I appreciate the insightful remarks by Dr Hsing et al. Unfortunately, I do not agree with them.

First, the statement that radiotherapy (RT) alone or combined with neck dissection (ND) results in a high probability of cure is not meant to imply that all patients should be treated with RT alone. Patients with an incomplete response to RT have an improved neck control rate if an ND is performed; the likelihood of an incomplete response increases with N-stage . Patients who have a complete response after RT have a very low chance of an isolated recurrence in the neck, and the ND may be safely withheld regardless of the initial neck stage . The proposal that all patients should undergo an ND to “stage the neck” is interesting but would mean that many patients would undergo an unnecessary ND and most would receive RT and concomitant chemotherapy regardless of the pathologic findings. An initial ND is indicated only for the small subset of patients with N ₁ disease who might be suitable for an ND alone and close follow-up and for those in whom a diagnosis cannot be obtained after several fine needle aspirates of the neck node(s). Otherwise, evaluation for a planned ND should take place after RT because many patients will not need to undergo the procedure and those who do may require a more limited dissection .

Dr. Hsing et al state that we did not provide data showing that eliminating the larynx and hypopharynx from the RT fields reduces toxicity, and indeed, we did not provide such data. However, the likelihood of an occult primary cancer in these locations is remote, at least in the United States, and it makes no sense to unnecessarily irradiate these mucosal sites that do not require treatment . Indeed, none of the 28 patients treated to the oropharynx and nasopharynx with larynx-sparing RT developed a mucosal failure . One might suspect that irradiating larger volumes of mucosa would increase toxicity, and this is likely the case.

As to the diagnostic evaluation, we routinely perform a tonsillectomy at direct laryngoscopy if there is enough tonsillar tissue to warrant the procedure. This study spans a significant period, and some patients treated earlier in our series did not undergo the procedure. In addition, many patients who have undergone a tonsillectomy as a child have essentially no tonsillar tissue, and a tonsillectomy is not warranted in those circumstances. We have not found positron emission tomography (PET) to be particularly useful after a thorough physical examination and computed tomography (CT). PET–CT has a high likelihood of a false positive in the oropharynx, the very site where the occult cancer is located in any event. In our experience, very few patients will have the primary lesion diagnosed only on the PET scan in the absence of other findings . PET–CT may be useful to identify distant metastases in the subset of patients with advanced neck disease particularly for those with positive nodes in the low neck .

In conclusion, RT alone or combined with concomitant chemotherapy is a safe alternative for patients who have a complete response after treatment. Adding an ND does not improve outcome and only increases toxicity. Larynx-sparing RT is likely less morbid than techniques that include the larynx and hypopharynx in the RT fields and results in a low probability of mucosal failure. Tonsillectomy should be performed at the time of direct laryngoscopy, if there is enough tissue present to warrant the procedure. Positron emission tomography–computed tomography is not necessary as a routine part of the diagnostic evaluation for many patients. PET scans rarely detect the occult primary cancer in the absence of suspicious findings on physical examination and/or contrast-enhanced CT. On the other hand, they will increase the cost of the diagnostic evaluation in every patient in whom the PET scan is obtained.