A Meta-analysis of the randomized controlled trials on elective neck dissection versus therapeutic neck dissection in oral cavity cancers with clinically node negative neck

摘要

There is no greater controversy on the management of oral cancers than the optimal treatment for clinical N0 necks. Researchers have however demonstrated that these clinical N0 neck have shown evidence of occult metastases in about 30% or higher, depending on the size, site of primary tumor and the histological diagnostic methods. The greatest challenge that is being faced by the head and neck oncologists and surgeons is the correct identification of the subset of these patients with cervical nodal micro metastases that will require elective neck treatment. Clinical palpation of the neck is grossly inadequate. Although the available radiological investigative tools have shown some improvement in the detection of neck metastasis but the sensitivity rates have been reported to be in the range of about 70 – 80%. Despite the increase in knowledge and advancement in cancer management, there is still no method to determine correctly the real micro metastatic disease free neck. Although squamous cell carcinoma of head and neck regions is a locally aggressive disease with a great tendency for loco-regional and distant metastasis, researchers have shown that not all the head and neck tumors metastasize, especially at the early stage. Treating the neck which is actually node negative means incurring unnecessary costs, prolong hospital stay and causing avoidable morbidity. However, when the neck is not included in the management plan for the primary tumor in a clinically N0 neck but with unidentified micro metastases, the implication of this is poor treatment outcome with increased morbidity and mortality rate. The reality is that some patients with a clinical N0 neck indeed have no cancer cells in the cervical lymphatics and their neck must not be over treated. In employing proper oncologic therapy for the neck, one must balance the desire to preserve the present function of the neck with the wish to prevent future morbidity or loss of neck function. This requires that all persons involved in the multimodality treatment of oral carcinomas; surgeons, radiation oncologists, and medical oncologists must have a unified therapeutic modality that may achieve the desired goal, while minimizing morbidity. Although there are many available retrospective studies on oral cancers patients with clinical N0 necks and modalities of therapies but there is no consensus on the unique therapeutic approach. The benefits of elective neck dissection in patients with early oral cavity tumors have remained obscure. Few prospective studies are available but there is still inconclusive evidence on whether elective neck dissection is of any value over therapeutic neck dissection in oral cancers with N0 neck. A systematic review of prospective randomized controlled trials is needed to answer these questions owing to the inherently biased nature of the available studies. Only few of such randomized controlled trials are available in the literature and none of these studies have a study population above eighty patients. This study therefore systematically reviewed the existing published randomized controlled trials on the unresolved questions of elective versus therapeutic neck dissection in the clinically N0 neck of oral carcinoma and performed a meta-analysis of their data. The PRISMA (Preferred Reporting Items for Systematic reviews and Meta-Analyses) guideline for randomized trials was followed. The objectives were to evaluate the effectiveness of elective neck dissection in the successful reduction of neck node recurrence in oral carcinomas with clinically N0 neck, to determine and compare the disease-specific death rate of elective neck dissection to the policy of observation in early oral squamous cell carcinoma with N0 neck and to compare the survival outcome of elective neck dissection to the policy of observation in oral squamous cell carcinomas. Out of the 613 studies identified during the comprehensive search, only 4 randomized controlled trials met the criteria and were included in the metaanalysis. The total number of patients from the studies was 283. All the studies had their patients randomized into two groups; END group and OBS group. There was no statistical difference between these two groups in terms of sex and age of patients, histologic type and staging. All the trials reported on the patients´ pathologic distributions, neck recurrences and metastasis, survival and death outcome and follow-up. Despite the intention to include other factors as the primary outcome measures in this meta-analysis, the only clinically meaningful endpoint to measure the outcome benefit of elective neck dissection is the disease-specific death rate. The meta-analyses of these studies showed that elective neck dissection can effectively reduce the risk of death from the disease (disease-specific death) thereby, increasing the chance of survival {Fixed effect model RR=0.57, 95% CI of 0.36 - 0.89, p=0.014} or {Random effects model RR=0.59, 95% CI of 0.37 - 0.96, p=0.034}. It is possible that this observed pooled effect in the metaanalysis between END and OBS might have been largely influenced by the older studies. Perhaps, if the studies are conducted now that there are better investigative tools to identify and better stage neck node metastasis, this observed difference may be absent. There was also a significant evidence of reduction in neck nodal recurrences when elective neck dissection was performed. A few retrospective studies have reported on the survival benefit of elective neck dissection in early stage oral carcinoma. Only the study by Kligerman et al from this systematic review showed statistical significant evidence of disease-free survival rates benefit of elective neck dissection over observation. However, this systematic review did not show any significant survival outcome benefit of elective neck dissection over the policy of observation. In conclusion, the benefits of statistical significant reduction in disease-specific death rates and neck node recurrences may justify the need for elective neck dissection in oral carcinomas with clinically N0 neck.