Gore MR, Zanation AM. Salvage Treatment of Late Neck Metastasis in EsthesioneuroblastomaA Meta-analysis. Arch Otolaryngol Head Neck Surg. 2009;135(10):1030-1034. doi:10.1001/archoto.2009.143
Esthesioneuroblastoma (ENB) is an uncommon tumor of the sinonasal region with a 20% rate of neck metastases. To our knowledge, the rate of neck metastases occurring 6 or more months after diagnosis has not been well characterized. The rate of successful salvage of these late neck metastases, defined in this study as disease-free survival for at least 1 year, has not been previously reported.
Meta-analysis examining 33 articles published since 1990.
A total of 678 patients with ENB with 79 patients with neck metastases occurring 6 or more months after the initial diagnosis.
Patients were grouped according to treatment with surgery, radiotherapy, or combined surgery and radiotherapy.
Main Outcome Measures
The rate of successful salvage of late neck metastases, defined as disease-free survival for at least 1 year, was compared for the 3 treatment groups.
The rate of cervical metastases was 20.2%, with a 12.4% rate of late neck metastases. The combined successful salvage rate for late neck metastases with surgery, radiation, or combined therapy was 31.2%. An odds ratio (OR) analysis revealed that surgery plus radiation provided a statistically significant increase in the rate of successful salvage in patients with late neck metastases, with an OR of 8.6 vs single modality therapy and a number-needed-to-treat of 3. We found no difference in the OR for successful salvage for surgery alone vs radiation alone (OR, 1.5).
Treatment of neck metastases occurring 6 or more months after an initial diagnosis of ENB with combined surgery and radiotherapy provides a statistically significant survival advantage vs single-modality therapy.
Esthesioneuroblastoma (ENB) is an uncommon tumor of the sinonasal region that accounts for approximately 3% to 6%1,2 of all primary malignant sinus tumors and less than 1% of all head and neck cancers.2 Esthesioneuroblastoma, sometimes referred to as olfactory neuroblastoma,3 was first described by Berger et al4 in 1924. There remains some controversy as to the cell of origin of ENB, although the most widely accepted opinion is that ENB arises from the olfactory epithelium,5 which would account for the intimate relationship of most in situ ENBs with the cribiform plate, the midline superior nasal structures, and the anterior skull base. Many authors have examined the rate of neck metastasis in patients with ENB, with larger series typically reporting a rate of neck metastasis of 15% through 30%.2 Ferlito and Micheau2 found an overall rate of 23.5% in a review of articles about ENB written from 1990 through 2003.
Many of these authors reported that a substantial proportion of these neck metastases occur late (ie, ≥6 months after diagnosis of the primary tumor). To our knowledge, no study has examined the overall rate of success in salvaging late neck metastases with surgery, radiotherapy, or combined surgery and radiation. There remains controversy regarding the treatment of neck metastases in ENB. Approximately 5% of patients with ENB present with evidence of disease in the neck,2 and it is generally accepted that treatment of the neck in these patients is indicated. Most centers do not routinely perform neck dissection or radiation treatment of the neck at the time of resection of the primary tumor unless there is evidence of cervical disease.
The difficulty in establishing the optimal approach to salvaging neck metastases in the patient with ENB stems from several factors, including the low incidence of the disease, which makes it difficult to perform randomized trials or to assemble retrospective cohorts with large numbers of patients, and the high proportion of late neck metastases, which can present several years after treatment of the primary tumor.
Given the significant rate of cervical metastases, particularly late neck metastasis, insight into the value of salvage therapy for late neck metastases is invaluable in developing a more standardized approach to the neck in the patient with ENB. There may be value to elective treatment of the neck with neck dissection, radiotherapy, or both, because cervical metastases have been shown to drastically reduce survival rates in ENB.
