients (6%). Grade 3 esophageal toxicity was found in two patients (4%), where one patient required continuous use of PEG and another patient with an esophageal stricture required dilation. The authors concluded that IMRT for MUP produces excellent outcome and suggested that CCRT might not be necessary when RT is delivered to the pharyngeal axis.

Our treatment approach for MUP was similar to Frank et al.'s study in terms of utilizing targeted RT with chemotherapy and IMRT, except our studies mainly employed CCRT (56%), with less use of IMRT (40%) while Frank et al. mainly utilized IMRT (100%) and less CCRT (37%). Of note, the proportion of patients with grade 1 toxicity was higher in our patient population than that of Frank et al. However, grade 2 and higher complications were more prevalent in their patient population, while our patient population basically did not experience any toxicity above grade 1, except for one patient with HIV who experienced grade 4 dysphagia. While a direct comparison cannot be made, it is plausible that the differences in the incidence and severity of toxicities between the two studies are due to difference in RT coverage of mucosal site that leads to infrequent but more severe toxicity. As a result of ongoing controversy regarding the best treatment modality in treating MUP, we were interested in assessing our unique treatment strategies and their impact on outcomes in a relatively homogeneous population treated for MUP over the past 13 years. Table II compares recent studies (17-19), as well as our own experience. The studies selected were chosen to be representative of the various approaches taken towards treatment at present. Some caveats, in comparison, include differing end-points and inclusion criteria among different studies.

Additional consideration should be given to the fact that the etiology of head and neck cancer is changing. Studies have shown that SCC of the oropharynx is increasingly associated with human papilloma virus (HPV), more common in those with a history of smoking or alcohol abuse while its incidence is also increasing among non-smokers (20-22). There is, however, a higher incidence of p53 gene mutation in smokers and drinkers compared to patients with no such habits (21, 22). In the present study, 12 (18%) of the patients were non-smokers. Furthermore, ethnicity becomes an important consideration in the management of MUP since non-Asian patients are more likely to develop SCC of the oropharynx associated with HPV, while Asian patients are more likely to develop SCC of the nasopharynx associated with Epstein–Barr virus (EBV). At the time our patients were treated, the role of HPV and EBV status in the prognosis was not clear; as such, it was not feasible for us to analyze all original tumor samples for HPV or EBV. This represents a limitation of our study. Hence, the pathological specimens of all patients in the present study will be analyzed for viral status in a future study to determine whether the improvement in outcomes was more closely linked to HPV- or EBV-positive disease.

Finally, the strengths of our study include: single-Institute experience and RT by two radiation oncologists (LBH and KSH). The caveats include: a non-randomized retrospective analysis, relatively small number of patients, dependent upon accuracy of documentation, vulnerability to selection bias, and unknown (HPV-P16 or EBV) tumoral status and poorly-documented smoking history.