We read with great interest the well-written article by Zanoletti et al., 1 which systematically compare the predictive power of the American Joint Committee on Cancer (AJCC) staging system (8th edition) with the revised Pittsburgh staging system (rPSS) in terms of prognosis for patients with temporal bone squamous cell carcinoma (TBSCC). However, we found that some of the conclusions seemed arbitrary, and further discussion is necessary.First, primary TBSCCs are actually heterogeneous carcinomas with varying origins, mainly including external auditory canal squamous cell carcinoma (EACSCC) and middle ear squamous cell carcinoma (MESCC), which are commonly investigated as a whole due to their rarity. 2,3 However, the authors declared "All patients diagnosed with primary SCC originated from the external auditory canal; peri-auricular area and adjoining sites were excluded," indicating that primary TBSCCs included in this study might mainly originate from subsites like middle ear and mastoid cavity, rather than external auditory canal. Actually, rPSS proposed by Moody et al. in 2000 4 was derived from a retrospective analysis of 32 EACSCC cases and not specifically designed for TBSCC originated from other subsites. Therefore, whether the predictive performance of rPSS validated in this study could be generalized to the whole TBSCC population still needs further discussion.Besides, the authors goes with "Statistical analysis found also significantly lower DFS, DSS, and OS for patients classified as T3-4 versus T1-2, according to the rPSS." Nevertheless, the original tumor subsite, an important confounding factor, was ignored while drawing these conclusions. Several studies have illustrated that MESCC showed worse prognosis than TBSCC originated from other subsites. [5][6][7] According to Moody's classification, all MESCCs should be classified into T3-4 status; therefore, the survival rates of advanced-stage (T3-4) TBSCC might be significantly pulled down, posing potential bias to studies taking MESCC and TBSCC originated from other subsites as a whole.Overall, we think that special attention should be paid to the original tumor subsite, an important confounding factor, in further TBSCC-related studies. At least MESCC and TBSCC originated from other subsites should be separately analyzed instead of combining them as a whole, especially when using the modified Pittsburgh classification. Besides, a proper staging scheme that specifically designed for MESCC is also needed.
DATA AVAILABILITY STATEMENTData sharing not applicable to this article as no datasets were generated or analysed during the current study.