We reviewed the largest ENB series, to our knowledge, reported since 1990.1,2,5- 37 The case series included patients in whom the diagnosis of ENB was supported by immunohistochemical, histochemical, or histologic analysis. Our inclusion criteria were patients with confirmed ENB with late neck metastases, which we defined as metastases occurring 6 or more months after diagnosis of the primary tumor, from studies since 1990 from which data were available on the rate of late neck metastases. We examined the rate of salvage of late neck failures with surgery, radiotherapy, or combined treatment with surgery and radiotherapy. Successful salvage therapy was defined as disease-free survival at 1 year after salvage therapy because follow-up data were available for patients with neck salvage failure from at least 12 months after salvage therapy. Cases of patients with late neck salvage failures treated exclusively with chemotherapy were not included in the salvage data. The rate of failure among patients treated with surgery, radiotherapy, and combined surgery and radiotherapy among patients in which late neck salvage failed vs successful late neck salvage was examined. We also examined the location of salvage failure for patients for whom these data were reported, as well as available data on the treatment of subsequent salvage failure following attempted salvage of late neck metastasis. The odds ratios (ORs) and confidence intervals (CIs) were calculated using OR = ad/bc, standard error (SE) of (log)e OR = √(1/a + 1/b + 1/c + 1/d), Y = loge OR − [N1-α/2 × SE (loge OR)], Z = loge OR + [N1-α/2 × SE (loge OR)] (where α = .05 to give the 95% CI), and CI = eY to eZ.38 The Fisher exact test was then used to calculate the P value for the association between groups and outcomes, with P ≤ .05 considered statistically significant.
A total of 32 studies, involving 687 patients (Table 1), published from 1990 through 2007 were examined.1,5,5- 37 Ten studies did not include explicit data on salvage in patients with late neck metastases, leaving a total of 476 patients included in the remaining studies. The overall rate of neck metastases was 20.2% for all 678 patients (137 of 678). The rate of late neck metastases for the 637 patients for whom data on time of diagnosis of cervical metastasis were available was 12.4% (79 of 637). Of the total number of patients with cervical metastases for whom data on time of diagnosis of neck disease were available, 61.7% (79 of 128) of these cervical metastases occurred 6 or more months after diagnosis of the primary ENB. The rate of attempted neck salvage with surgery, radiotherapy, or both was similar among patients with early and late neck metastases, with 45 of 62 patients (70%) with late neck metastases undergoing attempted neck salvage compared with 36 of 49 patients (73%) with early neck metastases undergoing attempted salvage therapy. Of 36 patients with neck metastases diagnosed at less than 6 months from the time of diagnosis of the primary tumor, neck salvage failed in 25 patients (69%). This was similar to the failure rate of attempted neck salvage in 45 patients with late neck metastases, with salvage failure in 31 patients (a rate of 69%).
Data on the specific type of salvage therapy used were available for each of the 45 patients with late neck metastases who underwent attempted salvage therapy (Table 2). Patients who underwent chemotherapy in addition to another modality were included with that modality (eg, patients who underwent salvage therapy with chemoradiation were included in the radiation treatment group). Patients who underwent chemotherapy alone were not included in the attempted salvage group. In patients with late neck metastases who underwent attempted salvage and failed treatment, most (16 of 31 [52%]) had undergone salvage therapy consisting of surgery alone. Treatment in the remaining patients was evenly divided between attempted salvage with radiation alone (8 of 31 [16%]) and surgery plus radiation (7 of 31 [23%]). For patients who underwent successful salvage of the neck (ie, experienced disease-free survival for at least 1 year), most had undergone salvage treatment with a combination of surgery and radiation (10 of 14 [71%]). Salvage therapy in 3 of 14 patients had been successful using surgery alone (21%), and salvage was successful in 1 of 14 patients treated with radiation to the neck alone (7%).
Table 3 shows the results of an OR analysis for combined surgery and radiotherapy vs surgery alone, radiotherapy alone, and the combined data for surgery alone plus radiotherapy alone. Because the incidence of ENB is much less than 10% in the general population, the OR is expected to approximate the relative risk well. The outcome was successful salvage therapy of late neck metastasis, with success defined as disease-free survival for at least 1 year. Combined surgery plus radiotherapy increased the rate of successful neck salvage over surgery alone (OR, 7.6; 95% CI, 1.1-24.2; 2-tailed Fisher exact test P = .01; NNT, 3; 95% CI of NNT, 1.4-6.9), radiotherapy alone (OR, 11.4; 95% CI, 1.2-113.2; 2-tailed Fisher exact test P = .04; NNT, 3; 95% CI of NNT, 1.3-6.0), and surgery alone plus radiotherapy alone (OR, 8.6; 95% CI, 2.0-35.9; 2-tailed Fisher exact test P = .003; NNT, 3; 95% CI of NNT, 1.4-5.6). The rate of successful salvage was equivalent for surgery alone and radiotherapy alone (OR, 1.5; 95% CI, 0.1-16.8; 2-tailed Fisher exact test P > .99; NNT, 22).
