Letter to the editor regarding ACR appropriateness criteria for ipsilateral radiation for squamous cell carcinoma of the tonsil

To the Editor: We write to dispute one of the key recommendations in the article entitled "ACR Appropriateness Criteria R ipsilateral radiation for squamous cell carcinoma of the tonsil'' recently published in Head & Neck. As stated in the abstract of the article and on the ACR website, ACR Appropriateness Criteria are designed to provide evidence-based guidelines to assist providers in making the most appropriate treatment decision for a specific clinical condition. Bearing in mind this definition, we are at a loss to understand how the ACR panel arrived at their recommendation that "For nodal stage N2b or greater, we recommend bilateral neck irradiation, regardless of the extent of soft palate or base of tongue invasion.'' The associated box headed "Variant 5'' reinforces this recommendation. In it, bilateral neck treatment of a patient with T1N2b disease is accorded a rating of 7 (usually appropriate) whereas ipsilateral treatment is rated 3 (usually inappropriate). In contrast to the strength of the panel's recommendation, the only data cited in the article that bear directly on this matter do not support it. In the prospective study of Rusthoven et al, none of 13 patients with neck stage 2b who were treated ipsilaterally failed in the contralateral neck. No other study specifically addresses the question and the ACR panel provides no evidence at all to support the need for contralateral neck treatment for patients with well-lateralized primaries based solely on their neck stage. Given the paucity of data, we agree that there might be an increased risk of contralateral failure in such patients, but it is fatuous to claim an evidence-base for the ACR panel's recommendation. Surely, the most appropriate conclusion of the panel in this regard would have been that there is no direct evidence of the need for contralateral treatment but that the matter is unresolved. At issue is whether the laterality of the primary is a more powerful determinant of contralateral subclinical neck involvement than advanced ipsilateral nodal stage. It is significant that in the large series of O'Sullivan et al, all 3 contralateral failures (with primary controlled) occurred in patients whose primary tumors extended close to midline and none of 39 patients with N2 to 3 nodal disease failed in the contralateral neck. In the study of Lim et al, more advanced T classification (a crude surrogate for laterality) was the only statistically significant predictor for contralateral nodal involvement on surgical dissection. Ipsilateral nodal stage was not a significant factor. The larger surgical series reported by Olzowy et al showed a similar strong T classification dependency for the risk of contralateral nodal involvement for all subsites including tonsillar cancer. When all oropharyngeal sites were combined, the risk of contralateral nodal involvement was significantly increased when 2 or more ipsilateral nodes were involved. This is not surprising, however, given that the tongue base, soft palate, and pharyngeal wall primaries nearly always approach or cross the midline. No analysis of the relevance of ipsilateral nodal stage was reported specifically for tonsillar cancer. Our own unpublished data support the position that ipsilateral nodal stage is not an independently powerful predictor of contralateral subclinical disease in tonsillar cancer. Of 57 patients who were treated unilaterally for various reasons between 1990 and 2002, 47 had documented lateralized tumors ( 1 cm from midline). Nineteen patients had nodal stage N2 (14) or N3 (5) and none recurred in the opposite neck with a median follow-up of 8.5 years. The question of whether to treat the contralateral neck is not a trivial one in terms of patient morbidity. Even with intensity modulated radiation therapy, both acute and late toxicity are significantly increased with bilateral irradiation. With longterm survival currently good for tonsillar cancer, it is incumbent on us not to inflict unnecessary morbidity on our patients. Given the high proportion of patients who present with ipsilateral neck involvement, we believe that a formal prospective study of the safety of unilateral irradiation in patients with lateralized primaries is justified. Such a study would also shed light on the relevance of human papillomavirus status, which the ACR panel quite "appropriately'' refrained from using as the basis for any recommendation. In the meantime, it would seem to us eminently appropriate to offer unilateral treatment to patients with lateralized primaries regardless of neck stage in circumstances in which the patient's ability or willingness to tolerate more comprehensive treatment is in question.