Optimization of radiotherapy for neck carcinoma metastasis from unknown primary sites: a meta-analysis

This meta-analysis was designed to evaluate radiotherapy (RT) options preferable for neck cancer metastases from unknown primary sites (NCUP). Relevant articles published up through September 2015 were selected from EMBASE, Cochrane, PubMed and Web of Science. Thirty-three articles were identified, and relative risks (RRs) and 95% CIs for all pre-specified endpoints were calculated. Surgery plus RT showed an advantage for 5-year overall survival (OS) (RR 0.66, 95% CI 0.52–0.83, p = 0.0004) and neck recurrence (NR) (RR = 0.74, 95% CI 0.59–0.92, p = 0.008) compared to RT alone. The RRs for NR, primary tumor emergence (PTE), and 5-year disease free survival (DFS) for bilateral neck compared to ipsilateral neck irradiation were 0.61 (95% CI 0.41–0.91, p = 0.01), 0.44(95% CI 0.26–0.77, p = 0.004), and 0.81 (95% CI 0.64–1.03, p = 0.09), respectively. Irradiation of the neck plus potential primary tumor sites (PPTS) showed a benefit for 5-year DFS (RR 0.75, 95% CI 0.61–0.92, p = 0.005), NR (RR = 0.72, 95% CI 0.56–0.92, p = 0.009), and PTE (RR = 0.23, 95% CI 0.12–0.45, p < 0.0001) compared to neck-only irradiation. Adverse events occurred more frequently with bilateral neck plus PPTS irradiation. For NCUP, surgery plus RT of the bilateral neck and PPTS was associated with greater improvement of clinical outcomes.

NCUP is thought to be potentially curable [2], but the data addressing the therapeutic protocols and outcomes of NCUP treatment are limited and controversial.The proposed treatment modalities include surgery alone, radiotherapy (RT) alone, and a combination of RT and surgery.Opinions on the field design for RT also vary.Some investigators have recommended involved-field irradiation, such as ipsilateral neck irradiation only [7,[11][12][13][14], while others suggest extended field irradiation, including prophylactic irradiation of potential head and neck mucosal sites and both sides of the neck [1,4,6,15].Differences in treatment strategy and patient selection have led to inconsistent results.Consequently, the reported 5-year overall survival (OS) rates for patients with NCUP range from 16% to 86%, and the local control rates range from 37% to 91%.The present meta-analysis was performed in an effort to identify the optimal treatment regimen for NCUP, focusing in particular on the optimal way to schedule RT.

Description of selected studies and quality assessment
A total of 787 articles were identified, of which 33 articles qualified for inclusion.The flow diagram for Research Paper www.impactjournals.com/oncotargetstudy selection is shown in Figure 1.The characteristics of the included studies are summarized in Tables 1 and 2.Although methods for managing missing data are not adequately described in some studies, none of the included studies had a NOS < 6, which suggests all were of high quality.

Toxic effects of different radiotherapeutic regimens
Only two of the included studies [11,15] compared toxicities between ipsilateral neck irradiation and bilateral neck plus PPTS irradiation, and the toxicity data could only be assessed for severe acute toxicity and xerostomia.We found there was an increased risk of severe acute toxicity (RR = 1.91, 95% CI 1.26-2.88,p = 0.002) (Supplementary Figure S1A) and xerostomia (RR = 6.82, 95% CI 0.96-48.55,p = 0.06) (Supplementary Figure S1B) in the group with bilateral neck and PPTS irradiation.