Owing to the relatively low incidence of ENB and the difficulty in assembling large case series or prospective trials, treatment of primary disease and regional spread remain controversial. In addition, the treatment of this disease has changed greatly over the years, meaning that patients included in case series can span several decades over which treatment modalities have changed greatly. One particular aspect of ENB that is potentially troublesome is the tendency of cervical metastases to present late. Although this pattern has been noted by several authors, to our knowledge, standardization of the evaluation and treatment of these late neck metastases has not been specifically addressed. The rate of cervical metastasis observed overall in this work, 20.2%, was similar to the rates reported previously in the literature. Interestingly, of all patients with cervical metastases, a slight majority, 61.4%, presented with neck disease at 6 or more months after diagnosis of the primary tumor. We found that the rate of attempted salvage (roughly 70% in both groups) as well as the rate of failed salvage of the neck with surgery, radiotherapy, or a combination of the 2 (roughly 69% in both groups) was similar between patients who presented with early and late neck metastases. The site of failure after neck salvage was specified in 17 patients, and most of these failures were locoregional,11 whereas 6 failures occurred in distant sites. Perhaps most interestingly, most patients (71%) who were alive and disease free for at least 1 year following attempted salvage of late neck metastases were treated with combined surgery and radiation therapy. This is in contrast to the group of patients who underwent failed salvage therapy, a small majority of whom (52%) had been treated with surgery alone, with similar numbers being treated with radiation alone or combined radiotherapy and surgery (26% and 23%, respectively). The OR analysis reveals that combined surgery and radiotherapy is clearly superior to surgery alone or radiotherapy alone, with the OR for successful salvage for combined surgery with radiotherapy vs surgery alone or radiotherapy alone being greater than 7.0 and statistically significant for each. In addition, the number of patients needed to treat (NNT) with combined therapy vs single modality therapy was only 3. The clear increase in the odds of successful salvage as well as the low NNT in order to benefit 1 patient illustrate that combined modality therapy clearly increases the odds of successful salvage and is superior to single modality therapy in providing disease-free survival for at least 1 year.
Although care must be taken when interpreting these numbers given the differences in use of chemotherapy, treatment of the original primary lesion, and the relatively large time span over which data were collected, there is a clear predominance in the use of surgery plus radiotherapy in the treatment of patients who experienced disease-free survival, whereas to a lesser degree there was a clear predominance of patients treated with surgery alone among those in whom salvage therapy failed. Although it is difficult to posit a causal relationship with retrospective data, it seems clear that surgery plus radiotherapy, with or without chemotherapy, is the treatment of choice in salvage of late neck metastases in patients with ENB. This finding is further strengthened by the low incidence of ENB in the general population, which predicts that the OR should approximate relative risk well. In addition, the rate of successful salvage of late neck metastases in this data set, approximately 30%, is greater than the rate of successful salvage for neck recurrence in patients with head and neck squamous cell carcinoma treated with surgery, radiotherapy, or combined modality treatment42 and similar to the rate of successful salvage of isolated neck recurrences in patients treated with definitive radiotherapy for node-positive head and neck cancer.43 Although late neck recurrence in ENB still carries a relatively poor prognosis, 30% is an encouraging rate of successful salvage for such advanced disease. Given the high rate of late neck metastases, it is reasonable to screen patients whose neck tumor is classified as N0 at the time of diagnosis of their primary ENB for recurrence 6 months to 1 year after diagnosis with computed tomographic scans of the neck. It is also reasonable to regularly examine the neck for clinical signs of metastasis when patients are seen in follow-up over the months to years following diagnosis, because patients have presented with regional spread many years after initial treatment.18 In addition, we recommend attempted neck salvage with a combination of radiotherapy and neck dissection in patients who present with late neck metastases and are both willing and sufficiently healthy to undergo salvage treatment of the neck.
Correspondence: Adam M. Zanation, MD, Department of Otolaryngology–Head and Neck Surgery, University of North Carolina at Chapel Hill School of Medicine, CB No. 7070, Chapel Hill, NC 27599-7070 (firstname.lastname@example.org).
Submitted for Publication: January 19, 2009; final revision received March 29, 2009; accepted April 5, 2009.
Author Contributions: All authors had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Gore and Zanation. Acquisition of data: Gore and Zanation. Analysis and interpretation of data: Gore and Zanation. Drafting of the manuscript: Gore and Zanation. Critical revision of the manuscript for important intellectual content: Gore and Zanation. Statistical analysis: Gore. Administrative, technical, and material support: Zanation. Study supervision: Zanation.
Financial Disclosure: None reported.
Previous Presentation: This study was a poster presentation at the Seventh International Conference on Head and Neck Cancer; July 20, 2008; San Francisco, California.