Is a laryngectomy justified for all histological subtypes of thyroid cancers as a part of gross total excision?

Abstract

To the Editor, We congratulate Flukes et al. from Memorial Sloan Kettering Cancer Center for publishing this novel retrospective study depicting outcomes for laryngectomy in locally advanced thyroid cancers. The cohort describes 30 patients with thyroid cancer who underwent total laryngectomy with the aim of gross total excision. Two‐thirds of these patients had undergone previous surgery and therefore had recurrent disease. Nearly half of the cohort had poorly differentiated or anaplastic carcinoma (47%). We stand neutral about the rationale of undertaking such extensive resections in a cohort with these demographics only striving for local control, or if this can be replicated in other centers globally. The Kaplan–Meier (K–M) survival curves are unclear. Survival statistics were calculated from the time of surgery to the occurrence of an event. Considering that two‐thirds of the cohort had received previous treatment, this significantly biases the survival outcomes and impairs their readability. In the K–M curves reported, the 5‐year overall survival plot for well‐differentiated tumors drops below that of poorly differentiated tumors at a point between 4 and 5 years. This seems inconsistent with clinical practice and in the real world where well‐differentiated tumors fare better. The authors mention that at the initial presentation, 16.7% or five patients had distant metastases, and a further 40% or 12 patients developed distant metastases during the study. Also, three patients had failed locally, and two more patients developed nodal recurrences. That means that 22 out of 30 patients had evidence of local or distant failure. Therefore, it would have been interesting to know the disease‐free survival curves for the entire cohort and not just disease‐specific survival. The authors report 5‐year overall survival (OS) and disease‐specific survival (DSS) to be 87.5% and 67.5% for well‐differentiated tumors, 77.8% and 53.3% for poorly differentiated tumors, and 37.5% and 0% for anaplastic tumors respectively. OS includes an event (death) due to any cause, while disease‐specific survival considers death (event) due to the particular disease. It follows that OS can, at best, match DSS but not exceed it. Therefore, a higher OS than DSS is unlikely to be well understood. Some of these issues with survival analysis may have been inadvertently due to the problem of calculating survival outcomes not from the date of the first diagnosis but the date of laryngectomy. While the idea of curative therapy for anaplastic thyroid carcinoma remains controversial and median survival averages a few months,2‐4 the authors should be commended that they report a 5‐year OS of 37.5% in the anaplastic subgroup alone. Even then, the five events in the five patients in the anaplastic subgroup are reported to have occurred just after 20 months. Therefore the conclusion requires further clarification. Lastly, the authors also mention that eight patients received systemic therapy under the broad heading of adjuvant therapy, where, in fact this was given due to distant metastasis. To the best of our knowledge, nonmetastatic thyroid cancers may benefit from adjuvant radioactive iodine (RAI) or radiotherapy, and systemic therapy is usually reserved for RAI‐refractory advanced or metastatic cases.