DISCUSSION
Optimal treatment for patients with NCUP remains uncertain.The incidence of NCUP is about 3 cases per 1,000,000 per year.Its rarity makes randomized and prospective studies unavailable, and leaves clinicians with only small retrospective studies for clinical decision making.To the best of our knowledge, this study is the first meta-analysis with a focus on comparing the therapeutic efficacies of different treatment regimens, and on providing a higher level of evidence for optimizing the RT schedule in NCUP.
Some studies held that surgery plus RT resulted in a higher probability of cure [2,5,9,27,28], while others reported that the outcome of surgery plus RT were similar to those of definitive RT alone, but with a higher risk of severe complications [19,25,29] .We therefore first performed a comparison of RT alone with RT plus surgery.The pooled analysis demonstrated that RT plus surgery was associated with a greater 5-year OS rate than RT alone.Moreover, there was a beneficial trend toward a higher 5-year DFS, though the effect was not significant.The higher CR rate and lower NR rate is also consistent with a survival benefit from the combination of surgery and RT.
For neck-irradiation settings, the current guidelines suggest treating the involved lymph node field [34].However, some reports indicate that patients administered RT to the bilateral neck nodes appeared have greater local control and higher survival rates than those who received only ipsilateral irradiation [1,4,6,15].In the present pooled analysis, significantly less contralateral cervical recurrence or emergence of a primary tumor was noted in patients receiving bilateral irradiation, and there was trend toward increased 5-year OS and DFS.These findings suggest that current guidelines recommending the involved lymph node field as the standard RT schedule may need to be re-evaluated for the NCUP setting.
As to the value of irradiation of the PPTS, although current guidelines recommend it as routine consideration for inclusion in the target volume [34], conclusions drawn from currently available evidence are controversial.Some studies have shown a higher 5-year OS rate and better regional control with addition of irradiation of potential head and neck mucosal sites of cancer growth [1, 4-6, 8, 13, 15], whereas, other trials observed that mucosal irradiation reduced both the emergence of primary tumors and regional recurrence, but did not affect OS [3,7,12].In the pooled analysis of all these trials, not only was an advantage for regional control validated, so was a survival benefit in patients treated with irradiation of the neck and PPTS.
We also evaluated the toxic effects of different RT regimens.Because the data were limited, we only assessed severe acute toxicities and xerostomia.We found a significantly higher risk of these adverse events in the RT to the bilateral neck plus PPTS group.However, it is believed that these severe acute toxicities are clinically manageable, and xerostomia could be minimized by application of intensity modulated RT (IMRT) [11,15].There were several limitations to this study.First, all the included studies were retrospective and the sample groups were small.There was not sufficient data to perform subgroup analyses based on lymph node levels, lymph node stages, histological types, sequence of surgery and RT, or radiation dosage.Second, these studies were performed over a long time-span.Consequently, the techniques for delivering RT were varied, and precision RT techniques, such as 3-D conformal RT and IMRT, were not yet broadly applied.This could result in an underestimation of the actuarial effect of RT.To address these issues, future multicenter RCTs are needed.

Literature search and selection
Two authors (X.M.L. and X.X.Z.) independently carried out systematic literature searches of EMBASE, Cochrane, Pubmed and Web of Science before September 25, 2015.The following terms were used: occult primary, unknown primary, neck lymph node, cervical lymph node, metastatic, metastases, cancer, neoplasm, tumor, carcinoma, radiotherapy, irradiation, radiation.
Studies meeting the following selection criteria were included.(1) Study population: patients with cervical lymph node metastases from unknown primary sites, and with no cancer history.(2) Study design: comparative studies comparing RT alone with a combination of RT and surgery (radical neck dissection, selective neck dissection, or excisional biopsy); comparing ipsilateral irradiation with bilateral irradiation; or comparing neckonly irradiation with neck and potential primary tumor site (PPTS) (nasopharynx, oropharynx, larynx, and/ or hypopharynx) irradiation.(3) Language: English.(4) Studies with available data on at least one of the pre-specified endpoints: 5-year OS, 5-year disease free survival (DFS), neck recurrence (NR), complete response (CR), primary tumor emergence (PTE), ipsilateral neck recurrence (INR), contralateral neck recurrence (CNR), severe acute toxicity (RTOG grade≥3) and xerostomia.Editorials, letters to the editor, and review articles were excluded (Figure 1).

Data extraction
The following items were extracted independently by the two authors (X.M.L. and D.H.L.) from the published articles: year of publication, first author, country, study period, demographic and clinical information on the study patients (age, gender, histology, N stage, N level), schedule of treatment, number of patients, outcome results, and follow-up.Any disagreement was resolved through further discussion and including a third author.

Figure 1 :
Figure 1: Flow diagram of studies identification and selection.