Can Adjuvant Neck Dissection be Deferred in Locally Advanced Head and Neck Cancer Patients wtih Complete Response to Definitive Chemoradiotherapy?

Reviewer: Christine Hill, MD
Abramson Cancer Center of the University of Pennsylvania
Last Modified: October 29, 2007

Presenter: S. Yovino
Presenter's Affiliation: University of Maryland Medical System, Baltimore, MD
Type of Session: Scientific


  • With recent advances in radiotherapy techniques and their use with concurrent chemotherapy, patients with locally advanced squamous cell carcinoma of the head and neck who are felt to be at risk for significant functional deficit after surgery are often treated with definitive chemoradiotherapy as a primary treatment approach.
  • A significant portion of these patients achieve a complete response (CR) to chemoradiotherapy, and have no further detectable disease on direct visualization/ palpation of the primary tumor site, as well as radiographic evaluation of the head and neck. Whether these patients may be observed without subsequent neck dissection is controversial.
    • In many centers, patients undergo neck dissection following chemoradiotherapy, an approach that carries risk of significant morbidity, including fibrosis, wound healing problems, lymphedema, pain sydromes, and peripheral nerve damage.
    • Other centers, including the University of Maryland, reserve neck dissection for patients with recurrent or residual disease, and employ observation alone for patients who achieve a CR.
    • These two contrasting approaches have never been evaluated in a prospective, randomized setting. Several groups have completed retrospective evaluations, and recommendations from these data sets have remained contrasting, controversial, and confusing.
    • This study is a retrospective evaluation of a cohort of patients treated within the University of Maryland Medical System.

Materials and Methods

  • The patient population consisted of 120 patients treated for locally advanced squamous cell carcinoma of the head and neck with definitive chemoradiation.
  • All patients had palpable nodal disease at presentation, and 92% of patients had stage IV disease (73% had N2 or N3 disease at presentation, and 72% had T3 or T4 classification tumors). The most common primary disease site was oropharynx (75%).
  • All patients received 70.2 Gray (Gy) in 1.8 Gy fractions to known areas of gross disease, 59.4 Gy in 1.8 Gy fractions to involved nodal chains, and 50.4 Gy in 1.8 Gy fractions to uninvolved nodal chains. All patients received concurrent platinum-based chemotherapy.
  • All patients with CR, defined as absence of radiographic disease (on CT/ PET scan) and absence of palpable disease at the primary site and neck nodes, were followed with observation. Patients with partial clinically or radiographically detectable residual disease underwent neck dissection following concurrent chemoradiotherapy.


  • Median follow-up in this study was 2.5 years, and median overall survival was 53 months.
  • Complete response was achieved by 91 patients (76%) after definitive chemoradiotherapy. The achievement of CR was a significant prognostic factor for prolonged overall survival (p < 0.001).
    • Of the 91 patients who were observed after CR, 3% developed subsequent neck failures. Of these 67% (2% of the total cohort) developed isolated neck failures, and the remainder developed neck failures simultaneously with failure either distantly or at the primary site.
    • Distant metastatic disease was the most common form of first failure, and occurred in 11% of patients.
    • Advanced nodal disease at presentation was not a predictive factor for neck failure, but did predict for both failure at the primary site and development of distant metastatic disease.
  • The remainder of patients, those with detectable residual disease, underwent neck dissection. Of those who were found to have no pathologic evidence of residual disease within the neck, 80% are currently alive without evidence of disease. Of those with residual pathologic nodal disease, 80% have experienced disease recurrence.
    • The most common form of failure in patients with residual disease was the local site, followed by development of distant metastatic disease.
    • More advanced nodal stage at presentation predicted for both local failure and development of distant metastases, but was not predictive of failure within the neck.
  • Finally, of patients who achieved CR and were followed with observation, eight went on to neck dissection because of concern for recurrent disease.
    • Four of these were found to have no evidence of malignancy, and remain alive without disease.
    • The remaining four were found to have recurrent disease and have subsequently died of squamous cell carcinoma.

Author's Conclusions

  • The authors conclude that these findings support their current practice of observation of the neck for patients achieving a CR to definitive chemoradiation. 
  • Because the most common form of failure in these patients is distant metastatic disease, the authors feel that surgical intervention in the neck would be unlikely to alter outcomes significantly in these patients. 
  • The authors go on to conclude that neck dissection is still necessary in patients who do not achieve a CR, as predicting for the presence of residual pathologic disease in these patients remains difficult.

Clinical/Scientific Implications

  • The approach to the neck after definitive chemoradiotherapy remains controversial, although several groups have demonstrated that observation is likely appropriate for patients who achieve a CR. 
  • This approach is supported by this study, in which the rate of failure within the neck was < 5%. 
  • Avoiding neck dissection can spare patients significant surgical morbidity, and is likely an appropriate approach given this cost/ benefit ratio.
  • Additionally, the data presented here with regards to patients with residual disease raises the point that neck dissection after definitive chemoradiotherapy may serve more of a prognostic/ diagnostic purpose than a therapeutic – this is evidenced by the fact that the majority of patients found to have residual neck disease on neck dissection ultimately failed treatment, while those who did not have remained disease-free. 
  • Given the extremely low rate of neck failure in patients with a CR, avoiding neck dissection seems certainly to be reasonable. Furthur studies examining clinical outcomes in patients with a PR who undergo neck dissection would likely shed further light on the delivery of optimal care for this subset of patients.
  • Those patients that have a lymph node dissection for residual disease should be considered for further clinical trials due to the high failure rate.

Partially funded by an unrestricted educational grant from Bristol-Myers Squibb.