BRAF Mutation and Thyroid Cancer Recurrence

We recently published a large multicenter study that unequivocally demonstrated an independent association between BRAF mutation and the recurrence of papillary thyroid cancer (PTC) in all PTCs, conventional PTC, and follicular-variant PTC, as exemplified by a hazard ratio of 3.20 (95% CI, 1.46 to 7.02) in follicularvariant PTC after a full adjustment for conventional clinicopathologic risk factors. 1 Although the significance of this study was generally appreciated in the accompanying editorial by Yarchoan et al, 2 the authors raised several issues that require clarification.Although Yarchoan et al 2 recognized the strong association of BRAF mutation with PTC-related mortality that was demonstrated in another study of ours, 3 they echoed Cappola and Mandel 4 in questioning the prognostic value of BRAF mutation for PTC-related mortality because of the loss of this association on multivariable adjustment for aggressive tumor behaviors (eg, extrathyroidal invasion and lymph node metastasis).This view originates from the lack of understanding that, as discussed previously, 5 this prognostic value of BRAF mutation lies in its accelerating mortality by promoting and synergizing aggressive tumor behaviors through various molecular mechanisms, 6 resulting in shortened patient survival; as such, this mortality-promoting effect of BRAF mutation, albeit important, cannot be independent of aggressive tumor behaviors.Notably, the effect of BRAF mutation on mortality remained strongly significant after adjustment for patient age, sex, and stratification for medical center variations, and within virtually every conventional risk category, BRAF mutation caused a significantly higher mortality beyond the conventional risk. 3This represents an important prognostic value of BRAF mutation.Given the dependence of BRAF mutation on tumor behaviors in affecting mortality, Yarchoan et al 2 seemed to be puzzled by the independent association of BRAF mutation with PTC recurrence. 1 The explanation is simple: patient death as a result of thyroid cancer requires tumor progression to a significant extent, which takes time; consequently, mortality was appreciable within the follow-up times of most studies mainly in those patients who already had clinically aggressive tumor behaviors at initial surgery.In contrast, PTC recurrence occurs earlier, often representing a progression from initially occult disease (eg, lymph node metastasis).The effect of BRAF mutation on such recurrence can thus be independent of overt initial aggressive tumor behaviors in conventional multivariable models. 1 Theoretically, recurrent disease in patients with PTC whose tumors do not demonstrate overt initial aggressive behavior, if given enough time, could progress to a sufficient extent to result in patient death, thereby displaying a so-called independent impact of BRAF mutation on mortality.Yarchoan et al 2 suggested that the prognostic value of BRAF mutation could not be established without prospective studies.Indeed, prospective studies are often superior to retrospective studies in