In Response to Grégoire et al.

To the Editors,

We thank Grégoire et al for their comments on our study and for their impact in advancing the field of head and neck cancer (HNC) radiation therapy throughout the past 2 decades. Their group has made one of the most significant contributions in HNC by developing standardized radiation therapy approaches in the most anatomically complex tumor site treated in radiation oncology.

Grégoire et al raise the concern that we are confusing the definition of clinical target volume (CTV) and planning target volume (PTV). To clarify, we are not doing so: We do not suggest that the PTV margin be deliberately used to cover microscopic disease. In our clinically delivered radiation therapy plans, we used a 5 mm high-dose CTV margin for microscopic disease, with a separate 5 mm PTV margin. The definitions of CTV and PTV are well ingrained in our specialty, and we agree that they must be kept distinct.

Instead, our article asked a very narrow question: When using the aforementioned CTV and PTV margins, is it dosimetrically necessary to place a low-dose CTV 5 mm (or less) from a high-dose CTV? Our study suggests it is not. Of course, radiation oncologists are welcome to do so, but it probably does not affect the final plan. Consider the following analogy, albeit an extreme one: If researchers were to discover microscopic cancer cells 5 mm outside the high-dose CTV that were highly radiation sensitive and only required 5 Gy for eradication, would we start creating a CTV5Gy immediately outside the CTV70 Gy? A purist can argue that we should as per International Commission on Radiation Units and Measurements definitions, but perhaps a pragmatic radiation oncologist would recognize the CTV5Gy cannot be underdosed in such close proximity, and the extra contours would waste time and may lead to errors.

As acknowledged in our article, our findings may not apply to proton therapy or smaller PTV margins and may be affected by the degree of elective nodal coverage, which contributes dose in the proximity of the primary tumor.

The larger question, not addressed in our article but raised in the letter, is how to decide on CTV margins in general. Grégoire et al take a very reasoned approach, basing CTV margins on the distance of microscopic extension found after surgery, a rationale that we summarized in our article. However, to our knowledge, no other tumor site treated with external beam radiation therapy uses separate high- and low-dose CTVs around a primary tumor. In lung cancer, for example, guideline authors recognized that surgical series demonstrate microscopic extension of 6 to 8 mm or more, but they concluded that the clinical relevance of those findings is uncertain and recommended a 5 mm CTV margin. This leads to a discrepancy within our specialty in the rationalization of CTV margins.

Which approach is correct? The answer is unknown. Also unknown are the trade-offs that occur as CTVs get smaller: At the extreme, omitting all CTVs around the primary tumor HNC may lead to higher local recurrence rates but could substantially improve long-term quality of life. This trade-off could be quantified in a randomized trial. Such a trial could provide strong evidence to test the CTV concept, which we have widely adopted in radiation oncology based on a reasonable scientific rationale but without high-level evidence.